821 files changed, 81058 insertions, 12256 deletions

diff --git a/.gitignore b/.gitignore
index fd3a35592543..b17dcd217d83 100644
--- a/.gitignore
+++ b/.gitignore
@@ -33,6 +33,7 @@
 *.lzo
 *.patch
 *.gcno
+*.ll
 modules.builtin
 Module.symvers
 *.dwo
@@ -112,3 +113,6 @@ all.config
 
 # Kdevelop4
 *.kdev4
+
+# fetched Android config fragments
+android/configs/android-*.cfg
diff --git a/Documentation/ABI/testing/sysfs-class-dual-role-usb b/Documentation/ABI/testing/sysfs-class-dual-role-usb
new file mode 100644
index 000000000000..a900fd75430c
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-class-dual-role-usb
@@ -0,0 +1,71 @@
+What:		/sys/class/dual_role_usb/.../
+Date:		June 2015
+Contact:	Badhri Jagan Sridharan<badhri@google.com>
+Description:
+		Provide a generic interface to monitor and change
+		the state of dual role usb ports. The name here
+		refers to the name mentioned in the
+		dual_role_phy_desc that is passed while registering
+		the dual_role_phy_intstance through
+		devm_dual_role_instance_register.
+
+What:           /sys/class/dual_role_usb/.../supported_modes
+Date:           June 2015
+Contact:        Badhri Jagan Sridharan<badhri@google.com>
+Description:
+		This is a static node, once initialized this
+		is not expected to change during runtime. "dfp"
+		refers to "downstream facing port" i.e. port can
+		only act as host. "ufp" refers to "upstream
+		facing port" i.e. port can only act as device.
+		"dfp ufp" refers to "dual role port" i.e. the port
+		can either be a host port or a device port.
+
+What:		/sys/class/dual_role_usb/.../mode
+Date:		June 2015
+Contact:	Badhri Jagan Sridharan<badhri@google.com>
+Description:
+		The mode node refers to the current mode in which the
+		port is operating. "dfp" for host ports. "ufp" for device
+		ports and "none" when cable is not connected.
+
+		On devices where the USB mode is software-controllable,
+		userspace can change the mode by writing "dfp" or "ufp".
+		On devices where the USB mode is fixed in hardware,
+		this attribute is read-only.
+
+What:		/sys/class/dual_role_usb/.../power_role
+Date:		June 2015
+Contact:	Badhri Jagan Sridharan<badhri@google.com>
+Description:
+		The power_role node mentions whether the port
+		is "sink"ing or "source"ing power. "none" if
+		they are not connected.
+
+		On devices implementing USB Power Delivery,
+		userspace can control the power role by writing "sink" or
+		"source". On devices without USB-PD, this attribute is
+		read-only.
+
+What:		/sys/class/dual_role_usb/.../data_role
+Date:		June 2015
+Contact:	Badhri Jagan Sridharan<badhri@google.com>
+Description:
+		The data_role node mentions whether the port
+		is acting as "host" or "device" for USB data connection.
+		"none" if there is no active data link.
+
+		On devices implementing USB Power Delivery, userspace
+		can control the data role by writing "host" or "device".
+		On devices without USB-PD, this attribute is read-only
+
+What:		/sys/class/dual_role_usb/.../powers_vconn
+Date:		June 2015
+Contact:	Badhri Jagan Sridharan<badhri@google.com>
+Description:
+		The powers_vconn node mentions whether the port
+		is supplying power for VCONN pin.
+
+		On devices with software control of VCONN,
+		userspace can disable the power supply to VCONN by writing "n",
+		or enable the power supply by writing "y".
diff --git a/Documentation/ABI/testing/sysfs-fs-f2fs b/Documentation/ABI/testing/sysfs-fs-f2fs
index 0345f2d1c727..500c60403653 100644
--- a/Documentation/ABI/testing/sysfs-fs-f2fs
+++ b/Documentation/ABI/testing/sysfs-fs-f2fs
@@ -57,6 +57,15 @@ Contact:	"Jaegeuk Kim" <jaegeuk.kim@samsung.com>
 Description:
 		 Controls the issue rate of small discard commands.
 
+What:          /sys/fs/f2fs/<disk>/discard_granularity
+Date:          July 2017
+Contact:       "Chao Yu" <yuchao0@huawei.com>
+Description:
+		Controls discard granularity of inner discard thread, inner thread
+		will not issue discards with size that is smaller than granularity.
+		The unit size is one block, now only support configuring in range
+		of [1, 512].
+
 What:		/sys/fs/f2fs/<disk>/max_victim_search
 Date:		January 2014
 Contact:	"Jaegeuk Kim" <jaegeuk.kim@samsung.com>
@@ -92,3 +101,47 @@ Date:		October 2015
 Contact:	"Chao Yu" <chao2.yu@samsung.com>
 Description:
 		 Controls the count of nid pages to be readaheaded.
+
+What:		/sys/fs/f2fs/<disk>/dirty_nats_ratio
+Date:		January 2016
+Contact:	"Chao Yu" <chao2.yu@samsung.com>
+Description:
+		 Controls dirty nat entries ratio threshold, if current
+		 ratio exceeds configured threshold, checkpoint will
+		 be triggered for flushing dirty nat entries.
+
+What:		/sys/fs/f2fs/<disk>/lifetime_write_kbytes
+Date:		January 2016
+Contact:	"Shuoran Liu" <liushuoran@huawei.com>
+Description:
+		 Shows total written kbytes issued to disk.
+
+What:		/sys/fs/f2fs/<disk>/inject_rate
+Date:		May 2016
+Contact:	"Sheng Yong" <shengyong1@huawei.com>
+Description:
+		 Controls the injection rate.
+
+What:		/sys/fs/f2fs/<disk>/inject_type
+Date:		May 2016
+Contact:	"Sheng Yong" <shengyong1@huawei.com>
+Description:
+		 Controls the injection type.
+
+What:		/sys/fs/f2fs/<disk>/reserved_blocks
+Date:		June 2017
+Contact:	"Chao Yu" <yuchao0@huawei.com>
+Description:
+		 Controls current reserved blocks in system.
+
+What:		/sys/fs/f2fs/<disk>/gc_urgent
+Date:		August 2017
+Contact:	"Jaegeuk Kim" <jaegeuk@kernel.org>
+Description:
+		 Do background GC agressively
+
+What:		/sys/fs/f2fs/<disk>/gc_urgent_sleep_time
+Date:		August 2017
+Contact:	"Jaegeuk Kim" <jaegeuk@kernel.org>
+Description:
+		 Controls sleep time of GC urgent mode
diff --git a/Documentation/ABI/testing/sysfs-kernel-wakeup_reasons b/Documentation/ABI/testing/sysfs-kernel-wakeup_reasons
new file mode 100644
index 000000000000..acb19b91c192
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-kernel-wakeup_reasons
@@ -0,0 +1,16 @@
+What:		/sys/kernel/wakeup_reasons/last_resume_reason
+Date:		February 2014
+Contact:	Ruchi Kandoi <kandoiruchi@google.com>
+Description:
+		The /sys/kernel/wakeup_reasons/last_resume_reason is
+		used to report wakeup reasons after system exited suspend.
+
+What:		/sys/kernel/wakeup_reasons/last_suspend_time
+Date:		March 2015
+Contact:	jinqian <jinqian@google.com>
+Description:
+		The /sys/kernel/wakeup_reasons/last_suspend_time is
+		used to report time spent in last suspend cycle. It contains
+		two numbers (in seconds) separated by space. First number is
+		the time spent in suspend and resume processes. Second number
+		is the time spent in sleep state.
\ No newline at end of file
diff --git a/Documentation/android.txt b/Documentation/android.txt
new file mode 100644
index 000000000000..0f40a78b045f
--- /dev/null
+++ b/Documentation/android.txt
@@ -0,0 +1,121 @@
+				=============
+				A N D R O I D
+				=============
+
+Copyright (C) 2009 Google, Inc.
+Written by Mike Chan <mike@android.com>
+
+CONTENTS:
+---------
+
+1. Android
+  1.1 Required enabled config options
+  1.2 Required disabled config options
+  1.3 Recommended enabled config options
+2. Contact
+
+
+1. Android
+==========
+
+Android (www.android.com) is an open source operating system for mobile devices.
+This document describes configurations needed to run the Android framework on
+top of the Linux kernel.
+
+To see a working defconfig look at msm_defconfig or goldfish_defconfig
+which can be found at http://android.git.kernel.org in kernel/common.git
+and kernel/msm.git
+
+
+1.1 Required enabled config options
+-----------------------------------
+After building a standard defconfig, ensure that these options are enabled in
+your .config or defconfig if they are not already. Based off the msm_defconfig.
+You should keep the rest of the default options enabled in the defconfig
+unless you know what you are doing.
+
+ANDROID_PARANOID_NETWORK
+ASHMEM
+CONFIG_FB_MODE_HELPERS
+CONFIG_FONT_8x16
+CONFIG_FONT_8x8
+CONFIG_YAFFS_SHORT_NAMES_IN_RAM
+DAB
+EARLYSUSPEND
+FB
+FB_CFB_COPYAREA
+FB_CFB_FILLRECT
+FB_CFB_IMAGEBLIT
+FB_DEFERRED_IO
+FB_TILEBLITTING
+HIGH_RES_TIMERS
+INOTIFY
+INOTIFY_USER
+INPUT_EVDEV
+INPUT_GPIO
+INPUT_MISC
+LEDS_CLASS
+LEDS_GPIO
+LOCK_KERNEL
+LkOGGER
+LOW_MEMORY_KILLER
+MISC_DEVICES
+NEW_LEDS
+NO_HZ
+POWER_SUPPLY
+PREEMPT
+RAMFS
+RTC_CLASS
+RTC_LIB
+SWITCH
+SWITCH_GPIO
+TMPFS
+UID_STAT
+UID16
+USB_FUNCTION
+USB_FUNCTION_ADB
+USER_WAKELOCK
+VIDEO_OUTPUT_CONTROL
+WAKELOCK
+YAFFS_AUTO_YAFFS2
+YAFFS_FS
+YAFFS_YAFFS1
+YAFFS_YAFFS2
+
+
+1.2 Required disabled config options
+------------------------------------
+CONFIG_YAFFS_DISABLE_LAZY_LOAD
+DNOTIFY
+
+
+1.3 Recommended enabled config options
+------------------------------
+ANDROID_PMEM
+PSTORE_CONSOLE
+PSTORE_RAM
+SCHEDSTATS
+DEBUG_PREEMPT
+DEBUG_MUTEXES
+DEBUG_SPINLOCK_SLEEP
+DEBUG_INFO
+FRAME_POINTER
+CPU_FREQ
+CPU_FREQ_TABLE
+CPU_FREQ_DEFAULT_GOV_ONDEMAND
+CPU_FREQ_GOV_ONDEMAND
+CRC_CCITT
+EMBEDDED
+INPUT_TOUCHSCREEN
+I2C
+I2C_BOARDINFO
+LOG_BUF_SHIFT=17
+SERIAL_CORE
+SERIAL_CORE_CONSOLE
+
+
+2. Contact
+==========
+website: http://android.git.kernel.org
+
+mailing-lists: android-kernel@googlegroups.com
diff --git a/Documentation/block/00-INDEX b/Documentation/block/00-INDEX
index e840b47613f7..bc5148757edb 100644
--- a/Documentation/block/00-INDEX
+++ b/Documentation/block/00-INDEX
@@ -26,3 +26,9 @@ switching-sched.txt
 	- Switching I/O schedulers at runtime
 writeback_cache_control.txt
 	- Control of volatile write back caches
+mmc-max-speed.txt
+	- eMMC layer speed simulation, related to /sys/block/mmcblk*/
+          attributes:
+            max_read_speed
+            max_write_speed
+            cache_size
diff --git a/Documentation/block/mmc-max-speed.txt b/Documentation/block/mmc-max-speed.txt
new file mode 100644
index 000000000000..3f052b9fb999
--- /dev/null
+++ b/Documentation/block/mmc-max-speed.txt
@@ -0,0 +1,38 @@
+eMMC Block layer simulation speed controls in /sys/block/mmcblk*/
+===============================================
+
+Turned on with CONFIG_MMC_SIMULATE_MAX_SPEED which enables MMC device speed
+limiting. Used to test and simulate the behavior of the system when
+confronted with a slow MMC.
+
+Enables max_read_speed, max_write_speed and cache_size attributes and module
+default parameters to control the write or read maximum KB/second speed
+behaviors.
+
+NB: There is room for improving the algorithm for aspects tied directly to
+eMMC specific behavior. For instance, wear leveling and stalls from an
+exhausted erase pool. We would expect that if there was a need to provide
+similar speed simulation controls to other types of block devices, aspects of
+their behavior are modelled separately (e.g. head seek times, heat assist,
+shingling and rotational latency).
+
+/sys/block/mmcblk0/max_read_speed:
+
+Number of KB/second reads allowed to the block device. Used to test and
+simulate the behavior of the system when confronted with a slow reading MMC.
+Set to 0 or "off" to place no speed limit.
+
+/sys/block/mmcblk0/max_write_speed:
+
+Number of KB/second writes allowed to the block device. Used to test and
+simulate the behavior of the system when confronted with a slow writing MMC.
+Set to 0 or "off" to place no speed limit.
+
+/sys/block/mmcblk0/cache_size:
+
+Number of MB of high speed memory or high speed SLC cache expected on the
+eMMC device being simulated. Used to help simulate the write-back behavior
+more accurately. The assumption is the cache has no delay, but draws down
+in the background to the MLC/TLC primary store at the max_write_speed rate.
+Any write speed delays will show up when the cache is full, or when an I/O
+request to flush is issued.
diff --git a/Documentation/cpu-freq/governors.txt b/Documentation/cpu-freq/governors.txt
index c15aa75f5227..ac8a37e0c76a 100644
--- a/Documentation/cpu-freq/governors.txt
+++ b/Documentation/cpu-freq/governors.txt
@@ -28,6 +28,7 @@ Contents:
 2.3  Userspace
 2.4  Ondemand
 2.5  Conservative
+2.6  Interactive
 
 3.   The Governor Interface in the CPUfreq Core
 
@@ -218,6 +219,90 @@ a decision on when to decrease the frequency while running in any
 speed. Load for frequency increase is still evaluated every
 sampling rate.
 
+2.6 Interactive
+---------------
+
+The CPUfreq governor "interactive" is designed for latency-sensitive,
+interactive workloads. This governor sets the CPU speed depending on
+usage, similar to "ondemand" and "conservative" governors, but with a
+different set of configurable behaviors.
+
+The tuneable values for this governor are:
+
+target_loads: CPU load values used to adjust speed to influence the
+current CPU load toward that value.  In general, the lower the target
+load, the more often the governor will raise CPU speeds to bring load
+below the target.  The format is a single target load, optionally
+followed by pairs of CPU speeds and CPU loads to target at or above
+those speeds.  Colons can be used between the speeds and associated
+target loads for readability.  For example:
+
+   85 1000000:90 1700000:99
+
+targets CPU load 85% below speed 1GHz, 90% at or above 1GHz, until
+1.7GHz and above, at which load 99% is targeted.  If speeds are
+specified these must appear in ascending order.  Higher target load
+values are typically specified for higher speeds, that is, target load
+values also usually appear in an ascending order. The default is
+target load 90% for all speeds.
+
+min_sample_time: The minimum amount of time to spend at the current
+frequency before ramping down. Default is 80000 uS.
+
+hispeed_freq: An intermediate "hi speed" at which to initially ramp
+when CPU load hits the value specified in go_hispeed_load.  If load
+stays high for the amount of time specified in above_hispeed_delay,
+then speed may be bumped higher.  Default is the maximum speed
+allowed by the policy at governor initialization time.
+
+go_hispeed_load: The CPU load at which to ramp to hispeed_freq.
+Default is 99%.
+
+above_hispeed_delay: When speed is at or above hispeed_freq, wait for
+this long before raising speed in response to continued high load.
+The format is a single delay value, optionally followed by pairs of
+CPU speeds and the delay to use at or above those speeds.  Colons can
+be used between the speeds and associated delays for readability.  For
+example:
+
+   80000 1300000:200000 1500000:40000
+
+uses delay 80000 uS until CPU speed 1.3 GHz, at which speed delay
+200000 uS is used until speed 1.5 GHz, at which speed (and above)
+delay 40000 uS is used.  If speeds are specified these must appear in
+ascending order.  Default is 20000 uS.
+
+timer_rate: Sample rate for reevaluating CPU load when the CPU is not
+idle.  A deferrable timer is used, such that the CPU will not be woken
+from idle to service this timer until something else needs to run.
+(The maximum time to allow deferring this timer when not running at
+minimum speed is configurable via timer_slack.)  Default is 20000 uS.
+
+timer_slack: Maximum additional time to defer handling the governor
+sampling timer beyond timer_rate when running at speeds above the
+minimum.  For platforms that consume additional power at idle when
+CPUs are running at speeds greater than minimum, this places an upper
+bound on how long the timer will be deferred prior to re-evaluating
+load and dropping speed.  For example, if timer_rate is 20000uS and
+timer_slack is 10000uS then timers will be deferred for up to 30msec
+when not at lowest speed.  A value of -1 means defer timers
+indefinitely at all speeds.  Default is 80000 uS.
+
+boost: If non-zero, immediately boost speed of all CPUs to at least
+hispeed_freq until zero is written to this attribute.  If zero, allow
+CPU speeds to drop below hispeed_freq according to load as usual.
+Default is zero.
+
+boostpulse: On each write, immediately boost speed of all CPUs to
+hispeed_freq for at least the period of time specified by
+boostpulse_duration, after which speeds are allowed to drop below
+hispeed_freq according to load as usual.
+
+boostpulse_duration: Length of time to hold CPU speed at hispeed_freq
+on a write to boostpulse, before allowing speed to drop according to
+load as usual.  Default is 80000 uS.
+
+
 3. The Governor Interface in the CPUfreq Core
 =============================================
 
diff --git a/Documentation/device-mapper/boot.txt b/Documentation/device-mapper/boot.txt
new file mode 100644
index 000000000000..adcaad5e5e32
--- /dev/null
+++ b/Documentation/device-mapper/boot.txt
@@ -0,0 +1,42 @@
+Boot time creation of mapped devices
+===================================
+
+It is possible to configure a device mapper device to act as the root
+device for your system in two ways.
+
+The first is to build an initial ramdisk which boots to a minimal
+userspace which configures the device, then pivot_root(8) in to it.
+
+For simple device mapper configurations, it is possible to boot directly
+using the following kernel command line:
+
+dm="<name> <uuid> <ro>,table line 1,...,table line n"
+
+name = the name to associate with the device
+	after boot, udev, if used, will use that name to label
+	the device node.
+uuid = may be 'none' or the UUID desired for the device.
+ro = may be "ro" or "rw".  If "ro", the device and device table will be
+	marked read-only.
+
+Each table line may be as normal when using the dmsetup tool except for
+two variations:
+1. Any use of commas will be interpreted as a newline
+2. Quotation marks cannot be escaped and cannot be used without
+   terminating the dm= argument.
+
+Unless renamed by udev, the device node created will be dm-0 as the
+first minor number for the device-mapper is used during early creation.
+
+Example
+=======
+
+- Booting to a linear array made up of user-mode linux block devices:
+
+  dm="lroot none 0, 0 4096 linear 98:16 0, 4096 4096 linear 98:32 0" \
+  root=/dev/dm-0
+
+Will boot to a rw dm-linear target of 8192 sectors split across two
+block devices identified by their major:minor numbers.  After boot, udev
+will rename this target to /dev/mapper/lroot (depending on the rules).
+No uuid was assigned.
diff --git a/Documentation/device-mapper/verity.txt b/Documentation/device-mapper/verity.txt
index e15bc1a0fb98..89fd8f9a259f 100644
--- a/Documentation/device-mapper/verity.txt
+++ b/Documentation/device-mapper/verity.txt
@@ -18,11 +18,11 @@ Construction Parameters
 
     0 is the original format used in the Chromium OS.
       The salt is appended when hashing, digests are stored continuously and
-      the rest of the block is padded with zeros.
+      the rest of the block is padded with zeroes.
 
     1 is the current format that should be used for new devices.
       The salt is prepended when hashing and each digest is
-      padded with zeros to the power of two.
+      padded with zeroes to the power of two.
 
 <dev>
     This is the device containing data, the integrity of which needs to be
@@ -79,6 +79,37 @@ restart_on_corruption
     not compatible with ignore_corruption and requires user space support to
     avoid restart loops.
 
+ignore_zero_blocks
+    Do not verify blocks that are expected to contain zeroes and always return
+    zeroes instead. This may be useful if the partition contains unused blocks
+    that are not guaranteed to contain zeroes.
+
+use_fec_from_device <fec_dev>
+    Use forward error correction (FEC) to recover from corruption if hash
+    verification fails. Use encoding data from the specified device. This
+    may be the same device where data and hash blocks reside, in which case
+    fec_start must be outside data and hash areas.
+
+    If the encoding data covers additional metadata, it must be accessible
+    on the hash device after the hash blocks.
+
+    Note: block sizes for data and hash devices must match. Also, if the
+    verity <dev> is encrypted the <fec_dev> should be too.
+
+fec_roots <num>
+    Number of generator roots. This equals to the number of parity bytes in
+    the encoding data. For example, in RS(M, N) encoding, the number of roots
+    is M-N.
+
+fec_blocks <num>
+    The number of encoding data blocks on the FEC device. The block size for
+    the FEC device is <data_block_size>.
+
+fec_start <offset>
+    This is the offset, in <data_block_size> blocks, from the start of the
+    FEC device to the beginning of the encoding data.
+
+
 Theory of operation
 ===================
 
@@ -98,6 +129,11 @@ per-block basis. This allows for a lightweight hash computation on first read
 into the page cache. Block hashes are stored linearly, aligned to the nearest
 block size.
 
+If forward error correction (FEC) support is enabled any recovery of
+corrupted data will be verified using the cryptographic hash of the
+corresponding data. This is why combining error correction with
+integrity checking is essential.
+
 Hash Tree
 ---------
 
diff --git a/Documentation/devicetree/bindings/goldfish/audio.txt b/Documentation/devicetree/bindings/goldfish/audio.txt
new file mode 100644
index 000000000000..d043fda433ba
--- /dev/null
+++ b/Documentation/devicetree/bindings/goldfish/audio.txt
@@ -0,0 +1,17 @@
+Android Goldfish Audio
+
+Android goldfish audio device generated by android emulator.
+
+Required properties:
+
+- compatible : should contain "google,goldfish-audio" to match emulator
+- reg        : <registers mapping>
+- interrupts : <interrupt mapping>
+
+Example:
+
+	goldfish_audio@9030000 {
+		compatible = "google,goldfish-audio";
+		reg = <0x9030000 0x100>;
+		interrupts = <0x4>;
+	};
diff --git a/Documentation/devicetree/bindings/goldfish/battery.txt b/Documentation/devicetree/bindings/goldfish/battery.txt
new file mode 100644
index 000000000000..4fb613933214
--- /dev/null
+++ b/Documentation/devicetree/bindings/goldfish/battery.txt
@@ -0,0 +1,17 @@
+Android Goldfish Battery
+
+Android goldfish battery device generated by android emulator.
+
+Required properties:
+
+- compatible : should contain "google,goldfish-battery" to match emulator
+- reg        : <registers mapping>
+- interrupts : <interrupt mapping>
+
+Example:
+
+	goldfish_battery@9020000 {
+		compatible = "google,goldfish-battery";
+		reg = <0x9020000 0x1000>;
+		interrupts = <0x3>;
+	};
diff --git a/Documentation/devicetree/bindings/goldfish/events.txt b/Documentation/devicetree/bindings/goldfish/events.txt
new file mode 100644
index 000000000000..5babf46317a4
--- /dev/null
+++ b/Documentation/devicetree/bindings/goldfish/events.txt
@@ -0,0 +1,17 @@
+Android Goldfish Events Keypad
+
+Android goldfish events keypad device generated by android emulator.
+
+Required properties:
+
+- compatible : should contain "google,goldfish-events-keypad" to match emulator
+- reg        : <registers mapping>
+- interrupts : <interrupt mapping>
+
+Example:
+
+	goldfish-events@9040000 {
+		compatible = "google,goldfish-events-keypad";
+		reg = <0x9040000 0x1000>;
+		interrupts = <0x5>;
+	};
diff --git a/Documentation/devicetree/bindings/goldfish/tty.txt b/Documentation/devicetree/bindings/goldfish/tty.txt
new file mode 100644
index 000000000000..82648278da77
--- /dev/null
+++ b/Documentation/devicetree/bindings/goldfish/tty.txt
@@ -0,0 +1,17 @@
+Android Goldfish TTY
+
+Android goldfish tty device generated by android emulator.
+
+Required properties:
+
+- compatible : should contain "google,goldfish-tty" to match emulator
+- reg        : <registers mapping>
+- interrupts : <interrupt mapping>
+
+Example:
+
+	goldfish_tty@1f004000 {
+		compatible = "google,goldfish-tty";
+		reg = <0x1f004000 0x1000>;
+		interrupts = <0xc>;
+	};
diff --git a/Documentation/devicetree/bindings/misc/memory-state-time.txt b/Documentation/devicetree/bindings/misc/memory-state-time.txt
new file mode 100644
index 000000000000..c99a506c030d
--- /dev/null
+++ b/Documentation/devicetree/bindings/misc/memory-state-time.txt
@@ -0,0 +1,8 @@
+Memory bandwidth and frequency state tracking
+
+Required properties:
+- compatible : should be:
+       "memory-state-time"
+- freq-tbl: Should contain entries with each frequency in Hz.
+- bw-buckets: Should contain upper-bound limits for each bandwidth bucket in Mbps.
+       Must match the framework power_profile.xml for the device.
diff --git a/Documentation/devicetree/bindings/misc/ramoops.txt b/Documentation/devicetree/bindings/misc/ramoops.txt
new file mode 100644
index 000000000000..5a475fae4aab
--- /dev/null
+++ b/Documentation/devicetree/bindings/misc/ramoops.txt
@@ -0,0 +1,43 @@
+Ramoops oops/panic logger
+=========================
+
+ramoops provides persistent RAM storage for oops and panics, so they can be
+recovered after a reboot.
+
+Parts of this storage may be set aside for other persistent log buffers, such
+as kernel log messages, or for optional ECC error-correction data.  The total
+size of these optional buffers must fit in the reserved region.
+
+Any remaining space will be used for a circular buffer of oops and panic
+records.  These records have a configurable size, with a size of 0 indicating
+that they should be disabled.
+
+
+Required properties:
+
+- compatible: must be "ramoops"
+
+- memory-region: phandle to a region of memory that is preserved between reboots
+
+
+Optional properties:
+
+- ecc-size: enables ECC support and specifies ECC buffer size in bytes
+  (defaults to no ECC)
+
+- record-size: maximum size in bytes of each dump done on oops/panic
+  (defaults to 0)
+
+- console-size: size in bytes of log buffer reserved for kernel messages
+  (defaults to 0)
+
+- ftrace-size: size in bytes of log buffer reserved for function tracing and
+  profiling (defaults to 0)
+
+- pmsg-size: size in bytes of log buffer reserved for userspace messages
+  (defaults to 0)
+
+- unbuffered: if present, use unbuffered mappings to map the reserved region
+  (defaults to buffered mappings)
+
+- no-dump-oops: if present, only dump panics (defaults to panics and oops)
diff --git a/Documentation/devicetree/bindings/scheduler/sched-energy-costs.txt b/Documentation/devicetree/bindings/scheduler/sched-energy-costs.txt
new file mode 100644
index 000000000000..11216f09e596
--- /dev/null
+++ b/Documentation/devicetree/bindings/scheduler/sched-energy-costs.txt
@@ -0,0 +1,360 @@
+===========================================================
+Energy cost bindings for Energy Aware Scheduling
+===========================================================
+
+===========================================================
+1 - Introduction
+===========================================================
+
+This note specifies bindings required for energy-aware scheduling
+(EAS)[1]. Historically, the scheduler's primary objective has been
+performance.  EAS aims to provide an alternative objective - energy
+efficiency. EAS relies on a simple platform energy cost model to
+guide scheduling decisions.  The model only considers the CPU
+subsystem.
+
+This note is aligned with the definition of the layout of physical
+CPUs in the system as described in the ARM topology binding
+description [2]. The concept is applicable to any system so long as
+the cost model data is provided for those processing elements in
+that system's topology that EAS is required to service.
+
+Processing elements refer to hardware threads, CPUs and clusters of
+related CPUs in increasing order of hierarchy.
+
+EAS requires two key cost metrics - busy costs and idle costs. Busy
+costs comprise of a list of compute capacities for the processing
+element in question and the corresponding power consumption at that
+capacity.  Idle costs comprise of a list of power consumption values
+for each idle state [C-state] that the processing element supports.
+For a detailed description of these metrics, their derivation and
+their use see [3].
+
+These cost metrics are required for processing elements in all
+scheduling domain levels that EAS is required to service.
+
+===========================================================
+2 - energy-costs node
+===========================================================
+
+Energy costs for the processing elements in scheduling domains that
+EAS is required to service are defined in the energy-costs node
+which acts as a container for the actual per processing element cost
+nodes. A single energy-costs node is required for a given system.
+
+- energy-costs node
+
+	Usage: Required
+
+	Description: The energy-costs node is a container node and
+	it's sub-nodes describe costs for each processing element at
+	all scheduling domain levels that EAS is required to
+	service.
+
+	Node name must be "energy-costs".
+
+	The energy-costs node's parent node must be the cpus node.
+
+	The energy-costs node's child nodes can be:
+
+	- one or more cost nodes.
+
+	Any other configuration is considered invalid.
+
+The energy-costs node can only contain a single type of child node
+whose bindings are described in paragraph 4.
+
+===========================================================
+3 - energy-costs node child nodes naming convention
+===========================================================
+
+energy-costs child nodes must follow a naming convention where the
+node name must be "thread-costN", "core-costN", "cluster-costN"
+depending on whether the costs in the node are for a thread, core or
+cluster.  N (where N = {0, 1, ...}) is the node number and has no
+bearing to the OS' logical thread, core or cluster index.
+
+===========================================================
+4 - cost node bindings
+===========================================================
+
+Bindings for cost nodes are defined as follows:
+
+- cluster-cost node
+
+	Description: must be declared within an energy-costs node. A
+	system can contain multiple clusters and each cluster
+	serviced by EAS	must have a corresponding cluster-costs
+	node.
+
+	The cluster-cost node name must be "cluster-costN" as
+	described in 3 above.
+
+	A cluster-cost node must be a leaf node with no children.
+
+	Properties for cluster-cost nodes are described in paragraph
+	5 below.
+
+	Any other configuration is considered invalid.
+
+- core-cost node
+
+	Description: must be declared within an energy-costs node. A
+	system can contain multiple cores and each core serviced by
+	EAS must have a corresponding core-cost node.
+
+	The core-cost node name must be "core-costN" as described in
+	3 above.
+
+	A core-cost node must be a leaf node with no children.
+
+	Properties for core-cost nodes are described in paragraph
+	5 below.
+
+	Any other configuration is considered invalid.
+
+- thread-cost node
+
+	Description: must be declared within an energy-costs node. A
+	system can contain cores with multiple hardware threads and
+	each thread serviced by EAS must have a corresponding
+	thread-cost node.
+
+	The core-cost node name must be "core-costN" as described in
+	3 above.
+
+	A core-cost node must be a leaf node with no children.
+
+	Properties for thread-cost nodes are described in paragraph
+	5 below.
+
+	Any other configuration is considered invalid.
+
+===========================================================
+5 - Cost node properties
+==========================================================
+
+All cost node types must have only the following properties:
+
+- busy-cost-data
+
+	Usage: required
+	Value type: An array of 2-item tuples. Each item is of type
+	u32.
+	Definition: The first item in the tuple is the capacity
+	value as described in [3]. The second item in the tuple is
+	the energy cost value as described in [3].
+
+- idle-cost-data
+
+	Usage: required
+	Value type: An array of 1-item tuples. The item is of type
+	u32.
+	Definition: The item in the tuple is the energy cost value
+	as described in [3].
+
+===========================================================
+4 - Extensions to the cpu node
+===========================================================
+
+The cpu node is extended with a property that establishes the
+connection between the processing element represented by the cpu
+node and the cost-nodes associated with this processing element.
+
+The connection is expressed in line with the topological hierarchy
+that this processing element belongs to starting with the level in
+the hierarchy that this processing element itself belongs to through
+to the highest level that EAS is required to service.  The
+connection cannot be sparse and must be contiguous from the
+processing element's level through to the highest desired level. The
+highest desired level must be the same for all processing elements.
+
+Example: Given that a cpu node may represent a thread that is a part
+of a core, this property may contain multiple elements which
+associate the thread with cost nodes describing the costs for the
+thread itself, the core the thread belongs to, the cluster the core
+belongs to and so on. The elements must be ordered from the lowest
+level nodes to the highest desired level that EAS must service. The
+highest desired level must be the same for all cpu nodes. The
+elements must not be sparse: there must be elements for the current
+thread, the next level of hierarchy (core) and so on without any
+'holes'.
+
+Example: Given that a cpu node may represent a core that is a part
+of a cluster of related cpus this property may contain multiple
+elements which associate the core with cost nodes describing the
+costs for the core itself, the cluster the core belongs to and so
+on. The elements must be ordered from the lowest level nodes to the
+highest desired level that EAS must service. The highest desired
+level must be the same for all cpu nodes. The elements must not be
+sparse: there must be elements for the current thread, the next
+level of hierarchy (core) and so on without any 'holes'.
+
+If the system comprises of hierarchical clusters of clusters, this
+property will contain multiple associations with the relevant number
+of cluster elements in hierarchical order.
+
+Property added to the cpu node:
+
+- sched-energy-costs
+
+	Usage: required
+	Value type: List of phandles
+	Definition: a list of phandles to specific cost nodes in the
+	energy-costs parent node that correspond to the processing
+	element represented by this cpu node in hierarchical order
+	of topology.
+
+	The order of phandles in the list is significant. The first
+	phandle is to the current processing element's own cost
+	node.  Subsequent phandles are to higher hierarchical level
+	cost nodes up until the maximum level that EAS is to
+	service.
+
+	All cpu nodes must have the same highest level cost node.
+
+	The phandle list must not be sparsely populated with handles
+	to non-contiguous hierarchical levels. See commentary above
+	for clarity.
+
+	Any other configuration is invalid.
+
+===========================================================
+5 - Example dts
+===========================================================
+
+Example 1 (ARM 64-bit, 6-cpu system, two clusters of cpus, one
+cluster of 2 Cortex-A57 cpus, one cluster of 4 Cortex-A53 cpus):
+
+cpus {
+	#address-cells = <2>;
+	#size-cells = <0>;
+	.
+	.
+	.
+	A57_0: cpu@0 {
+		compatible = "arm,cortex-a57","arm,armv8";
+		reg = <0x0 0x0>;
+		device_type = "cpu";
+		enable-method = "psci";
+		next-level-cache = <&A57_L2>;
+		clocks = <&scpi_dvfs 0>;
+		cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+		sched-energy-costs = <&CPU_COST_0 &CLUSTER_COST_0>;
+	};
+
+	A57_1: cpu@1 {
+		compatible = "arm,cortex-a57","arm,armv8";
+		reg = <0x0 0x1>;
+		device_type = "cpu";
+		enable-method = "psci";
+		next-level-cache = <&A57_L2>;
+		clocks = <&scpi_dvfs 0>;
+		cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+		sched-energy-costs = <&CPU_COST_0 &CLUSTER_COST_0>;
+	};
+
+	A53_0: cpu@100 {
+		compatible = "arm,cortex-a53","arm,armv8";
+		reg = <0x0 0x100>;
+		device_type = "cpu";
+		enable-method = "psci";
+		next-level-cache = <&A53_L2>;
+		clocks = <&scpi_dvfs 1>;
+		cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+		sched-energy-costs = <&CPU_COST_1 &CLUSTER_COST_1>;
+	};
+
+	A53_1: cpu@101 {
+		compatible = "arm,cortex-a53","arm,armv8";
+		reg = <0x0 0x101>;
+		device_type = "cpu";
+		enable-method = "psci";
+		next-level-cache = <&A53_L2>;
+		clocks = <&scpi_dvfs 1>;
+		cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+		sched-energy-costs = <&CPU_COST_1 &CLUSTER_COST_1>;
+	};
+
+	A53_2: cpu@102 {
+		compatible = "arm,cortex-a53","arm,armv8";
+		reg = <0x0 0x102>;
+		device_type = "cpu";
+		enable-method = "psci";
+		next-level-cache = <&A53_L2>;
+		clocks = <&scpi_dvfs 1>;
+		cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+		sched-energy-costs = <&CPU_COST_1 &CLUSTER_COST_1>;
+	};
+
+	A53_3: cpu@103 {
+		compatible = "arm,cortex-a53","arm,armv8";
+		reg = <0x0 0x103>;
+		device_type = "cpu";
+		enable-method = "psci";
+		next-level-cache = <&A53_L2>;
+		clocks = <&scpi_dvfs 1>;
+		cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+		sched-energy-costs = <&CPU_COST_1 &CLUSTER_COST_1>;
+	};
+
+	energy-costs {
+		CPU_COST_0: core-cost0 {
+			busy-cost-data = <
+				417   168
+				579   251
+				744   359
+				883   479
+				1024  616
+			>;
+			idle-cost-data = <
+				15
+				0
+			>;
+		};
+		CPU_COST_1: core-cost1 {
+			busy-cost-data = <
+				235 33
+				302 46
+				368 61
+				406 76
+				447 93
+			>;
+			idle-cost-data = <
+				6
+				0
+			>;
+		};
+		CLUSTER_COST_0: cluster-cost0 {
+			busy-cost-data = <
+				417   24
+				579   32
+				744   43
+				883   49
+				1024  64
+			>;
+			idle-cost-data = <
+				65
+				24
+			>;
+		};
+		CLUSTER_COST_1: cluster-cost1 {
+			busy-cost-data = <
+				235 26
+				303 30
+				368 39
+				406 47
+				447 57
+			>;
+			idle-cost-data = <
+				56
+				17
+			>;
+		};
+	};
+};
+
+===============================================================================
+[1] https://lkml.org/lkml/2015/5/12/728
+[2] Documentation/devicetree/bindings/topology.txt
+[3] Documentation/scheduler/sched-energy.txt
diff --git a/Documentation/filesystems/f2fs.txt b/Documentation/filesystems/f2fs.txt
index b102b436563e..6cf9ad12c57f 100644
--- a/Documentation/filesystems/f2fs.txt
+++ b/Documentation/filesystems/f2fs.txt
@@ -102,14 +102,16 @@ background_gc=%s       Turn on/off cleaning operations, namely garbage
                        collection, triggered in background when I/O subsystem is
                        idle. If background_gc=on, it will turn on the garbage
                        collection and if background_gc=off, garbage collection
-                       will be truned off. If background_gc=sync, it will turn
+                       will be turned off. If background_gc=sync, it will turn
                        on synchronous garbage collection running in background.
                        Default value for this option is on. So garbage
                        collection is on by default.
 disable_roll_forward   Disable the roll-forward recovery routine
 norecovery             Disable the roll-forward recovery routine, mounted read-
                        only (i.e., -o ro,disable_roll_forward)
-discard                Issue discard/TRIM commands when a segment is cleaned.
+discard/nodiscard      Enable/disable real-time discard in f2fs, if discard is
+                       enabled, f2fs will issue discard/TRIM commands when a
+		       segment is cleaned.
 no_heap                Disable heap-style segment allocation which finds free
                        segments for data from the beginning of main area, while
 		       for node from the end of main area.
@@ -123,12 +125,14 @@ active_logs=%u         Support configuring the number of active logs. In the
 disable_ext_identify   Disable the extension list configured by mkfs, so f2fs
                        does not aware of cold files such as media files.
 inline_xattr           Enable the inline xattrs feature.
+noinline_xattr         Disable the inline xattrs feature.
 inline_data            Enable the inline data feature: New created small(<~3.4k)
                        files can be written into inode block.
 inline_dentry          Enable the inline dir feature: data in new created
                        directory entries can be written into inode block. The
                        space of inode block which is used to store inline
                        dentries is limited to ~3.4k.
+noinline_dentry        Diable the inline dentry feature.
 flush_merge	       Merge concurrent cache_flush commands as much as possible
                        to eliminate redundant command issues. If the underlying
 		       device handles the cache_flush command relatively slowly,
@@ -145,10 +149,29 @@ extent_cache           Enable an extent cache based on rb-tree, it can cache
                        as many as extent which map between contiguous logical
                        address and physical address per inode, resulting in
                        increasing the cache hit ratio. Set by default.
-noextent_cache         Diable an extent cache based on rb-tree explicitly, see
+noextent_cache         Disable an extent cache based on rb-tree explicitly, see
                        the above extent_cache mount option.
 noinline_data          Disable the inline data feature, inline data feature is
                        enabled by default.
+data_flush             Enable data flushing before checkpoint in order to
+                       persist data of regular and symlink.
+mode=%s                Control block allocation mode which supports "adaptive"
+                       and "lfs". In "lfs" mode, there should be no random
+                       writes towards main area.
+io_bits=%u             Set the bit size of write IO requests. It should be set
+                       with "mode=lfs".
+usrquota               Enable plain user disk quota accounting.
+grpquota               Enable plain group disk quota accounting.
+prjquota               Enable plain project quota accounting.
+usrjquota=<file>       Appoint specified file and type during mount, so that quota
+grpjquota=<file>       information can be properly updated during recovery flow,
+prjjquota=<file>       <quota file>: must be in root directory;
+jqfmt=<quota type>     <quota type>: [vfsold,vfsv0,vfsv1].
+offusrjquota           Turn off user journelled quota.
+offgrpjquota           Turn off group journelled quota.
+offprjjquota           Turn off project journelled quota.
+quota                  Enable plain user disk quota accounting.
+noquota                Disable all plain disk quota option.
 
 ================================================================================
 DEBUGFS ENTRIES
@@ -192,7 +215,16 @@ Files in /sys/fs/f2fs/<devname>
                               policy for garbage collection. Setting gc_idle = 0
                               (default) will disable this option. Setting
                               gc_idle = 1 will select the Cost Benefit approach
-                              & setting gc_idle = 2 will select the greedy aproach.
+                              & setting gc_idle = 2 will select the greedy approach.
+
+ gc_urgent                    This parameter controls triggering background GCs
+                              urgently or not. Setting gc_urgent = 0 [default]
+                              makes back to default behavior, while if it is set
+                              to 1, background thread starts to do GC by given
+                              gc_urgent_sleep_time interval.
+
+ gc_urgent_sleep_time         This parameter controls sleep time for gc_urgent.
+                              500 ms is set by default. See above gc_urgent.
 
  reclaim_segments             This parameter controls the number of prefree
                               segments to be reclaimed. If the number of prefree
@@ -298,7 +330,7 @@ The dump.f2fs shows the information of specific inode and dumps SSA and SIT to
 file. Each file is dump_ssa and dump_sit.
 
 The dump.f2fs is used to debug on-disk data structures of the f2fs filesystem.
-It shows on-disk inode information reconized by a given inode number, and is
+It shows on-disk inode information recognized by a given inode number, and is
 able to dump all the SSA and SIT entries into predefined files, ./dump_ssa and
 ./dump_sit respectively.
 
diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt
index 6716413c17ba..fea4777a5695 100644
--- a/Documentation/filesystems/proc.txt
+++ b/Documentation/filesystems/proc.txt
@@ -43,6 +43,7 @@ Table of Contents
   3.7   /proc/<pid>/task/<tid>/children - Information about task children
   3.8   /proc/<pid>/fdinfo/<fd> - Information about opened file
   3.9   /proc/<pid>/map_files - Information about memory mapped files
+  3.10  /proc/<pid>/timerslack_ns - Task timerslack value
 
   4	Configuring procfs
   4.1	Mount options
@@ -380,6 +381,8 @@ is not associated with a file:
  [stack]                  = the stack of the main process
  [vdso]                   = the "virtual dynamic shared object",
                             the kernel system call handler
+ [anon:<name>]            = an anonymous mapping that has been
+                            named by userspace
 
  or if empty, the mapping is anonymous.
 
@@ -432,6 +435,7 @@ KernelPageSize:        4 kB
 MMUPageSize:           4 kB
 Locked:                0 kB
 VmFlags: rd ex mr mw me dw
+Name:           name from userspace
 
 the first of these lines shows the same information as is displayed for the
 mapping in /proc/PID/maps.  The remaining lines show the size of the mapping
@@ -494,6 +498,9 @@ Note that there is no guarantee that every flag and associated mnemonic will
 be present in all further kernel releases. Things get changed, the flags may
 be vanished or the reverse -- new added.
 
+The "Name" field will only be present on a mapping that has been named by
+userspace, and will show the name passed in by userspace.
+
 This file is only present if the CONFIG_MMU kernel configuration option is
 enabled.
 
@@ -1847,6 +1854,23 @@ time one can open(2) mappings from the listings of two processes and
 comparing their inode numbers to figure out which anonymous memory areas
 are actually shared.
 
+3.10	/proc/<pid>/timerslack_ns - Task timerslack value
+---------------------------------------------------------
+This file provides the value of the task's timerslack value in nanoseconds.
+This value specifies a amount of time that normal timers may be deferred
+in order to coalesce timers and avoid unnecessary wakeups.
+
+This allows a task's interactivity vs power consumption trade off to be
+adjusted.
+
+Writing 0 to the file will set the tasks timerslack to the default value.
+
+Valid values are from 0 - ULLONG_MAX
+
+An application setting the value must have PTRACE_MODE_ATTACH_FSCREDS level
+permissions on the task specified to change its timerslack_ns value.
+
+
 ------------------------------------------------------------------------------
 Configuring procfs
 ------------------------------------------------------------------------------
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index 7c74bc412c3d..21321b9acfb0 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -56,6 +56,7 @@ parameter is applicable:
 	BLACKFIN Blackfin architecture is enabled.
 	CLK	Common clock infrastructure is enabled.
 	CMA	Contiguous Memory Area support is enabled.
+	DM	Device mapper support is enabled.
 	DRM	Direct Rendering Management support is enabled.
 	DYNAMIC_DEBUG Build in debug messages and enable them at runtime
 	EDD	BIOS Enhanced Disk Drive Services (EDD) is enabled
@@ -915,6 +916,11 @@ bytes respectively. Such letter suffixes can also be entirely omitted.
 
 	dis_ucode_ldr	[X86] Disable the microcode loader.
 
+	dm=		[DM] Allows early creation of a device-mapper device.
+			See Documentation/device-mapper/boot.txt.
+
+	dmasound=	[HW,OSS] Sound subsystem buff
+
 	dma_debug=off	If the kernel is compiled with DMA_API_DEBUG support,
 			this option disables the debugging code at boot.
 
diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt
index 2ea4c45cf1c8..5f1ea84ed72b 100644
--- a/Documentation/networking/ip-sysctl.txt
+++ b/Documentation/networking/ip-sysctl.txt
@@ -584,6 +584,16 @@ tcp_fastopen - INTEGER
 
 	See include/net/tcp.h and the code for more details.
 
+tcp_fwmark_accept - BOOLEAN
+	If set, incoming connections to listening sockets that do not have a
+	socket mark will set the mark of the accepting socket to the fwmark of
+	the incoming SYN packet. This will cause all packets on that connection
+	(starting from the first SYNACK) to be sent with that fwmark. The
+	listening socket's mark is unchanged. Listening sockets that already
+	have a fwmark set via setsockopt(SOL_SOCKET, SO_MARK, ...) are
+	unaffected.
+	Default: 0
+
 tcp_syn_retries - INTEGER
 	Number of times initial SYNs for an active TCP connection attempt
 	will be retransmitted. Should not be higher than 255. Default value
@@ -1403,11 +1413,20 @@ accept_ra_pinfo - BOOLEAN
 	Functional default: enabled if accept_ra is enabled.
 			    disabled if accept_ra is disabled.
 
+accept_ra_rt_info_min_plen - INTEGER
+	Minimum prefix length of Route Information in RA.
+
+	Route Information w/ prefix smaller than this variable shall
+	be ignored.
+
+	Functional default: 0 if accept_ra_rtr_pref is enabled.
+			    -1 if accept_ra_rtr_pref is disabled.
+
 accept_ra_rt_info_max_plen - INTEGER
 	Maximum prefix length of Route Information in RA.
 
-	Route Information w/ prefix larger than or equal to this
-	variable shall be ignored.
+	Route Information w/ prefix larger than this variable shall
+	be ignored.
 
 	Functional default: 0 if accept_ra_rtr_pref is enabled.
 			    -1 if accept_ra_rtr_pref is disabled.
diff --git a/Documentation/ramoops.txt b/Documentation/ramoops.txt
index 5d8675615e59..9264bcab4099 100644
--- a/Documentation/ramoops.txt
+++ b/Documentation/ramoops.txt
@@ -45,7 +45,7 @@ corrupt, but usually it is restorable.
 
 2. Setting the parameters
 
-Setting the ramoops parameters can be done in 2 different manners:
+Setting the ramoops parameters can be done in 3 different manners:
  1. Use the module parameters (which have the names of the variables described
  as before).
  For quick debugging, you can also reserve parts of memory during boot
@@ -54,7 +54,9 @@ Setting the ramoops parameters can be done in 2 different manners:
  kernel to use only the first 128 MB of memory, and place ECC-protected ramoops
  region at 128 MB boundary:
  "mem=128M ramoops.mem_address=0x8000000 ramoops.ecc=1"
- 2. Use a platform device and set the platform data. The parameters can then
+ 2. Use Device Tree bindings, as described in
+ Documentation/device-tree/bindings/misc/ramoops.txt.
+ 3. Use a platform device and set the platform data. The parameters can then
  be set through that platform data. An example of doing that is:
 
 #include <linux/pstore_ram.h>
diff --git a/Documentation/scheduler/sched-energy.txt b/Documentation/scheduler/sched-energy.txt
new file mode 100644
index 000000000000..dab2f9088b33
--- /dev/null
+++ b/Documentation/scheduler/sched-energy.txt
@@ -0,0 +1,362 @@
+Energy cost model for energy-aware scheduling (EXPERIMENTAL)
+
+Introduction
+=============
+
+The basic energy model uses platform energy data stored in sched_group_energy
+data structures attached to the sched_groups in the sched_domain hierarchy. The
+energy cost model offers two functions that can be used to guide scheduling
+decisions:
+
+1.	static unsigned int sched_group_energy(struct energy_env *eenv)
+2.	static int energy_diff(struct energy_env *eenv)
+
+sched_group_energy() estimates the energy consumed by all cpus in a specific
+sched_group including any shared resources owned exclusively by this group of
+cpus. Resources shared with other cpus are excluded (e.g. later level caches).
+
+energy_diff() estimates the total energy impact of a utilization change. That
+is, adding, removing, or migrating utilization (tasks).
+
+Both functions use a struct energy_env to specify the scenario to be evaluated:
+
+	struct energy_env {
+		struct sched_group      *sg_top;
+		struct sched_group      *sg_cap;
+		int                     cap_idx;
+		int                     util_delta;
+		int                     src_cpu;
+		int                     dst_cpu;
+		int                     energy;
+	};
+
+sg_top: sched_group to be evaluated. Not used by energy_diff().
+
+sg_cap: sched_group covering the cpus in the same frequency domain. Set by
+sched_group_energy().
+
+cap_idx: Capacity state to be used for energy calculations. Set by
+find_new_capacity().
+
+util_delta: Amount of utilization to be added, removed, or migrated.
+
+src_cpu: Source cpu from where 'util_delta' utilization is removed. Should be
+-1 if no source (e.g. task wake-up).
+
+dst_cpu: Destination cpu where 'util_delta' utilization is added. Should be -1
+if utilization is removed (e.g. terminating tasks).
+
+energy: Result of sched_group_energy().
+
+The metric used to represent utilization is the actual per-entity running time
+averaged over time using a geometric series. Very similar to the existing
+per-entity load-tracking, but _not_ scaled by task priority and capped by the
+capacity of the cpu. The latter property does mean that utilization may
+underestimate the compute requirements for task on fully/over utilized cpus.
+The greatest potential for energy savings without affecting performance too much
+is scenarios where the system isn't fully utilized. If the system is deemed
+fully utilized load-balancing should be done with task load (includes task
+priority) instead in the interest of fairness and performance.
+
+
+Background and Terminology
+===========================
+
+To make it clear from the start:
+
+energy = [joule] (resource like a battery on powered devices)
+power = energy/time = [joule/second] = [watt]
+
+The goal of energy-aware scheduling is to minimize energy, while still getting
+the job done. That is, we want to maximize:
+
+	performance [inst/s]
+	--------------------
+	    power [W]
+
+which is equivalent to minimizing:
+
+	energy [J]
+	-----------
+	instruction
+
+while still getting 'good' performance. It is essentially an alternative
+optimization objective to the current performance-only objective for the
+scheduler. This alternative considers two objectives: energy-efficiency and
+performance. Hence, there needs to be a user controllable knob to switch the
+objective. Since it is early days, this is currently a sched_feature
+(ENERGY_AWARE).
+
+The idea behind introducing an energy cost model is to allow the scheduler to
+evaluate the implications of its decisions rather than applying energy-saving
+techniques blindly that may only have positive effects on some platforms. At
+the same time, the energy cost model must be as simple as possible to minimize
+the scheduler latency impact.
+
+Platform topology
+------------------
+
+The system topology (cpus, caches, and NUMA information, not peripherals) is
+represented in the scheduler by the sched_domain hierarchy which has
+sched_groups attached at each level that covers one or more cpus (see
+sched-domains.txt for more details). To add energy awareness to the scheduler
+we need to consider power and frequency domains.
+
+Power domain:
+
+A power domain is a part of the system that can be powered on/off
+independently. Power domains are typically organized in a hierarchy where you
+may be able to power down just a cpu or a group of cpus along with any
+associated resources (e.g.  shared caches). Powering up a cpu means that all
+power domains it is a part of in the hierarchy must be powered up. Hence, it is
+more expensive to power up the first cpu that belongs to a higher level power
+domain than powering up additional cpus in the same high level domain. Two
+level power domain hierarchy example:
+
+		Power source
+		         +-------------------------------+----...
+per group PD		 G                               G
+		         |           +----------+        |
+		    +--------+-------| Shared   |  (other groups)
+per-cpu PD	    G        G       | resource |
+		    |        |       +----------+
+		+-------+ +-------+
+		| CPU 0 | | CPU 1 |
+		+-------+ +-------+
+
+Frequency domain:
+
+Frequency domains (P-states) typically cover the same group of cpus as one of
+the power domain levels. That is, there might be several smaller power domains
+sharing the same frequency (P-state) or there might be a power domain spanning
+multiple frequency domains.
+
+From a scheduling point of view there is no need to know the actual frequencies
+[Hz]. All the scheduler cares about is the compute capacity available at the
+current state (P-state) the cpu is in and any other available states. For that
+reason, and to also factor in any cpu micro-architecture differences, compute
+capacity scaling states are called 'capacity states' in this document. For SMP
+systems this is equivalent to P-states. For mixed micro-architecture systems
+(like ARM big.LITTLE) it is P-states scaled according to the micro-architecture
+performance relative to the other cpus in the system.
+
+Energy modelling:
+------------------
+
+Due to the hierarchical nature of the power domains, the most obvious way to
+model energy costs is therefore to associate power and energy costs with
+domains (groups of cpus). Energy costs of shared resources are associated with
+the group of cpus that share the resources, only the cost of powering the
+cpu itself and any private resources (e.g. private L1 caches) is associated
+with the per-cpu groups (lowest level).
+
+For example, for an SMP system with per-cpu power domains and a cluster level
+(group of cpus) power domain we get the overall energy costs to be:
+
+	energy = energy_cluster + n * energy_cpu
+
+where 'n' is the number of cpus powered up and energy_cluster is the cost paid
+as soon as any cpu in the cluster is powered up.
+
+The power and frequency domains can naturally be mapped onto the existing
+sched_domain hierarchy and sched_groups by adding the necessary data to the
+existing data structures.
+
+The energy model considers energy consumption from two contributors (shown in
+the illustration below):
+
+1. Busy energy: Energy consumed while a cpu and the higher level groups that it
+belongs to are busy running tasks. Busy energy is associated with the state of
+the cpu, not an event. The time the cpu spends in this state varies. Thus, the
+most obvious platform parameter for this contribution is busy power
+(energy/time).
+
+2. Idle energy: Energy consumed while a cpu and higher level groups that it
+belongs to are idle (in a C-state). Like busy energy, idle energy is associated
+with the state of the cpu. Thus, the platform parameter for this contribution
+is idle power (energy/time).
+
+Energy consumed during transitions from an idle-state (C-state) to a busy state
+(P-state) or going the other way is ignored by the model to simplify the energy
+model calculations.
+
+
+	Power
+	^
+	|            busy->idle             idle->busy
+	|            transition             transition
+	|
+	|                _                      __
+	|               / \                    /  \__________________
+	|______________/   \                  /
+	|                   \                /
+	|  Busy              \    Idle      /        Busy
+	|  low P-state        \____________/         high P-state
+	|
+	+------------------------------------------------------------> time
+
+Busy    |--------------|                          |-----------------|
+
+Wakeup                 |------|            |------|
+
+Idle                          |------------|
+
+
+The basic algorithm
+====================
+
+The basic idea is to determine the total energy impact when utilization is
+added or removed by estimating the impact at each level in the sched_domain
+hierarchy starting from the bottom (sched_group contains just a single cpu).
+The energy cost comes from busy time (sched_group is awake because one or more
+cpus are busy) and idle time (in an idle-state). Energy model numbers account
+for energy costs associated with all cpus in the sched_group as a group.
+
+	for_each_domain(cpu, sd) {
+		sg = sched_group_of(cpu)
+		energy_before = curr_util(sg) * busy_power(sg)
+				+ (1-curr_util(sg)) * idle_power(sg)
+		energy_after = new_util(sg) * busy_power(sg)
+				+ (1-new_util(sg)) * idle_power(sg)
+		energy_diff += energy_before - energy_after
+
+	}
+
+	return energy_diff
+
+{curr, new}_util: The cpu utilization at the lowest level and the overall
+non-idle time for the entire group for higher levels. Utilization is in the
+range 0.0 to 1.0 in the pseudo-code.
+
+busy_power: The power consumption of the sched_group.
+
+idle_power: The power consumption of the sched_group when idle.
+
+Note: It is a fundamental assumption that the utilization is (roughly) scale
+invariant. Task utilization tracking factors in any frequency scaling and
+performance scaling differences due to difference cpu microarchitectures such
+that task utilization can be used across the entire system.
+
+
+Platform energy data
+=====================
+
+struct sched_group_energy can be attached to sched_groups in the sched_domain
+hierarchy and has the following members:
+
+cap_states:
+	List of struct capacity_state representing the supported capacity states
+	(P-states). struct capacity_state has two members: cap and power, which
+	represents the compute capacity and the busy_power of the state. The
+	list must be ordered by capacity low->high.
+
+nr_cap_states:
+	Number of capacity states in cap_states list.
+
+idle_states:
+	List of struct idle_state containing idle_state power cost for each
+	idle-state supported by the system orderd by shallowest state first.
+	All states must be included at all level in the hierarchy, i.e. a
+	sched_group spanning just a single cpu must also include coupled
+	idle-states (cluster states). In addition to the cpuidle idle-states,
+	the list must also contain an entry for the idling using the arch
+	default idle (arch_idle_cpu()). Despite this state may not be a true
+	hardware idle-state it is considered the shallowest idle-state in the
+	energy model and must be the first entry. cpus may enter this state
+	(possibly 'active idling') if cpuidle decides not enter a cpuidle
+	idle-state. Default idle may not be used when cpuidle is enabled.
+	In this case, it should just be a copy of the first cpuidle idle-state.
+
+nr_idle_states:
+	Number of idle states in idle_states list.
+
+There are no unit requirements for the energy cost data. Data can be normalized
+with any reference, however, the normalization must be consistent across all
+energy cost data. That is, one bogo-joule/watt must be the same quantity for
+data, but we don't care what it is.
+
+A recipe for platform characterization
+=======================================
+
+Obtaining the actual model data for a particular platform requires some way of
+measuring power/energy. There isn't a tool to help with this (yet). This
+section provides a recipe for use as reference. It covers the steps used to
+characterize the ARM TC2 development platform. This sort of measurements is
+expected to be done anyway when tuning cpuidle and cpufreq for a given
+platform.
+
+The energy model needs two types of data (struct sched_group_energy holds
+these) for each sched_group where energy costs should be taken into account:
+
+1. Capacity state information
+
+A list containing the compute capacity and power consumption when fully
+utilized attributed to the group as a whole for each available capacity state.
+At the lowest level (group contains just a single cpu) this is the power of the
+cpu alone without including power consumed by resources shared with other cpus.
+It basically needs to fit the basic modelling approach described in "Background
+and Terminology" section:
+
+	energy_system = energy_shared + n * energy_cpu
+
+for a system containing 'n' busy cpus. Only 'energy_cpu' should be included at
+the lowest level. 'energy_shared' is included at the next level which
+represents the group of cpus among which the resources are shared.
+
+This model is, of course, a simplification of reality. Thus, power/energy
+attributions might not always exactly represent how the hardware is designed.
+Also, busy power is likely to depend on the workload. It is therefore
+recommended to use a representative mix of workloads when characterizing the
+capacity states.
+
+If the group has no capacity scaling support, the list will contain a single
+state where power is the busy power attributed to the group. The capacity
+should be set to a default value (1024).
+
+When frequency domains include multiple power domains, the group representing
+the frequency domain and all child groups share capacity states. This must be
+indicated by setting the SD_SHARE_CAP_STATES sched_domain flag. All groups at
+all levels that share the capacity state must have the list of capacity states
+with the power set to the contribution of the individual group.
+
+2. Idle power information
+
+Stored in the idle_states list. The power number is the group idle power
+consumption in each idle state as well when the group is idle but has not
+entered an idle-state ('active idle' as mentioned earlier). Due to the way the
+energy model is defined, the idle power of the deepest group idle state can
+alternatively be accounted for in the parent group busy power. In that case the
+group idle state power values are offset such that the idle power of the
+deepest state is zero. It is less intuitive, but it is easier to measure as
+idle power consumed by the group and the busy/idle power of the parent group
+cannot be distinguished without per group measurement points.
+
+Measuring capacity states and idle power:
+
+The capacity states' capacity and power can be estimated by running a benchmark
+workload at each available capacity state. By restricting the benchmark to run
+on subsets of cpus it is possible to extrapolate the power consumption of
+shared resources.
+
+ARM TC2 has two clusters of two and three cpus respectively. Each cluster has a
+shared L2 cache. TC2 has on-chip energy counters per cluster. Running a
+benchmark workload on just one cpu in a cluster means that power is consumed in
+the cluster (higher level group) and a single cpu (lowest level group). Adding
+another benchmark task to another cpu increases the power consumption by the
+amount consumed by the additional cpu. Hence, it is possible to extrapolate the
+cluster busy power.
+
+For platforms that don't have energy counters or equivalent instrumentation
+built-in, it may be possible to use an external DAQ to acquire similar data.
+
+If the benchmark includes some performance score (for example sysbench cpu
+benchmark), this can be used to record the compute capacity.
+
+Measuring idle power requires insight into the idle state implementation on the
+particular platform. Specifically, if the platform has coupled idle-states (or
+package states). To measure non-coupled per-cpu idle-states it is necessary to
+keep one cpu busy to keep any shared resources alive to isolate the idle power
+of the cpu from idle/busy power of the shared resources. The cpu can be tricked
+into different per-cpu idle states by disabling the other states. Based on
+various combinations of measurements with specific cpus busy and disabling
+idle-states it is possible to extrapolate the idle-state power.
diff --git a/Documentation/scheduler/sched-tune.txt b/Documentation/scheduler/sched-tune.txt
new file mode 100644
index 000000000000..9bd2231c01b1
--- /dev/null
+++ b/Documentation/scheduler/sched-tune.txt
@@ -0,0 +1,366 @@
+             Central, scheduler-driven, power-performance control
+                               (EXPERIMENTAL)
+
+Abstract
+========
+
+The topic of a single simple power-performance tunable, that is wholly
+scheduler centric, and has well defined and predictable properties has come up
+on several occasions in the past [1,2]. With techniques such as a scheduler
+driven DVFS [3], we now have a good framework for implementing such a tunable.
+This document describes the overall ideas behind its design and implementation.
+
+
+Table of Contents
+=================
+
+1. Motivation
+2. Introduction
+3. Signal Boosting Strategy
+4. OPP selection using boosted CPU utilization
+5. Per task group boosting
+6. Question and Answers
+   - What about "auto" mode?
+   - What about boosting on a congested system?
+   - How CPUs are boosted when we have tasks with multiple boost values?
+7. References
+
+
+1. Motivation
+=============
+
+Sched-DVFS [3] is a new event-driven cpufreq governor which allows the
+scheduler to select the optimal DVFS operating point (OPP) for running a task
+allocated to a CPU. The introduction of sched-DVFS enables running workloads at
+the most energy efficient OPPs.
+
+However, sometimes it may be desired to intentionally boost the performance of
+a workload even if that could imply a reasonable increase in energy
+consumption. For example, in order to reduce the response time of a task, we
+may want to run the task at a higher OPP than the one that is actually required
+by it's CPU bandwidth demand.
+
+This last requirement is especially important if we consider that one of the
+main goals of the sched-DVFS component is to replace all currently available
+CPUFreq policies. Since sched-DVFS is event based, as opposed to the sampling
+driven governors we currently have, it is already more responsive at selecting
+the optimal OPP to run tasks allocated to a CPU. However, just tracking the
+actual task load demand may not be enough from a performance standpoint.  For
+example, it is not possible to get behaviors similar to those provided by the
+"performance" and "interactive" CPUFreq governors.
+
+This document describes an implementation of a tunable, stacked on top of the
+sched-DVFS which extends its functionality to support task performance
+boosting.
+
+By "performance boosting" we mean the reduction of the time required to
+complete a task activation, i.e. the time elapsed from a task wakeup to its
+next deactivation (e.g. because it goes back to sleep or it terminates).  For
+example, if we consider a simple periodic task which executes the same workload
+for 5[s] every 20[s] while running at a certain OPP, a boosted execution of
+that task must complete each of its activations in less than 5[s].
+
+A previous attempt [5] to introduce such a boosting feature has not been
+successful mainly because of the complexity of the proposed solution.  The
+approach described in this document exposes a single simple interface to
+user-space.  This single tunable knob allows the tuning of system wide
+scheduler behaviours ranging from energy efficiency at one end through to
+incremental performance boosting at the other end.  This first tunable affects
+all tasks. However, a more advanced extension of the concept is also provided
+which uses CGroups to boost the performance of only selected tasks while using
+the energy efficient default for all others.
+
+The rest of this document introduces in more details the proposed solution
+which has been named SchedTune.
+
+
+2. Introduction
+===============
+
+SchedTune exposes a simple user-space interface with a single power-performance
+tunable:
+
+  /proc/sys/kernel/sched_cfs_boost
+
+This permits expressing a boost value as an integer in the range [0..100].
+
+A value of 0 (default) configures the CFS scheduler for maximum energy
+efficiency. This means that sched-DVFS runs the tasks at the minimum OPP
+required to satisfy their workload demand.
+A value of 100 configures scheduler for maximum performance, which translates
+to the selection of the maximum OPP on that CPU.
+
+The range between 0 and 100 can be set to satisfy other scenarios suitably. For
+example to satisfy interactive response or depending on other system events
+(battery level etc).
+
+A CGroup based extension is also provided, which permits further user-space
+defined task classification to tune the scheduler for different goals depending
+on the specific nature of the task, e.g. background vs interactive vs
+low-priority.
+
+The overall design of the SchedTune module is built on top of "Per-Entity Load
+Tracking" (PELT) signals and sched-DVFS by introducing a bias on the Operating
+Performance Point (OPP) selection.
+Each time a task is allocated on a CPU, sched-DVFS has the opportunity to tune
+the operating frequency of that CPU to better match the workload demand. The
+selection of the actual OPP being activated is influenced by the global boost
+value, or the boost value for the task CGroup when in use.
+
+This simple biasing approach leverages existing frameworks, which means minimal
+modifications to the scheduler, and yet it allows to achieve a range of
+different behaviours all from a single simple tunable knob.
+The only new concept introduced is that of signal boosting.
+
+
+3. Signal Boosting Strategy
+===========================
+
+The whole PELT machinery works based on the value of a few load tracking signals
+which basically track the CPU bandwidth requirements for tasks and the capacity
+of CPUs. The basic idea behind the SchedTune knob is to artificially inflate
+some of these load tracking signals to make a task or RQ appears more demanding
+that it actually is.
+
+Which signals have to be inflated depends on the specific "consumer".  However,
+independently from the specific (signal, consumer) pair, it is important to
+define a simple and possibly consistent strategy for the concept of boosting a
+signal.
+
+A boosting strategy defines how the "abstract" user-space defined
+sched_cfs_boost value is translated into an internal "margin" value to be added
+to a signal to get its inflated value:
+
+  margin         := boosting_strategy(sched_cfs_boost, signal)
+  boosted_signal := signal + margin
+
+Different boosting strategies were identified and analyzed before selecting the
+one found to be most effective.
+
+Signal Proportional Compensation (SPC)
+--------------------------------------
+
+In this boosting strategy the sched_cfs_boost value is used to compute a
+margin which is proportional to the complement of the original signal.
+When a signal has a maximum possible value, its complement is defined as
+the delta from the actual value and its possible maximum.
+
+Since the tunable implementation uses signals which have SCHED_LOAD_SCALE as
+the maximum possible value, the margin becomes:
+
+	margin := sched_cfs_boost * (SCHED_LOAD_SCALE - signal)
+
+Using this boosting strategy:
+- a 100% sched_cfs_boost means that the signal is scaled to the maximum value
+- each value in the range of sched_cfs_boost effectively inflates the signal in
+  question by a quantity which is proportional to the maximum value.
+
+For example, by applying the SPC boosting strategy to the selection of the OPP
+to run a task it is possible to achieve these behaviors:
+
+-   0% boosting: run the task at the minimum OPP required by its workload
+- 100% boosting: run the task at the maximum OPP available for the CPU
+-  50% boosting: run at the half-way OPP between minimum and maximum
+
+Which means that, at 50% boosting, a task will be scheduled to run at half of
+the maximum theoretically achievable performance on the specific target
+platform.
+
+A graphical representation of an SPC boosted signal is represented in the
+following figure where:
+ a) "-" represents the original signal
+ b) "b" represents a  50% boosted signal
+ c) "p" represents a 100% boosted signal
+
+
+   ^
+   |  SCHED_LOAD_SCALE
+   +-----------------------------------------------------------------+
+   |pppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppppp
+   |
+   |                                             boosted_signal
+   |                                          bbbbbbbbbbbbbbbbbbbbbbbb
+   |
+   |                                            original signal
+   |                  bbbbbbbbbbbbbbbbbbbbbbbb+----------------------+
+   |                                          |
+   |bbbbbbbbbbbbbbbbbb                        |
+   |                                          |
+   |                                          |
+   |                                          |
+   |                  +-----------------------+
+   |                  |
+   |                  |
+   |                  |
+   |------------------+
+   |
+   |
+   +----------------------------------------------------------------------->
+
+The plot above shows a ramped load signal (titled 'original_signal') and it's
+boosted equivalent. For each step of the original signal the boosted signal
+corresponding to a 50% boost is midway from the original signal and the upper
+bound. Boosting by 100% generates a boosted signal which is always saturated to
+the upper bound.
+
+
+4. OPP selection using boosted CPU utilization
+==============================================
+
+It is worth calling out that the implementation does not introduce any new load
+signals. Instead, it provides an API to tune existing signals. This tuning is
+done on demand and only in scheduler code paths where it is sensible to do so.
+The new API calls are defined to return either the default signal or a boosted
+one, depending on the value of sched_cfs_boost. This is a clean an non invasive
+modification of the existing existing code paths.
+
+The signal representing a CPU's utilization is boosted according to the
+previously described SPC boosting strategy. To sched-DVFS, this allows a CPU
+(ie CFS run-queue) to appear more used then it actually is.
+
+Thus, with the sched_cfs_boost enabled we have the following main functions to
+get the current utilization of a CPU:
+
+  cpu_util()
+  boosted_cpu_util()
+
+The new boosted_cpu_util() is similar to the first but returns a boosted
+utilization signal which is a function of the sched_cfs_boost value.
+
+This function is used in the CFS scheduler code paths where sched-DVFS needs to
+decide the OPP to run a CPU at.
+For example, this allows selecting the highest OPP for a CPU which has
+the boost value set to 100%.
+
+
+5. Per task group boosting
+==========================
+
+The availability of a single knob which is used to boost all tasks in the
+system is certainly a simple solution but it quite likely doesn't fit many
+utilization scenarios, especially in the mobile device space.
+
+For example, on battery powered devices there usually are many background
+services which are long running and need energy efficient scheduling. On the
+other hand, some applications are more performance sensitive and require an
+interactive response and/or maximum performance, regardless of the energy cost.
+To better service such scenarios, the SchedTune implementation has an extension
+that provides a more fine grained boosting interface.
+
+A new CGroup controller, namely "schedtune", could be enabled which allows to
+defined and configure task groups with different boosting values.
+Tasks that require special performance can be put into separate CGroups.
+The value of the boost associated with the tasks in this group can be specified
+using a single knob exposed by the CGroup controller:
+
+   schedtune.boost
+
+This knob allows the definition of a boost value that is to be used for
+SPC boosting of all tasks attached to this group.
+
+The current schedtune controller implementation is really simple and has these
+main characteristics:
+
+  1) It is only possible to create 1 level depth hierarchies
+
+     The root control groups define the system-wide boost value to be applied
+     by default to all tasks. Its direct subgroups are named "boost groups" and
+     they define the boost value for specific set of tasks.
+     Further nested subgroups are not allowed since they do not have a sensible
+     meaning from a user-space standpoint.
+
+  2) It is possible to define only a limited number of "boost groups"
+
+     This number is defined at compile time and by default configured to 16.
+     This is a design decision motivated by two main reasons:
+     a) In a real system we do not expect utilization scenarios with more then few
+	boost groups. For example, a reasonable collection of groups could be
+        just "background", "interactive" and "performance".
+     b) It simplifies the implementation considerably, especially for the code
+	which has to compute the per CPU boosting once there are multiple
+        RUNNABLE tasks with different boost values.
+
+Such a simple design should allow servicing the main utilization scenarios identified
+so far. It provides a simple interface which can be used to manage the
+power-performance of all tasks or only selected tasks.
+Moreover, this interface can be easily integrated by user-space run-times (e.g.
+Android, ChromeOS) to implement a QoS solution for task boosting based on tasks
+classification, which has been a long standing requirement.
+
+Setup and usage
+---------------
+
+0. Use a kernel with CGROUP_SCHEDTUNE support enabled
+
+1. Check that the "schedtune" CGroup controller is available:
+
+   root@linaro-nano:~# cat /proc/cgroups
+   #subsys_name	hierarchy	num_cgroups	enabled
+   cpuset  	0		1		1
+   cpu     	0		1		1
+   schedtune	0		1		1
+
+2. Mount a tmpfs to create the CGroups mount point (Optional)
+
+   root@linaro-nano:~# sudo mount -t tmpfs cgroups /sys/fs/cgroup
+
+3. Mount the "schedtune" controller
+
+   root@linaro-nano:~# mkdir /sys/fs/cgroup/stune
+   root@linaro-nano:~# sudo mount -t cgroup -o schedtune stune /sys/fs/cgroup/stune
+
+4. Setup the system-wide boost value (Optional)
+
+   If not configured the root control group has a 0% boost value, which
+   basically disables boosting for all tasks in the system thus running in
+   an energy-efficient mode.
+
+   root@linaro-nano:~# echo $SYSBOOST > /sys/fs/cgroup/stune/schedtune.boost
+
+5. Create task groups and configure their specific boost value (Optional)
+
+   For example here we create a "performance" boost group configure to boost
+   all its tasks to 100%
+
+   root@linaro-nano:~# mkdir /sys/fs/cgroup/stune/performance
+   root@linaro-nano:~# echo 100 > /sys/fs/cgroup/stune/performance/schedtune.boost
+
+6. Move tasks into the boost group
+
+   For example, the following moves the tasks with PID $TASKPID (and all its
+   threads) into the "performance" boost group.
+
+   root@linaro-nano:~# echo "TASKPID > /sys/fs/cgroup/stune/performance/cgroup.procs
+
+This simple configuration allows only the threads of the $TASKPID task to run,
+when needed, at the highest OPP in the most capable CPU of the system.
+
+
+6. Question and Answers
+=======================
+
+What about "auto" mode?
+-----------------------
+
+The 'auto' mode as described in [5] can be implemented by interfacing SchedTune
+with some suitable user-space element. This element could use the exposed
+system-wide or cgroup based interface.
+
+How are multiple groups of tasks with different boost values managed?
+---------------------------------------------------------------------
+
+The current SchedTune implementation keeps track of the boosted RUNNABLE tasks
+on a CPU. Once sched-DVFS selects the OPP to run a CPU at, the CPU utilization
+is boosted with a value which is the maximum of the boost values of the
+currently RUNNABLE tasks in its RQ.
+
+This allows sched-DVFS to boost a CPU only while there are boosted tasks ready
+to run and switch back to the energy efficient mode as soon as the last boosted
+task is dequeued.
+
+
+7. References
+=============
+[1] http://lwn.net/Articles/552889
+[2] http://lkml.org/lkml/2012/5/18/91
+[3] http://lkml.org/lkml/2015/6/26/620
diff --git a/Documentation/sync.txt b/Documentation/sync.txt
new file mode 100644
index 000000000000..a2d05e7fa193
--- /dev/null
+++ b/Documentation/sync.txt
@@ -0,0 +1,75 @@
+Motivation:
+
+In complicated DMA pipelines such as graphics (multimedia, camera, gpu, display)
+a consumer of a buffer needs to know when the producer has finished producing
+it.  Likewise the producer needs to know when the consumer is finished with the
+buffer so it can reuse it.  A particular buffer may be consumed by multiple
+consumers which will retain the buffer for different amounts of time.  In
+addition, a consumer may consume multiple buffers atomically.
+The sync framework adds an API which allows synchronization between the
+producers and consumers in a generic way while also allowing platforms which
+have shared hardware synchronization primitives to exploit them.
+
+Goals:
+	* provide a generic API for expressing synchronization dependencies
+	* allow drivers to exploit hardware synchronization between hardware
+	  blocks
+	* provide a userspace API that allows a compositor to manage
+	  dependencies.
+	* provide rich telemetry data to allow debugging slowdowns and stalls of
+	   the graphics pipeline.
+
+Objects:
+	* sync_timeline
+	* sync_pt
+	* sync_fence
+
+sync_timeline:
+
+A sync_timeline is an abstract monotonically increasing counter. In general,
+each driver/hardware block context will have one of these.  They can be backed
+by the appropriate hardware or rely on the generic sw_sync implementation.
+Timelines are only ever created through their specific implementations
+(i.e. sw_sync.)
+
+sync_pt:
+
+A sync_pt is an abstract value which marks a point on a sync_timeline. Sync_pts
+have a single timeline parent.  They have 3 states: active, signaled, and error.
+They start in active state and transition, once, to either signaled (when the
+timeline counter advances beyond the sync_pt’s value) or error state.
+
+sync_fence:
+
+Sync_fences are the primary primitives used by drivers to coordinate
+synchronization of their buffers.  They are a collection of sync_pts which may
+or may not have the same timeline parent.  A sync_pt can only exist in one fence
+and the fence's list of sync_pts is immutable once created.  Fences can be
+waited on synchronously or asynchronously.  Two fences can also be merged to
+create a third fence containing a copy of the two fences’ sync_pts.  Fences are
+backed by file descriptors to allow userspace to coordinate the display pipeline
+dependencies.
+
+Use:
+
+A driver implementing sync support should have a work submission function which:
+     * takes a fence argument specifying when to begin work
+     * asynchronously queues that work to kick off when the fence is signaled
+     * returns a fence to indicate when its work will be done.
+     * signals the returned fence once the work is completed.
+
+Consider an imaginary display driver that has the following API:
+/*
+ * assumes buf is ready to be displayed.
+ * blocks until the buffer is on screen.
+ */
+    void display_buffer(struct dma_buf *buf);
+
+The new API will become:
+/*
+ * will display buf when fence is signaled.
+ * returns immediately with a fence that will signal when buf
+ * is no longer displayed.
+ */
+struct sync_fence* display_buffer(struct dma_buf *buf,
+                                 struct sync_fence *fence);
diff --git a/Documentation/sysctl/kernel.txt b/Documentation/sysctl/kernel.txt
index be61d53e997f..e7018fc62ffa 100644
--- a/Documentation/sysctl/kernel.txt
+++ b/Documentation/sysctl/kernel.txt
@@ -58,6 +58,8 @@ show up in /proc/sys/kernel:
 - panic_on_stackoverflow
 - panic_on_unrecovered_nmi
 - panic_on_warn
+- perf_cpu_time_max_percent
+- perf_event_paranoid
 - pid_max
 - powersave-nap               [ PPC only ]
 - printk
@@ -624,6 +626,19 @@ allowed to execute.
 
 ==============================================================
 
+perf_event_paranoid:
+
+Controls use of the performance events system by unprivileged
+users (without CAP_SYS_ADMIN).  The default value is 3 if
+CONFIG_SECURITY_PERF_EVENTS_RESTRICT is set, or 1 otherwise.
+
+ -1: Allow use of (almost) all events by all users
+>=0: Disallow raw tracepoint access by users without CAP_IOC_LOCK
+>=1: Disallow CPU event access by users without CAP_SYS_ADMIN
+>=2: Disallow kernel profiling by users without CAP_SYS_ADMIN
+>=3: Disallow all event access by users without CAP_SYS_ADMIN
+
+==============================================================
 
 pid_max:
 
diff --git a/Documentation/sysctl/vm.txt b/Documentation/sysctl/vm.txt
index f72370b440b1..c0397afe6a1d 100644
--- a/Documentation/sysctl/vm.txt
+++ b/Documentation/sysctl/vm.txt
@@ -30,6 +30,7 @@ Currently, these files are in /proc/sys/vm:
 - dirty_writeback_centisecs
 - drop_caches
 - extfrag_threshold
+- extra_free_kbytes
 - hugepages_treat_as_movable
 - hugetlb_shm_group
 - laptop_mode
@@ -42,6 +43,8 @@ Currently, these files are in /proc/sys/vm:
 - min_slab_ratio
 - min_unmapped_ratio
 - mmap_min_addr
+- mmap_rnd_bits
+- mmap_rnd_compat_bits
 - nr_hugepages
 - nr_overcommit_hugepages
 - nr_trim_pages         (only if CONFIG_MMU=n)
@@ -236,6 +239,21 @@ fragmentation index is <= extfrag_threshold. The default value is 500.
 
 ==============================================================
 
+extra_free_kbytes
+
+This parameter tells the VM to keep extra free memory between the threshold
+where background reclaim (kswapd) kicks in, and the threshold where direct
+reclaim (by allocating processes) kicks in.
+
+This is useful for workloads that require low latency memory allocations
+and have a bounded burstiness in memory allocations, for example a
+realtime application that receives and transmits network traffic
+(causing in-kernel memory allocations) with a maximum total message burst
+size of 200MB may need 200MB of extra free memory to avoid direct reclaim
+related latencies.
+
+==============================================================
+
 hugepages_treat_as_movable
 
 This parameter controls whether we can allocate hugepages from ZONE_MOVABLE
@@ -485,6 +503,33 @@ against future potential kernel bugs.
 
 ==============================================================
 
+mmap_rnd_bits:
+
+This value can be used to select the number of bits to use to
+determine the random offset to the base address of vma regions
+resulting from mmap allocations on architectures which support
+tuning address space randomization.  This value will be bounded
+by the architecture's minimum and maximum supported values.
+
+This value can be changed after boot using the
+/proc/sys/vm/mmap_rnd_bits tunable
+
+==============================================================
+
+mmap_rnd_compat_bits:
+
+This value can be used to select the number of bits to use to
+determine the random offset to the base address of vma regions
+resulting from mmap allocations for applications run in
+compatibility mode on architectures which support tuning address
+space randomization.  This value will be bounded by the
+architecture's minimum and maximum supported values.
+
+This value can be changed after boot using the
+/proc/sys/vm/mmap_rnd_compat_bits tunable
+
+==============================================================
+
 nr_hugepages
 
 Change the minimum size of the hugepage pool.
diff --git a/Documentation/trace/events-power.txt b/Documentation/trace/events-power.txt
index 21d514ced212..4d817d5acc40 100644
--- a/Documentation/trace/events-power.txt
+++ b/Documentation/trace/events-power.txt
@@ -25,6 +25,7 @@ cpufreq.
 
 cpu_idle		"state=%lu cpu_id=%lu"
 cpu_frequency		"state=%lu cpu_id=%lu"
+cpu_frequency_limits	"min=%lu max=%lu cpu_id=%lu"
 
 A suspend event is used to indicate the system going in and out of the
 suspend mode:
diff --git a/Documentation/trace/ftrace.txt b/Documentation/trace/ftrace.txt
index f52f297cb406..fa16fb2302a5 100644
--- a/Documentation/trace/ftrace.txt
+++ b/Documentation/trace/ftrace.txt
@@ -357,6 +357,26 @@ of ftrace. Here is a list of some of the key files:
 		  to correlate events across hypervisor/guest if
 		  tb_offset is known.
 
+	  mono: This uses the fast monotonic clock (CLOCK_MONOTONIC)
+		which is monotonic and is subject to NTP rate adjustments.
+
+	  mono_raw:
+		This is the raw monotonic clock (CLOCK_MONOTONIC_RAW)
+		which is montonic but is not subject to any rate adjustments
+		and ticks at the same rate as the hardware clocksource.
+
+	  boot: This is the boot clock (CLOCK_BOOTTIME) and is based on the
+		fast monotonic clock, but also accounts for time spent in
+		suspend. Since the clock access is designed for use in
+		tracing in the suspend path, some side effects are possible
+		if clock is accessed after the suspend time is accounted before
+		the fast mono clock is updated. In this case, the clock update
+		appears to happen slightly sooner than it normally would have.
+		Also on 32-bit systems, its possible that the 64-bit boot offset
+		sees a partial update. These effects are rare and post
+		processing should be able to handle them. See comments on
+		ktime_get_boot_fast_ns function for more information.
+
 	To set a clock, simply echo the clock name into this file.
 
 	  echo global > trace_clock
@@ -2088,6 +2108,35 @@ will produce:
  1)   1.449 us    |             }
 
 
+You can disable the hierarchical function call formatting and instead print a
+flat list of function entry and return events.  This uses the format described
+in the Output Formatting section and respects all the trace options that
+control that formatting.  Hierarchical formatting is the default.
+
+	hierachical: echo nofuncgraph-flat > trace_options
+	flat: echo funcgraph-flat > trace_options
+
+  ie:
+
+  # tracer: function_graph
+  #
+  # entries-in-buffer/entries-written: 68355/68355   #P:2
+  #
+  #                              _-----=> irqs-off
+  #                             / _----=> need-resched
+  #                            | / _---=> hardirq/softirq
+  #                            || / _--=> preempt-depth
+  #                            ||| /     delay
+  #           TASK-PID   CPU#  ||||    TIMESTAMP  FUNCTION
+  #              | |       |   ||||       |         |
+                sh-1806  [001] d...   198.843443: graph_ent: func=_raw_spin_lock
+                sh-1806  [001] d...   198.843445: graph_ent: func=__raw_spin_lock
+                sh-1806  [001] d..1   198.843447: graph_ret: func=__raw_spin_lock
+                sh-1806  [001] d..1   198.843449: graph_ret: func=_raw_spin_lock
+                sh-1806  [001] d..1   198.843451: graph_ent: func=_raw_spin_unlock_irqrestore
+                sh-1806  [001] d...   198.843453: graph_ret: func=_raw_spin_unlock_irqrestore
+
+
 You might find other useful features for this tracer in the
 following "dynamic ftrace" section such as tracing only specific
 functions or tasks.
diff --git a/Kbuild b/Kbuild
index f55cefd9bf29..f56ed561a284 100644
--- a/Kbuild
+++ b/Kbuild
@@ -6,31 +6,6 @@
 # 3) Generate asm-offsets.h (may need bounds.h and timeconst.h)
 # 4) Check for missing system calls
 
-# Default sed regexp - multiline due to syntax constraints
-define sed-y
-	"/^->/{s:->#\(.*\):/* \1 */:; \
-	s:^->\([^ ]*\) [\$$#]*\([-0-9]*\) \(.*\):#define \1 \2 /* \3 */:; \
-	s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; \
-	s:->::; p;}"
-endef
-
-# Use filechk to avoid rebuilds when a header changes, but the resulting file
-# does not
-define filechk_offsets
-	(set -e; \
-	 echo "#ifndef $2"; \
-	 echo "#define $2"; \
-	 echo "/*"; \
-	 echo " * DO NOT MODIFY."; \
-	 echo " *"; \
-	 echo " * This file was generated by Kbuild"; \
-	 echo " */"; \
-	 echo ""; \
-	 sed -ne $(sed-y); \
-	 echo ""; \
-	 echo "#endif" )
-endef
-
 #####
 # 1) Generate bounds.h
 
diff --git a/Makefile b/Makefile
index 9e036fac9c04..c5a13e031c85 100644
--- a/Makefile
+++ b/Makefile
@@ -146,7 +146,7 @@ PHONY += $(MAKECMDGOALS) sub-make
 $(filter-out _all sub-make $(CURDIR)/Makefile, $(MAKECMDGOALS)) _all: sub-make
 	@:
 
-sub-make: FORCE
+sub-make:
 	$(Q)$(MAKE) -C $(KBUILD_OUTPUT) KBUILD_SRC=$(CURDIR) \
 	-f $(CURDIR)/Makefile $(filter-out _all sub-make,$(MAKECMDGOALS))
 
@@ -301,7 +301,7 @@ CONFIG_SHELL := $(shell if [ -x "$$BASH" ]; then echo $$BASH; \
 
 HOSTCC       = gcc
 HOSTCXX      = g++
-HOSTCFLAGS   = -Wall -Wmissing-prototypes -Wstrict-prototypes -O2 -fomit-frame-pointer -std=gnu89
+HOSTCFLAGS   := -Wall -Wmissing-prototypes -Wstrict-prototypes -O2 -fomit-frame-pointer -std=gnu89
 HOSTCXXFLAGS = -O2
 
 ifeq ($(shell $(HOSTCC) -v 2>&1 | grep -c "clang version"), 1)
@@ -624,7 +624,7 @@ KBUILD_CFLAGS	+= $(call cc-disable-warning, format-overflow)
 KBUILD_CFLAGS	+= $(call cc-disable-warning, int-in-bool-context)
 
 ifdef CONFIG_CC_OPTIMIZE_FOR_SIZE
-KBUILD_CFLAGS	+= -Os
+KBUILD_CFLAGS	+= $(call cc-option,-Oz,-Os)
 else
 ifdef CONFIG_PROFILE_ALL_BRANCHES
 KBUILD_CFLAGS	+= -O2
@@ -694,11 +694,22 @@ endif
 KBUILD_CFLAGS += $(stackp-flag)
 
 ifeq ($(cc-name),clang)
+ifneq ($(CROSS_COMPILE),)
+CLANG_TRIPLE    ?= $(CROSS_COMPILE)
+CLANG_TARGET	:= --target=$(notdir $(CLANG_TRIPLE:%-=%))
+GCC_TOOLCHAIN	:= $(realpath $(dir $(shell which $(LD)))/..)
+endif
+ifneq ($(GCC_TOOLCHAIN),)
+CLANG_GCC_TC	:= --gcc-toolchain=$(GCC_TOOLCHAIN)
+endif
+KBUILD_CFLAGS += $(CLANG_TARGET) $(CLANG_GCC_TC)
+KBUILD_AFLAGS += $(CLANG_TARGET) $(CLANG_GCC_TC)
 KBUILD_CPPFLAGS += $(call cc-option,-Qunused-arguments,)
-KBUILD_CPPFLAGS += $(call cc-option,-Wno-unknown-warning-option,)
 KBUILD_CFLAGS += $(call cc-disable-warning, unused-variable)
 KBUILD_CFLAGS += $(call cc-disable-warning, format-invalid-specifier)
 KBUILD_CFLAGS += $(call cc-disable-warning, gnu)
+KBUILD_CFLAGS += $(call cc-disable-warning, address-of-packed-member)
+KBUILD_CFLAGS += $(call cc-disable-warning, duplicate-decl-specifier)
 # Quiet clang warning: comparison of unsigned expression < 0 is always false
 KBUILD_CFLAGS += $(call cc-disable-warning, tautological-compare)
 # CLANG uses a _MergedGlobals as optimization, but this breaks modpost, as the
@@ -706,6 +717,8 @@ KBUILD_CFLAGS += $(call cc-disable-warning, tautological-compare)
 # See modpost pattern 2
 KBUILD_CFLAGS += $(call cc-option, -mno-global-merge,)
 KBUILD_CFLAGS += $(call cc-option, -fcatch-undefined-behavior)
+KBUILD_CFLAGS += $(call cc-option, -no-integrated-as)
+KBUILD_AFLAGS += $(call cc-option, -no-integrated-as)
 else
 
 # These warnings generated too much noise in a regular build.
@@ -1003,7 +1016,7 @@ prepare1: prepare2 $(version_h) include/generated/utsrelease.h \
 
 archprepare: archheaders archscripts prepare1 scripts_basic
 
-prepare0: archprepare FORCE
+prepare0: archprepare
 	$(Q)$(MAKE) $(build)=.
 
 # All the preparing..
@@ -1048,7 +1061,7 @@ INSTALL_FW_PATH=$(INSTALL_MOD_PATH)/lib/firmware
 export INSTALL_FW_PATH
 
 PHONY += firmware_install
-firmware_install: FORCE
+firmware_install:
 	@mkdir -p $(objtree)/firmware
 	$(Q)$(MAKE) -f $(srctree)/scripts/Makefile.fwinst obj=firmware __fw_install
 
@@ -1068,7 +1081,7 @@ PHONY += archscripts
 archscripts:
 
 PHONY += __headers
-__headers: $(version_h) scripts_basic asm-generic archheaders archscripts FORCE
+__headers: $(version_h) scripts_basic asm-generic archheaders archscripts
 	$(Q)$(MAKE) $(build)=scripts build_unifdef
 
 PHONY += headers_install_all
@@ -1281,6 +1294,8 @@ help:
 	@echo  '                    (default: $$(INSTALL_MOD_PATH)/lib/firmware)'
 	@echo  '  dir/            - Build all files in dir and below'
 	@echo  '  dir/file.[ois]  - Build specified target only'
+	@echo  '  dir/file.ll     - Build the LLVM assembly file'
+	@echo  '                    (requires compiler support for LLVM assembly generation)'
 	@echo  '  dir/file.lst    - Build specified mixed source/assembly target only'
 	@echo  '                    (requires a recent binutils and recent build (System.map))'
 	@echo  '  dir/file.ko     - Build module including final link'
@@ -1456,6 +1471,7 @@ clean: $(clean-dirs)
 		-o -name '.*.d' -o -name '.*.tmp' -o -name '*.mod.c' \
 		-o -name '*.symtypes' -o -name 'modules.order' \
 		-o -name modules.builtin -o -name '.tmp_*.o.*' \
+		-o -name '*.ll' \
 		-o -name '*.gcno' \) -type f -print | xargs rm -f
 
 # Generate tags for editors
@@ -1559,6 +1575,8 @@ endif
 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
 %.symtypes: %.c prepare scripts FORCE
 	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
+%.ll: %.c prepare scripts FORCE
+	$(Q)$(MAKE) $(build)=$(build-dir) $(target-dir)$(notdir $@)
 
 # Modules
 /: prepare scripts FORCE
diff --git a/android/configs/android-fetch-configs.sh b/android/configs/android-fetch-configs.sh
new file mode 100755
index 000000000000..9915c1356ed3
--- /dev/null
+++ b/android/configs/android-fetch-configs.sh
@@ -0,0 +1,4 @@
+#!/bin/sh
+
+curl https://android.googlesource.com/kernel/configs/+archive/master/android-4.4.tar.gz | tar xzv
+
diff --git a/arch/Kconfig b/arch/Kconfig
index d4d9845530f1..ed2539c590bf 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -225,8 +225,8 @@ config ARCH_INIT_TASK
 config ARCH_TASK_STRUCT_ALLOCATOR
 	bool
 
-# Select if arch has its private alloc_thread_info() function
-config ARCH_THREAD_INFO_ALLOCATOR
+# Select if arch has its private alloc_thread_stack() function
+config ARCH_THREAD_STACK_ALLOCATOR
 	bool
 
 # Select if arch wants to size task_struct dynamically via arch_task_struct_size:
@@ -527,6 +527,74 @@ config HAVE_COPY_THREAD_TLS
 	  normal C parameter passing, rather than extracting the syscall
 	  argument from pt_regs.
 
+config HAVE_ARCH_MMAP_RND_BITS
+	bool
+	help
+	  An arch should select this symbol if it supports setting a variable
+	  number of bits for use in establishing the base address for mmap
+	  allocations, has MMU enabled and provides values for both:
+	  - ARCH_MMAP_RND_BITS_MIN
+	  - ARCH_MMAP_RND_BITS_MAX
+
+config ARCH_MMAP_RND_BITS_MIN
+	int
+
+config ARCH_MMAP_RND_BITS_MAX
+	int
+
+config ARCH_MMAP_RND_BITS_DEFAULT
+	int
+
+config ARCH_MMAP_RND_BITS
+	int "Number of bits to use for ASLR of mmap base address" if EXPERT
+	range ARCH_MMAP_RND_BITS_MIN ARCH_MMAP_RND_BITS_MAX
+	default ARCH_MMAP_RND_BITS_DEFAULT if ARCH_MMAP_RND_BITS_DEFAULT
+	default ARCH_MMAP_RND_BITS_MIN
+	depends on HAVE_ARCH_MMAP_RND_BITS
+	help
+	  This value can be used to select the number of bits to use to
+	  determine the random offset to the base address of vma regions
+	  resulting from mmap allocations. This value will be bounded
+	  by the architecture's minimum and maximum supported values.
+
+	  This value can be changed after boot using the
+	  /proc/sys/vm/mmap_rnd_bits tunable
+
+config HAVE_ARCH_MMAP_RND_COMPAT_BITS
+	bool
+	help
+	  An arch should select this symbol if it supports running applications
+	  in compatibility mode, supports setting a variable number of bits for
+	  use in establishing the base address for mmap allocations, has MMU
+	  enabled and provides values for both:
+	  - ARCH_MMAP_RND_COMPAT_BITS_MIN
+	  - ARCH_MMAP_RND_COMPAT_BITS_MAX
+
+config ARCH_MMAP_RND_COMPAT_BITS_MIN
+	int
+
+config ARCH_MMAP_RND_COMPAT_BITS_MAX
+	int
+
+config ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
+	int
+
+config ARCH_MMAP_RND_COMPAT_BITS
+	int "Number of bits to use for ASLR of mmap base address for compatible applications" if EXPERT
+	range ARCH_MMAP_RND_COMPAT_BITS_MIN ARCH_MMAP_RND_COMPAT_BITS_MAX
+	default ARCH_MMAP_RND_COMPAT_BITS_DEFAULT if ARCH_MMAP_RND_COMPAT_BITS_DEFAULT
+	default ARCH_MMAP_RND_COMPAT_BITS_MIN
+	depends on HAVE_ARCH_MMAP_RND_COMPAT_BITS
+	help
+	  This value can be used to select the number of bits to use to
+	  determine the random offset to the base address of vma regions
+	  resulting from mmap allocations for compatible applications This
+	  value will be bounded by the architecture's minimum and maximum
+	  supported values.
+
+	  This value can be changed after boot using the
+	  /proc/sys/vm/mmap_rnd_compat_bits tunable
+
 #
 # ABI hall of shame
 #
diff --git a/arch/alpha/kernel/module.c b/arch/alpha/kernel/module.c
index 2fd00b7077e4..936bc8f89a67 100644
--- a/arch/alpha/kernel/module.c
+++ b/arch/alpha/kernel/module.c
@@ -160,7 +160,7 @@ apply_relocate_add(Elf64_Shdr *sechdrs, const char *strtab,
 
 	/* The small sections were sorted to the end of the segment.
 	   The following should definitely cover them.  */
-	gp = (u64)me->module_core + me->core_size - 0x8000;
+	gp = (u64)me->core_layout.base + me->core_layout.size - 0x8000;
 	got = sechdrs[me->arch.gotsecindex].sh_addr;
 
 	for (i = 0; i < n; i++) {
diff --git a/arch/arc/kernel/unwind.c b/arch/arc/kernel/unwind.c
index 5eb707640e9c..0587bf121d11 100644
--- a/arch/arc/kernel/unwind.c
+++ b/arch/arc/kernel/unwind.c
@@ -385,8 +385,8 @@ void *unwind_add_table(struct module *module, const void *table_start,
 		return NULL;
 
 	init_unwind_table(table, module->name,
-			  module->module_core, module->core_size,
-			  module->module_init, module->init_size,
+			  module->core_layout.base, module->core_layout.size,
+			  module->init_layout.base, module->init_layout.size,
 			  table_start, table_size,
 			  NULL, 0);
 
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
index 9049ac023bee..625765fb805a 100644
--- a/arch/arm/Kconfig
+++ b/arch/arm/Kconfig
@@ -35,6 +35,7 @@ config ARM
 	select HAVE_ARCH_BITREVERSE if (CPU_32v7M || CPU_32v7) && !CPU_32v6
 	select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL && !CPU_ENDIAN_BE32
 	select HAVE_ARCH_KGDB if !CPU_ENDIAN_BE32
+	select HAVE_ARCH_MMAP_RND_BITS if MMU
 	select HAVE_ARCH_HARDENED_USERCOPY
 	select HAVE_ARCH_SECCOMP_FILTER if (AEABI && !OABI_COMPAT)
 	select HAVE_ARCH_TRACEHOOK
@@ -310,6 +311,14 @@ config MMU
 	  Select if you want MMU-based virtualised addressing space
 	  support by paged memory management. If unsure, say 'Y'.
 
+config ARCH_MMAP_RND_BITS_MIN
+	default 8
+
+config ARCH_MMAP_RND_BITS_MAX
+	default 14 if PAGE_OFFSET=0x40000000
+	default 15 if PAGE_OFFSET=0x80000000
+	default 16
+
 #
 # The "ARM system type" choice list is ordered alphabetically by option
 # text.  Please add new entries in the option alphabetic order.
@@ -1817,6 +1826,15 @@ config XEN
 	help
 	  Say Y if you want to run Linux in a Virtual Machine on Xen on ARM.
 
+config ARM_FLUSH_CONSOLE_ON_RESTART
+	bool "Force flush the console on restart"
+	help
+	  If the console is locked while the system is rebooted, the messages
+	  in the temporary logbuffer would not have propogated to all the
+	  console drivers. This option forces the console lock to be
+	  released if it failed to be acquired, which will cause all the
+	  pending messages to be flushed.
+
 endmenu
 
 menu "Boot options"
@@ -1845,6 +1863,21 @@ config DEPRECATED_PARAM_STRUCT
 	  This was deprecated in 2001 and announced to live on for 5 years.
 	  Some old boot loaders still use this way.
 
+config BUILD_ARM_APPENDED_DTB_IMAGE
+	bool "Build a concatenated zImage/dtb by default"
+	depends on OF
+	help
+	  Enabling this option will cause a concatenated zImage and list of
+	  DTBs to be built by default (instead of a standalone zImage.)
+	  The image will built in arch/arm/boot/zImage-dtb
+
+config BUILD_ARM_APPENDED_DTB_IMAGE_NAMES
+	string "Default dtb names"
+	depends on BUILD_ARM_APPENDED_DTB_IMAGE
+	help
+	  Space separated list of names of dtbs to append when
+	  building a concatenated zImage-dtb.
+
 # Compressed boot loader in ROM.  Yes, we really want to ask about
 # TEXT and BSS so we preserve their values in the config files.
 config ZBOOT_ROM_TEXT
diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug
index ddbb361267d8..f08e4fd600a3 100644
--- a/arch/arm/Kconfig.debug
+++ b/arch/arm/Kconfig.debug
@@ -1618,6 +1618,14 @@ config EARLY_PRINTK
 	  kernel low-level debugging functions. Add earlyprintk to your
 	  kernel parameters to enable this console.
 
+config EARLY_PRINTK_DIRECT
+	bool "Early printk direct"
+	depends on DEBUG_LL
+	help
+	  Say Y here if you want to have an early console using the
+	  kernel low-level debugging functions and EARLY_PRINTK is
+	  not early enough.
+
 config ARM_KPROBES_TEST
 	tristate "Kprobes test module"
 	depends on KPROBES && MODULES
diff --git a/arch/arm/Makefile b/arch/arm/Makefile
index 2c2b28ee4811..88e479c8bcef 100644
--- a/arch/arm/Makefile
+++ b/arch/arm/Makefile
@@ -296,6 +296,8 @@ libs-y				:= arch/arm/lib/ $(libs-y)
 # Default target when executing plain make
 ifeq ($(CONFIG_XIP_KERNEL),y)
 KBUILD_IMAGE := xipImage
+else ifeq ($(CONFIG_BUILD_ARM_APPENDED_DTB_IMAGE),y)
+KBUILD_IMAGE := zImage-dtb
 else
 KBUILD_IMAGE := zImage
 endif
@@ -346,6 +348,9 @@ ifeq ($(CONFIG_VDSO),y)
 	$(Q)$(MAKE) $(build)=arch/arm/vdso $@
 endif
 
+zImage-dtb: vmlinux scripts dtbs
+	$(Q)$(MAKE) $(build)=$(boot) MACHINE=$(MACHINE) $(boot)/$@
+
 # We use MRPROPER_FILES and CLEAN_FILES now
 archclean:
 	$(Q)$(MAKE) $(clean)=$(boot)
diff --git a/arch/arm/boot/.gitignore b/arch/arm/boot/.gitignore
index 3c79f85975aa..ad7a0253ea96 100644
--- a/arch/arm/boot/.gitignore
+++ b/arch/arm/boot/.gitignore
@@ -4,3 +4,4 @@ xipImage
 bootpImage
 uImage
 *.dtb
+zImage-dtb
\ No newline at end of file
diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile
index 9eca7aee927f..4a04ed3daf97 100644
--- a/arch/arm/boot/Makefile
+++ b/arch/arm/boot/Makefile
@@ -14,6 +14,7 @@
 ifneq ($(MACHINE),)
 include $(MACHINE)/Makefile.boot
 endif
+include $(srctree)/arch/arm/boot/dts/Makefile
 
 # Note: the following conditions must always be true:
 #   ZRELADDR == virt_to_phys(PAGE_OFFSET + TEXT_OFFSET)
@@ -27,6 +28,14 @@ export ZRELADDR INITRD_PHYS PARAMS_PHYS
 
 targets := Image zImage xipImage bootpImage uImage
 
+DTB_NAMES := $(subst $\",,$(CONFIG_BUILD_ARM_APPENDED_DTB_IMAGE_NAMES))
+ifneq ($(DTB_NAMES),)
+DTB_LIST := $(addsuffix .dtb,$(DTB_NAMES))
+else
+DTB_LIST := $(dtb-y)
+endif
+DTB_OBJS := $(addprefix $(obj)/dts/,$(DTB_LIST))
+
 ifeq ($(CONFIG_XIP_KERNEL),y)
 
 $(obj)/xipImage: vmlinux FORCE
@@ -55,6 +64,10 @@ $(obj)/zImage:	$(obj)/compressed/vmlinux FORCE
 	$(call if_changed,objcopy)
 	@$(kecho) '  Kernel: $@ is ready'
 
+$(obj)/zImage-dtb:	$(obj)/zImage $(DTB_OBJS) FORCE
+	$(call if_changed,cat)
+	@echo '  Kernel: $@ is ready'
+
 endif
 
 ifneq ($(LOADADDR),)
diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S
index 856913705169..d2e43b053d9b 100644
--- a/arch/arm/boot/compressed/head.S
+++ b/arch/arm/boot/compressed/head.S
@@ -778,6 +778,8 @@ __armv7_mmu_cache_on:
 		bic     r6, r6, #1 << 31        @ 32-bit translation system
 		bic     r6, r6, #(7 << 0) | (1 << 4)	@ use only ttbr0
 		mcrne	p15, 0, r3, c2, c0, 0	@ load page table pointer
+		mcrne	p15, 0, r0, c8, c7, 0	@ flush I,D TLBs
+		mcr	p15, 0, r0, c7, c5, 4	@ ISB
 		mcrne	p15, 0, r1, c3, c0, 0	@ load domain access control
 		mcrne   p15, 0, r6, c2, c0, 2   @ load ttb control
 #endif
diff --git a/arch/arm/boot/dts/Makefile b/arch/arm/boot/dts/Makefile
index 30bbc3746130..97d1b3719c75 100644
--- a/arch/arm/boot/dts/Makefile
+++ b/arch/arm/boot/dts/Makefile
@@ -782,5 +782,15 @@ endif
 dtstree		:= $(srctree)/$(src)
 dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(wildcard $(dtstree)/*.dts))
 
-always		:= $(dtb-y)
+DTB_NAMES := $(subst $\",,$(CONFIG_BUILD_ARM_APPENDED_DTB_IMAGE_NAMES))
+ifneq ($(DTB_NAMES),)
+DTB_LIST := $(addsuffix .dtb,$(DTB_NAMES))
+else
+DTB_LIST := $(dtb-y)
+endif
+
+targets += dtbs dtbs_install
+targets += $(DTB_LIST)
+
+always		:= $(DTB_LIST)
 clean-files	:= *.dtb
diff --git a/arch/arm/common/Kconfig b/arch/arm/common/Kconfig
index 9353184d730d..ce01364a96e3 100644
--- a/arch/arm/common/Kconfig
+++ b/arch/arm/common/Kconfig
@@ -17,3 +17,7 @@ config SHARP_PARAM
 
 config SHARP_SCOOP
 	bool
+
+config FIQ_GLUE
+	bool
+	select FIQ
diff --git a/arch/arm/common/Makefile b/arch/arm/common/Makefile
index 27f23b15b1ea..04aca896b338 100644
--- a/arch/arm/common/Makefile
+++ b/arch/arm/common/Makefile
@@ -4,6 +4,7 @@
 
 obj-y				+= firmware.o
 
+obj-$(CONFIG_FIQ_GLUE)		+= fiq_glue.o fiq_glue_setup.o
 obj-$(CONFIG_ICST)		+= icst.o
 obj-$(CONFIG_SA1111)		+= sa1111.o
 obj-$(CONFIG_DMABOUNCE)		+= dmabounce.o
diff --git a/arch/arm/common/fiq_glue.S b/arch/arm/common/fiq_glue.S
new file mode 100644
index 000000000000..24b42cec4813
--- /dev/null
+++ b/arch/arm/common/fiq_glue.S
@@ -0,0 +1,118 @@
+/*
+ * Copyright (C) 2008 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+		.text
+
+		.global fiq_glue_end
+
+		/* fiq stack: r0-r15,cpsr,spsr of interrupted mode */
+
+ENTRY(fiq_glue)
+		/* store pc, cpsr from previous mode, reserve space for spsr */
+		mrs	r12, spsr
+		sub	lr, lr, #4
+		subs	r10, #1
+		bne	nested_fiq
+
+		str	r12, [sp, #-8]!
+		str	lr, [sp, #-4]!
+
+		/* store r8-r14 from previous mode */
+		sub	sp, sp, #(7 * 4)
+		stmia	sp, {r8-r14}^
+		nop
+
+		/* store r0-r7 from previous mode */
+		stmfd	sp!, {r0-r7}
+
+		/* setup func(data,regs) arguments */
+		mov	r0, r9
+		mov	r1, sp
+		mov	r3, r8
+
+		mov	r7, sp
+
+		/* Get sp and lr from non-user modes */
+		and	r4, r12, #MODE_MASK
+		cmp	r4, #USR_MODE
+		beq	fiq_from_usr_mode
+
+		mov	r7, sp
+		orr	r4, r4, #(PSR_I_BIT | PSR_F_BIT)
+		msr	cpsr_c, r4
+		str	sp, [r7, #(4 * 13)]
+		str	lr, [r7, #(4 * 14)]
+		mrs	r5, spsr
+		str	r5, [r7, #(4 * 17)]
+
+		cmp	r4, #(SVC_MODE | PSR_I_BIT | PSR_F_BIT)
+		/* use fiq stack if we reenter this mode */
+		subne	sp, r7, #(4 * 3)
+
+fiq_from_usr_mode:
+		msr	cpsr_c, #(SVC_MODE | PSR_I_BIT | PSR_F_BIT)
+		mov	r2, sp
+		sub	sp, r7, #12
+		stmfd	sp!, {r2, ip, lr}
+		/* call func(data,regs) */
+		blx	r3
+		ldmfd	sp, {r2, ip, lr}
+		mov	sp, r2
+
+		/* restore/discard saved state */
+		cmp	r4, #USR_MODE
+		beq	fiq_from_usr_mode_exit
+
+		msr	cpsr_c, r4
+		ldr	sp, [r7, #(4 * 13)]
+		ldr	lr, [r7, #(4 * 14)]
+		msr	spsr_cxsf, r5
+
+fiq_from_usr_mode_exit:
+		msr	cpsr_c, #(FIQ_MODE | PSR_I_BIT | PSR_F_BIT)
+
+		ldmfd	sp!, {r0-r7}
+		ldr	lr, [sp, #(4 * 7)]
+		ldr	r12, [sp, #(4 * 8)]
+		add	sp, sp, #(10 * 4)
+exit_fiq:
+		msr	spsr_cxsf, r12
+		add	r10, #1
+		cmp	r11, #0
+		moveqs	pc, lr
+		bx	r11 /* jump to custom fiq return function */
+
+nested_fiq:
+		orr	r12, r12, #(PSR_F_BIT)
+		b	exit_fiq
+
+fiq_glue_end:
+
+ENTRY(fiq_glue_setup) /* func, data, sp, smc call number */
+		stmfd		sp!, {r4}
+		mrs		r4, cpsr
+		msr		cpsr_c, #(FIQ_MODE | PSR_I_BIT | PSR_F_BIT)
+		movs		r8, r0
+		mov		r9, r1
+		mov		sp, r2
+		mov		r11, r3
+		moveq		r10, #0
+		movne		r10, #1
+		msr		cpsr_c, r4
+		ldmfd		sp!, {r4}
+		bx		lr
+
diff --git a/arch/arm/common/fiq_glue_setup.c b/arch/arm/common/fiq_glue_setup.c
new file mode 100644
index 000000000000..8cb1b611c6d5
--- /dev/null
+++ b/arch/arm/common/fiq_glue_setup.c
@@ -0,0 +1,147 @@
+/*
+ * Copyright (C) 2010 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <linux/slab.h>
+#include <asm/fiq.h>
+#include <asm/fiq_glue.h>
+
+extern unsigned char fiq_glue, fiq_glue_end;
+extern void fiq_glue_setup(void *func, void *data, void *sp,
+			   fiq_return_handler_t fiq_return_handler);
+
+static struct fiq_handler fiq_debbuger_fiq_handler = {
+	.name = "fiq_glue",
+};
+DEFINE_PER_CPU(void *, fiq_stack);
+static struct fiq_glue_handler *current_handler;
+static fiq_return_handler_t fiq_return_handler;
+static DEFINE_MUTEX(fiq_glue_lock);
+
+static void fiq_glue_setup_helper(void *info)
+{
+	struct fiq_glue_handler *handler = info;
+	fiq_glue_setup(handler->fiq, handler,
+		__get_cpu_var(fiq_stack) + THREAD_START_SP,
+		fiq_return_handler);
+}
+
+int fiq_glue_register_handler(struct fiq_glue_handler *handler)
+{
+	int ret;
+	int cpu;
+
+	if (!handler || !handler->fiq)
+		return -EINVAL;
+
+	mutex_lock(&fiq_glue_lock);
+	if (fiq_stack) {
+		ret = -EBUSY;
+		goto err_busy;
+	}
+
+	for_each_possible_cpu(cpu) {
+		void *stack;
+		stack = (void *)__get_free_pages(GFP_KERNEL, THREAD_SIZE_ORDER);
+		if (WARN_ON(!stack)) {
+			ret = -ENOMEM;
+			goto err_alloc_fiq_stack;
+		}
+		per_cpu(fiq_stack, cpu) = stack;
+	}
+
+	ret = claim_fiq(&fiq_debbuger_fiq_handler);
+	if (WARN_ON(ret))
+		goto err_claim_fiq;
+
+	current_handler = handler;
+	on_each_cpu(fiq_glue_setup_helper, handler, true);
+	set_fiq_handler(&fiq_glue, &fiq_glue_end - &fiq_glue);
+
+	mutex_unlock(&fiq_glue_lock);
+	return 0;
+
+err_claim_fiq:
+err_alloc_fiq_stack:
+	for_each_possible_cpu(cpu) {
+		__free_pages(per_cpu(fiq_stack, cpu), THREAD_SIZE_ORDER);
+		per_cpu(fiq_stack, cpu) = NULL;
+	}
+err_busy:
+	mutex_unlock(&fiq_glue_lock);
+	return ret;
+}
+
+static void fiq_glue_update_return_handler(void (*fiq_return)(void))
+{
+	fiq_return_handler = fiq_return;
+	if (current_handler)
+		on_each_cpu(fiq_glue_setup_helper, current_handler, true);
+}
+
+int fiq_glue_set_return_handler(void (*fiq_return)(void))
+{
+	int ret;
+
+	mutex_lock(&fiq_glue_lock);
+	if (fiq_return_handler) {
+		ret = -EBUSY;
+		goto err_busy;
+	}
+	fiq_glue_update_return_handler(fiq_return);
+	ret = 0;
+err_busy:
+	mutex_unlock(&fiq_glue_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL(fiq_glue_set_return_handler);
+
+int fiq_glue_clear_return_handler(void (*fiq_return)(void))
+{
+	int ret;
+
+	mutex_lock(&fiq_glue_lock);
+	if (WARN_ON(fiq_return_handler != fiq_return)) {
+		ret = -EINVAL;
+		goto err_inval;
+	}
+	fiq_glue_update_return_handler(NULL);
+	ret = 0;
+err_inval:
+	mutex_unlock(&fiq_glue_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL(fiq_glue_clear_return_handler);
+
+/**
+ * fiq_glue_resume - Restore fiqs after suspend or low power idle states
+ *
+ * This must be called before calling local_fiq_enable after returning from a
+ * power state where the fiq mode registers were lost. If a driver provided
+ * a resume hook when it registered the handler it will be called.
+ */
+
+void fiq_glue_resume(void)
+{
+	if (!current_handler)
+		return;
+	fiq_glue_setup(current_handler->fiq, current_handler,
+		__get_cpu_var(fiq_stack) + THREAD_START_SP,
+		fiq_return_handler);
+	if (current_handler->resume)
+		current_handler->resume(current_handler);
+}
+
diff --git a/arch/arm/configs/ranchu_defconfig b/arch/arm/configs/ranchu_defconfig
new file mode 100644
index 000000000000..49e7bbd5825a
--- /dev/null
+++ b/arch/arm/configs/ranchu_defconfig
@@ -0,0 +1,316 @@
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_AUDIT=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_TASKSTATS=y
+CONFIG_TASK_DELAY_ACCT=y
+CONFIG_TASK_XACCT=y
+CONFIG_TASK_IO_ACCOUNTING=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_CGROUPS=y
+CONFIG_CGROUP_DEBUG=y
+CONFIG_CGROUP_FREEZER=y
+CONFIG_CPUSETS=y
+CONFIG_CGROUP_CPUACCT=y
+CONFIG_CGROUP_SCHED=y
+CONFIG_RT_GROUP_SCHED=y
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_KALLSYMS_ALL=y
+CONFIG_EMBEDDED=y
+CONFIG_PROFILING=y
+CONFIG_OPROFILE=y
+CONFIG_ARCH_MMAP_RND_BITS=16
+# CONFIG_BLK_DEV_BSG is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+# CONFIG_IOSCHED_CFQ is not set
+CONFIG_ARCH_VIRT=y
+CONFIG_ARM_KERNMEM_PERMS=y
+CONFIG_SMP=y
+CONFIG_PREEMPT=y
+CONFIG_AEABI=y
+CONFIG_HIGHMEM=y
+CONFIG_KSM=y
+CONFIG_SECCOMP=y
+CONFIG_CMDLINE="console=ttyAMA0"
+CONFIG_VFP=y
+CONFIG_NEON=y
+# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
+CONFIG_PM_AUTOSLEEP=y
+CONFIG_PM_WAKELOCKS=y
+CONFIG_PM_WAKELOCKS_LIMIT=0
+# CONFIG_PM_WAKELOCKS_GC is not set
+CONFIG_PM_DEBUG=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_XFRM_USER=y
+CONFIG_NET_KEY=y
+CONFIG_INET=y
+CONFIG_INET_DIAG_DESTROY=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_INET_ESP=y
+# CONFIG_INET_LRO is not set
+CONFIG_IPV6_ROUTER_PREF=y
+CONFIG_IPV6_ROUTE_INFO=y
+CONFIG_IPV6_OPTIMISTIC_DAD=y
+CONFIG_INET6_AH=y
+CONFIG_INET6_ESP=y
+CONFIG_INET6_IPCOMP=y
+CONFIG_IPV6_MIP6=y
+CONFIG_IPV6_MULTIPLE_TABLES=y
+CONFIG_NETFILTER=y
+CONFIG_NF_CONNTRACK=y
+CONFIG_NF_CONNTRACK_SECMARK=y
+CONFIG_NF_CONNTRACK_EVENTS=y
+CONFIG_NF_CT_PROTO_DCCP=y
+CONFIG_NF_CT_PROTO_SCTP=y
+CONFIG_NF_CT_PROTO_UDPLITE=y
+CONFIG_NF_CONNTRACK_AMANDA=y
+CONFIG_NF_CONNTRACK_FTP=y
+CONFIG_NF_CONNTRACK_H323=y
+CONFIG_NF_CONNTRACK_IRC=y
+CONFIG_NF_CONNTRACK_NETBIOS_NS=y
+CONFIG_NF_CONNTRACK_PPTP=y
+CONFIG_NF_CONNTRACK_SANE=y
+CONFIG_NF_CONNTRACK_TFTP=y
+CONFIG_NF_CT_NETLINK=y
+CONFIG_NETFILTER_XT_TARGET_CLASSIFY=y
+CONFIG_NETFILTER_XT_TARGET_CONNMARK=y
+CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=y
+CONFIG_NETFILTER_XT_TARGET_IDLETIMER=y
+CONFIG_NETFILTER_XT_TARGET_MARK=y
+CONFIG_NETFILTER_XT_TARGET_NFLOG=y
+CONFIG_NETFILTER_XT_TARGET_NFQUEUE=y
+CONFIG_NETFILTER_XT_TARGET_TPROXY=y
+CONFIG_NETFILTER_XT_TARGET_TRACE=y
+CONFIG_NETFILTER_XT_TARGET_SECMARK=y
+CONFIG_NETFILTER_XT_TARGET_TCPMSS=y
+CONFIG_NETFILTER_XT_MATCH_COMMENT=y
+CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_CONNMARK=y
+CONFIG_NETFILTER_XT_MATCH_CONNTRACK=y
+CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_HELPER=y
+CONFIG_NETFILTER_XT_MATCH_IPRANGE=y
+CONFIG_NETFILTER_XT_MATCH_LENGTH=y
+CONFIG_NETFILTER_XT_MATCH_LIMIT=y
+CONFIG_NETFILTER_XT_MATCH_MAC=y
+CONFIG_NETFILTER_XT_MATCH_MARK=y
+CONFIG_NETFILTER_XT_MATCH_POLICY=y
+CONFIG_NETFILTER_XT_MATCH_PKTTYPE=y
+CONFIG_NETFILTER_XT_MATCH_QTAGUID=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA2=y
+CONFIG_NETFILTER_XT_MATCH_SOCKET=y
+CONFIG_NETFILTER_XT_MATCH_STATE=y
+CONFIG_NETFILTER_XT_MATCH_STATISTIC=y
+CONFIG_NETFILTER_XT_MATCH_STRING=y
+CONFIG_NETFILTER_XT_MATCH_TIME=y
+CONFIG_NETFILTER_XT_MATCH_U32=y
+CONFIG_NF_CONNTRACK_IPV4=y
+CONFIG_IP_NF_IPTABLES=y
+CONFIG_IP_NF_MATCH_AH=y
+CONFIG_IP_NF_MATCH_ECN=y
+CONFIG_IP_NF_MATCH_TTL=y
+CONFIG_IP_NF_FILTER=y
+CONFIG_IP_NF_TARGET_REJECT=y
+CONFIG_IP_NF_MANGLE=y
+CONFIG_IP_NF_RAW=y
+CONFIG_IP_NF_SECURITY=y
+CONFIG_IP_NF_ARPTABLES=y
+CONFIG_IP_NF_ARPFILTER=y
+CONFIG_IP_NF_ARP_MANGLE=y
+CONFIG_NF_CONNTRACK_IPV6=y
+CONFIG_IP6_NF_IPTABLES=y
+CONFIG_IP6_NF_FILTER=y
+CONFIG_IP6_NF_TARGET_REJECT=y
+CONFIG_IP6_NF_MANGLE=y
+CONFIG_IP6_NF_RAW=y
+CONFIG_BRIDGE=y
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_HTB=y
+CONFIG_NET_CLS_U32=y
+CONFIG_NET_EMATCH=y
+CONFIG_NET_EMATCH_U32=y
+CONFIG_NET_CLS_ACT=y
+# CONFIG_WIRELESS is not set
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_MTD=y
+CONFIG_MTD_CMDLINE_PARTS=y
+CONFIG_MTD_BLOCK=y
+CONFIG_MTD_CFI=y
+CONFIG_MTD_CFI_INTELEXT=y
+CONFIG_MTD_CFI_AMDSTD=y
+CONFIG_BLK_DEV_LOOP=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_VIRTIO_BLK=y
+CONFIG_MD=y
+CONFIG_BLK_DEV_DM=y
+CONFIG_DM_CRYPT=y
+CONFIG_DM_UEVENT=y
+CONFIG_DM_VERITY=y
+CONFIG_DM_VERITY_FEC=y
+CONFIG_NETDEVICES=y
+CONFIG_TUN=y
+CONFIG_VIRTIO_NET=y
+CONFIG_SMSC911X=y
+CONFIG_PPP=y
+CONFIG_PPP_BSDCOMP=y
+CONFIG_PPP_DEFLATE=y
+CONFIG_PPP_MPPE=y
+CONFIG_PPPOLAC=y
+CONFIG_PPPOPNS=y
+CONFIG_USB_USBNET=y
+# CONFIG_WLAN is not set
+CONFIG_INPUT_EVDEV=y
+CONFIG_INPUT_KEYRESET=y
+CONFIG_KEYBOARD_GOLDFISH_EVENTS=y
+# CONFIG_INPUT_MOUSE is not set
+CONFIG_INPUT_JOYSTICK=y
+CONFIG_JOYSTICK_XPAD=y
+CONFIG_JOYSTICK_XPAD_FF=y
+CONFIG_JOYSTICK_XPAD_LEDS=y
+CONFIG_INPUT_TABLET=y
+CONFIG_TABLET_USB_ACECAD=y
+CONFIG_TABLET_USB_AIPTEK=y
+CONFIG_TABLET_USB_GTCO=y
+CONFIG_TABLET_USB_HANWANG=y
+CONFIG_TABLET_USB_KBTAB=y
+CONFIG_INPUT_MISC=y
+CONFIG_INPUT_KEYCHORD=y
+CONFIG_INPUT_UINPUT=y
+CONFIG_INPUT_GPIO=y
+# CONFIG_SERIO_SERPORT is not set
+CONFIG_SERIO_AMBAKMI=y
+# CONFIG_VT is not set
+# CONFIG_LEGACY_PTYS is not set
+# CONFIG_DEVMEM is not set
+# CONFIG_DEVKMEM is not set
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+CONFIG_VIRTIO_CONSOLE=y
+# CONFIG_HW_RANDOM is not set
+# CONFIG_HWMON is not set
+CONFIG_MEDIA_SUPPORT=y
+CONFIG_FB=y
+CONFIG_FB_GOLDFISH=y
+CONFIG_FB_SIMPLE=y
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LOGO=y
+# CONFIG_LOGO_LINUX_MONO is not set
+# CONFIG_LOGO_LINUX_VGA16 is not set
+CONFIG_SOUND=y
+CONFIG_SND=y
+CONFIG_HIDRAW=y
+CONFIG_UHID=y
+CONFIG_HID_A4TECH=y
+CONFIG_HID_ACRUX=y
+CONFIG_HID_ACRUX_FF=y
+CONFIG_HID_APPLE=y
+CONFIG_HID_BELKIN=y
+CONFIG_HID_CHERRY=y
+CONFIG_HID_CHICONY=y
+CONFIG_HID_PRODIKEYS=y
+CONFIG_HID_CYPRESS=y
+CONFIG_HID_DRAGONRISE=y
+CONFIG_DRAGONRISE_FF=y
+CONFIG_HID_EMS_FF=y
+CONFIG_HID_ELECOM=y
+CONFIG_HID_EZKEY=y
+CONFIG_HID_HOLTEK=y
+CONFIG_HID_KEYTOUCH=y
+CONFIG_HID_KYE=y
+CONFIG_HID_UCLOGIC=y
+CONFIG_HID_WALTOP=y
+CONFIG_HID_GYRATION=y
+CONFIG_HID_TWINHAN=y
+CONFIG_HID_KENSINGTON=y
+CONFIG_HID_LCPOWER=y
+CONFIG_HID_LOGITECH=y
+CONFIG_HID_LOGITECH_DJ=y
+CONFIG_LOGITECH_FF=y
+CONFIG_LOGIRUMBLEPAD2_FF=y
+CONFIG_LOGIG940_FF=y
+CONFIG_HID_MAGICMOUSE=y
+CONFIG_HID_MICROSOFT=y
+CONFIG_HID_MONTEREY=y
+CONFIG_HID_MULTITOUCH=y
+CONFIG_HID_NTRIG=y
+CONFIG_HID_ORTEK=y
+CONFIG_HID_PANTHERLORD=y
+CONFIG_PANTHERLORD_FF=y
+CONFIG_HID_PETALYNX=y
+CONFIG_HID_PICOLCD=y
+CONFIG_HID_PRIMAX=y
+CONFIG_HID_ROCCAT=y
+CONFIG_HID_SAITEK=y
+CONFIG_HID_SAMSUNG=y
+CONFIG_HID_SONY=y
+CONFIG_HID_SPEEDLINK=y
+CONFIG_HID_SUNPLUS=y
+CONFIG_HID_GREENASIA=y
+CONFIG_GREENASIA_FF=y
+CONFIG_HID_SMARTJOYPLUS=y
+CONFIG_SMARTJOYPLUS_FF=y
+CONFIG_HID_TIVO=y
+CONFIG_HID_TOPSEED=y
+CONFIG_HID_THRUSTMASTER=y
+CONFIG_HID_WACOM=y
+CONFIG_HID_WIIMOTE=y
+CONFIG_HID_ZEROPLUS=y
+CONFIG_HID_ZYDACRON=y
+CONFIG_USB_HIDDEV=y
+CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
+CONFIG_USB_EHCI_HCD=y
+CONFIG_USB_OTG_WAKELOCK=y
+CONFIG_RTC_CLASS=y
+CONFIG_RTC_DRV_PL031=y
+CONFIG_VIRTIO_MMIO=y
+CONFIG_STAGING=y
+CONFIG_ASHMEM=y
+CONFIG_ANDROID_LOW_MEMORY_KILLER=y
+CONFIG_SYNC=y
+CONFIG_SW_SYNC=y
+CONFIG_SW_SYNC_USER=y
+CONFIG_ION=y
+CONFIG_GOLDFISH_AUDIO=y
+CONFIG_GOLDFISH=y
+CONFIG_GOLDFISH_PIPE=y
+CONFIG_ANDROID=y
+CONFIG_ANDROID_BINDER_IPC=y
+CONFIG_EXT4_FS=y
+CONFIG_EXT4_FS_SECURITY=y
+CONFIG_QUOTA=y
+CONFIG_FUSE_FS=y
+CONFIG_CUSE=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_TMPFS=y
+CONFIG_TMPFS_POSIX_ACL=y
+CONFIG_PSTORE=y
+CONFIG_PSTORE_CONSOLE=y
+CONFIG_PSTORE_RAM=y
+CONFIG_NFS_FS=y
+CONFIG_ROOT_NFS=y
+CONFIG_NLS_CODEPAGE_437=y
+CONFIG_NLS_ISO8859_1=y
+CONFIG_DEBUG_INFO=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_DETECT_HUNG_TASK=y
+CONFIG_PANIC_TIMEOUT=5
+# CONFIG_SCHED_DEBUG is not set
+CONFIG_SCHEDSTATS=y
+CONFIG_TIMER_STATS=y
+CONFIG_ENABLE_DEFAULT_TRACERS=y
+CONFIG_SECURITY=y
+CONFIG_SECURITY_NETWORK=y
+CONFIG_SECURITY_SELINUX=y
+CONFIG_VIRTUALIZATION=y
diff --git a/arch/arm/crypto/aes-ce-glue.c b/arch/arm/crypto/aes-ce-glue.c
index 679c589c4828..1f7b98e1a00d 100644
--- a/arch/arm/crypto/aes-ce-glue.c
+++ b/arch/arm/crypto/aes-ce-glue.c
@@ -369,7 +369,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_blkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= ce_aes_setkey,
 		.encrypt	= ecb_encrypt,
 		.decrypt	= ecb_decrypt,
@@ -446,7 +446,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_ablkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= ablk_set_key,
 		.encrypt	= ablk_encrypt,
 		.decrypt	= ablk_decrypt,
diff --git a/arch/arm/include/asm/elf.h b/arch/arm/include/asm/elf.h
index f13ae153fb24..d2315ffd8f12 100644
--- a/arch/arm/include/asm/elf.h
+++ b/arch/arm/include/asm/elf.h
@@ -112,8 +112,12 @@ int dump_task_regs(struct task_struct *t, elf_gregset_t *elfregs);
 #define CORE_DUMP_USE_REGSET
 #define ELF_EXEC_PAGESIZE	4096
 
-/* This is the base location for PIE (ET_DYN with INTERP) loads. */
-#define ELF_ET_DYN_BASE		0x400000UL
+/* This is the location that an ET_DYN program is loaded if exec'ed.  Typical
+   use of this is to invoke "./ld.so someprog" to test out a new version of
+   the loader.  We need to make sure that it is out of the way of the program
+   that it will "exec", and that there is sufficient room for the brk.  */
+
+#define ELF_ET_DYN_BASE	(TASK_SIZE / 3 * 2)
 
 /* When the program starts, a1 contains a pointer to a function to be 
    registered with atexit, as per the SVR4 ABI.  A value of 0 means we 
diff --git a/arch/arm/include/asm/fiq_glue.h b/arch/arm/include/asm/fiq_glue.h
new file mode 100644
index 000000000000..a9e244f9f197
--- /dev/null
+++ b/arch/arm/include/asm/fiq_glue.h
@@ -0,0 +1,33 @@
+/*
+ * Copyright (C) 2010 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __ASM_FIQ_GLUE_H
+#define __ASM_FIQ_GLUE_H
+
+struct fiq_glue_handler {
+	void (*fiq)(struct fiq_glue_handler *h, void *regs, void *svc_sp);
+	void (*resume)(struct fiq_glue_handler *h);
+};
+typedef void (*fiq_return_handler_t)(void);
+
+int fiq_glue_register_handler(struct fiq_glue_handler *handler);
+int fiq_glue_set_return_handler(fiq_return_handler_t fiq_return);
+int fiq_glue_clear_return_handler(fiq_return_handler_t fiq_return);
+
+#ifdef CONFIG_FIQ_GLUE
+void fiq_glue_resume(void);
+#else
+static inline void fiq_glue_resume(void) {}
+#endif
+
+#endif
diff --git a/arch/arm/include/asm/pgtable-3level.h b/arch/arm/include/asm/pgtable-3level.h
index fd929b5ded9e..9459ca85bd20 100644
--- a/arch/arm/include/asm/pgtable-3level.h
+++ b/arch/arm/include/asm/pgtable-3level.h
@@ -250,6 +250,7 @@ PMD_BIT_FUNC(mkold,	&= ~PMD_SECT_AF);
 PMD_BIT_FUNC(mksplitting, |= L_PMD_SECT_SPLITTING);
 PMD_BIT_FUNC(mkwrite,   &= ~L_PMD_SECT_RDONLY);
 PMD_BIT_FUNC(mkdirty,   |= L_PMD_SECT_DIRTY);
+PMD_BIT_FUNC(mkclean,   &= ~L_PMD_SECT_DIRTY);
 PMD_BIT_FUNC(mkyoung,   |= PMD_SECT_AF);
 
 #define pmd_mkhuge(pmd)		(__pmd(pmd_val(pmd) & ~PMD_TABLE_BIT))
diff --git a/arch/arm/include/asm/topology.h b/arch/arm/include/asm/topology.h
index 370f7a732900..d06064120694 100644
--- a/arch/arm/include/asm/topology.h
+++ b/arch/arm/include/asm/topology.h
@@ -3,6 +3,7 @@
 
 #ifdef CONFIG_ARM_CPU_TOPOLOGY
 
+#include <linux/cpufreq.h>
 #include <linux/cpumask.h>
 
 struct cputopo_arm {
@@ -24,6 +25,12 @@ void init_cpu_topology(void);
 void store_cpu_topology(unsigned int cpuid);
 const struct cpumask *cpu_coregroup_mask(int cpu);
 
+#ifdef CONFIG_CPU_FREQ
+#define arch_scale_freq_capacity cpufreq_scale_freq_capacity
+#endif
+#define arch_scale_cpu_capacity scale_cpu_capacity
+extern unsigned long scale_cpu_capacity(struct sched_domain *sd, int cpu);
+
 #else
 
 static inline void init_cpu_topology(void) { }
diff --git a/arch/arm/kernel/asm-offsets.c b/arch/arm/kernel/asm-offsets.c
index 871b8267d211..a586cfe7b4e4 100644
--- a/arch/arm/kernel/asm-offsets.c
+++ b/arch/arm/kernel/asm-offsets.c
@@ -10,7 +10,6 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
-#include <linux/compiler.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/dma-mapping.h>
@@ -41,19 +40,10 @@
  * GCC 3.2.x: miscompiles NEW_AUX_ENT in fs/binfmt_elf.c
  *            (http://gcc.gnu.org/PR8896) and incorrect structure
  *	      initialisation in fs/jffs2/erase.c
- * GCC 4.8.0-4.8.2: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58854
- *	      miscompiles find_get_entry(), and can result in EXT3 and EXT4
- *	      filesystem corruption (possibly other FS too).
  */
-#ifdef __GNUC__
 #if (__GNUC__ == 3 && __GNUC_MINOR__ < 3)
 #error Your compiler is too buggy; it is known to miscompile kernels.
-#error    Known good compilers: 3.3, 4.x
-#endif
-#if GCC_VERSION >= 40800 && GCC_VERSION < 40803
-#error Your compiler is too buggy; it is known to miscompile kernels
-#error and result in filesystem corruption and oopses.
-#endif
+#error    Known good compilers: 3.3
 #endif
 
 int main(void)
diff --git a/arch/arm/kernel/kgdb.c b/arch/arm/kernel/kgdb.c
index 9232caee7060..f3c662299531 100644
--- a/arch/arm/kernel/kgdb.c
+++ b/arch/arm/kernel/kgdb.c
@@ -140,6 +140,8 @@ int kgdb_arch_handle_exception(int exception_vector, int signo,
 
 static int kgdb_brk_fn(struct pt_regs *regs, unsigned int instr)
 {
+	if (user_mode(regs))
+		return -1;
 	kgdb_handle_exception(1, SIGTRAP, 0, regs);
 
 	return 0;
@@ -147,6 +149,8 @@ static int kgdb_brk_fn(struct pt_regs *regs, unsigned int instr)
 
 static int kgdb_compiled_brk_fn(struct pt_regs *regs, unsigned int instr)
 {
+	if (user_mode(regs))
+		return -1;
 	compiled_break = 1;
 	kgdb_handle_exception(1, SIGTRAP, 0, regs);
 
diff --git a/arch/arm/kernel/module-plts.c b/arch/arm/kernel/module-plts.c
index 097e2e201b9f..0c7efc3446c0 100644
--- a/arch/arm/kernel/module-plts.c
+++ b/arch/arm/kernel/module-plts.c
@@ -32,7 +32,7 @@ struct plt_entries {
 
 static bool in_init(const struct module *mod, u32 addr)
 {
-	return addr - (u32)mod->module_init < mod->init_size;
+	return addr - (u32)mod->init_layout.base < mod->init_layout.size;
 }
 
 u32 get_module_plt(struct module *mod, unsigned long loc, Elf32_Addr val)
diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
index 4adfb46e3ee9..0017b3edfea8 100644
--- a/arch/arm/kernel/process.c
+++ b/arch/arm/kernel/process.c
@@ -80,6 +80,7 @@ void arch_cpu_idle_prepare(void)
 
 void arch_cpu_idle_enter(void)
 {
+	idle_notifier_call_chain(IDLE_START);
 	ledtrig_cpu(CPU_LED_IDLE_START);
 #ifdef CONFIG_PL310_ERRATA_769419
 	wmb();
@@ -89,6 +90,78 @@ void arch_cpu_idle_enter(void)
 void arch_cpu_idle_exit(void)
 {
 	ledtrig_cpu(CPU_LED_IDLE_END);
+	idle_notifier_call_chain(IDLE_END);
+}
+
+/*
+ * dump a block of kernel memory from around the given address
+ */
+static void show_data(unsigned long addr, int nbytes, const char *name)
+{
+	int	i, j;
+	int	nlines;
+	u32	*p;
+
+	/*
+	 * don't attempt to dump non-kernel addresses or
+	 * values that are probably just small negative numbers
+	 */
+	if (addr < PAGE_OFFSET || addr > -256UL)
+		return;
+
+	printk("\n%s: %#lx:\n", name, addr);
+
+	/*
+	 * round address down to a 32 bit boundary
+	 * and always dump a multiple of 32 bytes
+	 */
+	p = (u32 *)(addr & ~(sizeof(u32) - 1));
+	nbytes += (addr & (sizeof(u32) - 1));
+	nlines = (nbytes + 31) / 32;
+
+
+	for (i = 0; i < nlines; i++) {
+		/*
+		 * just display low 16 bits of address to keep
+		 * each line of the dump < 80 characters
+		 */
+		printk("%04lx ", (unsigned long)p & 0xffff);
+		for (j = 0; j < 8; j++) {
+			u32	data;
+			if (probe_kernel_address(p, data)) {
+				printk(" ********");
+			} else {
+				printk(" %08x", data);
+			}
+			++p;
+		}
+		printk("\n");
+	}
+}
+
+static void show_extra_register_data(struct pt_regs *regs, int nbytes)
+{
+	mm_segment_t fs;
+
+	fs = get_fs();
+	set_fs(KERNEL_DS);
+	show_data(regs->ARM_pc - nbytes, nbytes * 2, "PC");
+	show_data(regs->ARM_lr - nbytes, nbytes * 2, "LR");
+	show_data(regs->ARM_sp - nbytes, nbytes * 2, "SP");
+	show_data(regs->ARM_ip - nbytes, nbytes * 2, "IP");
+	show_data(regs->ARM_fp - nbytes, nbytes * 2, "FP");
+	show_data(regs->ARM_r0 - nbytes, nbytes * 2, "R0");
+	show_data(regs->ARM_r1 - nbytes, nbytes * 2, "R1");
+	show_data(regs->ARM_r2 - nbytes, nbytes * 2, "R2");
+	show_data(regs->ARM_r3 - nbytes, nbytes * 2, "R3");
+	show_data(regs->ARM_r4 - nbytes, nbytes * 2, "R4");
+	show_data(regs->ARM_r5 - nbytes, nbytes * 2, "R5");
+	show_data(regs->ARM_r6 - nbytes, nbytes * 2, "R6");
+	show_data(regs->ARM_r7 - nbytes, nbytes * 2, "R7");
+	show_data(regs->ARM_r8 - nbytes, nbytes * 2, "R8");
+	show_data(regs->ARM_r9 - nbytes, nbytes * 2, "R9");
+	show_data(regs->ARM_r10 - nbytes, nbytes * 2, "R10");
+	set_fs(fs);
 }
 
 void __show_regs(struct pt_regs *regs)
@@ -178,6 +251,8 @@ void __show_regs(struct pt_regs *regs)
 		printk("Control: %08x%s\n", ctrl, buf);
 	}
 #endif
+
+	show_extra_register_data(regs, 128);
 }
 
 void show_regs(struct pt_regs * regs)
diff --git a/arch/arm/kernel/reboot.c b/arch/arm/kernel/reboot.c
index 38269358fd25..1a06da8f0366 100644
--- a/arch/arm/kernel/reboot.c
+++ b/arch/arm/kernel/reboot.c
@@ -6,6 +6,7 @@
  * it under the terms of the GNU General Public License version 2 as
  * published by the Free Software Foundation.
  */
+#include <linux/console.h>
 #include <linux/cpu.h>
 #include <linux/delay.h>
 #include <linux/reboot.h>
@@ -124,6 +125,31 @@ void machine_power_off(void)
 		pm_power_off();
 }
 
+#ifdef CONFIG_ARM_FLUSH_CONSOLE_ON_RESTART
+void arm_machine_flush_console(void)
+{
+	printk("\n");
+	pr_emerg("Restarting %s\n", linux_banner);
+	if (console_trylock()) {
+		console_unlock();
+		return;
+	}
+
+	mdelay(50);
+
+	local_irq_disable();
+	if (!console_trylock())
+		pr_emerg("arm_restart: Console was locked! Busting\n");
+	else
+		pr_emerg("arm_restart: Console was locked!\n");
+	console_unlock();
+}
+#else
+void arm_machine_flush_console(void)
+{
+}
+#endif
+
 /*
  * Restart requires that the secondary CPUs stop performing any activity
  * while the primary CPU resets the system. Systems with a single CPU can
@@ -140,6 +166,10 @@ void machine_restart(char *cmd)
 	local_irq_disable();
 	smp_send_stop();
 
+	/* Flush the console to make sure all the relevant messages make it
+	 * out to the console drivers */
+	arm_machine_flush_console();
+
 	if (arm_pm_restart)
 		arm_pm_restart(reboot_mode, cmd);
 	else
diff --git a/arch/arm/kernel/setup.c b/arch/arm/kernel/setup.c
index 20edd349d379..bf63b4693457 100644
--- a/arch/arm/kernel/setup.c
+++ b/arch/arm/kernel/setup.c
@@ -772,7 +772,7 @@ static void __init request_standard_resources(const struct machine_desc *mdesc)
 	struct resource *res;
 
 	kernel_code.start   = virt_to_phys(_text);
-	kernel_code.end     = virt_to_phys(_etext - 1);
+	kernel_code.end     = virt_to_phys(__init_begin - 1);
 	kernel_data.start   = virt_to_phys(_sdata);
 	kernel_data.end     = virt_to_phys(_end - 1);
 
diff --git a/arch/arm/kernel/topology.c b/arch/arm/kernel/topology.c
index 08b7847bf912..4f2c51ef162d 100644
--- a/arch/arm/kernel/topology.c
+++ b/arch/arm/kernel/topology.c
@@ -42,9 +42,15 @@
  */
 static DEFINE_PER_CPU(unsigned long, cpu_scale);
 
-unsigned long arch_scale_cpu_capacity(struct sched_domain *sd, int cpu)
+unsigned long scale_cpu_capacity(struct sched_domain *sd, int cpu)
 {
+#ifdef CONFIG_CPU_FREQ
+	unsigned long max_freq_scale = cpufreq_scale_max_freq_capacity(cpu);
+
+	return per_cpu(cpu_scale, cpu) * max_freq_scale >> SCHED_CAPACITY_SHIFT;
+#else
 	return per_cpu(cpu_scale, cpu);
+#endif
 }
 
 static void set_capacity_scale(unsigned int cpu, unsigned long capacity)
@@ -153,6 +159,8 @@ static void __init parse_dt_topology(void)
 
 }
 
+static const struct sched_group_energy * const cpu_core_energy(int cpu);
+
 /*
  * Look for a customed capacity of a CPU in the cpu_capacity table during the
  * boot. The update of all CPUs is in O(n^2) for heteregeneous system but the
@@ -160,10 +168,14 @@ static void __init parse_dt_topology(void)
  */
 static void update_cpu_capacity(unsigned int cpu)
 {
-	if (!cpu_capacity(cpu))
-		return;
+	unsigned long capacity = SCHED_CAPACITY_SCALE;
+
+	if (cpu_core_energy(cpu)) {
+		int max_cap_idx = cpu_core_energy(cpu)->nr_cap_states - 1;
+		capacity = cpu_core_energy(cpu)->cap_states[max_cap_idx].cap;
+	}
 
-	set_capacity_scale(cpu, cpu_capacity(cpu) / middle_capacity);
+	set_capacity_scale(cpu, capacity);
 
 	pr_info("CPU%u: update cpu_capacity %lu\n",
 		cpu, arch_scale_cpu_capacity(NULL, cpu));
@@ -275,17 +287,138 @@ void store_cpu_topology(unsigned int cpuid)
 		cpu_topology[cpuid].socket_id, mpidr);
 }
 
+/*
+ * ARM TC2 specific energy cost model data. There are no unit requirements for
+ * the data. Data can be normalized to any reference point, but the
+ * normalization must be consistent. That is, one bogo-joule/watt must be the
+ * same quantity for all data, but we don't care what it is.
+ */
+static struct idle_state idle_states_cluster_a7[] = {
+	 { .power = 25 }, /* arch_cpu_idle() (active idle) = WFI */
+	 { .power = 25 }, /* WFI */
+	 { .power = 10 }, /* cluster-sleep-l */
+	};
+
+static struct idle_state idle_states_cluster_a15[] = {
+	 { .power = 70 }, /* arch_cpu_idle() (active idle) = WFI */
+	 { .power = 70 }, /* WFI */
+	 { .power = 25 }, /* cluster-sleep-b */
+	};
+
+static struct capacity_state cap_states_cluster_a7[] = {
+	/* Cluster only power */
+	 { .cap =  150, .power = 2967, }, /*  350 MHz */
+	 { .cap =  172, .power = 2792, }, /*  400 MHz */
+	 { .cap =  215, .power = 2810, }, /*  500 MHz */
+	 { .cap =  258, .power = 2815, }, /*  600 MHz */
+	 { .cap =  301, .power = 2919, }, /*  700 MHz */
+	 { .cap =  344, .power = 2847, }, /*  800 MHz */
+	 { .cap =  387, .power = 3917, }, /*  900 MHz */
+	 { .cap =  430, .power = 4905, }, /* 1000 MHz */
+	};
+
+static struct capacity_state cap_states_cluster_a15[] = {
+	/* Cluster only power */
+	 { .cap =  426, .power =  7920, }, /*  500 MHz */
+	 { .cap =  512, .power =  8165, }, /*  600 MHz */
+	 { .cap =  597, .power =  8172, }, /*  700 MHz */
+	 { .cap =  682, .power =  8195, }, /*  800 MHz */
+	 { .cap =  768, .power =  8265, }, /*  900 MHz */
+	 { .cap =  853, .power =  8446, }, /* 1000 MHz */
+	 { .cap =  938, .power = 11426, }, /* 1100 MHz */
+	 { .cap = 1024, .power = 15200, }, /* 1200 MHz */
+	};
+
+static struct sched_group_energy energy_cluster_a7 = {
+	  .nr_idle_states = ARRAY_SIZE(idle_states_cluster_a7),
+	  .idle_states    = idle_states_cluster_a7,
+	  .nr_cap_states  = ARRAY_SIZE(cap_states_cluster_a7),
+	  .cap_states     = cap_states_cluster_a7,
+};
+
+static struct sched_group_energy energy_cluster_a15 = {
+	  .nr_idle_states = ARRAY_SIZE(idle_states_cluster_a15),
+	  .idle_states    = idle_states_cluster_a15,
+	  .nr_cap_states  = ARRAY_SIZE(cap_states_cluster_a15),
+	  .cap_states     = cap_states_cluster_a15,
+};
+
+static struct idle_state idle_states_core_a7[] = {
+	 { .power = 0 }, /* arch_cpu_idle (active idle) = WFI */
+	 { .power = 0 }, /* WFI */
+	 { .power = 0 }, /* cluster-sleep-l */
+	};
+
+static struct idle_state idle_states_core_a15[] = {
+	 { .power = 0 }, /* arch_cpu_idle (active idle) = WFI */
+	 { .power = 0 }, /* WFI */
+	 { .power = 0 }, /* cluster-sleep-b */
+	};
+
+static struct capacity_state cap_states_core_a7[] = {
+	/* Power per cpu */
+	 { .cap =  150, .power =  187, }, /*  350 MHz */
+	 { .cap =  172, .power =  275, }, /*  400 MHz */
+	 { .cap =  215, .power =  334, }, /*  500 MHz */
+	 { .cap =  258, .power =  407, }, /*  600 MHz */
+	 { .cap =  301, .power =  447, }, /*  700 MHz */
+	 { .cap =  344, .power =  549, }, /*  800 MHz */
+	 { .cap =  387, .power =  761, }, /*  900 MHz */
+	 { .cap =  430, .power = 1024, }, /* 1000 MHz */
+	};
+
+static struct capacity_state cap_states_core_a15[] = {
+	/* Power per cpu */
+	 { .cap =  426, .power = 2021, }, /*  500 MHz */
+	 { .cap =  512, .power = 2312, }, /*  600 MHz */
+	 { .cap =  597, .power = 2756, }, /*  700 MHz */
+	 { .cap =  682, .power = 3125, }, /*  800 MHz */
+	 { .cap =  768, .power = 3524, }, /*  900 MHz */
+	 { .cap =  853, .power = 3846, }, /* 1000 MHz */
+	 { .cap =  938, .power = 5177, }, /* 1100 MHz */
+	 { .cap = 1024, .power = 6997, }, /* 1200 MHz */
+	};
+
+static struct sched_group_energy energy_core_a7 = {
+	  .nr_idle_states = ARRAY_SIZE(idle_states_core_a7),
+	  .idle_states    = idle_states_core_a7,
+	  .nr_cap_states  = ARRAY_SIZE(cap_states_core_a7),
+	  .cap_states     = cap_states_core_a7,
+};
+
+static struct sched_group_energy energy_core_a15 = {
+	  .nr_idle_states = ARRAY_SIZE(idle_states_core_a15),
+	  .idle_states    = idle_states_core_a15,
+	  .nr_cap_states  = ARRAY_SIZE(cap_states_core_a15),
+	  .cap_states     = cap_states_core_a15,
+};
+
+/* sd energy functions */
+static inline
+const struct sched_group_energy * const cpu_cluster_energy(int cpu)
+{
+	return cpu_topology[cpu].socket_id ? &energy_cluster_a7 :
+			&energy_cluster_a15;
+}
+
+static inline
+const struct sched_group_energy * const cpu_core_energy(int cpu)
+{
+	return cpu_topology[cpu].socket_id ? &energy_core_a7 :
+			&energy_core_a15;
+}
+
 static inline int cpu_corepower_flags(void)
 {
-	return SD_SHARE_PKG_RESOURCES  | SD_SHARE_POWERDOMAIN;
+	return SD_SHARE_PKG_RESOURCES  | SD_SHARE_POWERDOMAIN | \
+	       SD_SHARE_CAP_STATES;
 }
 
 static struct sched_domain_topology_level arm_topology[] = {
 #ifdef CONFIG_SCHED_MC
-	{ cpu_corepower_mask, cpu_corepower_flags, SD_INIT_NAME(GMC) },
-	{ cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) },
+	{ cpu_coregroup_mask, cpu_corepower_flags, cpu_core_energy, SD_INIT_NAME(MC) },
 #endif
-	{ cpu_cpu_mask, SD_INIT_NAME(DIE) },
+	{ cpu_cpu_mask, NULL, cpu_cluster_energy, SD_INIT_NAME(DIE) },
 	{ NULL, },
 };
 
diff --git a/arch/arm/kernel/vdso.c b/arch/arm/kernel/vdso.c
index 54a5aeab988d..bbbffe946122 100644
--- a/arch/arm/kernel/vdso.c
+++ b/arch/arm/kernel/vdso.c
@@ -17,6 +17,7 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
+#include <linux/cache.h>
 #include <linux/elf.h>
 #include <linux/err.h>
 #include <linux/kernel.h>
@@ -39,7 +40,7 @@
 static struct page **vdso_text_pagelist;
 
 /* Total number of pages needed for the data and text portions of the VDSO. */
-unsigned int vdso_total_pages __read_mostly;
+unsigned int vdso_total_pages __ro_after_init;
 
 /*
  * The VDSO data page.
@@ -47,13 +48,13 @@ unsigned int vdso_total_pages __read_mostly;
 static union vdso_data_store vdso_data_store __page_aligned_data;
 static struct vdso_data *vdso_data = &vdso_data_store.data;
 
-static struct page *vdso_data_page;
-static struct vm_special_mapping vdso_data_mapping = {
+static struct page *vdso_data_page __ro_after_init;
+static const struct vm_special_mapping vdso_data_mapping = {
 	.name = "[vvar]",
 	.pages = &vdso_data_page,
 };
 
-static struct vm_special_mapping vdso_text_mapping = {
+static struct vm_special_mapping vdso_text_mapping __ro_after_init = {
 	.name = "[vdso]",
 };
 
@@ -67,7 +68,7 @@ struct elfinfo {
 /* Cached result of boot-time check for whether the arch timer exists,
  * and if so, whether the virtual counter is useable.
  */
-static bool cntvct_ok __read_mostly;
+static bool cntvct_ok __ro_after_init;
 
 static bool __init cntvct_functional(void)
 {
@@ -224,7 +225,7 @@ static int install_vvar(struct mm_struct *mm, unsigned long addr)
 				       VM_READ | VM_MAYREAD,
 				       &vdso_data_mapping);
 
-	return IS_ERR(vma) ? PTR_ERR(vma) : 0;
+	return PTR_ERR_OR_ZERO(vma);
 }
 
 /* assumes mmap_sem is write-locked */
diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
index 8b60fde5ce48..be2ab6d3b91f 100644
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -120,6 +120,8 @@ SECTIONS
 #ifdef CONFIG_DEBUG_RODATA
 	. = ALIGN(1<<SECTION_SHIFT);
 #endif
+	_etext = .;			/* End of text section */
+
 	RO_DATA(PAGE_SIZE)
 
 	. = ALIGN(4);
@@ -150,8 +152,6 @@ SECTIONS
 
 	NOTES
 
-	_etext = .;			/* End of text and rodata section */
-
 #ifndef CONFIG_XIP_KERNEL
 # ifdef CONFIG_ARM_KERNMEM_PERMS
 	. = ALIGN(1<<SECTION_SHIFT);
diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S
index 24659952c278..11da0f50a1fe 100644
--- a/arch/arm/mm/cache-v6.S
+++ b/arch/arm/mm/cache-v6.S
@@ -270,6 +270,11 @@ v6_dma_clean_range:
  *	- end     - virtual end address of region
  */
 ENTRY(v6_dma_flush_range)
+#ifdef CONFIG_CACHE_FLUSH_RANGE_LIMIT
+	sub	r2, r1, r0
+	cmp	r2, #CONFIG_CACHE_FLUSH_RANGE_LIMIT
+	bhi	v6_dma_flush_dcache_all
+#endif
 #ifdef CONFIG_DMA_CACHE_RWFO
 	ldrb	r2, [r0]		@ read for ownership
 	strb	r2, [r0]		@ write for ownership
@@ -292,6 +297,18 @@ ENTRY(v6_dma_flush_range)
 	mcr	p15, 0, r0, c7, c10, 4		@ drain write buffer
 	ret	lr
 
+#ifdef CONFIG_CACHE_FLUSH_RANGE_LIMIT
+v6_dma_flush_dcache_all:
+	mov	r0, #0
+#ifdef HARVARD_CACHE
+	mcr	p15, 0, r0, c7, c14, 0		@ D cache clean+invalidate
+#else
+	mcr	p15, 0, r0, c7, c15, 0		@ Cache clean+invalidate
+#endif
+	mcr	p15, 0, r0, c7, c10, 4		@ drain write buffer
+	mov	pc, lr
+#endif
+
 /*
  *	dma_map_area(start, size, dir)
  *	- start	- kernel virtual start address
diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
index 0d20cd594017..83519afe3254 100644
--- a/arch/arm/mm/fault.c
+++ b/arch/arm/mm/fault.c
@@ -273,10 +273,10 @@ do_page_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
 		local_irq_enable();
 
 	/*
-	 * If we're in an interrupt or have no user
+	 * If we're in an interrupt, or have no irqs, or have no user
 	 * context, we must not take the fault..
 	 */
-	if (faulthandler_disabled() || !mm)
+	if (faulthandler_disabled() || irqs_disabled() || !mm)
 		goto no_context;
 
 	if (user_mode(regs))
diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c
index c469c0665752..641334ebf46d 100644
--- a/arch/arm/mm/mmap.c
+++ b/arch/arm/mm/mmap.c
@@ -173,8 +173,7 @@ unsigned long arch_mmap_rnd(void)
 {
 	unsigned long rnd;
 
-	/* 8 bits of randomness in 20 address space bits */
-	rnd = (unsigned long)get_random_int() % (1 << 8);
+	rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c
index e47cffd25c6c..aead23f15213 100644
--- a/arch/arm/mm/mmu.c
+++ b/arch/arm/mm/mmu.c
@@ -572,7 +572,7 @@ static void __init build_mem_type_table(void)
 	 * in the Short-descriptor translation table format descriptors.
 	 */
 	if (cpu_arch == CPU_ARCH_ARMv7 &&
-		(read_cpuid_ext(CPUID_EXT_MMFR0) & 0xF) == 4) {
+		(read_cpuid_ext(CPUID_EXT_MMFR0) & 0xF) >= 4) {
 		user_pmd_table |= PMD_PXNTABLE;
 	}
 #endif
diff --git a/arch/arm/vdso/Makefile b/arch/arm/vdso/Makefile
index 1160434eece0..59a8fa7b8a3b 100644
--- a/arch/arm/vdso/Makefile
+++ b/arch/arm/vdso/Makefile
@@ -74,5 +74,5 @@ $(MODLIB)/vdso: FORCE
 	@mkdir -p $(MODLIB)/vdso
 
 PHONY += vdso_install
-vdso_install: $(obj)/vdso.so.dbg $(MODLIB)/vdso FORCE
+vdso_install: $(obj)/vdso.so.dbg $(MODLIB)/vdso
 	$(call cmd,vdso_install)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 703b7f921806..190471794853 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -54,6 +54,8 @@ config ARM64
 	select HAVE_ARCH_JUMP_LABEL
 	select HAVE_ARCH_KASAN if SPARSEMEM_VMEMMAP && !(ARM64_16K_PAGES && ARM64_VA_BITS_48)
 	select HAVE_ARCH_KGDB
+	select HAVE_ARCH_MMAP_RND_BITS
+	select HAVE_ARCH_MMAP_RND_COMPAT_BITS if COMPAT
 	select HAVE_ARCH_SECCOMP_FILTER
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_BPF_JIT
@@ -100,6 +102,7 @@ config ARM64
 	select SYSCTL_EXCEPTION_TRACE
 	select HAVE_CONTEXT_TRACKING
 	select HAVE_ARM_SMCCC
+	select THREAD_INFO_IN_TASK
 	help
 	  ARM 64-bit (AArch64) Linux support.
 
@@ -112,6 +115,33 @@ config ARCH_PHYS_ADDR_T_64BIT
 config MMU
 	def_bool y
 
+config ARCH_MMAP_RND_BITS_MIN
+       default 14 if ARM64_64K_PAGES
+       default 16 if ARM64_16K_PAGES
+       default 18
+
+# max bits determined by the following formula:
+#  VA_BITS - PAGE_SHIFT - 3
+config ARCH_MMAP_RND_BITS_MAX
+       default 19 if ARM64_VA_BITS=36
+       default 24 if ARM64_VA_BITS=39
+       default 27 if ARM64_VA_BITS=42
+       default 30 if ARM64_VA_BITS=47
+       default 29 if ARM64_VA_BITS=48 && ARM64_64K_PAGES
+       default 31 if ARM64_VA_BITS=48 && ARM64_16K_PAGES
+       default 33 if ARM64_VA_BITS=48
+       default 14 if ARM64_64K_PAGES
+       default 16 if ARM64_16K_PAGES
+       default 18
+
+config ARCH_MMAP_RND_COMPAT_BITS_MIN
+       default 7 if ARM64_64K_PAGES
+       default 9 if ARM64_16K_PAGES
+       default 11
+
+config ARCH_MMAP_RND_COMPAT_BITS_MAX
+       default 16
+
 config ARM64_PAGE_SHIFT
 	int
 	default 16 if ARM64_64K_PAGES
@@ -127,6 +157,10 @@ config ARM64_CONT_SHIFT
 config NO_IOPORT_MAP
 	def_bool y if !PCI
 
+config ILLEGAL_POINTER_VALUE
+	hex
+	default 0xdead000000000000
+
 config STACKTRACE_SUPPORT
 	def_bool y
 
@@ -720,6 +754,14 @@ config SETEND_EMULATION
 	  If unsure, say Y
 endif
 
+config ARM64_SW_TTBR0_PAN
+	bool "Emulate Privileged Access Never using TTBR0_EL1 switching"
+	help
+	  Enabling this option prevents the kernel from accessing
+	  user-space memory directly by pointing TTBR0_EL1 to a reserved
+	  zeroed area and reserved ASID. The user access routines
+	  restore the valid TTBR0_EL1 temporarily.
+
 menu "ARMv8.1 architectural features"
 
 config ARM64_HW_AFDBM
@@ -828,7 +870,7 @@ config RANDOMIZE_BASE
 
 config RANDOMIZE_MODULE_REGION_FULL
 	bool "Randomize the module region independently from the core kernel"
-	depends on RANDOMIZE_BASE
+	depends on RANDOMIZE_BASE && !DYNAMIC_FTRACE
 	default y
 	help
 	  Randomizes the location of the module region without considering the
@@ -862,6 +904,23 @@ config CMDLINE
 	  entering them here. As a minimum, you should specify the the
 	  root device (e.g. root=/dev/nfs).
 
+choice
+	prompt "Kernel command line type" if CMDLINE != ""
+	default CMDLINE_FROM_BOOTLOADER
+
+config CMDLINE_FROM_BOOTLOADER
+	bool "Use bootloader kernel arguments if available"
+	help
+	  Uses the command-line options passed by the boot loader. If
+	  the boot loader doesn't provide any, the default kernel command
+	  string provided in CMDLINE will be used.
+
+config CMDLINE_EXTEND
+	bool "Extend bootloader kernel arguments"
+	help
+	  The command-line arguments provided by the boot loader will be
+	  appended to the default kernel command string.
+
 config CMDLINE_FORCE
 	bool "Always use the default kernel command string"
 	help
@@ -869,6 +928,7 @@ config CMDLINE_FORCE
 	  loader passes other arguments to the kernel.
 	  This is useful if you cannot or don't want to change the
 	  command-line options your boot loader passes to the kernel.
+endchoice
 
 config EFI_STUB
 	bool
@@ -901,6 +961,41 @@ config DMI
 	  However, even with this option, the resultant kernel should
 	  continue to boot on existing non-UEFI platforms.
 
+config BUILD_ARM64_APPENDED_DTB_IMAGE
+	bool "Build a concatenated Image.gz/dtb by default"
+	depends on OF
+	help
+	  Enabling this option will cause a concatenated Image.gz and list of
+	  DTBs to be built by default (instead of a standalone Image.gz.)
+	  The image will built in arch/arm64/boot/Image.gz-dtb
+
+choice
+	prompt "Appended DTB Kernel Image name"
+	depends on BUILD_ARM64_APPENDED_DTB_IMAGE
+	help
+	  Enabling this option will cause a specific kernel image Image or
+	  Image.gz to be used for final image creation.
+	  The image will built in arch/arm64/boot/IMAGE-NAME-dtb
+
+	config IMG_GZ_DTB
+		bool "Image.gz-dtb"
+	config IMG_DTB
+		bool "Image-dtb"
+endchoice
+
+config BUILD_ARM64_APPENDED_KERNEL_IMAGE_NAME
+	string
+	depends on BUILD_ARM64_APPENDED_DTB_IMAGE
+	default "Image.gz-dtb" if IMG_GZ_DTB
+	default "Image-dtb" if IMG_DTB
+
+config BUILD_ARM64_APPENDED_DTB_IMAGE_NAMES
+	string "Default dtb names"
+	depends on BUILD_ARM64_APPENDED_DTB_IMAGE
+	help
+	  Space separated list of names of dtbs to append when
+	  building a concatenated Image.gz-dtb.
+
 endmenu
 
 menu "Userspace binary formats"
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index ed9a181c4db5..f667885ede86 100644
--- a/arch/arm64/Makefile
+++ b/arch/arm64/Makefile
@@ -30,7 +30,15 @@ $(warning LSE atomics not supported by binutils)
   endif
 endif
 
-KBUILD_CFLAGS	+= -mgeneral-regs-only $(lseinstr)
+ifeq ($(cc-name),clang)
+# This is a workaround for https://bugs.llvm.org/show_bug.cgi?id=30792.
+# TODO: revert when this is fixed in LLVM.
+KBUILD_CFLAGS	+= -mno-implicit-float
+else
+KBUILD_CFLAGS	+= -mgeneral-regs-only
+endif
+KBUILD_CFLAGS	+= $(lseinstr)
+KBUILD_CFLAGS	+= -fno-pic
 KBUILD_CFLAGS	+= $(call cc-option, -mpc-relative-literal-loads)
 KBUILD_CFLAGS	+= -fno-asynchronous-unwind-tables
 KBUILD_AFLAGS	+= $(lseinstr)
@@ -67,6 +75,10 @@ else
 TEXT_OFFSET := 0x00080000
 endif
 
+ifeq ($(cc-name),clang)
+KBUILD_CFLAGS += $(call cc-disable-warning, asm-operand-widths)
+endif
+
 # KASAN_SHADOW_OFFSET = VA_START + (1 << (VA_BITS - 3)) - (1 << 61)
 # in 32-bit arithmetic
 KASAN_SHADOW_OFFSET := $(shell printf "0x%08x00000000\n" $$(( \
@@ -85,7 +97,12 @@ libs-y		:= arch/arm64/lib/ $(libs-y)
 core-$(CONFIG_EFI_STUB) += $(objtree)/drivers/firmware/efi/libstub/lib.a
 
 # Default target when executing plain make
+ifeq ($(CONFIG_BUILD_ARM64_APPENDED_DTB_IMAGE),y)
+KBUILD_IMAGE	:= $(subst $\",,$(CONFIG_BUILD_ARM64_APPENDED_KERNEL_IMAGE_NAME))
+else
 KBUILD_IMAGE	:= Image.gz
+endif
+
 KBUILD_DTBS	:= dtbs
 
 all:	$(KBUILD_IMAGE) $(KBUILD_DTBS)
@@ -112,6 +129,9 @@ dtbs: prepare scripts
 dtbs_install:
 	$(Q)$(MAKE) $(dtbinst)=$(boot)/dts
 
+Image-dtb Image.gz-dtb: vmlinux scripts dtbs
+	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+
 PHONY += vdso_install
 vdso_install:
 	$(Q)$(MAKE) $(build)=arch/arm64/kernel/vdso $@
@@ -121,6 +141,16 @@ archclean:
 	$(Q)$(MAKE) $(clean)=$(boot)
 	$(Q)$(MAKE) $(clean)=$(boot)/dts
 
+# We need to generate vdso-offsets.h before compiling certain files in kernel/.
+# In order to do that, we should use the archprepare target, but we can't since
+# asm-offsets.h is included in some files used to generate vdso-offsets.h, and
+# asm-offsets.h is built in prepare0, for which archprepare is a dependency.
+# Therefore we need to generate the header after prepare0 has been made, hence
+# this hack.
+prepare: vdso_prepare
+vdso_prepare: prepare0
+	$(Q)$(MAKE) $(build)=arch/arm64/kernel/vdso include/generated/vdso-offsets.h
+
 define archhelp
   echo  '* Image.gz      - Compressed kernel image (arch/$(ARCH)/boot/Image.gz)'
   echo  '  Image         - Uncompressed kernel image (arch/$(ARCH)/boot/Image)'
diff --git a/arch/arm64/boot/.gitignore b/arch/arm64/boot/.gitignore
index 8dab0bb6ae66..34e35209fc2e 100644
--- a/arch/arm64/boot/.gitignore
+++ b/arch/arm64/boot/.gitignore
@@ -1,2 +1,4 @@
 Image
+Image-dtb
 Image.gz
+Image.gz-dtb
diff --git a/arch/arm64/boot/Makefile b/arch/arm64/boot/Makefile
index abcbba2f01ba..7ab8e74cd83a 100644
--- a/arch/arm64/boot/Makefile
+++ b/arch/arm64/boot/Makefile
@@ -14,14 +14,27 @@
 # Based on the ia64 boot/Makefile.
 #
 
+include $(srctree)/arch/arm64/boot/dts/Makefile
+
 targets := Image Image.gz
 
+DTB_NAMES := $(subst $\",,$(CONFIG_BUILD_ARM64_APPENDED_DTB_IMAGE_NAMES))
+ifneq ($(DTB_NAMES),)
+DTB_LIST := $(addsuffix .dtb,$(DTB_NAMES))
+else
+DTB_LIST := $(dtb-y)
+endif
+DTB_OBJS := $(addprefix $(obj)/dts/,$(DTB_LIST))
+
 $(obj)/Image: vmlinux FORCE
 	$(call if_changed,objcopy)
 
 $(obj)/Image.bz2: $(obj)/Image FORCE
 	$(call if_changed,bzip2)
 
+$(obj)/Image-dtb: $(obj)/Image $(DTB_OBJS) FORCE
+	$(call if_changed,cat)
+
 $(obj)/Image.gz: $(obj)/Image FORCE
 	$(call if_changed,gzip)
 
@@ -34,6 +47,9 @@ $(obj)/Image.lzma: $(obj)/Image FORCE
 $(obj)/Image.lzo: $(obj)/Image FORCE
 	$(call if_changed,lzo)
 
+$(obj)/Image.gz-dtb: $(obj)/Image.gz $(DTB_OBJS) FORCE
+	$(call if_changed,cat)
+
 install: $(obj)/Image
 	$(CONFIG_SHELL) $(srctree)/$(src)/install.sh $(KERNELRELEASE) \
 	$(obj)/Image System.map "$(INSTALL_PATH)"
diff --git a/arch/arm64/boot/dts/Makefile b/arch/arm64/boot/dts/Makefile
index eb3c42d97175..062edb210993 100644
--- a/arch/arm64/boot/dts/Makefile
+++ b/arch/arm64/boot/dts/Makefile
@@ -21,3 +21,17 @@ dtstree		:= $(srctree)/$(src)
 dtb-$(CONFIG_OF_ALL_DTBS) := $(patsubst $(dtstree)/%.dts,%.dtb, $(foreach d,$(dts-dirs), $(wildcard $(dtstree)/$(d)/*.dts)))
 
 always		:= $(dtb-y)
+
+targets += dtbs
+
+DTB_NAMES := $(subst $\",,$(CONFIG_BUILD_ARM64_APPENDED_DTB_IMAGE_NAMES))
+ifneq ($(DTB_NAMES),)
+DTB_LIST := $(addsuffix .dtb,$(DTB_NAMES))
+else
+DTB_LIST := $(dtb-y)
+endif
+targets += $(DTB_LIST)
+
+dtbs: $(addprefix $(obj)/, $(DTB_LIST))
+
+clean-files := dts/*.dtb *.dtb
diff --git a/arch/arm64/boot/dts/arm/juno-r1.dts b/arch/arm64/boot/dts/arm/juno-r1.dts
index 93bc3d7d51c0..29315af22147 100644
--- a/arch/arm64/boot/dts/arm/juno-r1.dts
+++ b/arch/arm64/boot/dts/arm/juno-r1.dts
@@ -60,6 +60,28 @@
 			};
 		};
 
+		idle-states {
+			entry-method = "arm,psci";
+
+			CPU_SLEEP_0: cpu-sleep-0 {
+				compatible = "arm,idle-state";
+				arm,psci-suspend-param = <0x0010000>;
+				local-timer-stop;
+				entry-latency-us = <300>;
+				exit-latency-us = <1200>;
+				min-residency-us = <2000>;
+			};
+
+			CLUSTER_SLEEP_0: cluster-sleep-0 {
+				compatible = "arm,idle-state";
+				arm,psci-suspend-param = <0x1010000>;
+				local-timer-stop;
+				entry-latency-us = <400>;
+				exit-latency-us = <1200>;
+				min-residency-us = <2500>;
+			};
+		};
+
 		A57_0: cpu@0 {
 			compatible = "arm,cortex-a57","arm,armv8";
 			reg = <0x0 0x0>;
@@ -67,6 +89,7 @@
 			enable-method = "psci";
 			next-level-cache = <&A57_L2>;
 			clocks = <&scpi_dvfs 0>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
 		};
 
 		A57_1: cpu@1 {
@@ -76,6 +99,7 @@
 			enable-method = "psci";
 			next-level-cache = <&A57_L2>;
 			clocks = <&scpi_dvfs 0>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
 		};
 
 		A53_0: cpu@100 {
@@ -85,6 +109,7 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
 		};
 
 		A53_1: cpu@101 {
@@ -94,6 +119,7 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
 		};
 
 		A53_2: cpu@102 {
@@ -103,6 +129,7 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
 		};
 
 		A53_3: cpu@103 {
@@ -112,6 +139,7 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
 		};
 
 		A57_L2: l2-cache0 {
diff --git a/arch/arm64/boot/dts/arm/juno-sched-energy.dtsi b/arch/arm64/boot/dts/arm/juno-sched-energy.dtsi
new file mode 100644
index 000000000000..38207e4391ab
--- /dev/null
+++ b/arch/arm64/boot/dts/arm/juno-sched-energy.dtsi
@@ -0,0 +1,147 @@
+/*
+ * ARM JUNO specific energy cost model data. There are no unit requirements for
+ * the data. Data can be normalized to any reference point, but the
+ * normalization must be consistent. That is, one bogo-joule/watt must be the
+ * same quantity for all data, but we don't care what it is.
+ */
+
+/* static struct idle_state idle_states_cluster_a53[] = { */
+/*        { .power = 56 }, /\* arch_cpu_idle() (active idle) = WFI *\/ */
+/*        { .power = 56 }, /\* WFI *\/ */
+/*        { .power = 56 }, /\* cpu-sleep-0 *\/ */
+/*        { .power = 17 }, /\* cluster-sleep-0 *\/ */
+/* }; */
+
+/* static struct idle_state idle_states_cluster_a57[] = { */
+/*        { .power = 65 }, /\* arch_cpu_idle() (active idle) = WFI *\/ */
+/*        { .power = 65 }, /\* WFI *\/ */
+/*        { .power = 65 }, /\* cpu-sleep-0 *\/ */
+/*        { .power = 24 }, /\* cluster-sleep-0 *\/ */
+/* }; */
+
+/* static struct capacity_state cap_states_cluster_a53[] = { */
+/*         /\* Power per cluster *\/ */
+/*        { .cap =  235, .power = 26, }, /\*  450 MHz *\/ */
+/*        { .cap =  303, .power = 30, }, /\*  575 MHz *\/ */
+/*        { .cap =  368, .power = 39, }, /\*  700 MHz *\/ */
+/*        { .cap =  406, .power = 47, }, /\*  775 MHz *\/ */
+/*        { .cap =  447, .power = 57, }, /\*  850 Mhz *\/ */
+/* }; */
+
+/* static struct capacity_state cap_states_cluster_a57[] = { */
+/*         /\* Power per cluster *\/ */
+/*        { .cap =  417, .power = 24, }, /\*  450 MHz *\/ */
+/*        { .cap =  579, .power = 32, }, /\*  625 MHz *\/ */
+/*        { .cap =  744, .power = 43, }, /\*  800 MHz *\/ */
+/*        { .cap =  883, .power = 49, }, /\*  950 MHz *\/ */
+/*        { .cap = 1024, .power = 64, }, /\* 1100 MHz *\/ */
+/* }; */
+
+/* static struct sched_group_energy energy_cluster_a53 = { */
+/*        .nr_idle_states = ARRAY_SIZE(idle_states_cluster_a53), */
+/*        .idle_states    = idle_states_cluster_a53, */
+/*        .nr_cap_states  = ARRAY_SIZE(cap_states_cluster_a53), */
+/*        .cap_states     = cap_states_cluster_a53, */
+/* }; */
+
+/* static struct sched_group_energy energy_cluster_a57 = { */
+/*        .nr_idle_states = ARRAY_SIZE(idle_states_cluster_a57), */
+/*        .idle_states    = idle_states_cluster_a57, */
+/*        .nr_cap_states  = ARRAY_SIZE(cap_states_cluster_a57), */
+/*        .cap_states     = cap_states_cluster_a57, */
+/* }; */
+
+/* static struct idle_state idle_states_core_a53[] = { */
+/*        { .power = 6 }, /\* arch_cpu_idle() (active idle) = WFI *\/ */
+/*        { .power = 6 }, /\* WFI *\/ */
+/*        { .power = 0 }, /\* cpu-sleep-0 *\/ */
+/*        { .power = 0 }, /\* cluster-sleep-0 *\/ */
+/* }; */
+
+/* static struct idle_state idle_states_core_a57[] = { */
+/*        { .power = 15 }, /\* arch_cpu_idle() (active idle) = WFI *\/ */
+/*        { .power = 15 }, /\* WFI *\/ */
+/*        { .power = 0  }, /\* cpu-sleep-0 *\/ */
+/*        { .power = 0  }, /\* cluster-sleep-0 *\/ */
+/* }; */
+
+/* static struct capacity_state cap_states_core_a53[] = { */
+/*         /\* Power per cpu *\/ */
+/*        { .cap =  235, .power =  33, }, /\*  450 MHz *\/ */
+/*        { .cap =  302, .power =  46, }, /\*  575 MHz *\/ */
+/*        { .cap =  368, .power =  61, }, /\*  700 MHz *\/ */
+/*        { .cap =  406, .power =  76, }, /\*  775 MHz *\/ */
+/*        { .cap =  447, .power =  93, }, /\*  850 Mhz *\/ */
+/* }; */
+
+/* static struct capacity_state cap_states_core_a57[] = { */
+/*         /\* Power per cpu *\/ */
+/*        { .cap =  417, .power = 168, }, /\*  450 MHz *\/ */
+/*        { .cap =  579, .power = 251, }, /\*  625 MHz *\/ */
+/*        { .cap =  744, .power = 359, }, /\*  800 MHz *\/ */
+/*        { .cap =  883, .power = 479, }, /\*  950 MHz *\/ */
+/*        { .cap = 1024, .power = 616, }, /\* 1100 MHz *\/ */
+/* }; */
+
+energy-costs {
+	CPU_COST_A57: core-cost0 {
+		busy-cost-data = <
+			417   168
+			579   251
+			744   359
+			883   479
+		       1023   616
+		>;
+		idle-cost-data = <
+		      15
+		      15
+		       0
+		       0
+		>;
+	};
+	CPU_COST_A53: core-cost1 {
+		busy-cost-data = <
+			235    33
+			302    46
+			368    61
+			406    76
+			447    93
+		>;
+		idle-cost-data = <
+		      6
+		      6
+		      0
+		      0
+		>;
+	};
+	CLUSTER_COST_A57: cluster-cost0 {
+		busy-cost-data = <
+			417    24
+			579    32
+			744    43
+			883    49
+		       1024    64
+		>;
+		idle-cost-data = <
+			 65
+			 65
+			 65
+			 24
+		>;
+	};
+	CLUSTER_COST_A53: cluster-cost1 {
+		busy-cost-data = <
+			235    26
+			303    30
+			368    39
+			406    47
+			447    57
+		>;
+		idle-cost-data = <
+			56
+			56
+			56
+			17
+		>;
+	};
+};
diff --git a/arch/arm64/boot/dts/arm/juno.dts b/arch/arm64/boot/dts/arm/juno.dts
index 3e1a84b01b50..68816f71fa51 100644
--- a/arch/arm64/boot/dts/arm/juno.dts
+++ b/arch/arm64/boot/dts/arm/juno.dts
@@ -60,6 +60,28 @@
 			};
 		};
 
+		idle-states {
+			entry-method = "arm,psci";
+
+			CPU_SLEEP_0: cpu-sleep-0 {
+				compatible = "arm,idle-state";
+				arm,psci-suspend-param = <0x0010000>;
+				local-timer-stop;
+				entry-latency-us = <300>;
+				exit-latency-us = <1200>;
+				min-residency-us = <2000>;
+			};
+
+			CLUSTER_SLEEP_0: cluster-sleep-0 {
+				compatible = "arm,idle-state";
+				arm,psci-suspend-param = <0x1010000>;
+				local-timer-stop;
+				entry-latency-us = <400>;
+				exit-latency-us = <1200>;
+				min-residency-us = <2500>;
+			};
+		};
+
 		A57_0: cpu@0 {
 			compatible = "arm,cortex-a57","arm,armv8";
 			reg = <0x0 0x0>;
@@ -67,6 +89,8 @@
 			enable-method = "psci";
 			next-level-cache = <&A57_L2>;
 			clocks = <&scpi_dvfs 0>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+			sched-energy-costs = <&CPU_COST_A57 &CLUSTER_COST_A57>;
 		};
 
 		A57_1: cpu@1 {
@@ -76,6 +100,8 @@
 			enable-method = "psci";
 			next-level-cache = <&A57_L2>;
 			clocks = <&scpi_dvfs 0>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+			sched-energy-costs = <&CPU_COST_A57 &CLUSTER_COST_A57>;
 		};
 
 		A53_0: cpu@100 {
@@ -85,6 +111,8 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+			sched-energy-costs = <&CPU_COST_A53 &CLUSTER_COST_A53>;
 		};
 
 		A53_1: cpu@101 {
@@ -94,6 +122,8 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+			sched-energy-costs = <&CPU_COST_A53 &CLUSTER_COST_A53>;
 		};
 
 		A53_2: cpu@102 {
@@ -103,6 +133,8 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+			sched-energy-costs = <&CPU_COST_A53 &CLUSTER_COST_A53>;
 		};
 
 		A53_3: cpu@103 {
@@ -112,6 +144,8 @@
 			enable-method = "psci";
 			next-level-cache = <&A53_L2>;
 			clocks = <&scpi_dvfs 1>;
+			cpu-idle-states = <&CPU_SLEEP_0 &CLUSTER_SLEEP_0>;
+			sched-energy-costs = <&CPU_COST_A53 &CLUSTER_COST_A53>;
 		};
 
 		A57_L2: l2-cache0 {
@@ -121,6 +155,8 @@
 		A53_L2: l2-cache1 {
 			compatible = "cache";
 		};
+
+		/include/ "juno-sched-energy.dtsi"
 	};
 
 	pmu_a57 {
diff --git a/arch/arm64/configs/ranchu64_defconfig b/arch/arm64/configs/ranchu64_defconfig
new file mode 100644
index 000000000000..fc55008d8c4c
--- /dev/null
+++ b/arch/arm64/configs/ranchu64_defconfig
@@ -0,0 +1,312 @@
+# CONFIG_LOCALVERSION_AUTO is not set
+# CONFIG_SWAP is not set
+CONFIG_POSIX_MQUEUE=y
+CONFIG_AUDIT=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_BSD_PROCESS_ACCT=y
+CONFIG_BSD_PROCESS_ACCT_V3=y
+CONFIG_TASKSTATS=y
+CONFIG_TASK_DELAY_ACCT=y
+CONFIG_TASK_XACCT=y
+CONFIG_TASK_IO_ACCOUNTING=y
+CONFIG_IKCONFIG=y
+CONFIG_IKCONFIG_PROC=y
+CONFIG_LOG_BUF_SHIFT=14
+CONFIG_CGROUP_DEBUG=y
+CONFIG_CGROUP_FREEZER=y
+CONFIG_CGROUP_CPUACCT=y
+CONFIG_RT_GROUP_SCHED=y
+CONFIG_SCHED_AUTOGROUP=y
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_KALLSYMS_ALL=y
+CONFIG_EMBEDDED=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_PROFILING=y
+CONFIG_ARCH_MMAP_RND_BITS=24
+CONFIG_ARCH_MMAP_RND_COMPAT_BITS=16
+# CONFIG_BLK_DEV_BSG is not set
+# CONFIG_IOSCHED_DEADLINE is not set
+CONFIG_ARCH_VEXPRESS=y
+CONFIG_NR_CPUS=4
+CONFIG_PREEMPT=y
+CONFIG_KSM=y
+CONFIG_SECCOMP=y
+CONFIG_ARMV8_DEPRECATED=y
+CONFIG_SWP_EMULATION=y
+CONFIG_CP15_BARRIER_EMULATION=y
+CONFIG_SETEND_EMULATION=y
+CONFIG_CMDLINE="console=ttyAMA0"
+# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
+CONFIG_COMPAT=y
+CONFIG_PM_AUTOSLEEP=y
+CONFIG_PM_WAKELOCKS=y
+CONFIG_PM_WAKELOCKS_LIMIT=0
+# CONFIG_PM_WAKELOCKS_GC is not set
+CONFIG_PM_DEBUG=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_XFRM_USER=y
+CONFIG_NET_KEY=y
+CONFIG_INET=y
+CONFIG_INET_DIAG_DESTROY=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_INET_ESP=y
+# CONFIG_INET_LRO is not set
+CONFIG_IPV6_ROUTER_PREF=y
+CONFIG_IPV6_ROUTE_INFO=y
+CONFIG_IPV6_OPTIMISTIC_DAD=y
+CONFIG_INET6_AH=y
+CONFIG_INET6_ESP=y
+CONFIG_INET6_IPCOMP=y
+CONFIG_IPV6_MIP6=y
+CONFIG_IPV6_MULTIPLE_TABLES=y
+CONFIG_NETFILTER=y
+CONFIG_NF_CONNTRACK=y
+CONFIG_NF_CONNTRACK_SECMARK=y
+CONFIG_NF_CONNTRACK_EVENTS=y
+CONFIG_NF_CT_PROTO_DCCP=y
+CONFIG_NF_CT_PROTO_SCTP=y
+CONFIG_NF_CT_PROTO_UDPLITE=y
+CONFIG_NF_CONNTRACK_AMANDA=y
+CONFIG_NF_CONNTRACK_FTP=y
+CONFIG_NF_CONNTRACK_H323=y
+CONFIG_NF_CONNTRACK_IRC=y
+CONFIG_NF_CONNTRACK_NETBIOS_NS=y
+CONFIG_NF_CONNTRACK_PPTP=y
+CONFIG_NF_CONNTRACK_SANE=y
+CONFIG_NF_CONNTRACK_TFTP=y
+CONFIG_NF_CT_NETLINK=y
+CONFIG_NETFILTER_XT_TARGET_CLASSIFY=y
+CONFIG_NETFILTER_XT_TARGET_CONNMARK=y
+CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=y
+CONFIG_NETFILTER_XT_TARGET_IDLETIMER=y
+CONFIG_NETFILTER_XT_TARGET_MARK=y
+CONFIG_NETFILTER_XT_TARGET_NFLOG=y
+CONFIG_NETFILTER_XT_TARGET_NFQUEUE=y
+CONFIG_NETFILTER_XT_TARGET_TPROXY=y
+CONFIG_NETFILTER_XT_TARGET_TRACE=y
+CONFIG_NETFILTER_XT_TARGET_SECMARK=y
+CONFIG_NETFILTER_XT_TARGET_TCPMSS=y
+CONFIG_NETFILTER_XT_MATCH_COMMENT=y
+CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_CONNMARK=y
+CONFIG_NETFILTER_XT_MATCH_CONNTRACK=y
+CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_HELPER=y
+CONFIG_NETFILTER_XT_MATCH_IPRANGE=y
+CONFIG_NETFILTER_XT_MATCH_LENGTH=y
+CONFIG_NETFILTER_XT_MATCH_LIMIT=y
+CONFIG_NETFILTER_XT_MATCH_MAC=y
+CONFIG_NETFILTER_XT_MATCH_MARK=y
+CONFIG_NETFILTER_XT_MATCH_POLICY=y
+CONFIG_NETFILTER_XT_MATCH_PKTTYPE=y
+CONFIG_NETFILTER_XT_MATCH_QTAGUID=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA2=y
+CONFIG_NETFILTER_XT_MATCH_SOCKET=y
+CONFIG_NETFILTER_XT_MATCH_STATE=y
+CONFIG_NETFILTER_XT_MATCH_STATISTIC=y
+CONFIG_NETFILTER_XT_MATCH_STRING=y
+CONFIG_NETFILTER_XT_MATCH_TIME=y
+CONFIG_NETFILTER_XT_MATCH_U32=y
+CONFIG_NF_CONNTRACK_IPV4=y
+CONFIG_IP_NF_IPTABLES=y
+CONFIG_IP_NF_MATCH_AH=y
+CONFIG_IP_NF_MATCH_ECN=y
+CONFIG_IP_NF_MATCH_RPFILTER=y
+CONFIG_IP_NF_MATCH_TTL=y
+CONFIG_IP_NF_FILTER=y
+CONFIG_IP_NF_TARGET_REJECT=y
+CONFIG_IP_NF_MANGLE=y
+CONFIG_IP_NF_TARGET_ECN=y
+CONFIG_IP_NF_TARGET_TTL=y
+CONFIG_IP_NF_RAW=y
+CONFIG_IP_NF_SECURITY=y
+CONFIG_IP_NF_ARPTABLES=y
+CONFIG_IP_NF_ARPFILTER=y
+CONFIG_IP_NF_ARP_MANGLE=y
+CONFIG_NF_CONNTRACK_IPV6=y
+CONFIG_IP6_NF_IPTABLES=y
+CONFIG_IP6_NF_MATCH_AH=y
+CONFIG_IP6_NF_MATCH_EUI64=y
+CONFIG_IP6_NF_MATCH_FRAG=y
+CONFIG_IP6_NF_MATCH_OPTS=y
+CONFIG_IP6_NF_MATCH_HL=y
+CONFIG_IP6_NF_MATCH_IPV6HEADER=y
+CONFIG_IP6_NF_MATCH_MH=y
+CONFIG_IP6_NF_MATCH_RT=y
+CONFIG_IP6_NF_TARGET_HL=y
+CONFIG_IP6_NF_FILTER=y
+CONFIG_IP6_NF_TARGET_REJECT=y
+CONFIG_IP6_NF_MANGLE=y
+CONFIG_IP6_NF_RAW=y
+CONFIG_BRIDGE=y
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_HTB=y
+CONFIG_NET_CLS_U32=y
+CONFIG_NET_EMATCH=y
+CONFIG_NET_EMATCH_U32=y
+CONFIG_NET_CLS_ACT=y
+# CONFIG_WIRELESS is not set
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_BLK_DEV_LOOP=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_VIRTIO_BLK=y
+CONFIG_SCSI=y
+# CONFIG_SCSI_PROC_FS is not set
+CONFIG_BLK_DEV_SD=y
+# CONFIG_SCSI_LOWLEVEL is not set
+CONFIG_MD=y
+CONFIG_BLK_DEV_DM=y
+CONFIG_DM_CRYPT=y
+CONFIG_DM_UEVENT=y
+CONFIG_DM_VERITY=y
+CONFIG_DM_VERITY_FEC=y
+CONFIG_NETDEVICES=y
+CONFIG_TUN=y
+CONFIG_VIRTIO_NET=y
+CONFIG_SMC91X=y
+CONFIG_PPP=y
+CONFIG_PPP_BSDCOMP=y
+CONFIG_PPP_DEFLATE=y
+CONFIG_PPP_MPPE=y
+CONFIG_PPPOLAC=y
+CONFIG_PPPOPNS=y
+# CONFIG_WLAN is not set
+CONFIG_INPUT_EVDEV=y
+CONFIG_INPUT_KEYRESET=y
+CONFIG_KEYBOARD_GOLDFISH_EVENTS=y
+# CONFIG_INPUT_MOUSE is not set
+CONFIG_INPUT_JOYSTICK=y
+CONFIG_INPUT_TABLET=y
+CONFIG_INPUT_MISC=y
+CONFIG_INPUT_KEYCHORD=y
+CONFIG_INPUT_UINPUT=y
+CONFIG_INPUT_GPIO=y
+# CONFIG_SERIO_SERPORT is not set
+# CONFIG_VT is not set
+# CONFIG_LEGACY_PTYS is not set
+# CONFIG_DEVMEM is not set
+# CONFIG_DEVKMEM is not set
+CONFIG_SERIAL_AMBA_PL011=y
+CONFIG_SERIAL_AMBA_PL011_CONSOLE=y
+CONFIG_VIRTIO_CONSOLE=y
+# CONFIG_HW_RANDOM is not set
+CONFIG_BATTERY_GOLDFISH=y
+# CONFIG_HWMON is not set
+CONFIG_MEDIA_SUPPORT=y
+CONFIG_FB=y
+CONFIG_FB_GOLDFISH=y
+CONFIG_FB_SIMPLE=y
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+CONFIG_LOGO=y
+# CONFIG_LOGO_LINUX_MONO is not set
+# CONFIG_LOGO_LINUX_VGA16 is not set
+CONFIG_SOUND=y
+CONFIG_SND=y
+CONFIG_HIDRAW=y
+CONFIG_UHID=y
+CONFIG_HID_A4TECH=y
+CONFIG_HID_ACRUX=y
+CONFIG_HID_ACRUX_FF=y
+CONFIG_HID_APPLE=y
+CONFIG_HID_BELKIN=y
+CONFIG_HID_CHERRY=y
+CONFIG_HID_CHICONY=y
+CONFIG_HID_PRODIKEYS=y
+CONFIG_HID_CYPRESS=y
+CONFIG_HID_DRAGONRISE=y
+CONFIG_DRAGONRISE_FF=y
+CONFIG_HID_EMS_FF=y
+CONFIG_HID_ELECOM=y
+CONFIG_HID_EZKEY=y
+CONFIG_HID_KEYTOUCH=y
+CONFIG_HID_KYE=y
+CONFIG_HID_WALTOP=y
+CONFIG_HID_GYRATION=y
+CONFIG_HID_TWINHAN=y
+CONFIG_HID_KENSINGTON=y
+CONFIG_HID_LCPOWER=y
+CONFIG_HID_LOGITECH=y
+CONFIG_HID_LOGITECH_DJ=y
+CONFIG_LOGITECH_FF=y
+CONFIG_LOGIRUMBLEPAD2_FF=y
+CONFIG_LOGIG940_FF=y
+CONFIG_HID_MAGICMOUSE=y
+CONFIG_HID_MICROSOFT=y
+CONFIG_HID_MONTEREY=y
+CONFIG_HID_MULTITOUCH=y
+CONFIG_HID_ORTEK=y
+CONFIG_HID_PANTHERLORD=y
+CONFIG_PANTHERLORD_FF=y
+CONFIG_HID_PETALYNX=y
+CONFIG_HID_PICOLCD=y
+CONFIG_HID_PRIMAX=y
+CONFIG_HID_SAITEK=y
+CONFIG_HID_SAMSUNG=y
+CONFIG_HID_SPEEDLINK=y
+CONFIG_HID_SUNPLUS=y
+CONFIG_HID_GREENASIA=y
+CONFIG_GREENASIA_FF=y
+CONFIG_HID_SMARTJOYPLUS=y
+CONFIG_SMARTJOYPLUS_FF=y
+CONFIG_HID_TIVO=y
+CONFIG_HID_TOPSEED=y
+CONFIG_HID_THRUSTMASTER=y
+CONFIG_HID_WACOM=y
+CONFIG_HID_WIIMOTE=y
+CONFIG_HID_ZEROPLUS=y
+CONFIG_HID_ZYDACRON=y
+# CONFIG_USB_SUPPORT is not set
+CONFIG_RTC_CLASS=y
+CONFIG_VIRTIO_MMIO=y
+CONFIG_STAGING=y
+CONFIG_ASHMEM=y
+CONFIG_ANDROID_TIMED_GPIO=y
+CONFIG_ANDROID_LOW_MEMORY_KILLER=y
+CONFIG_SYNC=y
+CONFIG_SW_SYNC=y
+CONFIG_SW_SYNC_USER=y
+CONFIG_ION=y
+CONFIG_GOLDFISH_AUDIO=y
+CONFIG_GOLDFISH=y
+CONFIG_GOLDFISH_PIPE=y
+# CONFIG_IOMMU_SUPPORT is not set
+CONFIG_ANDROID=y
+CONFIG_ANDROID_BINDER_IPC=y
+CONFIG_EXT2_FS=y
+CONFIG_EXT4_FS=y
+CONFIG_EXT4_FS_SECURITY=y
+CONFIG_QUOTA=y
+CONFIG_FUSE_FS=y
+CONFIG_CUSE=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_TMPFS=y
+CONFIG_TMPFS_POSIX_ACL=y
+# CONFIG_MISC_FILESYSTEMS is not set
+CONFIG_NFS_FS=y
+CONFIG_ROOT_NFS=y
+CONFIG_NLS_CODEPAGE_437=y
+CONFIG_NLS_ISO8859_1=y
+CONFIG_DEBUG_INFO=y
+CONFIG_DEBUG_FS=y
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_PANIC_TIMEOUT=5
+# CONFIG_SCHED_DEBUG is not set
+CONFIG_SCHEDSTATS=y
+CONFIG_TIMER_STATS=y
+# CONFIG_FTRACE is not set
+CONFIG_ATOMIC64_SELFTEST=y
+CONFIG_DEBUG_RODATA=y
+CONFIG_SECURITY=y
+CONFIG_SECURITY_NETWORK=y
+CONFIG_SECURITY_SELINUX=y
diff --git a/arch/arm64/crypto/Kconfig b/arch/arm64/crypto/Kconfig
index 2cf32e9887e1..de1aab4b5da8 100644
--- a/arch/arm64/crypto/Kconfig
+++ b/arch/arm64/crypto/Kconfig
@@ -23,6 +23,11 @@ config CRYPTO_GHASH_ARM64_CE
 	depends on ARM64 && KERNEL_MODE_NEON
 	select CRYPTO_HASH
 
+config CRYPTO_POLY_HASH_ARM64_CE
+	tristate "poly_hash (for HEH encryption mode) using ARMv8 Crypto Extensions"
+	depends on ARM64 && KERNEL_MODE_NEON
+	select CRYPTO_HASH
+
 config CRYPTO_AES_ARM64_CE
 	tristate "AES core cipher using ARMv8 Crypto Extensions"
 	depends on ARM64 && KERNEL_MODE_NEON
diff --git a/arch/arm64/crypto/Makefile b/arch/arm64/crypto/Makefile
index abb79b3cfcfe..f0a8f2475ea3 100644
--- a/arch/arm64/crypto/Makefile
+++ b/arch/arm64/crypto/Makefile
@@ -17,6 +17,9 @@ sha2-ce-y := sha2-ce-glue.o sha2-ce-core.o
 obj-$(CONFIG_CRYPTO_GHASH_ARM64_CE) += ghash-ce.o
 ghash-ce-y := ghash-ce-glue.o ghash-ce-core.o
 
+obj-$(CONFIG_CRYPTO_POLY_HASH_ARM64_CE) += poly-hash-ce.o
+poly-hash-ce-y := poly-hash-ce-glue.o poly-hash-ce-core.o
+
 obj-$(CONFIG_CRYPTO_AES_ARM64_CE) += aes-ce-cipher.o
 CFLAGS_aes-ce-cipher.o += -march=armv8-a+crypto
 
diff --git a/arch/arm64/crypto/aes-glue.c b/arch/arm64/crypto/aes-glue.c
index 6a51dfccfe71..448b874a4826 100644
--- a/arch/arm64/crypto/aes-glue.c
+++ b/arch/arm64/crypto/aes-glue.c
@@ -294,7 +294,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_blkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= aes_setkey,
 		.encrypt	= ecb_encrypt,
 		.decrypt	= ecb_decrypt,
@@ -371,7 +371,7 @@ static struct crypto_alg aes_algs[] = { {
 	.cra_ablkcipher = {
 		.min_keysize	= AES_MIN_KEY_SIZE,
 		.max_keysize	= AES_MAX_KEY_SIZE,
-		.ivsize		= AES_BLOCK_SIZE,
+		.ivsize		= 0,
 		.setkey		= ablk_set_key,
 		.encrypt	= ablk_encrypt,
 		.decrypt	= ablk_decrypt,
diff --git a/arch/arm64/crypto/poly-hash-ce-core.S b/arch/arm64/crypto/poly-hash-ce-core.S
new file mode 100644
index 000000000000..8ccb544c5526
--- /dev/null
+++ b/arch/arm64/crypto/poly-hash-ce-core.S
@@ -0,0 +1,163 @@
+/*
+ * Accelerated poly_hash implementation with ARMv8 PMULL instructions.
+ *
+ * Based on ghash-ce-core.S.
+ *
+ * Copyright (C) 2014 Linaro Ltd. <ard.biesheuvel@linaro.org>
+ * Copyright (C) 2017 Google, Inc. <ebiggers@google.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ */
+
+#include <linux/linkage.h>
+#include <asm/assembler.h>
+
+	KEY	.req	v0
+	KEY2	.req	v1
+	T1	.req	v2
+	T2	.req	v3
+	GSTAR	.req	v4
+	XL	.req	v5
+	XM	.req	v6
+	XH	.req	v7
+
+	.text
+	.arch		armv8-a+crypto
+
+	/* 16-byte aligned (2**4 = 16); not required, but might as well */
+	.align		4
+.Lgstar:
+	.quad		0x87, 0x87
+
+/*
+ * void pmull_poly_hash_update(le128 *digest, const le128 *key,
+ *			       const u8 *src, unsigned int blocks,
+ *			       unsigned int partial);
+ */
+ENTRY(pmull_poly_hash_update)
+
+	/* Load digest into XL */
+	ld1		{XL.16b}, [x0]
+
+	/* Load key into KEY */
+	ld1		{KEY.16b}, [x1]
+
+	/* Load g*(x) = g(x) + x^128 = x^7 + x^2 + x + 1 into both halves of
+	 * GSTAR */
+	adr		x1, .Lgstar
+	ld1		{GSTAR.2d}, [x1]
+
+	/* Set KEY2 to (KEY[1]+KEY[0]):(KEY[1]+KEY[0]).  This is needed for
+	 * Karatsuba multiplication. */
+	ext		KEY2.16b, KEY.16b, KEY.16b, #8
+	eor		KEY2.16b, KEY2.16b, KEY.16b
+
+	/* If 'partial' is nonzero, then we're finishing a pending block and
+	 * should go right to the multiplication. */
+	cbnz		w4, 1f
+
+0:
+	/* Add the next block from 'src' to the digest */
+	ld1		{T1.16b}, [x2], #16
+	eor		XL.16b, XL.16b, T1.16b
+	sub		w3, w3, #1
+
+1:
+	/*
+	 * Multiply the current 128-bit digest (a1:a0, in XL) by the 128-bit key
+	 * (b1:b0, in KEY) using Karatsuba multiplication.
+	 */
+
+	/* T1 = (a1+a0):(a1+a0) */
+	ext		T1.16b, XL.16b, XL.16b, #8
+	eor		T1.16b, T1.16b, XL.16b
+
+	/* XH = a1 * b1 */
+	pmull2		XH.1q, XL.2d, KEY.2d
+
+	/* XL = a0 * b0 */
+	pmull		XL.1q, XL.1d, KEY.1d
+
+	/* XM = (a1+a0) * (b1+b0) */
+	pmull		XM.1q, T1.1d, KEY2.1d
+
+	/* XM += (XH[0]:XL[1]) + XL + XH */
+	ext		T1.16b, XL.16b, XH.16b, #8
+	eor		T2.16b, XL.16b, XH.16b
+	eor		XM.16b, XM.16b, T1.16b
+	eor		XM.16b, XM.16b, T2.16b
+
+	/*
+	 * Now the 256-bit product is in XH[1]:XM:XL[0].  It represents a
+	 * polynomial over GF(2) with degree as large as 255.  We need to
+	 * compute its remainder modulo g(x) = x^128+x^7+x^2+x+1.  For this it
+	 * is sufficient to compute the remainder of the high half 'c(x)x^128'
+	 * add it to the low half.  To reduce the high half we use the Barrett
+	 * reduction method.  The basic idea is that we can express the
+	 * remainder p(x) as g(x)q(x) mod x^128, where q(x) = (c(x)x^128)/g(x).
+	 * As detailed in [1], to avoid having to divide by g(x) at runtime the
+	 * following equivalent expression can be derived:
+	 *
+	 *	p(x) = [ g*(x)((c(x)q+(x))/x^128) ] mod x^128
+	 *
+	 * where g*(x) = x^128+g(x) = x^7+x^2+x+1, and q+(x) = x^256/g(x) = g(x)
+	 * in this case.  This is also equivalent to:
+	 *
+	 *	p(x) = [ g*(x)((c(x)(x^128 + g*(x)))/x^128) ] mod x^128
+	 *	     = [ g*(x)(c(x) + (c(x)g*(x))/x^128) ] mod x^128
+	 *
+	 * Since deg g*(x) < 64:
+	 *
+	 *	p(x) = [ g*(x)(c(x) + ((c(x)/x^64)g*(x))/x^64) ] mod x^128
+	 *	     = [ g*(x)((c(x)/x^64)x^64 + (c(x) mod x^64) +
+	 *				((c(x)/x^64)g*(x))/x^64) ] mod x^128
+	 *
+	 * Letting t(x) = g*(x)(c(x)/x^64):
+	 *
+	 *	p(x) = [ t(x)x^64 + g*(x)((c(x) mod x^64) + t(x)/x^64) ] mod x^128
+	 *
+	 * Therefore, to do the reduction we only need to issue two 64-bit =>
+	 * 128-bit carryless multiplications: g*(x) times c(x)/x^64, and g*(x)
+	 * times ((c(x) mod x^64) + t(x)/x^64).  (Multiplication by x^64 doesn't
+	 * count since it is simply a shift or move.)
+	 *
+	 * An alternate reduction method, also based on Barrett reduction and
+	 * described in [1], uses only shifts and XORs --- no multiplications.
+	 * However, the method with multiplications requires fewer instructions
+	 * and is faster on processors with fast carryless multiplication.
+	 *
+	 * [1] "Intel Carry-Less Multiplication Instruction and its Usage for
+	 * Computing the GCM Mode",
+	 * https://software.intel.com/sites/default/files/managed/72/cc/clmul-wp-rev-2.02-2014-04-20.pdf
+	 */
+
+	/* 256-bit product is XH[1]:XM:XL[0], so c(x) is XH[1]:XM[1] */
+
+	/* T1 = t(x) = g*(x)(c(x)/x^64) */
+	pmull2		T1.1q, GSTAR.2d, XH.2d
+
+	/* T2 = g*(x)((c(x) mod x^64) + t(x)/x^64) */
+	eor		T2.16b, XM.16b, T1.16b
+	pmull2		T2.1q, GSTAR.2d, T2.2d
+
+	/* Make XL[0] be the low half of the 128-bit result by adding the low 64
+	 * bits of the T2 term to what was already there.  The 't(x)x^64' term
+	 * makes no difference, so skip it. */
+	eor		XL.16b, XL.16b, T2.16b
+
+	/* Make XL[1] be the high half of the 128-bit result by adding the high
+	 * 64 bits of the 't(x)x^64' and T2 terms to what was already in XM[0],
+	 * then moving XM[0] to XL[1]. */
+	eor		XM.16b, XM.16b, T1.16b
+	ext		T2.16b, T2.16b, T2.16b, #8
+	eor		XM.16b, XM.16b, T2.16b
+	mov		XL.d[1], XM.d[0]
+
+	/* If more blocks remain, then loop back to process the next block;
+	 * else, store the digest and return. */
+	cbnz		w3, 0b
+	st1		{XL.16b}, [x0]
+	ret
+ENDPROC(pmull_poly_hash_update)
diff --git a/arch/arm64/crypto/poly-hash-ce-glue.c b/arch/arm64/crypto/poly-hash-ce-glue.c
new file mode 100644
index 000000000000..e195740c9ecf
--- /dev/null
+++ b/arch/arm64/crypto/poly-hash-ce-glue.c
@@ -0,0 +1,166 @@
+/*
+ * Accelerated poly_hash implementation with ARMv8 PMULL instructions.
+ *
+ * Based on ghash-ce-glue.c.
+ *
+ * poly_hash is part of the HEH (Hash-Encrypt-Hash) encryption mode, proposed in
+ * Internet Draft https://tools.ietf.org/html/draft-cope-heh-01.
+ *
+ * poly_hash is very similar to GHASH: both algorithms are keyed hashes which
+ * interpret their input data as coefficients of a polynomial over GF(2^128),
+ * then calculate a hash value by evaluating that polynomial at the point given
+ * by the key, e.g. using Horner's rule.  The difference is that poly_hash uses
+ * the more natural "ble" convention to represent GF(2^128) elements, whereas
+ * GHASH uses the less natural "lle" convention (see include/crypto/gf128mul.h).
+ * The ble convention makes it simpler to implement GF(2^128) multiplication.
+ *
+ * Copyright (C) 2014 Linaro Ltd. <ard.biesheuvel@linaro.org>
+ * Copyright (C) 2017 Google Inc. <ebiggers@google.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License version 2 as published
+ * by the Free Software Foundation.
+ */
+
+#include <asm/neon.h>
+#include <crypto/b128ops.h>
+#include <crypto/internal/hash.h>
+#include <linux/cpufeature.h>
+#include <linux/crypto.h>
+#include <linux/module.h>
+
+/*
+ * Note: in this algorithm we currently use 'le128' to represent GF(2^128)
+ * elements, even though poly_hash-generic uses 'be128'.  Both types are
+ * actually "wrong" because the elements are actually in 'ble' format, and there
+ * should be a ble type to represent this --- as well as lle, bbe, and lbe types
+ * for the other conventions for representing GF(2^128) elements.  But
+ * practically it doesn't matter which type we choose here, so we just use le128
+ * since it's arguably more accurate, while poly_hash-generic still has to use
+ * be128 because the generic GF(2^128) multiplication functions all take be128.
+ */
+
+struct poly_hash_desc_ctx {
+	le128 digest;
+	unsigned int count;
+};
+
+asmlinkage void pmull_poly_hash_update(le128 *digest, const le128 *key,
+				       const u8 *src, unsigned int blocks,
+				       unsigned int partial);
+
+static int poly_hash_setkey(struct crypto_shash *tfm,
+			    const u8 *key, unsigned int keylen)
+{
+	if (keylen != sizeof(le128)) {
+		crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
+		return -EINVAL;
+	}
+
+	memcpy(crypto_shash_ctx(tfm), key, sizeof(le128));
+	return 0;
+}
+
+static int poly_hash_init(struct shash_desc *desc)
+{
+	struct poly_hash_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	ctx->digest = (le128) { 0 };
+	ctx->count = 0;
+	return 0;
+}
+
+static int poly_hash_update(struct shash_desc *desc, const u8 *src,
+			    unsigned int len)
+{
+	struct poly_hash_desc_ctx *ctx = shash_desc_ctx(desc);
+	unsigned int partial = ctx->count % sizeof(le128);
+	u8 *dst = (u8 *)&ctx->digest + partial;
+
+	ctx->count += len;
+
+	/* Finishing at least one block? */
+	if (partial + len >= sizeof(le128)) {
+		const le128 *key = crypto_shash_ctx(desc->tfm);
+
+		if (partial) {
+			/* Finish the pending block. */
+			unsigned int n = sizeof(le128) - partial;
+
+			len -= n;
+			do {
+				*dst++ ^= *src++;
+			} while (--n);
+		}
+
+		/*
+		 * Do the real work.  If 'partial' is nonzero, this starts by
+		 * multiplying 'digest' by 'key'.  Then for each additional full
+		 * block it adds the block to 'digest' and multiplies by 'key'.
+		 */
+		kernel_neon_begin_partial(8);
+		pmull_poly_hash_update(&ctx->digest, key, src,
+				       len / sizeof(le128), partial);
+		kernel_neon_end();
+
+		src += len - (len % sizeof(le128));
+		len %= sizeof(le128);
+		dst = (u8 *)&ctx->digest;
+	}
+
+	/* Continue adding the next block to 'digest'. */
+	while (len--)
+		*dst++ ^= *src++;
+	return 0;
+}
+
+static int poly_hash_final(struct shash_desc *desc, u8 *out)
+{
+	struct poly_hash_desc_ctx *ctx = shash_desc_ctx(desc);
+	unsigned int partial = ctx->count % sizeof(le128);
+
+	/* Finish the last block if needed. */
+	if (partial) {
+		const le128 *key = crypto_shash_ctx(desc->tfm);
+
+		kernel_neon_begin_partial(8);
+		pmull_poly_hash_update(&ctx->digest, key, NULL, 0, partial);
+		kernel_neon_end();
+	}
+
+	memcpy(out, &ctx->digest, sizeof(le128));
+	return 0;
+}
+
+static struct shash_alg poly_hash_alg = {
+	.digestsize	= sizeof(le128),
+	.init		= poly_hash_init,
+	.update		= poly_hash_update,
+	.final		= poly_hash_final,
+	.setkey		= poly_hash_setkey,
+	.descsize	= sizeof(struct poly_hash_desc_ctx),
+	.base		= {
+		.cra_name		= "poly_hash",
+		.cra_driver_name	= "poly_hash-ce",
+		.cra_priority		= 300,
+		.cra_ctxsize		= sizeof(le128),
+		.cra_module		= THIS_MODULE,
+	},
+};
+
+static int __init poly_hash_ce_mod_init(void)
+{
+	return crypto_register_shash(&poly_hash_alg);
+}
+
+static void __exit poly_hash_ce_mod_exit(void)
+{
+	crypto_unregister_shash(&poly_hash_alg);
+}
+
+MODULE_DESCRIPTION("Polynomial evaluation hash using ARMv8 Crypto Extensions");
+MODULE_AUTHOR("Eric Biggers <ebiggers@google.com>");
+MODULE_LICENSE("GPL v2");
+
+module_cpu_feature_match(PMULL, poly_hash_ce_mod_init);
+module_exit(poly_hash_ce_mod_exit);
diff --git a/arch/arm64/crypto/sha1-ce-core.S b/arch/arm64/crypto/sha1-ce-core.S
index c98e7e849f06..8550408735a0 100644
--- a/arch/arm64/crypto/sha1-ce-core.S
+++ b/arch/arm64/crypto/sha1-ce-core.S
@@ -82,7 +82,8 @@ ENTRY(sha1_ce_transform)
 	ldr		dgb, [x0, #16]
 
 	/* load sha1_ce_state::finalize */
-	ldr		w4, [x0, #:lo12:sha1_ce_offsetof_finalize]
+	ldr_l		w4, sha1_ce_offsetof_finalize, x4
+	ldr		w4, [x0, x4]
 
 	/* load input */
 0:	ld1		{v8.4s-v11.4s}, [x1], #64
@@ -132,7 +133,8 @@ CPU_LE(	rev32		v11.16b, v11.16b	)
 	 * the padding is handled by the C code in that case.
 	 */
 	cbz		x4, 3f
-	ldr		x4, [x0, #:lo12:sha1_ce_offsetof_count]
+	ldr_l		w4, sha1_ce_offsetof_count, x4
+	ldr		x4, [x0, x4]
 	movi		v9.2d, #0
 	mov		x8, #0x80000000
 	movi		v10.2d, #0
diff --git a/arch/arm64/crypto/sha1-ce-glue.c b/arch/arm64/crypto/sha1-ce-glue.c
index aefda9868627..ea319c055f5d 100644
--- a/arch/arm64/crypto/sha1-ce-glue.c
+++ b/arch/arm64/crypto/sha1-ce-glue.c
@@ -17,9 +17,6 @@
 #include <linux/crypto.h>
 #include <linux/module.h>
 
-#define ASM_EXPORT(sym, val) \
-	asm(".globl " #sym "; .set " #sym ", %0" :: "I"(val));
-
 MODULE_DESCRIPTION("SHA1 secure hash using ARMv8 Crypto Extensions");
 MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
 MODULE_LICENSE("GPL v2");
@@ -32,6 +29,9 @@ struct sha1_ce_state {
 asmlinkage void sha1_ce_transform(struct sha1_ce_state *sst, u8 const *src,
 				  int blocks);
 
+const u32 sha1_ce_offsetof_count = offsetof(struct sha1_ce_state, sst.count);
+const u32 sha1_ce_offsetof_finalize = offsetof(struct sha1_ce_state, finalize);
+
 static int sha1_ce_update(struct shash_desc *desc, const u8 *data,
 			  unsigned int len)
 {
@@ -52,11 +52,6 @@ static int sha1_ce_finup(struct shash_desc *desc, const u8 *data,
 	struct sha1_ce_state *sctx = shash_desc_ctx(desc);
 	bool finalize = !sctx->sst.count && !(len % SHA1_BLOCK_SIZE);
 
-	ASM_EXPORT(sha1_ce_offsetof_count,
-		   offsetof(struct sha1_ce_state, sst.count));
-	ASM_EXPORT(sha1_ce_offsetof_finalize,
-		   offsetof(struct sha1_ce_state, finalize));
-
 	/*
 	 * Allow the asm code to perform the finalization if there is no
 	 * partial data and the input is a round multiple of the block size.
diff --git a/arch/arm64/crypto/sha2-ce-core.S b/arch/arm64/crypto/sha2-ce-core.S
index 01cfee066837..679c6c002f4f 100644
--- a/arch/arm64/crypto/sha2-ce-core.S
+++ b/arch/arm64/crypto/sha2-ce-core.S
@@ -88,7 +88,8 @@ ENTRY(sha2_ce_transform)
 	ld1		{dgav.4s, dgbv.4s}, [x0]
 
 	/* load sha256_ce_state::finalize */
-	ldr		w4, [x0, #:lo12:sha256_ce_offsetof_finalize]
+	ldr_l		w4, sha256_ce_offsetof_finalize, x4
+	ldr		w4, [x0, x4]
 
 	/* load input */
 0:	ld1		{v16.4s-v19.4s}, [x1], #64
@@ -136,7 +137,8 @@ CPU_LE(	rev32		v19.16b, v19.16b	)
 	 * the padding is handled by the C code in that case.
 	 */
 	cbz		x4, 3f
-	ldr		x4, [x0, #:lo12:sha256_ce_offsetof_count]
+	ldr_l		w4, sha256_ce_offsetof_count, x4
+	ldr		x4, [x0, x4]
 	movi		v17.2d, #0
 	mov		x8, #0x80000000
 	movi		v18.2d, #0
diff --git a/arch/arm64/crypto/sha2-ce-glue.c b/arch/arm64/crypto/sha2-ce-glue.c
index 7cd587564a41..0ed9486f75dd 100644
--- a/arch/arm64/crypto/sha2-ce-glue.c
+++ b/arch/arm64/crypto/sha2-ce-glue.c
@@ -17,9 +17,6 @@
 #include <linux/crypto.h>
 #include <linux/module.h>
 
-#define ASM_EXPORT(sym, val) \
-	asm(".globl " #sym "; .set " #sym ", %0" :: "I"(val));
-
 MODULE_DESCRIPTION("SHA-224/SHA-256 secure hash using ARMv8 Crypto Extensions");
 MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>");
 MODULE_LICENSE("GPL v2");
@@ -32,6 +29,11 @@ struct sha256_ce_state {
 asmlinkage void sha2_ce_transform(struct sha256_ce_state *sst, u8 const *src,
 				  int blocks);
 
+const u32 sha256_ce_offsetof_count = offsetof(struct sha256_ce_state,
+					      sst.count);
+const u32 sha256_ce_offsetof_finalize = offsetof(struct sha256_ce_state,
+						 finalize);
+
 static int sha256_ce_update(struct shash_desc *desc, const u8 *data,
 			    unsigned int len)
 {
@@ -52,11 +54,6 @@ static int sha256_ce_finup(struct shash_desc *desc, const u8 *data,
 	struct sha256_ce_state *sctx = shash_desc_ctx(desc);
 	bool finalize = !sctx->sst.count && !(len % SHA256_BLOCK_SIZE);
 
-	ASM_EXPORT(sha256_ce_offsetof_count,
-		   offsetof(struct sha256_ce_state, sst.count));
-	ASM_EXPORT(sha256_ce_offsetof_finalize,
-		   offsetof(struct sha256_ce_state, finalize));
-
 	/*
 	 * Allow the asm code to perform the finalization if there is no
 	 * partial data and the input is a round multiple of the block size.
diff --git a/arch/arm64/include/asm/Kbuild b/arch/arm64/include/asm/Kbuild
index 70fd9ffb58cf..232984266611 100644
--- a/arch/arm64/include/asm/Kbuild
+++ b/arch/arm64/include/asm/Kbuild
@@ -36,7 +36,6 @@ generic-y += poll.h
 generic-y += preempt.h
 generic-y += resource.h
 generic-y += rwsem.h
-generic-y += sections.h
 generic-y += segment.h
 generic-y += sembuf.h
 generic-y += serial.h
diff --git a/arch/arm64/include/asm/alternative.h b/arch/arm64/include/asm/alternative.h
index beccbdefa106..55101bd86b98 100644
--- a/arch/arm64/include/asm/alternative.h
+++ b/arch/arm64/include/asm/alternative.h
@@ -2,6 +2,7 @@
 #define __ASM_ALTERNATIVE_H
 
 #include <asm/cpufeature.h>
+#include <asm/insn.h>
 
 #ifndef __ASSEMBLY__
 
@@ -90,26 +91,15 @@ void apply_alternatives(void *start, size_t length);
 .endm
 
 /*
- * Begin an alternative code sequence.
+ * Alternative sequences
  *
- * The code that follows this macro will be assembled and linked as
- * normal. There are no restrictions on this code.
- */
-.macro alternative_if_not cap, enable = 1
-	.if \enable
-	.pushsection .altinstructions, "a"
-	altinstruction_entry 661f, 663f, \cap, 662f-661f, 664f-663f
-	.popsection
-661:
-	.endif
-.endm
-
-/*
- * Provide the alternative code sequence.
+ * The code for the case where the capability is not present will be
+ * assembled and linked as normal. There are no restrictions on this
+ * code.
  *
- * The code that follows this macro is assembled into a special
- * section to be used for dynamic patching. Code that follows this
- * macro must:
+ * The code for the case where the capability is present will be
+ * assembled into a special section to be used for dynamic patching.
+ * Code for that case must:
  *
  * 1. Be exactly the same length (in bytes) as the default code
  *    sequence.
@@ -118,27 +108,71 @@ void apply_alternatives(void *start, size_t length);
  *    alternative sequence it is defined in (branches into an
  *    alternative sequence are not fixed up).
  */
-.macro alternative_else, enable = 1
-	.if \enable
-662:	.pushsection .altinstr_replacement, "ax"
-663:
+
+/*
+ * Begin an alternative code sequence.
+ */
+.macro alternative_if_not cap
+	.set .Lasm_alt_mode, 0
+	.pushsection .altinstructions, "a"
+	altinstruction_entry 661f, 663f, \cap, 662f-661f, 664f-663f
+	.popsection
+661:
+.endm
+
+.macro alternative_if cap
+	.set .Lasm_alt_mode, 1
+	.pushsection .altinstructions, "a"
+	altinstruction_entry 663f, 661f, \cap, 664f-663f, 662f-661f
+	.popsection
+	.pushsection .altinstr_replacement, "ax"
+	.align 2	/* So GAS knows label 661 is suitably aligned */
+661:
+.endm
+
+/*
+ * Provide the other half of the alternative code sequence.
+ */
+.macro alternative_else
+662:
+	.if .Lasm_alt_mode==0
+	.pushsection .altinstr_replacement, "ax"
+	.else
+	.popsection
 	.endif
+663:
 .endm
 
 /*
  * Complete an alternative code sequence.
  */
-.macro alternative_endif, enable = 1
-	.if \enable
-664:	.popsection
+.macro alternative_endif
+664:
+	.if .Lasm_alt_mode==0
+	.popsection
+	.endif
 	.org	. - (664b-663b) + (662b-661b)
 	.org	. - (662b-661b) + (664b-663b)
-	.endif
+.endm
+
+/*
+ * Provides a trivial alternative or default sequence consisting solely
+ * of NOPs. The number of NOPs is chosen automatically to match the
+ * previous case.
+ */
+.macro alternative_else_nop_endif
+alternative_else
+	nops	(662b-661b) / AARCH64_INSN_SIZE
+alternative_endif
 .endm
 
 #define _ALTERNATIVE_CFG(insn1, insn2, cap, cfg, ...)	\
 	alternative_insn insn1, insn2, cap, IS_ENABLED(cfg)
 
+.macro user_alt, label, oldinstr, newinstr, cond
+9999:	alternative_insn "\oldinstr", "\newinstr", \cond
+	_ASM_EXTABLE 9999b, \label
+.endm
 
 /*
  * Generate the assembly for UAO alternatives with exception table entries.
diff --git a/arch/arm64/include/asm/assembler.h b/arch/arm64/include/asm/assembler.h
index 290e13428f4a..e450bb6d21bd 100644
--- a/arch/arm64/include/asm/assembler.h
+++ b/arch/arm64/include/asm/assembler.h
@@ -24,6 +24,7 @@
 #define __ASM_ASSEMBLER_H
 
 #include <asm/asm-offsets.h>
+#include <asm/cpufeature.h>
 #include <asm/page.h>
 #include <asm/pgtable-hwdef.h>
 #include <asm/ptrace.h>
@@ -52,6 +53,15 @@
 	msr	daifclr, #2
 	.endm
 
+	.macro	save_and_disable_irq, flags
+	mrs	\flags, daif
+	msr	daifset, #2
+	.endm
+
+	.macro	restore_irq, flags
+	msr	daif, \flags
+	.endm
+
 /*
  * Enable and disable debug exceptions.
  */
@@ -98,6 +108,15 @@
 	.endm
 
 /*
+ * NOP sequence
+ */
+	.macro	nops, num
+	.rept	\num
+	nop
+	.endr
+	.endm
+
+/*
  * Emit an entry into the exception table
  */
 	.macro		_asm_extable, from, to
@@ -204,14 +223,25 @@ lr	.req	x30		// link register
 	.endm
 
 	/*
+	 * @dst: Result of per_cpu(sym, smp_processor_id())
 	 * @sym: The name of the per-cpu variable
-	 * @reg: Result of per_cpu(sym, smp_processor_id())
 	 * @tmp: scratch register
 	 */
-	.macro this_cpu_ptr, sym, reg, tmp
-	adr_l	\reg, \sym
+	.macro adr_this_cpu, dst, sym, tmp
+	adr_l	\dst, \sym
 	mrs	\tmp, tpidr_el1
-	add	\reg, \reg, \tmp
+	add	\dst, \dst, \tmp
+	.endm
+
+	/*
+	 * @dst: Result of READ_ONCE(per_cpu(sym, smp_processor_id()))
+	 * @sym: The name of the per-cpu variable
+	 * @tmp: scratch register
+	 */
+	.macro ldr_this_cpu dst, sym, tmp
+	adr_l	\dst, \sym
+	mrs	\tmp, tpidr_el1
+	ldr	\dst, [\dst, \tmp]
 	.endm
 
 /*
@@ -273,7 +303,16 @@ lr	.req	x30		// link register
 	add	\size, \kaddr, \size
 	sub	\tmp2, \tmp1, #1
 	bic	\kaddr, \kaddr, \tmp2
-9998:	dc	\op, \kaddr
+9998:
+	.if	(\op == cvau || \op == cvac)
+alternative_if_not ARM64_WORKAROUND_CLEAN_CACHE
+	dc	\op, \kaddr
+alternative_else
+	dc	civac, \kaddr
+alternative_endif
+	.else
+	dc	\op, \kaddr
+	.endif
 	add	\kaddr, \kaddr, \tmp1
 	cmp	\kaddr, \size
 	b.lo	9998b
@@ -352,4 +391,24 @@ lr	.req	x30		// link register
 	movk	\reg, :abs_g0_nc:\val
 	.endm
 
+/*
+ * Return the current thread_info.
+ */
+	.macro	get_thread_info, rd
+	mrs	\rd, sp_el0
+	.endm
+
+/*
+ * Errata workaround post TTBR0_EL1 update.
+ */
+	.macro	post_ttbr0_update_workaround
+#ifdef CONFIG_CAVIUM_ERRATUM_27456
+alternative_if ARM64_WORKAROUND_CAVIUM_27456
+	ic	iallu
+	dsb	nsh
+	isb
+alternative_else_nop_endif
+#endif
+	.endm
+
 #endif	/* __ASM_ASSEMBLER_H */
diff --git a/arch/arm64/include/asm/barrier.h b/arch/arm64/include/asm/barrier.h
index f2d2c0bbe21b..0671711b46ab 100644
--- a/arch/arm64/include/asm/barrier.h
+++ b/arch/arm64/include/asm/barrier.h
@@ -20,6 +20,9 @@
 
 #ifndef __ASSEMBLY__
 
+#define __nops(n)	".rept	" #n "\nnop\n.endr\n"
+#define nops(n)		asm volatile(__nops(n))
+
 #define sev()		asm volatile("sev" : : : "memory")
 #define wfe()		asm volatile("wfe" : : : "memory")
 #define wfi()		asm volatile("wfi" : : : "memory")
diff --git a/arch/arm64/include/asm/cpufeature.h b/arch/arm64/include/asm/cpufeature.h
index 1695f77d8bf2..8e1f826caf99 100644
--- a/arch/arm64/include/asm/cpufeature.h
+++ b/arch/arm64/include/asm/cpufeature.h
@@ -189,6 +189,12 @@ static inline bool system_supports_mixed_endian_el0(void)
 	return id_aa64mmfr0_mixed_endian_el0(read_system_reg(SYS_ID_AA64MMFR0_EL1));
 }
 
+static inline bool system_uses_ttbr0_pan(void)
+{
+	return IS_ENABLED(CONFIG_ARM64_SW_TTBR0_PAN) &&
+		!cpus_have_cap(ARM64_HAS_PAN);
+}
+
 #endif /* __ASSEMBLY__ */
 
 #endif
diff --git a/arch/arm64/include/asm/current.h b/arch/arm64/include/asm/current.h
new file mode 100644
index 000000000000..483a6c9d3e10
--- /dev/null
+++ b/arch/arm64/include/asm/current.h
@@ -0,0 +1,35 @@
+#ifndef __ASM_CURRENT_H
+#define __ASM_CURRENT_H
+
+#include <linux/compiler.h>
+
+#include <asm/sysreg.h>
+
+#ifndef __ASSEMBLY__
+
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+struct task_struct;
+
+/*
+ * We don't use read_sysreg() as we want the compiler to cache the value where
+ * possible.
+ */
+static __always_inline struct task_struct *get_current(void)
+{
+	unsigned long sp_el0;
+
+	asm ("mrs %0, sp_el0" : "=r" (sp_el0));
+
+	return (struct task_struct *)sp_el0;
+}
+#define current get_current()
+#else
+#include <linux/thread_info.h>
+#define get_current() (current_thread_info()->task)
+#define current get_current()
+#endif
+
+#endif /* __ASSEMBLY__ */
+
+#endif /* __ASM_CURRENT_H */
+
diff --git a/arch/arm64/include/asm/efi.h b/arch/arm64/include/asm/efi.h
index ef572206f1c3..932f5a56d1a6 100644
--- a/arch/arm64/include/asm/efi.h
+++ b/arch/arm64/include/asm/efi.h
@@ -1,8 +1,11 @@
 #ifndef _ASM_EFI_H
 #define _ASM_EFI_H
 
+#include <asm/cpufeature.h>
 #include <asm/io.h>
+#include <asm/mmu_context.h>
 #include <asm/neon.h>
+#include <asm/tlbflush.h>
 
 #ifdef CONFIG_EFI
 extern void efi_init(void);
@@ -10,6 +13,8 @@ extern void efi_init(void);
 #define efi_init()
 #endif
 
+int efi_create_mapping(struct mm_struct *mm, efi_memory_desc_t *md);
+
 #define efi_call_virt(f, ...)						\
 ({									\
 	efi_##f##_t *__f;						\
@@ -63,6 +68,34 @@ extern void efi_init(void);
  *   Services are enabled and the EFI_RUNTIME_SERVICES bit set.
  */
 
+static inline void efi_set_pgd(struct mm_struct *mm)
+{
+	__switch_mm(mm);
+
+	if (system_uses_ttbr0_pan()) {
+		if (mm != current->active_mm) {
+			/*
+			 * Update the current thread's saved ttbr0 since it is
+			 * restored as part of a return from exception. Set
+			 * the hardware TTBR0_EL1 using cpu_switch_mm()
+			 * directly to enable potential errata workarounds.
+			 */
+			update_saved_ttbr0(current, mm);
+			cpu_switch_mm(mm->pgd, mm);
+		} else {
+			/*
+			 * Defer the switch to the current thread's TTBR0_EL1
+			 * until uaccess_enable(). Restore the current
+			 * thread's saved ttbr0 corresponding to its active_mm
+			 * (if different from init_mm).
+			 */
+			cpu_set_reserved_ttbr0();
+			if (current->active_mm != &init_mm)
+				update_saved_ttbr0(current, current->active_mm);
+		}
+	}
+}
+
 void efi_virtmap_load(void);
 void efi_virtmap_unload(void);
 
diff --git a/arch/arm64/include/asm/elf.h b/arch/arm64/include/asm/elf.h
index cd3dfa346649..f9d64ed4fd2b 100644
--- a/arch/arm64/include/asm/elf.h
+++ b/arch/arm64/include/asm/elf.h
@@ -169,7 +169,7 @@ extern int arch_setup_additional_pages(struct linux_binprm *bprm,
 #ifdef CONFIG_COMPAT
 
 /* PIE load location for compat arm. Must match ARM ELF_ET_DYN_BASE. */
-#define COMPAT_ELF_ET_DYN_BASE		0x000400000UL
+#define COMPAT_ELF_ET_DYN_BASE		(2 * TASK_SIZE_32 / 3)
 
 /* AArch32 registers. */
 #define COMPAT_ELF_NGREG		18
diff --git a/arch/arm64/include/asm/esr.h b/arch/arm64/include/asm/esr.h
index 77eeb2cc648f..f772e15c4766 100644
--- a/arch/arm64/include/asm/esr.h
+++ b/arch/arm64/include/asm/esr.h
@@ -74,6 +74,7 @@
 
 #define ESR_ELx_EC_SHIFT	(26)
 #define ESR_ELx_EC_MASK		(UL(0x3F) << ESR_ELx_EC_SHIFT)
+#define ESR_ELx_EC(esr)		(((esr) & ESR_ELx_EC_MASK) >> ESR_ELx_EC_SHIFT)
 
 #define ESR_ELx_IL		(UL(1) << 25)
 #define ESR_ELx_ISS_MASK	(ESR_ELx_IL - 1)
diff --git a/arch/arm64/include/asm/exec.h b/arch/arm64/include/asm/exec.h
index db0563c23482..f7865dd9d868 100644
--- a/arch/arm64/include/asm/exec.h
+++ b/arch/arm64/include/asm/exec.h
@@ -18,6 +18,9 @@
 #ifndef __ASM_EXEC_H
 #define __ASM_EXEC_H
 
+#include <linux/sched.h>
+
 extern unsigned long arch_align_stack(unsigned long sp);
+void uao_thread_switch(struct task_struct *next);
 
 #endif	/* __ASM_EXEC_H */
diff --git a/arch/arm64/include/asm/futex.h b/arch/arm64/include/asm/futex.h
index f2585cdd32c2..85c4a8981d47 100644
--- a/arch/arm64/include/asm/futex.h
+++ b/arch/arm64/include/asm/futex.h
@@ -21,15 +21,12 @@
 #include <linux/futex.h>
 #include <linux/uaccess.h>
 
-#include <asm/alternative.h>
-#include <asm/cpufeature.h>
 #include <asm/errno.h>
-#include <asm/sysreg.h>
 
 #define __futex_atomic_op(insn, ret, oldval, uaddr, tmp, oparg)		\
+do {									\
+	uaccess_enable();						\
 	asm volatile(							\
-	ALTERNATIVE("nop", SET_PSTATE_PAN(0), ARM64_HAS_PAN,		\
-		    CONFIG_ARM64_PAN)					\
 "	prfm	pstl1strm, %2\n"					\
 "1:	ldxr	%w1, %2\n"						\
 	insn "\n"							\
@@ -44,11 +41,11 @@
 "	.popsection\n"							\
 	_ASM_EXTABLE(1b, 4b)						\
 	_ASM_EXTABLE(2b, 4b)						\
-	ALTERNATIVE("nop", SET_PSTATE_PAN(1), ARM64_HAS_PAN,		\
-		    CONFIG_ARM64_PAN)					\
 	: "=&r" (ret), "=&r" (oldval), "+Q" (*uaddr), "=&r" (tmp)	\
 	: "r" (oparg), "Ir" (-EFAULT)					\
-	: "memory")
+	: "memory");							\
+	uaccess_disable();						\
+} while (0)
 
 static inline int
 futex_atomic_op_inuser (int encoded_op, u32 __user *uaddr)
@@ -118,8 +115,8 @@ futex_atomic_cmpxchg_inatomic(u32 *uval, u32 __user *uaddr,
 	if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32)))
 		return -EFAULT;
 
+	uaccess_enable();
 	asm volatile("// futex_atomic_cmpxchg_inatomic\n"
-ALTERNATIVE("nop", SET_PSTATE_PAN(0), ARM64_HAS_PAN, CONFIG_ARM64_PAN)
 "	prfm	pstl1strm, %2\n"
 "1:	ldxr	%w1, %2\n"
 "	sub	%w3, %w1, %w4\n"
@@ -134,10 +131,10 @@ ALTERNATIVE("nop", SET_PSTATE_PAN(0), ARM64_HAS_PAN, CONFIG_ARM64_PAN)
 "	.popsection\n"
 	_ASM_EXTABLE(1b, 4b)
 	_ASM_EXTABLE(2b, 4b)
-ALTERNATIVE("nop", SET_PSTATE_PAN(1), ARM64_HAS_PAN, CONFIG_ARM64_PAN)
 	: "+r" (ret), "=&r" (val), "+Q" (*uaddr), "=&r" (tmp)
 	: "r" (oldval), "r" (newval), "Ir" (-EFAULT)
 	: "memory");
+	uaccess_disable();
 
 	*uval = val;
 	return ret;
diff --git a/arch/arm64/include/asm/hw_breakpoint.h b/arch/arm64/include/asm/hw_breakpoint.h
index 9732908bfc8a..c72b8e201ab4 100644
--- a/arch/arm64/include/asm/hw_breakpoint.h
+++ b/arch/arm64/include/asm/hw_breakpoint.h
@@ -68,7 +68,11 @@ static inline void decode_ctrl_reg(u32 reg,
 /* Lengths */
 #define ARM_BREAKPOINT_LEN_1	0x1
 #define ARM_BREAKPOINT_LEN_2	0x3
+#define ARM_BREAKPOINT_LEN_3	0x7
 #define ARM_BREAKPOINT_LEN_4	0xf
+#define ARM_BREAKPOINT_LEN_5	0x1f
+#define ARM_BREAKPOINT_LEN_6	0x3f
+#define ARM_BREAKPOINT_LEN_7	0x7f
 #define ARM_BREAKPOINT_LEN_8	0xff
 
 /* Kernel stepping */
@@ -110,7 +114,7 @@ struct perf_event;
 struct pmu;
 
 extern int arch_bp_generic_fields(struct arch_hw_breakpoint_ctrl ctrl,
-				  int *gen_len, int *gen_type);
+				  int *gen_len, int *gen_type, int *offset);
 extern int arch_check_bp_in_kernelspace(struct perf_event *bp);
 extern int arch_validate_hwbkpt_settings(struct perf_event *bp);
 extern int hw_breakpoint_exceptions_notify(struct notifier_block *unused,
diff --git a/arch/arm64/include/asm/kernel-pgtable.h b/arch/arm64/include/asm/kernel-pgtable.h
index 5c6375d8528b..7803343e5881 100644
--- a/arch/arm64/include/asm/kernel-pgtable.h
+++ b/arch/arm64/include/asm/kernel-pgtable.h
@@ -19,6 +19,8 @@
 #ifndef __ASM_KERNEL_PGTABLE_H
 #define __ASM_KERNEL_PGTABLE_H
 
+#include <asm/pgtable.h>
+#include <asm/sparsemem.h>
 
 /*
  * The linear mapping and the start of memory are both 2M aligned (per
@@ -53,6 +55,12 @@
 #define SWAPPER_DIR_SIZE	(SWAPPER_PGTABLE_LEVELS * PAGE_SIZE)
 #define IDMAP_DIR_SIZE		(IDMAP_PGTABLE_LEVELS * PAGE_SIZE)
 
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+#define RESERVED_TTBR0_SIZE	(PAGE_SIZE)
+#else
+#define RESERVED_TTBR0_SIZE	(0)
+#endif
+
 /* Initial memory map size */
 #if ARM64_SWAPPER_USES_SECTION_MAPS
 #define SWAPPER_BLOCK_SHIFT	SECTION_SHIFT
@@ -86,10 +94,24 @@
  * (64k granule), or a multiple that can be mapped using contiguous bits
  * in the page tables: 32 * PMD_SIZE (16k granule)
  */
-#ifdef CONFIG_ARM64_64K_PAGES
-#define ARM64_MEMSTART_ALIGN	SZ_512M
+#if defined(CONFIG_ARM64_4K_PAGES)
+#define ARM64_MEMSTART_SHIFT		PUD_SHIFT
+#elif defined(CONFIG_ARM64_16K_PAGES)
+#define ARM64_MEMSTART_SHIFT		(PMD_SHIFT + 5)
+#else
+#define ARM64_MEMSTART_SHIFT		PMD_SHIFT
+#endif
+
+/*
+ * sparsemem vmemmap imposes an additional requirement on the alignment of
+ * memstart_addr, due to the fact that the base of the vmemmap region
+ * has a direct correspondence, and needs to appear sufficiently aligned
+ * in the virtual address space.
+ */
+#if defined(CONFIG_SPARSEMEM_VMEMMAP) && ARM64_MEMSTART_SHIFT < SECTION_SIZE_BITS
+#define ARM64_MEMSTART_ALIGN	(1UL << SECTION_SIZE_BITS)
 #else
-#define ARM64_MEMSTART_ALIGN	SZ_1G
+#define ARM64_MEMSTART_ALIGN	(1UL << ARM64_MEMSTART_SHIFT)
 #endif
 
 #endif	/* __ASM_KERNEL_PGTABLE_H */
diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h
index d776037d199f..ae11e8fdbfd2 100644
--- a/arch/arm64/include/asm/memory.h
+++ b/arch/arm64/include/asm/memory.h
@@ -196,7 +196,11 @@ static inline void *phys_to_virt(phys_addr_t x)
 #define ARCH_PFN_OFFSET		((unsigned long)PHYS_PFN_OFFSET)
 
 #define virt_to_page(kaddr)	pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
-#define	virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+#define _virt_addr_valid(kaddr)	pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
+
+#define _virt_addr_is_linear(kaddr)	(((u64)(kaddr)) >= PAGE_OFFSET)
+#define virt_addr_valid(kaddr)		(_virt_addr_is_linear(kaddr) && \
+					 _virt_addr_valid(kaddr))
 
 #endif
 
diff --git a/arch/arm64/include/asm/mmu_context.h b/arch/arm64/include/asm/mmu_context.h
index a00f7cf35bbd..e53d30c6f779 100644
--- a/arch/arm64/include/asm/mmu_context.h
+++ b/arch/arm64/include/asm/mmu_context.h
@@ -23,6 +23,7 @@
 #include <linux/sched.h>
 
 #include <asm/cacheflush.h>
+#include <asm/cpufeature.h>
 #include <asm/proc-fns.h>
 #include <asm-generic/mm_hooks.h>
 #include <asm/cputype.h>
@@ -113,7 +114,7 @@ static inline void cpu_uninstall_idmap(void)
 	local_flush_tlb_all();
 	cpu_set_default_tcr_t0sz();
 
-	if (mm != &init_mm)
+	if (mm != &init_mm && !system_uses_ttbr0_pan())
 		cpu_switch_mm(mm->pgd, mm);
 }
 
@@ -173,20 +174,26 @@ enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
 {
 }
 
-/*
- * This is the actual mm switch as far as the scheduler
- * is concerned.  No registers are touched.  We avoid
- * calling the CPU specific function when the mm hasn't
- * actually changed.
- */
-static inline void
-switch_mm(struct mm_struct *prev, struct mm_struct *next,
-	  struct task_struct *tsk)
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+static inline void update_saved_ttbr0(struct task_struct *tsk,
+				      struct mm_struct *mm)
 {
-	unsigned int cpu = smp_processor_id();
+	if (system_uses_ttbr0_pan()) {
+		BUG_ON(mm->pgd == swapper_pg_dir);
+		task_thread_info(tsk)->ttbr0 =
+			virt_to_phys(mm->pgd) | ASID(mm) << 48;
+	}
+}
+#else
+static inline void update_saved_ttbr0(struct task_struct *tsk,
+				      struct mm_struct *mm)
+{
+}
+#endif
 
-	if (prev == next)
-		return;
+static inline void __switch_mm(struct mm_struct *next)
+{
+	unsigned int cpu = smp_processor_id();
 
 	/*
 	 * init_mm.pgd does not contain any user mappings and it is always
@@ -200,7 +207,25 @@ switch_mm(struct mm_struct *prev, struct mm_struct *next,
 	check_and_switch_context(next, cpu);
 }
 
+static inline void
+switch_mm(struct mm_struct *prev, struct mm_struct *next,
+	  struct task_struct *tsk)
+{
+	if (prev != next)
+		__switch_mm(next);
+
+	/*
+	 * Update the saved TTBR0_EL1 of the scheduled-in task as the previous
+	 * value may have not been initialised yet (activate_mm caller) or the
+	 * ASID has changed since the last run (following the context switch
+	 * of another thread of the same process). Avoid setting the reserved
+	 * TTBR0_EL1 to swapper_pg_dir (init_mm; e.g. via idle_task_exit).
+	 */
+	if (next != &init_mm)
+		update_saved_ttbr0(tsk, next);
+}
+
 #define deactivate_mm(tsk,mm)	do { } while (0)
-#define activate_mm(prev,next)	switch_mm(prev, next, NULL)
+#define activate_mm(prev,next)	switch_mm(prev, next, current)
 
 #endif
diff --git a/arch/arm64/include/asm/module.h b/arch/arm64/include/asm/module.h
index 06ff7fd9e81f..b6c6fa29fe56 100644
--- a/arch/arm64/include/asm/module.h
+++ b/arch/arm64/include/asm/module.h
@@ -22,14 +22,19 @@
 #define MODULE_ARCH_VERMAGIC	"aarch64"
 
 #ifdef CONFIG_ARM64_MODULE_PLTS
-struct mod_arch_specific {
+struct mod_plt_sec {
 	struct elf64_shdr	*plt;
 	int			plt_num_entries;
 	int			plt_max_entries;
 };
+
+struct mod_arch_specific {
+	struct mod_plt_sec	core;
+	struct mod_plt_sec	init;
+};
 #endif
 
-u64 module_emit_plt_entry(struct module *mod, const Elf64_Rela *rela,
+u64 module_emit_plt_entry(struct module *mod, void *loc, const Elf64_Rela *rela,
 			  Elf64_Sym *sym);
 
 #ifdef CONFIG_RANDOMIZE_BASE
diff --git a/arch/arm64/include/asm/percpu.h b/arch/arm64/include/asm/percpu.h
index 8a336852eeba..2ce1a0262a59 100644
--- a/arch/arm64/include/asm/percpu.h
+++ b/arch/arm64/include/asm/percpu.h
@@ -16,6 +16,8 @@
 #ifndef __ASM_PERCPU_H
 #define __ASM_PERCPU_H
 
+#include <asm/stack_pointer.h>
+
 static inline void set_my_cpu_offset(unsigned long off)
 {
 	asm volatile("msr tpidr_el1, %0" :: "r" (off) : "memory");
diff --git a/arch/arm64/include/asm/perf_event.h b/arch/arm64/include/asm/perf_event.h
index 7bd3cdb533ea..91b6be092ce2 100644
--- a/arch/arm64/include/asm/perf_event.h
+++ b/arch/arm64/include/asm/perf_event.h
@@ -17,6 +17,8 @@
 #ifndef __ASM_PERF_EVENT_H
 #define __ASM_PERF_EVENT_H
 
+#include <asm/stack_pointer.h>
+
 #ifdef CONFIG_PERF_EVENTS
 struct pt_regs;
 extern unsigned long perf_instruction_pointer(struct pt_regs *regs);
diff --git a/arch/arm64/include/asm/pgtable.h b/arch/arm64/include/asm/pgtable.h
index 9a09ccf7122d..49fec68238c7 100644
--- a/arch/arm64/include/asm/pgtable.h
+++ b/arch/arm64/include/asm/pgtable.h
@@ -222,6 +222,16 @@ static inline pte_t pte_mknoncont(pte_t pte)
 	return clear_pte_bit(pte, __pgprot(PTE_CONT));
 }
 
+static inline pte_t pte_clear_rdonly(pte_t pte)
+{
+	return clear_pte_bit(pte, __pgprot(PTE_RDONLY));
+}
+
+static inline pte_t pte_mkpresent(pte_t pte)
+{
+	return set_pte_bit(pte, __pgprot(PTE_VALID));
+}
+
 static inline pmd_t pmd_mkcont(pmd_t pmd)
 {
 	return __pmd(pmd_val(pmd) | PMD_SECT_CONT);
diff --git a/arch/arm64/include/asm/sections.h b/arch/arm64/include/asm/sections.h
new file mode 100644
index 000000000000..4e7e7067afdb
--- /dev/null
+++ b/arch/arm64/include/asm/sections.h
@@ -0,0 +1,30 @@
+/*
+ * Copyright (C) 2016 ARM Limited
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#ifndef __ASM_SECTIONS_H
+#define __ASM_SECTIONS_H
+
+#include <asm-generic/sections.h>
+
+extern char __alt_instructions[], __alt_instructions_end[];
+extern char __exception_text_start[], __exception_text_end[];
+extern char __hibernate_exit_text_start[], __hibernate_exit_text_end[];
+extern char __hyp_idmap_text_start[], __hyp_idmap_text_end[];
+extern char __hyp_text_start[], __hyp_text_end[];
+extern char __idmap_text_start[], __idmap_text_end[];
+extern char __irqentry_text_start[], __irqentry_text_end[];
+extern char __mmuoff_data_start[], __mmuoff_data_end[];
+
+#endif /* __ASM_SECTIONS_H */
diff --git a/arch/arm64/include/asm/signal32.h b/arch/arm64/include/asm/signal32.h
index eeaa97559bab..81abea0b7650 100644
--- a/arch/arm64/include/asm/signal32.h
+++ b/arch/arm64/include/asm/signal32.h
@@ -22,8 +22,6 @@
 
 #define AARCH32_KERN_SIGRET_CODE_OFFSET	0x500
 
-extern const compat_ulong_t aarch32_sigret_code[6];
-
 int compat_setup_frame(int usig, struct ksignal *ksig, sigset_t *set,
 		       struct pt_regs *regs);
 int compat_setup_rt_frame(int usig, struct ksignal *ksig, sigset_t *set,
diff --git a/arch/arm64/include/asm/smp.h b/arch/arm64/include/asm/smp.h
index 1d3ff7e4a6c2..d909746cd421 100644
--- a/arch/arm64/include/asm/smp.h
+++ b/arch/arm64/include/asm/smp.h
@@ -29,11 +29,22 @@
 
 #ifndef __ASSEMBLY__
 
+#include <asm/percpu.h>
+
 #include <linux/threads.h>
 #include <linux/cpumask.h>
 #include <linux/thread_info.h>
 
-#define raw_smp_processor_id() (current_thread_info()->cpu)
+DECLARE_PER_CPU_READ_MOSTLY(int, cpu_number);
+
+/*
+ * We don't use this_cpu_read(cpu_number) as that has implicit writes to
+ * preempt_count, and associated (compiler) barriers, that we'd like to avoid
+ * the expense of. If we're preemptible, the value can be stale at use anyway.
+ * And we can't use this_cpu_ptr() either, as that winds up recursing back
+ * here under CONFIG_DEBUG_PREEMPT=y.
+ */
+#define raw_smp_processor_id() (*raw_cpu_ptr(&cpu_number))
 
 struct seq_file;
 
@@ -73,6 +84,9 @@ asmlinkage void secondary_start_kernel(void);
  */
 struct secondary_data {
 	void *stack;
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	struct task_struct *task;
+#endif
 	long status;
 };
 
diff --git a/arch/arm64/include/asm/stack_pointer.h b/arch/arm64/include/asm/stack_pointer.h
new file mode 100644
index 000000000000..ffcdf742cddf
--- /dev/null
+++ b/arch/arm64/include/asm/stack_pointer.h
@@ -0,0 +1,9 @@
+#ifndef __ASM_STACK_POINTER_H
+#define __ASM_STACK_POINTER_H
+
+/*
+ * how to get the current stack pointer from C
+ */
+register unsigned long current_stack_pointer asm ("sp");
+
+#endif /* __ASM_STACK_POINTER_H */
diff --git a/arch/arm64/include/asm/suspend.h b/arch/arm64/include/asm/suspend.h
index 024d623f662e..92d6a628e478 100644
--- a/arch/arm64/include/asm/suspend.h
+++ b/arch/arm64/include/asm/suspend.h
@@ -1,7 +1,7 @@
 #ifndef __ASM_SUSPEND_H
 #define __ASM_SUSPEND_H
 
-#define NR_CTX_REGS 10
+#define NR_CTX_REGS 12
 #define NR_CALLEE_SAVED_REGS 12
 
 /*
diff --git a/arch/arm64/include/asm/thread_info.h b/arch/arm64/include/asm/thread_info.h
index abd64bd1f6d9..67dd228c3f17 100644
--- a/arch/arm64/include/asm/thread_info.h
+++ b/arch/arm64/include/asm/thread_info.h
@@ -36,22 +36,36 @@
 
 struct task_struct;
 
+#include <asm/stack_pointer.h>
 #include <asm/types.h>
 
 typedef unsigned long mm_segment_t;
 
 /*
  * low level task data that entry.S needs immediate access to.
- * __switch_to() assumes cpu_context follows immediately after cpu_domain.
  */
 struct thread_info {
 	unsigned long		flags;		/* low level flags */
 	mm_segment_t		addr_limit;	/* address limit */
+#ifndef CONFIG_THREAD_INFO_IN_TASK
 	struct task_struct	*task;		/* main task structure */
+#endif
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+	u64			ttbr0;		/* saved TTBR0_EL1 */
+#endif
 	int			preempt_count;	/* 0 => preemptable, <0 => bug */
+#ifndef CONFIG_THREAD_INFO_IN_TASK
 	int			cpu;		/* cpu */
+#endif
 };
 
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+#define INIT_THREAD_INFO(tsk)						\
+{									\
+	.preempt_count	= INIT_PREEMPT_COUNT,				\
+	.addr_limit	= KERNEL_DS,					\
+}
+#else
 #define INIT_THREAD_INFO(tsk)						\
 {									\
 	.task		= &tsk,						\
@@ -60,14 +74,6 @@ struct thread_info {
 	.addr_limit	= KERNEL_DS,					\
 }
 
-#define init_thread_info	(init_thread_union.thread_info)
-#define init_stack		(init_thread_union.stack)
-
-/*
- * how to get the current stack pointer from C
- */
-register unsigned long current_stack_pointer asm ("sp");
-
 /*
  * how to get the thread information struct from C
  */
@@ -85,6 +91,11 @@ static inline struct thread_info *current_thread_info(void)
 	return (struct thread_info *)sp_el0;
 }
 
+#define init_thread_info	(init_thread_union.thread_info)
+#endif
+
+#define init_stack		(init_thread_union.stack)
+
 #define thread_saved_pc(tsk)	\
 	((unsigned long)(tsk->thread.cpu_context.pc))
 #define thread_saved_sp(tsk)	\
diff --git a/arch/arm64/include/asm/topology.h b/arch/arm64/include/asm/topology.h
index a3e9d6fdbf21..bbd362cd1ed1 100644
--- a/arch/arm64/include/asm/topology.h
+++ b/arch/arm64/include/asm/topology.h
@@ -22,6 +22,15 @@ void init_cpu_topology(void);
 void store_cpu_topology(unsigned int cpuid);
 const struct cpumask *cpu_coregroup_mask(int cpu);
 
+struct sched_domain;
+#ifdef CONFIG_CPU_FREQ
+#define arch_scale_freq_capacity cpufreq_scale_freq_capacity
+extern unsigned long cpufreq_scale_freq_capacity(struct sched_domain *sd, int cpu);
+extern unsigned long cpufreq_scale_max_freq_capacity(int cpu);
+#endif
+#define arch_scale_cpu_capacity scale_cpu_capacity
+extern unsigned long scale_cpu_capacity(struct sched_domain *sd, int cpu);
+
 #include <asm-generic/topology.h>
 
 #endif /* _ASM_ARM_TOPOLOGY_H */
diff --git a/arch/arm64/include/asm/traps.h b/arch/arm64/include/asm/traps.h
index 0cc2f29bf9da..ddee92bdb584 100644
--- a/arch/arm64/include/asm/traps.h
+++ b/arch/arm64/include/asm/traps.h
@@ -19,6 +19,7 @@
 #define __ASM_TRAP_H
 
 #include <linux/list.h>
+#include <asm/sections.h>
 
 struct pt_regs;
 
@@ -37,9 +38,6 @@ void unregister_undef_hook(struct undef_hook *hook);
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 static inline int __in_irqentry_text(unsigned long ptr)
 {
-	extern char __irqentry_text_start[];
-	extern char __irqentry_text_end[];
-
 	return ptr >= (unsigned long)&__irqentry_text_start &&
 	       ptr < (unsigned long)&__irqentry_text_end;
 }
@@ -52,8 +50,6 @@ static inline int __in_irqentry_text(unsigned long ptr)
 
 static inline int in_exception_text(unsigned long ptr)
 {
-	extern char __exception_text_start[];
-	extern char __exception_text_end[];
 	int in;
 
 	in = ptr >= (unsigned long)&__exception_text_start &&
diff --git a/arch/arm64/include/asm/uaccess.h b/arch/arm64/include/asm/uaccess.h
index 22d0f6e8623c..064cef9ae2d1 100644
--- a/arch/arm64/include/asm/uaccess.h
+++ b/arch/arm64/include/asm/uaccess.h
@@ -18,6 +18,12 @@
 #ifndef __ASM_UACCESS_H
 #define __ASM_UACCESS_H
 
+#include <asm/alternative.h>
+#include <asm/kernel-pgtable.h>
+#include <asm/sysreg.h>
+
+#ifndef __ASSEMBLY__
+
 /*
  * User space memory access functions
  */
@@ -25,10 +31,8 @@
 #include <linux/string.h>
 #include <linux/thread_info.h>
 
-#include <asm/alternative.h>
 #include <asm/cpufeature.h>
 #include <asm/ptrace.h>
-#include <asm/sysreg.h>
 #include <asm/errno.h>
 #include <asm/memory.h>
 #include <asm/compiler.h>
@@ -133,6 +137,99 @@ static inline void set_fs(mm_segment_t fs)
 	"	.popsection\n"
 
 /*
+ * User access enabling/disabling.
+ */
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+static inline void __uaccess_ttbr0_disable(void)
+{
+	unsigned long ttbr;
+
+	/* reserved_ttbr0 placed at the end of swapper_pg_dir */
+	ttbr = read_sysreg(ttbr1_el1) + SWAPPER_DIR_SIZE;
+	write_sysreg(ttbr, ttbr0_el1);
+	isb();
+}
+
+static inline void __uaccess_ttbr0_enable(void)
+{
+	unsigned long flags;
+
+	/*
+	 * Disable interrupts to avoid preemption between reading the 'ttbr0'
+	 * variable and the MSR. A context switch could trigger an ASID
+	 * roll-over and an update of 'ttbr0'.
+	 */
+	local_irq_save(flags);
+	write_sysreg(current_thread_info()->ttbr0, ttbr0_el1);
+	isb();
+	local_irq_restore(flags);
+}
+
+static inline bool uaccess_ttbr0_disable(void)
+{
+	if (!system_uses_ttbr0_pan())
+		return false;
+	__uaccess_ttbr0_disable();
+	return true;
+}
+
+static inline bool uaccess_ttbr0_enable(void)
+{
+	if (!system_uses_ttbr0_pan())
+		return false;
+	__uaccess_ttbr0_enable();
+	return true;
+}
+#else
+static inline bool uaccess_ttbr0_disable(void)
+{
+	return false;
+}
+
+static inline bool uaccess_ttbr0_enable(void)
+{
+	return false;
+}
+#endif
+
+#define __uaccess_disable(alt)						\
+do {									\
+	if (!uaccess_ttbr0_disable())					\
+		asm(ALTERNATIVE("nop", SET_PSTATE_PAN(1), alt,		\
+				CONFIG_ARM64_PAN));			\
+} while (0)
+
+#define __uaccess_enable(alt)						\
+do {									\
+	if (!uaccess_ttbr0_enable())					\
+		asm(ALTERNATIVE("nop", SET_PSTATE_PAN(0), alt,		\
+				CONFIG_ARM64_PAN));			\
+} while (0)
+
+static inline void uaccess_disable(void)
+{
+	__uaccess_disable(ARM64_HAS_PAN);
+}
+
+static inline void uaccess_enable(void)
+{
+	__uaccess_enable(ARM64_HAS_PAN);
+}
+
+/*
+ * These functions are no-ops when UAO is present.
+ */
+static inline void uaccess_disable_not_uao(void)
+{
+	__uaccess_disable(ARM64_ALT_PAN_NOT_UAO);
+}
+
+static inline void uaccess_enable_not_uao(void)
+{
+	__uaccess_enable(ARM64_ALT_PAN_NOT_UAO);
+}
+
+/*
  * The "__xxx" versions of the user access functions do not verify the address
  * space - it must have been done previously with a separate "access_ok()"
  * call.
@@ -159,8 +256,7 @@ static inline void set_fs(mm_segment_t fs)
 do {									\
 	unsigned long __gu_val;						\
 	__chk_user_ptr(ptr);						\
-	asm(ALTERNATIVE("nop", SET_PSTATE_PAN(0), ARM64_ALT_PAN_NOT_UAO,\
-			CONFIG_ARM64_PAN));				\
+	uaccess_enable_not_uao();					\
 	switch (sizeof(*(ptr))) {					\
 	case 1:								\
 		__get_user_asm("ldrb", "ldtrb", "%w", __gu_val, (ptr),  \
@@ -181,9 +277,8 @@ do {									\
 	default:							\
 		BUILD_BUG();						\
 	}								\
+	uaccess_disable_not_uao();					\
 	(x) = (__force __typeof__(*(ptr)))__gu_val;			\
-	asm(ALTERNATIVE("nop", SET_PSTATE_PAN(1), ARM64_ALT_PAN_NOT_UAO,\
-			CONFIG_ARM64_PAN));				\
 } while (0)
 
 #define __get_user(x, ptr)						\
@@ -228,8 +323,7 @@ do {									\
 do {									\
 	__typeof__(*(ptr)) __pu_val = (x);				\
 	__chk_user_ptr(ptr);						\
-	asm(ALTERNATIVE("nop", SET_PSTATE_PAN(0), ARM64_ALT_PAN_NOT_UAO,\
-			CONFIG_ARM64_PAN));				\
+	uaccess_enable_not_uao();					\
 	switch (sizeof(*(ptr))) {					\
 	case 1:								\
 		__put_user_asm("strb", "sttrb", "%w", __pu_val, (ptr),	\
@@ -250,8 +344,7 @@ do {									\
 	default:							\
 		BUILD_BUG();						\
 	}								\
-	asm(ALTERNATIVE("nop", SET_PSTATE_PAN(1), ARM64_ALT_PAN_NOT_UAO,\
-			CONFIG_ARM64_PAN));				\
+	uaccess_disable_not_uao();					\
 } while (0)
 
 #define __put_user(x, ptr)						\
@@ -336,4 +429,66 @@ extern long strncpy_from_user(char *dest, const char __user *src, long count);
 extern __must_check long strlen_user(const char __user *str);
 extern __must_check long strnlen_user(const char __user *str, long n);
 
+#else	/* __ASSEMBLY__ */
+
+#include <asm/assembler.h>
+
+/*
+ * User access enabling/disabling macros.
+ */
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+	.macro	__uaccess_ttbr0_disable, tmp1
+	mrs	\tmp1, ttbr1_el1		// swapper_pg_dir
+	add	\tmp1, \tmp1, #SWAPPER_DIR_SIZE	// reserved_ttbr0 at the end of swapper_pg_dir
+	msr	ttbr0_el1, \tmp1		// set reserved TTBR0_EL1
+	isb
+	.endm
+
+	.macro	__uaccess_ttbr0_enable, tmp1
+	get_thread_info \tmp1
+	ldr	\tmp1, [\tmp1, #TSK_TI_TTBR0]	// load saved TTBR0_EL1
+	msr	ttbr0_el1, \tmp1		// set the non-PAN TTBR0_EL1
+	isb
+	.endm
+
+	.macro	uaccess_ttbr0_disable, tmp1
+alternative_if_not ARM64_HAS_PAN
+	__uaccess_ttbr0_disable \tmp1
+alternative_else_nop_endif
+	.endm
+
+	.macro	uaccess_ttbr0_enable, tmp1, tmp2
+alternative_if_not ARM64_HAS_PAN
+	save_and_disable_irq \tmp2		// avoid preemption
+	__uaccess_ttbr0_enable \tmp1
+	restore_irq \tmp2
+alternative_else_nop_endif
+	.endm
+#else
+	.macro	uaccess_ttbr0_disable, tmp1
+	.endm
+
+	.macro	uaccess_ttbr0_enable, tmp1, tmp2
+	.endm
+#endif
+
+/*
+ * These macros are no-ops when UAO is present.
+ */
+	.macro	uaccess_disable_not_uao, tmp1
+	uaccess_ttbr0_disable \tmp1
+alternative_if ARM64_ALT_PAN_NOT_UAO
+	SET_PSTATE_PAN(1)
+alternative_else_nop_endif
+	.endm
+
+	.macro	uaccess_enable_not_uao, tmp1, tmp2
+	uaccess_ttbr0_enable \tmp1, \tmp2
+alternative_if ARM64_ALT_PAN_NOT_UAO
+	SET_PSTATE_PAN(0)
+alternative_else_nop_endif
+	.endm
+
+#endif	/* __ASSEMBLY__ */
+
 #endif /* __ASM_UACCESS_H */
diff --git a/arch/arm64/include/asm/vdso_datapage.h b/arch/arm64/include/asm/vdso_datapage.h
index de66199673d7..2b9a63771eda 100644
--- a/arch/arm64/include/asm/vdso_datapage.h
+++ b/arch/arm64/include/asm/vdso_datapage.h
@@ -22,6 +22,8 @@
 
 struct vdso_data {
 	__u64 cs_cycle_last;	/* Timebase at clocksource init */
+	__u64 raw_time_sec;	/* Raw time */
+	__u64 raw_time_nsec;
 	__u64 xtime_clock_sec;	/* Kernel time */
 	__u64 xtime_clock_nsec;
 	__u64 xtime_coarse_sec;	/* Coarse time */
@@ -29,8 +31,10 @@ struct vdso_data {
 	__u64 wtm_clock_sec;	/* Wall to monotonic time */
 	__u64 wtm_clock_nsec;
 	__u32 tb_seq_count;	/* Timebase sequence counter */
-	__u32 cs_mult;		/* Clocksource multiplier */
-	__u32 cs_shift;		/* Clocksource shift */
+	/* cs_* members must be adjacent and in this order (ldp accesses) */
+	__u32 cs_mono_mult;	/* NTP-adjusted clocksource multiplier */
+	__u32 cs_shift;		/* Clocksource shift (mono = raw) */
+	__u32 cs_raw_mult;	/* Raw clocksource multiplier */
 	__u32 tz_minuteswest;	/* Whacky timezone stuff */
 	__u32 tz_dsttime;
 	__u32 use_syscall;
diff --git a/arch/arm64/include/asm/virt.h b/arch/arm64/include/asm/virt.h
index e6c27b8ce311..db5739413677 100644
--- a/arch/arm64/include/asm/virt.h
+++ b/arch/arm64/include/asm/virt.h
@@ -45,6 +45,7 @@
 #ifndef __ASSEMBLY__
 
 #include <asm/ptrace.h>
+#include <asm/sections.h>
 
 /*
  * __boot_cpu_mode records what mode CPUs were booted in.
@@ -81,9 +82,11 @@ static inline bool is_kernel_in_hyp_mode(void)
 	return el == CurrentEL_EL2;
 }
 
-/* The section containing the hypervisor text */
-extern char __hyp_text_start[];
-extern char __hyp_text_end[];
+#ifdef CONFIG_ARM64_VHE
+extern void verify_cpu_run_el(void);
+#else
+static inline void verify_cpu_run_el(void) {}
+#endif
 
 #endif /* __ASSEMBLY__ */
 
diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile
index 96ae961ada72..01c0b3881f88 100644
--- a/arch/arm64/kernel/Makefile
+++ b/arch/arm64/kernel/Makefile
@@ -52,7 +52,3 @@ obj-y					+= $(arm64-obj-y) vdso/ probes/
 obj-m					+= $(arm64-obj-m)
 head-y					:= head.o
 extra-y					+= $(head-y) vmlinux.lds
-
-# vDSO - this must be built first to generate the symbol offsets
-$(call objectify,$(arm64-obj-y)): $(obj)/vdso/vdso-offsets.h
-$(obj)/vdso/vdso-offsets.h: $(obj)/vdso
diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
index d2ee1b21a10d..4434dabde898 100644
--- a/arch/arm64/kernel/alternative.c
+++ b/arch/arm64/kernel/alternative.c
@@ -25,14 +25,13 @@
 #include <asm/alternative.h>
 #include <asm/cpufeature.h>
 #include <asm/insn.h>
+#include <asm/sections.h>
 #include <linux/stop_machine.h>
 
 #define __ALT_PTR(a,f)		(u32 *)((void *)&(a)->f + (a)->f)
 #define ALT_ORIG_PTR(a)		__ALT_PTR(a, orig_offset)
 #define ALT_REPL_PTR(a)		__ALT_PTR(a, alt_offset)
 
-extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
-
 struct alt_region {
 	struct alt_instr *begin;
 	struct alt_instr *end;
@@ -124,8 +123,8 @@ static int __apply_alternatives_multi_stop(void *unused)
 {
 	static int patched = 0;
 	struct alt_region region = {
-		.begin	= __alt_instructions,
-		.end	= __alt_instructions_end,
+		.begin	= (struct alt_instr *)__alt_instructions,
+		.end	= (struct alt_instr *)__alt_instructions_end,
 	};
 
 	/* We always have a CPU 0 at this point (__init) */
diff --git a/arch/arm64/kernel/armv8_deprecated.c b/arch/arm64/kernel/armv8_deprecated.c
index d03e311b6f3c..10d3642deb7c 100644
--- a/arch/arm64/kernel/armv8_deprecated.c
+++ b/arch/arm64/kernel/armv8_deprecated.c
@@ -14,7 +14,6 @@
 #include <linux/slab.h>
 #include <linux/sysctl.h>
 
-#include <asm/alternative.h>
 #include <asm/cpufeature.h>
 #include <asm/insn.h>
 #include <asm/opcodes.h>
@@ -281,9 +280,9 @@ static void __init register_insn_emulation_sysctl(struct ctl_table *table)
  * Error-checking SWP macros implemented using ldxr{b}/stxr{b}
  */
 #define __user_swpX_asm(data, addr, res, temp, B)		\
+do {								\
+	uaccess_enable();					\
 	__asm__ __volatile__(					\
-	ALTERNATIVE("nop", SET_PSTATE_PAN(0), ARM64_HAS_PAN,	\
-		    CONFIG_ARM64_PAN)				\
 	"0:	ldxr"B"		%w2, [%3]\n"			\
 	"1:	stxr"B"		%w0, %w1, [%3]\n"		\
 	"	cbz		%w0, 2f\n"			\
@@ -299,12 +298,12 @@ static void __init register_insn_emulation_sysctl(struct ctl_table *table)
 	"	.popsection"					\
 	_ASM_EXTABLE(0b, 4b)					\
 	_ASM_EXTABLE(1b, 4b)					\
-	ALTERNATIVE("nop", SET_PSTATE_PAN(1), ARM64_HAS_PAN,	\
-		CONFIG_ARM64_PAN)				\
 	: "=&r" (res), "+r" (data), "=&r" (temp)		\
 	: "r" ((unsigned long)addr), "i" (-EAGAIN),		\
 	  "i" (-EFAULT)						\
-	: "memory")
+	: "memory");						\
+	uaccess_disable();					\
+} while (0)
 
 #define __user_swp_asm(data, addr, res, temp) \
 	__user_swpX_asm(data, addr, res, temp, "")
diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c
index a506ddc270c0..67ebe708e30c 100644
--- a/arch/arm64/kernel/asm-offsets.c
+++ b/arch/arm64/kernel/asm-offsets.c
@@ -35,11 +35,19 @@ int main(void)
 {
   DEFINE(TSK_ACTIVE_MM,		offsetof(struct task_struct, active_mm));
   BLANK();
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+  DEFINE(TSK_TI_FLAGS,		offsetof(struct task_struct, thread_info.flags));
+  DEFINE(TSK_TI_PREEMPT,	offsetof(struct task_struct, thread_info.preempt_count));
+  DEFINE(TSK_TI_ADDR_LIMIT,	offsetof(struct task_struct, thread_info.addr_limit));
+  DEFINE(TSK_STACK,		offsetof(struct task_struct, stack));
+#else
   DEFINE(TI_FLAGS,		offsetof(struct thread_info, flags));
   DEFINE(TI_PREEMPT,		offsetof(struct thread_info, preempt_count));
   DEFINE(TI_ADDR_LIMIT,		offsetof(struct thread_info, addr_limit));
-  DEFINE(TI_TASK,		offsetof(struct thread_info, task));
-  DEFINE(TI_CPU,		offsetof(struct thread_info, cpu));
+#endif
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+  DEFINE(TSK_TI_TTBR0,		offsetof(struct thread_info, ttbr0));
+#endif
   BLANK();
   DEFINE(THREAD_CPU_CONTEXT,	offsetof(struct task_struct, thread.cpu_context));
   BLANK();
@@ -89,6 +97,7 @@ int main(void)
   BLANK();
   DEFINE(CLOCK_REALTIME,	CLOCK_REALTIME);
   DEFINE(CLOCK_MONOTONIC,	CLOCK_MONOTONIC);
+  DEFINE(CLOCK_MONOTONIC_RAW,	CLOCK_MONOTONIC_RAW);
   DEFINE(CLOCK_REALTIME_RES,	MONOTONIC_RES_NSEC);
   DEFINE(CLOCK_REALTIME_COARSE,	CLOCK_REALTIME_COARSE);
   DEFINE(CLOCK_MONOTONIC_COARSE,CLOCK_MONOTONIC_COARSE);
@@ -96,6 +105,8 @@ int main(void)
   DEFINE(NSEC_PER_SEC,		NSEC_PER_SEC);
   BLANK();
   DEFINE(VDSO_CS_CYCLE_LAST,	offsetof(struct vdso_data, cs_cycle_last));
+  DEFINE(VDSO_RAW_TIME_SEC,	offsetof(struct vdso_data, raw_time_sec));
+  DEFINE(VDSO_RAW_TIME_NSEC,	offsetof(struct vdso_data, raw_time_nsec));
   DEFINE(VDSO_XTIME_CLK_SEC,	offsetof(struct vdso_data, xtime_clock_sec));
   DEFINE(VDSO_XTIME_CLK_NSEC,	offsetof(struct vdso_data, xtime_clock_nsec));
   DEFINE(VDSO_XTIME_CRS_SEC,	offsetof(struct vdso_data, xtime_coarse_sec));
@@ -103,7 +114,8 @@ int main(void)
   DEFINE(VDSO_WTM_CLK_SEC,	offsetof(struct vdso_data, wtm_clock_sec));
   DEFINE(VDSO_WTM_CLK_NSEC,	offsetof(struct vdso_data, wtm_clock_nsec));
   DEFINE(VDSO_TB_SEQ_COUNT,	offsetof(struct vdso_data, tb_seq_count));
-  DEFINE(VDSO_CS_MULT,		offsetof(struct vdso_data, cs_mult));
+  DEFINE(VDSO_CS_MONO_MULT,	offsetof(struct vdso_data, cs_mono_mult));
+  DEFINE(VDSO_CS_RAW_MULT,	offsetof(struct vdso_data, cs_raw_mult));
   DEFINE(VDSO_CS_SHIFT,		offsetof(struct vdso_data, cs_shift));
   DEFINE(VDSO_TZ_MINWEST,	offsetof(struct vdso_data, tz_minuteswest));
   DEFINE(VDSO_TZ_DSTTIME,	offsetof(struct vdso_data, tz_dsttime));
@@ -119,6 +131,11 @@ int main(void)
   BLANK();
   DEFINE(CPU_BOOT_STACK,	offsetof(struct secondary_data, stack));
   BLANK();
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+  DEFINE(CPU_BOOT_STACK,	offsetof(struct secondary_data, stack));
+  DEFINE(CPU_BOOT_TASK,		offsetof(struct secondary_data, task));
+  BLANK();
+#endif
 #ifdef CONFIG_KVM_ARM_HOST
   DEFINE(VCPU_CONTEXT,		offsetof(struct kvm_vcpu, arch.ctxt));
   DEFINE(CPU_GP_REGS,		offsetof(struct kvm_cpu_context, gp_regs));
diff --git a/arch/arm64/kernel/cpufeature.c b/arch/arm64/kernel/cpufeature.c
index 476fb5caa361..53fab76d3c39 100644
--- a/arch/arm64/kernel/cpufeature.c
+++ b/arch/arm64/kernel/cpufeature.c
@@ -46,6 +46,7 @@ unsigned int compat_elf_hwcap2 __read_mostly;
 #endif
 
 DECLARE_BITMAP(cpu_hwcaps, ARM64_NCAPS);
+EXPORT_SYMBOL(cpu_hwcaps);
 
 #define __ARM64_FTR_BITS(SIGNED, STRICT, TYPE, SHIFT, WIDTH, SAFE_VAL) \
 	{						\
diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c
index 40270c1fa947..b60da856fc19 100644
--- a/arch/arm64/kernel/efi.c
+++ b/arch/arm64/kernel/efi.c
@@ -11,317 +11,34 @@
  *
  */
 
-#include <linux/atomic.h>
 #include <linux/dmi.h>
 #include <linux/efi.h>
-#include <linux/export.h>
-#include <linux/memblock.h>
-#include <linux/mm_types.h>
-#include <linux/bootmem.h>
-#include <linux/of.h>
-#include <linux/of_fdt.h>
-#include <linux/preempt.h>
-#include <linux/rbtree.h>
-#include <linux/rwsem.h>
-#include <linux/sched.h>
-#include <linux/slab.h>
-#include <linux/spinlock.h>
+#include <linux/init.h>
 
-#include <asm/cacheflush.h>
 #include <asm/efi.h>
-#include <asm/tlbflush.h>
-#include <asm/mmu_context.h>
-#include <asm/mmu.h>
-#include <asm/pgtable.h>
 
-struct efi_memory_map memmap;
-
-static u64 efi_system_table;
-
-static pgd_t efi_pgd[PTRS_PER_PGD] __page_aligned_bss;
-
-static struct mm_struct efi_mm = {
-	.mm_rb			= RB_ROOT,
-	.pgd			= efi_pgd,
-	.mm_users		= ATOMIC_INIT(2),
-	.mm_count		= ATOMIC_INIT(1),
-	.mmap_sem		= __RWSEM_INITIALIZER(efi_mm.mmap_sem),
-	.page_table_lock	= __SPIN_LOCK_UNLOCKED(efi_mm.page_table_lock),
-	.mmlist			= LIST_HEAD_INIT(efi_mm.mmlist),
-};
-
-static int __init is_normal_ram(efi_memory_desc_t *md)
-{
-	if (md->attribute & EFI_MEMORY_WB)
-		return 1;
-	return 0;
-}
-
-/*
- * Translate a EFI virtual address into a physical address: this is necessary,
- * as some data members of the EFI system table are virtually remapped after
- * SetVirtualAddressMap() has been called.
- */
-static phys_addr_t efi_to_phys(unsigned long addr)
+int __init efi_create_mapping(struct mm_struct *mm, efi_memory_desc_t *md)
 {
-	efi_memory_desc_t *md;
-
-	for_each_efi_memory_desc(&memmap, md) {
-		if (!(md->attribute & EFI_MEMORY_RUNTIME))
-			continue;
-		if (md->virt_addr == 0)
-			/* no virtual mapping has been installed by the stub */
-			break;
-		if (md->virt_addr <= addr &&
-		    (addr - md->virt_addr) < (md->num_pages << EFI_PAGE_SHIFT))
-			return md->phys_addr + addr - md->virt_addr;
-	}
-	return addr;
-}
-
-static int __init uefi_init(void)
-{
-	efi_char16_t *c16;
-	void *config_tables;
-	u64 table_size;
-	char vendor[100] = "unknown";
-	int i, retval;
-
-	efi.systab = early_memremap(efi_system_table,
-				    sizeof(efi_system_table_t));
-	if (efi.systab == NULL) {
-		pr_warn("Unable to map EFI system table.\n");
-		return -ENOMEM;
-	}
-
-	set_bit(EFI_BOOT, &efi.flags);
-	set_bit(EFI_64BIT, &efi.flags);
+	pteval_t prot_val;
 
 	/*
-	 * Verify the EFI Table
+	 * Only regions of type EFI_RUNTIME_SERVICES_CODE need to be
+	 * executable, everything else can be mapped with the XN bits
+	 * set.
 	 */
-	if (efi.systab->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE) {
-		pr_err("System table signature incorrect\n");
-		retval = -EINVAL;
-		goto out;
-	}
-	if ((efi.systab->hdr.revision >> 16) < 2)
-		pr_warn("Warning: EFI system table version %d.%02d, expected 2.00 or greater\n",
-			efi.systab->hdr.revision >> 16,
-			efi.systab->hdr.revision & 0xffff);
-
-	/* Show what we know for posterity */
-	c16 = early_memremap(efi_to_phys(efi.systab->fw_vendor),
-			     sizeof(vendor) * sizeof(efi_char16_t));
-	if (c16) {
-		for (i = 0; i < (int) sizeof(vendor) - 1 && *c16; ++i)
-			vendor[i] = c16[i];
-		vendor[i] = '\0';
-		early_memunmap(c16, sizeof(vendor) * sizeof(efi_char16_t));
-	}
-
-	pr_info("EFI v%u.%.02u by %s\n",
-		efi.systab->hdr.revision >> 16,
-		efi.systab->hdr.revision & 0xffff, vendor);
-
-	table_size = sizeof(efi_config_table_64_t) * efi.systab->nr_tables;
-	config_tables = early_memremap(efi_to_phys(efi.systab->tables),
-				       table_size);
-	if (config_tables == NULL) {
-		pr_warn("Unable to map EFI config table array.\n");
-		retval = -ENOMEM;
-		goto out;
-	}
-	retval = efi_config_parse_tables(config_tables, efi.systab->nr_tables,
-					 sizeof(efi_config_table_64_t), NULL);
-
-	early_memunmap(config_tables, table_size);
-out:
-	early_memunmap(efi.systab,  sizeof(efi_system_table_t));
-	return retval;
-}
-
-/*
- * Return true for RAM regions we want to permanently reserve.
- */
-static __init int is_reserve_region(efi_memory_desc_t *md)
-{
-	switch (md->type) {
-	case EFI_LOADER_CODE:
-	case EFI_LOADER_DATA:
-	case EFI_BOOT_SERVICES_CODE:
-	case EFI_BOOT_SERVICES_DATA:
-	case EFI_CONVENTIONAL_MEMORY:
-	case EFI_PERSISTENT_MEMORY:
-		return 0;
-	default:
-		break;
-	}
-	return is_normal_ram(md);
-}
-
-static __init void reserve_regions(void)
-{
-	efi_memory_desc_t *md;
-	u64 paddr, npages, size;
-
-	if (efi_enabled(EFI_DBG))
-		pr_info("Processing EFI memory map:\n");
-
-	for_each_efi_memory_desc(&memmap, md) {
-		paddr = md->phys_addr;
-		npages = md->num_pages;
-
-		if (efi_enabled(EFI_DBG)) {
-			char buf[64];
-
-			pr_info("  0x%012llx-0x%012llx %s",
-				paddr, paddr + (npages << EFI_PAGE_SHIFT) - 1,
-				efi_md_typeattr_format(buf, sizeof(buf), md));
-		}
-
-		memrange_efi_to_native(&paddr, &npages);
-		size = npages << PAGE_SHIFT;
-
-		if (is_normal_ram(md))
-			early_init_dt_add_memory_arch(paddr, size);
-
-		if (is_reserve_region(md)) {
-			memblock_reserve(paddr, size);
-			if (efi_enabled(EFI_DBG))
-				pr_cont("*");
-		}
-
-		if (efi_enabled(EFI_DBG))
-			pr_cont("\n");
-	}
-
-	set_bit(EFI_MEMMAP, &efi.flags);
-}
-
-void __init efi_init(void)
-{
-	struct efi_fdt_params params;
-
-	/* Grab UEFI information placed in FDT by stub */
-	if (!efi_get_fdt_params(&params))
-		return;
-
-	efi_system_table = params.system_table;
-
-	memblock_reserve(params.mmap & PAGE_MASK,
-			 PAGE_ALIGN(params.mmap_size + (params.mmap & ~PAGE_MASK)));
-	memmap.phys_map = params.mmap;
-	memmap.map = early_memremap(params.mmap, params.mmap_size);
-	if (memmap.map == NULL) {
-		/*
-		* If we are booting via UEFI, the UEFI memory map is the only
-		* description of memory we have, so there is little point in
-		* proceeding if we cannot access it.
-		*/
-		panic("Unable to map EFI memory map.\n");
-	}
-	memmap.map_end = memmap.map + params.mmap_size;
-	memmap.desc_size = params.desc_size;
-	memmap.desc_version = params.desc_ver;
-
-	if (uefi_init() < 0)
-		return;
-
-	reserve_regions();
-	early_memunmap(memmap.map, params.mmap_size);
-}
-
-static bool __init efi_virtmap_init(void)
-{
-	efi_memory_desc_t *md;
-
-	init_new_context(NULL, &efi_mm);
-
-	for_each_efi_memory_desc(&memmap, md) {
-		pgprot_t prot;
-
-		if (!(md->attribute & EFI_MEMORY_RUNTIME))
-			continue;
-		if (md->virt_addr == 0)
-			return false;
-
-		pr_info("  EFI remap 0x%016llx => %p\n",
-			md->phys_addr, (void *)md->virt_addr);
-
-		/*
-		 * Only regions of type EFI_RUNTIME_SERVICES_CODE need to be
-		 * executable, everything else can be mapped with the XN bits
-		 * set.
-		 */
-		if (!is_normal_ram(md))
-			prot = __pgprot(PROT_DEVICE_nGnRE);
-		else if (md->type == EFI_RUNTIME_SERVICES_CODE ||
-			 !PAGE_ALIGNED(md->phys_addr))
-			prot = PAGE_KERNEL_EXEC;
-		else
-			prot = PAGE_KERNEL;
-
-		create_pgd_mapping(&efi_mm, md->phys_addr, md->virt_addr,
-				   md->num_pages << EFI_PAGE_SHIFT, 
-				   __pgprot(pgprot_val(prot) | PTE_NG), true);
-	}
-	return true;
-}
-
-/*
- * Enable the UEFI Runtime Services if all prerequisites are in place, i.e.,
- * non-early mapping of the UEFI system table and virtual mappings for all
- * EFI_MEMORY_RUNTIME regions.
- */
-static int __init arm64_enable_runtime_services(void)
-{
-	u64 mapsize;
-
-	if (!efi_enabled(EFI_BOOT)) {
-		pr_info("EFI services will not be available.\n");
-		return 0;
-	}
-
-	if (efi_runtime_disabled()) {
-		pr_info("EFI runtime services will be disabled.\n");
-		return 0;
-	}
-
-	pr_info("Remapping and enabling EFI services.\n");
-
-	mapsize = memmap.map_end - memmap.map;
-	memmap.map = (__force void *)ioremap_cache(memmap.phys_map,
-						   mapsize);
-	if (!memmap.map) {
-		pr_err("Failed to remap EFI memory map\n");
-		return -ENOMEM;
-	}
-	memmap.map_end = memmap.map + mapsize;
-	efi.memmap = &memmap;
-
-	efi.systab = (__force void *)ioremap_cache(efi_system_table,
-						   sizeof(efi_system_table_t));
-	if (!efi.systab) {
-		pr_err("Failed to remap EFI System Table\n");
-		return -ENOMEM;
-	}
-	set_bit(EFI_SYSTEM_TABLES, &efi.flags);
-
-	if (!efi_virtmap_init()) {
-		pr_err("No UEFI virtual mapping was installed -- runtime services will not be available\n");
-		return -ENOMEM;
-	}
-
-	/* Set up runtime services function pointers */
-	efi_native_runtime_setup();
-	set_bit(EFI_RUNTIME_SERVICES, &efi.flags);
-
-	efi.runtime_version = efi.systab->hdr.revision;
-
+	if ((md->attribute & EFI_MEMORY_WB) == 0)
+		prot_val = PROT_DEVICE_nGnRE;
+	else if (md->type == EFI_RUNTIME_SERVICES_CODE ||
+		 !PAGE_ALIGNED(md->phys_addr))
+		prot_val = pgprot_val(PAGE_KERNEL_EXEC);
+	else
+		prot_val = pgprot_val(PAGE_KERNEL);
+
+	create_pgd_mapping(mm, md->phys_addr, md->virt_addr,
+			   md->num_pages << EFI_PAGE_SHIFT,
+			   __pgprot(prot_val | PTE_NG), true);
 	return 0;
 }
-early_initcall(arm64_enable_runtime_services);
 
 static int __init arm64_dmi_init(void)
 {
@@ -337,23 +54,6 @@ static int __init arm64_dmi_init(void)
 }
 core_initcall(arm64_dmi_init);
 
-static void efi_set_pgd(struct mm_struct *mm)
-{
-	switch_mm(NULL, mm, NULL);
-}
-
-void efi_virtmap_load(void)
-{
-	preempt_disable();
-	efi_set_pgd(&efi_mm);
-}
-
-void efi_virtmap_unload(void)
-{
-	efi_set_pgd(current->active_mm);
-	preempt_enable();
-}
-
 /*
  * UpdateCapsule() depends on the system being shutdown via
  * ResetSystem().
diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
index 6f69650c8b7b..b5119f7f5cd6 100644
--- a/arch/arm64/kernel/entry.S
+++ b/arch/arm64/kernel/entry.S
@@ -29,7 +29,9 @@
 #include <asm/esr.h>
 #include <asm/irq.h>
 #include <asm/memory.h>
+#include <asm/ptrace.h>
 #include <asm/thread_info.h>
+#include <asm/uaccess.h>
 #include <asm/asm-uaccess.h>
 #include <asm/unistd.h>
 
@@ -91,9 +93,14 @@
 
 	.if	\el == 0
 	mrs	x21, sp_el0
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr_this_cpu	tsk, __entry_task, x20	// Ensure MDSCR_EL1.SS is clear,
+	ldr	x19, [tsk, #TSK_TI_FLAGS]	// since we can unmask debug
+#else
 	mov	tsk, sp
 	and	tsk, tsk, #~(THREAD_SIZE - 1)	// Ensure MDSCR_EL1.SS is clear,
 	ldr	x19, [tsk, #TI_FLAGS]		// since we can unmask debug
+#endif
 	disable_step_tsk x19, x20		// exceptions when scheduling.
 
 	mov	x29, xzr			// fp pointed to user-space
@@ -101,14 +108,48 @@
 	add	x21, sp, #S_FRAME_SIZE
 	get_thread_info tsk
 	/* Save the task's original addr_limit and set USER_DS (TASK_SIZE_64) */
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	x20, [tsk, #TSK_TI_ADDR_LIMIT]
+#else
 	ldr	x20, [tsk, #TI_ADDR_LIMIT]
+#endif
 	str	x20, [sp, #S_ORIG_ADDR_LIMIT]
 	mov	x20, #TASK_SIZE_64
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	str	x20, [tsk, #TSK_TI_ADDR_LIMIT]
+#else
 	str	x20, [tsk, #TI_ADDR_LIMIT]
+#endif
 	.endif /* \el == 0 */
 	mrs	x22, elr_el1
 	mrs	x23, spsr_el1
 	stp	lr, x21, [sp, #S_LR]
+
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+	/*
+	 * Set the TTBR0 PAN bit in SPSR. When the exception is taken from
+	 * EL0, there is no need to check the state of TTBR0_EL1 since
+	 * accesses are always enabled.
+	 * Note that the meaning of this bit differs from the ARMv8.1 PAN
+	 * feature as all TTBR0_EL1 accesses are disabled, not just those to
+	 * user mappings.
+	 */
+alternative_if ARM64_HAS_PAN
+	b	1f				// skip TTBR0 PAN
+alternative_else_nop_endif
+
+	.if	\el != 0
+	mrs	x21, ttbr0_el1
+	tst	x21, #0xffff << 48		// Check for the reserved ASID
+	orr	x23, x23, #PSR_PAN_BIT		// Set the emulated PAN in the saved SPSR
+	b.eq	1f				// TTBR0 access already disabled
+	and	x23, x23, #~PSR_PAN_BIT		// Clear the emulated PAN in the saved SPSR
+	.endif
+
+	__uaccess_ttbr0_disable x21
+1:
+#endif
+
 	stp	x22, x23, [sp, #S_PC]
 
 	/*
@@ -139,12 +180,50 @@
 	.if	\el != 0
 	/* Restore the task's original addr_limit. */
 	ldr	x20, [sp, #S_ORIG_ADDR_LIMIT]
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	str	x20, [tsk, #TSK_TI_ADDR_LIMIT]
+#else
 	str	x20, [tsk, #TI_ADDR_LIMIT]
+#endif
 	.endif
 
 	ldp	x21, x22, [sp, #S_PC]		// load ELR, SPSR
 	.if	\el == 0
 	ct_user_enter
+	.endif
+
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+	/*
+	 * Restore access to TTBR0_EL1. If returning to EL0, no need for SPSR
+	 * PAN bit checking.
+	 */
+alternative_if ARM64_HAS_PAN
+	b	2f				// skip TTBR0 PAN
+alternative_else_nop_endif
+
+	.if	\el != 0
+	tbnz	x22, #22, 1f			// Skip re-enabling TTBR0 access if the PSR_PAN_BIT is set
+	.endif
+
+	__uaccess_ttbr0_enable x0
+
+	.if	\el == 0
+	/*
+	 * Enable errata workarounds only if returning to user. The only
+	 * workaround currently required for TTBR0_EL1 changes are for the
+	 * Cavium erratum 27456 (broadcast TLBI instructions may cause I-cache
+	 * corruption).
+	 */
+	post_ttbr0_update_workaround
+	.endif
+1:
+	.if	\el != 0
+	and	x22, x22, #~PSR_PAN_BIT		// ARMv8.0 CPUs do not understand this bit
+	.endif
+2:
+#endif
+
+	.if	\el == 0
 	ldr	x23, [sp, #S_SP]		// load return stack pointer
 	msr	sp_el0, x23
 #ifdef CONFIG_ARM64_ERRATUM_845719
@@ -166,6 +245,7 @@ alternative_else
 alternative_endif
 #endif
 	.endif
+
 	msr	elr_el1, x21			// set up the return data
 	msr	spsr_el1, x22
 	ldp	x0, x1, [sp, #16 * 0]
@@ -188,23 +268,26 @@ alternative_endif
 	eret					// return to kernel
 	.endm
 
-	.macro	get_thread_info, rd
-	mrs	\rd, sp_el0
-	.endm
-
 	.macro	irq_stack_entry
 	mov	x19, sp			// preserve the original sp
 
 	/*
-	 * Compare sp with the current thread_info, if the top
-	 * ~(THREAD_SIZE - 1) bits match, we are on a task stack, and
-	 * should switch to the irq stack.
+	 * Compare sp with the base of the task stack.
+	 * If the top ~(THREAD_SIZE - 1) bits match, we are on a task stack,
+	 * and should switch to the irq stack.
 	 */
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	x25, [tsk, TSK_STACK]
+	eor	x25, x25, x19
+	and	x25, x25, #~(THREAD_SIZE - 1)
+	cbnz	x25, 9998f
+#else
 	and	x25, x19, #~(THREAD_SIZE - 1)
 	cmp	x25, tsk
 	b.ne	9998f
+#endif
 
-	this_cpu_ptr irq_stack, x25, x26
+	adr_this_cpu x25, irq_stack, x26
 	mov	x26, #IRQ_STACK_START_SP
 	add	x26, x25, x26
 
@@ -292,7 +375,7 @@ END(vectors)
  * Invalid mode handlers
  */
 	.macro	inv_entry, el, reason, regsize = 64
-	kernel_entry el, \regsize
+	kernel_entry \el, \regsize
 	mov	x0, sp
 	mov	x1, #\reason
 	mrs	x2, esr_el1
@@ -351,6 +434,8 @@ el1_sync:
 	lsr	x24, x1, #ESR_ELx_EC_SHIFT	// exception class
 	cmp	x24, #ESR_ELx_EC_DABT_CUR	// data abort in EL1
 	b.eq	el1_da
+	cmp	x24, #ESR_ELx_EC_IABT_CUR	// instruction abort in EL1
+	b.eq	el1_ia
 	cmp	x24, #ESR_ELx_EC_SYS64		// configurable trap
 	b.eq	el1_undef
 	cmp	x24, #ESR_ELx_EC_SP_ALIGN	// stack alignment exception
@@ -362,6 +447,11 @@ el1_sync:
 	cmp	x24, #ESR_ELx_EC_BREAKPT_CUR	// debug exception in EL1
 	b.ge	el1_dbg
 	b	el1_inv
+
+el1_ia:
+	/*
+	 * Fall through to the Data abort case
+	 */
 el1_da:
 	/*
 	 * Data abort handling
@@ -426,9 +516,17 @@ el1_irq:
 	irq_handler
 
 #ifdef CONFIG_PREEMPT
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	w24, [tsk, #TSK_TI_PREEMPT]	// get preempt count
+#else
 	ldr	w24, [tsk, #TI_PREEMPT]		// get preempt count
+#endif
 	cbnz	w24, 1f				// preempt count != 0
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	x0, [tsk, #TSK_TI_FLAGS]	// get flags
+#else
 	ldr	x0, [tsk, #TI_FLAGS]		// get flags
+#endif
 	tbz	x0, #TIF_NEED_RESCHED, 1f	// needs rescheduling?
 	bl	el1_preempt
 1:
@@ -443,7 +541,11 @@ ENDPROC(el1_irq)
 el1_preempt:
 	mov	x24, lr
 1:	bl	preempt_schedule_irq		// irq en/disable is done inside
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	x0, [tsk, #TSK_TI_FLAGS]	// get new tasks TI_FLAGS
+#else
 	ldr	x0, [tsk, #TI_FLAGS]		// get new tasks TI_FLAGS
+#endif
 	tbnz	x0, #TIF_NEED_RESCHED, 1b	// needs rescheduling?
 	ret	x24
 #endif
@@ -548,7 +650,7 @@ el0_ia:
 	enable_dbg_and_irq
 	ct_user_exit
 	mov	x0, x26
-	orr	x1, x25, #1 << 24		// use reserved ISS bit for instruction aborts
+	mov	x1, x25
 	mov	x2, sp
 	bl	do_mem_abort
 	b	ret_to_user
@@ -663,8 +765,12 @@ ENTRY(cpu_switch_to)
 	ldp	x29, x9, [x8], #16
 	ldr	lr, [x8]
 	mov	sp, x9
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	msr	sp_el0, x1
+#else
 	and	x9, x9, #~(THREAD_SIZE - 1)
 	msr	sp_el0, x9
+#endif
 	ret
 ENDPROC(cpu_switch_to)
 
@@ -675,7 +781,11 @@ ENDPROC(cpu_switch_to)
 ret_fast_syscall:
 	disable_irq				// disable interrupts
 	str	x0, [sp, #S_X0]			// returned x0
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	x1, [tsk, #TSK_TI_FLAGS]	// re-check for syscall tracing
+#else
 	ldr	x1, [tsk, #TI_FLAGS]		// re-check for syscall tracing
+#endif
 	and	x2, x1, #_TIF_SYSCALL_WORK
 	cbnz	x2, ret_fast_syscall_trace
 	and	x2, x1, #_TIF_WORK_MASK
@@ -707,7 +817,11 @@ work_resched:
  */
 ret_to_user:
 	disable_irq				// disable interrupts
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	x1, [tsk, #TSK_TI_FLAGS]
+#else
 	ldr	x1, [tsk, #TI_FLAGS]
+#endif
 	and	x2, x1, #_TIF_WORK_MASK
 	cbnz	x2, work_pending
 	enable_step_tsk x1, x2
@@ -739,7 +853,11 @@ el0_svc_naked:					// compat entry point
 	enable_dbg_and_irq
 	ct_user_exit 1
 
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	ldr	x16, [tsk, #TSK_TI_FLAGS]	// check for syscall hooks
+#else
 	ldr	x16, [tsk, #TI_FLAGS]		// check for syscall hooks
+#endif
 	tst	x16, #_TIF_SYSCALL_WORK
 	b.ne	__sys_trace
 	cmp     scno, sc_nr                     // check upper syscall limit
diff --git a/arch/arm64/kernel/head.S b/arch/arm64/kernel/head.S
index 23cf8e9a295f..3e99814b6463 100644
--- a/arch/arm64/kernel/head.S
+++ b/arch/arm64/kernel/head.S
@@ -319,14 +319,14 @@ __create_page_tables:
 	 * dirty cache lines being evicted.
 	 */
 	mov	x0, x25
-	add	x1, x26, #SWAPPER_DIR_SIZE
+	add	x1, x26, #SWAPPER_DIR_SIZE + RESERVED_TTBR0_SIZE
 	bl	__inval_cache_range
 
 	/*
 	 * Clear the idmap and swapper page tables.
 	 */
 	mov	x0, x25
-	add	x6, x26, #SWAPPER_DIR_SIZE
+	add	x6, x26, #SWAPPER_DIR_SIZE + RESERVED_TTBR0_SIZE
 1:	stp	xzr, xzr, [x0], #16
 	stp	xzr, xzr, [x0], #16
 	stp	xzr, xzr, [x0], #16
@@ -405,7 +405,7 @@ __create_page_tables:
 	 * tables again to remove any speculatively loaded cache lines.
 	 */
 	mov	x0, x25
-	add	x1, x26, #SWAPPER_DIR_SIZE
+	add	x1, x26, #SWAPPER_DIR_SIZE + RESERVED_TTBR0_SIZE
 	dmb	sy
 	bl	__inval_cache_range
 
@@ -419,6 +419,7 @@ ENDPROC(__create_page_tables)
 	.set	initial_sp, init_thread_union + THREAD_START_SP
 __primary_switched:
 	mov	x28, lr				// preserve LR
+
 	adr_l	x8, vectors			// load VBAR_EL1 with virtual
 	msr	vbar_el1, x8			// vector table address
 	isb
@@ -431,10 +432,18 @@ __primary_switched:
 	bl	__pi_memset
 	dsb	ishst				// Make zero page visible to PTW
 
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+        adrp    x4, init_thread_union
+        add     sp, x4, #THREAD_SIZE
+        adr_l   x5, init_task
+        msr     sp_el0, x5                      // Save thread_info
+#else
 	adr_l	sp, initial_sp, x4
 	mov	x4, sp
 	and	x4, x4, #~(THREAD_SIZE - 1)
 	msr	sp_el0, x4			// Save thread_info
+#endif
+
 	str_l	x21, __fdt_pointer, x5		// Save FDT pointer
 
 	ldr_l	x4, kimage_vaddr		// Save the offset between
@@ -464,7 +473,7 @@ ENDPROC(__primary_switched)
  * end early head section, begin head code that is also used for
  * hotplug and needs to have the same protections as the text region
  */
-	.section ".text","ax"
+	.section ".idmap.text","ax"
 
 ENTRY(kimage_vaddr)
 	.quad		_text - TEXT_OFFSET
@@ -587,17 +596,23 @@ set_cpu_boot_mode_flag:
 ENDPROC(set_cpu_boot_mode_flag)
 
 /*
+ * These values are written with the MMU off, but read with the MMU on.
+ * Writers will invalidate the corresponding address, discarding up to a
+ * 'Cache Writeback Granule' (CWG) worth of data. The linker script ensures
+ * sufficient alignment that the CWG doesn't overlap another section.
+ */
+	.pushsection ".mmuoff.data.write", "aw"
+/*
  * We need to find out the CPU boot mode long after boot, so we need to
  * store it in a writable variable.
  *
  * This is not in .bss, because we set it sufficiently early that the boot-time
  * zeroing of .bss would clobber it.
  */
-	.pushsection	.data..cacheline_aligned
-	.align	L1_CACHE_SHIFT
 ENTRY(__boot_cpu_mode)
 	.long	BOOT_CPU_MODE_EL2
 	.long	BOOT_CPU_MODE_EL1
+
 	.popsection
 
 	/*
@@ -644,12 +659,19 @@ __secondary_switched:
 	adr_l	x5, vectors
 	msr	vbar_el1, x5
 	isb
-
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	adr_l	x0, secondary_data
+	ldr	x1, [x0, #CPU_BOOT_STACK]	// get secondary_data.stack
+	mov	sp, x1
+	ldr	x2, [x0, #CPU_BOOT_TASK]
+	msr	sp_el0, x2
+#else
 	adr_l	x0, secondary_data
 	ldr	x0, [x0, #CPU_BOOT_STACK]	// get secondary_data.stack
 	mov	sp, x0
 	and	x0, x0, #~(THREAD_SIZE - 1)
 	msr	sp_el0, x0			// save thread_info
+#endif
 	mov	x29, #0
 	b	secondary_start_kernel
 ENDPROC(__secondary_switched)
@@ -688,7 +710,6 @@ ENTRY(__early_cpu_boot_status)
  * Checks if the selected granule size is supported by the CPU.
  * If it isn't, park the CPU
  */
-	.section	".idmap.text", "ax"
 ENTRY(__enable_mmu)
 	mrs	x18, sctlr_el1			// preserve old SCTLR_EL1 value
 	mrs	x1, ID_AA64MMFR0_EL1
diff --git a/arch/arm64/kernel/hibernate.c b/arch/arm64/kernel/hibernate.c
index 35a33d705536..4284ced551fe 100644
--- a/arch/arm64/kernel/hibernate.c
+++ b/arch/arm64/kernel/hibernate.c
@@ -54,12 +54,6 @@ extern int in_suspend;
 /* Do we need to reset el2? */
 #define el2_reset_needed() (is_hyp_mode_available() && !is_kernel_in_hyp_mode())
 
-/*
- * Start/end of the hibernate exit code, this must be copied to a 'safe'
- * location in memory, and executed from there.
- */
-extern char __hibernate_exit_text_start[], __hibernate_exit_text_end[];
-
 /* temporary el2 vectors in the __hibernate_exit_text section. */
 extern char hibernate_el2_vectors[];
 
@@ -242,6 +236,7 @@ out:
 	return rc;
 }
 
+#define dcache_clean_range(start, end)	__flush_dcache_area(start, (end - start))
 
 int swsusp_arch_suspend(void)
 {
@@ -257,8 +252,13 @@ int swsusp_arch_suspend(void)
 
 		ret = swsusp_save();
 	} else {
-		/* Clean kernel to PoC for secondary core startup */
-		__flush_dcache_area(LMADDR(KERNEL_START), KERNEL_END - KERNEL_START);
+		/* Clean kernel core startup/idle code to PoC*/
+		dcache_clean_range(__mmuoff_data_start, __mmuoff_data_end);
+		dcache_clean_range(__idmap_text_start, __idmap_text_end);
+
+		/* Clean kvm setup code to PoC? */
+		if (el2_reset_needed())
+			dcache_clean_range(__hyp_idmap_text_start, __hyp_idmap_text_end);
 
 		/* make the crash dump kernel image protected again */
 		crash_post_resume();
@@ -277,6 +277,33 @@ int swsusp_arch_suspend(void)
 	return ret;
 }
 
+static void _copy_pte(pte_t *dst_pte, pte_t *src_pte, unsigned long addr)
+{
+	pte_t pte = *src_pte;
+
+	if (pte_valid(pte)) {
+		/*
+		 * Resume will overwrite areas that may be marked
+		 * read only (code, rodata). Clear the RDONLY bit from
+		 * the temporary mappings we use during restore.
+		 */
+		set_pte(dst_pte, pte_clear_rdonly(pte));
+	} else if (debug_pagealloc_enabled() && !pte_none(pte)) {
+		/*
+		 * debug_pagealloc will removed the PTE_VALID bit if
+		 * the page isn't in use by the resume kernel. It may have
+		 * been in use by the original kernel, in which case we need
+		 * to put it back in our copy to do the restore.
+		 *
+		 * Before marking this entry valid, check the pfn should
+		 * be mapped.
+		 */
+		BUG_ON(!pfn_valid(pte_pfn(pte)));
+
+		set_pte(dst_pte, pte_mkpresent(pte_clear_rdonly(pte)));
+	}
+}
+
 static int copy_pte(pmd_t *dst_pmd, pmd_t *src_pmd, unsigned long start,
 		    unsigned long end)
 {
@@ -292,13 +319,7 @@ static int copy_pte(pmd_t *dst_pmd, pmd_t *src_pmd, unsigned long start,
 
 	src_pte = pte_offset_kernel(src_pmd, start);
 	do {
-		if (!pte_none(*src_pte))
-			/*
-			 * Resume will overwrite areas that may be marked
-			 * read only (code, rodata). Clear the RDONLY bit from
-			 * the temporary mappings we use during restore.
-			 */
-			set_pte(dst_pte, __pte(pte_val(*src_pte) & ~PTE_RDONLY));
+		_copy_pte(dst_pte, src_pte, addr);
 	} while (dst_pte++, src_pte++, addr += PAGE_SIZE, addr != end);
 
 	return 0;
diff --git a/arch/arm64/kernel/hw_breakpoint.c b/arch/arm64/kernel/hw_breakpoint.c
index 22543cb1a8e1..1c694f3c643c 100644
--- a/arch/arm64/kernel/hw_breakpoint.c
+++ b/arch/arm64/kernel/hw_breakpoint.c
@@ -318,9 +318,21 @@ static int get_hbp_len(u8 hbp_len)
 	case ARM_BREAKPOINT_LEN_2:
 		len_in_bytes = 2;
 		break;
+	case ARM_BREAKPOINT_LEN_3:
+		len_in_bytes = 3;
+		break;
 	case ARM_BREAKPOINT_LEN_4:
 		len_in_bytes = 4;
 		break;
+	case ARM_BREAKPOINT_LEN_5:
+		len_in_bytes = 5;
+		break;
+	case ARM_BREAKPOINT_LEN_6:
+		len_in_bytes = 6;
+		break;
+	case ARM_BREAKPOINT_LEN_7:
+		len_in_bytes = 7;
+		break;
 	case ARM_BREAKPOINT_LEN_8:
 		len_in_bytes = 8;
 		break;
@@ -350,7 +362,7 @@ int arch_check_bp_in_kernelspace(struct perf_event *bp)
  * to generic breakpoint descriptions.
  */
 int arch_bp_generic_fields(struct arch_hw_breakpoint_ctrl ctrl,
-			   int *gen_len, int *gen_type)
+			   int *gen_len, int *gen_type, int *offset)
 {
 	/* Type */
 	switch (ctrl.type) {
@@ -370,17 +382,33 @@ int arch_bp_generic_fields(struct arch_hw_breakpoint_ctrl ctrl,
 		return -EINVAL;
 	}
 
+	if (!ctrl.len)
+		return -EINVAL;
+	*offset = __ffs(ctrl.len);
+
 	/* Len */
-	switch (ctrl.len) {
+	switch (ctrl.len >> *offset) {
 	case ARM_BREAKPOINT_LEN_1:
 		*gen_len = HW_BREAKPOINT_LEN_1;
 		break;
 	case ARM_BREAKPOINT_LEN_2:
 		*gen_len = HW_BREAKPOINT_LEN_2;
 		break;
+	case ARM_BREAKPOINT_LEN_3:
+		*gen_len = HW_BREAKPOINT_LEN_3;
+		break;
 	case ARM_BREAKPOINT_LEN_4:
 		*gen_len = HW_BREAKPOINT_LEN_4;
 		break;
+	case ARM_BREAKPOINT_LEN_5:
+		*gen_len = HW_BREAKPOINT_LEN_5;
+		break;
+	case ARM_BREAKPOINT_LEN_6:
+		*gen_len = HW_BREAKPOINT_LEN_6;
+		break;
+	case ARM_BREAKPOINT_LEN_7:
+		*gen_len = HW_BREAKPOINT_LEN_7;
+		break;
 	case ARM_BREAKPOINT_LEN_8:
 		*gen_len = HW_BREAKPOINT_LEN_8;
 		break;
@@ -424,9 +452,21 @@ static int arch_build_bp_info(struct perf_event *bp)
 	case HW_BREAKPOINT_LEN_2:
 		info->ctrl.len = ARM_BREAKPOINT_LEN_2;
 		break;
+	case HW_BREAKPOINT_LEN_3:
+		info->ctrl.len = ARM_BREAKPOINT_LEN_3;
+		break;
 	case HW_BREAKPOINT_LEN_4:
 		info->ctrl.len = ARM_BREAKPOINT_LEN_4;
 		break;
+	case HW_BREAKPOINT_LEN_5:
+		info->ctrl.len = ARM_BREAKPOINT_LEN_5;
+		break;
+	case HW_BREAKPOINT_LEN_6:
+		info->ctrl.len = ARM_BREAKPOINT_LEN_6;
+		break;
+	case HW_BREAKPOINT_LEN_7:
+		info->ctrl.len = ARM_BREAKPOINT_LEN_7;
+		break;
 	case HW_BREAKPOINT_LEN_8:
 		info->ctrl.len = ARM_BREAKPOINT_LEN_8;
 		break;
@@ -518,18 +558,17 @@ int arch_validate_hwbkpt_settings(struct perf_event *bp)
 		default:
 			return -EINVAL;
 		}
-
-		info->address &= ~alignment_mask;
-		info->ctrl.len <<= offset;
 	} else {
 		if (info->ctrl.type == ARM_BREAKPOINT_EXECUTE)
 			alignment_mask = 0x3;
 		else
 			alignment_mask = 0x7;
-		if (info->address & alignment_mask)
-			return -EINVAL;
+		offset = info->address & alignment_mask;
 	}
 
+	info->address &= ~alignment_mask;
+	info->ctrl.len <<= offset;
+
 	/*
 	 * Disallow per-task kernel breakpoints since these would
 	 * complicate the stepping code.
@@ -662,12 +701,47 @@ unlock:
 }
 NOKPROBE_SYMBOL(breakpoint_handler);
 
+/*
+ * Arm64 hardware does not always report a watchpoint hit address that matches
+ * one of the watchpoints set. It can also report an address "near" the
+ * watchpoint if a single instruction access both watched and unwatched
+ * addresses. There is no straight-forward way, short of disassembling the
+ * offending instruction, to map that address back to the watchpoint. This
+ * function computes the distance of the memory access from the watchpoint as a
+ * heuristic for the likelyhood that a given access triggered the watchpoint.
+ *
+ * See Section D2.10.5 "Determining the memory location that caused a Watchpoint
+ * exception" of ARMv8 Architecture Reference Manual for details.
+ *
+ * The function returns the distance of the address from the bytes watched by
+ * the watchpoint. In case of an exact match, it returns 0.
+ */
+static u64 get_distance_from_watchpoint(unsigned long addr, u64 val,
+					struct arch_hw_breakpoint_ctrl *ctrl)
+{
+	u64 wp_low, wp_high;
+	u32 lens, lene;
+
+	lens = __ffs(ctrl->len);
+	lene = __fls(ctrl->len);
+
+	wp_low = val + lens;
+	wp_high = val + lene;
+	if (addr < wp_low)
+		return wp_low - addr;
+	else if (addr > wp_high)
+		return addr - wp_high;
+	else
+		return 0;
+}
+
 static int watchpoint_handler(unsigned long addr, unsigned int esr,
 			      struct pt_regs *regs)
 {
-	int i, step = 0, *kernel_step, access;
+	int i, step = 0, *kernel_step, access, closest_match = 0;
+	u64 min_dist = -1, dist;
 	u32 ctrl_reg;
-	u64 val, alignment_mask;
+	u64 val;
 	struct perf_event *wp, **slots;
 	struct debug_info *debug_info;
 	struct arch_hw_breakpoint *info;
@@ -676,35 +750,15 @@ static int watchpoint_handler(unsigned long addr, unsigned int esr,
 	slots = this_cpu_ptr(wp_on_reg);
 	debug_info = &current->thread.debug;
 
+	/*
+	 * Find all watchpoints that match the reported address. If no exact
+	 * match is found. Attribute the hit to the closest watchpoint.
+	 */
+	rcu_read_lock();
 	for (i = 0; i < core_num_wrps; ++i) {
-		rcu_read_lock();
-
 		wp = slots[i];
-
 		if (wp == NULL)
-			goto unlock;
-
-		info = counter_arch_bp(wp);
-		/* AArch32 watchpoints are either 4 or 8 bytes aligned. */
-		if (is_compat_task()) {
-			if (info->ctrl.len == ARM_BREAKPOINT_LEN_8)
-				alignment_mask = 0x7;
-			else
-				alignment_mask = 0x3;
-		} else {
-			alignment_mask = 0x7;
-		}
-
-		/* Check if the watchpoint value matches. */
-		val = read_wb_reg(AARCH64_DBG_REG_WVR, i);
-		if (val != (untagged_addr(addr) & ~alignment_mask))
-			goto unlock;
-
-		/* Possible match, check the byte address select to confirm. */
-		ctrl_reg = read_wb_reg(AARCH64_DBG_REG_WCR, i);
-		decode_ctrl_reg(ctrl_reg, &ctrl);
-		if (!((1 << (addr & alignment_mask)) & ctrl.len))
-			goto unlock;
+			continue;
 
 		/*
 		 * Check that the access type matches.
@@ -713,18 +767,41 @@ static int watchpoint_handler(unsigned long addr, unsigned int esr,
 		access = (esr & AARCH64_ESR_ACCESS_MASK) ? HW_BREAKPOINT_W :
 			 HW_BREAKPOINT_R;
 		if (!(access & hw_breakpoint_type(wp)))
-			goto unlock;
+			continue;
 
+		/* Check if the watchpoint value and byte select match. */
+		val = read_wb_reg(AARCH64_DBG_REG_WVR, i);
+		ctrl_reg = read_wb_reg(AARCH64_DBG_REG_WCR, i);
+		decode_ctrl_reg(ctrl_reg, &ctrl);
+		dist = get_distance_from_watchpoint(addr, val, &ctrl);
+		if (dist < min_dist) {
+			min_dist = dist;
+			closest_match = i;
+		}
+		/* Is this an exact match? */
+		if (dist != 0)
+			continue;
+
+		info = counter_arch_bp(wp);
 		info->trigger = addr;
 		perf_bp_event(wp, regs);
 
 		/* Do we need to handle the stepping? */
 		if (!wp->overflow_handler)
 			step = 1;
+	}
+	if (min_dist > 0 && min_dist != -1) {
+		/* No exact match found. */
+		wp = slots[closest_match];
+		info = counter_arch_bp(wp);
+		info->trigger = addr;
+		perf_bp_event(wp, regs);
 
-unlock:
-		rcu_read_unlock();
+		/* Do we need to handle the stepping? */
+		if (!wp->overflow_handler)
+			step = 1;
 	}
+	rcu_read_unlock();
 
 	if (!step)
 		return 0;
diff --git a/arch/arm64/kernel/io.c b/arch/arm64/kernel/io.c
index 354be2a872ae..79b17384effa 100644
--- a/arch/arm64/kernel/io.c
+++ b/arch/arm64/kernel/io.c
@@ -25,8 +25,7 @@
  */
 void __memcpy_fromio(void *to, const volatile void __iomem *from, size_t count)
 {
-	while (count && (!IS_ALIGNED((unsigned long)from, 8) ||
-			 !IS_ALIGNED((unsigned long)to, 8))) {
+	while (count && !IS_ALIGNED((unsigned long)from, 8)) {
 		*(u8 *)to = __raw_readb(from);
 		from++;
 		to++;
@@ -54,23 +53,22 @@ EXPORT_SYMBOL(__memcpy_fromio);
  */
 void __memcpy_toio(volatile void __iomem *to, const void *from, size_t count)
 {
-	while (count && (!IS_ALIGNED((unsigned long)to, 8) ||
-			 !IS_ALIGNED((unsigned long)from, 8))) {
-		__raw_writeb(*(volatile u8 *)from, to);
+	while (count && !IS_ALIGNED((unsigned long)to, 8)) {
+		__raw_writeb(*(u8 *)from, to);
 		from++;
 		to++;
 		count--;
 	}
 
 	while (count >= 8) {
-		__raw_writeq(*(volatile u64 *)from, to);
+		__raw_writeq(*(u64 *)from, to);
 		from += 8;
 		to += 8;
 		count -= 8;
 	}
 
 	while (count) {
-		__raw_writeb(*(volatile u8 *)from, to);
+		__raw_writeb(*(u8 *)from, to);
 		from++;
 		to++;
 		count--;
diff --git a/arch/arm64/kernel/module-plts.c b/arch/arm64/kernel/module-plts.c
index 1ce90d8450ae..d05dbe658409 100644
--- a/arch/arm64/kernel/module-plts.c
+++ b/arch/arm64/kernel/module-plts.c
@@ -1,5 +1,5 @@
 /*
- * Copyright (C) 2014-2016 Linaro Ltd. <ard.biesheuvel@linaro.org>
+ * Copyright (C) 2014-2017 Linaro Ltd. <ard.biesheuvel@linaro.org>
  *
  * This program is free software; you can redistribute it and/or modify
  * it under the terms of the GNU General Public License version 2 as
@@ -26,35 +26,21 @@ struct plt_entry {
 	__le32	br;	/* br	x16				*/
 };
 
-u64 module_emit_plt_entry(struct module *mod, const Elf64_Rela *rela,
+static bool in_init(const struct module *mod, void *loc)
+{
+	return (u64)loc - (u64)mod->init_layout.base < mod->init_layout.size;
+}
+
+u64 module_emit_plt_entry(struct module *mod, void *loc, const Elf64_Rela *rela,
 			  Elf64_Sym *sym)
 {
-	struct plt_entry *plt = (struct plt_entry *)mod->arch.plt->sh_addr;
-	int i = mod->arch.plt_num_entries;
+	struct mod_plt_sec *pltsec = !in_init(mod, loc) ? &mod->arch.core :
+							  &mod->arch.init;
+	struct plt_entry *plt = (struct plt_entry *)pltsec->plt->sh_addr;
+	int i = pltsec->plt_num_entries;
 	u64 val = sym->st_value + rela->r_addend;
 
 	/*
-	 * We only emit PLT entries against undefined (SHN_UNDEF) symbols,
-	 * which are listed in the ELF symtab section, but without a type
-	 * or a size.
-	 * So, similar to how the module loader uses the Elf64_Sym::st_value
-	 * field to store the resolved addresses of undefined symbols, let's
-	 * borrow the Elf64_Sym::st_size field (whose value is never used by
-	 * the module loader, even for symbols that are defined) to record
-	 * the address of a symbol's associated PLT entry as we emit it for a
-	 * zero addend relocation (which is the only kind we have to deal with
-	 * in practice). This allows us to find duplicates without having to
-	 * go through the table every time.
-	 */
-	if (rela->r_addend == 0 && sym->st_size != 0) {
-		BUG_ON(sym->st_size < (u64)plt || sym->st_size >= (u64)&plt[i]);
-		return sym->st_size;
-	}
-
-	mod->arch.plt_num_entries++;
-	BUG_ON(mod->arch.plt_num_entries > mod->arch.plt_max_entries);
-
-	/*
 	 * MOVK/MOVN/MOVZ opcode:
 	 * +--------+------------+--------+-----------+-------------+---------+
 	 * | sf[31] | opc[30:29] | 100101 | hw[22:21] | imm16[20:5] | Rd[4:0] |
@@ -72,8 +58,19 @@ u64 module_emit_plt_entry(struct module *mod, const Elf64_Rela *rela,
 		cpu_to_le32(0xd61f0200)
 	};
 
-	if (rela->r_addend == 0)
-		sym->st_size = (u64)&plt[i];
+	/*
+	 * Check if the entry we just created is a duplicate. Given that the
+	 * relocations are sorted, this will be the last entry we allocated.
+	 * (if one exists).
+	 */
+	if (i > 0 &&
+	    plt[i].mov0 == plt[i - 1].mov0 &&
+	    plt[i].mov1 == plt[i - 1].mov1 &&
+	    plt[i].mov2 == plt[i - 1].mov2)
+		return (u64)&plt[i - 1];
+
+	pltsec->plt_num_entries++;
+	BUG_ON(pltsec->plt_num_entries > pltsec->plt_max_entries);
 
 	return (u64)&plt[i];
 }
@@ -104,7 +101,8 @@ static bool duplicate_rel(const Elf64_Rela *rela, int num)
 	return num > 0 && cmp_rela(rela + num, rela + num - 1) == 0;
 }
 
-static unsigned int count_plts(Elf64_Sym *syms, Elf64_Rela *rela, int num)
+static unsigned int count_plts(Elf64_Sym *syms, Elf64_Rela *rela, int num,
+			       Elf64_Word dstidx)
 {
 	unsigned int ret = 0;
 	Elf64_Sym *s;
@@ -116,13 +114,17 @@ static unsigned int count_plts(Elf64_Sym *syms, Elf64_Rela *rela, int num)
 		case R_AARCH64_CALL26:
 			/*
 			 * We only have to consider branch targets that resolve
-			 * to undefined symbols. This is not simply a heuristic,
-			 * it is a fundamental limitation, since the PLT itself
-			 * is part of the module, and needs to be within 128 MB
-			 * as well, so modules can never grow beyond that limit.
+			 * to symbols that are defined in a different section.
+			 * This is not simply a heuristic, it is a fundamental
+			 * limitation, since there is no guaranteed way to emit
+			 * PLT entries sufficiently close to the branch if the
+			 * section size exceeds the range of a branch
+			 * instruction. So ignore relocations against defined
+			 * symbols if they live in the same section as the
+			 * relocation target.
 			 */
 			s = syms + ELF64_R_SYM(rela[i].r_info);
-			if (s->st_shndx != SHN_UNDEF)
+			if (s->st_shndx == dstidx)
 				break;
 
 			/*
@@ -149,7 +151,8 @@ static unsigned int count_plts(Elf64_Sym *syms, Elf64_Rela *rela, int num)
 int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
 			      char *secstrings, struct module *mod)
 {
-	unsigned long plt_max_entries = 0;
+	unsigned long core_plts = 0;
+	unsigned long init_plts = 0;
 	Elf64_Sym *syms = NULL;
 	int i;
 
@@ -158,14 +161,16 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
 	 * entries. Record the symtab address as well.
 	 */
 	for (i = 0; i < ehdr->e_shnum; i++) {
-		if (strcmp(".plt", secstrings + sechdrs[i].sh_name) == 0)
-			mod->arch.plt = sechdrs + i;
+		if (!strcmp(secstrings + sechdrs[i].sh_name, ".plt"))
+			mod->arch.core.plt = sechdrs + i;
+		else if (!strcmp(secstrings + sechdrs[i].sh_name, ".init.plt"))
+			mod->arch.init.plt = sechdrs + i;
 		else if (sechdrs[i].sh_type == SHT_SYMTAB)
 			syms = (Elf64_Sym *)sechdrs[i].sh_addr;
 	}
 
-	if (!mod->arch.plt) {
-		pr_err("%s: module PLT section missing\n", mod->name);
+	if (!mod->arch.core.plt || !mod->arch.init.plt) {
+		pr_err("%s: module PLT section(s) missing\n", mod->name);
 		return -ENOEXEC;
 	}
 	if (!syms) {
@@ -188,14 +193,27 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
 		/* sort by type, symbol index and addend */
 		sort(rels, numrels, sizeof(Elf64_Rela), cmp_rela, NULL);
 
-		plt_max_entries += count_plts(syms, rels, numrels);
+		if (strncmp(secstrings + dstsec->sh_name, ".init", 5) != 0)
+			core_plts += count_plts(syms, rels, numrels,
+						sechdrs[i].sh_info);
+		else
+			init_plts += count_plts(syms, rels, numrels,
+						sechdrs[i].sh_info);
 	}
 
-	mod->arch.plt->sh_type = SHT_NOBITS;
-	mod->arch.plt->sh_flags = SHF_EXECINSTR | SHF_ALLOC;
-	mod->arch.plt->sh_addralign = L1_CACHE_BYTES;
-	mod->arch.plt->sh_size = plt_max_entries * sizeof(struct plt_entry);
-	mod->arch.plt_num_entries = 0;
-	mod->arch.plt_max_entries = plt_max_entries;
+	mod->arch.core.plt->sh_type = SHT_NOBITS;
+	mod->arch.core.plt->sh_flags = SHF_EXECINSTR | SHF_ALLOC;
+	mod->arch.core.plt->sh_addralign = L1_CACHE_BYTES;
+	mod->arch.core.plt->sh_size = (core_plts  + 1) * sizeof(struct plt_entry);
+	mod->arch.core.plt_num_entries = 0;
+	mod->arch.core.plt_max_entries = core_plts;
+
+	mod->arch.init.plt->sh_type = SHT_NOBITS;
+	mod->arch.init.plt->sh_flags = SHF_EXECINSTR | SHF_ALLOC;
+	mod->arch.init.plt->sh_addralign = L1_CACHE_BYTES;
+	mod->arch.init.plt->sh_size = (init_plts + 1) * sizeof(struct plt_entry);
+	mod->arch.init.plt_num_entries = 0;
+	mod->arch.init.plt_max_entries = init_plts;
+
 	return 0;
 }
diff --git a/arch/arm64/kernel/module.c b/arch/arm64/kernel/module.c
index 7f316982ce00..c9a2ab446dc6 100644
--- a/arch/arm64/kernel/module.c
+++ b/arch/arm64/kernel/module.c
@@ -380,7 +380,7 @@ int apply_relocate_add(Elf64_Shdr *sechdrs,
 
 			if (IS_ENABLED(CONFIG_ARM64_MODULE_PLTS) &&
 			    ovf == -ERANGE) {
-				val = module_emit_plt_entry(me, &rel[i], sym);
+				val = module_emit_plt_entry(me, loc, &rel[i], sym);
 				ovf = reloc_insn_imm(RELOC_OP_PREL, loc, val, 2,
 						     26, AARCH64_INSN_IMM_26);
 			}
diff --git a/arch/arm64/kernel/module.lds b/arch/arm64/kernel/module.lds
index 8949f6c6f729..f7c9781a9d48 100644
--- a/arch/arm64/kernel/module.lds
+++ b/arch/arm64/kernel/module.lds
@@ -1,3 +1,4 @@
 SECTIONS {
 	.plt (NOLOAD) : { BYTE(0) }
+	.init.plt (NOLOAD) : { BYTE(0) }
 }
diff --git a/arch/arm64/kernel/probes/decode-insn.c b/arch/arm64/kernel/probes/decode-insn.c
index f7931d900bca..37e47a9d617e 100644
--- a/arch/arm64/kernel/probes/decode-insn.c
+++ b/arch/arm64/kernel/probes/decode-insn.c
@@ -157,10 +157,10 @@ arm_kprobe_decode_insn(kprobe_opcode_t *addr, struct arch_specific_insn *asi)
 		mod = __module_address((unsigned long)addr);
 		if (mod && within_module_init((unsigned long)addr, mod) &&
 			!within_module_init((unsigned long)scan_end, mod))
-			scan_end = (kprobe_opcode_t *)mod->module_init;
+			scan_end = (kprobe_opcode_t *)mod->init_layout.base;
 		else if (mod && within_module_core((unsigned long)addr, mod) &&
 			!within_module_core((unsigned long)scan_end, mod))
-			scan_end = (kprobe_opcode_t *)mod->module_core;
+			scan_end = (kprobe_opcode_t *)mod->core_layout.base;
 		preempt_enable();
 	}
 #endif
diff --git a/arch/arm64/kernel/probes/kprobes.c b/arch/arm64/kernel/probes/kprobes.c
index 1ee93c7c5a75..17a6449c96a2 100644
--- a/arch/arm64/kernel/probes/kprobes.c
+++ b/arch/arm64/kernel/probes/kprobes.c
@@ -31,7 +31,7 @@
 #include <asm/insn.h>
 #include <asm/uaccess.h>
 #include <asm/irq.h>
-#include <asm-generic/sections.h>
+#include <asm/sections.h>
 
 #include "decode-insn.h"
 
@@ -543,8 +543,6 @@ int __kprobes longjmp_break_handler(struct kprobe *p, struct pt_regs *regs)
 
 bool arch_within_kprobe_blacklist(unsigned long addr)
 {
-	extern char __idmap_text_start[], __idmap_text_end[];
-
 	if ((addr >= (unsigned long)__kprobes_text_start &&
 	    addr < (unsigned long)__kprobes_text_end) ||
 	    (addr >= (unsigned long)__entry_text_start &&
diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c
index 80624829db61..e34bcf3f2c35 100644
--- a/arch/arm64/kernel/process.c
+++ b/arch/arm64/kernel/process.c
@@ -45,10 +45,14 @@
 #include <linux/personality.h>
 #include <linux/notifier.h>
 #include <trace/events/power.h>
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+#include <linux/percpu.h>
+#endif
 
 #include <asm/alternative.h>
 #include <asm/compat.h>
 #include <asm/cacheflush.h>
+#include <asm/exec.h>
 #include <asm/fpsimd.h>
 #include <asm/mmu_context.h>
 #include <asm/processor.h>
@@ -165,6 +169,70 @@ void machine_restart(char *cmd)
 	while (1);
 }
 
+/*
+ * dump a block of kernel memory from around the given address
+ */
+static void show_data(unsigned long addr, int nbytes, const char *name)
+{
+	int	i, j;
+	int	nlines;
+	u32	*p;
+
+	/*
+	 * don't attempt to dump non-kernel addresses or
+	 * values that are probably just small negative numbers
+	 */
+	if (addr < PAGE_OFFSET || addr > -256UL)
+		return;
+
+	printk("\n%s: %#lx:\n", name, addr);
+
+	/*
+	 * round address down to a 32 bit boundary
+	 * and always dump a multiple of 32 bytes
+	 */
+	p = (u32 *)(addr & ~(sizeof(u32) - 1));
+	nbytes += (addr & (sizeof(u32) - 1));
+	nlines = (nbytes + 31) / 32;
+
+
+	for (i = 0; i < nlines; i++) {
+		/*
+		 * just display low 16 bits of address to keep
+		 * each line of the dump < 80 characters
+		 */
+		printk("%04lx ", (unsigned long)p & 0xffff);
+		for (j = 0; j < 8; j++) {
+			u32	data;
+			if (probe_kernel_address(p, data)) {
+				printk(" ********");
+			} else {
+				printk(" %08x", data);
+			}
+			++p;
+		}
+		printk("\n");
+	}
+}
+
+static void show_extra_register_data(struct pt_regs *regs, int nbytes)
+{
+	mm_segment_t fs;
+	unsigned int i;
+
+	fs = get_fs();
+	set_fs(KERNEL_DS);
+	show_data(regs->pc - nbytes, nbytes * 2, "PC");
+	show_data(regs->regs[30] - nbytes, nbytes * 2, "LR");
+	show_data(regs->sp - nbytes, nbytes * 2, "SP");
+	for (i = 0; i < 30; i++) {
+		char name[4];
+		snprintf(name, sizeof(name), "X%u", i);
+		show_data(regs->regs[i] - nbytes, nbytes * 2, name);
+	}
+	set_fs(fs);
+}
+
 void __show_regs(struct pt_regs *regs)
 {
 	int i, top_reg;
@@ -191,6 +259,8 @@ void __show_regs(struct pt_regs *regs)
 		if (i % 2 == 0)
 			printk("\n");
 	}
+	if (!user_mode(regs))
+		show_extra_register_data(regs, 128);
 	printk("\n");
 }
 
@@ -313,7 +383,7 @@ static void tls_thread_switch(struct task_struct *next)
 }
 
 /* Restore the UAO state depending on next's addr_limit */
-static void uao_thread_switch(struct task_struct *next)
+void uao_thread_switch(struct task_struct *next)
 {
 	if (IS_ENABLED(CONFIG_ARM64_UAO)) {
 		if (task_thread_info(next)->addr_limit == KERNEL_DS)
@@ -323,6 +393,22 @@ static void uao_thread_switch(struct task_struct *next)
 	}
 }
 
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+/*
+ * We store our current task in sp_el0, which is clobbered by userspace. Keep a
+ * shadow copy so that we can restore this upon entry from userspace.
+ *
+ * This is *only* for exception entry from EL0, and is not valid until we
+ * __switch_to() a user task.
+ */
+DEFINE_PER_CPU(struct task_struct *, __entry_task);
+
+static void entry_task_switch(struct task_struct *next)
+{
+	__this_cpu_write(__entry_task, next);
+}
+#endif
+
 /*
  * Thread switching.
  */
@@ -335,6 +421,9 @@ struct task_struct *__switch_to(struct task_struct *prev,
 	tls_thread_switch(next);
 	hw_breakpoint_thread_switch(next);
 	contextidr_thread_switch(next);
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	entry_task_switch(next);
+#endif
 	uao_thread_switch(next);
 
 	/*
@@ -352,27 +441,35 @@ struct task_struct *__switch_to(struct task_struct *prev,
 unsigned long get_wchan(struct task_struct *p)
 {
 	struct stackframe frame;
-	unsigned long stack_page;
+	unsigned long stack_page, ret = 0;
 	int count = 0;
 	if (!p || p == current || p->state == TASK_RUNNING)
 		return 0;
 
+	stack_page = (unsigned long)try_get_task_stack(p);
+	if (!stack_page)
+		return 0;
+
 	frame.fp = thread_saved_fp(p);
 	frame.sp = thread_saved_sp(p);
 	frame.pc = thread_saved_pc(p);
 #ifdef CONFIG_FUNCTION_GRAPH_TRACER
 	frame.graph = p->curr_ret_stack;
 #endif
-	stack_page = (unsigned long)task_stack_page(p);
 	do {
 		if (frame.sp < stack_page ||
 		    frame.sp >= stack_page + THREAD_SIZE ||
 		    unwind_frame(p, &frame))
-			return 0;
-		if (!in_sched_functions(frame.pc))
-			return frame.pc;
+			goto out;
+		if (!in_sched_functions(frame.pc)) {
+			ret = frame.pc;
+			goto out;
+		}
 	} while (count ++ < 16);
-	return 0;
+
+out:
+	put_task_stack(p);
+	return ret;
 }
 
 unsigned long arch_align_stack(unsigned long sp)
diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c
index c5ef05959813..6204b7600d1b 100644
--- a/arch/arm64/kernel/ptrace.c
+++ b/arch/arm64/kernel/ptrace.c
@@ -327,13 +327,13 @@ static int ptrace_hbp_fill_attr_ctrl(unsigned int note_type,
 				     struct arch_hw_breakpoint_ctrl ctrl,
 				     struct perf_event_attr *attr)
 {
-	int err, len, type, disabled = !ctrl.enabled;
+	int err, len, type, offset, disabled = !ctrl.enabled;
 
 	attr->disabled = disabled;
 	if (disabled)
 		return 0;
 
-	err = arch_bp_generic_fields(ctrl, &len, &type);
+	err = arch_bp_generic_fields(ctrl, &len, &type, &offset);
 	if (err)
 		return err;
 
@@ -352,6 +352,7 @@ static int ptrace_hbp_fill_attr_ctrl(unsigned int note_type,
 
 	attr->bp_len	= len;
 	attr->bp_type	= type;
+	attr->bp_addr	+= offset;
 
 	return 0;
 }
@@ -404,7 +405,7 @@ static int ptrace_hbp_get_addr(unsigned int note_type,
 	if (IS_ERR(bp))
 		return PTR_ERR(bp);
 
-	*addr = bp ? bp->attr.bp_addr : 0;
+	*addr = bp ? counter_arch_bp(bp)->address : 0;
 	return 0;
 }
 
diff --git a/arch/arm64/kernel/return_address.c b/arch/arm64/kernel/return_address.c
index 1718706fde83..12a87f2600f2 100644
--- a/arch/arm64/kernel/return_address.c
+++ b/arch/arm64/kernel/return_address.c
@@ -12,6 +12,7 @@
 #include <linux/export.h>
 #include <linux/ftrace.h>
 
+#include <asm/stack_pointer.h>
 #include <asm/stacktrace.h>
 
 struct return_address_data {
diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
index 566f3b9b1588..f1fae3cbc4ca 100644
--- a/arch/arm64/kernel/setup.c
+++ b/arch/arm64/kernel/setup.c
@@ -200,7 +200,7 @@ static void __init request_standard_resources(void)
 	struct resource *res;
 
 	kernel_code.start   = virt_to_phys(_text);
-	kernel_code.end     = virt_to_phys(_etext - 1);
+	kernel_code.end     = virt_to_phys(__init_begin - 1);
 	kernel_data.start   = virt_to_phys(_sdata);
 	kernel_data.end     = virt_to_phys(_end - 1);
 
@@ -351,6 +351,19 @@ void __init setup_arch(char **cmdline_p)
 	smp_init_cpus();
 	smp_build_mpidr_hash();
 
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+	/*
+	 * Make sure thread_info.ttbr0 always generates translation
+	 * faults in case uaccess_enable() is inadvertently called by the init
+	 * thread.
+	 */
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	init_task.thread_info.ttbr0 = virt_to_phys(empty_zero_page);
+#else
+	init_thread_info.ttbr0 = virt_to_phys(empty_zero_page);
+#endif
+#endif
+
 #ifdef CONFIG_VT
 #if defined(CONFIG_VGA_CONSOLE)
 	conswitchp = &vga_con;
diff --git a/arch/arm64/kernel/sleep.S b/arch/arm64/kernel/sleep.S
index c2bf5a58039f..9696c5239a3a 100644
--- a/arch/arm64/kernel/sleep.S
+++ b/arch/arm64/kernel/sleep.S
@@ -97,6 +97,7 @@ ENTRY(__cpu_suspend_enter)
 ENDPROC(__cpu_suspend_enter)
 	.ltorg
 
+	.pushsection ".idmap.text", "ax"
 ENTRY(cpu_resume)
 	bl	el2_setup		// if in EL2 drop to EL1 cleanly
 	/* enable the MMU early - so we can access sleep_save_stash by va */
@@ -124,9 +125,6 @@ ENTRY(_cpu_resume)
 	/* load sp from context */
 	ldr	x2, [x0, #CPU_CTX_SP]
 	mov	sp, x2
-	/* save thread_info */
-	and	x2, x2, #~(THREAD_SIZE - 1)
-	msr	sp_el0, x2
 	/*
 	 * cpu_do_resume expects x0 to contain context address pointer
 	 */
diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
index a1d06fc42048..fb3a21e17bff 100644
--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -58,6 +58,9 @@
 #define CREATE_TRACE_POINTS
 #include <trace/events/ipi.h>
 
+DEFINE_PER_CPU_READ_MOSTLY(int, cpu_number);
+EXPORT_PER_CPU_SYMBOL(cpu_number);
+
 /*
  * as from 2.5, kernels no longer have an init_tasks structure
  * so we need some other way of telling a new secondary core
@@ -110,6 +113,9 @@ int __cpu_up(unsigned int cpu, struct task_struct *idle)
 	 * We need to tell the secondary core where to find its stack and the
 	 * page tables.
 	 */
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	secondary_data.task = idle;
+#endif
 	secondary_data.stack = task_stack_page(idle) + THREAD_START_SP;
 	update_cpu_boot_status(CPU_MMU_OFF);
 	__flush_dcache_area(&secondary_data, sizeof(secondary_data));
@@ -134,6 +140,9 @@ int __cpu_up(unsigned int cpu, struct task_struct *idle)
 		pr_err("CPU%u: failed to boot: %d\n", cpu, ret);
 	}
 
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	secondary_data.task = NULL;
+#endif
 	secondary_data.stack = NULL;
 	status = READ_ONCE(secondary_data.status);
 	if (ret && status) {
@@ -177,7 +186,10 @@ static void smp_store_cpu_info(unsigned int cpuid)
 asmlinkage void secondary_start_kernel(void)
 {
 	struct mm_struct *mm = &init_mm;
-	unsigned int cpu = smp_processor_id();
+	unsigned int cpu;
+
+	cpu = task_cpu(current);
+	set_my_cpu_offset(per_cpu_offset(cpu));
 
 	/*
 	 * All kernel threads share the same mm context; grab a
@@ -186,8 +198,6 @@ asmlinkage void secondary_start_kernel(void)
 	atomic_inc(&mm->mm_count);
 	current->active_mm = mm;
 
-	set_my_cpu_offset(per_cpu_offset(smp_processor_id()));
-
 	/*
 	 * TTBR0 is only used for the identity mapping at this stage. Make it
 	 * point to zero page to avoid speculatively fetching new entries.
@@ -676,6 +686,8 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 		if (max_cpus == 0)
 			break;
 
+		per_cpu(cpu_number, cpu) = cpu;
+
 		if (cpu == smp_processor_id())
 			continue;
 
diff --git a/arch/arm64/kernel/smp_spin_table.c b/arch/arm64/kernel/smp_spin_table.c
index aef3605a8c47..4b4b0ad6ebde 100644
--- a/arch/arm64/kernel/smp_spin_table.c
+++ b/arch/arm64/kernel/smp_spin_table.c
@@ -29,7 +29,8 @@
 #include <asm/smp_plat.h>
 
 extern void secondary_holding_pen(void);
-volatile unsigned long secondary_holding_pen_release = INVALID_HWID;
+volatile unsigned long __section(".mmuoff.data.read")
+secondary_holding_pen_release = INVALID_HWID;
 
 static phys_addr_t cpu_release_addr[NR_CPUS];
 
diff --git a/arch/arm64/kernel/stacktrace.c b/arch/arm64/kernel/stacktrace.c
index a99eff9afc1f..95b915d17afc 100644
--- a/arch/arm64/kernel/stacktrace.c
+++ b/arch/arm64/kernel/stacktrace.c
@@ -22,6 +22,7 @@
 #include <linux/stacktrace.h>
 
 #include <asm/irq.h>
+#include <asm/stack_pointer.h>
 #include <asm/stacktrace.h>
 
 /*
@@ -128,7 +129,6 @@ void notrace walk_stackframe(struct task_struct *tsk, struct stackframe *frame,
 			break;
 	}
 }
-EXPORT_SYMBOL(walk_stackframe);
 
 #ifdef CONFIG_STACKTRACE
 struct stack_trace_data {
@@ -160,6 +160,9 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 	struct stack_trace_data data;
 	struct stackframe frame;
 
+	if (!try_get_task_stack(tsk))
+		return;
+
 	data.trace = trace;
 	data.skip = trace->skip;
 
@@ -181,6 +184,8 @@ void save_stack_trace_tsk(struct task_struct *tsk, struct stack_trace *trace)
 	walk_stackframe(tsk, &frame, save_trace, &data);
 	if (trace->nr_entries < trace->max_entries)
 		trace->entries[trace->nr_entries++] = ULONG_MAX;
+
+	put_task_stack(tsk);
 }
 
 void save_stack_trace(struct stack_trace *trace)
diff --git a/arch/arm64/kernel/suspend.c b/arch/arm64/kernel/suspend.c
index 5a0b1088c17c..468b939f3471 100644
--- a/arch/arm64/kernel/suspend.c
+++ b/arch/arm64/kernel/suspend.c
@@ -5,6 +5,7 @@
 #include <asm/cacheflush.h>
 #include <asm/cpufeature.h>
 #include <asm/debug-monitors.h>
+#include <asm/exec.h>
 #include <asm/pgtable.h>
 #include <asm/memory.h>
 #include <asm/mmu_context.h>
@@ -44,12 +45,6 @@ void notrace __cpu_suspend_exit(void)
 	cpu_uninstall_idmap();
 
 	/*
-	 * Restore per-cpu offset before any kernel
-	 * subsystem relying on it has a chance to run.
-	 */
-	set_my_cpu_offset(per_cpu_offset(smp_processor_id()));
-
-	/*
 	 * Restore HW breakpoint registers to sane values
 	 * before debug exceptions are possibly reenabled
 	 * through local_dbg_restore.
@@ -95,6 +90,7 @@ int cpu_suspend(unsigned long arg, int (*fn)(unsigned long))
 		 */
 		asm(ALTERNATIVE("nop", SET_PSTATE_PAN(1), ARM64_HAS_PAN,
 				CONFIG_ARM64_PAN));
+		uao_thread_switch(current);
 
 		/*
 		 * Restore HW breakpoint registers to sane values
diff --git a/arch/arm64/kernel/topology.c b/arch/arm64/kernel/topology.c
index 694f6deedbab..7758f7ff131b 100644
--- a/arch/arm64/kernel/topology.c
+++ b/arch/arm64/kernel/topology.c
@@ -19,10 +19,30 @@
 #include <linux/nodemask.h>
 #include <linux/of.h>
 #include <linux/sched.h>
+#include <linux/sched.h>
+#include <linux/sched_energy.h>
 
 #include <asm/cputype.h>
 #include <asm/topology.h>
 
+static DEFINE_PER_CPU(unsigned long, cpu_scale) = SCHED_CAPACITY_SCALE;
+
+unsigned long scale_cpu_capacity(struct sched_domain *sd, int cpu)
+{
+#ifdef CONFIG_CPU_FREQ
+	unsigned long max_freq_scale = cpufreq_scale_max_freq_capacity(cpu);
+
+	return per_cpu(cpu_scale, cpu) * max_freq_scale >> SCHED_CAPACITY_SHIFT;
+#else
+	return per_cpu(cpu_scale, cpu);
+#endif
+}
+
+static void set_capacity_scale(unsigned int cpu, unsigned long capacity)
+{
+	per_cpu(cpu_scale, cpu) = capacity;
+}
+
 static int __init get_cpu_for_node(struct device_node *node)
 {
 	struct device_node *cpu_node;
@@ -206,11 +226,72 @@ out:
 struct cpu_topology cpu_topology[NR_CPUS];
 EXPORT_SYMBOL_GPL(cpu_topology);
 
+/* sd energy functions */
+static inline
+const struct sched_group_energy * const cpu_cluster_energy(int cpu)
+{
+	struct sched_group_energy *sge = sge_array[cpu][SD_LEVEL1];
+
+	if (!sge) {
+		pr_warn("Invalid sched_group_energy for Cluster%d\n", cpu);
+		return NULL;
+	}
+
+	return sge;
+}
+
+static inline
+const struct sched_group_energy * const cpu_core_energy(int cpu)
+{
+	struct sched_group_energy *sge = sge_array[cpu][SD_LEVEL0];
+
+	if (!sge) {
+		pr_warn("Invalid sched_group_energy for CPU%d\n", cpu);
+		return NULL;
+	}
+
+	return sge;
+}
+
 const struct cpumask *cpu_coregroup_mask(int cpu)
 {
 	return &cpu_topology[cpu].core_sibling;
 }
 
+static int cpu_cpu_flags(void)
+{
+	return SD_ASYM_CPUCAPACITY;
+}
+
+static inline int cpu_corepower_flags(void)
+{
+	return SD_SHARE_PKG_RESOURCES  | SD_SHARE_POWERDOMAIN | \
+	       SD_SHARE_CAP_STATES;
+}
+
+static struct sched_domain_topology_level arm64_topology[] = {
+#ifdef CONFIG_SCHED_MC
+	{ cpu_coregroup_mask, cpu_corepower_flags, cpu_core_energy, SD_INIT_NAME(MC) },
+#endif
+	{ cpu_cpu_mask, cpu_cpu_flags, cpu_cluster_energy, SD_INIT_NAME(DIE) },
+	{ NULL, },
+};
+
+static void update_cpu_capacity(unsigned int cpu)
+{
+	unsigned long capacity = SCHED_CAPACITY_SCALE;
+
+	if (cpu_core_energy(cpu)) {
+		int max_cap_idx = cpu_core_energy(cpu)->nr_cap_states - 1;
+		capacity = cpu_core_energy(cpu)->cap_states[max_cap_idx].cap;
+	}
+
+	set_capacity_scale(cpu, capacity);
+
+	pr_info("CPU%d: update cpu_capacity %lu\n",
+		cpu, arch_scale_cpu_capacity(NULL, cpu));
+}
+
 static void update_siblings_masks(unsigned int cpuid)
 {
 	struct cpu_topology *cpu_topo, *cpuid_topo = &cpu_topology[cpuid];
@@ -272,6 +353,7 @@ void store_cpu_topology(unsigned int cpuid)
 
 topology_populated:
 	update_siblings_masks(cpuid);
+	update_cpu_capacity(cpuid);
 }
 
 static void __init reset_cpu_topology(void)
@@ -302,4 +384,8 @@ void __init init_cpu_topology(void)
 	 */
 	if (of_have_populated_dt() && parse_dt_topology())
 		reset_cpu_topology();
+	else
+		set_sched_topology(arm64_topology);
+
+	init_sched_energy_costs();
 }
diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c
index b62ec1e8a843..6543cd7bcb93 100644
--- a/arch/arm64/kernel/traps.c
+++ b/arch/arm64/kernel/traps.c
@@ -38,6 +38,7 @@
 #include <asm/esr.h>
 #include <asm/insn.h>
 #include <asm/traps.h>
+#include <asm/stack_pointer.h>
 #include <asm/stacktrace.h>
 #include <asm/exception.h>
 #include <asm/system_misc.h>
@@ -154,6 +155,14 @@ static void dump_backtrace(struct pt_regs *regs, struct task_struct *tsk)
 	if (!tsk)
 		tsk = current;
 
+	pr_debug("%s(regs = %p tsk = %p)\n", __func__, regs, tsk);
+
+	if (!tsk)
+		tsk = current;
+
+	if (!try_get_task_stack(tsk))
+		return;
+
 	/*
 	 * Switching between stacks is valid when tracing current and in
 	 * non-preemptible context.
@@ -219,6 +228,8 @@ static void dump_backtrace(struct pt_regs *regs, struct task_struct *tsk)
 				 stack + sizeof(struct pt_regs), false);
 		}
 	}
+
+	put_task_stack(tsk);
 }
 
 void show_stack(struct task_struct *tsk, unsigned long *sp)
@@ -234,10 +245,9 @@ void show_stack(struct task_struct *tsk, unsigned long *sp)
 #endif
 #define S_SMP " SMP"
 
-static int __die(const char *str, int err, struct thread_info *thread,
-		 struct pt_regs *regs)
+static int __die(const char *str, int err, struct pt_regs *regs)
 {
-	struct task_struct *tsk = thread->task;
+	struct task_struct *tsk = current;
 	static int die_counter;
 	int ret;
 
@@ -252,7 +262,8 @@ static int __die(const char *str, int err, struct thread_info *thread,
 	print_modules();
 	__show_regs(regs);
 	pr_emerg("Process %.*s (pid: %d, stack limit = 0x%p)\n",
-		 TASK_COMM_LEN, tsk->comm, task_pid_nr(tsk), thread + 1);
+		 TASK_COMM_LEN, tsk->comm, task_pid_nr(tsk),
+		 end_of_stack(tsk));
 
 	if (!user_mode(regs) || in_interrupt()) {
 		dump_mem(KERN_EMERG, "Stack: ", regs->sp,
@@ -272,7 +283,6 @@ static DEFINE_RAW_SPINLOCK(die_lock);
  */
 void die(const char *str, struct pt_regs *regs, int err)
 {
-	struct thread_info *thread = current_thread_info();
 	int ret;
 
 	oops_enter();
@@ -280,9 +290,9 @@ void die(const char *str, struct pt_regs *regs, int err)
 	raw_spin_lock_irq(&die_lock);
 	console_verbose();
 	bust_spinlocks(1);
-	ret = __die(str, err, thread, regs);
+	ret = __die(str, err, regs);
 
-	if (regs && kexec_should_crash(thread->task))
+	if (regs && kexec_should_crash(current))
 		crash_kexec(regs);
 
 	bust_spinlocks(0);
@@ -465,7 +475,7 @@ static const char *esr_class_str[] = {
 
 const char *esr_get_class_string(u32 esr)
 {
-	return esr_class_str[esr >> ESR_ELx_EC_SHIFT];
+	return esr_class_str[ESR_ELx_EC(esr)];
 }
 
 /*
diff --git a/arch/arm64/kernel/vdso.c b/arch/arm64/kernel/vdso.c
index 97bc68f4c689..3b8acfae7797 100644
--- a/arch/arm64/kernel/vdso.c
+++ b/arch/arm64/kernel/vdso.c
@@ -55,7 +55,7 @@ struct vdso_data *vdso_data = &vdso_data_store.data;
  */
 static struct page *vectors_page[1];
 
-static int alloc_vectors_page(void)
+static int __init alloc_vectors_page(void)
 {
 	extern char __kuser_helper_start[], __kuser_helper_end[];
 	extern char __aarch32_sigret_code_start[], __aarch32_sigret_code_end[];
@@ -88,7 +88,7 @@ int aarch32_setup_vectors_page(struct linux_binprm *bprm, int uses_interp)
 {
 	struct mm_struct *mm = current->mm;
 	unsigned long addr = AARCH32_VECTORS_BASE;
-	static struct vm_special_mapping spec = {
+	static const struct vm_special_mapping spec = {
 		.name	= "[vectors]",
 		.pages	= vectors_page,
 
@@ -212,10 +212,16 @@ void update_vsyscall(struct timekeeper *tk)
 	vdso_data->wtm_clock_nsec		= tk->wall_to_monotonic.tv_nsec;
 
 	if (!use_syscall) {
+		/* tkr_mono.cycle_last == tkr_raw.cycle_last */
 		vdso_data->cs_cycle_last	= tk->tkr_mono.cycle_last;
+		vdso_data->raw_time_sec		= tk->raw_time.tv_sec;
+		vdso_data->raw_time_nsec	= tk->raw_time.tv_nsec;
 		vdso_data->xtime_clock_sec	= tk->xtime_sec;
 		vdso_data->xtime_clock_nsec	= tk->tkr_mono.xtime_nsec;
-		vdso_data->cs_mult		= tk->tkr_mono.mult;
+		/* tkr_raw.xtime_nsec == 0 */
+		vdso_data->cs_mono_mult		= tk->tkr_mono.mult;
+		vdso_data->cs_raw_mult		= tk->tkr_raw.mult;
+		/* tkr_mono.shift == tkr_raw.shift */
 		vdso_data->cs_shift		= tk->tkr_mono.shift;
 	}
 
diff --git a/arch/arm64/kernel/vdso/Makefile b/arch/arm64/kernel/vdso/Makefile
index b467fd0a384b..62c84f7cb01b 100644
--- a/arch/arm64/kernel/vdso/Makefile
+++ b/arch/arm64/kernel/vdso/Makefile
@@ -23,7 +23,7 @@ GCOV_PROFILE := n
 ccflags-y += -Wl,-shared
 
 obj-y += vdso.o
-extra-y += vdso.lds vdso-offsets.h
+extra-y += vdso.lds
 CPPFLAGS_vdso.lds += -P -C -U$(ARCH)
 
 # Force dependency (incbin is bad)
@@ -42,11 +42,10 @@ $(obj)/%.so: $(obj)/%.so.dbg FORCE
 gen-vdsosym := $(srctree)/$(src)/gen_vdso_offsets.sh
 quiet_cmd_vdsosym = VDSOSYM $@
 define cmd_vdsosym
-	$(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@ && \
-	cp $@ include/generated/
+	$(NM) $< | $(gen-vdsosym) | LC_ALL=C sort > $@
 endef
 
-$(obj)/vdso-offsets.h: $(obj)/vdso.so.dbg FORCE
+include/generated/vdso-offsets.h: $(obj)/vdso.so.dbg FORCE
 	$(call if_changed,vdsosym)
 
 # Assembly rules for the .S files
diff --git a/arch/arm64/kernel/vdso/gettimeofday.S b/arch/arm64/kernel/vdso/gettimeofday.S
index efa79e8d4196..e00b4671bd7c 100644
--- a/arch/arm64/kernel/vdso/gettimeofday.S
+++ b/arch/arm64/kernel/vdso/gettimeofday.S
@@ -26,24 +26,109 @@
 #define NSEC_PER_SEC_HI16	0x3b9a
 
 vdso_data	.req	x6
-use_syscall	.req	w7
-seqcnt		.req	w8
+seqcnt		.req	w7
+w_tmp		.req	w8
+x_tmp		.req	x8
+
+/*
+ * Conventions for macro arguments:
+ * - An argument is write-only if its name starts with "res".
+ * - All other arguments are read-only, unless otherwise specified.
+ */
 
 	.macro	seqcnt_acquire
 9999:	ldr	seqcnt, [vdso_data, #VDSO_TB_SEQ_COUNT]
 	tbnz	seqcnt, #0, 9999b
 	dmb	ishld
-	ldr	use_syscall, [vdso_data, #VDSO_USE_SYSCALL]
 	.endm
 
-	.macro	seqcnt_read, cnt
+	.macro	seqcnt_check fail
 	dmb	ishld
-	ldr	\cnt, [vdso_data, #VDSO_TB_SEQ_COUNT]
+	ldr	w_tmp, [vdso_data, #VDSO_TB_SEQ_COUNT]
+	cmp	w_tmp, seqcnt
+	b.ne	\fail
 	.endm
 
-	.macro	seqcnt_check, cnt, fail
-	cmp	\cnt, seqcnt
-	b.ne	\fail
+	.macro	syscall_check fail
+	ldr	w_tmp, [vdso_data, #VDSO_USE_SYSCALL]
+	cbnz	w_tmp, \fail
+	.endm
+
+	.macro get_nsec_per_sec res
+	mov	\res, #NSEC_PER_SEC_LO16
+	movk	\res, #NSEC_PER_SEC_HI16, lsl #16
+	.endm
+
+	/*
+	 * Returns the clock delta, in nanoseconds left-shifted by the clock
+	 * shift.
+	 */
+	.macro	get_clock_shifted_nsec res, cycle_last, mult
+	/* Read the virtual counter. */
+	isb
+	mrs	x_tmp, cntvct_el0
+	/* Calculate cycle delta and convert to ns. */
+	sub	\res, x_tmp, \cycle_last
+	/* We can only guarantee 56 bits of precision. */
+	movn	x_tmp, #0xff00, lsl #48
+	and	\res, x_tmp, \res
+	mul	\res, \res, \mult
+	.endm
+
+	/*
+	 * Returns in res_{sec,nsec} the REALTIME timespec, based on the
+	 * "wall time" (xtime) and the clock_mono delta.
+	 */
+	.macro	get_ts_realtime res_sec, res_nsec, \
+			clock_nsec, xtime_sec, xtime_nsec, nsec_to_sec
+	add	\res_nsec, \clock_nsec, \xtime_nsec
+	udiv	x_tmp, \res_nsec, \nsec_to_sec
+	add	\res_sec, \xtime_sec, x_tmp
+	msub	\res_nsec, x_tmp, \nsec_to_sec, \res_nsec
+	.endm
+
+	/*
+	 * Returns in res_{sec,nsec} the timespec based on the clock_raw delta,
+	 * used for CLOCK_MONOTONIC_RAW.
+	 */
+	.macro	get_ts_clock_raw res_sec, res_nsec, clock_nsec, nsec_to_sec
+	udiv	\res_sec, \clock_nsec, \nsec_to_sec
+	msub	\res_nsec, \res_sec, \nsec_to_sec, \clock_nsec
+	.endm
+
+	/* sec and nsec are modified in place. */
+	.macro add_ts sec, nsec, ts_sec, ts_nsec, nsec_to_sec
+	/* Add timespec. */
+	add	\sec, \sec, \ts_sec
+	add	\nsec, \nsec, \ts_nsec
+
+	/* Normalise the new timespec. */
+	cmp	\nsec, \nsec_to_sec
+	b.lt	9999f
+	sub	\nsec, \nsec, \nsec_to_sec
+	add	\sec, \sec, #1
+9999:
+	cmp	\nsec, #0
+	b.ge	9998f
+	add	\nsec, \nsec, \nsec_to_sec
+	sub	\sec, \sec, #1
+9998:
+	.endm
+
+	.macro clock_gettime_return, shift=0
+	.if \shift == 1
+	lsr	x11, x11, x12
+	.endif
+	stp	x10, x11, [x1, #TSPEC_TV_SEC]
+	mov	x0, xzr
+	ret
+	.endm
+
+	.macro jump_slot jumptable, index, label
+	.if (. - \jumptable) != 4 * (\index)
+	.error "Jump slot index mismatch"
+	.endif
+	b	\label
 	.endm
 
 	.text
@@ -51,18 +136,25 @@ seqcnt		.req	w8
 /* int __kernel_gettimeofday(struct timeval *tv, struct timezone *tz); */
 ENTRY(__kernel_gettimeofday)
 	.cfi_startproc
-	mov	x2, x30
-	.cfi_register x30, x2
-
-	/* Acquire the sequence counter and get the timespec. */
 	adr	vdso_data, _vdso_data
-1:	seqcnt_acquire
-	cbnz	use_syscall, 4f
-
 	/* If tv is NULL, skip to the timezone code. */
 	cbz	x0, 2f
-	bl	__do_get_tspec
-	seqcnt_check w9, 1b
+
+	/* Compute the time of day. */
+1:	seqcnt_acquire
+	syscall_check fail=4f
+	ldr	x10, [vdso_data, #VDSO_CS_CYCLE_LAST]
+	/* w11 = cs_mono_mult, w12 = cs_shift */
+	ldp	w11, w12, [vdso_data, #VDSO_CS_MONO_MULT]
+	ldp	x13, x14, [vdso_data, #VDSO_XTIME_CLK_SEC]
+	seqcnt_check fail=1b
+
+	get_nsec_per_sec res=x9
+	lsl	x9, x9, x12
+
+	get_clock_shifted_nsec res=x15, cycle_last=x10, mult=x11
+	get_ts_realtime res_sec=x10, res_nsec=x11, \
+		clock_nsec=x15, xtime_sec=x13, xtime_nsec=x14, nsec_to_sec=x9
 
 	/* Convert ns to us. */
 	mov	x13, #1000
@@ -76,95 +168,126 @@ ENTRY(__kernel_gettimeofday)
 	stp	w4, w5, [x1, #TZ_MINWEST]
 3:
 	mov	x0, xzr
-	ret	x2
+	ret
 4:
 	/* Syscall fallback. */
 	mov	x8, #__NR_gettimeofday
 	svc	#0
-	ret	x2
+	ret
 	.cfi_endproc
 ENDPROC(__kernel_gettimeofday)
 
+#define JUMPSLOT_MAX CLOCK_MONOTONIC_COARSE
+
 /* int __kernel_clock_gettime(clockid_t clock_id, struct timespec *tp); */
 ENTRY(__kernel_clock_gettime)
 	.cfi_startproc
-	cmp	w0, #CLOCK_REALTIME
-	ccmp	w0, #CLOCK_MONOTONIC, #0x4, ne
-	b.ne	2f
+	cmp	w0, #JUMPSLOT_MAX
+	b.hi	syscall
+	adr	vdso_data, _vdso_data
+	adr	x_tmp, jumptable
+	add	x_tmp, x_tmp, w0, uxtw #2
+	br	x_tmp
+
+	ALIGN
+jumptable:
+	jump_slot jumptable, CLOCK_REALTIME, realtime
+	jump_slot jumptable, CLOCK_MONOTONIC, monotonic
+	b	syscall
+	b	syscall
+	jump_slot jumptable, CLOCK_MONOTONIC_RAW, monotonic_raw
+	jump_slot jumptable, CLOCK_REALTIME_COARSE, realtime_coarse
+	jump_slot jumptable, CLOCK_MONOTONIC_COARSE, monotonic_coarse
+
+	.if (. - jumptable) != 4 * (JUMPSLOT_MAX + 1)
+	.error	"Wrong jumptable size"
+	.endif
+
+	ALIGN
+realtime:
+	seqcnt_acquire
+	syscall_check fail=syscall
+	ldr	x10, [vdso_data, #VDSO_CS_CYCLE_LAST]
+	/* w11 = cs_mono_mult, w12 = cs_shift */
+	ldp	w11, w12, [vdso_data, #VDSO_CS_MONO_MULT]
+	ldp	x13, x14, [vdso_data, #VDSO_XTIME_CLK_SEC]
+	seqcnt_check fail=realtime
 
-	mov	x2, x30
-	.cfi_register x30, x2
+	/* All computations are done with left-shifted nsecs. */
+	get_nsec_per_sec res=x9
+	lsl	x9, x9, x12
 
-	/* Get kernel timespec. */
-	adr	vdso_data, _vdso_data
-1:	seqcnt_acquire
-	cbnz	use_syscall, 7f
+	get_clock_shifted_nsec res=x15, cycle_last=x10, mult=x11
+	get_ts_realtime res_sec=x10, res_nsec=x11, \
+		clock_nsec=x15, xtime_sec=x13, xtime_nsec=x14, nsec_to_sec=x9
+	clock_gettime_return, shift=1
 
-	bl	__do_get_tspec
-	seqcnt_check w9, 1b
+	ALIGN
+monotonic:
+	seqcnt_acquire
+	syscall_check fail=syscall
+	ldr	x10, [vdso_data, #VDSO_CS_CYCLE_LAST]
+	/* w11 = cs_mono_mult, w12 = cs_shift */
+	ldp	w11, w12, [vdso_data, #VDSO_CS_MONO_MULT]
+	ldp	x13, x14, [vdso_data, #VDSO_XTIME_CLK_SEC]
+	ldp	x3, x4, [vdso_data, #VDSO_WTM_CLK_SEC]
+	seqcnt_check fail=monotonic
 
-	mov	x30, x2
+	/* All computations are done with left-shifted nsecs. */
+	lsl	x4, x4, x12
+	get_nsec_per_sec res=x9
+	lsl	x9, x9, x12
 
-	cmp	w0, #CLOCK_MONOTONIC
-	b.ne	6f
+	get_clock_shifted_nsec res=x15, cycle_last=x10, mult=x11
+	get_ts_realtime res_sec=x10, res_nsec=x11, \
+		clock_nsec=x15, xtime_sec=x13, xtime_nsec=x14, nsec_to_sec=x9
 
-	/* Get wtm timespec. */
-	ldp	x13, x14, [vdso_data, #VDSO_WTM_CLK_SEC]
+	add_ts sec=x10, nsec=x11, ts_sec=x3, ts_nsec=x4, nsec_to_sec=x9
+	clock_gettime_return, shift=1
 
-	/* Check the sequence counter. */
-	seqcnt_read w9
-	seqcnt_check w9, 1b
-	b	4f
-2:
-	cmp	w0, #CLOCK_REALTIME_COARSE
-	ccmp	w0, #CLOCK_MONOTONIC_COARSE, #0x4, ne
-	b.ne	8f
+	ALIGN
+monotonic_raw:
+	seqcnt_acquire
+	syscall_check fail=syscall
+	ldr	x10, [vdso_data, #VDSO_CS_CYCLE_LAST]
+	/* w11 = cs_raw_mult, w12 = cs_shift */
+	ldp	w12, w11, [vdso_data, #VDSO_CS_SHIFT]
+	ldp	x13, x14, [vdso_data, #VDSO_RAW_TIME_SEC]
+	seqcnt_check fail=monotonic_raw
 
-	/* xtime_coarse_nsec is already right-shifted */
-	mov	x12, #0
+	/* All computations are done with left-shifted nsecs. */
+	lsl	x14, x14, x12
+	get_nsec_per_sec res=x9
+	lsl	x9, x9, x12
 
-	/* Get coarse timespec. */
-	adr	vdso_data, _vdso_data
-3:	seqcnt_acquire
+	get_clock_shifted_nsec res=x15, cycle_last=x10, mult=x11
+	get_ts_clock_raw res_sec=x10, res_nsec=x11, \
+		clock_nsec=x15, nsec_to_sec=x9
+
+	add_ts sec=x10, nsec=x11, ts_sec=x13, ts_nsec=x14, nsec_to_sec=x9
+	clock_gettime_return, shift=1
+
+	ALIGN
+realtime_coarse:
+	seqcnt_acquire
 	ldp	x10, x11, [vdso_data, #VDSO_XTIME_CRS_SEC]
+	seqcnt_check fail=realtime_coarse
+	clock_gettime_return
 
-	/* Get wtm timespec. */
+	ALIGN
+monotonic_coarse:
+	seqcnt_acquire
+	ldp	x10, x11, [vdso_data, #VDSO_XTIME_CRS_SEC]
 	ldp	x13, x14, [vdso_data, #VDSO_WTM_CLK_SEC]
+	seqcnt_check fail=monotonic_coarse
 
-	/* Check the sequence counter. */
-	seqcnt_read w9
-	seqcnt_check w9, 3b
+	/* Computations are done in (non-shifted) nsecs. */
+	get_nsec_per_sec res=x9
+	add_ts sec=x10, nsec=x11, ts_sec=x13, ts_nsec=x14, nsec_to_sec=x9
+	clock_gettime_return
 
-	cmp	w0, #CLOCK_MONOTONIC_COARSE
-	b.ne	6f
-4:
-	/* Add on wtm timespec. */
-	add	x10, x10, x13
-	lsl	x14, x14, x12
-	add	x11, x11, x14
-
-	/* Normalise the new timespec. */
-	mov	x15, #NSEC_PER_SEC_LO16
-	movk	x15, #NSEC_PER_SEC_HI16, lsl #16
-	lsl	x15, x15, x12
-	cmp	x11, x15
-	b.lt	5f
-	sub	x11, x11, x15
-	add	x10, x10, #1
-5:
-	cmp	x11, #0
-	b.ge	6f
-	add	x11, x11, x15
-	sub	x10, x10, #1
-
-6:	/* Store to the user timespec. */
-	lsr	x11, x11, x12
-	stp	x10, x11, [x1, #TSPEC_TV_SEC]
-	mov	x0, xzr
-	ret
-7:
-	mov	x30, x2
-8:	/* Syscall fallback. */
+	ALIGN
+syscall: /* Syscall fallback. */
 	mov	x8, #__NR_clock_gettime
 	svc	#0
 	ret
@@ -176,6 +299,7 @@ ENTRY(__kernel_clock_getres)
 	.cfi_startproc
 	cmp	w0, #CLOCK_REALTIME
 	ccmp	w0, #CLOCK_MONOTONIC, #0x4, ne
+	ccmp	w0, #CLOCK_MONOTONIC_RAW, #0x4, ne
 	b.ne	1f
 
 	ldr	x2, 5f
@@ -203,46 +327,3 @@ ENTRY(__kernel_clock_getres)
 	.quad	CLOCK_COARSE_RES
 	.cfi_endproc
 ENDPROC(__kernel_clock_getres)
-
-/*
- * Read the current time from the architected counter.
- * Expects vdso_data to be initialised.
- * Clobbers the temporary registers (x9 - x15).
- * Returns:
- *  - w9		= vDSO sequence counter
- *  - (x10, x11)	= (ts->tv_sec, shifted ts->tv_nsec)
- *  - w12		= cs_shift
- */
-ENTRY(__do_get_tspec)
-	.cfi_startproc
-
-	/* Read from the vDSO data page. */
-	ldr	x10, [vdso_data, #VDSO_CS_CYCLE_LAST]
-	ldp	x13, x14, [vdso_data, #VDSO_XTIME_CLK_SEC]
-	ldp	w11, w12, [vdso_data, #VDSO_CS_MULT]
-	seqcnt_read w9
-
-	/* Read the virtual counter. */
-	isb
-	mrs	x15, cntvct_el0
-
-	/* Calculate cycle delta and convert to ns. */
-	sub	x10, x15, x10
-	/* We can only guarantee 56 bits of precision. */
-	movn	x15, #0xff00, lsl #48
-	and	x10, x15, x10
-	mul	x10, x10, x11
-
-	/* Use the kernel time to calculate the new timespec. */
-	mov	x11, #NSEC_PER_SEC_LO16
-	movk	x11, #NSEC_PER_SEC_HI16, lsl #16
-	lsl	x11, x11, x12
-	add	x15, x10, x14
-	udiv	x14, x15, x11
-	add	x10, x13, x14
-	mul	x13, x14, x11
-	sub	x11, x15, x13
-
-	ret
-	.cfi_endproc
-ENDPROC(__do_get_tspec)
diff --git a/arch/arm64/kernel/vmlinux.lds.S b/arch/arm64/kernel/vmlinux.lds.S
index 623532f44323..f2d34d4c0a6b 100644
--- a/arch/arm64/kernel/vmlinux.lds.S
+++ b/arch/arm64/kernel/vmlinux.lds.S
@@ -133,12 +133,13 @@ SECTIONS
 	}
 
 	. = ALIGN(SEGMENT_ALIGN);
-	RO_DATA(PAGE_SIZE)		/* everything from this point to */
-	EXCEPTION_TABLE(8)		/* _etext will be marked RO NX   */
+	_etext = .;			/* End of text section */
+
+	RO_DATA(PAGE_SIZE)		/* everything from this point to     */
+	EXCEPTION_TABLE(8)		/* __init_begin will be marked RO NX */
 	NOTES
 
 	. = ALIGN(SEGMENT_ALIGN);
-	_etext = .;			/* End of text and rodata section */
 	__init_begin = .;
 
 	INIT_TEXT_SECTION(8)
@@ -182,6 +183,25 @@ SECTIONS
 	_data = .;
 	_sdata = .;
 	RW_DATA_SECTION(L1_CACHE_BYTES, PAGE_SIZE, THREAD_SIZE)
+
+	/*
+	 * Data written with the MMU off but read with the MMU on requires
+	 * cache lines to be invalidated, discarding up to a Cache Writeback
+	 * Granule (CWG) of data from the cache. Keep the section that
+	 * requires this type of maintenance to be in its own Cache Writeback
+	 * Granule (CWG) area so the cache maintenance operations don't
+	 * interfere with adjacent data.
+	 */
+	.mmuoff.data.write : ALIGN(SZ_2K) {
+		__mmuoff_data_start = .;
+		*(.mmuoff.data.write)
+	}
+	. = ALIGN(SZ_2K);
+	.mmuoff.data.read : {
+		*(.mmuoff.data.read)
+		__mmuoff_data_end = .;
+	}
+
 	PECOFF_EDATA_PADDING
 	_edata = .;
 
@@ -193,6 +213,11 @@ SECTIONS
 	swapper_pg_dir = .;
 	. += SWAPPER_DIR_SIZE;
 
+#ifdef CONFIG_ARM64_SW_TTBR0_PAN
+	reserved_ttbr0 = .;
+	. += RESERVED_TTBR0_SIZE;
+#endif
+
 	_end = .;
 
 	STABS_DEBUG
diff --git a/arch/arm64/lib/clear_user.S b/arch/arm64/lib/clear_user.S
index 5d1cad3ce6d6..d7150e30438a 100644
--- a/arch/arm64/lib/clear_user.S
+++ b/arch/arm64/lib/clear_user.S
@@ -17,10 +17,7 @@
  */
 #include <linux/linkage.h>
 
-#include <asm/alternative.h>
-#include <asm/assembler.h>
-#include <asm/cpufeature.h>
-#include <asm/sysreg.h>
+#include <asm/uaccess.h>
 
 	.text
 
@@ -33,8 +30,7 @@
  * Alignment fixed up by hardware.
  */
 ENTRY(__clear_user)
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(0)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_enable_not_uao x2, x3
 	mov	x2, x1			// save the size for fixup return
 	subs	x1, x1, #8
 	b.mi	2f
@@ -54,8 +50,7 @@ uao_user_alternative 9f, strh, sttrh, wzr, x0, 2
 	b.mi	5f
 uao_user_alternative 9f, strb, sttrb, wzr, x0, 0
 5:	mov	x0, #0
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(1)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_disable_not_uao x2
 	ret
 ENDPROC(__clear_user)
 
diff --git a/arch/arm64/lib/copy_from_user.S b/arch/arm64/lib/copy_from_user.S
index 0b90497d4424..90154f3f7f2a 100644
--- a/arch/arm64/lib/copy_from_user.S
+++ b/arch/arm64/lib/copy_from_user.S
@@ -16,11 +16,8 @@
 
 #include <linux/linkage.h>
 
-#include <asm/alternative.h>
-#include <asm/assembler.h>
 #include <asm/cache.h>
-#include <asm/cpufeature.h>
-#include <asm/sysreg.h>
+#include <asm/uaccess.h>
 
 /*
  * Copy from user space to a kernel buffer (alignment handled by the hardware)
@@ -67,12 +64,10 @@
 
 end	.req	x5
 ENTRY(__arch_copy_from_user)
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(0)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_enable_not_uao x3, x4
 	add	end, x0, x2
 #include "copy_template.S"
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(1)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_disable_not_uao x3
 	mov	x0, #0				// Nothing to copy
 	ret
 ENDPROC(__arch_copy_from_user)
diff --git a/arch/arm64/lib/copy_in_user.S b/arch/arm64/lib/copy_in_user.S
index f7292dd08c84..718b1c4e2f85 100644
--- a/arch/arm64/lib/copy_in_user.S
+++ b/arch/arm64/lib/copy_in_user.S
@@ -18,11 +18,8 @@
 
 #include <linux/linkage.h>
 
-#include <asm/alternative.h>
-#include <asm/assembler.h>
 #include <asm/cache.h>
-#include <asm/cpufeature.h>
-#include <asm/sysreg.h>
+#include <asm/uaccess.h>
 
 /*
  * Copy from user space to user space (alignment handled by the hardware)
@@ -68,12 +65,10 @@
 
 end	.req	x5
 ENTRY(__copy_in_user)
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(0)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_enable_not_uao x3, x4
 	add	end, x0, x2
 #include "copy_template.S"
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(1)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_disable_not_uao x3
 	mov	x0, #0
 	ret
 ENDPROC(__copy_in_user)
diff --git a/arch/arm64/lib/copy_to_user.S b/arch/arm64/lib/copy_to_user.S
index 7a7efe255034..e99e31c9acac 100644
--- a/arch/arm64/lib/copy_to_user.S
+++ b/arch/arm64/lib/copy_to_user.S
@@ -16,11 +16,8 @@
 
 #include <linux/linkage.h>
 
-#include <asm/alternative.h>
-#include <asm/assembler.h>
 #include <asm/cache.h>
-#include <asm/cpufeature.h>
-#include <asm/sysreg.h>
+#include <asm/uaccess.h>
 
 /*
  * Copy to user space from a kernel buffer (alignment handled by the hardware)
@@ -66,12 +63,10 @@
 
 end	.req	x5
 ENTRY(__arch_copy_to_user)
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(0)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_enable_not_uao x3, x4
 	add	end, x0, x2
 #include "copy_template.S"
-ALTERNATIVE("nop", __stringify(SET_PSTATE_PAN(1)), ARM64_ALT_PAN_NOT_UAO, \
-	    CONFIG_ARM64_PAN)
+	uaccess_disable_not_uao x3
 	mov	x0, #0
 	ret
 ENDPROC(__arch_copy_to_user)
diff --git a/arch/arm64/mm/cache.S b/arch/arm64/mm/cache.S
index 50ff9ba3a236..3be2cda5dbda 100644
--- a/arch/arm64/mm/cache.S
+++ b/arch/arm64/mm/cache.S
@@ -23,6 +23,7 @@
 #include <asm/assembler.h>
 #include <asm/cpufeature.h>
 #include <asm/alternative.h>
+#include <asm/uaccess.h>
 
 /*
  *	flush_icache_range(start,end)
@@ -48,11 +49,12 @@ ENTRY(flush_icache_range)
  *	- end     - virtual end address of region
  */
 ENTRY(__flush_cache_user_range)
+	uaccess_ttbr0_enable x2, x3
 	dcache_line_size x2, x3
 	sub	x3, x2, #1
 	bic	x4, x0, x3
 1:
-USER(9f, dc	cvau, x4	)		// clean D line to PoU
+user_alt 9f, "dc cvau, x4",  "dc civac, x4",  ARM64_WORKAROUND_CLEAN_CACHE
 	add	x4, x4, x2
 	cmp	x4, x1
 	b.lo	1b
@@ -69,10 +71,12 @@ USER(9f, ic	ivau, x4	)		// invalidate I line PoU
 	dsb	ish
 	isb
 	mov	x0, #0
+1:
+	uaccess_ttbr0_disable x1
 	ret
 9:
 	mov	x0, #-EFAULT
-	ret
+	b	1b
 ENDPROC(flush_icache_range)
 ENDPROC(__flush_cache_user_range)
 
diff --git a/arch/arm64/mm/context.c b/arch/arm64/mm/context.c
index 7275628ba59f..25128089c386 100644
--- a/arch/arm64/mm/context.c
+++ b/arch/arm64/mm/context.c
@@ -182,7 +182,12 @@ void check_and_switch_context(struct mm_struct *mm, unsigned int cpu)
 	raw_spin_unlock_irqrestore(&cpu_asid_lock, flags);
 
 switch_mm_fastpath:
-	cpu_switch_mm(mm->pgd, mm);
+	/*
+	 * Defer TTBR0_EL1 setting for user threads to uaccess_enable() when
+	 * emulating PAN.
+	 */
+	if (!system_uses_ttbr0_pan())
+		cpu_switch_mm(mm->pgd, mm);
 }
 
 static int asids_init(void)
diff --git a/arch/arm64/mm/dma-mapping.c b/arch/arm64/mm/dma-mapping.c
index a6e757cbab77..925b2b3a06f8 100644
--- a/arch/arm64/mm/dma-mapping.c
+++ b/arch/arm64/mm/dma-mapping.c
@@ -170,7 +170,7 @@ static void *__dma_alloc(struct device *dev, size_t size,
 	/* create a coherent mapping */
 	page = virt_to_page(ptr);
 	coherent_ptr = dma_common_contiguous_remap(page, size, VM_USERMAP,
-						   prot, NULL);
+						   prot, __builtin_return_address(0));
 	if (!coherent_ptr)
 		goto no_map;
 
diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
index 5cc1e418cb79..5613c5a29fd9 100644
--- a/arch/arm64/mm/fault.c
+++ b/arch/arm64/mm/fault.c
@@ -155,6 +155,11 @@ int ptep_set_access_flags(struct vm_area_struct *vma,
 }
 #endif
 
+static bool is_el1_instruction_abort(unsigned int esr)
+{
+	return ESR_ELx_EC(esr) == ESR_ELx_EC_IABT_CUR;
+}
+
 /*
  * The kernel tried to access some page that wasn't present.
  */
@@ -163,8 +168,9 @@ static void __do_kernel_fault(struct mm_struct *mm, unsigned long addr,
 {
 	/*
 	 * Are we prepared to handle this kernel fault?
+	 * We are almost certainly not prepared to handle instruction faults.
 	 */
-	if (fixup_exception(regs))
+	if (!is_el1_instruction_abort(esr) && fixup_exception(regs))
 		return;
 
 	/*
@@ -226,8 +232,6 @@ static void do_bad_area(unsigned long addr, unsigned int esr, struct pt_regs *re
 #define VM_FAULT_BADMAP		0x010000
 #define VM_FAULT_BADACCESS	0x020000
 
-#define ESR_LNX_EXEC		(1 << 24)
-
 static int __do_page_fault(struct mm_struct *mm, unsigned long addr,
 			   unsigned int mm_flags, unsigned long vm_flags,
 			   struct task_struct *tsk)
@@ -266,12 +270,24 @@ out:
 	return fault;
 }
 
-static inline int permission_fault(unsigned int esr)
+static inline bool is_permission_fault(unsigned int esr, struct pt_regs *regs)
 {
-	unsigned int ec       = (esr & ESR_ELx_EC_MASK) >> ESR_ELx_EC_SHIFT;
+	unsigned int ec       = ESR_ELx_EC(esr);
 	unsigned int fsc_type = esr & ESR_ELx_FSC_TYPE;
 
-	return (ec == ESR_ELx_EC_DABT_CUR && fsc_type == ESR_ELx_FSC_PERM);
+	if (ec != ESR_ELx_EC_DABT_CUR && ec != ESR_ELx_EC_IABT_CUR)
+		return false;
+
+	if (system_uses_ttbr0_pan())
+		return fsc_type == ESR_ELx_FSC_FAULT &&
+			(regs->pstate & PSR_PAN_BIT);
+	else
+		return fsc_type == ESR_ELx_FSC_PERM;
+}
+
+static bool is_el0_instruction_abort(unsigned int esr)
+{
+	return ESR_ELx_EC(esr) == ESR_ELx_EC_IABT_LOW;
 }
 
 static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
@@ -303,17 +319,20 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
 	if (user_mode(regs))
 		mm_flags |= FAULT_FLAG_USER;
 
-	if (esr & ESR_LNX_EXEC) {
+	if (is_el0_instruction_abort(esr)) {
 		vm_flags = VM_EXEC;
 	} else if ((esr & ESR_ELx_WNR) && !(esr & ESR_ELx_CM)) {
 		vm_flags = VM_WRITE;
 		mm_flags |= FAULT_FLAG_WRITE;
 	}
 
-	if (permission_fault(esr) && (addr < USER_DS)) {
+	if (addr < USER_DS && is_permission_fault(esr, regs)) {
 		if (get_fs() == KERNEL_DS)
 			die("Accessing user space memory with fs=KERNEL_DS", regs, esr);
 
+		if (is_el1_instruction_abort(esr))
+			die("Attempting to execute userspace memory", regs, esr);
+
 		if (!search_exception_tables(regs->pc))
 			die("Accessing user space memory outside uaccess.h routines", regs, esr);
 	}
@@ -468,7 +487,7 @@ static int do_bad(unsigned long addr, unsigned int esr, struct pt_regs *regs)
 	return 1;
 }
 
-static struct fault_info {
+static const struct fault_info {
 	int	(*fn)(unsigned long addr, unsigned int esr, struct pt_regs *regs);
 	int	sig;
 	int	code;
diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
index b2adff43bdaa..74ac8a90ba3f 100644
--- a/arch/arm64/mm/init.c
+++ b/arch/arm64/mm/init.c
@@ -266,9 +266,11 @@ static void __init zone_sizes_init(unsigned long min, unsigned long max)
 }
 
 #ifdef CONFIG_HAVE_ARCH_PFN_VALID
+#define PFN_MASK ((1UL << (64 - PAGE_SHIFT)) - 1)
+
 int pfn_valid(unsigned long pfn)
 {
-	return memblock_is_memory(pfn << PAGE_SHIFT);
+	return (pfn & PFN_MASK) == pfn && memblock_is_map_memory(pfn << PAGE_SHIFT);
 }
 EXPORT_SYMBOL(pfn_valid);
 #endif
@@ -364,8 +366,12 @@ void __init arm64_memblock_init(void)
 	 */
 	memblock_remove(max_t(u64, memstart_addr + linear_region_size, __pa(_end)),
 			ULLONG_MAX);
-	if (memblock_end_of_DRAM() > linear_region_size)
-		memblock_remove(0, memblock_end_of_DRAM() - linear_region_size);
+	if (memstart_addr + linear_region_size < memblock_end_of_DRAM()) {
+		/* ensure that memstart_addr remains sufficiently aligned */
+		memstart_addr = round_up(memblock_end_of_DRAM() - linear_region_size,
+					 ARM64_MEMSTART_ALIGN);
+		memblock_remove(0, memstart_addr);
+	}
 
 	/*
 	 * Apply the memory limit if it was set. Since the kernel may be loaded
@@ -566,8 +572,8 @@ void __init mem_init(void)
 		  MLM(MODULES_VADDR, MODULES_END),
 		  MLG(VMALLOC_START, VMALLOC_END),
 		  MLK_ROUNDUP(__init_begin, __init_end),
-		  MLK_ROUNDUP(_text, __start_rodata),
-		  MLK_ROUNDUP(__start_rodata, _etext),
+		  MLK_ROUNDUP(_text, _etext),
+		  MLK_ROUNDUP(__start_rodata, __init_begin),
 		  MLK_ROUNDUP(_sdata, _edata),
 #ifdef CONFIG_SPARSEMEM_VMEMMAP
 		  MLG(VMEMMAP_START,
diff --git a/arch/arm64/mm/mmap.c b/arch/arm64/mm/mmap.c
index ed177475dd8c..232f787a088a 100644
--- a/arch/arm64/mm/mmap.c
+++ b/arch/arm64/mm/mmap.c
@@ -51,8 +51,12 @@ unsigned long arch_mmap_rnd(void)
 {
 	unsigned long rnd;
 
-	rnd = (unsigned long)get_random_int() & STACK_RND_MASK;
-
+#ifdef CONFIG_COMPAT
+	if (test_thread_flag(TIF_32BIT))
+		rnd = get_random_long() & ((1UL << mmap_rnd_compat_bits) - 1);
+	else
+#endif
+		rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
 	return rnd << PAGE_SHIFT;
 }
 
diff --git a/arch/arm64/mm/mmu.c b/arch/arm64/mm/mmu.c
index d23342804230..2d2b6643a31e 100644
--- a/arch/arm64/mm/mmu.c
+++ b/arch/arm64/mm/mmu.c
@@ -381,7 +381,7 @@ static void __init __map_memblock(pgd_t *pgd, phys_addr_t start,
 static void __init map_mem(pgd_t *pgd)
 {
 	unsigned long kernel_start = __pa(_text);
-	unsigned long kernel_end = __pa(_etext);
+	unsigned long kernel_end = __pa(__init_begin);
 	struct memblock_region *reg;
 
 	/*
@@ -412,7 +412,7 @@ static void __init map_mem(pgd_t *pgd)
 	}
 
 	/*
-	 * Map the linear alias of the [_text, _etext) interval as
+	 * Map the linear alias of the [_text, __init_begin) interval as
 	 * read-only/non-executable. This makes the contents of the
 	 * region accessible to subsystems such as hibernate, but
 	 * protects it from inadvertent modification or execution.
@@ -440,14 +440,14 @@ void mark_rodata_ro(void)
 {
 	unsigned long section_size;
 
-	section_size = (unsigned long)__start_rodata - (unsigned long)_text;
+	section_size = (unsigned long)_etext - (unsigned long)_text;
 	create_mapping_late(__pa(_text), (unsigned long)_text,
 			    section_size, PAGE_KERNEL_ROX);
 	/*
-	 * mark .rodata as read only. Use _etext rather than __end_rodata to
-	 * cover NOTES and EXCEPTION_TABLE.
+	 * mark .rodata as read only. Use __init_begin rather than __end_rodata
+	 * to cover NOTES and EXCEPTION_TABLE.
 	 */
-	section_size = (unsigned long)_etext - (unsigned long)__start_rodata;
+	section_size = (unsigned long)__init_begin - (unsigned long)__start_rodata;
 	create_mapping_late(__pa(__start_rodata), (unsigned long)__start_rodata,
 			    section_size, PAGE_KERNEL_RO);
 }
@@ -490,8 +490,8 @@ static void __init map_kernel(pgd_t *pgd)
 {
 	static struct vm_struct vmlinux_text, vmlinux_rodata, vmlinux_init, vmlinux_data;
 
-	map_kernel_segment(pgd, _text, __start_rodata, PAGE_KERNEL_EXEC, &vmlinux_text);
-	map_kernel_segment(pgd, __start_rodata, _etext, PAGE_KERNEL, &vmlinux_rodata);
+	map_kernel_segment(pgd, _text, _etext, PAGE_KERNEL_EXEC, &vmlinux_text);
+	map_kernel_segment(pgd, __start_rodata, __init_begin, PAGE_KERNEL, &vmlinux_rodata);
 	map_kernel_segment(pgd, __init_begin, __init_end, PAGE_KERNEL_EXEC,
 			   &vmlinux_init);
 	map_kernel_segment(pgd, _data, _end, PAGE_KERNEL, &vmlinux_data);
diff --git a/arch/arm64/mm/pageattr.c b/arch/arm64/mm/pageattr.c
index f4e39dbdd36b..9ef0b013e6d4 100644
--- a/arch/arm64/mm/pageattr.c
+++ b/arch/arm64/mm/pageattr.c
@@ -152,4 +152,43 @@ void __kernel_map_pages(struct page *page, int numpages, int enable)
 					__pgprot(0),
 					__pgprot(PTE_VALID));
 }
-#endif
+#ifdef CONFIG_HIBERNATION
+/*
+ * When built with CONFIG_DEBUG_PAGEALLOC and CONFIG_HIBERNATION, this function
+ * is used to determine if a linear map page has been marked as not-valid by
+ * CONFIG_DEBUG_PAGEALLOC. Walk the page table and check the PTE_VALID bit.
+ * This is based on kern_addr_valid(), which almost does what we need.
+ *
+ * Because this is only called on the kernel linear map,  p?d_sect() implies
+ * p?d_present(). When debug_pagealloc is enabled, sections mappings are
+ * disabled.
+ */
+bool kernel_page_present(struct page *page)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+	unsigned long addr = (unsigned long)page_address(page);
+
+	pgd = pgd_offset_k(addr);
+	if (pgd_none(*pgd))
+		return false;
+
+	pud = pud_offset(pgd, addr);
+	if (pud_none(*pud))
+		return false;
+	if (pud_sect(*pud))
+		return true;
+
+	pmd = pmd_offset(pud, addr);
+	if (pmd_none(*pmd))
+		return false;
+	if (pmd_sect(*pmd))
+		return true;
+
+	pte = pte_offset_kernel(pmd, addr);
+	return pte_valid(*pte);
+}
+#endif /* CONFIG_HIBERNATION */
+#endif /* CONFIG_DEBUG_PAGEALLOC */
diff --git a/arch/arm64/mm/proc-macros.S b/arch/arm64/mm/proc-macros.S
deleted file mode 100644
index 984edcda1850..000000000000
--- a/arch/arm64/mm/proc-macros.S
+++ /dev/null
@@ -1,98 +0,0 @@
-/*
- * Based on arch/arm/mm/proc-macros.S
- *
- * Copyright (C) 2012 ARM Ltd.
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- *
- * This program is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
- *
- * You should have received a copy of the GNU General Public License
- * along with this program.  If not, see <http://www.gnu.org/licenses/>.
- */
-
-#include <asm/asm-offsets.h>
-#include <asm/thread_info.h>
-
-/*
- * vma_vm_mm - get mm pointer from vma pointer (vma->vm_mm)
- */
-	.macro	vma_vm_mm, rd, rn
-	ldr	\rd, [\rn, #VMA_VM_MM]
-	.endm
-
-/*
- * mmid - get context id from mm pointer (mm->context.id)
- */
-	.macro	mmid, rd, rn
-	ldr	\rd, [\rn, #MM_CONTEXT_ID]
-	.endm
-
-/*
- * dcache_line_size - get the minimum D-cache line size from the CTR register.
- */
-	.macro	dcache_line_size, reg, tmp
-	mrs	\tmp, ctr_el0			// read CTR
-	ubfm	\tmp, \tmp, #16, #19		// cache line size encoding
-	mov	\reg, #4			// bytes per word
-	lsl	\reg, \reg, \tmp		// actual cache line size
-	.endm
-
-/*
- * icache_line_size - get the minimum I-cache line size from the CTR register.
- */
-	.macro	icache_line_size, reg, tmp
-	mrs	\tmp, ctr_el0			// read CTR
-	and	\tmp, \tmp, #0xf		// cache line size encoding
-	mov	\reg, #4			// bytes per word
-	lsl	\reg, \reg, \tmp		// actual cache line size
-	.endm
-
-/*
- * tcr_set_idmap_t0sz - update TCR.T0SZ so that we can load the ID map
- */
-	.macro	tcr_set_idmap_t0sz, valreg, tmpreg
-#ifndef CONFIG_ARM64_VA_BITS_48
-	ldr_l	\tmpreg, idmap_t0sz
-	bfi	\valreg, \tmpreg, #TCR_T0SZ_OFFSET, #TCR_TxSZ_WIDTH
-#endif
-	.endm
-
-/*
- * reset_pmuserenr_el0 - reset PMUSERENR_EL0 if PMUv3 present
- */
-	.macro	reset_pmuserenr_el0, tmpreg
-	mrs	\tmpreg, id_aa64dfr0_el1	// Check ID_AA64DFR0_EL1 PMUVer
-	sbfx	\tmpreg, \tmpreg, #8, #4
-	cmp	\tmpreg, #1			// Skip if no PMU present
-	b.lt	9000f
-	msr	pmuserenr_el0, xzr		// Disable PMU access from EL0
-9000:
-	.endm
-
-/*
- * Macro to perform a data cache maintenance for the interval
- * [kaddr, kaddr + size)
- *
- * 	op:		operation passed to dc instruction
- * 	domain:		domain used in dsb instruciton
- * 	kaddr:		starting virtual address of the region
- * 	size:		size of the region
- * 	Corrupts: 	kaddr, size, tmp1, tmp2
- */
-	.macro dcache_by_line_op op, domain, kaddr, size, tmp1, tmp2
-	dcache_line_size \tmp1, \tmp2
-	add	\size, \kaddr, \size
-	sub	\tmp2, \tmp1, #1
-	bic	\kaddr, \kaddr, \tmp2
-9998:	dc	\op, \kaddr
-	add	\kaddr, \kaddr, \tmp1
-	cmp	\kaddr, \size
-	b.lo	9998b
-	dsb	\domain
-	.endm
diff --git a/arch/arm64/mm/proc.S b/arch/arm64/mm/proc.S
index 9d37e967fa19..32682be978e0 100644
--- a/arch/arm64/mm/proc.S
+++ b/arch/arm64/mm/proc.S
@@ -70,11 +70,14 @@ ENTRY(cpu_do_suspend)
 	mrs	x8, mdscr_el1
 	mrs	x9, oslsr_el1
 	mrs	x10, sctlr_el1
+	mrs	x11, tpidr_el1
+	mrs	x12, sp_el0
 	stp	x2, x3, [x0]
 	stp	x4, xzr, [x0, #16]
 	stp	x5, x6, [x0, #32]
 	stp	x7, x8, [x0, #48]
 	stp	x9, x10, [x0, #64]
+	stp	x11, x12, [x0, #80]
 	ret
 ENDPROC(cpu_do_suspend)
 
@@ -83,12 +86,14 @@ ENDPROC(cpu_do_suspend)
  *
  * x0: Address of context pointer
  */
+	.pushsection ".idmap.text", "ax"
 ENTRY(cpu_do_resume)
 	ldp	x2, x3, [x0]
 	ldp	x4, x5, [x0, #16]
 	ldp	x6, x8, [x0, #32]
 	ldp	x9, x10, [x0, #48]
 	ldp	x11, x12, [x0, #64]
+	ldp	x13, x14, [x0, #80]
 	msr	tpidr_el0, x2
 	msr	tpidrro_el0, x3
 	msr	contextidr_el1, x4
@@ -111,6 +116,8 @@ ENTRY(cpu_do_resume)
 	msr	mdscr_el1, x10
 
 	msr	sctlr_el1, x12
+	msr	tpidr_el1, x13
+	msr	sp_el0, x14
 	/*
 	 * Restore oslsr_el1 by writing oslar_el1
 	 */
@@ -120,6 +127,7 @@ ENTRY(cpu_do_resume)
 	isb
 	ret
 ENDPROC(cpu_do_resume)
+	.popsection
 #endif
 
 /*
@@ -134,17 +142,8 @@ ENTRY(cpu_do_switch_mm)
 	bfi	x0, x1, #48, #16		// set the ASID
 	msr	ttbr0_el1, x0			// set TTBR0
 	isb
-alternative_if_not ARM64_WORKAROUND_CAVIUM_27456
+	post_ttbr0_update_workaround
 	ret
-	nop
-	nop
-	nop
-alternative_else
-	ic	iallu
-	dsb	nsh
-	isb
-	ret
-alternative_endif
 ENDPROC(cpu_do_switch_mm)
 
 	.pushsection ".idmap.text", "ax"
@@ -181,6 +180,7 @@ ENDPROC(idmap_cpu_replace_ttbr1)
  *	Initialise the processor for turning the MMU on.  Return in x0 the
  *	value of the SCTLR_EL1 register.
  */
+	.pushsection ".idmap.text", "ax"
 ENTRY(__cpu_setup)
 	tlbi	vmalle1				// Invalidate local TLB
 	dsb	nsh
@@ -266,3 +266,4 @@ ENDPROC(__cpu_setup)
 crval:
 	.word	0xfcffffff			// clear
 	.word	0x34d5d91d			// set
+	.popsection
diff --git a/arch/arm64/xen/hypercall.S b/arch/arm64/xen/hypercall.S
index 8bbe9401f4f0..b96db5dafec4 100644
--- a/arch/arm64/xen/hypercall.S
+++ b/arch/arm64/xen/hypercall.S
@@ -49,6 +49,7 @@
 
 #include <linux/linkage.h>
 #include <asm/assembler.h>
+#include <asm/uaccess.h>
 #include <xen/interface/xen.h>
 
 
@@ -89,6 +90,20 @@ ENTRY(privcmd_call)
 	mov x2, x3
 	mov x3, x4
 	mov x4, x5
+	/*
+	 * Privcmd calls are issued by the userspace. The kernel needs to
+	 * enable access to TTBR0_EL1 as the hypervisor would issue stage 1
+	 * translations to user memory via AT instructions. Since AT
+	 * instructions are not affected by the PAN bit (ARMv8.1), we only
+	 * need the explicit uaccess_enable/disable if the TTBR0 PAN emulation
+	 * is enabled (it implies that hardware UAO and PAN disabled).
+	 */
+	uaccess_ttbr0_enable x6, x7
 	hvc XEN_IMM
+
+	/*
+	 * Disable userspace access from kernel once the hyp call completed.
+	 */
+	uaccess_ttbr0_disable x6
 	ret
 ENDPROC(privcmd_call);
diff --git a/arch/avr32/kernel/module.c b/arch/avr32/kernel/module.c
index 164efa009e5b..2b4c54c04cb6 100644
--- a/arch/avr32/kernel/module.c
+++ b/arch/avr32/kernel/module.c
@@ -118,9 +118,9 @@ int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
 	 * Increase core size to make room for GOT and set start
 	 * offset for GOT.
 	 */
-	module->core_size = ALIGN(module->core_size, 4);
-	module->arch.got_offset = module->core_size;
-	module->core_size += module->arch.got_size;
+	module->core_layout.size = ALIGN(module->core_layout.size, 4);
+	module->arch.got_offset = module->core_layout.size;
+	module->core_layout.size += module->arch.got_size;
 
 	return 0;
 
@@ -177,7 +177,7 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab,
 			if (!info->got_initialized) {
 				Elf32_Addr *gotent;
 
-				gotent = (module->module_core
+				gotent = (module->core_layout.base
 					  + module->arch.got_offset
 					  + info->got_offset);
 				*gotent = relocation;
@@ -255,8 +255,8 @@ int apply_relocate_add(Elf32_Shdr *sechdrs, const char *strtab,
 			 */
 			pr_debug("GOTPC: PC=0x%x, got_offset=0x%lx, core=0x%p\n",
 				 relocation, module->arch.got_offset,
-				 module->module_core);
-			relocation -= ((unsigned long)module->module_core
+				 module->core_layout.base);
+			relocation -= ((unsigned long)module->core_layout.base
 				       + module->arch.got_offset);
 			*location = relocation;
 			break;
diff --git a/arch/ia64/Kconfig b/arch/ia64/Kconfig
index 2c86a4ef6742..7091a367eeda 100644
--- a/arch/ia64/Kconfig
+++ b/arch/ia64/Kconfig
@@ -45,7 +45,7 @@ config IA64
 	select GENERIC_SMP_IDLE_THREAD
 	select ARCH_INIT_TASK
 	select ARCH_TASK_STRUCT_ALLOCATOR
-	select ARCH_THREAD_INFO_ALLOCATOR
+	select ARCH_THREAD_STACK_ALLOCATOR
 	select ARCH_CLOCKSOURCE_DATA
 	select GENERIC_TIME_VSYSCALL_OLD
 	select SYSCTL_ARCH_UNALIGN_NO_WARN
diff --git a/arch/ia64/Makefile b/arch/ia64/Makefile
index 970d0bd99621..648f1cef33fa 100644
--- a/arch/ia64/Makefile
+++ b/arch/ia64/Makefile
@@ -95,8 +95,8 @@ define archhelp
   echo '* unwcheck	- Check vmlinux for invalid unwind info'
 endef
 
-archprepare: make_nr_irqs_h FORCE
+archprepare: make_nr_irqs_h
 PHONY += make_nr_irqs_h FORCE
 
-make_nr_irqs_h: FORCE
+make_nr_irqs_h:
 	$(Q)$(MAKE) $(build)=arch/ia64/kernel include/generated/nr-irqs.h
diff --git a/arch/ia64/include/asm/early_ioremap.h b/arch/ia64/include/asm/early_ioremap.h
new file mode 100644
index 000000000000..eec9e1d1b833
--- /dev/null
+++ b/arch/ia64/include/asm/early_ioremap.h
@@ -0,0 +1,10 @@
+#ifndef _ASM_IA64_EARLY_IOREMAP_H
+#define _ASM_IA64_EARLY_IOREMAP_H
+
+extern void __iomem * early_ioremap (unsigned long phys_addr, unsigned long size);
+#define early_memremap(phys_addr, size)        early_ioremap(phys_addr, size)
+
+extern void early_iounmap (volatile void __iomem *addr, unsigned long size);
+#define early_memunmap(addr, size)             early_iounmap(addr, size)
+
+#endif
diff --git a/arch/ia64/include/asm/io.h b/arch/ia64/include/asm/io.h
index 8fdb9c7eeb66..5de673ac9cb1 100644
--- a/arch/ia64/include/asm/io.h
+++ b/arch/ia64/include/asm/io.h
@@ -20,6 +20,7 @@
  */
 
 #include <asm/unaligned.h>
+#include <asm/early_ioremap.h>
 
 /* We don't use IO slowdowns on the ia64, but.. */
 #define __SLOW_DOWN_IO	do { } while (0)
@@ -427,10 +428,6 @@ __writeq (unsigned long val, volatile void __iomem *addr)
 extern void __iomem * ioremap(unsigned long offset, unsigned long size);
 extern void __iomem * ioremap_nocache (unsigned long offset, unsigned long size);
 extern void iounmap (volatile void __iomem *addr);
-extern void __iomem * early_ioremap (unsigned long phys_addr, unsigned long size);
-#define early_memremap(phys_addr, size)        early_ioremap(phys_addr, size)
-extern void early_iounmap (volatile void __iomem *addr, unsigned long size);
-#define early_memunmap(addr, size)             early_iounmap(addr, size)
 static inline void __iomem * ioremap_cache (unsigned long phys_addr, unsigned long size)
 {
 	return ioremap(phys_addr, size);
diff --git a/arch/ia64/include/asm/thread_info.h b/arch/ia64/include/asm/thread_info.h
index aa995b67c3f5..d1212b84fb83 100644
--- a/arch/ia64/include/asm/thread_info.h
+++ b/arch/ia64/include/asm/thread_info.h
@@ -48,15 +48,15 @@ struct thread_info {
 #ifndef ASM_OFFSETS_C
 /* how to get the thread information struct from C */
 #define current_thread_info()	((struct thread_info *) ((char *) current + IA64_TASK_SIZE))
-#define alloc_thread_info_node(tsk, node)	\
-		((struct thread_info *) ((char *) (tsk) + IA64_TASK_SIZE))
+#define alloc_thread_stack_node(tsk, node)	\
+		((unsigned long *) ((char *) (tsk) + IA64_TASK_SIZE))
 #define task_thread_info(tsk)	((struct thread_info *) ((char *) (tsk) + IA64_TASK_SIZE))
 #else
 #define current_thread_info()	((struct thread_info *) 0)
-#define alloc_thread_info_node(tsk, node)	((struct thread_info *) 0)
+#define alloc_thread_stack_node(tsk, node)	((unsigned long *) 0)
 #define task_thread_info(tsk)	((struct thread_info *) 0)
 #endif
-#define free_thread_info(ti)	/* nothing */
+#define free_thread_stack(ti)	/* nothing */
 #define task_stack_page(tsk)	((void *)(tsk))
 
 #define __HAVE_THREAD_FUNCTIONS
diff --git a/arch/ia64/kernel/Makefile b/arch/ia64/kernel/Makefile
index 3686d6abafde..9edda5466020 100644
--- a/arch/ia64/kernel/Makefile
+++ b/arch/ia64/kernel/Makefile
@@ -50,32 +50,10 @@ CFLAGS_traps.o  += -mfixed-range=f2-f5,f16-f31
 # The gate DSO image is built using a special linker script.
 include $(src)/Makefile.gate
 
-# Calculate NR_IRQ = max(IA64_NATIVE_NR_IRQS, XEN_NR_IRQS, ...) based on config
-define sed-y
-	"/^->/{s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; s:->::; p;}"
-endef
-quiet_cmd_nr_irqs = GEN     $@
-define cmd_nr_irqs
-	(set -e; \
-	 echo "#ifndef __ASM_NR_IRQS_H__"; \
-	 echo "#define __ASM_NR_IRQS_H__"; \
-	 echo "/*"; \
-	 echo " * DO NOT MODIFY."; \
-	 echo " *"; \
-	 echo " * This file was generated by Kbuild"; \
-	 echo " *"; \
-	 echo " */"; \
-	 echo ""; \
-	 sed -ne $(sed-y) $<; \
-	 echo ""; \
-	 echo "#endif" ) > $@
-endef
-
 # We use internal kbuild rules to avoid the "is up to date" message from make
 arch/$(SRCARCH)/kernel/nr-irqs.s: arch/$(SRCARCH)/kernel/nr-irqs.c
 	$(Q)mkdir -p $(dir $@)
 	$(call if_changed_dep,cc_s_c)
 
-include/generated/nr-irqs.h: arch/$(SRCARCH)/kernel/nr-irqs.s
-	$(Q)mkdir -p $(dir $@)
-	$(call cmd,nr_irqs)
+include/generated/nr-irqs.h: arch/$(SRCARCH)/kernel/nr-irqs.s FORCE
+	$(call filechk,offsets,__ASM_NR_IRQS_H__)
diff --git a/arch/ia64/kernel/init_task.c b/arch/ia64/kernel/init_task.c
index f9efe9739d3f..0eaa89f3defd 100644
--- a/arch/ia64/kernel/init_task.c
+++ b/arch/ia64/kernel/init_task.c
@@ -26,6 +26,7 @@ static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
  * handled. This is done by having a special ".data..init_task" section...
  */
 #define init_thread_info	init_task_mem.s.thread_info
+#define init_stack		init_task_mem.stack
 
 union {
 	struct {
diff --git a/arch/ia64/kernel/module.c b/arch/ia64/kernel/module.c
index b15933c31b2f..6ab0ae7d6535 100644
--- a/arch/ia64/kernel/module.c
+++ b/arch/ia64/kernel/module.c
@@ -486,13 +486,13 @@ module_frob_arch_sections (Elf_Ehdr *ehdr, Elf_Shdr *sechdrs, char *secstrings,
 static inline int
 in_init (const struct module *mod, uint64_t addr)
 {
-	return addr - (uint64_t) mod->module_init < mod->init_size;
+	return addr - (uint64_t) mod->init_layout.base < mod->init_layout.size;
 }
 
 static inline int
 in_core (const struct module *mod, uint64_t addr)
 {
-	return addr - (uint64_t) mod->module_core < mod->core_size;
+	return addr - (uint64_t) mod->core_layout.base < mod->core_layout.size;
 }
 
 static inline int
@@ -675,7 +675,7 @@ do_reloc (struct module *mod, uint8_t r_type, Elf64_Sym *sym, uint64_t addend,
 		break;
 
 	      case RV_BDREL:
-		val -= (uint64_t) (in_init(mod, val) ? mod->module_init : mod->module_core);
+		val -= (uint64_t) (in_init(mod, val) ? mod->init_layout.base : mod->core_layout.base);
 		break;
 
 	      case RV_LTV:
@@ -810,15 +810,15 @@ apply_relocate_add (Elf64_Shdr *sechdrs, const char *strtab, unsigned int symind
 		 *     addresses have been selected...
 		 */
 		uint64_t gp;
-		if (mod->core_size > MAX_LTOFF)
+		if (mod->core_layout.size > MAX_LTOFF)
 			/*
 			 * This takes advantage of fact that SHF_ARCH_SMALL gets allocated
 			 * at the end of the module.
 			 */
-			gp = mod->core_size - MAX_LTOFF / 2;
+			gp = mod->core_layout.size - MAX_LTOFF / 2;
 		else
-			gp = mod->core_size / 2;
-		gp = (uint64_t) mod->module_core + ((gp + 7) & -8);
+			gp = mod->core_layout.size / 2;
+		gp = (uint64_t) mod->core_layout.base + ((gp + 7) & -8);
 		mod->arch.gp = gp;
 		DEBUGP("%s: placing gp at 0x%lx\n", __func__, gp);
 	}
diff --git a/arch/metag/kernel/module.c b/arch/metag/kernel/module.c
index 986331cd0a52..bb8dfba9a763 100644
--- a/arch/metag/kernel/module.c
+++ b/arch/metag/kernel/module.c
@@ -176,8 +176,8 @@ static uint32_t do_plt_call(void *location, Elf32_Addr val,
 	tramp[1] = 0xac000001 | ((val & 0x0000ffff) << 3);
 
 	/* Init, or core PLT? */
-	if (location >= mod->module_core
-	    && location < mod->module_core + mod->core_size)
+	if (location >= mod->core_layout.base
+	    && location < mod->core_layout.base + mod->core_layout.size)
 		entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr;
 	else
 		entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr;
diff --git a/arch/mips/kernel/vpe.c b/arch/mips/kernel/vpe.c
index 9067b651c7a2..544ea21bfef9 100644
--- a/arch/mips/kernel/vpe.c
+++ b/arch/mips/kernel/vpe.c
@@ -205,11 +205,11 @@ static void layout_sections(struct module *mod, const Elf_Ehdr *hdr,
 			    || s->sh_entsize != ~0UL)
 				continue;
 			s->sh_entsize =
-				get_offset((unsigned long *)&mod->core_size, s);
+				get_offset((unsigned long *)&mod->core_layout.size, s);
 		}
 
 		if (m == 0)
-			mod->core_text_size = mod->core_size;
+			mod->core_layout.text_size = mod->core_layout.size;
 
 	}
 }
@@ -641,7 +641,7 @@ static int vpe_elfload(struct vpe *v)
 		layout_sections(&mod, hdr, sechdrs, secstrings);
 	}
 
-	v->load_addr = alloc_progmem(mod.core_size);
+	v->load_addr = alloc_progmem(mod.core_layout.size);
 	if (!v->load_addr)
 		return -ENOMEM;
 
diff --git a/arch/mips/mm/mmap.c b/arch/mips/mm/mmap.c
index 025cb31aa0a2..c5fdea5debe5 100644
--- a/arch/mips/mm/mmap.c
+++ b/arch/mips/mm/mmap.c
@@ -146,7 +146,7 @@ unsigned long arch_mmap_rnd(void)
 {
 	unsigned long rnd;
 
-	rnd = (unsigned long)get_random_int();
+	rnd = get_random_long();
 	rnd <<= PAGE_SHIFT;
 	if (TASK_IS_32BIT_ADDR)
 		rnd &= 0xfffffful;
@@ -174,7 +174,7 @@ void arch_pick_mmap_layout(struct mm_struct *mm)
 
 static inline unsigned long brk_rnd(void)
 {
-	unsigned long rnd = get_random_int();
+	unsigned long rnd = get_random_long();
 
 	rnd = rnd << PAGE_SHIFT;
 	/* 8MB for 32bit, 256MB for 64bit */
diff --git a/arch/mn10300/include/asm/thread_info.h b/arch/mn10300/include/asm/thread_info.h
index 4861a78c7160..f5f90bbf019d 100644
--- a/arch/mn10300/include/asm/thread_info.h
+++ b/arch/mn10300/include/asm/thread_info.h
@@ -115,7 +115,7 @@ static inline unsigned long current_stack_pointer(void)
 }
 
 #ifndef CONFIG_KGDB
-void arch_release_thread_info(struct thread_info *ti);
+void arch_release_thread_stack(unsigned long *stack);
 #endif
 #define get_thread_info(ti)	get_task_struct((ti)->task)
 #define put_thread_info(ti)	put_task_struct((ti)->task)
diff --git a/arch/mn10300/kernel/kgdb.c b/arch/mn10300/kernel/kgdb.c
index 99770823451a..2d7986c386fe 100644
--- a/arch/mn10300/kernel/kgdb.c
+++ b/arch/mn10300/kernel/kgdb.c
@@ -397,8 +397,9 @@ static bool kgdb_arch_undo_singlestep(struct pt_regs *regs)
  * single-step state is cleared.  At this point the breakpoints should have
  * been removed by __switch_to().
  */
-void arch_release_thread_info(struct thread_info *ti)
+void arch_release_thread_stack(unsigned long *stack)
 {
+	struct thread_info *ti = (void *)stack;
 	if (kgdb_sstep_thread == ti) {
 		kgdb_sstep_thread = NULL;
 
diff --git a/arch/parisc/kernel/module.c b/arch/parisc/kernel/module.c
index 3c63a820fcda..b9d75d9fa9ac 100644
--- a/arch/parisc/kernel/module.c
+++ b/arch/parisc/kernel/module.c
@@ -42,9 +42,9 @@
  *      We are not doing SEGREL32 handling correctly. According to the ABI, we
  *      should do a value offset, like this:
  *			if (in_init(me, (void *)val))
- *				val -= (uint32_t)me->module_init;
+ *				val -= (uint32_t)me->init_layout.base;
  *			else
- *				val -= (uint32_t)me->module_core;
+ *				val -= (uint32_t)me->core_layout.base;
  *	However, SEGREL32 is used only for PARISC unwind entries, and we want
  *	those entries to have an absolute address, and not just an offset.
  *
@@ -100,14 +100,14 @@
  * or init pieces the location is */
 static inline int in_init(struct module *me, void *loc)
 {
-	return (loc >= me->module_init &&
-		loc <= (me->module_init + me->init_size));
+	return (loc >= me->init_layout.base &&
+		loc <= (me->init_layout.base + me->init_layout.size));
 }
 
 static inline int in_core(struct module *me, void *loc)
 {
-	return (loc >= me->module_core &&
-		loc <= (me->module_core + me->core_size));
+	return (loc >= me->core_layout.base &&
+		loc <= (me->core_layout.base + me->core_layout.size));
 }
 
 static inline int in_local(struct module *me, void *loc)
@@ -367,13 +367,13 @@ int module_frob_arch_sections(CONST Elf_Ehdr *hdr,
 	}
 
 	/* align things a bit */
-	me->core_size = ALIGN(me->core_size, 16);
-	me->arch.got_offset = me->core_size;
-	me->core_size += gots * sizeof(struct got_entry);
+	me->core_layout.size = ALIGN(me->core_layout.size, 16);
+	me->arch.got_offset = me->core_layout.size;
+	me->core_layout.size += gots * sizeof(struct got_entry);
 
-	me->core_size = ALIGN(me->core_size, 16);
-	me->arch.fdesc_offset = me->core_size;
-	me->core_size += fdescs * sizeof(Elf_Fdesc);
+	me->core_layout.size = ALIGN(me->core_layout.size, 16);
+	me->arch.fdesc_offset = me->core_layout.size;
+	me->core_layout.size += fdescs * sizeof(Elf_Fdesc);
 
 	me->arch.got_max = gots;
 	me->arch.fdesc_max = fdescs;
@@ -391,7 +391,7 @@ static Elf64_Word get_got(struct module *me, unsigned long value, long addend)
 
 	BUG_ON(value == 0);
 
-	got = me->module_core + me->arch.got_offset;
+	got = me->core_layout.base + me->arch.got_offset;
 	for (i = 0; got[i].addr; i++)
 		if (got[i].addr == value)
 			goto out;
@@ -409,7 +409,7 @@ static Elf64_Word get_got(struct module *me, unsigned long value, long addend)
 #ifdef CONFIG_64BIT
 static Elf_Addr get_fdesc(struct module *me, unsigned long value)
 {
-	Elf_Fdesc *fdesc = me->module_core + me->arch.fdesc_offset;
+	Elf_Fdesc *fdesc = me->core_layout.base + me->arch.fdesc_offset;
 
 	if (!value) {
 		printk(KERN_ERR "%s: zero OPD requested!\n", me->name);
@@ -427,7 +427,7 @@ static Elf_Addr get_fdesc(struct module *me, unsigned long value)
 
 	/* Create new one */
 	fdesc->addr = value;
-	fdesc->gp = (Elf_Addr)me->module_core + me->arch.got_offset;
+	fdesc->gp = (Elf_Addr)me->core_layout.base + me->arch.got_offset;
 	return (Elf_Addr)fdesc;
 }
 #endif /* CONFIG_64BIT */
@@ -839,7 +839,7 @@ register_unwind_table(struct module *me,
 
 	table = (unsigned char *)sechdrs[me->arch.unwind_section].sh_addr;
 	end = table + sechdrs[me->arch.unwind_section].sh_size;
-	gp = (Elf_Addr)me->module_core + me->arch.got_offset;
+	gp = (Elf_Addr)me->core_layout.base + me->arch.got_offset;
 
 	DEBUGP("register_unwind_table(), sect = %d at 0x%p - 0x%p (gp=0x%lx)\n",
 	       me->arch.unwind_section, table, end, gp);
diff --git a/arch/powerpc/kernel/module_32.c b/arch/powerpc/kernel/module_32.c
index c94d2e018d84..2c01665eb410 100644
--- a/arch/powerpc/kernel/module_32.c
+++ b/arch/powerpc/kernel/module_32.c
@@ -188,8 +188,8 @@ static uint32_t do_plt_call(void *location,
 
 	pr_debug("Doing plt for call to 0x%x at 0x%x\n", val, (unsigned int)location);
 	/* Init, or core PLT? */
-	if (location >= mod->module_core
-	    && location < mod->module_core + mod->core_size)
+	if (location >= mod->core_layout.base
+	    && location < mod->core_layout.base + mod->core_layout.size)
 		entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr;
 	else
 		entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr;
@@ -296,7 +296,7 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
 	}
 #ifdef CONFIG_DYNAMIC_FTRACE
 	module->arch.tramp =
-		do_plt_call(module->module_core,
+		do_plt_call(module->core_layout.base,
 			    (unsigned long)ftrace_caller,
 			    sechdrs, module);
 #endif
diff --git a/arch/powerpc/kernel/process.c b/arch/powerpc/kernel/process.c
index cf788d7d7e56..b7abf3cd2a67 100644
--- a/arch/powerpc/kernel/process.c
+++ b/arch/powerpc/kernel/process.c
@@ -1651,9 +1651,9 @@ static inline unsigned long brk_rnd(void)
 
 	/* 8MB for 32bit, 1GB for 64bit */
 	if (is_32bit_task())
-		rnd = (long)(get_random_int() % (1<<(23-PAGE_SHIFT)));
+		rnd = (get_random_long() % (1UL<<(23-PAGE_SHIFT)));
 	else
-		rnd = (long)(get_random_int() % (1<<(30-PAGE_SHIFT)));
+		rnd = (get_random_long() % (1UL<<(30-PAGE_SHIFT)));
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/arch/powerpc/mm/mmap.c b/arch/powerpc/mm/mmap.c
index 0f0502e12f6c..4087705ba90f 100644
--- a/arch/powerpc/mm/mmap.c
+++ b/arch/powerpc/mm/mmap.c
@@ -59,9 +59,9 @@ unsigned long arch_mmap_rnd(void)
 
 	/* 8MB for 32bit, 1GB for 64bit */
 	if (is_32bit_task())
-		rnd = (unsigned long)get_random_int() % (1<<(23-PAGE_SHIFT));
+		rnd = get_random_long() % (1<<(23-PAGE_SHIFT));
 	else
-		rnd = (unsigned long)get_random_int() % (1<<(30-PAGE_SHIFT));
+		rnd = get_random_long() % (1UL<<(30-PAGE_SHIFT));
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/arch/s390/kernel/module.c b/arch/s390/kernel/module.c
index 0c1a679314dd..7873e171457c 100644
--- a/arch/s390/kernel/module.c
+++ b/arch/s390/kernel/module.c
@@ -159,11 +159,11 @@ int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs,
 
 	/* Increase core size by size of got & plt and set start
 	   offsets for got and plt. */
-	me->core_size = ALIGN(me->core_size, 4);
-	me->arch.got_offset = me->core_size;
-	me->core_size += me->arch.got_size;
-	me->arch.plt_offset = me->core_size;
-	me->core_size += me->arch.plt_size;
+	me->core_layout.size = ALIGN(me->core_layout.size, 4);
+	me->arch.got_offset = me->core_layout.size;
+	me->core_layout.size += me->arch.got_size;
+	me->arch.plt_offset = me->core_layout.size;
+	me->core_layout.size += me->arch.plt_size;
 	return 0;
 }
 
@@ -279,7 +279,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
 		if (info->got_initialized == 0) {
 			Elf_Addr *gotent;
 
-			gotent = me->module_core + me->arch.got_offset +
+			gotent = me->core_layout.base + me->arch.got_offset +
 				info->got_offset;
 			*gotent = val;
 			info->got_initialized = 1;
@@ -302,7 +302,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
 			rc = apply_rela_bits(loc, val, 0, 64, 0);
 		else if (r_type == R_390_GOTENT ||
 			 r_type == R_390_GOTPLTENT) {
-			val += (Elf_Addr) me->module_core - loc;
+			val += (Elf_Addr) me->core_layout.base - loc;
 			rc = apply_rela_bits(loc, val, 1, 32, 1);
 		}
 		break;
@@ -315,7 +315,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
 	case R_390_PLTOFF64:	/* 16 bit offset from GOT to PLT. */
 		if (info->plt_initialized == 0) {
 			unsigned int *ip;
-			ip = me->module_core + me->arch.plt_offset +
+			ip = me->core_layout.base + me->arch.plt_offset +
 				info->plt_offset;
 			ip[0] = 0x0d10e310; /* basr 1,0; lg 1,10(1); br 1 */
 			ip[1] = 0x100a0004;
@@ -334,7 +334,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
 			       val - loc + 0xffffUL < 0x1ffffeUL) ||
 			      (r_type == R_390_PLT32DBL &&
 			       val - loc + 0xffffffffULL < 0x1fffffffeULL)))
-				val = (Elf_Addr) me->module_core +
+				val = (Elf_Addr) me->core_layout.base +
 					me->arch.plt_offset +
 					info->plt_offset;
 			val += rela->r_addend - loc;
@@ -356,7 +356,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
 	case R_390_GOTOFF32:	/* 32 bit offset to GOT.  */
 	case R_390_GOTOFF64:	/* 64 bit offset to GOT. */
 		val = val + rela->r_addend -
-			((Elf_Addr) me->module_core + me->arch.got_offset);
+			((Elf_Addr) me->core_layout.base + me->arch.got_offset);
 		if (r_type == R_390_GOTOFF16)
 			rc = apply_rela_bits(loc, val, 0, 16, 0);
 		else if (r_type == R_390_GOTOFF32)
@@ -366,7 +366,7 @@ static int apply_rela(Elf_Rela *rela, Elf_Addr base, Elf_Sym *symtab,
 		break;
 	case R_390_GOTPC:	/* 32 bit PC relative offset to GOT. */
 	case R_390_GOTPCDBL:	/* 32 bit PC rel. off. to GOT shifted by 1. */
-		val = (Elf_Addr) me->module_core + me->arch.got_offset +
+		val = (Elf_Addr) me->core_layout.base + me->arch.got_offset +
 			rela->r_addend - loc;
 		if (r_type == R_390_GOTPC)
 			rc = apply_rela_bits(loc, val, 1, 32, 0);
diff --git a/arch/sparc/kernel/sys_sparc_64.c b/arch/sparc/kernel/sys_sparc_64.c
index 7f0f7c01b297..98a5cf313d39 100644
--- a/arch/sparc/kernel/sys_sparc_64.c
+++ b/arch/sparc/kernel/sys_sparc_64.c
@@ -264,7 +264,7 @@ static unsigned long mmap_rnd(void)
 	unsigned long rnd = 0UL;
 
 	if (current->flags & PF_RANDOMIZE) {
-		unsigned long val = get_random_int();
+		unsigned long val = get_random_long();
 		if (test_thread_flag(TIF_32BIT))
 			rnd = (val % (1UL << (23UL-PAGE_SHIFT)));
 		else
diff --git a/arch/tile/include/asm/thread_info.h b/arch/tile/include/asm/thread_info.h
index dc1fb28d9636..489b15016303 100644
--- a/arch/tile/include/asm/thread_info.h
+++ b/arch/tile/include/asm/thread_info.h
@@ -78,7 +78,7 @@ struct thread_info {
 
 #ifndef __ASSEMBLY__
 
-void arch_release_thread_info(struct thread_info *info);
+void arch_release_thread_stack(unsigned long *stack);
 
 /* How to get the thread information struct from C. */
 register unsigned long stack_pointer __asm__("sp");
diff --git a/arch/tile/kernel/process.c b/arch/tile/kernel/process.c
index 7d5769310bef..a97ab1a69a90 100644
--- a/arch/tile/kernel/process.c
+++ b/arch/tile/kernel/process.c
@@ -73,8 +73,9 @@ void arch_cpu_idle(void)
 /*
  * Release a thread_info structure
  */
-void arch_release_thread_info(struct thread_info *info)
+void arch_release_thread_stack(unsigned long *stack)
 {
+	struct thread_info *info = (void *)stack;
 	struct single_step_state *step_state = info->step_state;
 
 	if (step_state) {
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index cb00b5f669c1..23b2767e423d 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -46,6 +46,8 @@ config X86
 	select ARCH_WANTS_DYNAMIC_TASK_STRUCT
 	select ARCH_WANT_FRAME_POINTERS
 	select ARCH_WANT_IPC_PARSE_VERSION	if X86_32
+	select HAVE_ARCH_MMAP_RND_BITS		if MMU
+	select HAVE_ARCH_MMAP_RND_COMPAT_BITS	if MMU && COMPAT
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select BUILDTIME_EXTABLE_SORT
 	select CLKEVT_I8253
@@ -184,6 +186,20 @@ config HAVE_LATENCYTOP_SUPPORT
 config MMU
 	def_bool y
 
+config ARCH_MMAP_RND_BITS_MIN
+	default 28 if 64BIT
+	default 8
+
+config ARCH_MMAP_RND_BITS_MAX
+	default 32 if 64BIT
+	default 16
+
+config ARCH_MMAP_RND_COMPAT_BITS_MIN
+	default 8
+
+config ARCH_MMAP_RND_COMPAT_BITS_MAX
+	default 16
+
 config SBUS
 	bool
 
@@ -279,11 +295,6 @@ config X86_32_LAZY_GS
 	def_bool y
 	depends on X86_32 && !CC_STACKPROTECTOR
 
-config ARCH_HWEIGHT_CFLAGS
-	string
-	default "-fcall-saved-ecx -fcall-saved-edx" if X86_32
-	default "-fcall-saved-rdi -fcall-saved-rsi -fcall-saved-rdx -fcall-saved-rcx -fcall-saved-r8 -fcall-saved-r9 -fcall-saved-r10 -fcall-saved-r11" if X86_64
-
 config ARCH_SUPPORTS_UPROBES
 	def_bool y
 
diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 4086abca0b32..0048d9886725 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -11,6 +11,16 @@ else
         KBUILD_DEFCONFIG := $(ARCH)_defconfig
 endif
 
+# For gcc stack alignment is specified with -mpreferred-stack-boundary,
+# clang has the option -mstack-alignment for that purpose.
+ifneq ($(call cc-option, -mpreferred-stack-boundary=4),)
+      cc_stack_align4 := -mpreferred-stack-boundary=2
+      cc_stack_align8 := -mpreferred-stack-boundary=3
+else ifneq ($(call cc-option, -mstack-alignment=16),)
+      cc_stack_align4 := -mstack-alignment=4
+      cc_stack_align8 := -mstack-alignment=8
+endif
+
 # How to compile the 16-bit code.  Note we always compile for -march=i386;
 # that way we can complain to the user if the CPU is insufficient.
 #
@@ -24,10 +34,11 @@ REALMODE_CFLAGS	:= $(M16_CFLAGS) -g -Os -D__KERNEL__ \
 		   -DDISABLE_BRANCH_PROFILING \
 		   -Wall -Wstrict-prototypes -march=i386 -mregparm=3 \
 		   -fno-strict-aliasing -fomit-frame-pointer -fno-pic \
-		   -mno-mmx -mno-sse \
-		   $(call cc-option, -ffreestanding) \
-		   $(call cc-option, -fno-stack-protector) \
-		   $(call cc-option, -mpreferred-stack-boundary=2)
+		   -mno-mmx -mno-sse
+
+REALMODE_CFLAGS += $(call __cc-option, $(CC), $(REALMODE_CFLAGS), -ffreestanding)
+REALMODE_CFLAGS += $(call __cc-option, $(CC), $(REALMODE_CFLAGS), -fno-stack-protector)
+REALMODE_CFLAGS += $(call __cc-option, $(CC), $(REALMODE_CFLAGS), $(cc_stack_align4))
 export REALMODE_CFLAGS
 
 # BITS is used as extension for files which are available in a 32 bit
@@ -64,8 +75,10 @@ ifeq ($(CONFIG_X86_32),y)
         # with nonstandard options
         KBUILD_CFLAGS += -fno-pic
 
-        # prevent gcc from keeping the stack 16 byte aligned
-        KBUILD_CFLAGS += $(call cc-option,-mpreferred-stack-boundary=2)
+        # Align the stack to the register width instead of using the default
+        # alignment of 16 bytes. This reduces stack usage and the number of
+        # alignment instructions.
+        KBUILD_CFLAGS += $(call cc-option,$(cc_stack_align4))
 
         # Disable unit-at-a-time mode on pre-gcc-4.0 compilers, it makes gcc use
         # a lot more stack due to the lack of sharing of stacklots:
@@ -88,17 +101,25 @@ else
         KBUILD_CFLAGS += -m64
 
         # Align jump targets to 1 byte, not the default 16 bytes:
-        KBUILD_CFLAGS += -falign-jumps=1
+        KBUILD_CFLAGS += $(call cc-option,-falign-jumps=1)
 
         # Pack loops tightly as well:
-        KBUILD_CFLAGS += -falign-loops=1
+        KBUILD_CFLAGS += $(call cc-option,-falign-loops=1)
 
         # Don't autogenerate traditional x87 instructions
         KBUILD_CFLAGS += $(call cc-option,-mno-80387)
         KBUILD_CFLAGS += $(call cc-option,-mno-fp-ret-in-387)
 
-	# Use -mpreferred-stack-boundary=3 if supported.
-	KBUILD_CFLAGS += $(call cc-option,-mpreferred-stack-boundary=3)
+        KBUILD_CFLAGS += -fno-pic
+
+        # By default gcc and clang use a stack alignment of 16 bytes for x86.
+        # However the standard kernel entry on x86-64 leaves the stack on an
+        # 8-byte boundary. If the compiler isn't informed about the actual
+        # alignment it will generate extra alignment instructions for the
+        # default alignment which keep the stack *mis*aligned.
+        # Furthermore an alignment to the register width reduces stack usage
+        # and the number of alignment instructions.
+        KBUILD_CFLAGS += $(call cc-option,$(cc_stack_align8))
 
 	# Use -mskip-rax-setup if supported.
 	KBUILD_CFLAGS += $(call cc-option,-mskip-rax-setup)
diff --git a/arch/x86/boot/compressed/aslr.c b/arch/x86/boot/compressed/aslr.c
index 6a9b96b4624d..4a5fbd2da658 100644
--- a/arch/x86/boot/compressed/aslr.c
+++ b/arch/x86/boot/compressed/aslr.c
@@ -94,7 +94,7 @@ static unsigned long get_random_long(void)
 	}
 
 	/* Circular multiply for better bit diffusion */
-	asm("mul %3"
+	asm(_ASM_MUL "%3"
 	    : "=a" (random), "=d" (raw)
 	    : "a" (random), "rm" (mix_const));
 	random += raw;
diff --git a/arch/x86/boot/string.c b/arch/x86/boot/string.c
index 06ceddb3a22e..1d56adea8a7c 100644
--- a/arch/x86/boot/string.c
+++ b/arch/x86/boot/string.c
@@ -16,6 +16,15 @@
 #include "ctype.h"
 #include "string.h"
 
+/*
+ * Undef these macros so that the functions that we provide
+ * here will have the correct names regardless of how string.h
+ * may have chosen to #define them.
+ */
+#undef memcpy
+#undef memset
+#undef memcmp
+
 int memcmp(const void *s1, const void *s2, size_t len)
 {
 	u8 diff;
diff --git a/arch/x86/configs/i386_ranchu_defconfig b/arch/x86/configs/i386_ranchu_defconfig
new file mode 100644
index 000000000000..65ed8c8f8444
--- /dev/null
+++ b/arch/x86/configs/i386_ranchu_defconfig
@@ -0,0 +1,423 @@
+# CONFIG_64BIT is not set
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_POSIX_MQUEUE=y
+CONFIG_AUDIT=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_BSD_PROCESS_ACCT=y
+CONFIG_TASKSTATS=y
+CONFIG_TASK_DELAY_ACCT=y
+CONFIG_TASK_XACCT=y
+CONFIG_TASK_IO_ACCOUNTING=y
+CONFIG_CGROUPS=y
+CONFIG_CGROUP_DEBUG=y
+CONFIG_CGROUP_FREEZER=y
+CONFIG_CGROUP_CPUACCT=y
+CONFIG_CGROUP_SCHED=y
+CONFIG_RT_GROUP_SCHED=y
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SYSCTL_SYSCALL=y
+CONFIG_KALLSYMS_ALL=y
+CONFIG_EMBEDDED=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_ARCH_MMAP_RND_BITS=16
+CONFIG_PARTITION_ADVANCED=y
+CONFIG_OSF_PARTITION=y
+CONFIG_AMIGA_PARTITION=y
+CONFIG_MAC_PARTITION=y
+CONFIG_BSD_DISKLABEL=y
+CONFIG_MINIX_SUBPARTITION=y
+CONFIG_SOLARIS_X86_PARTITION=y
+CONFIG_UNIXWARE_DISKLABEL=y
+CONFIG_SGI_PARTITION=y
+CONFIG_SUN_PARTITION=y
+CONFIG_KARMA_PARTITION=y
+CONFIG_SMP=y
+CONFIG_X86_BIGSMP=y
+CONFIG_MCORE2=y
+CONFIG_X86_GENERIC=y
+CONFIG_HPET_TIMER=y
+CONFIG_NR_CPUS=512
+CONFIG_PREEMPT=y
+# CONFIG_X86_MCE is not set
+CONFIG_X86_REBOOTFIXUPS=y
+CONFIG_X86_MSR=y
+CONFIG_X86_CPUID=y
+CONFIG_KSM=y
+CONFIG_CMA=y
+# CONFIG_MTRR_SANITIZER is not set
+CONFIG_EFI=y
+CONFIG_EFI_STUB=y
+CONFIG_HZ_100=y
+CONFIG_PHYSICAL_START=0x100000
+CONFIG_PM_AUTOSLEEP=y
+CONFIG_PM_WAKELOCKS=y
+CONFIG_PM_WAKELOCKS_LIMIT=0
+# CONFIG_PM_WAKELOCKS_GC is not set
+CONFIG_PM_DEBUG=y
+CONFIG_CPU_FREQ=y
+# CONFIG_CPU_FREQ_STAT is not set
+CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
+CONFIG_CPU_FREQ_GOV_USERSPACE=y
+CONFIG_PCIEPORTBUS=y
+# CONFIG_PCIEASPM is not set
+CONFIG_PCCARD=y
+CONFIG_YENTA=y
+CONFIG_HOTPLUG_PCI=y
+# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
+CONFIG_BINFMT_MISC=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_XFRM_USER=y
+CONFIG_NET_KEY=y
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IP_ROUTE_MULTIPATH=y
+CONFIG_IP_ROUTE_VERBOSE=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+CONFIG_IP_MROUTE=y
+CONFIG_IP_PIMSM_V1=y
+CONFIG_IP_PIMSM_V2=y
+CONFIG_SYN_COOKIES=y
+CONFIG_INET_ESP=y
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+CONFIG_INET_DIAG_DESTROY=y
+CONFIG_IPV6_ROUTER_PREF=y
+CONFIG_IPV6_ROUTE_INFO=y
+CONFIG_IPV6_OPTIMISTIC_DAD=y
+CONFIG_INET6_AH=y
+CONFIG_INET6_ESP=y
+CONFIG_INET6_IPCOMP=y
+CONFIG_IPV6_MIP6=y
+CONFIG_IPV6_MULTIPLE_TABLES=y
+CONFIG_NETLABEL=y
+CONFIG_NETFILTER=y
+CONFIG_NF_CONNTRACK=y
+CONFIG_NF_CONNTRACK_SECMARK=y
+CONFIG_NF_CONNTRACK_EVENTS=y
+CONFIG_NF_CT_PROTO_DCCP=y
+CONFIG_NF_CT_PROTO_SCTP=y
+CONFIG_NF_CT_PROTO_UDPLITE=y
+CONFIG_NF_CONNTRACK_AMANDA=y
+CONFIG_NF_CONNTRACK_FTP=y
+CONFIG_NF_CONNTRACK_H323=y
+CONFIG_NF_CONNTRACK_IRC=y
+CONFIG_NF_CONNTRACK_NETBIOS_NS=y
+CONFIG_NF_CONNTRACK_PPTP=y
+CONFIG_NF_CONNTRACK_SANE=y
+CONFIG_NF_CONNTRACK_TFTP=y
+CONFIG_NF_CT_NETLINK=y
+CONFIG_NETFILTER_XT_TARGET_CLASSIFY=y
+CONFIG_NETFILTER_XT_TARGET_CONNMARK=y
+CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=y
+CONFIG_NETFILTER_XT_TARGET_IDLETIMER=y
+CONFIG_NETFILTER_XT_TARGET_MARK=y
+CONFIG_NETFILTER_XT_TARGET_NFLOG=y
+CONFIG_NETFILTER_XT_TARGET_NFQUEUE=y
+CONFIG_NETFILTER_XT_TARGET_TPROXY=y
+CONFIG_NETFILTER_XT_TARGET_TRACE=y
+CONFIG_NETFILTER_XT_TARGET_SECMARK=y
+CONFIG_NETFILTER_XT_TARGET_TCPMSS=y
+CONFIG_NETFILTER_XT_MATCH_COMMENT=y
+CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_CONNMARK=y
+CONFIG_NETFILTER_XT_MATCH_CONNTRACK=y
+CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_HELPER=y
+CONFIG_NETFILTER_XT_MATCH_IPRANGE=y
+CONFIG_NETFILTER_XT_MATCH_LENGTH=y
+CONFIG_NETFILTER_XT_MATCH_LIMIT=y
+CONFIG_NETFILTER_XT_MATCH_MAC=y
+CONFIG_NETFILTER_XT_MATCH_MARK=y
+CONFIG_NETFILTER_XT_MATCH_POLICY=y
+CONFIG_NETFILTER_XT_MATCH_PKTTYPE=y
+CONFIG_NETFILTER_XT_MATCH_QTAGUID=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA2=y
+CONFIG_NETFILTER_XT_MATCH_SOCKET=y
+CONFIG_NETFILTER_XT_MATCH_STATE=y
+CONFIG_NETFILTER_XT_MATCH_STATISTIC=y
+CONFIG_NETFILTER_XT_MATCH_STRING=y
+CONFIG_NETFILTER_XT_MATCH_TIME=y
+CONFIG_NETFILTER_XT_MATCH_U32=y
+CONFIG_NF_CONNTRACK_IPV4=y
+CONFIG_IP_NF_IPTABLES=y
+CONFIG_IP_NF_MATCH_AH=y
+CONFIG_IP_NF_MATCH_ECN=y
+CONFIG_IP_NF_MATCH_TTL=y
+CONFIG_IP_NF_FILTER=y
+CONFIG_IP_NF_TARGET_REJECT=y
+CONFIG_IP_NF_MANGLE=y
+CONFIG_IP_NF_RAW=y
+CONFIG_IP_NF_SECURITY=y
+CONFIG_IP_NF_ARPTABLES=y
+CONFIG_IP_NF_ARPFILTER=y
+CONFIG_IP_NF_ARP_MANGLE=y
+CONFIG_NF_CONNTRACK_IPV6=y
+CONFIG_IP6_NF_IPTABLES=y
+CONFIG_IP6_NF_FILTER=y
+CONFIG_IP6_NF_TARGET_REJECT=y
+CONFIG_IP6_NF_MANGLE=y
+CONFIG_IP6_NF_RAW=y
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_HTB=y
+CONFIG_NET_CLS_U32=y
+CONFIG_NET_EMATCH=y
+CONFIG_NET_EMATCH_U32=y
+CONFIG_NET_CLS_ACT=y
+CONFIG_CFG80211=y
+CONFIG_MAC80211=y
+CONFIG_MAC80211_LEDS=y
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_DMA_CMA=y
+CONFIG_CMA_SIZE_MBYTES=16
+CONFIG_CONNECTOR=y
+CONFIG_BLK_DEV_LOOP=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_VIRTIO_BLK=y
+CONFIG_BLK_DEV_SD=y
+CONFIG_BLK_DEV_SR=y
+CONFIG_BLK_DEV_SR_VENDOR=y
+CONFIG_CHR_DEV_SG=y
+CONFIG_SCSI_CONSTANTS=y
+CONFIG_SCSI_SPI_ATTRS=y
+CONFIG_SCSI_ISCSI_ATTRS=y
+# CONFIG_SCSI_LOWLEVEL is not set
+CONFIG_ATA=y
+CONFIG_SATA_AHCI=y
+CONFIG_ATA_PIIX=y
+CONFIG_PATA_AMD=y
+CONFIG_PATA_OLDPIIX=y
+CONFIG_PATA_SCH=y
+CONFIG_PATA_MPIIX=y
+CONFIG_ATA_GENERIC=y
+CONFIG_MD=y
+CONFIG_BLK_DEV_MD=y
+CONFIG_BLK_DEV_DM=y
+CONFIG_DM_DEBUG=y
+CONFIG_DM_CRYPT=y
+CONFIG_DM_MIRROR=y
+CONFIG_DM_ZERO=y
+CONFIG_DM_UEVENT=y
+CONFIG_DM_VERITY=y
+CONFIG_DM_VERITY_FEC=y
+CONFIG_NETDEVICES=y
+CONFIG_NETCONSOLE=y
+CONFIG_TUN=y
+CONFIG_VIRTIO_NET=y
+CONFIG_BNX2=y
+CONFIG_TIGON3=y
+CONFIG_NET_TULIP=y
+CONFIG_E100=y
+CONFIG_E1000=y
+CONFIG_E1000E=y
+CONFIG_SKY2=y
+CONFIG_NE2K_PCI=y
+CONFIG_FORCEDETH=y
+CONFIG_8139TOO=y
+# CONFIG_8139TOO_PIO is not set
+CONFIG_R8169=y
+CONFIG_FDDI=y
+CONFIG_PPP=y
+CONFIG_PPP_BSDCOMP=y
+CONFIG_PPP_DEFLATE=y
+CONFIG_PPP_MPPE=y
+CONFIG_PPPOLAC=y
+CONFIG_PPPOPNS=y
+CONFIG_USB_USBNET=y
+CONFIG_INPUT_POLLDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_EVDEV=y
+CONFIG_INPUT_KEYRESET=y
+# CONFIG_KEYBOARD_ATKBD is not set
+CONFIG_KEYBOARD_GOLDFISH_EVENTS=y
+# CONFIG_INPUT_MOUSE is not set
+CONFIG_INPUT_JOYSTICK=y
+CONFIG_JOYSTICK_XPAD=y
+CONFIG_JOYSTICK_XPAD_FF=y
+CONFIG_JOYSTICK_XPAD_LEDS=y
+CONFIG_INPUT_TABLET=y
+CONFIG_TABLET_USB_ACECAD=y
+CONFIG_TABLET_USB_AIPTEK=y
+CONFIG_TABLET_USB_GTCO=y
+CONFIG_TABLET_USB_HANWANG=y
+CONFIG_TABLET_USB_KBTAB=y
+CONFIG_INPUT_TOUCHSCREEN=y
+CONFIG_INPUT_MISC=y
+CONFIG_INPUT_KEYCHORD=y
+CONFIG_INPUT_UINPUT=y
+CONFIG_INPUT_GPIO=y
+# CONFIG_SERIO is not set
+# CONFIG_VT is not set
+# CONFIG_LEGACY_PTYS is not set
+CONFIG_SERIAL_NONSTANDARD=y
+# CONFIG_DEVMEM is not set
+# CONFIG_DEVKMEM is not set
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_VIRTIO_CONSOLE=y
+CONFIG_NVRAM=y
+CONFIG_I2C_I801=y
+CONFIG_BATTERY_GOLDFISH=y
+CONFIG_WATCHDOG=y
+CONFIG_MEDIA_SUPPORT=y
+CONFIG_AGP=y
+CONFIG_AGP_AMD64=y
+CONFIG_AGP_INTEL=y
+CONFIG_DRM=y
+CONFIG_FB_MODE_HELPERS=y
+CONFIG_FB_TILEBLITTING=y
+CONFIG_FB_EFI=y
+CONFIG_FB_GOLDFISH=y
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+# CONFIG_LCD_CLASS_DEVICE is not set
+CONFIG_SOUND=y
+CONFIG_SND=y
+CONFIG_HIDRAW=y
+CONFIG_UHID=y
+CONFIG_HID_A4TECH=y
+CONFIG_HID_ACRUX=y
+CONFIG_HID_ACRUX_FF=y
+CONFIG_HID_APPLE=y
+CONFIG_HID_BELKIN=y
+CONFIG_HID_CHERRY=y
+CONFIG_HID_CHICONY=y
+CONFIG_HID_PRODIKEYS=y
+CONFIG_HID_CYPRESS=y
+CONFIG_HID_DRAGONRISE=y
+CONFIG_DRAGONRISE_FF=y
+CONFIG_HID_EMS_FF=y
+CONFIG_HID_ELECOM=y
+CONFIG_HID_EZKEY=y
+CONFIG_HID_HOLTEK=y
+CONFIG_HID_KEYTOUCH=y
+CONFIG_HID_KYE=y
+CONFIG_HID_UCLOGIC=y
+CONFIG_HID_WALTOP=y
+CONFIG_HID_GYRATION=y
+CONFIG_HID_TWINHAN=y
+CONFIG_HID_KENSINGTON=y
+CONFIG_HID_LCPOWER=y
+CONFIG_HID_LOGITECH=y
+CONFIG_HID_LOGITECH_DJ=y
+CONFIG_LOGITECH_FF=y
+CONFIG_LOGIRUMBLEPAD2_FF=y
+CONFIG_LOGIG940_FF=y
+CONFIG_HID_MAGICMOUSE=y
+CONFIG_HID_MICROSOFT=y
+CONFIG_HID_MONTEREY=y
+CONFIG_HID_MULTITOUCH=y
+CONFIG_HID_NTRIG=y
+CONFIG_HID_ORTEK=y
+CONFIG_HID_PANTHERLORD=y
+CONFIG_PANTHERLORD_FF=y
+CONFIG_HID_PETALYNX=y
+CONFIG_HID_PICOLCD=y
+CONFIG_HID_PRIMAX=y
+CONFIG_HID_ROCCAT=y
+CONFIG_HID_SAITEK=y
+CONFIG_HID_SAMSUNG=y
+CONFIG_HID_SONY=y
+CONFIG_HID_SPEEDLINK=y
+CONFIG_HID_SUNPLUS=y
+CONFIG_HID_GREENASIA=y
+CONFIG_GREENASIA_FF=y
+CONFIG_HID_SMARTJOYPLUS=y
+CONFIG_SMARTJOYPLUS_FF=y
+CONFIG_HID_TIVO=y
+CONFIG_HID_TOPSEED=y
+CONFIG_HID_THRUSTMASTER=y
+CONFIG_HID_WACOM=y
+CONFIG_HID_WIIMOTE=y
+CONFIG_HID_ZEROPLUS=y
+CONFIG_HID_ZYDACRON=y
+CONFIG_HID_PID=y
+CONFIG_USB_HIDDEV=y
+CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
+CONFIG_USB_MON=y
+CONFIG_USB_EHCI_HCD=y
+# CONFIG_USB_EHCI_TT_NEWSCHED is not set
+CONFIG_USB_OHCI_HCD=y
+CONFIG_USB_UHCI_HCD=y
+CONFIG_USB_PRINTER=y
+CONFIG_USB_STORAGE=y
+CONFIG_USB_OTG_WAKELOCK=y
+CONFIG_EDAC=y
+CONFIG_RTC_CLASS=y
+# CONFIG_RTC_HCTOSYS is not set
+CONFIG_DMADEVICES=y
+CONFIG_VIRTIO_PCI=y
+CONFIG_STAGING=y
+CONFIG_ASHMEM=y
+CONFIG_ANDROID_LOW_MEMORY_KILLER=y
+CONFIG_SYNC=y
+CONFIG_SW_SYNC=y
+CONFIG_ION=y
+CONFIG_GOLDFISH_AUDIO=y
+CONFIG_GOLDFISH_SYNC=y
+CONFIG_SND_HDA_INTEL=y
+CONFIG_GOLDFISH=y
+CONFIG_GOLDFISH_PIPE=y
+CONFIG_ANDROID=y
+CONFIG_ANDROID_BINDER_IPC=y
+CONFIG_ISCSI_IBFT_FIND=y
+CONFIG_EXT4_FS=y
+CONFIG_EXT4_FS_SECURITY=y
+CONFIG_QUOTA=y
+CONFIG_QUOTA_NETLINK_INTERFACE=y
+# CONFIG_PRINT_QUOTA_WARNING is not set
+CONFIG_FUSE_FS=y
+CONFIG_ISO9660_FS=y
+CONFIG_JOLIET=y
+CONFIG_ZISOFS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_TMPFS=y
+CONFIG_TMPFS_POSIX_ACL=y
+CONFIG_HUGETLBFS=y
+CONFIG_PSTORE=y
+CONFIG_PSTORE_CONSOLE=y
+CONFIG_PSTORE_RAM=y
+# CONFIG_NETWORK_FILESYSTEMS is not set
+CONFIG_NLS_DEFAULT="utf8"
+CONFIG_NLS_CODEPAGE_437=y
+CONFIG_NLS_ASCII=y
+CONFIG_NLS_ISO8859_1=y
+CONFIG_NLS_UTF8=y
+CONFIG_PRINTK_TIME=y
+CONFIG_DEBUG_INFO=y
+# CONFIG_ENABLE_WARN_DEPRECATED is not set
+# CONFIG_ENABLE_MUST_CHECK is not set
+CONFIG_FRAME_WARN=2048
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_MEMORY_INIT=y
+CONFIG_PANIC_TIMEOUT=5
+CONFIG_SCHEDSTATS=y
+CONFIG_TIMER_STATS=y
+CONFIG_SCHED_TRACER=y
+CONFIG_BLK_DEV_IO_TRACE=y
+CONFIG_PROVIDE_OHCI1394_DMA_INIT=y
+CONFIG_KEYS=y
+CONFIG_SECURITY=y
+CONFIG_SECURITY_NETWORK=y
+CONFIG_SECURITY_SELINUX=y
+CONFIG_CRYPTO_AES_586=y
+CONFIG_CRYPTO_TWOFISH=y
+CONFIG_ASYMMETRIC_KEY_TYPE=y
+CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=y
+CONFIG_X509_CERTIFICATE_PARSER=y
+CONFIG_PKCS7_MESSAGE_PARSER=y
+CONFIG_PKCS7_TEST_KEY=y
+# CONFIG_VIRTUALIZATION is not set
+CONFIG_CRC_T10DIF=y
diff --git a/arch/x86/configs/x86_64_ranchu_defconfig b/arch/x86/configs/x86_64_ranchu_defconfig
new file mode 100644
index 000000000000..d977bd91e390
--- /dev/null
+++ b/arch/x86/configs/x86_64_ranchu_defconfig
@@ -0,0 +1,418 @@
+# CONFIG_LOCALVERSION_AUTO is not set
+CONFIG_POSIX_MQUEUE=y
+CONFIG_AUDIT=y
+CONFIG_NO_HZ=y
+CONFIG_HIGH_RES_TIMERS=y
+CONFIG_BSD_PROCESS_ACCT=y
+CONFIG_TASKSTATS=y
+CONFIG_TASK_DELAY_ACCT=y
+CONFIG_TASK_XACCT=y
+CONFIG_TASK_IO_ACCOUNTING=y
+CONFIG_CGROUPS=y
+CONFIG_CGROUP_DEBUG=y
+CONFIG_CGROUP_FREEZER=y
+CONFIG_CGROUP_CPUACCT=y
+CONFIG_CGROUP_SCHED=y
+CONFIG_RT_GROUP_SCHED=y
+CONFIG_BLK_DEV_INITRD=y
+CONFIG_CC_OPTIMIZE_FOR_SIZE=y
+CONFIG_SYSCTL_SYSCALL=y
+CONFIG_KALLSYMS_ALL=y
+CONFIG_EMBEDDED=y
+# CONFIG_COMPAT_BRK is not set
+CONFIG_ARCH_MMAP_RND_BITS=32
+CONFIG_ARCH_MMAP_RND_COMPAT_BITS=16
+CONFIG_PARTITION_ADVANCED=y
+CONFIG_OSF_PARTITION=y
+CONFIG_AMIGA_PARTITION=y
+CONFIG_MAC_PARTITION=y
+CONFIG_BSD_DISKLABEL=y
+CONFIG_MINIX_SUBPARTITION=y
+CONFIG_SOLARIS_X86_PARTITION=y
+CONFIG_UNIXWARE_DISKLABEL=y
+CONFIG_SGI_PARTITION=y
+CONFIG_SUN_PARTITION=y
+CONFIG_KARMA_PARTITION=y
+CONFIG_SMP=y
+CONFIG_MCORE2=y
+CONFIG_MAXSMP=y
+CONFIG_PREEMPT=y
+# CONFIG_X86_MCE is not set
+CONFIG_X86_MSR=y
+CONFIG_X86_CPUID=y
+CONFIG_KSM=y
+CONFIG_CMA=y
+# CONFIG_MTRR_SANITIZER is not set
+CONFIG_EFI=y
+CONFIG_EFI_STUB=y
+CONFIG_HZ_100=y
+CONFIG_PHYSICAL_START=0x100000
+CONFIG_PM_AUTOSLEEP=y
+CONFIG_PM_WAKELOCKS=y
+CONFIG_PM_WAKELOCKS_LIMIT=0
+# CONFIG_PM_WAKELOCKS_GC is not set
+CONFIG_PM_DEBUG=y
+CONFIG_CPU_FREQ=y
+# CONFIG_CPU_FREQ_STAT is not set
+CONFIG_CPU_FREQ_DEFAULT_GOV_ONDEMAND=y
+CONFIG_CPU_FREQ_GOV_USERSPACE=y
+CONFIG_PCI_MMCONFIG=y
+CONFIG_PCIEPORTBUS=y
+# CONFIG_PCIEASPM is not set
+CONFIG_PCCARD=y
+CONFIG_YENTA=y
+CONFIG_HOTPLUG_PCI=y
+# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set
+CONFIG_BINFMT_MISC=y
+CONFIG_IA32_EMULATION=y
+CONFIG_NET=y
+CONFIG_PACKET=y
+CONFIG_UNIX=y
+CONFIG_XFRM_USER=y
+CONFIG_NET_KEY=y
+CONFIG_INET=y
+CONFIG_IP_MULTICAST=y
+CONFIG_IP_ADVANCED_ROUTER=y
+CONFIG_IP_MULTIPLE_TABLES=y
+CONFIG_IP_ROUTE_MULTIPATH=y
+CONFIG_IP_ROUTE_VERBOSE=y
+CONFIG_IP_PNP=y
+CONFIG_IP_PNP_DHCP=y
+CONFIG_IP_PNP_BOOTP=y
+CONFIG_IP_PNP_RARP=y
+CONFIG_IP_MROUTE=y
+CONFIG_IP_PIMSM_V1=y
+CONFIG_IP_PIMSM_V2=y
+CONFIG_SYN_COOKIES=y
+CONFIG_INET_ESP=y
+# CONFIG_INET_XFRM_MODE_BEET is not set
+# CONFIG_INET_LRO is not set
+CONFIG_INET_DIAG_DESTROY=y
+CONFIG_IPV6_ROUTER_PREF=y
+CONFIG_IPV6_ROUTE_INFO=y
+CONFIG_IPV6_OPTIMISTIC_DAD=y
+CONFIG_INET6_AH=y
+CONFIG_INET6_ESP=y
+CONFIG_INET6_IPCOMP=y
+CONFIG_IPV6_MIP6=y
+CONFIG_IPV6_MULTIPLE_TABLES=y
+CONFIG_NETLABEL=y
+CONFIG_NETFILTER=y
+CONFIG_NF_CONNTRACK=y
+CONFIG_NF_CONNTRACK_SECMARK=y
+CONFIG_NF_CONNTRACK_EVENTS=y
+CONFIG_NF_CT_PROTO_DCCP=y
+CONFIG_NF_CT_PROTO_SCTP=y
+CONFIG_NF_CT_PROTO_UDPLITE=y
+CONFIG_NF_CONNTRACK_AMANDA=y
+CONFIG_NF_CONNTRACK_FTP=y
+CONFIG_NF_CONNTRACK_H323=y
+CONFIG_NF_CONNTRACK_IRC=y
+CONFIG_NF_CONNTRACK_NETBIOS_NS=y
+CONFIG_NF_CONNTRACK_PPTP=y
+CONFIG_NF_CONNTRACK_SANE=y
+CONFIG_NF_CONNTRACK_TFTP=y
+CONFIG_NF_CT_NETLINK=y
+CONFIG_NETFILTER_XT_TARGET_CLASSIFY=y
+CONFIG_NETFILTER_XT_TARGET_CONNMARK=y
+CONFIG_NETFILTER_XT_TARGET_CONNSECMARK=y
+CONFIG_NETFILTER_XT_TARGET_IDLETIMER=y
+CONFIG_NETFILTER_XT_TARGET_MARK=y
+CONFIG_NETFILTER_XT_TARGET_NFLOG=y
+CONFIG_NETFILTER_XT_TARGET_NFQUEUE=y
+CONFIG_NETFILTER_XT_TARGET_TPROXY=y
+CONFIG_NETFILTER_XT_TARGET_TRACE=y
+CONFIG_NETFILTER_XT_TARGET_SECMARK=y
+CONFIG_NETFILTER_XT_TARGET_TCPMSS=y
+CONFIG_NETFILTER_XT_MATCH_COMMENT=y
+CONFIG_NETFILTER_XT_MATCH_CONNLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_CONNMARK=y
+CONFIG_NETFILTER_XT_MATCH_CONNTRACK=y
+CONFIG_NETFILTER_XT_MATCH_HASHLIMIT=y
+CONFIG_NETFILTER_XT_MATCH_HELPER=y
+CONFIG_NETFILTER_XT_MATCH_IPRANGE=y
+CONFIG_NETFILTER_XT_MATCH_LENGTH=y
+CONFIG_NETFILTER_XT_MATCH_LIMIT=y
+CONFIG_NETFILTER_XT_MATCH_MAC=y
+CONFIG_NETFILTER_XT_MATCH_MARK=y
+CONFIG_NETFILTER_XT_MATCH_POLICY=y
+CONFIG_NETFILTER_XT_MATCH_PKTTYPE=y
+CONFIG_NETFILTER_XT_MATCH_QTAGUID=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA=y
+CONFIG_NETFILTER_XT_MATCH_QUOTA2=y
+CONFIG_NETFILTER_XT_MATCH_SOCKET=y
+CONFIG_NETFILTER_XT_MATCH_STATE=y
+CONFIG_NETFILTER_XT_MATCH_STATISTIC=y
+CONFIG_NETFILTER_XT_MATCH_STRING=y
+CONFIG_NETFILTER_XT_MATCH_TIME=y
+CONFIG_NETFILTER_XT_MATCH_U32=y
+CONFIG_NF_CONNTRACK_IPV4=y
+CONFIG_IP_NF_IPTABLES=y
+CONFIG_IP_NF_MATCH_AH=y
+CONFIG_IP_NF_MATCH_ECN=y
+CONFIG_IP_NF_MATCH_TTL=y
+CONFIG_IP_NF_FILTER=y
+CONFIG_IP_NF_TARGET_REJECT=y
+CONFIG_IP_NF_MANGLE=y
+CONFIG_IP_NF_RAW=y
+CONFIG_IP_NF_SECURITY=y
+CONFIG_IP_NF_ARPTABLES=y
+CONFIG_IP_NF_ARPFILTER=y
+CONFIG_IP_NF_ARP_MANGLE=y
+CONFIG_NF_CONNTRACK_IPV6=y
+CONFIG_IP6_NF_IPTABLES=y
+CONFIG_IP6_NF_FILTER=y
+CONFIG_IP6_NF_TARGET_REJECT=y
+CONFIG_IP6_NF_MANGLE=y
+CONFIG_IP6_NF_RAW=y
+CONFIG_NET_SCHED=y
+CONFIG_NET_SCH_HTB=y
+CONFIG_NET_CLS_U32=y
+CONFIG_NET_EMATCH=y
+CONFIG_NET_EMATCH_U32=y
+CONFIG_NET_CLS_ACT=y
+CONFIG_CFG80211=y
+CONFIG_MAC80211=y
+CONFIG_MAC80211_LEDS=y
+CONFIG_UEVENT_HELPER_PATH="/sbin/hotplug"
+CONFIG_DMA_CMA=y
+CONFIG_CONNECTOR=y
+CONFIG_BLK_DEV_LOOP=y
+CONFIG_BLK_DEV_RAM=y
+CONFIG_BLK_DEV_RAM_SIZE=8192
+CONFIG_VIRTIO_BLK=y
+CONFIG_BLK_DEV_SD=y
+CONFIG_BLK_DEV_SR=y
+CONFIG_BLK_DEV_SR_VENDOR=y
+CONFIG_CHR_DEV_SG=y
+CONFIG_SCSI_CONSTANTS=y
+CONFIG_SCSI_SPI_ATTRS=y
+CONFIG_SCSI_ISCSI_ATTRS=y
+# CONFIG_SCSI_LOWLEVEL is not set
+CONFIG_ATA=y
+CONFIG_SATA_AHCI=y
+CONFIG_ATA_PIIX=y
+CONFIG_PATA_AMD=y
+CONFIG_PATA_OLDPIIX=y
+CONFIG_PATA_SCH=y
+CONFIG_PATA_MPIIX=y
+CONFIG_ATA_GENERIC=y
+CONFIG_MD=y
+CONFIG_BLK_DEV_MD=y
+CONFIG_BLK_DEV_DM=y
+CONFIG_DM_DEBUG=y
+CONFIG_DM_CRYPT=y
+CONFIG_DM_MIRROR=y
+CONFIG_DM_ZERO=y
+CONFIG_DM_UEVENT=y
+CONFIG_DM_VERITY=y
+CONFIG_DM_VERITY_FEC=y
+CONFIG_NETDEVICES=y
+CONFIG_NETCONSOLE=y
+CONFIG_TUN=y
+CONFIG_VIRTIO_NET=y
+CONFIG_BNX2=y
+CONFIG_TIGON3=y
+CONFIG_NET_TULIP=y
+CONFIG_E100=y
+CONFIG_E1000=y
+CONFIG_E1000E=y
+CONFIG_SKY2=y
+CONFIG_NE2K_PCI=y
+CONFIG_FORCEDETH=y
+CONFIG_8139TOO=y
+# CONFIG_8139TOO_PIO is not set
+CONFIG_R8169=y
+CONFIG_FDDI=y
+CONFIG_PPP=y
+CONFIG_PPP_BSDCOMP=y
+CONFIG_PPP_DEFLATE=y
+CONFIG_PPP_MPPE=y
+CONFIG_PPPOLAC=y
+CONFIG_PPPOPNS=y
+CONFIG_USB_USBNET=y
+CONFIG_INPUT_POLLDEV=y
+# CONFIG_INPUT_MOUSEDEV_PSAUX is not set
+CONFIG_INPUT_EVDEV=y
+CONFIG_INPUT_KEYRESET=y
+# CONFIG_KEYBOARD_ATKBD is not set
+CONFIG_KEYBOARD_GOLDFISH_EVENTS=y
+# CONFIG_INPUT_MOUSE is not set
+CONFIG_INPUT_JOYSTICK=y
+CONFIG_JOYSTICK_XPAD=y
+CONFIG_JOYSTICK_XPAD_FF=y
+CONFIG_JOYSTICK_XPAD_LEDS=y
+CONFIG_INPUT_TABLET=y
+CONFIG_TABLET_USB_ACECAD=y
+CONFIG_TABLET_USB_AIPTEK=y
+CONFIG_TABLET_USB_GTCO=y
+CONFIG_TABLET_USB_HANWANG=y
+CONFIG_TABLET_USB_KBTAB=y
+CONFIG_INPUT_TOUCHSCREEN=y
+CONFIG_INPUT_MISC=y
+CONFIG_INPUT_KEYCHORD=y
+CONFIG_INPUT_UINPUT=y
+CONFIG_INPUT_GPIO=y
+# CONFIG_SERIO is not set
+# CONFIG_VT is not set
+# CONFIG_LEGACY_PTYS is not set
+CONFIG_SERIAL_NONSTANDARD=y
+# CONFIG_DEVMEM is not set
+# CONFIG_DEVKMEM is not set
+CONFIG_SERIAL_8250=y
+CONFIG_SERIAL_8250_CONSOLE=y
+CONFIG_VIRTIO_CONSOLE=y
+CONFIG_NVRAM=y
+CONFIG_I2C_I801=y
+CONFIG_BATTERY_GOLDFISH=y
+CONFIG_WATCHDOG=y
+CONFIG_MEDIA_SUPPORT=y
+CONFIG_AGP=y
+CONFIG_AGP_AMD64=y
+CONFIG_AGP_INTEL=y
+CONFIG_DRM=y
+CONFIG_FB_MODE_HELPERS=y
+CONFIG_FB_TILEBLITTING=y
+CONFIG_FB_EFI=y
+CONFIG_FB_GOLDFISH=y
+CONFIG_BACKLIGHT_LCD_SUPPORT=y
+# CONFIG_LCD_CLASS_DEVICE is not set
+CONFIG_SOUND=y
+CONFIG_SND=y
+CONFIG_HIDRAW=y
+CONFIG_UHID=y
+CONFIG_HID_A4TECH=y
+CONFIG_HID_ACRUX=y
+CONFIG_HID_ACRUX_FF=y
+CONFIG_HID_APPLE=y
+CONFIG_HID_BELKIN=y
+CONFIG_HID_CHERRY=y
+CONFIG_HID_CHICONY=y
+CONFIG_HID_PRODIKEYS=y
+CONFIG_HID_CYPRESS=y
+CONFIG_HID_DRAGONRISE=y
+CONFIG_DRAGONRISE_FF=y
+CONFIG_HID_EMS_FF=y
+CONFIG_HID_ELECOM=y
+CONFIG_HID_EZKEY=y
+CONFIG_HID_HOLTEK=y
+CONFIG_HID_KEYTOUCH=y
+CONFIG_HID_KYE=y
+CONFIG_HID_UCLOGIC=y
+CONFIG_HID_WALTOP=y
+CONFIG_HID_GYRATION=y
+CONFIG_HID_TWINHAN=y
+CONFIG_HID_KENSINGTON=y
+CONFIG_HID_LCPOWER=y
+CONFIG_HID_LOGITECH=y
+CONFIG_HID_LOGITECH_DJ=y
+CONFIG_LOGITECH_FF=y
+CONFIG_LOGIRUMBLEPAD2_FF=y
+CONFIG_LOGIG940_FF=y
+CONFIG_HID_MAGICMOUSE=y
+CONFIG_HID_MICROSOFT=y
+CONFIG_HID_MONTEREY=y
+CONFIG_HID_MULTITOUCH=y
+CONFIG_HID_NTRIG=y
+CONFIG_HID_ORTEK=y
+CONFIG_HID_PANTHERLORD=y
+CONFIG_PANTHERLORD_FF=y
+CONFIG_HID_PETALYNX=y
+CONFIG_HID_PICOLCD=y
+CONFIG_HID_PRIMAX=y
+CONFIG_HID_ROCCAT=y
+CONFIG_HID_SAITEK=y
+CONFIG_HID_SAMSUNG=y
+CONFIG_HID_SONY=y
+CONFIG_HID_SPEEDLINK=y
+CONFIG_HID_SUNPLUS=y
+CONFIG_HID_GREENASIA=y
+CONFIG_GREENASIA_FF=y
+CONFIG_HID_SMARTJOYPLUS=y
+CONFIG_SMARTJOYPLUS_FF=y
+CONFIG_HID_TIVO=y
+CONFIG_HID_TOPSEED=y
+CONFIG_HID_THRUSTMASTER=y
+CONFIG_HID_WACOM=y
+CONFIG_HID_WIIMOTE=y
+CONFIG_HID_ZEROPLUS=y
+CONFIG_HID_ZYDACRON=y
+CONFIG_HID_PID=y
+CONFIG_USB_HIDDEV=y
+CONFIG_USB_ANNOUNCE_NEW_DEVICES=y
+CONFIG_USB_MON=y
+CONFIG_USB_EHCI_HCD=y
+# CONFIG_USB_EHCI_TT_NEWSCHED is not set
+CONFIG_USB_OHCI_HCD=y
+CONFIG_USB_UHCI_HCD=y
+CONFIG_USB_PRINTER=y
+CONFIG_USB_STORAGE=y
+CONFIG_USB_OTG_WAKELOCK=y
+CONFIG_EDAC=y
+CONFIG_RTC_CLASS=y
+# CONFIG_RTC_HCTOSYS is not set
+CONFIG_DMADEVICES=y
+CONFIG_VIRTIO_PCI=y
+CONFIG_STAGING=y
+CONFIG_ASHMEM=y
+CONFIG_ANDROID_LOW_MEMORY_KILLER=y
+CONFIG_SYNC=y
+CONFIG_SW_SYNC=y
+CONFIG_ION=y
+CONFIG_GOLDFISH_AUDIO=y
+CONFIG_GOLDFISH_SYNC=y
+CONFIG_SND_HDA_INTEL=y
+CONFIG_GOLDFISH=y
+CONFIG_GOLDFISH_PIPE=y
+CONFIG_ANDROID=y
+CONFIG_ANDROID_BINDER_IPC=y
+CONFIG_ISCSI_IBFT_FIND=y
+CONFIG_EXT4_FS=y
+CONFIG_EXT4_FS_SECURITY=y
+CONFIG_QUOTA=y
+CONFIG_QUOTA_NETLINK_INTERFACE=y
+# CONFIG_PRINT_QUOTA_WARNING is not set
+CONFIG_FUSE_FS=y
+CONFIG_ISO9660_FS=y
+CONFIG_JOLIET=y
+CONFIG_ZISOFS=y
+CONFIG_MSDOS_FS=y
+CONFIG_VFAT_FS=y
+CONFIG_PROC_KCORE=y
+CONFIG_TMPFS=y
+CONFIG_TMPFS_POSIX_ACL=y
+CONFIG_HUGETLBFS=y
+CONFIG_PSTORE=y
+CONFIG_PSTORE_CONSOLE=y
+CONFIG_PSTORE_RAM=y
+# CONFIG_NETWORK_FILESYSTEMS is not set
+CONFIG_NLS_DEFAULT="utf8"
+CONFIG_NLS_CODEPAGE_437=y
+CONFIG_NLS_ASCII=y
+CONFIG_NLS_ISO8859_1=y
+CONFIG_NLS_UTF8=y
+CONFIG_PRINTK_TIME=y
+CONFIG_DEBUG_INFO=y
+# CONFIG_ENABLE_WARN_DEPRECATED is not set
+# CONFIG_ENABLE_MUST_CHECK is not set
+# CONFIG_UNUSED_SYMBOLS is not set
+CONFIG_MAGIC_SYSRQ=y
+CONFIG_DEBUG_MEMORY_INIT=y
+CONFIG_PANIC_TIMEOUT=5
+CONFIG_SCHEDSTATS=y
+CONFIG_TIMER_STATS=y
+CONFIG_SCHED_TRACER=y
+CONFIG_BLK_DEV_IO_TRACE=y
+CONFIG_PROVIDE_OHCI1394_DMA_INIT=y
+CONFIG_KEYS=y
+CONFIG_SECURITY=y
+CONFIG_SECURITY_NETWORK=y
+CONFIG_SECURITY_SELINUX=y
+CONFIG_CRYPTO_TWOFISH=y
+CONFIG_ASYMMETRIC_KEY_TYPE=y
+CONFIG_ASYMMETRIC_PUBLIC_KEY_SUBTYPE=y
+CONFIG_X509_CERTIFICATE_PARSER=y
+CONFIG_PKCS7_MESSAGE_PARSER=y
+CONFIG_PKCS7_TEST_KEY=y
+# CONFIG_VIRTUALIZATION is not set
+CONFIG_CRC_T10DIF=y
diff --git a/arch/x86/crypto/aes_ctrby8_avx-x86_64.S b/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
index a916c4a61165..5f6a5af9c489 100644
--- a/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
+++ b/arch/x86/crypto/aes_ctrby8_avx-x86_64.S
@@ -65,7 +65,6 @@
 #include <linux/linkage.h>
 #include <asm/inst.h>
 
-#define CONCAT(a,b)	a##b
 #define VMOVDQ		vmovdqu
 
 #define xdata0		%xmm0
@@ -92,8 +91,6 @@
 #define num_bytes	%r8
 
 #define tmp		%r10
-#define	DDQ(i)		CONCAT(ddq_add_,i)
-#define	XMM(i)		CONCAT(%xmm, i)
 #define	DDQ_DATA	0
 #define	XDATA		1
 #define KEY_128		1
@@ -131,12 +128,12 @@ ddq_add_8:
 /* generate a unique variable for ddq_add_x */
 
 .macro setddq n
-	var_ddq_add = DDQ(\n)
+	var_ddq_add = ddq_add_\n
 .endm
 
 /* generate a unique variable for xmm register */
 .macro setxdata n
-	var_xdata = XMM(\n)
+	var_xdata = %xmm\n
 .endm
 
 /* club the numeric 'id' to the symbol 'name' */
diff --git a/arch/x86/entry/vdso/Makefile b/arch/x86/entry/vdso/Makefile
index 265c0ed68118..7af017a8958f 100644
--- a/arch/x86/entry/vdso/Makefile
+++ b/arch/x86/entry/vdso/Makefile
@@ -187,10 +187,10 @@ vdso_img_insttargets := $(vdso_img_sodbg:%.dbg=install_%)
 $(MODLIB)/vdso: FORCE
 	@mkdir -p $(MODLIB)/vdso
 
-$(vdso_img_insttargets): install_%: $(obj)/%.dbg $(MODLIB)/vdso FORCE
+$(vdso_img_insttargets): install_%: $(obj)/%.dbg $(MODLIB)/vdso
 	$(call cmd,vdso_install)
 
 PHONY += vdso_install $(vdso_img_insttargets)
-vdso_install: $(vdso_img_insttargets) FORCE
+vdso_install: $(vdso_img_insttargets)
 
 clean-files := vdso32.so vdso32.so.dbg vdso64* vdso-image-*.c vdsox32.so*
diff --git a/arch/x86/include/asm/arch_hweight.h b/arch/x86/include/asm/arch_hweight.h
index 259a7c1ef709..44f825c80ed5 100644
--- a/arch/x86/include/asm/arch_hweight.h
+++ b/arch/x86/include/asm/arch_hweight.h
@@ -2,8 +2,8 @@
 #define _ASM_X86_HWEIGHT_H
 
 #ifdef CONFIG_64BIT
-/* popcnt %edi, %eax -- redundant REX prefix for alignment */
-#define POPCNT32 ".byte 0xf3,0x40,0x0f,0xb8,0xc7"
+/* popcnt %edi, %eax */
+#define POPCNT32 ".byte 0xf3,0x0f,0xb8,0xc7"
 /* popcnt %rdi, %rax */
 #define POPCNT64 ".byte 0xf3,0x48,0x0f,0xb8,0xc7"
 #define REG_IN "D"
@@ -15,19 +15,15 @@
 #define REG_OUT "a"
 #endif
 
-/*
- * __sw_hweightXX are called from within the alternatives below
- * and callee-clobbered registers need to be taken care of. See
- * ARCH_HWEIGHT_CFLAGS in <arch/x86/Kconfig> for the respective
- * compiler switches.
- */
+#define __HAVE_ARCH_SW_HWEIGHT
+
 static __always_inline unsigned int __arch_hweight32(unsigned int w)
 {
-	unsigned int res = 0;
+	unsigned int res;
 
 	asm (ALTERNATIVE("call __sw_hweight32", POPCNT32, X86_FEATURE_POPCNT)
-		     : "="REG_OUT (res)
-		     : REG_IN (w));
+			 : "="REG_OUT (res)
+			 : REG_IN (w));
 
 	return res;
 }
@@ -51,11 +47,11 @@ static inline unsigned long __arch_hweight64(__u64 w)
 #else
 static __always_inline unsigned long __arch_hweight64(__u64 w)
 {
-	unsigned long res = 0;
+	unsigned long res;
 
 	asm (ALTERNATIVE("call __sw_hweight64", POPCNT64, X86_FEATURE_POPCNT)
-		     : "="REG_OUT (res)
-		     : REG_IN (w));
+			 : "="REG_OUT (res)
+			 : REG_IN (w));
 
 	return res;
 }
diff --git a/arch/x86/include/asm/asm.h b/arch/x86/include/asm/asm.h
index 189679aba703..1f16ec50abeb 100644
--- a/arch/x86/include/asm/asm.h
+++ b/arch/x86/include/asm/asm.h
@@ -32,6 +32,7 @@
 #define _ASM_ADD	__ASM_SIZE(add)
 #define _ASM_SUB	__ASM_SIZE(sub)
 #define _ASM_XADD	__ASM_SIZE(xadd)
+#define _ASM_MUL	__ASM_SIZE(mul)
 
 #define _ASM_AX		__ASM_REG(ax)
 #define _ASM_BX		__ASM_REG(bx)
diff --git a/arch/x86/include/asm/idle.h b/arch/x86/include/asm/idle.h
index c5d1785373ed..02bab09707f2 100644
--- a/arch/x86/include/asm/idle.h
+++ b/arch/x86/include/asm/idle.h
@@ -1,13 +1,6 @@
 #ifndef _ASM_X86_IDLE_H
 #define _ASM_X86_IDLE_H
 
-#define IDLE_START 1
-#define IDLE_END 2
-
-struct notifier_block;
-void idle_notifier_register(struct notifier_block *n);
-void idle_notifier_unregister(struct notifier_block *n);
-
 #ifdef CONFIG_X86_64
 void enter_idle(void);
 void exit_idle(void);
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index e9cab5c75865..6a07c05956a6 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -716,7 +716,7 @@ __copy_from_user_overflow(int size, unsigned long count)
 
 #endif
 
-static inline unsigned long __must_check
+static __always_inline unsigned long __must_check
 copy_from_user(void *to, const void __user *from, unsigned long n)
 {
 	int sz = __compiletime_object_size(to);
@@ -752,7 +752,7 @@ copy_from_user(void *to, const void __user *from, unsigned long n)
 	return n;
 }
 
-static inline unsigned long __must_check
+static __always_inline unsigned long __must_check
 copy_to_user(void __user *to, const void *from, unsigned long n)
 {
 	int sz = __compiletime_object_size(from);
diff --git a/arch/x86/kernel/i386_ksyms_32.c b/arch/x86/kernel/i386_ksyms_32.c
index 64341aa485ae..d40ee8a38fed 100644
--- a/arch/x86/kernel/i386_ksyms_32.c
+++ b/arch/x86/kernel/i386_ksyms_32.c
@@ -42,3 +42,5 @@ EXPORT_SYMBOL(empty_zero_page);
 EXPORT_SYMBOL(___preempt_schedule);
 EXPORT_SYMBOL(___preempt_schedule_notrace);
 #endif
+
+EXPORT_SYMBOL(__sw_hweight32);
diff --git a/arch/x86/kernel/livepatch.c b/arch/x86/kernel/livepatch.c
index d1d35ccffed3..bcc06e82a593 100644
--- a/arch/x86/kernel/livepatch.c
+++ b/arch/x86/kernel/livepatch.c
@@ -41,8 +41,8 @@ int klp_write_module_reloc(struct module *mod, unsigned long type,
 	int ret, numpages, size = 4;
 	bool readonly;
 	unsigned long val;
-	unsigned long core = (unsigned long)mod->module_core;
-	unsigned long core_size = mod->core_size;
+	unsigned long core = (unsigned long)mod->core_layout.base;
+	unsigned long core_size = mod->core_layout.size;
 
 	switch (type) {
 	case R_X86_64_NONE:
@@ -72,7 +72,7 @@ int klp_write_module_reloc(struct module *mod, unsigned long type,
 	readonly = false;
 
 #ifdef CONFIG_DEBUG_SET_MODULE_RONX
-	if (loc < core + mod->core_ro_size)
+	if (loc < core + mod->core_layout.ro_size)
 		readonly = true;
 #endif
 
diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c
index 9f7c21c22477..57eca132962f 100644
--- a/arch/x86/kernel/process.c
+++ b/arch/x86/kernel/process.c
@@ -62,19 +62,6 @@ EXPORT_PER_CPU_SYMBOL(cpu_tss);
 
 #ifdef CONFIG_X86_64
 static DEFINE_PER_CPU(unsigned char, is_idle);
-static ATOMIC_NOTIFIER_HEAD(idle_notifier);
-
-void idle_notifier_register(struct notifier_block *n)
-{
-	atomic_notifier_chain_register(&idle_notifier, n);
-}
-EXPORT_SYMBOL_GPL(idle_notifier_register);
-
-void idle_notifier_unregister(struct notifier_block *n)
-{
-	atomic_notifier_chain_unregister(&idle_notifier, n);
-}
-EXPORT_SYMBOL_GPL(idle_notifier_unregister);
 #endif
 
 /*
@@ -251,14 +238,14 @@ static inline void play_dead(void)
 void enter_idle(void)
 {
 	this_cpu_write(is_idle, 1);
-	atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
+	idle_notifier_call_chain(IDLE_START);
 }
 
 static void __exit_idle(void)
 {
 	if (x86_test_and_clear_bit_percpu(0, is_idle) == 0)
 		return;
-	atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL);
+	idle_notifier_call_chain(IDLE_END);
 }
 
 /* Called from interrupts to signify idle end */
diff --git a/arch/x86/kernel/x8664_ksyms_64.c b/arch/x86/kernel/x8664_ksyms_64.c
index a0695be19864..c7efd394c42b 100644
--- a/arch/x86/kernel/x8664_ksyms_64.c
+++ b/arch/x86/kernel/x8664_ksyms_64.c
@@ -42,6 +42,9 @@ EXPORT_SYMBOL(clear_page);
 
 EXPORT_SYMBOL(csum_partial);
 
+EXPORT_SYMBOL(__sw_hweight32);
+EXPORT_SYMBOL(__sw_hweight64);
+
 /*
  * Export string functions. We normally rely on gcc builtin for most of these,
  * but gcc sometimes decides not to inline them.
diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile
index f2587888d987..db4a1c6ea785 100644
--- a/arch/x86/lib/Makefile
+++ b/arch/x86/lib/Makefile
@@ -22,7 +22,7 @@ lib-y += memcpy_$(BITS).o
 lib-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem.o
 lib-$(CONFIG_INSTRUCTION_DECODER) += insn.o inat.o
 
-obj-y += msr.o msr-reg.o msr-reg-export.o
+obj-y += msr.o msr-reg.o msr-reg-export.o hweight.o
 
 ifeq ($(CONFIG_X86_32),y)
         obj-y += atomic64_32.o
diff --git a/arch/x86/lib/hweight.S b/arch/x86/lib/hweight.S
new file mode 100644
index 000000000000..8a602a1e404a
--- /dev/null
+++ b/arch/x86/lib/hweight.S
@@ -0,0 +1,79 @@
+#include <linux/linkage.h>
+
+#include <asm/asm.h>
+
+/*
+ * unsigned int __sw_hweight32(unsigned int w)
+ * %rdi: w
+ */
+ENTRY(__sw_hweight32)
+
+#ifdef CONFIG_X86_64
+	movl %edi, %eax				# w
+#endif
+	__ASM_SIZE(push,) %__ASM_REG(dx)
+	movl %eax, %edx				# w -> t
+	shrl %edx				# t >>= 1
+	andl $0x55555555, %edx			# t &= 0x55555555
+	subl %edx, %eax				# w -= t
+
+	movl %eax, %edx				# w -> t
+	shrl $2, %eax				# w_tmp >>= 2
+	andl $0x33333333, %edx			# t	&= 0x33333333
+	andl $0x33333333, %eax			# w_tmp &= 0x33333333
+	addl %edx, %eax				# w = w_tmp + t
+
+	movl %eax, %edx				# w -> t
+	shrl $4, %edx				# t >>= 4
+	addl %edx, %eax				# w_tmp += t
+	andl  $0x0f0f0f0f, %eax			# w_tmp &= 0x0f0f0f0f
+	imull $0x01010101, %eax, %eax		# w_tmp *= 0x01010101
+	shrl $24, %eax				# w = w_tmp >> 24
+	__ASM_SIZE(pop,) %__ASM_REG(dx)
+	ret
+ENDPROC(__sw_hweight32)
+
+ENTRY(__sw_hweight64)
+#ifdef CONFIG_X86_64
+	pushq   %rdi
+	pushq   %rdx
+
+	movq    %rdi, %rdx                      # w -> t
+	movabsq $0x5555555555555555, %rax
+	shrq    %rdx                            # t >>= 1
+	andq    %rdx, %rax                      # t &= 0x5555555555555555
+	movabsq $0x3333333333333333, %rdx
+	subq    %rax, %rdi                      # w -= t
+
+	movq    %rdi, %rax                      # w -> t
+	shrq    $2, %rdi                        # w_tmp >>= 2
+	andq    %rdx, %rax                      # t     &= 0x3333333333333333
+	andq    %rdi, %rdx                      # w_tmp &= 0x3333333333333333
+	addq    %rdx, %rax                      # w = w_tmp + t
+
+	movq    %rax, %rdx                      # w -> t
+	shrq    $4, %rdx                        # t >>= 4
+	addq    %rdx, %rax                      # w_tmp += t
+	movabsq $0x0f0f0f0f0f0f0f0f, %rdx
+	andq    %rdx, %rax                      # w_tmp &= 0x0f0f0f0f0f0f0f0f
+	movabsq $0x0101010101010101, %rdx
+	imulq   %rdx, %rax                      # w_tmp *= 0x0101010101010101
+	shrq    $56, %rax                       # w = w_tmp >> 56
+
+	popq    %rdx
+	popq    %rdi
+	ret
+#else /* CONFIG_X86_32 */
+	/* We're getting an u64 arg in (%eax,%edx): unsigned long hweight64(__u64 w) */
+	pushl   %ecx
+
+	call    __sw_hweight32
+	movl    %eax, %ecx                      # stash away result
+	movl    %edx, %eax                      # second part of input
+	call    __sw_hweight32
+	addl    %ecx, %eax                      # result
+
+	popl    %ecx
+	ret
+#endif
+ENDPROC(__sw_hweight64)
diff --git a/arch/x86/mm/mmap.c b/arch/x86/mm/mmap.c
index 307f60ecfc6d..d2dc0438d654 100644
--- a/arch/x86/mm/mmap.c
+++ b/arch/x86/mm/mmap.c
@@ -69,14 +69,14 @@ unsigned long arch_mmap_rnd(void)
 {
 	unsigned long rnd;
 
-	/*
-	 *  8 bits of randomness in 32bit mmaps, 20 address space bits
-	 * 28 bits of randomness in 64bit mmaps, 40 address space bits
-	 */
 	if (mmap_is_ia32())
-		rnd = (unsigned long)get_random_int() % (1<<8);
+#ifdef CONFIG_COMPAT
+		rnd = get_random_long() & ((1UL << mmap_rnd_compat_bits) - 1);
+#else
+		rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
+#endif
 	else
-		rnd = (unsigned long)get_random_int() % (1<<28);
+		rnd = get_random_long() & ((1UL << mmap_rnd_bits) - 1);
 
 	return rnd << PAGE_SHIFT;
 }
diff --git a/block/blk-core.c b/block/blk-core.c
index 119658534dfd..323cc15b155a 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -40,6 +40,8 @@
 #include "blk.h"
 #include "blk-mq.h"
 
+#include <linux/math64.h>
+
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_remap);
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_rq_remap);
 EXPORT_TRACEPOINT_SYMBOL_GPL(block_bio_complete);
@@ -3563,3 +3565,85 @@ int __init blk_dev_init(void)
 
 	return 0;
 }
+
+/*
+ * Blk IO latency support. We want this to be as cheap as possible, so doing
+ * this lockless (and avoiding atomics), a few off by a few errors in this
+ * code is not harmful, and we don't want to do anything that is
+ * perf-impactful.
+ * TODO : If necessary, we can make the histograms per-cpu and aggregate
+ * them when printing them out.
+ */
+void
+blk_zero_latency_hist(struct io_latency_state *s)
+{
+	memset(s->latency_y_axis_read, 0,
+	       sizeof(s->latency_y_axis_read));
+	memset(s->latency_y_axis_write, 0,
+	       sizeof(s->latency_y_axis_write));
+	s->latency_reads_elems = 0;
+	s->latency_writes_elems = 0;
+}
+EXPORT_SYMBOL(blk_zero_latency_hist);
+
+ssize_t
+blk_latency_hist_show(struct io_latency_state *s, char *buf)
+{
+	int i;
+	int bytes_written = 0;
+	u_int64_t num_elem, elem;
+	int pct;
+
+	num_elem = s->latency_reads_elems;
+	if (num_elem > 0) {
+		bytes_written += scnprintf(buf + bytes_written,
+			   PAGE_SIZE - bytes_written,
+			   "IO svc_time Read Latency Histogram (n = %llu):\n",
+			   num_elem);
+		for (i = 0;
+		     i < ARRAY_SIZE(latency_x_axis_us);
+		     i++) {
+			elem = s->latency_y_axis_read[i];
+			pct = div64_u64(elem * 100, num_elem);
+			bytes_written += scnprintf(buf + bytes_written,
+						   PAGE_SIZE - bytes_written,
+						   "\t< %5lluus%15llu%15d%%\n",
+						   latency_x_axis_us[i],
+						   elem, pct);
+		}
+		/* Last element in y-axis table is overflow */
+		elem = s->latency_y_axis_read[i];
+		pct = div64_u64(elem * 100, num_elem);
+		bytes_written += scnprintf(buf + bytes_written,
+					   PAGE_SIZE - bytes_written,
+					   "\t> %5dms%15llu%15d%%\n", 10,
+					   elem, pct);
+	}
+	num_elem = s->latency_writes_elems;
+	if (num_elem > 0) {
+		bytes_written += scnprintf(buf + bytes_written,
+			   PAGE_SIZE - bytes_written,
+			   "IO svc_time Write Latency Histogram (n = %llu):\n",
+			   num_elem);
+		for (i = 0;
+		     i < ARRAY_SIZE(latency_x_axis_us);
+		     i++) {
+			elem = s->latency_y_axis_write[i];
+			pct = div64_u64(elem * 100, num_elem);
+			bytes_written += scnprintf(buf + bytes_written,
+						   PAGE_SIZE - bytes_written,
+						   "\t< %5lluus%15llu%15d%%\n",
+						   latency_x_axis_us[i],
+						   elem, pct);
+		}
+		/* Last element in y-axis table is overflow */
+		elem = s->latency_y_axis_write[i];
+		pct = div64_u64(elem * 100, num_elem);
+		bytes_written += scnprintf(buf + bytes_written,
+					   PAGE_SIZE - bytes_written,
+					   "\t> %5dms%15llu%15d%%\n", 10,
+					   elem, pct);
+	}
+	return bytes_written;
+}
+EXPORT_SYMBOL(blk_latency_hist_show);
diff --git a/block/genhd.c b/block/genhd.c
index 3032453a89e6..a87e77626504 100644
--- a/block/genhd.c
+++ b/block/genhd.c
@@ -1117,6 +1117,22 @@ static void disk_release(struct device *dev)
 		blk_put_queue(disk->queue);
 	kfree(disk);
 }
+
+static int disk_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct gendisk *disk = dev_to_disk(dev);
+	struct disk_part_iter piter;
+	struct hd_struct *part;
+	int cnt = 0;
+
+	disk_part_iter_init(&piter, disk, 0);
+	while((part = disk_part_iter_next(&piter)))
+		cnt++;
+	disk_part_iter_exit(&piter);
+	add_uevent_var(env, "NPARTS=%u", cnt);
+	return 0;
+}
+
 struct class block_class = {
 	.name		= "block",
 };
@@ -1136,6 +1152,7 @@ static struct device_type disk_type = {
 	.groups		= disk_attr_groups,
 	.release	= disk_release,
 	.devnode	= block_devnode,
+	.uevent		= disk_uevent,
 };
 
 #ifdef CONFIG_PROC_FS
diff --git a/block/partition-generic.c b/block/partition-generic.c
index 3c062699f28b..19cf33b91a5a 100644
--- a/block/partition-generic.c
+++ b/block/partition-generic.c
@@ -216,10 +216,21 @@ static void part_release(struct device *dev)
 	kfree(p);
 }
 
+static int part_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct hd_struct *part = dev_to_part(dev);
+
+	add_uevent_var(env, "PARTN=%u", part->partno);
+	if (part->info && part->info->volname[0])
+		add_uevent_var(env, "PARTNAME=%s", part->info->volname);
+	return 0;
+}
+
 struct device_type part_type = {
 	.name		= "partition",
 	.groups		= part_attr_groups,
 	.release	= part_release,
+	.uevent		= part_uevent,
 };
 
 static void delete_partition_rcu_cb(struct rcu_head *head)
diff --git a/crypto/Kconfig b/crypto/Kconfig
index 7240821137fd..3240d394426c 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -289,6 +289,24 @@ config CRYPTO_CBC
 	  CBC: Cipher Block Chaining mode
 	  This block cipher algorithm is required for IPSec.
 
+config CRYPTO_HEH
+	tristate "HEH support"
+	select CRYPTO_CMAC
+	select CRYPTO_ECB
+	select CRYPTO_GF128MUL
+	select CRYPTO_MANAGER
+	select CRYPTO_POLY_HASH_ARM64_CE if ARM64 && KERNEL_MODE_NEON
+	help
+	  HEH: Hash-Encrypt-Hash mode
+	  HEH is a proposed block cipher mode of operation which extends the
+	  strong pseudo-random permutation (SPRP) property of block ciphers to
+	  arbitrary-length input strings.  This provides a stronger notion of
+	  security than existing block cipher modes of operation (e.g. CBC, CTR,
+	  XTS), though it is usually less performant.  Applications include disk
+	  encryption and encryption of file names and contents.  Currently, this
+	  implementation only provides a symmetric cipher interface, so it can't
+	  yet be used as an AEAD.
+
 config CRYPTO_CTR
 	tristate "CTR support"
 	select CRYPTO_BLKCIPHER
diff --git a/crypto/Makefile b/crypto/Makefile
index 03e66097eb0c..8507d1fab3ac 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -67,6 +67,7 @@ obj-$(CONFIG_CRYPTO_TGR192) += tgr192.o
 obj-$(CONFIG_CRYPTO_GF128MUL) += gf128mul.o
 obj-$(CONFIG_CRYPTO_ECB) += ecb.o
 obj-$(CONFIG_CRYPTO_CBC) += cbc.o
+obj-$(CONFIG_CRYPTO_HEH) += heh.o
 obj-$(CONFIG_CRYPTO_PCBC) += pcbc.o
 obj-$(CONFIG_CRYPTO_CTS) += cts.o
 obj-$(CONFIG_CRYPTO_LRW) += lrw.o
diff --git a/crypto/blkcipher.c b/crypto/blkcipher.c
index dca7bc87dad9..7bbfadc195a6 100644
--- a/crypto/blkcipher.c
+++ b/crypto/blkcipher.c
@@ -373,6 +373,27 @@ int blkcipher_aead_walk_virt_block(struct blkcipher_desc *desc,
 }
 EXPORT_SYMBOL_GPL(blkcipher_aead_walk_virt_block);
 
+/*
+ * This function allows ablkcipher algorithms to use the blkcipher_walk API to
+ * walk over their data.  The specified crypto_ablkcipher tfm is used to
+ * initialize the struct blkcipher_walk, and the crypto_blkcipher specified in
+ * desc->tfm is never used so it can be left NULL.  (Yes, this design is ugly,
+ * but it parallels blkcipher_aead_walk_virt_block() above.  In the 4.10 kernel
+ * this is starting to be cleaned up...)
+ */
+int blkcipher_ablkcipher_walk_virt(struct blkcipher_desc *desc,
+				   struct blkcipher_walk *walk,
+				   struct crypto_ablkcipher *tfm)
+{
+	walk->flags &= ~BLKCIPHER_WALK_PHYS;
+	walk->walk_blocksize = crypto_ablkcipher_blocksize(tfm);
+	walk->cipher_blocksize = walk->walk_blocksize;
+	walk->ivsize = crypto_ablkcipher_ivsize(tfm);
+	walk->alignmask = crypto_ablkcipher_alignmask(tfm);
+	return blkcipher_walk_first(desc, walk);
+}
+EXPORT_SYMBOL_GPL(blkcipher_ablkcipher_walk_virt);
+
 static int setkey_unaligned(struct crypto_tfm *tfm, const u8 *key,
 			    unsigned int keylen)
 {
diff --git a/crypto/gf128mul.c b/crypto/gf128mul.c
index 5276607c72d0..f3d9f6da0767 100644
--- a/crypto/gf128mul.c
+++ b/crypto/gf128mul.c
@@ -44,7 +44,7 @@
  ---------------------------------------------------------------------------
  Issue 31/01/2006
 
- This file provides fast multiplication in GF(128) as required by several
+ This file provides fast multiplication in GF(2^128) as required by several
  cryptographic authentication modes
 */
 
@@ -88,37 +88,52 @@
 	q(0xf8), q(0xf9), q(0xfa), q(0xfb), q(0xfc), q(0xfd), q(0xfe), q(0xff) \
 }
 
-/*	Given the value i in 0..255 as the byte overflow when a field element
-    in GHASH is multiplied by x^8, this function will return the values that
-    are generated in the lo 16-bit word of the field value by applying the
-    modular polynomial. The values lo_byte and hi_byte are returned via the
-    macro xp_fun(lo_byte, hi_byte) so that the values can be assembled into
-    memory as required by a suitable definition of this macro operating on
-    the table above
-*/
-
-#define xx(p, q)	0x##p##q
+/*
+ * Given a value i in 0..255 as the byte overflow when a field element
+ * in GF(2^128) is multiplied by x^8, the following macro returns the
+ * 16-bit value that must be XOR-ed into the low-degree end of the
+ * product to reduce it modulo the irreducible polynomial x^128 + x^7 +
+ * x^2 + x + 1.
+ *
+ * There are two versions of the macro, and hence two tables: one for
+ * the "be" convention where the highest-order bit is the coefficient of
+ * the highest-degree polynomial term, and one for the "le" convention
+ * where the highest-order bit is the coefficient of the lowest-degree
+ * polynomial term.  In both cases the values are stored in CPU byte
+ * endianness such that the coefficients are ordered consistently across
+ * bytes, i.e. in the "be" table bits 15..0 of the stored value
+ * correspond to the coefficients of x^15..x^0, and in the "le" table
+ * bits 15..0 correspond to the coefficients of x^0..x^15.
+ *
+ * Therefore, provided that the appropriate byte endianness conversions
+ * are done by the multiplication functions (and these must be in place
+ * anyway to support both little endian and big endian CPUs), the "be"
+ * table can be used for multiplications of both "bbe" and "ble"
+ * elements, and the "le" table can be used for multiplications of both
+ * "lle" and "lbe" elements.
+ */
 
-#define xda_bbe(i) ( \
-	(i & 0x80 ? xx(43, 80) : 0) ^ (i & 0x40 ? xx(21, c0) : 0) ^ \
-	(i & 0x20 ? xx(10, e0) : 0) ^ (i & 0x10 ? xx(08, 70) : 0) ^ \
-	(i & 0x08 ? xx(04, 38) : 0) ^ (i & 0x04 ? xx(02, 1c) : 0) ^ \
-	(i & 0x02 ? xx(01, 0e) : 0) ^ (i & 0x01 ? xx(00, 87) : 0) \
+#define xda_be(i) ( \
+	(i & 0x80 ? 0x4380 : 0) ^ (i & 0x40 ? 0x21c0 : 0) ^ \
+	(i & 0x20 ? 0x10e0 : 0) ^ (i & 0x10 ? 0x0870 : 0) ^ \
+	(i & 0x08 ? 0x0438 : 0) ^ (i & 0x04 ? 0x021c : 0) ^ \
+	(i & 0x02 ? 0x010e : 0) ^ (i & 0x01 ? 0x0087 : 0) \
 )
 
-#define xda_lle(i) ( \
-	(i & 0x80 ? xx(e1, 00) : 0) ^ (i & 0x40 ? xx(70, 80) : 0) ^ \
-	(i & 0x20 ? xx(38, 40) : 0) ^ (i & 0x10 ? xx(1c, 20) : 0) ^ \
-	(i & 0x08 ? xx(0e, 10) : 0) ^ (i & 0x04 ? xx(07, 08) : 0) ^ \
-	(i & 0x02 ? xx(03, 84) : 0) ^ (i & 0x01 ? xx(01, c2) : 0) \
+#define xda_le(i) ( \
+	(i & 0x80 ? 0xe100 : 0) ^ (i & 0x40 ? 0x7080 : 0) ^ \
+	(i & 0x20 ? 0x3840 : 0) ^ (i & 0x10 ? 0x1c20 : 0) ^ \
+	(i & 0x08 ? 0x0e10 : 0) ^ (i & 0x04 ? 0x0708 : 0) ^ \
+	(i & 0x02 ? 0x0384 : 0) ^ (i & 0x01 ? 0x01c2 : 0) \
 )
 
-static const u16 gf128mul_table_lle[256] = gf128mul_dat(xda_lle);
-static const u16 gf128mul_table_bbe[256] = gf128mul_dat(xda_bbe);
+static const u16 gf128mul_table_le[256] = gf128mul_dat(xda_le);
+static const u16 gf128mul_table_be[256] = gf128mul_dat(xda_be);
 
-/* These functions multiply a field element by x, by x^4 and by x^8
- * in the polynomial field representation. It uses 32-bit word operations
- * to gain speed but compensates for machine endianess and hence works
+/*
+ * The following functions multiply a field element by x or by x^8 in
+ * the polynomial field representation.  They use 64-bit word operations
+ * to gain speed but compensate for machine endianness and hence work
  * correctly on both styles of machine.
  */
 
@@ -126,7 +141,7 @@ static void gf128mul_x_lle(be128 *r, const be128 *x)
 {
 	u64 a = be64_to_cpu(x->a);
 	u64 b = be64_to_cpu(x->b);
-	u64 _tt = gf128mul_table_lle[(b << 7) & 0xff];
+	u64 _tt = gf128mul_table_le[(b << 7) & 0xff];
 
 	r->b = cpu_to_be64((b >> 1) | (a << 63));
 	r->a = cpu_to_be64((a >> 1) ^ (_tt << 48));
@@ -136,7 +151,7 @@ static void gf128mul_x_bbe(be128 *r, const be128 *x)
 {
 	u64 a = be64_to_cpu(x->a);
 	u64 b = be64_to_cpu(x->b);
-	u64 _tt = gf128mul_table_bbe[a >> 63];
+	u64 _tt = gf128mul_table_be[a >> 63];
 
 	r->a = cpu_to_be64((a << 1) | (b >> 63));
 	r->b = cpu_to_be64((b << 1) ^ _tt);
@@ -146,7 +161,7 @@ void gf128mul_x_ble(be128 *r, const be128 *x)
 {
 	u64 a = le64_to_cpu(x->a);
 	u64 b = le64_to_cpu(x->b);
-	u64 _tt = gf128mul_table_bbe[b >> 63];
+	u64 _tt = gf128mul_table_be[b >> 63];
 
 	r->a = cpu_to_le64((a << 1) ^ _tt);
 	r->b = cpu_to_le64((b << 1) | (a >> 63));
@@ -157,7 +172,7 @@ static void gf128mul_x8_lle(be128 *x)
 {
 	u64 a = be64_to_cpu(x->a);
 	u64 b = be64_to_cpu(x->b);
-	u64 _tt = gf128mul_table_lle[b & 0xff];
+	u64 _tt = gf128mul_table_le[b & 0xff];
 
 	x->b = cpu_to_be64((b >> 8) | (a << 56));
 	x->a = cpu_to_be64((a >> 8) ^ (_tt << 48));
@@ -167,12 +182,22 @@ static void gf128mul_x8_bbe(be128 *x)
 {
 	u64 a = be64_to_cpu(x->a);
 	u64 b = be64_to_cpu(x->b);
-	u64 _tt = gf128mul_table_bbe[a >> 56];
+	u64 _tt = gf128mul_table_be[a >> 56];
 
 	x->a = cpu_to_be64((a << 8) | (b >> 56));
 	x->b = cpu_to_be64((b << 8) ^ _tt);
 }
 
+static void gf128mul_x8_ble(be128 *x)
+{
+	u64 a = le64_to_cpu(x->b);
+	u64 b = le64_to_cpu(x->a);
+	u64 _tt = gf128mul_table_be[a >> 56];
+
+	x->b = cpu_to_le64((a << 8) | (b >> 56));
+	x->a = cpu_to_le64((b << 8) ^ _tt);
+}
+
 void gf128mul_lle(be128 *r, const be128 *b)
 {
 	be128 p[8];
@@ -249,9 +274,48 @@ void gf128mul_bbe(be128 *r, const be128 *b)
 }
 EXPORT_SYMBOL(gf128mul_bbe);
 
+void gf128mul_ble(be128 *r, const be128 *b)
+{
+	be128 p[8];
+	int i;
+
+	p[0] = *r;
+	for (i = 0; i < 7; ++i)
+		gf128mul_x_ble((be128 *)&p[i + 1], (be128 *)&p[i]);
+
+	memset(r, 0, sizeof(*r));
+	for (i = 0;;) {
+		u8 ch = ((u8 *)b)[15 - i];
+
+		if (ch & 0x80)
+			be128_xor(r, r, &p[7]);
+		if (ch & 0x40)
+			be128_xor(r, r, &p[6]);
+		if (ch & 0x20)
+			be128_xor(r, r, &p[5]);
+		if (ch & 0x10)
+			be128_xor(r, r, &p[4]);
+		if (ch & 0x08)
+			be128_xor(r, r, &p[3]);
+		if (ch & 0x04)
+			be128_xor(r, r, &p[2]);
+		if (ch & 0x02)
+			be128_xor(r, r, &p[1]);
+		if (ch & 0x01)
+			be128_xor(r, r, &p[0]);
+
+		if (++i >= 16)
+			break;
+
+		gf128mul_x8_ble(r);
+	}
+}
+EXPORT_SYMBOL(gf128mul_ble);
+
+
 /*      This version uses 64k bytes of table space.
     A 16 byte buffer has to be multiplied by a 16 byte key
-    value in GF(128).  If we consider a GF(128) value in
+    value in GF(2^128).  If we consider a GF(2^128) value in
     the buffer's lowest byte, we can construct a table of
     the 256 16 byte values that result from the 256 values
     of this byte.  This requires 4096 bytes. But we also
@@ -352,8 +416,8 @@ void gf128mul_free_64k(struct gf128mul_64k *t)
 	int i;
 
 	for (i = 0; i < 16; i++)
-		kfree(t->t[i]);
-	kfree(t);
+		kzfree(t->t[i]);
+	kzfree(t);
 }
 EXPORT_SYMBOL(gf128mul_free_64k);
 
@@ -385,7 +449,7 @@ EXPORT_SYMBOL(gf128mul_64k_bbe);
 
 /*      This version uses 4k bytes of table space.
     A 16 byte buffer has to be multiplied by a 16 byte key
-    value in GF(128).  If we consider a GF(128) value in a
+    value in GF(2^128).  If we consider a GF(2^128) value in a
     single byte, we can construct a table of the 256 16 byte
     values that result from the 256 values of this byte.
     This requires 4096 bytes. If we take the highest byte in
@@ -443,6 +507,28 @@ out:
 }
 EXPORT_SYMBOL(gf128mul_init_4k_bbe);
 
+struct gf128mul_4k *gf128mul_init_4k_ble(const be128 *g)
+{
+	struct gf128mul_4k *t;
+	int j, k;
+
+	t = kzalloc(sizeof(*t), GFP_KERNEL);
+	if (!t)
+		goto out;
+
+	t->t[1] = *g;
+	for (j = 1; j <= 64; j <<= 1)
+		gf128mul_x_ble(&t->t[j + j], &t->t[j]);
+
+	for (j = 2; j < 256; j += j)
+		for (k = 1; k < j; ++k)
+			be128_xor(&t->t[j + k], &t->t[j], &t->t[k]);
+
+out:
+	return t;
+}
+EXPORT_SYMBOL(gf128mul_init_4k_ble);
+
 void gf128mul_4k_lle(be128 *a, struct gf128mul_4k *t)
 {
 	u8 *ap = (u8 *)a;
@@ -473,5 +559,20 @@ void gf128mul_4k_bbe(be128 *a, struct gf128mul_4k *t)
 }
 EXPORT_SYMBOL(gf128mul_4k_bbe);
 
+void gf128mul_4k_ble(be128 *a, struct gf128mul_4k *t)
+{
+	u8 *ap = (u8 *)a;
+	be128 r[1];
+	int i = 15;
+
+	*r = t->t[ap[15]];
+	while (i--) {
+		gf128mul_x8_ble(r);
+		be128_xor(r, r, &t->t[ap[i]]);
+	}
+	*a = *r;
+}
+EXPORT_SYMBOL(gf128mul_4k_ble);
+
 MODULE_LICENSE("GPL");
 MODULE_DESCRIPTION("Functions for multiplying elements of GF(2^128)");
diff --git a/crypto/heh.c b/crypto/heh.c
new file mode 100644
index 000000000000..10c00aaf797e
--- /dev/null
+++ b/crypto/heh.c
@@ -0,0 +1,1033 @@
+/*
+ * HEH: Hash-Encrypt-Hash mode
+ *
+ * Copyright (c) 2016 Google Inc.
+ *
+ * Authors:
+ *	Alex Cope <alexcope@google.com>
+ *	Eric Biggers <ebiggers@google.com>
+ */
+
+/*
+ * Hash-Encrypt-Hash (HEH) is a proposed block cipher mode of operation which
+ * extends the strong pseudo-random permutation (SPRP) property of block ciphers
+ * (e.g. AES) to arbitrary length input strings.  It uses two keyed invertible
+ * hash functions with a layer of ECB encryption applied in-between.  The
+ * algorithm is specified by the following Internet Draft:
+ *
+ *	https://tools.ietf.org/html/draft-cope-heh-01
+ *
+ * Although HEH can be used as either a regular symmetric cipher or as an AEAD,
+ * currently this module only provides it as a symmetric cipher.  Additionally,
+ * only 16-byte nonces are supported.
+ */
+
+#include <crypto/gf128mul.h>
+#include <crypto/internal/hash.h>
+#include <crypto/internal/skcipher.h>
+#include <crypto/scatterwalk.h>
+#include <crypto/skcipher.h>
+#include "internal.h"
+
+/*
+ * The block size is the size of GF(2^128) elements and also the required block
+ * size of the underlying block cipher.
+ */
+#define HEH_BLOCK_SIZE		16
+
+struct heh_instance_ctx {
+	struct crypto_shash_spawn cmac;
+	struct crypto_shash_spawn poly_hash;
+	struct crypto_skcipher_spawn ecb;
+};
+
+struct heh_tfm_ctx {
+	struct crypto_shash *cmac;
+	struct crypto_shash *poly_hash; /* keyed with tau_key */
+	struct crypto_ablkcipher *ecb;
+};
+
+struct heh_cmac_data {
+	u8 nonce[HEH_BLOCK_SIZE];
+	__le32 nonce_length;
+	__le32 aad_length;
+	__le32 message_length;
+	__le32 padding;
+};
+
+struct heh_req_ctx { /* aligned to alignmask */
+	be128 beta1_key;
+	be128 beta2_key;
+	union {
+		struct {
+			struct heh_cmac_data data;
+			struct shash_desc desc;
+			/* + crypto_shash_descsize(cmac) */
+		} cmac;
+		struct {
+			struct shash_desc desc;
+			/* + crypto_shash_descsize(poly_hash) */
+		} poly_hash;
+		struct {
+			u8 keystream[HEH_BLOCK_SIZE];
+			u8 tmp[HEH_BLOCK_SIZE];
+			struct scatterlist tmp_sgl[2];
+			struct ablkcipher_request req;
+			/* + crypto_ablkcipher_reqsize(ecb) */
+		} ecb;
+	} u;
+};
+
+/*
+ * Get the offset in bytes to the last full block, or equivalently the length of
+ * all full blocks excluding the last
+ */
+static inline unsigned int get_tail_offset(unsigned int len)
+{
+	len -= len % HEH_BLOCK_SIZE;
+	return len - HEH_BLOCK_SIZE;
+}
+
+static inline struct heh_req_ctx *heh_req_ctx(struct ablkcipher_request *req)
+{
+	unsigned int alignmask = crypto_ablkcipher_alignmask(
+						crypto_ablkcipher_reqtfm(req));
+
+	return (void *)PTR_ALIGN((u8 *)ablkcipher_request_ctx(req),
+				 alignmask + 1);
+}
+
+static inline void async_done(struct crypto_async_request *areq, int err,
+			      int (*next_step)(struct ablkcipher_request *,
+					       u32))
+{
+	struct ablkcipher_request *req = areq->data;
+
+	if (err)
+		goto out;
+
+	err = next_step(req, req->base.flags & ~CRYPTO_TFM_REQ_MAY_SLEEP);
+	if (err == -EINPROGRESS ||
+	    (err == -EBUSY && (req->base.flags & CRYPTO_TFM_REQ_MAY_BACKLOG)))
+		return;
+out:
+	ablkcipher_request_complete(req, err);
+}
+
+/*
+ * Generate the per-message "beta" keys used by the hashing layers of HEH.  The
+ * first beta key is the CMAC of the nonce, the additional authenticated data
+ * (AAD), and the lengths in bytes of the nonce, AAD, and message.  The nonce
+ * and AAD are each zero-padded to the next 16-byte block boundary, and the
+ * lengths are serialized as 4-byte little endian integers and zero-padded to
+ * the next 16-byte block boundary.
+ * The second beta key is the first one interpreted as an element in GF(2^128)
+ * and multiplied by x.
+ *
+ * Note that because the nonce and AAD may, in general, be variable-length, the
+ * key generation must be done by a pseudo-random function (PRF) on
+ * variable-length inputs.  CBC-MAC does not satisfy this, as it is only a PRF
+ * on fixed-length inputs.  CMAC remedies this flaw.  Including the lengths of
+ * the nonce, AAD, and message is also critical to avoid collisions.
+ *
+ * That being said, this implementation does not yet operate as an AEAD and
+ * therefore there is never any AAD, nor are variable-length nonces supported.
+ */
+static int generate_betas(struct ablkcipher_request *req,
+			  be128 *beta1_key, be128 *beta2_key)
+{
+	struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
+	struct heh_tfm_ctx *ctx = crypto_ablkcipher_ctx(tfm);
+	struct heh_req_ctx *rctx = heh_req_ctx(req);
+	struct heh_cmac_data *data = &rctx->u.cmac.data;
+	struct shash_desc *desc = &rctx->u.cmac.desc;
+	int err;
+
+	BUILD_BUG_ON(sizeof(*data) != 2 * HEH_BLOCK_SIZE);
+	memcpy(data->nonce, req->info, HEH_BLOCK_SIZE);
+	data->nonce_length = cpu_to_le32(HEH_BLOCK_SIZE);
+	data->aad_length = cpu_to_le32(0);
+	data->message_length = cpu_to_le32(req->nbytes);
+	data->padding = cpu_to_le32(0);
+
+	desc->tfm = ctx->cmac;
+	desc->flags = req->base.flags;
+
+	err = crypto_shash_digest(desc, (const u8 *)data, sizeof(*data),
+				  (u8 *)beta1_key);
+	if (err)
+		return err;
+
+	gf128mul_x_ble(beta2_key, beta1_key);
+	return 0;
+}
+
+/*****************************************************************************/
+
+/*
+ * This is the generic version of poly_hash.  It does the GF(2^128)
+ * multiplication by 'tau_key' using a precomputed table, without using any
+ * special CPU instructions.  On some platforms, an accelerated version (with
+ * higher cra_priority) may be used instead.
+ */
+
+struct poly_hash_tfm_ctx {
+	struct gf128mul_4k *tau_key;
+};
+
+struct poly_hash_desc_ctx {
+	be128 digest;
+	unsigned int count;
+};
+
+static int poly_hash_setkey(struct crypto_shash *tfm,
+			    const u8 *key, unsigned int keylen)
+{
+	struct poly_hash_tfm_ctx *tctx = crypto_shash_ctx(tfm);
+	be128 key128;
+
+	if (keylen != HEH_BLOCK_SIZE) {
+		crypto_shash_set_flags(tfm, CRYPTO_TFM_RES_BAD_KEY_LEN);
+		return -EINVAL;
+	}
+
+	if (tctx->tau_key)
+		gf128mul_free_4k(tctx->tau_key);
+	memcpy(&key128, key, HEH_BLOCK_SIZE);
+	tctx->tau_key = gf128mul_init_4k_ble(&key128);
+	if (!tctx->tau_key)
+		return -ENOMEM;
+	return 0;
+}
+
+static int poly_hash_init(struct shash_desc *desc)
+{
+	struct poly_hash_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	ctx->digest = (be128) { 0 };
+	ctx->count = 0;
+	return 0;
+}
+
+static int poly_hash_update(struct shash_desc *desc, const u8 *src,
+			    unsigned int len)
+{
+	struct poly_hash_tfm_ctx *tctx = crypto_shash_ctx(desc->tfm);
+	struct poly_hash_desc_ctx *ctx = shash_desc_ctx(desc);
+	unsigned int partial = ctx->count % HEH_BLOCK_SIZE;
+	u8 *dst = (u8 *)&ctx->digest + partial;
+
+	ctx->count += len;
+
+	/* Finishing at least one block? */
+	if (partial + len >= HEH_BLOCK_SIZE) {
+
+		if (partial) {
+			/* Finish the pending block. */
+			unsigned int n = HEH_BLOCK_SIZE - partial;
+
+			len -= n;
+			do {
+				*dst++ ^= *src++;
+			} while (--n);
+
+			gf128mul_4k_ble(&ctx->digest, tctx->tau_key);
+		}
+
+		/* Process zero or more full blocks. */
+		while (len >= HEH_BLOCK_SIZE) {
+			be128 coeff;
+
+			memcpy(&coeff, src, HEH_BLOCK_SIZE);
+			be128_xor(&ctx->digest, &ctx->digest, &coeff);
+			src += HEH_BLOCK_SIZE;
+			len -= HEH_BLOCK_SIZE;
+			gf128mul_4k_ble(&ctx->digest, tctx->tau_key);
+		}
+		dst = (u8 *)&ctx->digest;
+	}
+
+	/* Continue adding the next block to 'digest'. */
+	while (len--)
+		*dst++ ^= *src++;
+	return 0;
+}
+
+static int poly_hash_final(struct shash_desc *desc, u8 *out)
+{
+	struct poly_hash_desc_ctx *ctx = shash_desc_ctx(desc);
+
+	/* Finish the last block if needed. */
+	if (ctx->count % HEH_BLOCK_SIZE) {
+		struct poly_hash_tfm_ctx *tctx = crypto_shash_ctx(desc->tfm);
+
+		gf128mul_4k_ble(&ctx->digest, tctx->tau_key);
+	}
+
+	memcpy(out, &ctx->digest, HEH_BLOCK_SIZE);
+	return 0;
+}
+
+static void poly_hash_exit(struct crypto_tfm *tfm)
+{
+	struct poly_hash_tfm_ctx *tctx = crypto_tfm_ctx(tfm);
+
+	gf128mul_free_4k(tctx->tau_key);
+}
+
+static struct shash_alg poly_hash_alg = {
+	.digestsize	= HEH_BLOCK_SIZE,
+	.init		= poly_hash_init,
+	.update		= poly_hash_update,
+	.final		= poly_hash_final,
+	.setkey		= poly_hash_setkey,
+	.descsize	= sizeof(struct poly_hash_desc_ctx),
+	.base		= {
+		.cra_name		= "poly_hash",
+		.cra_driver_name	= "poly_hash-generic",
+		.cra_priority		= 100,
+		.cra_ctxsize		= sizeof(struct poly_hash_tfm_ctx),
+		.cra_exit		= poly_hash_exit,
+		.cra_module		= THIS_MODULE,
+	},
+};
+
+/*****************************************************************************/
+
+/*
+ * Split the message into 16 byte blocks, padding out the last block, and use
+ * the blocks as coefficients in the evaluation of a polynomial over GF(2^128)
+ * at the secret point 'tau_key'. For ease of implementing the higher-level
+ * heh_hash_inv() function, the constant and degree-1 coefficients are swapped
+ * if there is a partial block.
+ *
+ * Mathematically, compute:
+ *   if (no partial block)
+ *     k^{N-1} * m_0 + ... + k * m_{N-2} + m_{N-1}
+ *   else if (partial block)
+ *     k^N * m_0 + ... + k^2 * m_{N-2} + k * m_N + m_{N-1}
+ *
+ * where:
+ *	t is tau_key
+ *	N is the number of full blocks in the message
+ *	m_i is the i-th full block in the message for i = 0 to N-1 inclusive
+ *	m_N is the partial block of the message zero-padded up to 16 bytes
+ *
+ * Note that most of this is now separated out into its own keyed hash
+ * algorithm, to allow optimized implementations.  However, we still handle the
+ * swapping of the last two coefficients here in the HEH template because this
+ * simplifies the poly_hash algorithms: they don't have to buffer an extra
+ * block, don't have to duplicate as much code, and are more similar to GHASH.
+ */
+static int poly_hash(struct ablkcipher_request *req, struct scatterlist *sgl,
+		     be128 *hash)
+{
+	struct heh_req_ctx *rctx = heh_req_ctx(req);
+	struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
+	struct heh_tfm_ctx *ctx = crypto_ablkcipher_ctx(tfm);
+	struct shash_desc *desc = &rctx->u.poly_hash.desc;
+	unsigned int tail_offset = get_tail_offset(req->nbytes);
+	unsigned int tail_len = req->nbytes - tail_offset;
+	be128 tail[2];
+	unsigned int i, n;
+	struct sg_mapping_iter miter;
+	int err;
+
+	desc->tfm = ctx->poly_hash;
+	desc->flags = req->base.flags;
+
+	/* Handle all full blocks except the last */
+	err = crypto_shash_init(desc);
+	sg_miter_start(&miter, sgl, sg_nents(sgl),
+		       SG_MITER_FROM_SG | SG_MITER_ATOMIC);
+	for (i = 0; i < tail_offset && !err; i += n) {
+		sg_miter_next(&miter);
+		n = min_t(unsigned int, miter.length, tail_offset - i);
+		err = crypto_shash_update(desc, miter.addr, n);
+	}
+	sg_miter_stop(&miter);
+	if (err)
+		return err;
+
+	/* Handle the last full block and the partial block */
+	scatterwalk_map_and_copy(tail, sgl, tail_offset, tail_len, 0);
+
+	if (tail_len != HEH_BLOCK_SIZE) {
+		/* handle the partial block */
+		memset((u8 *)tail + tail_len, 0, sizeof(tail) - tail_len);
+		err = crypto_shash_update(desc, (u8 *)&tail[1], HEH_BLOCK_SIZE);
+		if (err)
+			return err;
+	}
+	err = crypto_shash_final(desc, (u8 *)hash);
+	if (err)
+		return err;
+	be128_xor(hash, hash, &tail[0]);
+	return 0;
+}
+
+/*
+ * Transform all full blocks except the last.
+ * This is used by both the hash and inverse hash phases.
+ */
+static int heh_tfm_blocks(struct ablkcipher_request *req,
+			  struct scatterlist *src_sgl,
+			  struct scatterlist *dst_sgl, unsigned int len,
+			  const be128 *hash, const be128 *beta_key)
+{
+	struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
+	struct blkcipher_desc desc = { .flags = req->base.flags };
+	struct blkcipher_walk walk;
+	be128 e = *beta_key;
+	int err;
+	unsigned int nbytes;
+
+	blkcipher_walk_init(&walk, dst_sgl, src_sgl, len);
+
+	err = blkcipher_ablkcipher_walk_virt(&desc, &walk, tfm);
+
+	while ((nbytes = walk.nbytes)) {
+		const be128 *src = (be128 *)walk.src.virt.addr;
+		be128 *dst = (be128 *)walk.dst.virt.addr;
+
+		do {
+			gf128mul_x_ble(&e, &e);
+			be128_xor(dst, src, hash);
+			be128_xor(dst, dst, &e);
+			src++;
+			dst++;
+		} while ((nbytes -= HEH_BLOCK_SIZE) >= HEH_BLOCK_SIZE);
+		err = blkcipher_walk_done(&desc, &walk, nbytes);
+	}
+	return err;
+}
+
+/*
+ * The hash phase of HEH.  Given a message, compute:
+ *
+ *     (m_0 + H, ..., m_{N-2} + H, H, m_N) + (xb, x^2b, ..., x^{N-1}b, b, 0)
+ *
+ * where:
+ *	N is the number of full blocks in the message
+ *	m_i is the i-th full block in the message for i = 0 to N-1 inclusive
+ *	m_N is the unpadded partial block, possibly empty
+ *	H is the poly_hash() of the message, keyed by tau_key
+ *	b is beta_key
+ *	x is the element x in our representation of GF(2^128)
+ *
+ * Note that the partial block remains unchanged, but it does affect the result
+ * of poly_hash() and therefore the transformation of all the full blocks.
+ */
+static int heh_hash(struct ablkcipher_request *req, const be128 *beta_key)
+{
+	be128 hash;
+	unsigned int tail_offset = get_tail_offset(req->nbytes);
+	unsigned int partial_len = req->nbytes % HEH_BLOCK_SIZE;
+	int err;
+
+	/* poly_hash() the full message including the partial block */
+	err = poly_hash(req, req->src, &hash);
+	if (err)
+		return err;
+
+	/* Transform all full blocks except the last */
+	err = heh_tfm_blocks(req, req->src, req->dst, tail_offset, &hash,
+			     beta_key);
+	if (err)
+		return err;
+
+	/* Set the last full block to hash XOR beta_key */
+	be128_xor(&hash, &hash, beta_key);
+	scatterwalk_map_and_copy(&hash, req->dst, tail_offset, HEH_BLOCK_SIZE,
+				 1);
+
+	/* Copy the partial block if needed */
+	if (partial_len != 0 && req->src != req->dst) {
+		unsigned int offs = tail_offset + HEH_BLOCK_SIZE;
+
+		scatterwalk_map_and_copy(&hash, req->src, offs, partial_len, 0);
+		scatterwalk_map_and_copy(&hash, req->dst, offs, partial_len, 1);
+	}
+	return 0;
+}
+
+/*
+ * The inverse hash phase of HEH.  This undoes the result of heh_hash().
+ */
+static int heh_hash_inv(struct ablkcipher_request *req, const be128 *beta_key)
+{
+	be128 hash;
+	be128 tmp;
+	struct scatterlist tmp_sgl[2];
+	struct scatterlist *tail_sgl;
+	unsigned int tail_offset = get_tail_offset(req->nbytes);
+	struct scatterlist *sgl = req->dst;
+	int err;
+
+	/*
+	 * The last full block was computed as hash XOR beta_key, so XOR it with
+	 * beta_key to recover hash.
+	 */
+	tail_sgl = scatterwalk_ffwd(tmp_sgl, sgl, tail_offset);
+	scatterwalk_map_and_copy(&hash, tail_sgl, 0, HEH_BLOCK_SIZE, 0);
+	be128_xor(&hash, &hash, beta_key);
+
+	/* Transform all full blocks except the last */
+	err = heh_tfm_blocks(req, sgl, sgl, tail_offset, &hash, beta_key);
+	if (err)
+		return err;
+
+	/*
+	 * Recover the last full block.  We know 'hash', i.e. the poly_hash() of
+	 * the the original message.  The last full block was the constant term
+	 * of the polynomial.  To recover the last full block, temporarily zero
+	 * it, compute the poly_hash(), and take the difference from 'hash'.
+	 */
+	memset(&tmp, 0, sizeof(tmp));
+	scatterwalk_map_and_copy(&tmp, tail_sgl, 0, HEH_BLOCK_SIZE, 1);
+	err = poly_hash(req, sgl, &tmp);
+	if (err)
+		return err;
+	be128_xor(&tmp, &tmp, &hash);
+	scatterwalk_map_and_copy(&tmp, tail_sgl, 0, HEH_BLOCK_SIZE, 1);
+	return 0;
+}
+
+static int heh_hash_inv_step(struct ablkcipher_request *req, u32 flags)
+{
+	struct heh_req_ctx *rctx = heh_req_ctx(req);
+
+	return heh_hash_inv(req, &rctx->beta2_key);
+}
+
+static int heh_ecb_step_3(struct ablkcipher_request *req, u32 flags)
+{
+	struct heh_req_ctx *rctx = heh_req_ctx(req);
+	u8 partial_block[HEH_BLOCK_SIZE] __aligned(__alignof__(u32));
+	unsigned int tail_offset = get_tail_offset(req->nbytes);
+	unsigned int partial_offset = tail_offset + HEH_BLOCK_SIZE;
+	unsigned int partial_len = req->nbytes - partial_offset;
+
+	/*
+	 * Extract the pad in req->dst at tail_offset, and xor the partial block
+	 * with it to create encrypted partial block
+	 */
+	scatterwalk_map_and_copy(rctx->u.ecb.keystream, req->dst, tail_offset,
+				 HEH_BLOCK_SIZE, 0);
+	scatterwalk_map_and_copy(partial_block, req->dst, partial_offset,
+				 partial_len, 0);
+	crypto_xor(partial_block, rctx->u.ecb.keystream, partial_len);
+
+	/*
+	 * Store the encrypted final block and partial block back in dst_sg
+	 */
+	scatterwalk_map_and_copy(&rctx->u.ecb.tmp, req->dst, tail_offset,
+				 HEH_BLOCK_SIZE, 1);
+	scatterwalk_map_and_copy(partial_block, req->dst, partial_offset,
+				 partial_len, 1);
+
+	return heh_hash_inv_step(req, flags);
+}
+
+static void heh_ecb_step_2_done(struct crypto_async_request *areq, int err)
+{
+	return async_done(areq, err, heh_ecb_step_3);
+}
+
+static int heh_ecb_step_2(struct ablkcipher_request *req, u32 flags)
+{
+	struct heh_req_ctx *rctx = heh_req_ctx(req);
+	unsigned int partial_len = req->nbytes % HEH_BLOCK_SIZE;
+	struct scatterlist *tmp_sgl;
+	int err;
+	unsigned int tail_offset = get_tail_offset(req->nbytes);
+
+	if (partial_len == 0)
+		return heh_hash_inv_step(req, flags);
+
+	/*
+	 * Extract the final full block, store it in tmp, and then xor that with
+	 * the value saved in u.ecb.keystream
+	 */
+	scatterwalk_map_and_copy(rctx->u.ecb.tmp, req->dst, tail_offset,
+				 HEH_BLOCK_SIZE, 0);
+	crypto_xor(rctx->u.ecb.keystream, rctx->u.ecb.tmp, HEH_BLOCK_SIZE);
+
+	/*
+	 * Encrypt the value in rctx->u.ecb.keystream to create the pad for the
+	 * partial block.
+	 * We cannot encrypt stack buffers, so re-use the dst_sg to do this
+	 * encryption to avoid a malloc. The value at tail_offset is stored in
+	 * tmp, and will be restored later.
+	 */
+	scatterwalk_map_and_copy(rctx->u.ecb.keystream, req->dst, tail_offset,
+				 HEH_BLOCK_SIZE, 1);
+	tmp_sgl = scatterwalk_ffwd(rctx->u.ecb.tmp_sgl, req->dst, tail_offset);
+	ablkcipher_request_set_callback(&rctx->u.ecb.req, flags,
+					heh_ecb_step_2_done, req);
+	ablkcipher_request_set_crypt(&rctx->u.ecb.req, tmp_sgl, tmp_sgl,
+				     HEH_BLOCK_SIZE, NULL);
+	err = crypto_ablkcipher_encrypt(&rctx->u.ecb.req);
+	if (err)
+		return err;
+	return heh_ecb_step_3(req, flags);
+}
+
+static void heh_ecb_full_done(struct crypto_async_request *areq, int err)
+{
+	return async_done(areq, err, heh_ecb_step_2);
+}
+
+/*
+ * The encrypt phase of HEH.  This uses ECB encryption, with special handling
+ * for the partial block at the end if any.  The source data is already in
+ * req->dst, so the encryption happens in-place.
+ *
+ * After the encrypt phase we continue on to the inverse hash phase.  The
+ * functions calls are chained to support asynchronous ECB algorithms.
+ */
+static int heh_ecb(struct ablkcipher_request *req, bool decrypt)
+{
+	struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req);
+	struct heh_tfm_ctx *ctx = crypto_ablkcipher_ctx(tfm);
+	struct heh_req_ctx *rctx = heh_req_ctx(req);
+	struct ablkcipher_request *ecb_req = &rctx->u.ecb.req;
+	unsigned int tail_offset = get_tail_offset(req->nbytes);
+	unsigned int full_len = tail_offset + HEH_BLOCK_SIZE;
+	int err;
+
+	/*
+	 * Save the last full block before it is encrypted/decrypted. This will
+	 * be used later to encrypt/decrypt the partial block
+	 */
+	scatterwalk_map_and_copy(rctx->u.ecb.keystream, req->dst, tail_offset,
+				 HEH_BLOCK_SIZE, 0);
+
+	/* Encrypt/decrypt all full blocks */
+	ablkcipher_request_set_tfm(ecb_req, ctx->ecb);
+	ablkcipher_request_set_callback(ecb_req, req->base.flags,
+				      heh_ecb_full_done, req);
+	ablkcipher_request_set_crypt(ecb_req, req->dst, req->dst, full_len,
+				     NULL);
+	if (decrypt)
+		err = crypto_ablkcipher_decrypt(ecb_req);
+	else
+		err = crypto_ablkcipher_encrypt(ecb_req);
+	if (err)
+		return err;
+
+	return heh_ecb_step_2(req, req->base.flags);
+}
+
+static int heh_crypt(struct ablkcipher_request *req, bool decrypt)
+{
+	struct heh_req_ctx *rctx = heh_req_ctx(req);
+	int err;
+
+	/* Inputs must be at least one full block */
+	if (req->nbytes < HEH_BLOCK_SIZE)
+		return -EINVAL;
+
+	err = generate_betas(req, &rctx->beta1_key, &rctx->beta2_key);
+	if (err)
+		return err;
+
+	if (decrypt)
+		swap(rctx->beta1_key, rctx->beta2_key);
+
+	err = heh_hash(req, &rctx->beta1_key);
+	if (err)
+		return err;
+
+	return heh_ecb(req, decrypt);
+}
+
+static int heh_encrypt(struct ablkcipher_request *req)
+{
+	return heh_crypt(req, false);
+}
+
+static int heh_decrypt(struct ablkcipher_request *req)
+{
+	return heh_crypt(req, true);
+}
+
+static int heh_setkey(struct crypto_ablkcipher *parent, const u8 *key,
+		      unsigned int keylen)
+{
+	struct heh_tfm_ctx *ctx = crypto_ablkcipher_ctx(parent);
+	struct crypto_shash *cmac = ctx->cmac;
+	struct crypto_ablkcipher *ecb = ctx->ecb;
+	SHASH_DESC_ON_STACK(desc, cmac);
+	u8 *derived_keys;
+	u8 digest[HEH_BLOCK_SIZE];
+	unsigned int i;
+	int err;
+
+	/* set prf_key = key */
+	crypto_shash_clear_flags(cmac, CRYPTO_TFM_REQ_MASK);
+	crypto_shash_set_flags(cmac, crypto_ablkcipher_get_flags(parent) &
+				     CRYPTO_TFM_REQ_MASK);
+	err = crypto_shash_setkey(cmac, key, keylen);
+	crypto_ablkcipher_set_flags(parent, crypto_shash_get_flags(cmac) &
+					    CRYPTO_TFM_RES_MASK);
+	if (err)
+		return err;
+
+	/*
+	 * Generate tau_key and ecb_key as follows:
+	 * tau_key = cmac(prf_key, 0x00...01)
+	 * ecb_key = cmac(prf_key, 0x00...02) || cmac(prf_key, 0x00...03) || ...
+	 *           truncated to keylen bytes
+	 */
+	derived_keys = kzalloc(round_up(HEH_BLOCK_SIZE + keylen,
+					HEH_BLOCK_SIZE), GFP_KERNEL);
+	if (!derived_keys)
+		return -ENOMEM;
+	desc->tfm = cmac;
+	desc->flags = (crypto_shash_get_flags(cmac) & CRYPTO_TFM_REQ_MASK);
+	for (i = 0; i < keylen + HEH_BLOCK_SIZE; i += HEH_BLOCK_SIZE) {
+		derived_keys[i + HEH_BLOCK_SIZE - 1] =
+					0x01 + i / HEH_BLOCK_SIZE;
+		err = crypto_shash_digest(desc, derived_keys + i,
+					  HEH_BLOCK_SIZE, digest);
+		if (err)
+			goto out;
+		memcpy(derived_keys + i, digest, HEH_BLOCK_SIZE);
+	}
+
+	err = crypto_shash_setkey(ctx->poly_hash, derived_keys, HEH_BLOCK_SIZE);
+	if (err)
+		goto out;
+
+	crypto_ablkcipher_clear_flags(ecb, CRYPTO_TFM_REQ_MASK);
+	crypto_ablkcipher_set_flags(ecb, crypto_ablkcipher_get_flags(parent) &
+					 CRYPTO_TFM_REQ_MASK);
+	err = crypto_ablkcipher_setkey(ecb, derived_keys + HEH_BLOCK_SIZE,
+				       keylen);
+	crypto_ablkcipher_set_flags(parent, crypto_ablkcipher_get_flags(ecb) &
+					    CRYPTO_TFM_RES_MASK);
+out:
+	kzfree(derived_keys);
+	return err;
+}
+
+static int heh_init_tfm(struct crypto_tfm *tfm)
+{
+	struct crypto_instance *inst = crypto_tfm_alg_instance(tfm);
+	struct heh_instance_ctx *ictx = crypto_instance_ctx(inst);
+	struct heh_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+	struct crypto_shash *cmac;
+	struct crypto_shash *poly_hash;
+	struct crypto_ablkcipher *ecb;
+	unsigned int reqsize;
+	int err;
+
+	cmac = crypto_spawn_shash(&ictx->cmac);
+	if (IS_ERR(cmac))
+		return PTR_ERR(cmac);
+
+	poly_hash = crypto_spawn_shash(&ictx->poly_hash);
+	err = PTR_ERR(poly_hash);
+	if (IS_ERR(poly_hash))
+		goto err_free_cmac;
+
+	ecb = crypto_spawn_skcipher(&ictx->ecb);
+	err = PTR_ERR(ecb);
+	if (IS_ERR(ecb))
+		goto err_free_poly_hash;
+
+	ctx->cmac = cmac;
+	ctx->poly_hash = poly_hash;
+	ctx->ecb = ecb;
+
+	reqsize = crypto_tfm_alg_alignmask(tfm) &
+		  ~(crypto_tfm_ctx_alignment() - 1);
+	reqsize += max3(offsetof(struct heh_req_ctx, u.cmac.desc) +
+			  sizeof(struct shash_desc) +
+			  crypto_shash_descsize(cmac),
+			offsetof(struct heh_req_ctx, u.poly_hash.desc) +
+			  sizeof(struct shash_desc) +
+			  crypto_shash_descsize(poly_hash),
+			offsetof(struct heh_req_ctx, u.ecb.req) +
+			  sizeof(struct ablkcipher_request) +
+			  crypto_ablkcipher_reqsize(ecb));
+	tfm->crt_ablkcipher.reqsize = reqsize;
+
+	return 0;
+
+err_free_poly_hash:
+	crypto_free_shash(poly_hash);
+err_free_cmac:
+	crypto_free_shash(cmac);
+	return err;
+}
+
+static void heh_exit_tfm(struct crypto_tfm *tfm)
+{
+	struct heh_tfm_ctx *ctx = crypto_tfm_ctx(tfm);
+
+	crypto_free_shash(ctx->cmac);
+	crypto_free_shash(ctx->poly_hash);
+	crypto_free_ablkcipher(ctx->ecb);
+}
+
+static void heh_free_instance(struct crypto_instance *inst)
+{
+	struct heh_instance_ctx *ctx = crypto_instance_ctx(inst);
+
+	crypto_drop_shash(&ctx->cmac);
+	crypto_drop_shash(&ctx->poly_hash);
+	crypto_drop_skcipher(&ctx->ecb);
+	kfree(inst);
+}
+
+/*
+ * Create an instance of HEH as a ablkcipher.
+ *
+ * This relies on underlying CMAC and ECB algorithms, usually cmac(aes) and
+ * ecb(aes).  For performance reasons we support asynchronous ECB algorithms.
+ * However, we do not yet support asynchronous CMAC algorithms because CMAC is
+ * only used on a small fixed amount of data per request, independent of the
+ * request length.  This would change if AEAD or variable-length nonce support
+ * were to be exposed.
+ */
+static int heh_create_common(struct crypto_template *tmpl, struct rtattr **tb,
+			     const char *full_name, const char *cmac_name,
+			     const char *poly_hash_name, const char *ecb_name)
+{
+	struct crypto_attr_type *algt;
+	struct crypto_instance *inst;
+	struct heh_instance_ctx *ctx;
+	struct shash_alg *cmac;
+	struct shash_alg *poly_hash;
+	struct crypto_alg *ecb;
+	int err;
+
+	algt = crypto_get_attr_type(tb);
+	if (IS_ERR(algt))
+		return PTR_ERR(algt);
+
+	/* User must be asking for something compatible with ablkcipher */
+	if ((algt->type ^ CRYPTO_ALG_TYPE_ABLKCIPHER) & algt->mask)
+		return -EINVAL;
+
+	/* Allocate the ablkcipher instance */
+	inst = kzalloc(sizeof(*inst) + sizeof(*ctx), GFP_KERNEL);
+	if (!inst)
+		return -ENOMEM;
+
+	ctx = crypto_instance_ctx(inst);
+
+	/* Set up the cmac spawn */
+	ctx->cmac.base.inst = inst;
+	err = crypto_grab_shash(&ctx->cmac, cmac_name, 0, 0);
+	if (err)
+		goto err_free_inst;
+	cmac = crypto_spawn_shash_alg(&ctx->cmac);
+	err = -EINVAL;
+	if (cmac->digestsize != HEH_BLOCK_SIZE)
+		goto err_drop_cmac;
+
+	/* Set up the poly_hash spawn */
+	ctx->poly_hash.base.inst = inst;
+	err = crypto_grab_shash(&ctx->poly_hash, poly_hash_name, 0, 0);
+	if (err)
+		goto err_drop_cmac;
+	poly_hash = crypto_spawn_shash_alg(&ctx->poly_hash);
+	err = -EINVAL;
+	if (poly_hash->digestsize != HEH_BLOCK_SIZE)
+		goto err_drop_poly_hash;
+
+	/* Set up the ecb spawn */
+	ctx->ecb.base.inst = inst;
+	err = crypto_grab_skcipher(&ctx->ecb, ecb_name, 0,
+				   crypto_requires_sync(algt->type,
+							algt->mask));
+	if (err)
+		goto err_drop_poly_hash;
+	ecb = crypto_skcipher_spawn_alg(&ctx->ecb);
+
+	/* HEH only supports block ciphers with 16 byte block size */
+	err = -EINVAL;
+	if (ecb->cra_blocksize != HEH_BLOCK_SIZE)
+		goto err_drop_ecb;
+
+	/* The underlying "ECB" algorithm must not require an IV */
+	err = -EINVAL;
+	if ((ecb->cra_flags & CRYPTO_ALG_TYPE_MASK) == CRYPTO_ALG_TYPE_BLKCIPHER) {
+		if (ecb->cra_blkcipher.ivsize != 0)
+			goto err_drop_ecb;
+	} else {
+		if (ecb->cra_ablkcipher.ivsize != 0)
+			goto err_drop_ecb;
+	}
+
+	/* Set the instance names */
+	err = -ENAMETOOLONG;
+	if (snprintf(inst->alg.cra_driver_name, CRYPTO_MAX_ALG_NAME,
+		     "heh_base(%s,%s,%s)", cmac->base.cra_driver_name,
+		     poly_hash->base.cra_driver_name,
+		     ecb->cra_driver_name) >= CRYPTO_MAX_ALG_NAME)
+		goto err_drop_ecb;
+
+	err = -ENAMETOOLONG;
+	if (snprintf(inst->alg.cra_name, CRYPTO_MAX_ALG_NAME,
+		     "%s", full_name) >= CRYPTO_MAX_ALG_NAME)
+		goto err_drop_ecb;
+
+	/* Finish initializing the instance */
+
+	inst->alg.cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER |
+				(ecb->cra_flags & CRYPTO_ALG_ASYNC);
+	inst->alg.cra_blocksize = HEH_BLOCK_SIZE;
+	inst->alg.cra_ctxsize = sizeof(struct heh_tfm_ctx);
+	inst->alg.cra_alignmask = ecb->cra_alignmask | (__alignof__(be128) - 1);
+	inst->alg.cra_priority = ecb->cra_priority;
+	inst->alg.cra_type = &crypto_ablkcipher_type;
+	inst->alg.cra_init = heh_init_tfm;
+	inst->alg.cra_exit = heh_exit_tfm;
+
+	inst->alg.cra_ablkcipher.setkey = heh_setkey;
+	inst->alg.cra_ablkcipher.encrypt = heh_encrypt;
+	inst->alg.cra_ablkcipher.decrypt = heh_decrypt;
+	if ((ecb->cra_flags & CRYPTO_ALG_TYPE_MASK) == CRYPTO_ALG_TYPE_BLKCIPHER) {
+		inst->alg.cra_ablkcipher.min_keysize = ecb->cra_blkcipher.min_keysize;
+		inst->alg.cra_ablkcipher.max_keysize = ecb->cra_blkcipher.max_keysize;
+	} else {
+		inst->alg.cra_ablkcipher.min_keysize = ecb->cra_ablkcipher.min_keysize;
+		inst->alg.cra_ablkcipher.max_keysize = ecb->cra_ablkcipher.max_keysize;
+	}
+	inst->alg.cra_ablkcipher.ivsize = HEH_BLOCK_SIZE;
+
+	/* Register the instance */
+	err = crypto_register_instance(tmpl, inst);
+	if (err)
+		goto err_drop_ecb;
+	return 0;
+
+err_drop_ecb:
+	crypto_drop_skcipher(&ctx->ecb);
+err_drop_poly_hash:
+	crypto_drop_shash(&ctx->poly_hash);
+err_drop_cmac:
+	crypto_drop_shash(&ctx->cmac);
+err_free_inst:
+	kfree(inst);
+	return err;
+}
+
+static int heh_create(struct crypto_template *tmpl, struct rtattr **tb)
+{
+	const char *cipher_name;
+	char full_name[CRYPTO_MAX_ALG_NAME];
+	char cmac_name[CRYPTO_MAX_ALG_NAME];
+	char ecb_name[CRYPTO_MAX_ALG_NAME];
+
+	/* Get the name of the requested block cipher (e.g. aes) */
+	cipher_name = crypto_attr_alg_name(tb[1]);
+	if (IS_ERR(cipher_name))
+		return PTR_ERR(cipher_name);
+
+	if (snprintf(full_name, CRYPTO_MAX_ALG_NAME, "heh(%s)", cipher_name) >=
+	    CRYPTO_MAX_ALG_NAME)
+		return -ENAMETOOLONG;
+
+	if (snprintf(cmac_name, CRYPTO_MAX_ALG_NAME, "cmac(%s)", cipher_name) >=
+	    CRYPTO_MAX_ALG_NAME)
+		return -ENAMETOOLONG;
+
+	if (snprintf(ecb_name, CRYPTO_MAX_ALG_NAME, "ecb(%s)", cipher_name) >=
+	    CRYPTO_MAX_ALG_NAME)
+		return -ENAMETOOLONG;
+
+	return heh_create_common(tmpl, tb, full_name, cmac_name, "poly_hash",
+				 ecb_name);
+}
+
+static struct crypto_template heh_tmpl = {
+	.name = "heh",
+	.create = heh_create,
+	.free = heh_free_instance,
+	.module = THIS_MODULE,
+};
+
+static int heh_base_create(struct crypto_template *tmpl, struct rtattr **tb)
+{
+	char full_name[CRYPTO_MAX_ALG_NAME];
+	const char *cmac_name;
+	const char *poly_hash_name;
+	const char *ecb_name;
+
+	cmac_name = crypto_attr_alg_name(tb[1]);
+	if (IS_ERR(cmac_name))
+		return PTR_ERR(cmac_name);
+
+	poly_hash_name = crypto_attr_alg_name(tb[2]);
+	if (IS_ERR(poly_hash_name))
+		return PTR_ERR(poly_hash_name);
+
+	ecb_name = crypto_attr_alg_name(tb[3]);
+	if (IS_ERR(ecb_name))
+		return PTR_ERR(ecb_name);
+
+	if (snprintf(full_name, CRYPTO_MAX_ALG_NAME, "heh_base(%s,%s,%s)",
+		     cmac_name, poly_hash_name, ecb_name) >=
+	    CRYPTO_MAX_ALG_NAME)
+		return -ENAMETOOLONG;
+
+	return heh_create_common(tmpl, tb, full_name, cmac_name, poly_hash_name,
+				 ecb_name);
+}
+
+/*
+ * If HEH is instantiated as "heh_base" instead of "heh", then specific
+ * implementations of cmac, poly_hash, and ecb can be specified instead of just
+ * the cipher.
+ */
+static struct crypto_template heh_base_tmpl = {
+	.name = "heh_base",
+	.create = heh_base_create,
+	.free = heh_free_instance,
+	.module = THIS_MODULE,
+};
+
+static int __init heh_module_init(void)
+{
+	int err;
+
+	err = crypto_register_template(&heh_tmpl);
+	if (err)
+		return err;
+
+	err = crypto_register_template(&heh_base_tmpl);
+	if (err)
+		goto out_undo_heh;
+
+	err = crypto_register_shash(&poly_hash_alg);
+	if (err)
+		goto out_undo_heh_base;
+
+	return 0;
+
+out_undo_heh_base:
+	crypto_unregister_template(&heh_base_tmpl);
+out_undo_heh:
+	crypto_unregister_template(&heh_tmpl);
+	return err;
+}
+
+static void __exit heh_module_exit(void)
+{
+	crypto_unregister_template(&heh_tmpl);
+	crypto_unregister_template(&heh_base_tmpl);
+	crypto_unregister_shash(&poly_hash_alg);
+}
+
+module_init(heh_module_init);
+module_exit(heh_module_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("Hash-Encrypt-Hash block cipher mode");
+MODULE_ALIAS_CRYPTO("heh");
+MODULE_ALIAS_CRYPTO("heh_base");
diff --git a/crypto/shash.c b/crypto/shash.c
index b2cd109d9171..eba42e28af4c 100644
--- a/crypto/shash.c
+++ b/crypto/shash.c
@@ -685,6 +685,14 @@ void shash_free_instance(struct crypto_instance *inst)
 }
 EXPORT_SYMBOL_GPL(shash_free_instance);
 
+int crypto_grab_shash(struct crypto_shash_spawn *spawn,
+		      const char *name, u32 type, u32 mask)
+{
+	spawn->base.frontend = &crypto_shash_type;
+	return crypto_grab_spawn(&spawn->base, name, type, mask);
+}
+EXPORT_SYMBOL_GPL(crypto_grab_shash);
+
 int crypto_init_shash_spawn(struct crypto_shash_spawn *spawn,
 			    struct shash_alg *alg,
 			    struct crypto_instance *inst)
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index 5f15f45fcc9f..6d4da8fd24fd 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -3215,6 +3215,21 @@ static const struct alg_test_desc alg_test_descs[] = {
 			}
 		}
 	}, {
+		.alg = "heh(aes)",
+		.test = alg_test_skcipher,
+		.suite = {
+			.cipher = {
+				.enc = {
+					.vecs = aes_heh_enc_tv_template,
+					.count = AES_HEH_ENC_TEST_VECTORS
+				},
+				.dec = {
+					.vecs = aes_heh_dec_tv_template,
+					.count = AES_HEH_DEC_TEST_VECTORS
+				}
+			}
+		}
+	}, {
 		.alg = "hmac(crc32)",
 		.test = alg_test_hash,
 		.suite = {
diff --git a/crypto/testmgr.h b/crypto/testmgr.h
index 0e02c60a57b6..ba6530d8ba58 100644
--- a/crypto/testmgr.h
+++ b/crypto/testmgr.h
@@ -14139,6 +14139,8 @@ static struct cipher_testvec cast6_xts_dec_tv_template[] = {
 #define AES_DEC_TEST_VECTORS 4
 #define AES_CBC_ENC_TEST_VECTORS 5
 #define AES_CBC_DEC_TEST_VECTORS 5
+#define AES_HEH_ENC_TEST_VECTORS 4
+#define AES_HEH_DEC_TEST_VECTORS 4
 #define HMAC_MD5_ECB_CIPHER_NULL_ENC_TEST_VECTORS 2
 #define HMAC_MD5_ECB_CIPHER_NULL_DEC_TEST_VECTORS 2
 #define HMAC_SHA1_ECB_CIPHER_NULL_ENC_TEST_VEC 2
@@ -14511,6 +14513,198 @@ static struct cipher_testvec aes_dec_tv_template[] = {
 	},
 };
 
+static struct cipher_testvec aes_heh_enc_tv_template[] = {
+	{
+		.key    = "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.klen   = 16,
+		.iv	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00",
+		.input	= "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.ilen   = 16,
+		.result = "\xd8\xbd\x40\xbf\xca\xe5\xee\x81"
+			  "\x0f\x3d\x1f\x1f\xae\x89\x07\x55",
+		.rlen   = 16,
+		.also_non_np = 1,
+		.np	= 2,
+		.tap	= { 8, 8 },
+	}, {
+		.key    = "\xa8\xda\x24\x9b\x5e\xfa\x13\xc2"
+			  "\xc1\x94\xbf\x32\xba\x38\xa3\x77",
+		.klen   = 16,
+		.iv	= "\x4d\x47\x61\x37\x2b\x47\x86\xf0"
+			  "\xd6\x47\xb5\xc2\xe8\xcf\x85\x27",
+		.input	= "\xb8\xee\x29\xe4\xa5\xd1\xe7\x55"
+			  "\xd0\xfd\xe7\x22\x63\x76\x36\xe2"
+			  "\xf8\x0c\xf8\xfe\x65\x76\xe7\xca"
+			  "\xc1\x42\xf5\xca\x5a\xa8\xac\x2a",
+		.ilen   = 32,
+		.result = "\x59\xf2\x78\x4e\x10\x94\xf9\x5c"
+			  "\x22\x23\x78\x2a\x30\x48\x11\x97"
+			  "\xb1\xfe\x70\xc4\xef\xdf\x04\xef"
+			  "\x16\x39\x04\xcf\xc0\x95\x9a\x98",
+		.rlen   = 32,
+		.also_non_np = 1,
+		.np	= 3,
+		.tap	= { 16, 13, 3 },
+	}, {
+		.key    = "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.klen   = 16,
+		.iv	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00",
+		.input	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00",
+		.ilen   = 63,
+		.result = "\xe0\x40\xeb\xe9\x52\xbe\x65\x60"
+			  "\xe4\x68\x68\xa3\x73\x75\xb8\x52"
+			  "\xef\x38\x6a\x87\x25\x25\xf6\x04"
+			  "\xe5\x8e\xbe\x14\x8b\x02\x14\x1f"
+			  "\xa9\x73\xb7\xad\x15\xbe\x9c\xa0"
+			  "\xd2\x8a\x2c\xdc\xd4\xe3\x05\x55"
+			  "\x0a\xf5\xf8\x51\xee\xe5\x62\xa5"
+			  "\x71\xa7\x7c\x15\x5d\x7a\x9e",
+		.rlen   = 63,
+		.also_non_np = 1,
+		.np	= 8,
+		.tap	= { 20, 20, 10, 8, 2, 1, 1, 1 },
+	}, {
+		.key    = "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F"
+			  "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F"
+			  "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.klen   = 16,
+		.iv	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00",
+		.input	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x01"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00",
+		.ilen   = 63,
+		.result = "\x4b\x1a\x15\xa0\xaf\x08\x6d\x70"
+			  "\xf0\xa7\x97\xb5\x31\x4b\x8c\xc3"
+			  "\x4d\xf2\x7a\x9d\xdd\xd4\x15\x99"
+			  "\x57\xad\xc6\xb1\x35\x69\xf5\x6a"
+			  "\x2d\x70\xe4\x97\x49\xb2\x9f\x71"
+			  "\xde\x22\xb5\x70\x8c\x69\x24\xd3"
+			  "\xad\x80\x58\x48\x90\xe4\xed\xba"
+			  "\x76\x3d\x71\x7c\x57\x25\x87",
+		.rlen   = 63,
+		.also_non_np = 1,
+		.np	= 8,
+		.tap	= { 20, 20, 10, 8, 2, 1, 1, 1 },
+	}
+};
+
+static struct cipher_testvec aes_heh_dec_tv_template[] = {
+	{
+		.key    = "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.klen   = 16,
+		.iv	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00",
+		.input = "\xd8\xbd\x40\xbf\xca\xe5\xee\x81"
+			  "\x0f\x3d\x1f\x1f\xae\x89\x07\x55",
+		.ilen   = 16,
+		.result	= "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.rlen   = 16,
+		.also_non_np = 1,
+		.np	= 2,
+		.tap	= { 8, 8 },
+	}, {
+		.key    = "\xa8\xda\x24\x9b\x5e\xfa\x13\xc2"
+			  "\xc1\x94\xbf\x32\xba\x38\xa3\x77",
+		.klen   = 16,
+		.iv	= "\x4d\x47\x61\x37\x2b\x47\x86\xf0"
+			  "\xd6\x47\xb5\xc2\xe8\xcf\x85\x27",
+		.input = "\x59\xf2\x78\x4e\x10\x94\xf9\x5c"
+			  "\x22\x23\x78\x2a\x30\x48\x11\x97"
+			  "\xb1\xfe\x70\xc4\xef\xdf\x04\xef"
+			  "\x16\x39\x04\xcf\xc0\x95\x9a\x98",
+		.ilen   = 32,
+		.result	= "\xb8\xee\x29\xe4\xa5\xd1\xe7\x55"
+			  "\xd0\xfd\xe7\x22\x63\x76\x36\xe2"
+			  "\xf8\x0c\xf8\xfe\x65\x76\xe7\xca"
+			  "\xc1\x42\xf5\xca\x5a\xa8\xac\x2a",
+		.rlen   = 32,
+		.also_non_np = 1,
+		.np	= 3,
+		.tap	= { 16, 13, 3 },
+	}, {
+		.key    = "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.klen   = 16,
+		.iv	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00",
+		.input = "\xe0\x40\xeb\xe9\x52\xbe\x65\x60"
+			  "\xe4\x68\x68\xa3\x73\x75\xb8\x52"
+			  "\xef\x38\x6a\x87\x25\x25\xf6\x04"
+			  "\xe5\x8e\xbe\x14\x8b\x02\x14\x1f"
+			  "\xa9\x73\xb7\xad\x15\xbe\x9c\xa0"
+			  "\xd2\x8a\x2c\xdc\xd4\xe3\x05\x55"
+			  "\x0a\xf5\xf8\x51\xee\xe5\x62\xa5"
+			  "\x71\xa7\x7c\x15\x5d\x7a\x9e",
+		.ilen   = 63,
+		.result	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00",
+		.rlen   = 63,
+		.also_non_np = 1,
+		.np	= 8,
+		.tap	= { 20, 20, 10, 8, 2, 1, 1, 1 },
+	}, {
+		.key    = "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F"
+			  "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F"
+			  "\x00\x01\x02\x03\x04\x05\x06\x07"
+			  "\x08\x09\x0A\x0B\x0C\x0D\x0E\x0F",
+		.klen   = 16,
+		.iv	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00",
+		.input = "\x4b\x1a\x15\xa0\xaf\x08\x6d\x70"
+			  "\xf0\xa7\x97\xb5\x31\x4b\x8c\xc3"
+			  "\x4d\xf2\x7a\x9d\xdd\xd4\x15\x99"
+			  "\x57\xad\xc6\xb1\x35\x69\xf5\x6a"
+			  "\x2d\x70\xe4\x97\x49\xb2\x9f\x71"
+			  "\xde\x22\xb5\x70\x8c\x69\x24\xd3"
+			  "\xad\x80\x58\x48\x90\xe4\xed\xba"
+			  "\x76\x3d\x71\x7c\x57\x25\x87",
+		.ilen   = 63,
+		.result	= "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00\x01"
+			  "\x00\x00\x00\x00\x00\x00\x00\x00"
+			  "\x00\x00\x00\x00\x00\x00\x00",
+		.rlen   = 63,
+		.also_non_np = 1,
+		.np	= 8,
+		.tap	= { 20, 20, 10, 8, 2, 1, 1, 1 },
+	}
+};
+
 static struct cipher_testvec aes_cbc_enc_tv_template[] = {
 	{ /* From RFC 3602 */
 		.key    = "\x06\xa9\x21\x40\x36\xb8\xa1\x5b"
diff --git a/drivers/android/Kconfig b/drivers/android/Kconfig
index bdfc6c6f4f5a..01de42c8b74b 100644
--- a/drivers/android/Kconfig
+++ b/drivers/android/Kconfig
@@ -19,6 +19,18 @@ config ANDROID_BINDER_IPC
 	  Android process, using Binder to identify, invoke and pass arguments
 	  between said processes.
 
+config ANDROID_BINDER_DEVICES
+	string "Android Binder devices"
+	depends on ANDROID_BINDER_IPC
+	default "binder,hwbinder,vndbinder"
+	---help---
+	  Default value for the binder.devices parameter.
+
+	  The binder.devices parameter is a comma-separated list of strings
+	  that specifies the names of the binder device nodes that will be
+	  created. Each binder device has its own context manager, and is
+	  therefore logically separated from the other devices.
+
 config ANDROID_BINDER_IPC_32BIT
 	bool
 	depends on !64BIT && ANDROID_BINDER_IPC
@@ -32,6 +44,16 @@ config ANDROID_BINDER_IPC_32BIT
 
 	  Note that enabling this will break newer Android user-space.
 
+config ANDROID_BINDER_IPC_SELFTEST
+	bool "Android Binder IPC Driver Selftest"
+	depends on ANDROID_BINDER_IPC
+	---help---
+	  This feature allows binder selftest to run.
+
+	  Binder selftest checks the allocation and free of binder buffers
+	  exhaustively with combinations of various buffer sizes and
+	  alignments.
+
 endif # if ANDROID
 
 endmenu
diff --git a/drivers/android/Makefile b/drivers/android/Makefile
index 3b7e4b072c58..a01254c43ee3 100644
--- a/drivers/android/Makefile
+++ b/drivers/android/Makefile
@@ -1,3 +1,4 @@
 ccflags-y += -I$(src)			# needed for trace events
 
-obj-$(CONFIG_ANDROID_BINDER_IPC)	+= binder.o
+obj-$(CONFIG_ANDROID_BINDER_IPC)	+= binder.o binder_alloc.o
+obj-$(CONFIG_ANDROID_BINDER_IPC_SELFTEST) += binder_alloc_selftest.o
diff --git a/drivers/android/binder.c b/drivers/android/binder.c
index 5531f020e561..008e448536de 100644
--- a/drivers/android/binder.c
+++ b/drivers/android/binder.c
@@ -15,6 +15,40 @@
  *
  */
 
+/*
+ * Locking overview
+ *
+ * There are 3 main spinlocks which must be acquired in the
+ * order shown:
+ *
+ * 1) proc->outer_lock : protects binder_ref
+ *    binder_proc_lock() and binder_proc_unlock() are
+ *    used to acq/rel.
+ * 2) node->lock : protects most fields of binder_node.
+ *    binder_node_lock() and binder_node_unlock() are
+ *    used to acq/rel
+ * 3) proc->inner_lock : protects the thread and node lists
+ *    (proc->threads, proc->waiting_threads, proc->nodes)
+ *    and all todo lists associated with the binder_proc
+ *    (proc->todo, thread->todo, proc->delivered_death and
+ *    node->async_todo), as well as thread->transaction_stack
+ *    binder_inner_proc_lock() and binder_inner_proc_unlock()
+ *    are used to acq/rel
+ *
+ * Any lock under procA must never be nested under any lock at the same
+ * level or below on procB.
+ *
+ * Functions that require a lock held on entry indicate which lock
+ * in the suffix of the function name:
+ *
+ * foo_olocked() : requires node->outer_lock
+ * foo_nlocked() : requires node->lock
+ * foo_ilocked() : requires proc->inner_lock
+ * foo_oilocked(): requires proc->outer_lock and proc->inner_lock
+ * foo_nilocked(): requires node->lock and proc->inner_lock
+ * ...
+ */
+
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <asm/cacheflush.h>
@@ -24,7 +58,6 @@
 #include <linux/fs.h>
 #include <linux/list.h>
 #include <linux/miscdevice.h>
-#include <linux/mm.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
 #include <linux/nsproxy.h>
@@ -34,31 +67,31 @@
 #include <linux/sched.h>
 #include <linux/seq_file.h>
 #include <linux/uaccess.h>
-#include <linux/vmalloc.h>
-#include <linux/slab.h>
 #include <linux/pid_namespace.h>
 #include <linux/security.h>
+#include <linux/spinlock.h>
 
 #ifdef CONFIG_ANDROID_BINDER_IPC_32BIT
 #define BINDER_IPC_32BIT 1
 #endif
 
 #include <uapi/linux/android/binder.h>
+#include "binder_alloc.h"
 #include "binder_trace.h"
 
-static DEFINE_MUTEX(binder_main_lock);
+static HLIST_HEAD(binder_deferred_list);
 static DEFINE_MUTEX(binder_deferred_lock);
-static DEFINE_MUTEX(binder_mmap_lock);
 
+static HLIST_HEAD(binder_devices);
 static HLIST_HEAD(binder_procs);
-static HLIST_HEAD(binder_deferred_list);
+static DEFINE_MUTEX(binder_procs_lock);
+
 static HLIST_HEAD(binder_dead_nodes);
+static DEFINE_SPINLOCK(binder_dead_nodes_lock);
 
 static struct dentry *binder_debugfs_dir_entry_root;
 static struct dentry *binder_debugfs_dir_entry_proc;
-static struct binder_node *binder_context_mgr_node;
-static kuid_t binder_context_mgr_uid = INVALID_UID;
-static int binder_last_id;
+static atomic_t binder_last_id;
 static struct workqueue_struct *binder_deferred_workqueue;
 
 #define BINDER_DEBUG_ENTRY(name) \
@@ -105,16 +138,15 @@ enum {
 	BINDER_DEBUG_TRANSACTION_COMPLETE   = 1U << 10,
 	BINDER_DEBUG_FREE_BUFFER            = 1U << 11,
 	BINDER_DEBUG_INTERNAL_REFS          = 1U << 12,
-	BINDER_DEBUG_BUFFER_ALLOC           = 1U << 13,
-	BINDER_DEBUG_PRIORITY_CAP           = 1U << 14,
-	BINDER_DEBUG_BUFFER_ALLOC_ASYNC     = 1U << 15,
+	BINDER_DEBUG_PRIORITY_CAP           = 1U << 13,
+	BINDER_DEBUG_SPINLOCKS              = 1U << 14,
 };
 static uint32_t binder_debug_mask = BINDER_DEBUG_USER_ERROR |
 	BINDER_DEBUG_FAILED_TRANSACTION | BINDER_DEBUG_DEAD_TRANSACTION;
 module_param_named(debug_mask, binder_debug_mask, uint, S_IWUSR | S_IRUGO);
 
-static bool binder_debug_no_lock;
-module_param_named(proc_no_lock, binder_debug_no_lock, bool, S_IWUSR | S_IRUGO);
+static char *binder_devices_param = CONFIG_ANDROID_BINDER_DEVICES;
+module_param_named(devices, binder_devices_param, charp, S_IRUGO);
 
 static DECLARE_WAIT_QUEUE_HEAD(binder_user_error_wait);
 static int binder_stop_on_user_error;
@@ -146,6 +178,17 @@ module_param_call(stop_on_user_error, binder_set_stop_on_user_error,
 			binder_stop_on_user_error = 2; \
 	} while (0)
 
+#define to_flat_binder_object(hdr) \
+	container_of(hdr, struct flat_binder_object, hdr)
+
+#define to_binder_fd_object(hdr) container_of(hdr, struct binder_fd_object, hdr)
+
+#define to_binder_buffer_object(hdr) \
+	container_of(hdr, struct binder_buffer_object, hdr)
+
+#define to_binder_fd_array_object(hdr) \
+	container_of(hdr, struct binder_fd_array_object, hdr)
+
 enum binder_stat_types {
 	BINDER_STAT_PROC,
 	BINDER_STAT_THREAD,
@@ -158,26 +201,27 @@ enum binder_stat_types {
 };
 
 struct binder_stats {
-	int br[_IOC_NR(BR_FAILED_REPLY) + 1];
-	int bc[_IOC_NR(BC_DEAD_BINDER_DONE) + 1];
-	int obj_created[BINDER_STAT_COUNT];
-	int obj_deleted[BINDER_STAT_COUNT];
+	atomic_t br[_IOC_NR(BR_FAILED_REPLY) + 1];
+	atomic_t bc[_IOC_NR(BC_REPLY_SG) + 1];
+	atomic_t obj_created[BINDER_STAT_COUNT];
+	atomic_t obj_deleted[BINDER_STAT_COUNT];
 };
 
 static struct binder_stats binder_stats;
 
 static inline void binder_stats_deleted(enum binder_stat_types type)
 {
-	binder_stats.obj_deleted[type]++;
+	atomic_inc(&binder_stats.obj_deleted[type]);
 }
 
 static inline void binder_stats_created(enum binder_stat_types type)
 {
-	binder_stats.obj_created[type]++;
+	atomic_inc(&binder_stats.obj_created[type]);
 }
 
 struct binder_transaction_log_entry {
 	int debug_id;
+	int debug_id_done;
 	int call_type;
 	int from_proc;
 	int from_thread;
@@ -187,10 +231,14 @@ struct binder_transaction_log_entry {
 	int to_node;
 	int data_size;
 	int offsets_size;
+	int return_error_line;
+	uint32_t return_error;
+	uint32_t return_error_param;
+	const char *context_name;
 };
 struct binder_transaction_log {
-	int next;
-	int full;
+	atomic_t cur;
+	bool full;
 	struct binder_transaction_log_entry entry[32];
 };
 static struct binder_transaction_log binder_transaction_log;
@@ -200,22 +248,50 @@ static struct binder_transaction_log_entry *binder_transaction_log_add(
 	struct binder_transaction_log *log)
 {
 	struct binder_transaction_log_entry *e;
+	unsigned int cur = atomic_inc_return(&log->cur);
 
-	e = &log->entry[log->next];
-	memset(e, 0, sizeof(*e));
-	log->next++;
-	if (log->next == ARRAY_SIZE(log->entry)) {
-		log->next = 0;
+	if (cur >= ARRAY_SIZE(log->entry))
 		log->full = 1;
-	}
+	e = &log->entry[cur % ARRAY_SIZE(log->entry)];
+	WRITE_ONCE(e->debug_id_done, 0);
+	/*
+	 * write-barrier to synchronize access to e->debug_id_done.
+	 * We make sure the initialized 0 value is seen before
+	 * memset() other fields are zeroed by memset.
+	 */
+	smp_wmb();
+	memset(e, 0, sizeof(*e));
 	return e;
 }
 
+struct binder_context {
+	struct binder_node *binder_context_mgr_node;
+	struct mutex context_mgr_node_lock;
+
+	kuid_t binder_context_mgr_uid;
+	const char *name;
+};
+
+struct binder_device {
+	struct hlist_node hlist;
+	struct miscdevice miscdev;
+	struct binder_context context;
+};
+
+/**
+ * struct binder_work - work enqueued on a worklist
+ * @entry:             node enqueued on list
+ * @type:              type of work to be performed
+ *
+ * There are separate work lists for proc, thread, and node (async).
+ */
 struct binder_work {
 	struct list_head entry;
+
 	enum {
 		BINDER_WORK_TRANSACTION = 1,
 		BINDER_WORK_TRANSACTION_COMPLETE,
+		BINDER_WORK_RETURN_ERROR,
 		BINDER_WORK_NODE,
 		BINDER_WORK_DEAD_BINDER,
 		BINDER_WORK_DEAD_BINDER_AND_CLEAR,
@@ -223,8 +299,76 @@ struct binder_work {
 	} type;
 };
 
+struct binder_error {
+	struct binder_work work;
+	uint32_t cmd;
+};
+
+/**
+ * struct binder_node - binder node bookkeeping
+ * @debug_id:             unique ID for debugging
+ *                        (invariant after initialized)
+ * @lock:                 lock for node fields
+ * @work:                 worklist element for node work
+ *                        (protected by @proc->inner_lock)
+ * @rb_node:              element for proc->nodes tree
+ *                        (protected by @proc->inner_lock)
+ * @dead_node:            element for binder_dead_nodes list
+ *                        (protected by binder_dead_nodes_lock)
+ * @proc:                 binder_proc that owns this node
+ *                        (invariant after initialized)
+ * @refs:                 list of references on this node
+ *                        (protected by @lock)
+ * @internal_strong_refs: used to take strong references when
+ *                        initiating a transaction
+ *                        (protected by @proc->inner_lock if @proc
+ *                        and by @lock)
+ * @local_weak_refs:      weak user refs from local process
+ *                        (protected by @proc->inner_lock if @proc
+ *                        and by @lock)
+ * @local_strong_refs:    strong user refs from local process
+ *                        (protected by @proc->inner_lock if @proc
+ *                        and by @lock)
+ * @tmp_refs:             temporary kernel refs
+ *                        (protected by @proc->inner_lock while @proc
+ *                        is valid, and by binder_dead_nodes_lock
+ *                        if @proc is NULL. During inc/dec and node release
+ *                        it is also protected by @lock to provide safety
+ *                        as the node dies and @proc becomes NULL)
+ * @ptr:                  userspace pointer for node
+ *                        (invariant, no lock needed)
+ * @cookie:               userspace cookie for node
+ *                        (invariant, no lock needed)
+ * @has_strong_ref:       userspace notified of strong ref
+ *                        (protected by @proc->inner_lock if @proc
+ *                        and by @lock)
+ * @pending_strong_ref:   userspace has acked notification of strong ref
+ *                        (protected by @proc->inner_lock if @proc
+ *                        and by @lock)
+ * @has_weak_ref:         userspace notified of weak ref
+ *                        (protected by @proc->inner_lock if @proc
+ *                        and by @lock)
+ * @pending_weak_ref:     userspace has acked notification of weak ref
+ *                        (protected by @proc->inner_lock if @proc
+ *                        and by @lock)
+ * @has_async_transaction: async transaction to node in progress
+ *                        (protected by @lock)
+ * @sched_policy:         minimum scheduling policy for node
+ *                        (invariant after initialized)
+ * @accept_fds:           file descriptor operations supported for node
+ *                        (invariant after initialized)
+ * @min_priority:         minimum scheduling priority
+ *                        (invariant after initialized)
+ * @inherit_rt:           inherit RT scheduling policy from caller
+ *                        (invariant after initialized)
+ * @async_todo:           list of async work items
+ *                        (protected by @proc->inner_lock)
+ *
+ * Bookkeeping structure for binder nodes.
+ */
 struct binder_node {
 	int debug_id;
+	spinlock_t lock;
 	struct binder_work work;
 	union {
 		struct rb_node rb_node;
@@ -235,87 +379,181 @@ struct binder_node {
 	int internal_strong_refs;
 	int local_weak_refs;
 	int local_strong_refs;
+	int tmp_refs;
 	binder_uintptr_t ptr;
 	binder_uintptr_t cookie;
-	unsigned has_strong_ref:1;
-	unsigned pending_strong_ref:1;
-	unsigned has_weak_ref:1;
-	unsigned pending_weak_ref:1;
-	unsigned has_async_transaction:1;
-	unsigned accept_fds:1;
-	unsigned min_priority:8;
+	struct {
+		/*
+		 * bitfield elements protected by
+		 * proc inner_lock
+		 */
+		u8 has_strong_ref:1;
+		u8 pending_strong_ref:1;
+		u8 has_weak_ref:1;
+		u8 pending_weak_ref:1;
+	};
+	struct {
+		/*
+		 * invariant after initialization
+		 */
+		u8 sched_policy:2;
+		u8 inherit_rt:1;
+		u8 accept_fds:1;
+		u8 min_priority;
+	};
+	bool has_async_transaction;
 	struct list_head async_todo;
 };
 
 struct binder_ref_death {
+	/**
+	 * @work: worklist element for death notifications
+	 *        (protected by inner_lock of the proc that
+	 *        this ref belongs to)
+	 */
 	struct binder_work work;
 	binder_uintptr_t cookie;
 };
 
+/**
+ * struct binder_ref_data - binder_ref counts and id
+ * @debug_id:        unique ID for the ref
+ * @desc:            unique userspace handle for ref
+ * @strong:          strong ref count (debugging only if not locked)
+ * @weak:            weak ref count (debugging only if not locked)
+ *
+ * Structure to hold ref count and ref id information. Since
+ * the actual ref can only be accessed with a lock, this structure
+ * is used to return information about the ref to callers of
+ * ref inc/dec functions.
+ */
+struct binder_ref_data {
+	int debug_id;
+	uint32_t desc;
+	int strong;
+	int weak;
+};
+
+/**
+ * struct binder_ref - struct to track references on nodes
+ * @data:        binder_ref_data containing id, handle, and current refcounts
+ * @rb_node_desc: node for lookup by @data.desc in proc's rb_tree
+ * @rb_node_node: node for lookup by @node in proc's rb_tree
+ * @node_entry:  list entry for node->refs list in target node
+ *               (protected by @node->lock)
+ * @proc:        binder_proc containing ref
+ * @node:        binder_node of target node. When cleaning up a
+ *               ref for deletion in binder_cleanup_ref, a non-NULL
+ *               @node indicates the node must be freed
+ * @death:       pointer to death notification (ref_death) if requested
+ *               (protected by @node->lock)
+ *
+ * Structure to track references from procA to target node (on procB). This
+ * structure is unsafe to access without holding @proc->outer_lock.
+ */
 struct binder_ref {
 	/* Lookups needed: */
 	/*   node + proc => ref (transaction) */
 	/*   desc + proc => ref (transaction, inc/dec ref) */
 	/*   node => refs + procs (proc exit) */
-	int debug_id;
+	struct binder_ref_data data;
 	struct rb_node rb_node_desc;
 	struct rb_node rb_node_node;
 	struct hlist_node node_entry;
 	struct binder_proc *proc;
 	struct binder_node *node;
-	uint32_t desc;
-	int strong;
-	int weak;
 	struct binder_ref_death *death;
 };
 
-struct binder_buffer {
-	struct list_head entry; /* free and allocated entries by address */
-	struct rb_node rb_node; /* free entry by size or allocated entry */
-				/* by address */
-	unsigned free:1;
-	unsigned allow_user_free:1;
-	unsigned async_transaction:1;
-	unsigned debug_id:29;
-
-	struct binder_transaction *transaction;
-
-	struct binder_node *target_node;
-	size_t data_size;
-	size_t offsets_size;
-	uint8_t data[0];
+enum binder_deferred_state {
+	BINDER_DEFERRED_FLUSH        = 0x01,
+	BINDER_DEFERRED_RELEASE      = 0x02,
 };
 
-enum binder_deferred_state {
-	BINDER_DEFERRED_PUT_FILES    = 0x01,
-	BINDER_DEFERRED_FLUSH        = 0x02,
-	BINDER_DEFERRED_RELEASE      = 0x04,
+/**
+ * struct binder_priority - scheduler policy and priority
+ * @sched_policy            scheduler policy
+ * @prio                    [100..139] for SCHED_NORMAL, [0..99] for FIFO/RT
+ *
+ * The binder driver supports inheriting the following scheduler policies:
+ * SCHED_NORMAL
+ * SCHED_BATCH
+ * SCHED_FIFO
+ * SCHED_RR
+ */
+struct binder_priority {
+	unsigned int sched_policy;
+	int prio;
 };
 
+/**
+ * struct binder_proc - binder process bookkeeping
+ * @proc_node:            element for binder_procs list
+ * @threads:              rbtree of binder_threads in this proc
+ *                        (protected by @inner_lock)
+ * @nodes:                rbtree of binder nodes associated with
+ *                        this proc ordered by node->ptr
+ *                        (protected by @inner_lock)
+ * @refs_by_desc:         rbtree of refs ordered by ref->desc
+ *                        (protected by @outer_lock)
+ * @refs_by_node:         rbtree of refs ordered by ref->node
+ *                        (protected by @outer_lock)
+ * @waiting_threads:      threads currently waiting for proc work
+ *                        (protected by @inner_lock)
+ * @pid                   PID of group_leader of process
+ *                        (invariant after initialized)
+ * @tsk                   task_struct for group_leader of process
+ *                        (invariant after initialized)
+ * @deferred_work_node:   element for binder_deferred_list
+ *                        (protected by binder_deferred_lock)
+ * @deferred_work:        bitmap of deferred work to perform
+ *                        (protected by binder_deferred_lock)
+ * @is_dead:              process is dead and awaiting free
+ *                        when outstanding transactions are cleaned up
+ *                        (protected by @inner_lock)
+ * @todo:                 list of work for this process
+ *                        (protected by @inner_lock)
+ * @wait:                 wait queue head to wait for proc work
+ *                        (invariant after initialized)
+ * @stats:                per-process binder statistics
+ *                        (atomics, no lock needed)
+ * @delivered_death:      list of delivered death notification
+ *                        (protected by @inner_lock)
+ * @max_threads:          cap on number of binder threads
+ *                        (protected by @inner_lock)
+ * @requested_threads:    number of binder threads requested but not
+ *                        yet started. In current implementation, can
+ *                        only be 0 or 1.
+ *                        (protected by @inner_lock)
+ * @requested_threads_started: number binder threads started
+ *                        (protected by @inner_lock)
+ * @tmp_ref:              temporary reference to indicate proc is in use
+ *                        (protected by @inner_lock)
+ * @default_priority:     default scheduler priority
+ *                        (invariant after initialized)
+ * @debugfs_entry:        debugfs node
+ * @alloc:                binder allocator bookkeeping
+ * @context:              binder_context for this proc
+ *                        (invariant after initialized)
+ * @inner_lock:           can nest under outer_lock and/or node lock
+ * @outer_lock:           no nesting under innor or node lock
+ *                        Lock order: 1) outer, 2) node, 3) inner
+ *
+ * Bookkeeping structure for binder processes
+ */
 struct binder_proc {
 	struct hlist_node proc_node;
 	struct rb_root threads;
 	struct rb_root nodes;
 	struct rb_root refs_by_desc;
 	struct rb_root refs_by_node;
+	struct list_head waiting_threads;
 	int pid;
-	struct vm_area_struct *vma;
-	struct mm_struct *vma_vm_mm;
 	struct task_struct *tsk;
-	struct files_struct *files;
 	struct hlist_node deferred_work_node;
 	int deferred_work;
-	void *buffer;
-	ptrdiff_t user_buffer_offset;
-
-	struct list_head buffers;
-	struct rb_root free_buffers;
-	struct rb_root allocated_buffers;
-	size_t free_async_space;
+	bool is_dead;
 
-	struct page **pages;
-	size_t buffer_size;
-	uint32_t buffer_free;
 	struct list_head todo;
 	wait_queue_head_t wait;
 	struct binder_stats stats;
@@ -323,9 +561,13 @@ struct binder_proc {
 	int max_threads;
 	int requested_threads;
 	int requested_threads_started;
-	int ready_threads;
-	long default_priority;
+	int tmp_ref;
+	struct binder_priority default_priority;
 	struct dentry *debugfs_entry;
+	struct binder_alloc alloc;
+	struct binder_context *context;
+	spinlock_t inner_lock;
+	spinlock_t outer_lock;
 };
 
 enum {
@@ -334,22 +576,63 @@ enum {
 	BINDER_LOOPER_STATE_EXITED      = 0x04,
 	BINDER_LOOPER_STATE_INVALID     = 0x08,
 	BINDER_LOOPER_STATE_WAITING     = 0x10,
-	BINDER_LOOPER_STATE_NEED_RETURN = 0x20
+	BINDER_LOOPER_STATE_POLL        = 0x20,
 };
 
+/**
+ * struct binder_thread - binder thread bookkeeping
+ * @proc:                 binder process for this thread
+ *                        (invariant after initialization)
+ * @rb_node:              element for proc->threads rbtree
+ *                        (protected by @proc->inner_lock)
+ * @waiting_thread_node:  element for @proc->waiting_threads list
+ *                        (protected by @proc->inner_lock)
+ * @pid:                  PID for this thread
+ *                        (invariant after initialization)
+ * @looper:               bitmap of looping state
+ *                        (only accessed by this thread)
+ * @looper_needs_return:  looping thread needs to exit driver
+ *                        (no lock needed)
+ * @transaction_stack:    stack of in-progress transactions for this thread
+ *                        (protected by @proc->inner_lock)
+ * @todo:                 list of work to do for this thread
+ *                        (protected by @proc->inner_lock)
+ * @process_todo:         whether work in @todo should be processed
+ *                        (protected by @proc->inner_lock)
+ * @return_error:         transaction errors reported by this thread
+ *                        (only accessed by this thread)
+ * @reply_error:          transaction errors reported by target thread
+ *                        (protected by @proc->inner_lock)
+ * @wait:                 wait queue for thread work
+ * @stats:                per-thread statistics
+ *                        (atomics, no lock needed)
+ * @tmp_ref:              temporary reference to indicate thread is in use
+ *                        (atomic since @proc->inner_lock cannot
+ *                        always be acquired)
+ * @is_dead:              thread is dead and awaiting free
+ *                        when outstanding transactions are cleaned up
+ *                        (protected by @proc->inner_lock)
+ * @task:                 struct task_struct for this thread
+ *
+ * Bookkeeping structure for binder threads.
+ */
 struct binder_thread {
 	struct binder_proc *proc;
 	struct rb_node rb_node;
+	struct list_head waiting_thread_node;
 	int pid;
-	int looper;
+	int looper;              /* only modified by this thread */
+	bool looper_need_return; /* can be written by other thread */
 	struct binder_transaction *transaction_stack;
 	struct list_head todo;
-	uint32_t return_error; /* Write failed, return error code in read buf */
-	uint32_t return_error2; /* Write failed, return error code in read */
-		/* buffer. Used when sending a reply to a dead process that */
-		/* we are also waiting on */
+	bool process_todo;
+	struct binder_error return_error;
+	struct binder_error reply_error;
 	wait_queue_head_t wait;
 	struct binder_stats stats;
+	atomic_t tmp_ref;
+	bool is_dead;
+	struct task_struct *task;
 };
 
 struct binder_transaction {
@@ -366,30 +649,330 @@ struct binder_transaction {
 	struct binder_buffer *buffer;
 	unsigned int	code;
 	unsigned int	flags;
-	long	priority;
-	long	saved_priority;
+	struct binder_priority	priority;
+	struct binder_priority	saved_priority;
+	bool    set_priority_called;
 	kuid_t	sender_euid;
+	/**
+	 * @lock:  protects @from, @to_proc, and @to_thread
+	 *
+	 * @from, @to_proc, and @to_thread can be set to NULL
+	 * during thread teardown
+	 */
+	spinlock_t lock;
 };
 
+/**
+ * binder_proc_lock() - Acquire outer lock for given binder_proc
+ * @proc:         struct binder_proc to acquire
+ *
+ * Acquires proc->outer_lock. Used to protect binder_ref
+ * structures associated with the given proc.
+ */
+#define binder_proc_lock(proc) _binder_proc_lock(proc, __LINE__)
+static void
+_binder_proc_lock(struct binder_proc *proc, int line)
+{
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	spin_lock(&proc->outer_lock);
+}
+
+/**
+ * binder_proc_unlock() - Release spinlock for given binder_proc
+ * @proc:         struct binder_proc to acquire
+ *
+ * Release lock acquired via binder_proc_lock()
+ */
+#define binder_proc_unlock(_proc) _binder_proc_unlock(_proc, __LINE__)
+static void
+_binder_proc_unlock(struct binder_proc *proc, int line)
+{
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	spin_unlock(&proc->outer_lock);
+}
+
+/**
+ * binder_inner_proc_lock() - Acquire inner lock for given binder_proc
+ * @proc:         struct binder_proc to acquire
+ *
+ * Acquires proc->inner_lock. Used to protect todo lists
+ */
+#define binder_inner_proc_lock(proc) _binder_inner_proc_lock(proc, __LINE__)
+static void
+_binder_inner_proc_lock(struct binder_proc *proc, int line)
+{
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	spin_lock(&proc->inner_lock);
+}
+
+/**
+ * binder_inner_proc_unlock() - Release inner lock for given binder_proc
+ * @proc:         struct binder_proc to acquire
+ *
+ * Release lock acquired via binder_inner_proc_lock()
+ */
+#define binder_inner_proc_unlock(proc) _binder_inner_proc_unlock(proc, __LINE__)
+static void
+_binder_inner_proc_unlock(struct binder_proc *proc, int line)
+{
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	spin_unlock(&proc->inner_lock);
+}
+
+/**
+ * binder_node_lock() - Acquire spinlock for given binder_node
+ * @node:         struct binder_node to acquire
+ *
+ * Acquires node->lock. Used to protect binder_node fields
+ */
+#define binder_node_lock(node) _binder_node_lock(node, __LINE__)
+static void
+_binder_node_lock(struct binder_node *node, int line)
+{
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	spin_lock(&node->lock);
+}
+
+/**
+ * binder_node_unlock() - Release spinlock for given binder_proc
+ * @node:         struct binder_node to acquire
+ *
+ * Release lock acquired via binder_node_lock()
+ */
+#define binder_node_unlock(node) _binder_node_unlock(node, __LINE__)
+static void
+_binder_node_unlock(struct binder_node *node, int line)
+{
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	spin_unlock(&node->lock);
+}
+
+/**
+ * binder_node_inner_lock() - Acquire node and inner locks
+ * @node:         struct binder_node to acquire
+ *
+ * Acquires node->lock. If node->proc also acquires
+ * proc->inner_lock. Used to protect binder_node fields
+ */
+#define binder_node_inner_lock(node) _binder_node_inner_lock(node, __LINE__)
+static void
+_binder_node_inner_lock(struct binder_node *node, int line)
+{
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	spin_lock(&node->lock);
+	if (node->proc)
+		binder_inner_proc_lock(node->proc);
+}
+
+/**
+ * binder_node_unlock() - Release node and inner locks
+ * @node:         struct binder_node to acquire
+ *
+ * Release lock acquired via binder_node_lock()
+ */
+#define binder_node_inner_unlock(node) _binder_node_inner_unlock(node, __LINE__)
+static void
+_binder_node_inner_unlock(struct binder_node *node, int line)
+{
+	struct binder_proc *proc = node->proc;
+
+	binder_debug(BINDER_DEBUG_SPINLOCKS,
+		     "%s: line=%d\n", __func__, line);
+	if (proc)
+		binder_inner_proc_unlock(proc);
+	spin_unlock(&node->lock);
+}
+
+static bool binder_worklist_empty_ilocked(struct list_head *list)
+{
+	return list_empty(list);
+}
+
+/**
+ * binder_worklist_empty() - Check if no items on the work list
+ * @proc:       binder_proc associated with list
+ * @list:	list to check
+ *
+ * Return: true if there are no items on list, else false
+ */
+static bool binder_worklist_empty(struct binder_proc *proc,
+				  struct list_head *list)
+{
+	bool ret;
+
+	binder_inner_proc_lock(proc);
+	ret = binder_worklist_empty_ilocked(list);
+	binder_inner_proc_unlock(proc);
+	return ret;
+}
+
+/**
+ * binder_enqueue_work_ilocked() - Add an item to the work list
+ * @work:         struct binder_work to add to list
+ * @target_list:  list to add work to
+ *
+ * Adds the work to the specified list. Asserts that work
+ * is not already on a list.
+ *
+ * Requires the proc->inner_lock to be held.
+ */
+static void
+binder_enqueue_work_ilocked(struct binder_work *work,
+			   struct list_head *target_list)
+{
+	BUG_ON(target_list == NULL);
+	BUG_ON(work->entry.next && !list_empty(&work->entry));
+	list_add_tail(&work->entry, target_list);
+}
+
+/**
+ * binder_enqueue_deferred_thread_work_ilocked() - Add deferred thread work
+ * @thread:       thread to queue work to
+ * @work:         struct binder_work to add to list
+ *
+ * Adds the work to the todo list of the thread. Doesn't set the process_todo
+ * flag, which means that (if it wasn't already set) the thread will go to
+ * sleep without handling this work when it calls read.
+ *
+ * Requires the proc->inner_lock to be held.
+ */
+static void
+binder_enqueue_deferred_thread_work_ilocked(struct binder_thread *thread,
+					    struct binder_work *work)
+{
+	binder_enqueue_work_ilocked(work, &thread->todo);
+}
+
+/**
+ * binder_enqueue_thread_work_ilocked() - Add an item to the thread work list
+ * @thread:       thread to queue work to
+ * @work:         struct binder_work to add to list
+ *
+ * Adds the work to the todo list of the thread, and enables processing
+ * of the todo queue.
+ *
+ * Requires the proc->inner_lock to be held.
+ */
+static void
+binder_enqueue_thread_work_ilocked(struct binder_thread *thread,
+				   struct binder_work *work)
+{
+	binder_enqueue_work_ilocked(work, &thread->todo);
+	thread->process_todo = true;
+}
+
+/**
+ * binder_enqueue_thread_work() - Add an item to the thread work list
+ * @thread:       thread to queue work to
+ * @work:         struct binder_work to add to list
+ *
+ * Adds the work to the todo list of the thread, and enables processing
+ * of the todo queue.
+ */
+static void
+binder_enqueue_thread_work(struct binder_thread *thread,
+			   struct binder_work *work)
+{
+	binder_inner_proc_lock(thread->proc);
+	binder_enqueue_thread_work_ilocked(thread, work);
+	binder_inner_proc_unlock(thread->proc);
+}
+
+static void
+binder_dequeue_work_ilocked(struct binder_work *work)
+{
+	list_del_init(&work->entry);
+}
+
+/**
+ * binder_dequeue_work() - Removes an item from the work list
+ * @proc:         binder_proc associated with list
+ * @work:         struct binder_work to remove from list
+ *
+ * Removes the specified work item from whatever list it is on.
+ * Can safely be called if work is not on any list.
+ */
+static void
+binder_dequeue_work(struct binder_proc *proc, struct binder_work *work)
+{
+	binder_inner_proc_lock(proc);
+	binder_dequeue_work_ilocked(work);
+	binder_inner_proc_unlock(proc);
+}
+
+static struct binder_work *binder_dequeue_work_head_ilocked(
+					struct list_head *list)
+{
+	struct binder_work *w;
+
+	w = list_first_entry_or_null(list, struct binder_work, entry);
+	if (w)
+		list_del_init(&w->entry);
+	return w;
+}
+
+/**
+ * binder_dequeue_work_head() - Dequeues the item at head of list
+ * @proc:         binder_proc associated with list
+ * @list:         list to dequeue head
+ *
+ * Removes the head of the list if there are items on the list
+ *
+ * Return: pointer dequeued binder_work, NULL if list was empty
+ */
+static struct binder_work *binder_dequeue_work_head(
+					struct binder_proc *proc,
+					struct list_head *list)
+{
+	struct binder_work *w;
+
+	binder_inner_proc_lock(proc);
+	w = binder_dequeue_work_head_ilocked(list);
+	binder_inner_proc_unlock(proc);
+	return w;
+}
+
 static void
 binder_defer_work(struct binder_proc *proc, enum binder_deferred_state defer);
+static void binder_free_thread(struct binder_thread *thread);
+static void binder_free_proc(struct binder_proc *proc);
+static void binder_inc_node_tmpref_ilocked(struct binder_node *node);
+
+struct files_struct *binder_get_files_struct(struct binder_proc *proc)
+{
+	return get_files_struct(proc->tsk);
+}
 
 static int task_get_unused_fd_flags(struct binder_proc *proc, int flags)
 {
-	struct files_struct *files = proc->files;
+	struct files_struct *files;
 	unsigned long rlim_cur;
 	unsigned long irqs;
+	int ret;
 
+	files = binder_get_files_struct(proc);
 	if (files == NULL)
 		return -ESRCH;
 
-	if (!lock_task_sighand(proc->tsk, &irqs))
-		return -EMFILE;
+	if (!lock_task_sighand(proc->tsk, &irqs)) {
+		ret = -EMFILE;
+		goto err;
+	}
 
 	rlim_cur = task_rlimit(proc->tsk, RLIMIT_NOFILE);
 	unlock_task_sighand(proc->tsk, &irqs);
 
-	return __alloc_fd(files, 0, rlim_cur, flags);
+	ret = __alloc_fd(files, 0, rlim_cur, flags);
+err:
+	put_files_struct(files);
+	return ret;
 }
 
 /*
@@ -398,8 +981,12 @@ static int task_get_unused_fd_flags(struct binder_proc *proc, int flags)
 static void task_fd_install(
 	struct binder_proc *proc, unsigned int fd, struct file *file)
 {
-	if (proc->files)
-		__fd_install(proc->files, fd, file);
+	struct files_struct *files = binder_get_files_struct(proc);
+
+	if (files) {
+		__fd_install(files, fd, file);
+		put_files_struct(files);
+	}
 }
 
 /*
@@ -407,469 +994,300 @@ static void task_fd_install(
  */
 static long task_close_fd(struct binder_proc *proc, unsigned int fd)
 {
+	struct files_struct *files = binder_get_files_struct(proc);
 	int retval;
 
-	if (proc->files == NULL)
+	if (files == NULL)
 		return -ESRCH;
 
-	retval = __close_fd(proc->files, fd);
+	retval = __close_fd(files, fd);
 	/* can't restart close syscall because file table entry was cleared */
 	if (unlikely(retval == -ERESTARTSYS ||
 		     retval == -ERESTARTNOINTR ||
 		     retval == -ERESTARTNOHAND ||
 		     retval == -ERESTART_RESTARTBLOCK))
 		retval = -EINTR;
+	put_files_struct(files);
 
 	return retval;
 }
 
-static inline void binder_lock(const char *tag)
+static bool binder_has_work_ilocked(struct binder_thread *thread,
+				    bool do_proc_work)
 {
-	trace_binder_lock(tag);
-	mutex_lock(&binder_main_lock);
-	trace_binder_locked(tag);
+	return thread->process_todo ||
+		thread->looper_need_return ||
+		(do_proc_work &&
+		 !binder_worklist_empty_ilocked(&thread->proc->todo));
 }
 
-static inline void binder_unlock(const char *tag)
+static bool binder_has_work(struct binder_thread *thread, bool do_proc_work)
 {
-	trace_binder_unlock(tag);
-	mutex_unlock(&binder_main_lock);
-}
+	bool has_work;
 
-static void binder_set_nice(long nice)
-{
-	long min_nice;
+	binder_inner_proc_lock(thread->proc);
+	has_work = binder_has_work_ilocked(thread, do_proc_work);
+	binder_inner_proc_unlock(thread->proc);
 
-	if (can_nice(current, nice)) {
-		set_user_nice(current, nice);
-		return;
-	}
-	min_nice = rlimit_to_nice(current->signal->rlim[RLIMIT_NICE].rlim_cur);
-	binder_debug(BINDER_DEBUG_PRIORITY_CAP,
-		     "%d: nice value %ld not allowed use %ld instead\n",
-		      current->pid, nice, min_nice);
-	set_user_nice(current, min_nice);
-	if (min_nice <= MAX_NICE)
-		return;
-	binder_user_error("%d RLIMIT_NICE not set\n", current->pid);
+	return has_work;
 }
 
-static size_t binder_buffer_size(struct binder_proc *proc,
-				 struct binder_buffer *buffer)
+static bool binder_available_for_proc_work_ilocked(struct binder_thread *thread)
 {
-	if (list_is_last(&buffer->entry, &proc->buffers))
-		return proc->buffer + proc->buffer_size - (void *)buffer->data;
-	return (size_t)list_entry(buffer->entry.next,
-			  struct binder_buffer, entry) - (size_t)buffer->data;
+	return !thread->transaction_stack &&
+		binder_worklist_empty_ilocked(&thread->todo) &&
+		(thread->looper & (BINDER_LOOPER_STATE_ENTERED |
+				   BINDER_LOOPER_STATE_REGISTERED));
 }
 
-static void binder_insert_free_buffer(struct binder_proc *proc,
-				      struct binder_buffer *new_buffer)
+static void binder_wakeup_poll_threads_ilocked(struct binder_proc *proc,
+					       bool sync)
 {
-	struct rb_node **p = &proc->free_buffers.rb_node;
-	struct rb_node *parent = NULL;
-	struct binder_buffer *buffer;
-	size_t buffer_size;
-	size_t new_buffer_size;
-
-	BUG_ON(!new_buffer->free);
-
-	new_buffer_size = binder_buffer_size(proc, new_buffer);
-
-	binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-		     "%d: add free buffer, size %zd, at %p\n",
-		      proc->pid, new_buffer_size, new_buffer);
-
-	while (*p) {
-		parent = *p;
-		buffer = rb_entry(parent, struct binder_buffer, rb_node);
-		BUG_ON(!buffer->free);
-
-		buffer_size = binder_buffer_size(proc, buffer);
+	struct rb_node *n;
+	struct binder_thread *thread;
 
-		if (new_buffer_size < buffer_size)
-			p = &parent->rb_left;
-		else
-			p = &parent->rb_right;
+	for (n = rb_first(&proc->threads); n != NULL; n = rb_next(n)) {
+		thread = rb_entry(n, struct binder_thread, rb_node);
+		if (thread->looper & BINDER_LOOPER_STATE_POLL &&
+		    binder_available_for_proc_work_ilocked(thread)) {
+			if (sync)
+				wake_up_interruptible_sync(&thread->wait);
+			else
+				wake_up_interruptible(&thread->wait);
+		}
 	}
-	rb_link_node(&new_buffer->rb_node, parent, p);
-	rb_insert_color(&new_buffer->rb_node, &proc->free_buffers);
 }
 
-static void binder_insert_allocated_buffer(struct binder_proc *proc,
-					   struct binder_buffer *new_buffer)
+/**
+ * binder_select_thread_ilocked() - selects a thread for doing proc work.
+ * @proc:	process to select a thread from
+ *
+ * Note that calling this function moves the thread off the waiting_threads
+ * list, so it can only be woken up by the caller of this function, or a
+ * signal. Therefore, callers *should* always wake up the thread this function
+ * returns.
+ *
+ * Return:	If there's a thread currently waiting for process work,
+ *		returns that thread. Otherwise returns NULL.
+ */
+static struct binder_thread *
+binder_select_thread_ilocked(struct binder_proc *proc)
 {
-	struct rb_node **p = &proc->allocated_buffers.rb_node;
-	struct rb_node *parent = NULL;
-	struct binder_buffer *buffer;
+	struct binder_thread *thread;
 
-	BUG_ON(new_buffer->free);
+	assert_spin_locked(&proc->inner_lock);
+	thread = list_first_entry_or_null(&proc->waiting_threads,
+					  struct binder_thread,
+					  waiting_thread_node);
 
-	while (*p) {
-		parent = *p;
-		buffer = rb_entry(parent, struct binder_buffer, rb_node);
-		BUG_ON(buffer->free);
+	if (thread)
+		list_del_init(&thread->waiting_thread_node);
 
-		if (new_buffer < buffer)
-			p = &parent->rb_left;
-		else if (new_buffer > buffer)
-			p = &parent->rb_right;
-		else
-			BUG();
-	}
-	rb_link_node(&new_buffer->rb_node, parent, p);
-	rb_insert_color(&new_buffer->rb_node, &proc->allocated_buffers);
+	return thread;
 }
 
-static struct binder_buffer *binder_buffer_lookup(struct binder_proc *proc,
-						  uintptr_t user_ptr)
+/**
+ * binder_wakeup_thread_ilocked() - wakes up a thread for doing proc work.
+ * @proc:	process to wake up a thread in
+ * @thread:	specific thread to wake-up (may be NULL)
+ * @sync:	whether to do a synchronous wake-up
+ *
+ * This function wakes up a thread in the @proc process.
+ * The caller may provide a specific thread to wake-up in
+ * the @thread parameter. If @thread is NULL, this function
+ * will wake up threads that have called poll().
+ *
+ * Note that for this function to work as expected, callers
+ * should first call binder_select_thread() to find a thread
+ * to handle the work (if they don't have a thread already),
+ * and pass the result into the @thread parameter.
+ */
+static void binder_wakeup_thread_ilocked(struct binder_proc *proc,
+					 struct binder_thread *thread,
+					 bool sync)
 {
-	struct rb_node *n = proc->allocated_buffers.rb_node;
-	struct binder_buffer *buffer;
-	struct binder_buffer *kern_ptr;
-
-	kern_ptr = (struct binder_buffer *)(user_ptr - proc->user_buffer_offset
-		- offsetof(struct binder_buffer, data));
-
-	while (n) {
-		buffer = rb_entry(n, struct binder_buffer, rb_node);
-		BUG_ON(buffer->free);
+	assert_spin_locked(&proc->inner_lock);
 
-		if (kern_ptr < buffer)
-			n = n->rb_left;
-		else if (kern_ptr > buffer)
-			n = n->rb_right;
+	if (thread) {
+		if (sync)
+			wake_up_interruptible_sync(&thread->wait);
 		else
-			return buffer;
+			wake_up_interruptible(&thread->wait);
+		return;
 	}
-	return NULL;
+
+	/* Didn't find a thread waiting for proc work; this can happen
+	 * in two scenarios:
+	 * 1. All threads are busy handling transactions
+	 *    In that case, one of those threads should call back into
+	 *    the kernel driver soon and pick up this work.
+	 * 2. Threads are using the (e)poll interface, in which case
+	 *    they may be blocked on the waitqueue without having been
+	 *    added to waiting_threads. For this case, we just iterate
+	 *    over all threads not handling transaction work, and
+	 *    wake them all up. We wake all because we don't know whether
+	 *    a thread that called into (e)poll is handling non-binder
+	 *    work currently.
+	 */
+	binder_wakeup_poll_threads_ilocked(proc, sync);
 }
 
-static int binder_update_page_range(struct binder_proc *proc, int allocate,
-				    void *start, void *end,
-				    struct vm_area_struct *vma)
+static void binder_wakeup_proc_ilocked(struct binder_proc *proc)
 {
-	void *page_addr;
-	unsigned long user_page_addr;
-	struct page **page;
-	struct mm_struct *mm;
-
-	binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-		     "%d: %s pages %p-%p\n", proc->pid,
-		     allocate ? "allocate" : "free", start, end);
+	struct binder_thread *thread = binder_select_thread_ilocked(proc);
 
-	if (end <= start)
-		return 0;
+	binder_wakeup_thread_ilocked(proc, thread, /* sync = */false);
+}
 
-	trace_binder_update_page_range(proc, allocate, start, end);
+static bool is_rt_policy(int policy)
+{
+	return policy == SCHED_FIFO || policy == SCHED_RR;
+}
 
-	if (vma)
-		mm = NULL;
-	else
-		mm = get_task_mm(proc->tsk);
-
-	if (mm) {
-		down_write(&mm->mmap_sem);
-		vma = proc->vma;
-		if (vma && mm != proc->vma_vm_mm) {
-			pr_err("%d: vma mm and task mm mismatch\n",
-				proc->pid);
-			vma = NULL;
-		}
-	}
+static bool is_fair_policy(int policy)
+{
+	return policy == SCHED_NORMAL || policy == SCHED_BATCH;
+}
 
-	if (allocate == 0)
-		goto free_range;
+static bool binder_supported_policy(int policy)
+{
+	return is_fair_policy(policy) || is_rt_policy(policy);
+}
 
-	if (vma == NULL) {
-		pr_err("%d: binder_alloc_buf failed to map pages in userspace, no vma\n",
-			proc->pid);
-		goto err_no_vma;
-	}
+static int to_userspace_prio(int policy, int kernel_priority)
+{
+	if (is_fair_policy(policy))
+		return PRIO_TO_NICE(kernel_priority);
+	else
+		return MAX_USER_RT_PRIO - 1 - kernel_priority;
+}
 
-	for (page_addr = start; page_addr < end; page_addr += PAGE_SIZE) {
-		int ret;
+static int to_kernel_prio(int policy, int user_priority)
+{
+	if (is_fair_policy(policy))
+		return NICE_TO_PRIO(user_priority);
+	else
+		return MAX_USER_RT_PRIO - 1 - user_priority;
+}
 
-		page = &proc->pages[(page_addr - proc->buffer) / PAGE_SIZE];
+static void binder_do_set_priority(struct task_struct *task,
+				   struct binder_priority desired,
+				   bool verify)
+{
+	int priority; /* user-space prio value */
+	bool has_cap_nice;
+	unsigned int policy = desired.sched_policy;
 
-		BUG_ON(*page);
-		*page = alloc_page(GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO);
-		if (*page == NULL) {
-			pr_err("%d: binder_alloc_buf failed for page at %p\n",
-				proc->pid, page_addr);
-			goto err_alloc_page_failed;
-		}
-		ret = map_kernel_range_noflush((unsigned long)page_addr,
-					PAGE_SIZE, PAGE_KERNEL, page);
-		flush_cache_vmap((unsigned long)page_addr,
-				(unsigned long)page_addr + PAGE_SIZE);
-		if (ret != 1) {
-			pr_err("%d: binder_alloc_buf failed to map page at %p in kernel\n",
-			       proc->pid, page_addr);
-			goto err_map_kernel_failed;
-		}
-		user_page_addr =
-			(uintptr_t)page_addr + proc->user_buffer_offset;
-		ret = vm_insert_page(vma, user_page_addr, page[0]);
-		if (ret) {
-			pr_err("%d: binder_alloc_buf failed to map page at %lx in userspace\n",
-			       proc->pid, user_page_addr);
-			goto err_vm_insert_page_failed;
-		}
-		/* vm_insert_page does not seem to increment the refcount */
-	}
-	if (mm) {
-		up_write(&mm->mmap_sem);
-		mmput(mm);
-	}
-	return 0;
+	if (task->policy == policy && task->normal_prio == desired.prio)
+		return;
 
-free_range:
-	for (page_addr = end - PAGE_SIZE; page_addr >= start;
-	     page_addr -= PAGE_SIZE) {
-		page = &proc->pages[(page_addr - proc->buffer) / PAGE_SIZE];
-		if (vma)
-			zap_page_range(vma, (uintptr_t)page_addr +
-				proc->user_buffer_offset, PAGE_SIZE, NULL);
-err_vm_insert_page_failed:
-		unmap_kernel_range((unsigned long)page_addr, PAGE_SIZE);
-err_map_kernel_failed:
-		__free_page(*page);
-		*page = NULL;
-err_alloc_page_failed:
-		;
-	}
-err_no_vma:
-	if (mm) {
-		up_write(&mm->mmap_sem);
-		mmput(mm);
-	}
-	return -ENOMEM;
-}
-
-static struct binder_buffer *binder_alloc_buf(struct binder_proc *proc,
-					      size_t data_size,
-					      size_t offsets_size, int is_async)
-{
-	struct rb_node *n = proc->free_buffers.rb_node;
-	struct binder_buffer *buffer;
-	size_t buffer_size;
-	struct rb_node *best_fit = NULL;
-	void *has_page_addr;
-	void *end_page_addr;
-	size_t size;
-
-	if (proc->vma == NULL) {
-		pr_err("%d: binder_alloc_buf, no vma\n",
-		       proc->pid);
-		return NULL;
-	}
+	has_cap_nice = has_capability_noaudit(task, CAP_SYS_NICE);
 
-	size = ALIGN(data_size, sizeof(void *)) +
-		ALIGN(offsets_size, sizeof(void *));
+	priority = to_userspace_prio(policy, desired.prio);
 
-	if (size < data_size || size < offsets_size) {
-		binder_user_error("%d: got transaction with invalid size %zd-%zd\n",
-				proc->pid, data_size, offsets_size);
-		return NULL;
-	}
+	if (verify && is_rt_policy(policy) && !has_cap_nice) {
+		long max_rtprio = task_rlimit(task, RLIMIT_RTPRIO);
 
-	if (is_async &&
-	    proc->free_async_space < size + sizeof(struct binder_buffer)) {
-		binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-			     "%d: binder_alloc_buf size %zd failed, no async space left\n",
-			      proc->pid, size);
-		return NULL;
+		if (max_rtprio == 0) {
+			policy = SCHED_NORMAL;
+			priority = MIN_NICE;
+		} else if (priority > max_rtprio) {
+			priority = max_rtprio;
+		}
 	}
 
-	while (n) {
-		buffer = rb_entry(n, struct binder_buffer, rb_node);
-		BUG_ON(!buffer->free);
-		buffer_size = binder_buffer_size(proc, buffer);
+	if (verify && is_fair_policy(policy) && !has_cap_nice) {
+		long min_nice = rlimit_to_nice(task_rlimit(task, RLIMIT_NICE));
 
-		if (size < buffer_size) {
-			best_fit = n;
-			n = n->rb_left;
-		} else if (size > buffer_size)
-			n = n->rb_right;
-		else {
-			best_fit = n;
-			break;
+		if (min_nice > MAX_NICE) {
+			binder_user_error("%d RLIMIT_NICE not set\n",
+					  task->pid);
+			return;
+		} else if (priority < min_nice) {
+			priority = min_nice;
 		}
 	}
-	if (best_fit == NULL) {
-		pr_err("%d: binder_alloc_buf size %zd failed, no address space\n",
-			proc->pid, size);
-		return NULL;
-	}
-	if (n == NULL) {
-		buffer = rb_entry(best_fit, struct binder_buffer, rb_node);
-		buffer_size = binder_buffer_size(proc, buffer);
-	}
 
-	binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-		     "%d: binder_alloc_buf size %zd got buffer %p size %zd\n",
-		      proc->pid, size, buffer, buffer_size);
+	if (policy != desired.sched_policy ||
+	    to_kernel_prio(policy, priority) != desired.prio)
+		binder_debug(BINDER_DEBUG_PRIORITY_CAP,
+			     "%d: priority %d not allowed, using %d instead\n",
+			      task->pid, desired.prio,
+			      to_kernel_prio(policy, priority));
 
-	has_page_addr =
-		(void *)(((uintptr_t)buffer->data + buffer_size) & PAGE_MASK);
-	if (n == NULL) {
-		if (size + sizeof(struct binder_buffer) + 4 >= buffer_size)
-			buffer_size = size; /* no room for other buffers */
-		else
-			buffer_size = size + sizeof(struct binder_buffer);
-	}
-	end_page_addr =
-		(void *)PAGE_ALIGN((uintptr_t)buffer->data + buffer_size);
-	if (end_page_addr > has_page_addr)
-		end_page_addr = has_page_addr;
-	if (binder_update_page_range(proc, 1,
-	    (void *)PAGE_ALIGN((uintptr_t)buffer->data), end_page_addr, NULL))
-		return NULL;
+	trace_binder_set_priority(task->tgid, task->pid, task->normal_prio,
+				  to_kernel_prio(policy, priority),
+				  desired.prio);
 
-	rb_erase(best_fit, &proc->free_buffers);
-	buffer->free = 0;
-	binder_insert_allocated_buffer(proc, buffer);
-	if (buffer_size != size) {
-		struct binder_buffer *new_buffer = (void *)buffer->data + size;
+	/* Set the actual priority */
+	if (task->policy != policy || is_rt_policy(policy)) {
+		struct sched_param params;
 
-		list_add(&new_buffer->entry, &buffer->entry);
-		new_buffer->free = 1;
-		binder_insert_free_buffer(proc, new_buffer);
-	}
-	binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-		     "%d: binder_alloc_buf size %zd got %p\n",
-		      proc->pid, size, buffer);
-	buffer->data_size = data_size;
-	buffer->offsets_size = offsets_size;
-	buffer->async_transaction = is_async;
-	if (is_async) {
-		proc->free_async_space -= size + sizeof(struct binder_buffer);
-		binder_debug(BINDER_DEBUG_BUFFER_ALLOC_ASYNC,
-			     "%d: binder_alloc_buf size %zd async free %zd\n",
-			      proc->pid, size, proc->free_async_space);
-	}
-
-	return buffer;
-}
+		params.sched_priority = is_rt_policy(policy) ? priority : 0;
 
-static void *buffer_start_page(struct binder_buffer *buffer)
-{
-	return (void *)((uintptr_t)buffer & PAGE_MASK);
+		sched_setscheduler_nocheck(task,
+					   policy | SCHED_RESET_ON_FORK,
+					   &params);
+	}
+	if (is_fair_policy(policy))
+		set_user_nice(task, priority);
 }
 
-static void *buffer_end_page(struct binder_buffer *buffer)
+static void binder_set_priority(struct task_struct *task,
+				struct binder_priority desired)
 {
-	return (void *)(((uintptr_t)(buffer + 1) - 1) & PAGE_MASK);
+	binder_do_set_priority(task, desired, /* verify = */ true);
 }
 
-static void binder_delete_free_buffer(struct binder_proc *proc,
-				      struct binder_buffer *buffer)
+static void binder_restore_priority(struct task_struct *task,
+				    struct binder_priority desired)
 {
-	struct binder_buffer *prev, *next = NULL;
-	int free_page_end = 1;
-	int free_page_start = 1;
-
-	BUG_ON(proc->buffers.next == &buffer->entry);
-	prev = list_entry(buffer->entry.prev, struct binder_buffer, entry);
-	BUG_ON(!prev->free);
-	if (buffer_end_page(prev) == buffer_start_page(buffer)) {
-		free_page_start = 0;
-		if (buffer_end_page(prev) == buffer_end_page(buffer))
-			free_page_end = 0;
-		binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-			     "%d: merge free, buffer %p share page with %p\n",
-			      proc->pid, buffer, prev);
-	}
-
-	if (!list_is_last(&buffer->entry, &proc->buffers)) {
-		next = list_entry(buffer->entry.next,
-				  struct binder_buffer, entry);
-		if (buffer_start_page(next) == buffer_end_page(buffer)) {
-			free_page_end = 0;
-			if (buffer_start_page(next) ==
-			    buffer_start_page(buffer))
-				free_page_start = 0;
-			binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-				     "%d: merge free, buffer %p share page with %p\n",
-				      proc->pid, buffer, prev);
-		}
-	}
-	list_del(&buffer->entry);
-	if (free_page_start || free_page_end) {
-		binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-			     "%d: merge free, buffer %p do not share page%s%s with %p or %p\n",
-			     proc->pid, buffer, free_page_start ? "" : " end",
-			     free_page_end ? "" : " start", prev, next);
-		binder_update_page_range(proc, 0, free_page_start ?
-			buffer_start_page(buffer) : buffer_end_page(buffer),
-			(free_page_end ? buffer_end_page(buffer) :
-			buffer_start_page(buffer)) + PAGE_SIZE, NULL);
-	}
+	binder_do_set_priority(task, desired, /* verify = */ false);
 }
 
-static void binder_free_buf(struct binder_proc *proc,
-			    struct binder_buffer *buffer)
+static void binder_transaction_priority(struct task_struct *task,
+					struct binder_transaction *t,
+					struct binder_priority node_prio,
+					bool inherit_rt)
 {
-	size_t size, buffer_size;
-
-	buffer_size = binder_buffer_size(proc, buffer);
-
-	size = ALIGN(buffer->data_size, sizeof(void *)) +
-		ALIGN(buffer->offsets_size, sizeof(void *));
+	struct binder_priority desired_prio = t->priority;
 
-	binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-		     "%d: binder_free_buf %p size %zd buffer_size %zd\n",
-		      proc->pid, buffer, size, buffer_size);
-
-	BUG_ON(buffer->free);
-	BUG_ON(size > buffer_size);
-	BUG_ON(buffer->transaction != NULL);
-	BUG_ON((void *)buffer < proc->buffer);
-	BUG_ON((void *)buffer > proc->buffer + proc->buffer_size);
+	if (t->set_priority_called)
+		return;
 
-	if (buffer->async_transaction) {
-		proc->free_async_space += size + sizeof(struct binder_buffer);
+	t->set_priority_called = true;
+	t->saved_priority.sched_policy = task->policy;
+	t->saved_priority.prio = task->normal_prio;
 
-		binder_debug(BINDER_DEBUG_BUFFER_ALLOC_ASYNC,
-			     "%d: binder_free_buf size %zd async free %zd\n",
-			      proc->pid, size, proc->free_async_space);
+	if (!inherit_rt && is_rt_policy(desired_prio.sched_policy)) {
+		desired_prio.prio = NICE_TO_PRIO(0);
+		desired_prio.sched_policy = SCHED_NORMAL;
 	}
 
-	binder_update_page_range(proc, 0,
-		(void *)PAGE_ALIGN((uintptr_t)buffer->data),
-		(void *)(((uintptr_t)buffer->data + buffer_size) & PAGE_MASK),
-		NULL);
-	rb_erase(&buffer->rb_node, &proc->allocated_buffers);
-	buffer->free = 1;
-	if (!list_is_last(&buffer->entry, &proc->buffers)) {
-		struct binder_buffer *next = list_entry(buffer->entry.next,
-						struct binder_buffer, entry);
-
-		if (next->free) {
-			rb_erase(&next->rb_node, &proc->free_buffers);
-			binder_delete_free_buffer(proc, next);
-		}
+	if (node_prio.prio < t->priority.prio ||
+	    (node_prio.prio == t->priority.prio &&
+	     node_prio.sched_policy == SCHED_FIFO)) {
+		/*
+		 * In case the minimum priority on the node is
+		 * higher (lower value), use that priority. If
+		 * the priority is the same, but the node uses
+		 * SCHED_FIFO, prefer SCHED_FIFO, since it can
+		 * run unbounded, unlike SCHED_RR.
+		 */
+		desired_prio = node_prio;
 	}
-	if (proc->buffers.next != &buffer->entry) {
-		struct binder_buffer *prev = list_entry(buffer->entry.prev,
-						struct binder_buffer, entry);
 
-		if (prev->free) {
-			binder_delete_free_buffer(proc, buffer);
-			rb_erase(&prev->rb_node, &proc->free_buffers);
-			buffer = prev;
-		}
-	}
-	binder_insert_free_buffer(proc, buffer);
+	binder_set_priority(task, desired_prio);
 }
 
-static struct binder_node *binder_get_node(struct binder_proc *proc,
-					   binder_uintptr_t ptr)
+static struct binder_node *binder_get_node_ilocked(struct binder_proc *proc,
+						   binder_uintptr_t ptr)
 {
 	struct rb_node *n = proc->nodes.rb_node;
 	struct binder_node *node;
 
+	assert_spin_locked(&proc->inner_lock);
+
 	while (n) {
 		node = rb_entry(n, struct binder_node, rb_node);
 
@@ -877,21 +1295,47 @@ static struct binder_node *binder_get_node(struct binder_proc *proc,
 			n = n->rb_left;
 		else if (ptr > node->ptr)
 			n = n->rb_right;
-		else
+		else {
+			/*
+			 * take an implicit weak reference
+			 * to ensure node stays alive until
+			 * call to binder_put_node()
+			 */
+			binder_inc_node_tmpref_ilocked(node);
 			return node;
+		}
 	}
 	return NULL;
 }
 
-static struct binder_node *binder_new_node(struct binder_proc *proc,
-					   binder_uintptr_t ptr,
-					   binder_uintptr_t cookie)
+static struct binder_node *binder_get_node(struct binder_proc *proc,
+					   binder_uintptr_t ptr)
+{
+	struct binder_node *node;
+
+	binder_inner_proc_lock(proc);
+	node = binder_get_node_ilocked(proc, ptr);
+	binder_inner_proc_unlock(proc);
+	return node;
+}
+
+static struct binder_node *binder_init_node_ilocked(
+						struct binder_proc *proc,
+						struct binder_node *new_node,
+						struct flat_binder_object *fp)
 {
 	struct rb_node **p = &proc->nodes.rb_node;
 	struct rb_node *parent = NULL;
 	struct binder_node *node;
+	binder_uintptr_t ptr = fp ? fp->binder : 0;
+	binder_uintptr_t cookie = fp ? fp->cookie : 0;
+	__u32 flags = fp ? fp->flags : 0;
+	s8 priority;
+
+	assert_spin_locked(&proc->inner_lock);
 
 	while (*p) {
+
 		parent = *p;
 		node = rb_entry(parent, struct binder_node, rb_node);
 
@@ -899,39 +1343,86 @@ static struct binder_node *binder_new_node(struct binder_proc *proc,
 			p = &(*p)->rb_left;
 		else if (ptr > node->ptr)
 			p = &(*p)->rb_right;
-		else
-			return NULL;
+		else {
+			/*
+			 * A matching node is already in
+			 * the rb tree. Abandon the init
+			 * and return it.
+			 */
+			binder_inc_node_tmpref_ilocked(node);
+			return node;
+		}
 	}
-
-	node = kzalloc(sizeof(*node), GFP_KERNEL);
-	if (node == NULL)
-		return NULL;
+	node = new_node;
 	binder_stats_created(BINDER_STAT_NODE);
+	node->tmp_refs++;
 	rb_link_node(&node->rb_node, parent, p);
 	rb_insert_color(&node->rb_node, &proc->nodes);
-	node->debug_id = ++binder_last_id;
+	node->debug_id = atomic_inc_return(&binder_last_id);
 	node->proc = proc;
 	node->ptr = ptr;
 	node->cookie = cookie;
 	node->work.type = BINDER_WORK_NODE;
+	priority = flags & FLAT_BINDER_FLAG_PRIORITY_MASK;
+	node->sched_policy = (flags & FLAT_BINDER_FLAG_SCHED_POLICY_MASK) >>
+		FLAT_BINDER_FLAG_SCHED_POLICY_SHIFT;
+	node->min_priority = to_kernel_prio(node->sched_policy, priority);
+	node->accept_fds = !!(flags & FLAT_BINDER_FLAG_ACCEPTS_FDS);
+	node->inherit_rt = !!(flags & FLAT_BINDER_FLAG_INHERIT_RT);
+	spin_lock_init(&node->lock);
 	INIT_LIST_HEAD(&node->work.entry);
 	INIT_LIST_HEAD(&node->async_todo);
 	binder_debug(BINDER_DEBUG_INTERNAL_REFS,
 		     "%d:%d node %d u%016llx c%016llx created\n",
 		     proc->pid, current->pid, node->debug_id,
 		     (u64)node->ptr, (u64)node->cookie);
+
 	return node;
 }
 
-static int binder_inc_node(struct binder_node *node, int strong, int internal,
-			   struct list_head *target_list)
+static struct binder_node *binder_new_node(struct binder_proc *proc,
+					   struct flat_binder_object *fp)
+{
+	struct binder_node *node;
+	struct binder_node *new_node = kzalloc(sizeof(*node), GFP_KERNEL);
+
+	if (!new_node)
+		return NULL;
+	binder_inner_proc_lock(proc);
+	node = binder_init_node_ilocked(proc, new_node, fp);
+	binder_inner_proc_unlock(proc);
+	if (node != new_node)
+		/*
+		 * The node was already added by another thread
+		 */
+		kfree(new_node);
+
+	return node;
+}
+
+static void binder_free_node(struct binder_node *node)
+{
+	kfree(node);
+	binder_stats_deleted(BINDER_STAT_NODE);
+}
+
+static int binder_inc_node_nilocked(struct binder_node *node, int strong,
+				    int internal,
+				    struct list_head *target_list)
 {
+	struct binder_proc *proc = node->proc;
+
+	assert_spin_locked(&node->lock);
+	if (proc)
+		assert_spin_locked(&proc->inner_lock);
 	if (strong) {
 		if (internal) {
 			if (target_list == NULL &&
 			    node->internal_strong_refs == 0 &&
-			    !(node == binder_context_mgr_node &&
-			    node->has_strong_ref)) {
+			    !(node->proc &&
+			      node == node->proc->context->
+				      binder_context_mgr_node &&
+			      node->has_strong_ref)) {
 				pr_err("invalid inc strong node for %d\n",
 					node->debug_id);
 				return -EINVAL;
@@ -940,8 +1431,19 @@ static int binder_inc_node(struct binder_node *node, int strong, int internal,
 		} else
 			node->local_strong_refs++;
 		if (!node->has_strong_ref && target_list) {
-			list_del_init(&node->work.entry);
-			list_add_tail(&node->work.entry, target_list);
+			binder_dequeue_work_ilocked(&node->work);
+			/*
+			 * Note: this function is the only place where we queue
+			 * directly to a thread->todo without using the
+			 * corresponding binder_enqueue_thread_work() helper
+			 * functions; in this case it's ok to not set the
+			 * process_todo flag, since we know this node work will
+			 * always be followed by other work that starts queue
+			 * processing: in case of synchronous transactions, a
+			 * BR_REPLY or BR_ERROR; in case of oneway
+			 * transactions, a BR_TRANSACTION_COMPLETE.
+			 */
+			binder_enqueue_work_ilocked(&node->work, target_list);
 		}
 	} else {
 		if (!internal)
@@ -952,58 +1454,172 @@ static int binder_inc_node(struct binder_node *node, int strong, int internal,
 					node->debug_id);
 				return -EINVAL;
 			}
-			list_add_tail(&node->work.entry, target_list);
+			/*
+			 * See comment above
+			 */
+			binder_enqueue_work_ilocked(&node->work, target_list);
 		}
 	}
 	return 0;
 }
 
-static int binder_dec_node(struct binder_node *node, int strong, int internal)
+static int binder_inc_node(struct binder_node *node, int strong, int internal,
+			   struct list_head *target_list)
 {
+	int ret;
+
+	binder_node_inner_lock(node);
+	ret = binder_inc_node_nilocked(node, strong, internal, target_list);
+	binder_node_inner_unlock(node);
+
+	return ret;
+}
+
+static bool binder_dec_node_nilocked(struct binder_node *node,
+				     int strong, int internal)
+{
+	struct binder_proc *proc = node->proc;
+
+	assert_spin_locked(&node->lock);
+	if (proc)
+		assert_spin_locked(&proc->inner_lock);
 	if (strong) {
 		if (internal)
 			node->internal_strong_refs--;
 		else
 			node->local_strong_refs--;
 		if (node->local_strong_refs || node->internal_strong_refs)
-			return 0;
+			return false;
 	} else {
 		if (!internal)
 			node->local_weak_refs--;
-		if (node->local_weak_refs || !hlist_empty(&node->refs))
-			return 0;
+		if (node->local_weak_refs || node->tmp_refs ||
+				!hlist_empty(&node->refs))
+			return false;
 	}
-	if (node->proc && (node->has_strong_ref || node->has_weak_ref)) {
+
+	if (proc && (node->has_strong_ref || node->has_weak_ref)) {
 		if (list_empty(&node->work.entry)) {
-			list_add_tail(&node->work.entry, &node->proc->todo);
-			wake_up_interruptible(&node->proc->wait);
+			binder_enqueue_work_ilocked(&node->work, &proc->todo);
+			binder_wakeup_proc_ilocked(proc);
 		}
 	} else {
 		if (hlist_empty(&node->refs) && !node->local_strong_refs &&
-		    !node->local_weak_refs) {
-			list_del_init(&node->work.entry);
-			if (node->proc) {
-				rb_erase(&node->rb_node, &node->proc->nodes);
+		    !node->local_weak_refs && !node->tmp_refs) {
+			if (proc) {
+				binder_dequeue_work_ilocked(&node->work);
+				rb_erase(&node->rb_node, &proc->nodes);
 				binder_debug(BINDER_DEBUG_INTERNAL_REFS,
 					     "refless node %d deleted\n",
 					     node->debug_id);
 			} else {
+				BUG_ON(!list_empty(&node->work.entry));
+				spin_lock(&binder_dead_nodes_lock);
+				/*
+				 * tmp_refs could have changed so
+				 * check it again
+				 */
+				if (node->tmp_refs) {
+					spin_unlock(&binder_dead_nodes_lock);
+					return false;
+				}
 				hlist_del(&node->dead_node);
+				spin_unlock(&binder_dead_nodes_lock);
 				binder_debug(BINDER_DEBUG_INTERNAL_REFS,
 					     "dead node %d deleted\n",
 					     node->debug_id);
 			}
-			kfree(node);
-			binder_stats_deleted(BINDER_STAT_NODE);
+			return true;
 		}
 	}
+	return false;
+}
 
-	return 0;
+static void binder_dec_node(struct binder_node *node, int strong, int internal)
+{
+	bool free_node;
+
+	binder_node_inner_lock(node);
+	free_node = binder_dec_node_nilocked(node, strong, internal);
+	binder_node_inner_unlock(node);
+	if (free_node)
+		binder_free_node(node);
 }
 
+static void binder_inc_node_tmpref_ilocked(struct binder_node *node)
+{
+	/*
+	 * No call to binder_inc_node() is needed since we
+	 * don't need to inform userspace of any changes to
+	 * tmp_refs
+	 */
+	node->tmp_refs++;
+}
+
+/**
+ * binder_inc_node_tmpref() - take a temporary reference on node
+ * @node:	node to reference
+ *
+ * Take reference on node to prevent the node from being freed
+ * while referenced only by a local variable. The inner lock is
+ * needed to serialize with the node work on the queue (which
+ * isn't needed after the node is dead). If the node is dead
+ * (node->proc is NULL), use binder_dead_nodes_lock to protect
+ * node->tmp_refs against dead-node-only cases where the node
+ * lock cannot be acquired (eg traversing the dead node list to
+ * print nodes)
+ */
+static void binder_inc_node_tmpref(struct binder_node *node)
+{
+	binder_node_lock(node);
+	if (node->proc)
+		binder_inner_proc_lock(node->proc);
+	else
+		spin_lock(&binder_dead_nodes_lock);
+	binder_inc_node_tmpref_ilocked(node);
+	if (node->proc)
+		binder_inner_proc_unlock(node->proc);
+	else
+		spin_unlock(&binder_dead_nodes_lock);
+	binder_node_unlock(node);
+}
+
+/**
+ * binder_dec_node_tmpref() - remove a temporary reference on node
+ * @node:	node to reference
+ *
+ * Release temporary reference on node taken via binder_inc_node_tmpref()
+ */
+static void binder_dec_node_tmpref(struct binder_node *node)
+{
+	bool free_node;
+
+	binder_node_inner_lock(node);
+	if (!node->proc)
+		spin_lock(&binder_dead_nodes_lock);
+	node->tmp_refs--;
+	BUG_ON(node->tmp_refs < 0);
+	if (!node->proc)
+		spin_unlock(&binder_dead_nodes_lock);
+	/*
+	 * Call binder_dec_node() to check if all refcounts are 0
+	 * and cleanup is needed. Calling with strong=0 and internal=1
+	 * causes no actual reference to be released in binder_dec_node().
+	 * If that changes, a change is needed here too.
+	 */
+	free_node = binder_dec_node_nilocked(node, 0, 1);
+	binder_node_inner_unlock(node);
+	if (free_node)
+		binder_free_node(node);
+}
+
+static void binder_put_node(struct binder_node *node)
+{
+	binder_dec_node_tmpref(node);
+}
 
-static struct binder_ref *binder_get_ref(struct binder_proc *proc,
-					 u32 desc, bool need_strong_ref)
+static struct binder_ref *binder_get_ref_olocked(struct binder_proc *proc,
+						 u32 desc, bool need_strong_ref)
 {
 	struct rb_node *n = proc->refs_by_desc.rb_node;
 	struct binder_ref *ref;
@@ -1011,11 +1627,11 @@ static struct binder_ref *binder_get_ref(struct binder_proc *proc,
 	while (n) {
 		ref = rb_entry(n, struct binder_ref, rb_node_desc);
 
-		if (desc < ref->desc) {
+		if (desc < ref->data.desc) {
 			n = n->rb_left;
-		} else if (desc > ref->desc) {
+		} else if (desc > ref->data.desc) {
 			n = n->rb_right;
-		} else if (need_strong_ref && !ref->strong) {
+		} else if (need_strong_ref && !ref->data.strong) {
 			binder_user_error("tried to use weak ref as strong ref\n");
 			return NULL;
 		} else {
@@ -1025,13 +1641,34 @@ static struct binder_ref *binder_get_ref(struct binder_proc *proc,
 	return NULL;
 }
 
-static struct binder_ref *binder_get_ref_for_node(struct binder_proc *proc,
-						  struct binder_node *node)
+/**
+ * binder_get_ref_for_node_olocked() - get the ref associated with given node
+ * @proc:	binder_proc that owns the ref
+ * @node:	binder_node of target
+ * @new_ref:	newly allocated binder_ref to be initialized or %NULL
+ *
+ * Look up the ref for the given node and return it if it exists
+ *
+ * If it doesn't exist and the caller provides a newly allocated
+ * ref, initialize the fields of the newly allocated ref and insert
+ * into the given proc rb_trees and node refs list.
+ *
+ * Return:	the ref for node. It is possible that another thread
+ *		allocated/initialized the ref first in which case the
+ *		returned ref would be different than the passed-in
+ *		new_ref. new_ref must be kfree'd by the caller in
+ *		this case.
+ */
+static struct binder_ref *binder_get_ref_for_node_olocked(
+					struct binder_proc *proc,
+					struct binder_node *node,
+					struct binder_ref *new_ref)
 {
-	struct rb_node *n;
+	struct binder_context *context = proc->context;
 	struct rb_node **p = &proc->refs_by_node.rb_node;
 	struct rb_node *parent = NULL;
-	struct binder_ref *ref, *new_ref;
+	struct binder_ref *ref;
+	struct rb_node *n;
 
 	while (*p) {
 		parent = *p;
@@ -1044,22 +1681,22 @@ static struct binder_ref *binder_get_ref_for_node(struct binder_proc *proc,
 		else
 			return ref;
 	}
-	new_ref = kzalloc(sizeof(*ref), GFP_KERNEL);
-	if (new_ref == NULL)
+	if (!new_ref)
 		return NULL;
+
 	binder_stats_created(BINDER_STAT_REF);
-	new_ref->debug_id = ++binder_last_id;
+	new_ref->data.debug_id = atomic_inc_return(&binder_last_id);
 	new_ref->proc = proc;
 	new_ref->node = node;
 	rb_link_node(&new_ref->rb_node_node, parent, p);
 	rb_insert_color(&new_ref->rb_node_node, &proc->refs_by_node);
 
-	new_ref->desc = (node == binder_context_mgr_node) ? 0 : 1;
+	new_ref->data.desc = (node == context->binder_context_mgr_node) ? 0 : 1;
 	for (n = rb_first(&proc->refs_by_desc); n != NULL; n = rb_next(n)) {
 		ref = rb_entry(n, struct binder_ref, rb_node_desc);
-		if (ref->desc > new_ref->desc)
+		if (ref->data.desc > new_ref->data.desc)
 			break;
-		new_ref->desc = ref->desc + 1;
+		new_ref->data.desc = ref->data.desc + 1;
 	}
 
 	p = &proc->refs_by_desc.rb_node;
@@ -1067,121 +1704,423 @@ static struct binder_ref *binder_get_ref_for_node(struct binder_proc *proc,
 		parent = *p;
 		ref = rb_entry(parent, struct binder_ref, rb_node_desc);
 
-		if (new_ref->desc < ref->desc)
+		if (new_ref->data.desc < ref->data.desc)
 			p = &(*p)->rb_left;
-		else if (new_ref->desc > ref->desc)
+		else if (new_ref->data.desc > ref->data.desc)
 			p = &(*p)->rb_right;
 		else
 			BUG();
 	}
 	rb_link_node(&new_ref->rb_node_desc, parent, p);
 	rb_insert_color(&new_ref->rb_node_desc, &proc->refs_by_desc);
-	if (node) {
-		hlist_add_head(&new_ref->node_entry, &node->refs);
 
-		binder_debug(BINDER_DEBUG_INTERNAL_REFS,
-			     "%d new ref %d desc %d for node %d\n",
-			      proc->pid, new_ref->debug_id, new_ref->desc,
-			      node->debug_id);
-	} else {
-		binder_debug(BINDER_DEBUG_INTERNAL_REFS,
-			     "%d new ref %d desc %d for dead node\n",
-			      proc->pid, new_ref->debug_id, new_ref->desc);
-	}
+	binder_node_lock(node);
+	hlist_add_head(&new_ref->node_entry, &node->refs);
+
+	binder_debug(BINDER_DEBUG_INTERNAL_REFS,
+		     "%d new ref %d desc %d for node %d\n",
+		      proc->pid, new_ref->data.debug_id, new_ref->data.desc,
+		      node->debug_id);
+	binder_node_unlock(node);
 	return new_ref;
 }
 
-static void binder_delete_ref(struct binder_ref *ref)
+static void binder_cleanup_ref_olocked(struct binder_ref *ref)
 {
+	bool delete_node = false;
+
 	binder_debug(BINDER_DEBUG_INTERNAL_REFS,
 		     "%d delete ref %d desc %d for node %d\n",
-		      ref->proc->pid, ref->debug_id, ref->desc,
+		      ref->proc->pid, ref->data.debug_id, ref->data.desc,
 		      ref->node->debug_id);
 
 	rb_erase(&ref->rb_node_desc, &ref->proc->refs_by_desc);
 	rb_erase(&ref->rb_node_node, &ref->proc->refs_by_node);
-	if (ref->strong)
-		binder_dec_node(ref->node, 1, 1);
+
+	binder_node_inner_lock(ref->node);
+	if (ref->data.strong)
+		binder_dec_node_nilocked(ref->node, 1, 1);
+
 	hlist_del(&ref->node_entry);
-	binder_dec_node(ref->node, 0, 1);
+	delete_node = binder_dec_node_nilocked(ref->node, 0, 1);
+	binder_node_inner_unlock(ref->node);
+	/*
+	 * Clear ref->node unless we want the caller to free the node
+	 */
+	if (!delete_node) {
+		/*
+		 * The caller uses ref->node to determine
+		 * whether the node needs to be freed. Clear
+		 * it since the node is still alive.
+		 */
+		ref->node = NULL;
+	}
+
 	if (ref->death) {
 		binder_debug(BINDER_DEBUG_DEAD_BINDER,
 			     "%d delete ref %d desc %d has death notification\n",
-			      ref->proc->pid, ref->debug_id, ref->desc);
-		list_del(&ref->death->work.entry);
-		kfree(ref->death);
+			      ref->proc->pid, ref->data.debug_id,
+			      ref->data.desc);
+		binder_dequeue_work(ref->proc, &ref->death->work);
 		binder_stats_deleted(BINDER_STAT_DEATH);
 	}
-	kfree(ref);
 	binder_stats_deleted(BINDER_STAT_REF);
 }
 
-static int binder_inc_ref(struct binder_ref *ref, int strong,
-			  struct list_head *target_list)
+/**
+ * binder_inc_ref_olocked() - increment the ref for given handle
+ * @ref:         ref to be incremented
+ * @strong:      if true, strong increment, else weak
+ * @target_list: list to queue node work on
+ *
+ * Increment the ref. @ref->proc->outer_lock must be held on entry
+ *
+ * Return: 0, if successful, else errno
+ */
+static int binder_inc_ref_olocked(struct binder_ref *ref, int strong,
+				  struct list_head *target_list)
 {
 	int ret;
 
 	if (strong) {
-		if (ref->strong == 0) {
+		if (ref->data.strong == 0) {
 			ret = binder_inc_node(ref->node, 1, 1, target_list);
 			if (ret)
 				return ret;
 		}
-		ref->strong++;
+		ref->data.strong++;
 	} else {
-		if (ref->weak == 0) {
+		if (ref->data.weak == 0) {
 			ret = binder_inc_node(ref->node, 0, 1, target_list);
 			if (ret)
 				return ret;
 		}
-		ref->weak++;
+		ref->data.weak++;
 	}
 	return 0;
 }
 
-
-static int binder_dec_ref(struct binder_ref *ref, int strong)
+/**
+ * binder_dec_ref() - dec the ref for given handle
+ * @ref:	ref to be decremented
+ * @strong:	if true, strong decrement, else weak
+ *
+ * Decrement the ref.
+ *
+ * Return: true if ref is cleaned up and ready to be freed
+ */
+static bool binder_dec_ref_olocked(struct binder_ref *ref, int strong)
 {
 	if (strong) {
-		if (ref->strong == 0) {
+		if (ref->data.strong == 0) {
 			binder_user_error("%d invalid dec strong, ref %d desc %d s %d w %d\n",
-					  ref->proc->pid, ref->debug_id,
-					  ref->desc, ref->strong, ref->weak);
-			return -EINVAL;
-		}
-		ref->strong--;
-		if (ref->strong == 0) {
-			int ret;
-
-			ret = binder_dec_node(ref->node, strong, 1);
-			if (ret)
-				return ret;
+					  ref->proc->pid, ref->data.debug_id,
+					  ref->data.desc, ref->data.strong,
+					  ref->data.weak);
+			return false;
 		}
+		ref->data.strong--;
+		if (ref->data.strong == 0)
+			binder_dec_node(ref->node, strong, 1);
 	} else {
-		if (ref->weak == 0) {
+		if (ref->data.weak == 0) {
 			binder_user_error("%d invalid dec weak, ref %d desc %d s %d w %d\n",
-					  ref->proc->pid, ref->debug_id,
-					  ref->desc, ref->strong, ref->weak);
-			return -EINVAL;
+					  ref->proc->pid, ref->data.debug_id,
+					  ref->data.desc, ref->data.strong,
+					  ref->data.weak);
+			return false;
 		}
-		ref->weak--;
+		ref->data.weak--;
 	}
-	if (ref->strong == 0 && ref->weak == 0)
-		binder_delete_ref(ref);
-	return 0;
+	if (ref->data.strong == 0 && ref->data.weak == 0) {
+		binder_cleanup_ref_olocked(ref);
+		return true;
+	}
+	return false;
 }
 
-static void binder_pop_transaction(struct binder_thread *target_thread,
-				   struct binder_transaction *t)
+/**
+ * binder_get_node_from_ref() - get the node from the given proc/desc
+ * @proc:	proc containing the ref
+ * @desc:	the handle associated with the ref
+ * @need_strong_ref: if true, only return node if ref is strong
+ * @rdata:	the id/refcount data for the ref
+ *
+ * Given a proc and ref handle, return the associated binder_node
+ *
+ * Return: a binder_node or NULL if not found or not strong when strong required
+ */
+static struct binder_node *binder_get_node_from_ref(
+		struct binder_proc *proc,
+		u32 desc, bool need_strong_ref,
+		struct binder_ref_data *rdata)
 {
-	if (target_thread) {
-		BUG_ON(target_thread->transaction_stack != t);
-		BUG_ON(target_thread->transaction_stack->from != target_thread);
-		target_thread->transaction_stack =
-			target_thread->transaction_stack->from_parent;
-		t->from = NULL;
+	struct binder_node *node;
+	struct binder_ref *ref;
+
+	binder_proc_lock(proc);
+	ref = binder_get_ref_olocked(proc, desc, need_strong_ref);
+	if (!ref)
+		goto err_no_ref;
+	node = ref->node;
+	/*
+	 * Take an implicit reference on the node to ensure
+	 * it stays alive until the call to binder_put_node()
+	 */
+	binder_inc_node_tmpref(node);
+	if (rdata)
+		*rdata = ref->data;
+	binder_proc_unlock(proc);
+
+	return node;
+
+err_no_ref:
+	binder_proc_unlock(proc);
+	return NULL;
+}
+
+/**
+ * binder_free_ref() - free the binder_ref
+ * @ref:	ref to free
+ *
+ * Free the binder_ref. Free the binder_node indicated by ref->node
+ * (if non-NULL) and the binder_ref_death indicated by ref->death.
+ */
+static void binder_free_ref(struct binder_ref *ref)
+{
+	if (ref->node)
+		binder_free_node(ref->node);
+	kfree(ref->death);
+	kfree(ref);
+}
+
+/**
+ * binder_update_ref_for_handle() - inc/dec the ref for given handle
+ * @proc:	proc containing the ref
+ * @desc:	the handle associated with the ref
+ * @increment:	true=inc reference, false=dec reference
+ * @strong:	true=strong reference, false=weak reference
+ * @rdata:	the id/refcount data for the ref
+ *
+ * Given a proc and ref handle, increment or decrement the ref
+ * according to "increment" arg.
+ *
+ * Return: 0 if successful, else errno
+ */
+static int binder_update_ref_for_handle(struct binder_proc *proc,
+		uint32_t desc, bool increment, bool strong,
+		struct binder_ref_data *rdata)
+{
+	int ret = 0;
+	struct binder_ref *ref;
+	bool delete_ref = false;
+
+	binder_proc_lock(proc);
+	ref = binder_get_ref_olocked(proc, desc, strong);
+	if (!ref) {
+		ret = -EINVAL;
+		goto err_no_ref;
 	}
-	t->need_reply = 0;
+	if (increment)
+		ret = binder_inc_ref_olocked(ref, strong, NULL);
+	else
+		delete_ref = binder_dec_ref_olocked(ref, strong);
+
+	if (rdata)
+		*rdata = ref->data;
+	binder_proc_unlock(proc);
+
+	if (delete_ref)
+		binder_free_ref(ref);
+	return ret;
+
+err_no_ref:
+	binder_proc_unlock(proc);
+	return ret;
+}
+
+/**
+ * binder_dec_ref_for_handle() - dec the ref for given handle
+ * @proc:	proc containing the ref
+ * @desc:	the handle associated with the ref
+ * @strong:	true=strong reference, false=weak reference
+ * @rdata:	the id/refcount data for the ref
+ *
+ * Just calls binder_update_ref_for_handle() to decrement the ref.
+ *
+ * Return: 0 if successful, else errno
+ */
+static int binder_dec_ref_for_handle(struct binder_proc *proc,
+		uint32_t desc, bool strong, struct binder_ref_data *rdata)
+{
+	return binder_update_ref_for_handle(proc, desc, false, strong, rdata);
+}
+
+
+/**
+ * binder_inc_ref_for_node() - increment the ref for given proc/node
+ * @proc:	 proc containing the ref
+ * @node:	 target node
+ * @strong:	 true=strong reference, false=weak reference
+ * @target_list: worklist to use if node is incremented
+ * @rdata:	 the id/refcount data for the ref
+ *
+ * Given a proc and node, increment the ref. Create the ref if it
+ * doesn't already exist
+ *
+ * Return: 0 if successful, else errno
+ */
+static int binder_inc_ref_for_node(struct binder_proc *proc,
+			struct binder_node *node,
+			bool strong,
+			struct list_head *target_list,
+			struct binder_ref_data *rdata)
+{
+	struct binder_ref *ref;
+	struct binder_ref *new_ref = NULL;
+	int ret = 0;
+
+	binder_proc_lock(proc);
+	ref = binder_get_ref_for_node_olocked(proc, node, NULL);
+	if (!ref) {
+		binder_proc_unlock(proc);
+		new_ref = kzalloc(sizeof(*ref), GFP_KERNEL);
+		if (!new_ref)
+			return -ENOMEM;
+		binder_proc_lock(proc);
+		ref = binder_get_ref_for_node_olocked(proc, node, new_ref);
+	}
+	ret = binder_inc_ref_olocked(ref, strong, target_list);
+	*rdata = ref->data;
+	binder_proc_unlock(proc);
+	if (new_ref && ref != new_ref)
+		/*
+		 * Another thread created the ref first so
+		 * free the one we allocated
+		 */
+		kfree(new_ref);
+	return ret;
+}
+
+static void binder_pop_transaction_ilocked(struct binder_thread *target_thread,
+					   struct binder_transaction *t)
+{
+	BUG_ON(!target_thread);
+	assert_spin_locked(&target_thread->proc->inner_lock);
+	BUG_ON(target_thread->transaction_stack != t);
+	BUG_ON(target_thread->transaction_stack->from != target_thread);
+	target_thread->transaction_stack =
+		target_thread->transaction_stack->from_parent;
+	t->from = NULL;
+}
+
+/**
+ * binder_thread_dec_tmpref() - decrement thread->tmp_ref
+ * @thread:	thread to decrement
+ *
+ * A thread needs to be kept alive while being used to create or
+ * handle a transaction. binder_get_txn_from() is used to safely
+ * extract t->from from a binder_transaction and keep the thread
+ * indicated by t->from from being freed. When done with that
+ * binder_thread, this function is called to decrement the
+ * tmp_ref and free if appropriate (thread has been released
+ * and no transaction being processed by the driver)
+ */
+static void binder_thread_dec_tmpref(struct binder_thread *thread)
+{
+	/*
+	 * atomic is used to protect the counter value while
+	 * it cannot reach zero or thread->is_dead is false
+	 */
+	binder_inner_proc_lock(thread->proc);
+	atomic_dec(&thread->tmp_ref);
+	if (thread->is_dead && !atomic_read(&thread->tmp_ref)) {
+		binder_inner_proc_unlock(thread->proc);
+		binder_free_thread(thread);
+		return;
+	}
+	binder_inner_proc_unlock(thread->proc);
+}
+
+/**
+ * binder_proc_dec_tmpref() - decrement proc->tmp_ref
+ * @proc:	proc to decrement
+ *
+ * A binder_proc needs to be kept alive while being used to create or
+ * handle a transaction. proc->tmp_ref is incremented when
+ * creating a new transaction or the binder_proc is currently in-use
+ * by threads that are being released. When done with the binder_proc,
+ * this function is called to decrement the counter and free the
+ * proc if appropriate (proc has been released, all threads have
+ * been released and not currenly in-use to process a transaction).
+ */
+static void binder_proc_dec_tmpref(struct binder_proc *proc)
+{
+	binder_inner_proc_lock(proc);
+	proc->tmp_ref--;
+	if (proc->is_dead && RB_EMPTY_ROOT(&proc->threads) &&
+			!proc->tmp_ref) {
+		binder_inner_proc_unlock(proc);
+		binder_free_proc(proc);
+		return;
+	}
+	binder_inner_proc_unlock(proc);
+}
+
+/**
+ * binder_get_txn_from() - safely extract the "from" thread in transaction
+ * @t:	binder transaction for t->from
+ *
+ * Atomically return the "from" thread and increment the tmp_ref
+ * count for the thread to ensure it stays alive until
+ * binder_thread_dec_tmpref() is called.
+ *
+ * Return: the value of t->from
+ */
+static struct binder_thread *binder_get_txn_from(
+		struct binder_transaction *t)
+{
+	struct binder_thread *from;
+
+	spin_lock(&t->lock);
+	from = t->from;
+	if (from)
+		atomic_inc(&from->tmp_ref);
+	spin_unlock(&t->lock);
+	return from;
+}
+
+/**
+ * binder_get_txn_from_and_acq_inner() - get t->from and acquire inner lock
+ * @t:	binder transaction for t->from
+ *
+ * Same as binder_get_txn_from() except it also acquires the proc->inner_lock
+ * to guarantee that the thread cannot be released while operating on it.
+ * The caller must call binder_inner_proc_unlock() to release the inner lock
+ * as well as call binder_dec_thread_txn() to release the reference.
+ *
+ * Return: the value of t->from
+ */
+static struct binder_thread *binder_get_txn_from_and_acq_inner(
+		struct binder_transaction *t)
+{
+	struct binder_thread *from;
+
+	from = binder_get_txn_from(t);
+	if (!from)
+		return NULL;
+	binder_inner_proc_lock(from->proc);
+	if (t->from) {
+		BUG_ON(from != t->from);
+		return from;
+	}
+	binder_inner_proc_unlock(from->proc);
+	binder_thread_dec_tmpref(from);
+	return NULL;
+}
+
+static void binder_free_transaction(struct binder_transaction *t)
+{
 	if (t->buffer)
 		t->buffer->transaction = NULL;
 	kfree(t);
@@ -1196,30 +2135,28 @@ static void binder_send_failed_reply(struct binder_transaction *t,
 
 	BUG_ON(t->flags & TF_ONE_WAY);
 	while (1) {
-		target_thread = t->from;
+		target_thread = binder_get_txn_from_and_acq_inner(t);
 		if (target_thread) {
-			if (target_thread->return_error != BR_OK &&
-			   target_thread->return_error2 == BR_OK) {
-				target_thread->return_error2 =
-					target_thread->return_error;
-				target_thread->return_error = BR_OK;
-			}
-			if (target_thread->return_error == BR_OK) {
-				binder_debug(BINDER_DEBUG_FAILED_TRANSACTION,
-					     "send failed reply for transaction %d to %d:%d\n",
-					      t->debug_id,
-					      target_thread->proc->pid,
-					      target_thread->pid);
-
-				binder_pop_transaction(target_thread, t);
-				target_thread->return_error = error_code;
+			binder_debug(BINDER_DEBUG_FAILED_TRANSACTION,
+				     "send failed reply for transaction %d to %d:%d\n",
+				      t->debug_id,
+				      target_thread->proc->pid,
+				      target_thread->pid);
+
+			binder_pop_transaction_ilocked(target_thread, t);
+			if (target_thread->reply_error.cmd == BR_OK) {
+				target_thread->reply_error.cmd = error_code;
+				binder_enqueue_thread_work_ilocked(
+					target_thread,
+					&target_thread->reply_error.work);
 				wake_up_interruptible(&target_thread->wait);
 			} else {
-				pr_err("reply failed, target thread, %d:%d, has error code %d already\n",
-					target_thread->proc->pid,
-					target_thread->pid,
-					target_thread->return_error);
+				WARN(1, "Unexpected reply error: %u\n",
+						target_thread->reply_error.cmd);
 			}
+			binder_inner_proc_unlock(target_thread->proc);
+			binder_thread_dec_tmpref(target_thread);
+			binder_free_transaction(t);
 			return;
 		}
 		next = t->from_parent;
@@ -1228,7 +2165,7 @@ static void binder_send_failed_reply(struct binder_transaction *t,
 			     "send failed reply for transaction %d, target dead\n",
 			     t->debug_id);
 
-		binder_pop_transaction(target_thread, t);
+		binder_free_transaction(t);
 		if (next == NULL) {
 			binder_debug(BINDER_DEBUG_DEAD_BINDER,
 				     "reply failed, no target thread at root\n");
@@ -1241,11 +2178,178 @@ static void binder_send_failed_reply(struct binder_transaction *t,
 	}
 }
 
+/**
+ * binder_cleanup_transaction() - cleans up undelivered transaction
+ * @t:		transaction that needs to be cleaned up
+ * @reason:	reason the transaction wasn't delivered
+ * @error_code:	error to return to caller (if synchronous call)
+ */
+static void binder_cleanup_transaction(struct binder_transaction *t,
+				       const char *reason,
+				       uint32_t error_code)
+{
+	if (t->buffer->target_node && !(t->flags & TF_ONE_WAY)) {
+		binder_send_failed_reply(t, error_code);
+	} else {
+		binder_debug(BINDER_DEBUG_DEAD_TRANSACTION,
+			"undelivered transaction %d, %s\n",
+			t->debug_id, reason);
+		binder_free_transaction(t);
+	}
+}
+
+/**
+ * binder_validate_object() - checks for a valid metadata object in a buffer.
+ * @buffer:	binder_buffer that we're parsing.
+ * @offset:	offset in the buffer at which to validate an object.
+ *
+ * Return:	If there's a valid metadata object at @offset in @buffer, the
+ *		size of that object. Otherwise, it returns zero.
+ */
+static size_t binder_validate_object(struct binder_buffer *buffer, u64 offset)
+{
+	/* Check if we can read a header first */
+	struct binder_object_header *hdr;
+	size_t object_size = 0;
+
+	if (offset > buffer->data_size - sizeof(*hdr) ||
+	    buffer->data_size < sizeof(*hdr) ||
+	    !IS_ALIGNED(offset, sizeof(u32)))
+		return 0;
+
+	/* Ok, now see if we can read a complete object. */
+	hdr = (struct binder_object_header *)(buffer->data + offset);
+	switch (hdr->type) {
+	case BINDER_TYPE_BINDER:
+	case BINDER_TYPE_WEAK_BINDER:
+	case BINDER_TYPE_HANDLE:
+	case BINDER_TYPE_WEAK_HANDLE:
+		object_size = sizeof(struct flat_binder_object);
+		break;
+	case BINDER_TYPE_FD:
+		object_size = sizeof(struct binder_fd_object);
+		break;
+	case BINDER_TYPE_PTR:
+		object_size = sizeof(struct binder_buffer_object);
+		break;
+	case BINDER_TYPE_FDA:
+		object_size = sizeof(struct binder_fd_array_object);
+		break;
+	default:
+		return 0;
+	}
+	if (offset <= buffer->data_size - object_size &&
+	    buffer->data_size >= object_size)
+		return object_size;
+	else
+		return 0;
+}
+
+/**
+ * binder_validate_ptr() - validates binder_buffer_object in a binder_buffer.
+ * @b:		binder_buffer containing the object
+ * @index:	index in offset array at which the binder_buffer_object is
+ *		located
+ * @start:	points to the start of the offset array
+ * @num_valid:	the number of valid offsets in the offset array
+ *
+ * Return:	If @index is within the valid range of the offset array
+ *		described by @start and @num_valid, and if there's a valid
+ *		binder_buffer_object at the offset found in index @index
+ *		of the offset array, that object is returned. Otherwise,
+ *		%NULL is returned.
+ *		Note that the offset found in index @index itself is not
+ *		verified; this function assumes that @num_valid elements
+ *		from @start were previously verified to have valid offsets.
+ */
+static struct binder_buffer_object *binder_validate_ptr(struct binder_buffer *b,
+							binder_size_t index,
+							binder_size_t *start,
+							binder_size_t num_valid)
+{
+	struct binder_buffer_object *buffer_obj;
+	binder_size_t *offp;
+
+	if (index >= num_valid)
+		return NULL;
+
+	offp = start + index;
+	buffer_obj = (struct binder_buffer_object *)(b->data + *offp);
+	if (buffer_obj->hdr.type != BINDER_TYPE_PTR)
+		return NULL;
+
+	return buffer_obj;
+}
+
+/**
+ * binder_validate_fixup() - validates pointer/fd fixups happen in order.
+ * @b:			transaction buffer
+ * @objects_start	start of objects buffer
+ * @buffer:		binder_buffer_object in which to fix up
+ * @offset:		start offset in @buffer to fix up
+ * @last_obj:		last binder_buffer_object that we fixed up in
+ * @last_min_offset:	minimum fixup offset in @last_obj
+ *
+ * Return:		%true if a fixup in buffer @buffer at offset @offset is
+ *			allowed.
+ *
+ * For safety reasons, we only allow fixups inside a buffer to happen
+ * at increasing offsets; additionally, we only allow fixup on the last
+ * buffer object that was verified, or one of its parents.
+ *
+ * Example of what is allowed:
+ *
+ * A
+ *   B (parent = A, offset = 0)
+ *   C (parent = A, offset = 16)
+ *     D (parent = C, offset = 0)
+ *   E (parent = A, offset = 32) // min_offset is 16 (C.parent_offset)
+ *
+ * Examples of what is not allowed:
+ *
+ * Decreasing offsets within the same parent:
+ * A
+ *   C (parent = A, offset = 16)
+ *   B (parent = A, offset = 0) // decreasing offset within A
+ *
+ * Referring to a parent that wasn't the last object or any of its parents:
+ * A
+ *   B (parent = A, offset = 0)
+ *   C (parent = A, offset = 0)
+ *   C (parent = A, offset = 16)
+ *     D (parent = B, offset = 0) // B is not A or any of A's parents
+ */
+static bool binder_validate_fixup(struct binder_buffer *b,
+				  binder_size_t *objects_start,
+				  struct binder_buffer_object *buffer,
+				  binder_size_t fixup_offset,
+				  struct binder_buffer_object *last_obj,
+				  binder_size_t last_min_offset)
+{
+	if (!last_obj) {
+		/* Nothing to fix up in */
+		return false;
+	}
+
+	while (last_obj != buffer) {
+		/*
+		 * Safe to retrieve the parent of last_obj, since it
+		 * was already previously verified by the driver.
+		 */
+		if ((last_obj->flags & BINDER_BUFFER_FLAG_HAS_PARENT) == 0)
+			return false;
+		last_min_offset = last_obj->parent_offset + sizeof(uintptr_t);
+		last_obj = (struct binder_buffer_object *)
+			(b->data + *(objects_start + last_obj->parent));
+	}
+	return (fixup_offset >= last_min_offset);
+}
+
 static void binder_transaction_buffer_release(struct binder_proc *proc,
 					      struct binder_buffer *buffer,
 					      binder_size_t *failed_at)
 {
-	binder_size_t *offp, *off_end;
+	binder_size_t *offp, *off_start, *off_end;
 	int debug_id = buffer->debug_id;
 
 	binder_debug(BINDER_DEBUG_TRANSACTION,
@@ -1256,28 +2360,30 @@ static void binder_transaction_buffer_release(struct binder_proc *proc,
 	if (buffer->target_node)
 		binder_dec_node(buffer->target_node, 1, 0);
 
-	offp = (binder_size_t *)(buffer->data +
-				 ALIGN(buffer->data_size, sizeof(void *)));
+	off_start = (binder_size_t *)(buffer->data +
+				      ALIGN(buffer->data_size, sizeof(void *)));
 	if (failed_at)
 		off_end = failed_at;
 	else
-		off_end = (void *)offp + buffer->offsets_size;
-	for (; offp < off_end; offp++) {
-		struct flat_binder_object *fp;
+		off_end = (void *)off_start + buffer->offsets_size;
+	for (offp = off_start; offp < off_end; offp++) {
+		struct binder_object_header *hdr;
+		size_t object_size = binder_validate_object(buffer, *offp);
 
-		if (*offp > buffer->data_size - sizeof(*fp) ||
-		    buffer->data_size < sizeof(*fp) ||
-		    !IS_ALIGNED(*offp, sizeof(u32))) {
-			pr_err("transaction release %d bad offset %lld, size %zd\n",
+		if (object_size == 0) {
+			pr_err("transaction release %d bad object at offset %lld, size %zd\n",
 			       debug_id, (u64)*offp, buffer->data_size);
 			continue;
 		}
-		fp = (struct flat_binder_object *)(buffer->data + *offp);
-		switch (fp->type) {
+		hdr = (struct binder_object_header *)(buffer->data + *offp);
+		switch (hdr->type) {
 		case BINDER_TYPE_BINDER:
 		case BINDER_TYPE_WEAK_BINDER: {
-			struct binder_node *node = binder_get_node(proc, fp->binder);
+			struct flat_binder_object *fp;
+			struct binder_node *node;
 
+			fp = to_flat_binder_object(hdr);
+			node = binder_get_node(proc, fp->binder);
 			if (node == NULL) {
 				pr_err("transaction release %d bad node %016llx\n",
 				       debug_id, (u64)fp->binder);
@@ -1286,89 +2392,560 @@ static void binder_transaction_buffer_release(struct binder_proc *proc,
 			binder_debug(BINDER_DEBUG_TRANSACTION,
 				     "        node %d u%016llx\n",
 				     node->debug_id, (u64)node->ptr);
-			binder_dec_node(node, fp->type == BINDER_TYPE_BINDER, 0);
+			binder_dec_node(node, hdr->type == BINDER_TYPE_BINDER,
+					0);
+			binder_put_node(node);
 		} break;
 		case BINDER_TYPE_HANDLE:
 		case BINDER_TYPE_WEAK_HANDLE: {
-			struct binder_ref *ref;
+			struct flat_binder_object *fp;
+			struct binder_ref_data rdata;
+			int ret;
 
-			ref = binder_get_ref(proc, fp->handle,
-					     fp->type == BINDER_TYPE_HANDLE);
+			fp = to_flat_binder_object(hdr);
+			ret = binder_dec_ref_for_handle(proc, fp->handle,
+				hdr->type == BINDER_TYPE_HANDLE, &rdata);
 
-			if (ref == NULL) {
-				pr_err("transaction release %d bad handle %d\n",
-				 debug_id, fp->handle);
+			if (ret) {
+				pr_err("transaction release %d bad handle %d, ret = %d\n",
+				 debug_id, fp->handle, ret);
 				break;
 			}
 			binder_debug(BINDER_DEBUG_TRANSACTION,
-				     "        ref %d desc %d (node %d)\n",
-				     ref->debug_id, ref->desc, ref->node->debug_id);
-			binder_dec_ref(ref, fp->type == BINDER_TYPE_HANDLE);
+				     "        ref %d desc %d\n",
+				     rdata.debug_id, rdata.desc);
 		} break;
 
-		case BINDER_TYPE_FD:
+		case BINDER_TYPE_FD: {
+			struct binder_fd_object *fp = to_binder_fd_object(hdr);
+
 			binder_debug(BINDER_DEBUG_TRANSACTION,
-				     "        fd %d\n", fp->handle);
+				     "        fd %d\n", fp->fd);
 			if (failed_at)
-				task_close_fd(proc, fp->handle);
+				task_close_fd(proc, fp->fd);
+		} break;
+		case BINDER_TYPE_PTR:
+			/*
+			 * Nothing to do here, this will get cleaned up when the
+			 * transaction buffer gets freed
+			 */
 			break;
-
+		case BINDER_TYPE_FDA: {
+			struct binder_fd_array_object *fda;
+			struct binder_buffer_object *parent;
+			uintptr_t parent_buffer;
+			u32 *fd_array;
+			size_t fd_index;
+			binder_size_t fd_buf_size;
+
+			fda = to_binder_fd_array_object(hdr);
+			parent = binder_validate_ptr(buffer, fda->parent,
+						     off_start,
+						     offp - off_start);
+			if (!parent) {
+				pr_err("transaction release %d bad parent offset",
+				       debug_id);
+				continue;
+			}
+			/*
+			 * Since the parent was already fixed up, convert it
+			 * back to kernel address space to access it
+			 */
+			parent_buffer = parent->buffer -
+				binder_alloc_get_user_buffer_offset(
+						&proc->alloc);
+
+			fd_buf_size = sizeof(u32) * fda->num_fds;
+			if (fda->num_fds >= SIZE_MAX / sizeof(u32)) {
+				pr_err("transaction release %d invalid number of fds (%lld)\n",
+				       debug_id, (u64)fda->num_fds);
+				continue;
+			}
+			if (fd_buf_size > parent->length ||
+			    fda->parent_offset > parent->length - fd_buf_size) {
+				/* No space for all file descriptors here. */
+				pr_err("transaction release %d not enough space for %lld fds in buffer\n",
+				       debug_id, (u64)fda->num_fds);
+				continue;
+			}
+			fd_array = (u32 *)(parent_buffer + fda->parent_offset);
+			for (fd_index = 0; fd_index < fda->num_fds; fd_index++)
+				task_close_fd(proc, fd_array[fd_index]);
+		} break;
 		default:
 			pr_err("transaction release %d bad object type %x\n",
-				debug_id, fp->type);
+				debug_id, hdr->type);
 			break;
 		}
 	}
 }
 
+static int binder_translate_binder(struct flat_binder_object *fp,
+				   struct binder_transaction *t,
+				   struct binder_thread *thread)
+{
+	struct binder_node *node;
+	struct binder_proc *proc = thread->proc;
+	struct binder_proc *target_proc = t->to_proc;
+	struct binder_ref_data rdata;
+	int ret = 0;
+
+	node = binder_get_node(proc, fp->binder);
+	if (!node) {
+		node = binder_new_node(proc, fp);
+		if (!node)
+			return -ENOMEM;
+	}
+	if (fp->cookie != node->cookie) {
+		binder_user_error("%d:%d sending u%016llx node %d, cookie mismatch %016llx != %016llx\n",
+				  proc->pid, thread->pid, (u64)fp->binder,
+				  node->debug_id, (u64)fp->cookie,
+				  (u64)node->cookie);
+		ret = -EINVAL;
+		goto done;
+	}
+	if (security_binder_transfer_binder(proc->tsk, target_proc->tsk)) {
+		ret = -EPERM;
+		goto done;
+	}
+
+	ret = binder_inc_ref_for_node(target_proc, node,
+			fp->hdr.type == BINDER_TYPE_BINDER,
+			&thread->todo, &rdata);
+	if (ret)
+		goto done;
+
+	if (fp->hdr.type == BINDER_TYPE_BINDER)
+		fp->hdr.type = BINDER_TYPE_HANDLE;
+	else
+		fp->hdr.type = BINDER_TYPE_WEAK_HANDLE;
+	fp->binder = 0;
+	fp->handle = rdata.desc;
+	fp->cookie = 0;
+
+	trace_binder_transaction_node_to_ref(t, node, &rdata);
+	binder_debug(BINDER_DEBUG_TRANSACTION,
+		     "        node %d u%016llx -> ref %d desc %d\n",
+		     node->debug_id, (u64)node->ptr,
+		     rdata.debug_id, rdata.desc);
+done:
+	binder_put_node(node);
+	return ret;
+}
+
+static int binder_translate_handle(struct flat_binder_object *fp,
+				   struct binder_transaction *t,
+				   struct binder_thread *thread)
+{
+	struct binder_proc *proc = thread->proc;
+	struct binder_proc *target_proc = t->to_proc;
+	struct binder_node *node;
+	struct binder_ref_data src_rdata;
+	int ret = 0;
+
+	node = binder_get_node_from_ref(proc, fp->handle,
+			fp->hdr.type == BINDER_TYPE_HANDLE, &src_rdata);
+	if (!node) {
+		binder_user_error("%d:%d got transaction with invalid handle, %d\n",
+				  proc->pid, thread->pid, fp->handle);
+		return -EINVAL;
+	}
+	if (security_binder_transfer_binder(proc->tsk, target_proc->tsk)) {
+		ret = -EPERM;
+		goto done;
+	}
+
+	binder_node_lock(node);
+	if (node->proc == target_proc) {
+		if (fp->hdr.type == BINDER_TYPE_HANDLE)
+			fp->hdr.type = BINDER_TYPE_BINDER;
+		else
+			fp->hdr.type = BINDER_TYPE_WEAK_BINDER;
+		fp->binder = node->ptr;
+		fp->cookie = node->cookie;
+		if (node->proc)
+			binder_inner_proc_lock(node->proc);
+		binder_inc_node_nilocked(node,
+					 fp->hdr.type == BINDER_TYPE_BINDER,
+					 0, NULL);
+		if (node->proc)
+			binder_inner_proc_unlock(node->proc);
+		trace_binder_transaction_ref_to_node(t, node, &src_rdata);
+		binder_debug(BINDER_DEBUG_TRANSACTION,
+			     "        ref %d desc %d -> node %d u%016llx\n",
+			     src_rdata.debug_id, src_rdata.desc, node->debug_id,
+			     (u64)node->ptr);
+		binder_node_unlock(node);
+	} else {
+		struct binder_ref_data dest_rdata;
+
+		binder_node_unlock(node);
+		ret = binder_inc_ref_for_node(target_proc, node,
+				fp->hdr.type == BINDER_TYPE_HANDLE,
+				NULL, &dest_rdata);
+		if (ret)
+			goto done;
+
+		fp->binder = 0;
+		fp->handle = dest_rdata.desc;
+		fp->cookie = 0;
+		trace_binder_transaction_ref_to_ref(t, node, &src_rdata,
+						    &dest_rdata);
+		binder_debug(BINDER_DEBUG_TRANSACTION,
+			     "        ref %d desc %d -> ref %d desc %d (node %d)\n",
+			     src_rdata.debug_id, src_rdata.desc,
+			     dest_rdata.debug_id, dest_rdata.desc,
+			     node->debug_id);
+	}
+done:
+	binder_put_node(node);
+	return ret;
+}
+
+static int binder_translate_fd(int fd,
+			       struct binder_transaction *t,
+			       struct binder_thread *thread,
+			       struct binder_transaction *in_reply_to)
+{
+	struct binder_proc *proc = thread->proc;
+	struct binder_proc *target_proc = t->to_proc;
+	int target_fd;
+	struct file *file;
+	int ret;
+	bool target_allows_fd;
+
+	if (in_reply_to)
+		target_allows_fd = !!(in_reply_to->flags & TF_ACCEPT_FDS);
+	else
+		target_allows_fd = t->buffer->target_node->accept_fds;
+	if (!target_allows_fd) {
+		binder_user_error("%d:%d got %s with fd, %d, but target does not allow fds\n",
+				  proc->pid, thread->pid,
+				  in_reply_to ? "reply" : "transaction",
+				  fd);
+		ret = -EPERM;
+		goto err_fd_not_accepted;
+	}
+
+	file = fget(fd);
+	if (!file) {
+		binder_user_error("%d:%d got transaction with invalid fd, %d\n",
+				  proc->pid, thread->pid, fd);
+		ret = -EBADF;
+		goto err_fget;
+	}
+	ret = security_binder_transfer_file(proc->tsk, target_proc->tsk, file);
+	if (ret < 0) {
+		ret = -EPERM;
+		goto err_security;
+	}
+
+	target_fd = task_get_unused_fd_flags(target_proc, O_CLOEXEC);
+	if (target_fd < 0) {
+		ret = -ENOMEM;
+		goto err_get_unused_fd;
+	}
+	task_fd_install(target_proc, target_fd, file);
+	trace_binder_transaction_fd(t, fd, target_fd);
+	binder_debug(BINDER_DEBUG_TRANSACTION, "        fd %d -> %d\n",
+		     fd, target_fd);
+
+	return target_fd;
+
+err_get_unused_fd:
+err_security:
+	fput(file);
+err_fget:
+err_fd_not_accepted:
+	return ret;
+}
+
+static int binder_translate_fd_array(struct binder_fd_array_object *fda,
+				     struct binder_buffer_object *parent,
+				     struct binder_transaction *t,
+				     struct binder_thread *thread,
+				     struct binder_transaction *in_reply_to)
+{
+	binder_size_t fdi, fd_buf_size, num_installed_fds;
+	int target_fd;
+	uintptr_t parent_buffer;
+	u32 *fd_array;
+	struct binder_proc *proc = thread->proc;
+	struct binder_proc *target_proc = t->to_proc;
+
+	fd_buf_size = sizeof(u32) * fda->num_fds;
+	if (fda->num_fds >= SIZE_MAX / sizeof(u32)) {
+		binder_user_error("%d:%d got transaction with invalid number of fds (%lld)\n",
+				  proc->pid, thread->pid, (u64)fda->num_fds);
+		return -EINVAL;
+	}
+	if (fd_buf_size > parent->length ||
+	    fda->parent_offset > parent->length - fd_buf_size) {
+		/* No space for all file descriptors here. */
+		binder_user_error("%d:%d not enough space to store %lld fds in buffer\n",
+				  proc->pid, thread->pid, (u64)fda->num_fds);
+		return -EINVAL;
+	}
+	/*
+	 * Since the parent was already fixed up, convert it
+	 * back to the kernel address space to access it
+	 */
+	parent_buffer = parent->buffer -
+		binder_alloc_get_user_buffer_offset(&target_proc->alloc);
+	fd_array = (u32 *)(parent_buffer + fda->parent_offset);
+	if (!IS_ALIGNED((unsigned long)fd_array, sizeof(u32))) {
+		binder_user_error("%d:%d parent offset not aligned correctly.\n",
+				  proc->pid, thread->pid);
+		return -EINVAL;
+	}
+	for (fdi = 0; fdi < fda->num_fds; fdi++) {
+		target_fd = binder_translate_fd(fd_array[fdi], t, thread,
+						in_reply_to);
+		if (target_fd < 0)
+			goto err_translate_fd_failed;
+		fd_array[fdi] = target_fd;
+	}
+	return 0;
+
+err_translate_fd_failed:
+	/*
+	 * Failed to allocate fd or security error, free fds
+	 * installed so far.
+	 */
+	num_installed_fds = fdi;
+	for (fdi = 0; fdi < num_installed_fds; fdi++)
+		task_close_fd(target_proc, fd_array[fdi]);
+	return target_fd;
+}
+
+static int binder_fixup_parent(struct binder_transaction *t,
+			       struct binder_thread *thread,
+			       struct binder_buffer_object *bp,
+			       binder_size_t *off_start,
+			       binder_size_t num_valid,
+			       struct binder_buffer_object *last_fixup_obj,
+			       binder_size_t last_fixup_min_off)
+{
+	struct binder_buffer_object *parent;
+	u8 *parent_buffer;
+	struct binder_buffer *b = t->buffer;
+	struct binder_proc *proc = thread->proc;
+	struct binder_proc *target_proc = t->to_proc;
+
+	if (!(bp->flags & BINDER_BUFFER_FLAG_HAS_PARENT))
+		return 0;
+
+	parent = binder_validate_ptr(b, bp->parent, off_start, num_valid);
+	if (!parent) {
+		binder_user_error("%d:%d got transaction with invalid parent offset or type\n",
+				  proc->pid, thread->pid);
+		return -EINVAL;
+	}
+
+	if (!binder_validate_fixup(b, off_start,
+				   parent, bp->parent_offset,
+				   last_fixup_obj,
+				   last_fixup_min_off)) {
+		binder_user_error("%d:%d got transaction with out-of-order buffer fixup\n",
+				  proc->pid, thread->pid);
+		return -EINVAL;
+	}
+
+	if (parent->length < sizeof(binder_uintptr_t) ||
+	    bp->parent_offset > parent->length - sizeof(binder_uintptr_t)) {
+		/* No space for a pointer here! */
+		binder_user_error("%d:%d got transaction with invalid parent offset\n",
+				  proc->pid, thread->pid);
+		return -EINVAL;
+	}
+	parent_buffer = (u8 *)(parent->buffer -
+			binder_alloc_get_user_buffer_offset(
+				&target_proc->alloc));
+	*(binder_uintptr_t *)(parent_buffer + bp->parent_offset) = bp->buffer;
+
+	return 0;
+}
+
+/**
+ * binder_proc_transaction() - sends a transaction to a process and wakes it up
+ * @t:		transaction to send
+ * @proc:	process to send the transaction to
+ * @thread:	thread in @proc to send the transaction to (may be NULL)
+ *
+ * This function queues a transaction to the specified process. It will try
+ * to find a thread in the target process to handle the transaction and
+ * wake it up. If no thread is found, the work is queued to the proc
+ * waitqueue.
+ *
+ * If the @thread parameter is not NULL, the transaction is always queued
+ * to the waitlist of that specific thread.
+ *
+ * Return:	true if the transactions was successfully queued
+ *		false if the target process or thread is dead
+ */
+static bool binder_proc_transaction(struct binder_transaction *t,
+				    struct binder_proc *proc,
+				    struct binder_thread *thread)
+{
+	struct binder_node *node = t->buffer->target_node;
+	struct binder_priority node_prio;
+	bool oneway = !!(t->flags & TF_ONE_WAY);
+	bool pending_async = false;
+
+	BUG_ON(!node);
+	binder_node_lock(node);
+	node_prio.prio = node->min_priority;
+	node_prio.sched_policy = node->sched_policy;
+
+	if (oneway) {
+		BUG_ON(thread);
+		if (node->has_async_transaction) {
+			pending_async = true;
+		} else {
+			node->has_async_transaction = 1;
+		}
+	}
+
+	binder_inner_proc_lock(proc);
+
+	if (proc->is_dead || (thread && thread->is_dead)) {
+		binder_inner_proc_unlock(proc);
+		binder_node_unlock(node);
+		return false;
+	}
+
+	if (!thread && !pending_async)
+		thread = binder_select_thread_ilocked(proc);
+
+	if (thread) {
+		binder_transaction_priority(thread->task, t, node_prio,
+					    node->inherit_rt);
+		binder_enqueue_thread_work_ilocked(thread, &t->work);
+	} else if (!pending_async) {
+		binder_enqueue_work_ilocked(&t->work, &proc->todo);
+	} else {
+		binder_enqueue_work_ilocked(&t->work, &node->async_todo);
+	}
+
+	if (!pending_async)
+		binder_wakeup_thread_ilocked(proc, thread, !oneway /* sync */);
+
+	binder_inner_proc_unlock(proc);
+	binder_node_unlock(node);
+
+	return true;
+}
+
+/**
+ * binder_get_node_refs_for_txn() - Get required refs on node for txn
+ * @node:         struct binder_node for which to get refs
+ * @proc:         returns @node->proc if valid
+ * @error:        if no @proc then returns BR_DEAD_REPLY
+ *
+ * User-space normally keeps the node alive when creating a transaction
+ * since it has a reference to the target. The local strong ref keeps it
+ * alive if the sending process dies before the target process processes
+ * the transaction. If the source process is malicious or has a reference
+ * counting bug, relying on the local strong ref can fail.
+ *
+ * Since user-space can cause the local strong ref to go away, we also take
+ * a tmpref on the node to ensure it survives while we are constructing
+ * the transaction. We also need a tmpref on the proc while we are
+ * constructing the transaction, so we take that here as well.
+ *
+ * Return: The target_node with refs taken or NULL if no @node->proc is NULL.
+ * Also sets @proc if valid. If the @node->proc is NULL indicating that the
+ * target proc has died, @error is set to BR_DEAD_REPLY
+ */
+static struct binder_node *binder_get_node_refs_for_txn(
+		struct binder_node *node,
+		struct binder_proc **procp,
+		uint32_t *error)
+{
+	struct binder_node *target_node = NULL;
+
+	binder_node_inner_lock(node);
+	if (node->proc) {
+		target_node = node;
+		binder_inc_node_nilocked(node, 1, 0, NULL);
+		binder_inc_node_tmpref_ilocked(node);
+		node->proc->tmp_ref++;
+		*procp = node->proc;
+	} else
+		*error = BR_DEAD_REPLY;
+	binder_node_inner_unlock(node);
+
+	return target_node;
+}
+
 static void binder_transaction(struct binder_proc *proc,
 			       struct binder_thread *thread,
-			       struct binder_transaction_data *tr, int reply)
+			       struct binder_transaction_data *tr, int reply,
+			       binder_size_t extra_buffers_size)
 {
+	int ret;
 	struct binder_transaction *t;
 	struct binder_work *tcomplete;
-	binder_size_t *offp, *off_end;
-	struct binder_proc *target_proc;
+	binder_size_t *offp, *off_end, *off_start;
+	binder_size_t off_min;
+	u8 *sg_bufp, *sg_buf_end;
+	struct binder_proc *target_proc = NULL;
 	struct binder_thread *target_thread = NULL;
 	struct binder_node *target_node = NULL;
-	struct list_head *target_list;
-	wait_queue_head_t *target_wait;
 	struct binder_transaction *in_reply_to = NULL;
 	struct binder_transaction_log_entry *e;
-	uint32_t return_error;
+	uint32_t return_error = 0;
+	uint32_t return_error_param = 0;
+	uint32_t return_error_line = 0;
+	struct binder_buffer_object *last_fixup_obj = NULL;
+	binder_size_t last_fixup_min_off = 0;
+	struct binder_context *context = proc->context;
+	int t_debug_id = atomic_inc_return(&binder_last_id);
 
 	e = binder_transaction_log_add(&binder_transaction_log);
+	e->debug_id = t_debug_id;
 	e->call_type = reply ? 2 : !!(tr->flags & TF_ONE_WAY);
 	e->from_proc = proc->pid;
 	e->from_thread = thread->pid;
 	e->target_handle = tr->target.handle;
 	e->data_size = tr->data_size;
 	e->offsets_size = tr->offsets_size;
+	e->context_name = proc->context->name;
 
 	if (reply) {
+		binder_inner_proc_lock(proc);
 		in_reply_to = thread->transaction_stack;
 		if (in_reply_to == NULL) {
+			binder_inner_proc_unlock(proc);
 			binder_user_error("%d:%d got reply transaction with no transaction stack\n",
 					  proc->pid, thread->pid);
 			return_error = BR_FAILED_REPLY;
+			return_error_param = -EPROTO;
+			return_error_line = __LINE__;
 			goto err_empty_call_stack;
 		}
-		binder_set_nice(in_reply_to->saved_priority);
 		if (in_reply_to->to_thread != thread) {
+			spin_lock(&in_reply_to->lock);
 			binder_user_error("%d:%d got reply transaction with bad transaction stack, transaction %d has target %d:%d\n",
 				proc->pid, thread->pid, in_reply_to->debug_id,
 				in_reply_to->to_proc ?
 				in_reply_to->to_proc->pid : 0,
 				in_reply_to->to_thread ?
 				in_reply_to->to_thread->pid : 0);
+			spin_unlock(&in_reply_to->lock);
+			binder_inner_proc_unlock(proc);
 			return_error = BR_FAILED_REPLY;
+			return_error_param = -EPROTO;
+			return_error_line = __LINE__;
 			in_reply_to = NULL;
 			goto err_bad_call_stack;
 		}
 		thread->transaction_stack = in_reply_to->to_parent;
-		target_thread = in_reply_to->from;
+		binder_inner_proc_unlock(proc);
+		target_thread = binder_get_txn_from_and_acq_inner(in_reply_to);
 		if (target_thread == NULL) {
 			return_error = BR_DEAD_REPLY;
+			return_error_line = __LINE__;
 			goto err_dead_binder;
 		}
 		if (target_thread->transaction_stack != in_reply_to) {
@@ -1377,106 +2954,148 @@ static void binder_transaction(struct binder_proc *proc,
 				target_thread->transaction_stack ?
 				target_thread->transaction_stack->debug_id : 0,
 				in_reply_to->debug_id);
+			binder_inner_proc_unlock(target_thread->proc);
 			return_error = BR_FAILED_REPLY;
+			return_error_param = -EPROTO;
+			return_error_line = __LINE__;
 			in_reply_to = NULL;
 			target_thread = NULL;
 			goto err_dead_binder;
 		}
 		target_proc = target_thread->proc;
+		target_proc->tmp_ref++;
+		binder_inner_proc_unlock(target_thread->proc);
 	} else {
 		if (tr->target.handle) {
 			struct binder_ref *ref;
 
-			ref = binder_get_ref(proc, tr->target.handle, true);
-			if (ref == NULL) {
+			/*
+			 * There must already be a strong ref
+			 * on this node. If so, do a strong
+			 * increment on the node to ensure it
+			 * stays alive until the transaction is
+			 * done.
+			 */
+			binder_proc_lock(proc);
+			ref = binder_get_ref_olocked(proc, tr->target.handle,
+						     true);
+			if (ref) {
+				target_node = binder_get_node_refs_for_txn(
+						ref->node, &target_proc,
+						&return_error);
+			} else {
 				binder_user_error("%d:%d got transaction to invalid handle\n",
-					proc->pid, thread->pid);
+						  proc->pid, thread->pid);
 				return_error = BR_FAILED_REPLY;
-				goto err_invalid_target_handle;
 			}
-			target_node = ref->node;
+			binder_proc_unlock(proc);
 		} else {
-			target_node = binder_context_mgr_node;
-			if (target_node == NULL) {
+			mutex_lock(&context->context_mgr_node_lock);
+			target_node = context->binder_context_mgr_node;
+			if (target_node)
+				target_node = binder_get_node_refs_for_txn(
+						target_node, &target_proc,
+						&return_error);
+			else
 				return_error = BR_DEAD_REPLY;
-				goto err_no_context_mgr_node;
-			}
+			mutex_unlock(&context->context_mgr_node_lock);
 		}
-		e->to_node = target_node->debug_id;
-		target_proc = target_node->proc;
-		if (target_proc == NULL) {
-			return_error = BR_DEAD_REPLY;
+		if (!target_node) {
+			/*
+			 * return_error is set above
+			 */
+			return_error_param = -EINVAL;
+			return_error_line = __LINE__;
 			goto err_dead_binder;
 		}
+		e->to_node = target_node->debug_id;
 		if (security_binder_transaction(proc->tsk,
 						target_proc->tsk) < 0) {
 			return_error = BR_FAILED_REPLY;
+			return_error_param = -EPERM;
+			return_error_line = __LINE__;
 			goto err_invalid_target_handle;
 		}
+		binder_inner_proc_lock(proc);
 		if (!(tr->flags & TF_ONE_WAY) && thread->transaction_stack) {
 			struct binder_transaction *tmp;
 
 			tmp = thread->transaction_stack;
 			if (tmp->to_thread != thread) {
+				spin_lock(&tmp->lock);
 				binder_user_error("%d:%d got new transaction with bad transaction stack, transaction %d has target %d:%d\n",
 					proc->pid, thread->pid, tmp->debug_id,
 					tmp->to_proc ? tmp->to_proc->pid : 0,
 					tmp->to_thread ?
 					tmp->to_thread->pid : 0);
+				spin_unlock(&tmp->lock);
+				binder_inner_proc_unlock(proc);
 				return_error = BR_FAILED_REPLY;
+				return_error_param = -EPROTO;
+				return_error_line = __LINE__;
 				goto err_bad_call_stack;
 			}
 			while (tmp) {
-				if (tmp->from && tmp->from->proc == target_proc)
-					target_thread = tmp->from;
+				struct binder_thread *from;
+
+				spin_lock(&tmp->lock);
+				from = tmp->from;
+				if (from && from->proc == target_proc) {
+					atomic_inc(&from->tmp_ref);
+					target_thread = from;
+					spin_unlock(&tmp->lock);
+					break;
+				}
+				spin_unlock(&tmp->lock);
 				tmp = tmp->from_parent;
 			}
 		}
+		binder_inner_proc_unlock(proc);
 	}
-	if (target_thread) {
+	if (target_thread)
 		e->to_thread = target_thread->pid;
-		target_list = &target_thread->todo;
-		target_wait = &target_thread->wait;
-	} else {
-		target_list = &target_proc->todo;
-		target_wait = &target_proc->wait;
-	}
 	e->to_proc = target_proc->pid;
 
 	/* TODO: reuse incoming transaction for reply */
 	t = kzalloc(sizeof(*t), GFP_KERNEL);
 	if (t == NULL) {
 		return_error = BR_FAILED_REPLY;
+		return_error_param = -ENOMEM;
+		return_error_line = __LINE__;
 		goto err_alloc_t_failed;
 	}
 	binder_stats_created(BINDER_STAT_TRANSACTION);
+	spin_lock_init(&t->lock);
 
 	tcomplete = kzalloc(sizeof(*tcomplete), GFP_KERNEL);
 	if (tcomplete == NULL) {
 		return_error = BR_FAILED_REPLY;
+		return_error_param = -ENOMEM;
+		return_error_line = __LINE__;
 		goto err_alloc_tcomplete_failed;
 	}
 	binder_stats_created(BINDER_STAT_TRANSACTION_COMPLETE);
 
-	t->debug_id = ++binder_last_id;
-	e->debug_id = t->debug_id;
+	t->debug_id = t_debug_id;
 
 	if (reply)
 		binder_debug(BINDER_DEBUG_TRANSACTION,
-			     "%d:%d BC_REPLY %d -> %d:%d, data %016llx-%016llx size %lld-%lld\n",
+			     "%d:%d BC_REPLY %d -> %d:%d, data %016llx-%016llx size %lld-%lld-%lld\n",
 			     proc->pid, thread->pid, t->debug_id,
 			     target_proc->pid, target_thread->pid,
 			     (u64)tr->data.ptr.buffer,
 			     (u64)tr->data.ptr.offsets,
-			     (u64)tr->data_size, (u64)tr->offsets_size);
+			     (u64)tr->data_size, (u64)tr->offsets_size,
+			     (u64)extra_buffers_size);
 	else
 		binder_debug(BINDER_DEBUG_TRANSACTION,
-			     "%d:%d BC_TRANSACTION %d -> %d - node %d, data %016llx-%016llx size %lld-%lld\n",
+			     "%d:%d BC_TRANSACTION %d -> %d - node %d, data %016llx-%016llx size %lld-%lld-%lld\n",
 			     proc->pid, thread->pid, t->debug_id,
 			     target_proc->pid, target_node->debug_id,
 			     (u64)tr->data.ptr.buffer,
 			     (u64)tr->data.ptr.offsets,
-			     (u64)tr->data_size, (u64)tr->offsets_size);
+			     (u64)tr->data_size, (u64)tr->offsets_size,
+			     (u64)extra_buffers_size);
 
 	if (!reply && !(tr->flags & TF_ONE_WAY))
 		t->from = thread;
@@ -1487,14 +3106,30 @@ static void binder_transaction(struct binder_proc *proc,
 	t->to_thread = target_thread;
 	t->code = tr->code;
 	t->flags = tr->flags;
-	t->priority = task_nice(current);
+	if (!(t->flags & TF_ONE_WAY) &&
+	    binder_supported_policy(current->policy)) {
+		/* Inherit supported policies for synchronous transactions */
+		t->priority.sched_policy = current->policy;
+		t->priority.prio = current->normal_prio;
+	} else {
+		/* Otherwise, fall back to the default priority */
+		t->priority = target_proc->default_priority;
+	}
 
 	trace_binder_transaction(reply, t, target_node);
 
-	t->buffer = binder_alloc_buf(target_proc, tr->data_size,
-		tr->offsets_size, !reply && (t->flags & TF_ONE_WAY));
-	if (t->buffer == NULL) {
-		return_error = BR_FAILED_REPLY;
+	t->buffer = binder_alloc_new_buf(&target_proc->alloc, tr->data_size,
+		tr->offsets_size, extra_buffers_size,
+		!reply && (t->flags & TF_ONE_WAY));
+	if (IS_ERR(t->buffer)) {
+		/*
+		 * -ESRCH indicates VMA cleared. The target is dying.
+		 */
+		return_error_param = PTR_ERR(t->buffer);
+		return_error = return_error_param == -ESRCH ?
+			BR_DEAD_REPLY : BR_FAILED_REPLY;
+		return_error_line = __LINE__;
+		t->buffer = NULL;
 		goto err_binder_alloc_buf_failed;
 	}
 	t->buffer->allow_user_free = 0;
@@ -1502,17 +3137,17 @@ static void binder_transaction(struct binder_proc *proc,
 	t->buffer->transaction = t;
 	t->buffer->target_node = target_node;
 	trace_binder_transaction_alloc_buf(t->buffer);
-	if (target_node)
-		binder_inc_node(target_node, 1, 0, NULL);
-
-	offp = (binder_size_t *)(t->buffer->data +
-				 ALIGN(tr->data_size, sizeof(void *)));
+	off_start = (binder_size_t *)(t->buffer->data +
+				      ALIGN(tr->data_size, sizeof(void *)));
+	offp = off_start;
 
 	if (copy_from_user(t->buffer->data, (const void __user *)(uintptr_t)
 			   tr->data.ptr.buffer, tr->data_size)) {
 		binder_user_error("%d:%d got transaction with invalid data ptr\n",
 				proc->pid, thread->pid);
 		return_error = BR_FAILED_REPLY;
+		return_error_param = -EFAULT;
+		return_error_line = __LINE__;
 		goto err_copy_data_failed;
 	}
 	if (copy_from_user(offp, (const void __user *)(uintptr_t)
@@ -1520,225 +3155,253 @@ static void binder_transaction(struct binder_proc *proc,
 		binder_user_error("%d:%d got transaction with invalid offsets ptr\n",
 				proc->pid, thread->pid);
 		return_error = BR_FAILED_REPLY;
+		return_error_param = -EFAULT;
+		return_error_line = __LINE__;
 		goto err_copy_data_failed;
 	}
 	if (!IS_ALIGNED(tr->offsets_size, sizeof(binder_size_t))) {
 		binder_user_error("%d:%d got transaction with invalid offsets size, %lld\n",
 				proc->pid, thread->pid, (u64)tr->offsets_size);
 		return_error = BR_FAILED_REPLY;
+		return_error_param = -EINVAL;
+		return_error_line = __LINE__;
+		goto err_bad_offset;
+	}
+	if (!IS_ALIGNED(extra_buffers_size, sizeof(u64))) {
+		binder_user_error("%d:%d got transaction with unaligned buffers size, %lld\n",
+				  proc->pid, thread->pid,
+				  (u64)extra_buffers_size);
+		return_error = BR_FAILED_REPLY;
+		return_error_param = -EINVAL;
+		return_error_line = __LINE__;
 		goto err_bad_offset;
 	}
-	off_end = (void *)offp + tr->offsets_size;
+	off_end = (void *)off_start + tr->offsets_size;
+	sg_bufp = (u8 *)(PTR_ALIGN(off_end, sizeof(void *)));
+	sg_buf_end = sg_bufp + extra_buffers_size;
+	off_min = 0;
 	for (; offp < off_end; offp++) {
-		struct flat_binder_object *fp;
-
-		if (*offp > t->buffer->data_size - sizeof(*fp) ||
-		    t->buffer->data_size < sizeof(*fp) ||
-		    !IS_ALIGNED(*offp, sizeof(u32))) {
-			binder_user_error("%d:%d got transaction with invalid offset, %lld\n",
-					  proc->pid, thread->pid, (u64)*offp);
+		struct binder_object_header *hdr;
+		size_t object_size = binder_validate_object(t->buffer, *offp);
+
+		if (object_size == 0 || *offp < off_min) {
+			binder_user_error("%d:%d got transaction with invalid offset (%lld, min %lld max %lld) or object.\n",
+					  proc->pid, thread->pid, (u64)*offp,
+					  (u64)off_min,
+					  (u64)t->buffer->data_size);
 			return_error = BR_FAILED_REPLY;
+			return_error_param = -EINVAL;
+			return_error_line = __LINE__;
 			goto err_bad_offset;
 		}
-		fp = (struct flat_binder_object *)(t->buffer->data + *offp);
-		switch (fp->type) {
+
+		hdr = (struct binder_object_header *)(t->buffer->data + *offp);
+		off_min = *offp + object_size;
+		switch (hdr->type) {
 		case BINDER_TYPE_BINDER:
 		case BINDER_TYPE_WEAK_BINDER: {
-			struct binder_ref *ref;
-			struct binder_node *node = binder_get_node(proc, fp->binder);
+			struct flat_binder_object *fp;
 
-			if (node == NULL) {
-				node = binder_new_node(proc, fp->binder, fp->cookie);
-				if (node == NULL) {
-					return_error = BR_FAILED_REPLY;
-					goto err_binder_new_node_failed;
-				}
-				node->min_priority = fp->flags & FLAT_BINDER_FLAG_PRIORITY_MASK;
-				node->accept_fds = !!(fp->flags & FLAT_BINDER_FLAG_ACCEPTS_FDS);
-			}
-			if (fp->cookie != node->cookie) {
-				binder_user_error("%d:%d sending u%016llx node %d, cookie mismatch %016llx != %016llx\n",
-					proc->pid, thread->pid,
-					(u64)fp->binder, node->debug_id,
-					(u64)fp->cookie, (u64)node->cookie);
-				return_error = BR_FAILED_REPLY;
-				goto err_binder_get_ref_for_node_failed;
-			}
-			if (security_binder_transfer_binder(proc->tsk,
-							    target_proc->tsk)) {
+			fp = to_flat_binder_object(hdr);
+			ret = binder_translate_binder(fp, t, thread);
+			if (ret < 0) {
 				return_error = BR_FAILED_REPLY;
-				goto err_binder_get_ref_for_node_failed;
+				return_error_param = ret;
+				return_error_line = __LINE__;
+				goto err_translate_failed;
 			}
-			ref = binder_get_ref_for_node(target_proc, node);
-			if (ref == NULL) {
-				return_error = BR_FAILED_REPLY;
-				goto err_binder_get_ref_for_node_failed;
-			}
-			if (fp->type == BINDER_TYPE_BINDER)
-				fp->type = BINDER_TYPE_HANDLE;
-			else
-				fp->type = BINDER_TYPE_WEAK_HANDLE;
-			fp->binder = 0;
-			fp->handle = ref->desc;
-			fp->cookie = 0;
-			binder_inc_ref(ref, fp->type == BINDER_TYPE_HANDLE,
-				       &thread->todo);
-
-			trace_binder_transaction_node_to_ref(t, node, ref);
-			binder_debug(BINDER_DEBUG_TRANSACTION,
-				     "        node %d u%016llx -> ref %d desc %d\n",
-				     node->debug_id, (u64)node->ptr,
-				     ref->debug_id, ref->desc);
 		} break;
 		case BINDER_TYPE_HANDLE:
 		case BINDER_TYPE_WEAK_HANDLE: {
-			struct binder_ref *ref;
+			struct flat_binder_object *fp;
 
-			ref = binder_get_ref(proc, fp->handle,
-					     fp->type == BINDER_TYPE_HANDLE);
+			fp = to_flat_binder_object(hdr);
+			ret = binder_translate_handle(fp, t, thread);
+			if (ret < 0) {
+				return_error = BR_FAILED_REPLY;
+				return_error_param = ret;
+				return_error_line = __LINE__;
+				goto err_translate_failed;
+			}
+		} break;
 
-			if (ref == NULL) {
-				binder_user_error("%d:%d got transaction with invalid handle, %d\n",
-						proc->pid,
-						thread->pid, fp->handle);
+		case BINDER_TYPE_FD: {
+			struct binder_fd_object *fp = to_binder_fd_object(hdr);
+			int target_fd = binder_translate_fd(fp->fd, t, thread,
+							    in_reply_to);
+
+			if (target_fd < 0) {
 				return_error = BR_FAILED_REPLY;
-				goto err_binder_get_ref_failed;
+				return_error_param = target_fd;
+				return_error_line = __LINE__;
+				goto err_translate_failed;
 			}
-			if (security_binder_transfer_binder(proc->tsk,
-							    target_proc->tsk)) {
+			fp->pad_binder = 0;
+			fp->fd = target_fd;
+		} break;
+		case BINDER_TYPE_FDA: {
+			struct binder_fd_array_object *fda =
+				to_binder_fd_array_object(hdr);
+			struct binder_buffer_object *parent =
+				binder_validate_ptr(t->buffer, fda->parent,
+						    off_start,
+						    offp - off_start);
+			if (!parent) {
+				binder_user_error("%d:%d got transaction with invalid parent offset or type\n",
+						  proc->pid, thread->pid);
 				return_error = BR_FAILED_REPLY;
-				goto err_binder_get_ref_failed;
+				return_error_param = -EINVAL;
+				return_error_line = __LINE__;
+				goto err_bad_parent;
 			}
-			if (ref->node->proc == target_proc) {
-				if (fp->type == BINDER_TYPE_HANDLE)
-					fp->type = BINDER_TYPE_BINDER;
-				else
-					fp->type = BINDER_TYPE_WEAK_BINDER;
-				fp->binder = ref->node->ptr;
-				fp->cookie = ref->node->cookie;
-				binder_inc_node(ref->node, fp->type == BINDER_TYPE_BINDER, 0, NULL);
-				trace_binder_transaction_ref_to_node(t, ref);
-				binder_debug(BINDER_DEBUG_TRANSACTION,
-					     "        ref %d desc %d -> node %d u%016llx\n",
-					     ref->debug_id, ref->desc, ref->node->debug_id,
-					     (u64)ref->node->ptr);
-			} else {
-				struct binder_ref *new_ref;
-
-				new_ref = binder_get_ref_for_node(target_proc, ref->node);
-				if (new_ref == NULL) {
-					return_error = BR_FAILED_REPLY;
-					goto err_binder_get_ref_for_node_failed;
-				}
-				fp->binder = 0;
-				fp->handle = new_ref->desc;
-				fp->cookie = 0;
-				binder_inc_ref(new_ref, fp->type == BINDER_TYPE_HANDLE, NULL);
-				trace_binder_transaction_ref_to_ref(t, ref,
-								    new_ref);
-				binder_debug(BINDER_DEBUG_TRANSACTION,
-					     "        ref %d desc %d -> ref %d desc %d (node %d)\n",
-					     ref->debug_id, ref->desc, new_ref->debug_id,
-					     new_ref->desc, ref->node->debug_id);
+			if (!binder_validate_fixup(t->buffer, off_start,
+						   parent, fda->parent_offset,
+						   last_fixup_obj,
+						   last_fixup_min_off)) {
+				binder_user_error("%d:%d got transaction with out-of-order buffer fixup\n",
+						  proc->pid, thread->pid);
+				return_error = BR_FAILED_REPLY;
+				return_error_param = -EINVAL;
+				return_error_line = __LINE__;
+				goto err_bad_parent;
 			}
-		} break;
-
-		case BINDER_TYPE_FD: {
-			int target_fd;
-			struct file *file;
-
-			if (reply) {
-				if (!(in_reply_to->flags & TF_ACCEPT_FDS)) {
-					binder_user_error("%d:%d got reply with fd, %d, but target does not allow fds\n",
-						proc->pid, thread->pid, fp->handle);
-					return_error = BR_FAILED_REPLY;
-					goto err_fd_not_allowed;
-				}
-			} else if (!target_node->accept_fds) {
-				binder_user_error("%d:%d got transaction with fd, %d, but target does not allow fds\n",
-					proc->pid, thread->pid, fp->handle);
+			ret = binder_translate_fd_array(fda, parent, t, thread,
+							in_reply_to);
+			if (ret < 0) {
 				return_error = BR_FAILED_REPLY;
-				goto err_fd_not_allowed;
+				return_error_param = ret;
+				return_error_line = __LINE__;
+				goto err_translate_failed;
 			}
-
-			file = fget(fp->handle);
-			if (file == NULL) {
-				binder_user_error("%d:%d got transaction with invalid fd, %d\n",
-					proc->pid, thread->pid, fp->handle);
+			last_fixup_obj = parent;
+			last_fixup_min_off =
+				fda->parent_offset + sizeof(u32) * fda->num_fds;
+		} break;
+		case BINDER_TYPE_PTR: {
+			struct binder_buffer_object *bp =
+				to_binder_buffer_object(hdr);
+			size_t buf_left = sg_buf_end - sg_bufp;
+
+			if (bp->length > buf_left) {
+				binder_user_error("%d:%d got transaction with too large buffer\n",
+						  proc->pid, thread->pid);
 				return_error = BR_FAILED_REPLY;
-				goto err_fget_failed;
+				return_error_param = -EINVAL;
+				return_error_line = __LINE__;
+				goto err_bad_offset;
 			}
-			if (security_binder_transfer_file(proc->tsk,
-							  target_proc->tsk,
-							  file) < 0) {
-				fput(file);
+			if (copy_from_user(sg_bufp,
+					   (const void __user *)(uintptr_t)
+					   bp->buffer, bp->length)) {
+				binder_user_error("%d:%d got transaction with invalid offsets ptr\n",
+						  proc->pid, thread->pid);
+				return_error_param = -EFAULT;
 				return_error = BR_FAILED_REPLY;
-				goto err_get_unused_fd_failed;
+				return_error_line = __LINE__;
+				goto err_copy_data_failed;
 			}
-			target_fd = task_get_unused_fd_flags(target_proc, O_CLOEXEC);
-			if (target_fd < 0) {
-				fput(file);
+			/* Fixup buffer pointer to target proc address space */
+			bp->buffer = (uintptr_t)sg_bufp +
+				binder_alloc_get_user_buffer_offset(
+						&target_proc->alloc);
+			sg_bufp += ALIGN(bp->length, sizeof(u64));
+
+			ret = binder_fixup_parent(t, thread, bp, off_start,
+						  offp - off_start,
+						  last_fixup_obj,
+						  last_fixup_min_off);
+			if (ret < 0) {
 				return_error = BR_FAILED_REPLY;
-				goto err_get_unused_fd_failed;
+				return_error_param = ret;
+				return_error_line = __LINE__;
+				goto err_translate_failed;
 			}
-			task_fd_install(target_proc, target_fd, file);
-			trace_binder_transaction_fd(t, fp->handle, target_fd);
-			binder_debug(BINDER_DEBUG_TRANSACTION,
-				     "        fd %d -> %d\n", fp->handle, target_fd);
-			/* TODO: fput? */
-			fp->binder = 0;
-			fp->handle = target_fd;
+			last_fixup_obj = bp;
+			last_fixup_min_off = 0;
 		} break;
-
 		default:
 			binder_user_error("%d:%d got transaction with invalid object type, %x\n",
-				proc->pid, thread->pid, fp->type);
+				proc->pid, thread->pid, hdr->type);
 			return_error = BR_FAILED_REPLY;
+			return_error_param = -EINVAL;
+			return_error_line = __LINE__;
 			goto err_bad_object_type;
 		}
 	}
+	tcomplete->type = BINDER_WORK_TRANSACTION_COMPLETE;
+	t->work.type = BINDER_WORK_TRANSACTION;
+
 	if (reply) {
+		binder_enqueue_thread_work(thread, tcomplete);
+		binder_inner_proc_lock(target_proc);
+		if (target_thread->is_dead) {
+			binder_inner_proc_unlock(target_proc);
+			goto err_dead_proc_or_thread;
+		}
 		BUG_ON(t->buffer->async_transaction != 0);
-		binder_pop_transaction(target_thread, in_reply_to);
+		binder_pop_transaction_ilocked(target_thread, in_reply_to);
+		binder_enqueue_thread_work_ilocked(target_thread, &t->work);
+		binder_inner_proc_unlock(target_proc);
+		wake_up_interruptible_sync(&target_thread->wait);
+		binder_restore_priority(current, in_reply_to->saved_priority);
+		binder_free_transaction(in_reply_to);
 	} else if (!(t->flags & TF_ONE_WAY)) {
 		BUG_ON(t->buffer->async_transaction != 0);
+		binder_inner_proc_lock(proc);
+		/*
+		 * Defer the TRANSACTION_COMPLETE, so we don't return to
+		 * userspace immediately; this allows the target process to
+		 * immediately start processing this transaction, reducing
+		 * latency. We will then return the TRANSACTION_COMPLETE when
+		 * the target replies (or there is an error).
+		 */
+		binder_enqueue_deferred_thread_work_ilocked(thread, tcomplete);
 		t->need_reply = 1;
 		t->from_parent = thread->transaction_stack;
 		thread->transaction_stack = t;
+		binder_inner_proc_unlock(proc);
+		if (!binder_proc_transaction(t, target_proc, target_thread)) {
+			binder_inner_proc_lock(proc);
+			binder_pop_transaction_ilocked(thread, t);
+			binder_inner_proc_unlock(proc);
+			goto err_dead_proc_or_thread;
+		}
 	} else {
 		BUG_ON(target_node == NULL);
 		BUG_ON(t->buffer->async_transaction != 1);
-		if (target_node->has_async_transaction) {
-			target_list = &target_node->async_todo;
-			target_wait = NULL;
-		} else
-			target_node->has_async_transaction = 1;
-	}
-	t->work.type = BINDER_WORK_TRANSACTION;
-	list_add_tail(&t->work.entry, target_list);
-	tcomplete->type = BINDER_WORK_TRANSACTION_COMPLETE;
-	list_add_tail(&tcomplete->entry, &thread->todo);
-	if (target_wait) {
-		if (reply || !(t->flags & TF_ONE_WAY))
-			wake_up_interruptible_sync(target_wait);
-		else
-			wake_up_interruptible(target_wait);
+		binder_enqueue_thread_work(thread, tcomplete);
+		if (!binder_proc_transaction(t, target_proc, NULL))
+			goto err_dead_proc_or_thread;
 	}
+	if (target_thread)
+		binder_thread_dec_tmpref(target_thread);
+	binder_proc_dec_tmpref(target_proc);
+	if (target_node)
+		binder_dec_node_tmpref(target_node);
+	/*
+	 * write barrier to synchronize with initialization
+	 * of log entry
+	 */
+	smp_wmb();
+	WRITE_ONCE(e->debug_id_done, t_debug_id);
 	return;
 
-err_get_unused_fd_failed:
-err_fget_failed:
-err_fd_not_allowed:
-err_binder_get_ref_for_node_failed:
-err_binder_get_ref_failed:
-err_binder_new_node_failed:
+err_dead_proc_or_thread:
+	return_error = BR_DEAD_REPLY;
+	return_error_line = __LINE__;
+	binder_dequeue_work(proc, tcomplete);
+err_translate_failed:
 err_bad_object_type:
 err_bad_offset:
+err_bad_parent:
 err_copy_data_failed:
 	trace_binder_transaction_failed_buffer_release(t->buffer);
 	binder_transaction_buffer_release(target_proc, t->buffer, offp);
+	if (target_node)
+		binder_dec_node_tmpref(target_node);
+	target_node = NULL;
 	t->buffer->transaction = NULL;
-	binder_free_buf(target_proc, t->buffer);
+	binder_alloc_free_buf(&target_proc->alloc, t->buffer);
 err_binder_alloc_buf_failed:
 	kfree(tcomplete);
 	binder_stats_deleted(BINDER_STAT_TRANSACTION_COMPLETE);
@@ -1750,25 +3413,48 @@ err_bad_call_stack:
 err_empty_call_stack:
 err_dead_binder:
 err_invalid_target_handle:
-err_no_context_mgr_node:
+	if (target_thread)
+		binder_thread_dec_tmpref(target_thread);
+	if (target_proc)
+		binder_proc_dec_tmpref(target_proc);
+	if (target_node) {
+		binder_dec_node(target_node, 1, 0);
+		binder_dec_node_tmpref(target_node);
+	}
+
 	binder_debug(BINDER_DEBUG_FAILED_TRANSACTION,
-		     "%d:%d transaction failed %d, size %lld-%lld\n",
-		     proc->pid, thread->pid, return_error,
-		     (u64)tr->data_size, (u64)tr->offsets_size);
+		     "%d:%d transaction failed %d/%d, size %lld-%lld line %d\n",
+		     proc->pid, thread->pid, return_error, return_error_param,
+		     (u64)tr->data_size, (u64)tr->offsets_size,
+		     return_error_line);
 
 	{
 		struct binder_transaction_log_entry *fe;
 
+		e->return_error = return_error;
+		e->return_error_param = return_error_param;
+		e->return_error_line = return_error_line;
 		fe = binder_transaction_log_add(&binder_transaction_log_failed);
 		*fe = *e;
+		/*
+		 * write barrier to synchronize with initialization
+		 * of log entry
+		 */
+		smp_wmb();
+		WRITE_ONCE(e->debug_id_done, t_debug_id);
+		WRITE_ONCE(fe->debug_id_done, t_debug_id);
 	}
 
-	BUG_ON(thread->return_error != BR_OK);
+	BUG_ON(thread->return_error.cmd != BR_OK);
 	if (in_reply_to) {
-		thread->return_error = BR_TRANSACTION_COMPLETE;
+		binder_restore_priority(current, in_reply_to->saved_priority);
+		thread->return_error.cmd = BR_TRANSACTION_COMPLETE;
+		binder_enqueue_thread_work(thread, &thread->return_error.work);
 		binder_send_failed_reply(in_reply_to, return_error);
-	} else
-		thread->return_error = return_error;
+	} else {
+		thread->return_error.cmd = return_error;
+		binder_enqueue_thread_work(thread, &thread->return_error.work);
+	}
 }
 
 static int binder_thread_write(struct binder_proc *proc,
@@ -1777,19 +3463,22 @@ static int binder_thread_write(struct binder_proc *proc,
 			binder_size_t *consumed)
 {
 	uint32_t cmd;
+	struct binder_context *context = proc->context;
 	void __user *buffer = (void __user *)(uintptr_t)binder_buffer;
 	void __user *ptr = buffer + *consumed;
 	void __user *end = buffer + size;
 
-	while (ptr < end && thread->return_error == BR_OK) {
+	while (ptr < end && thread->return_error.cmd == BR_OK) {
+		int ret;
+
 		if (get_user(cmd, (uint32_t __user *)ptr))
 			return -EFAULT;
 		ptr += sizeof(uint32_t);
 		trace_binder_command(cmd);
 		if (_IOC_NR(cmd) < ARRAY_SIZE(binder_stats.bc)) {
-			binder_stats.bc[_IOC_NR(cmd)]++;
-			proc->stats.bc[_IOC_NR(cmd)]++;
-			thread->stats.bc[_IOC_NR(cmd)]++;
+			atomic_inc(&binder_stats.bc[_IOC_NR(cmd)]);
+			atomic_inc(&proc->stats.bc[_IOC_NR(cmd)]);
+			atomic_inc(&thread->stats.bc[_IOC_NR(cmd)]);
 		}
 		switch (cmd) {
 		case BC_INCREFS:
@@ -1797,53 +3486,61 @@ static int binder_thread_write(struct binder_proc *proc,
 		case BC_RELEASE:
 		case BC_DECREFS: {
 			uint32_t target;
-			struct binder_ref *ref;
 			const char *debug_string;
+			bool strong = cmd == BC_ACQUIRE || cmd == BC_RELEASE;
+			bool increment = cmd == BC_INCREFS || cmd == BC_ACQUIRE;
+			struct binder_ref_data rdata;
 
 			if (get_user(target, (uint32_t __user *)ptr))
 				return -EFAULT;
+
 			ptr += sizeof(uint32_t);
-			if (target == 0 && binder_context_mgr_node &&
-			    (cmd == BC_INCREFS || cmd == BC_ACQUIRE)) {
-				ref = binder_get_ref_for_node(proc,
-					       binder_context_mgr_node);
-				if (ref->desc != target) {
-					binder_user_error("%d:%d tried to acquire reference to desc 0, got %d instead\n",
-						proc->pid, thread->pid,
-						ref->desc);
-				}
-			} else
-				ref = binder_get_ref(proc, target,
-						     cmd == BC_ACQUIRE ||
-						     cmd == BC_RELEASE);
-			if (ref == NULL) {
-				binder_user_error("%d:%d refcount change on invalid ref %d\n",
-					proc->pid, thread->pid, target);
-				break;
+			ret = -1;
+			if (increment && !target) {
+				struct binder_node *ctx_mgr_node;
+				mutex_lock(&context->context_mgr_node_lock);
+				ctx_mgr_node = context->binder_context_mgr_node;
+				if (ctx_mgr_node)
+					ret = binder_inc_ref_for_node(
+							proc, ctx_mgr_node,
+							strong, NULL, &rdata);
+				mutex_unlock(&context->context_mgr_node_lock);
+			}
+			if (ret)
+				ret = binder_update_ref_for_handle(
+						proc, target, increment, strong,
+						&rdata);
+			if (!ret && rdata.desc != target) {
+				binder_user_error("%d:%d tried to acquire reference to desc %d, got %d instead\n",
+					proc->pid, thread->pid,
+					target, rdata.desc);
 			}
 			switch (cmd) {
 			case BC_INCREFS:
 				debug_string = "IncRefs";
-				binder_inc_ref(ref, 0, NULL);
 				break;
 			case BC_ACQUIRE:
 				debug_string = "Acquire";
-				binder_inc_ref(ref, 1, NULL);
 				break;
 			case BC_RELEASE:
 				debug_string = "Release";
-				binder_dec_ref(ref, 1);
 				break;
 			case BC_DECREFS:
 			default:
 				debug_string = "DecRefs";
-				binder_dec_ref(ref, 0);
+				break;
+			}
+			if (ret) {
+				binder_user_error("%d:%d %s %d refcount change on invalid ref %d ret %d\n",
+					proc->pid, thread->pid, debug_string,
+					strong, target, ret);
 				break;
 			}
 			binder_debug(BINDER_DEBUG_USER_REFS,
-				     "%d:%d %s ref %d desc %d s %d w %d for node %d\n",
-				     proc->pid, thread->pid, debug_string, ref->debug_id,
-				     ref->desc, ref->strong, ref->weak, ref->node->debug_id);
+				     "%d:%d %s ref %d desc %d s %d w %d\n",
+				     proc->pid, thread->pid, debug_string,
+				     rdata.debug_id, rdata.desc, rdata.strong,
+				     rdata.weak);
 			break;
 		}
 		case BC_INCREFS_DONE:
@@ -1851,6 +3548,7 @@ static int binder_thread_write(struct binder_proc *proc,
 			binder_uintptr_t node_ptr;
 			binder_uintptr_t cookie;
 			struct binder_node *node;
+			bool free_node;
 
 			if (get_user(node_ptr, (binder_uintptr_t __user *)ptr))
 				return -EFAULT;
@@ -1875,13 +3573,17 @@ static int binder_thread_write(struct binder_proc *proc,
 					"BC_INCREFS_DONE" : "BC_ACQUIRE_DONE",
 					(u64)node_ptr, node->debug_id,
 					(u64)cookie, (u64)node->cookie);
+				binder_put_node(node);
 				break;
 			}
+			binder_node_inner_lock(node);
 			if (cmd == BC_ACQUIRE_DONE) {
 				if (node->pending_strong_ref == 0) {
 					binder_user_error("%d:%d BC_ACQUIRE_DONE node %d has no pending acquire request\n",
 						proc->pid, thread->pid,
 						node->debug_id);
+					binder_node_inner_unlock(node);
+					binder_put_node(node);
 					break;
 				}
 				node->pending_strong_ref = 0;
@@ -1890,16 +3592,23 @@ static int binder_thread_write(struct binder_proc *proc,
 					binder_user_error("%d:%d BC_INCREFS_DONE node %d has no pending increfs request\n",
 						proc->pid, thread->pid,
 						node->debug_id);
+					binder_node_inner_unlock(node);
+					binder_put_node(node);
 					break;
 				}
 				node->pending_weak_ref = 0;
 			}
-			binder_dec_node(node, cmd == BC_ACQUIRE_DONE, 0);
+			free_node = binder_dec_node_nilocked(node,
+					cmd == BC_ACQUIRE_DONE, 0);
+			WARN_ON(free_node);
 			binder_debug(BINDER_DEBUG_USER_REFS,
-				     "%d:%d %s node %d ls %d lw %d\n",
+				     "%d:%d %s node %d ls %d lw %d tr %d\n",
 				     proc->pid, thread->pid,
 				     cmd == BC_INCREFS_DONE ? "BC_INCREFS_DONE" : "BC_ACQUIRE_DONE",
-				     node->debug_id, node->local_strong_refs, node->local_weak_refs);
+				     node->debug_id, node->local_strong_refs,
+				     node->local_weak_refs, node->tmp_refs);
+			binder_node_inner_unlock(node);
+			binder_put_node(node);
 			break;
 		}
 		case BC_ATTEMPT_ACQUIRE:
@@ -1917,7 +3626,8 @@ static int binder_thread_write(struct binder_proc *proc,
 				return -EFAULT;
 			ptr += sizeof(binder_uintptr_t);
 
-			buffer = binder_buffer_lookup(proc, data_ptr);
+			buffer = binder_alloc_prepare_to_free(&proc->alloc,
+							      data_ptr);
 			if (buffer == NULL) {
 				binder_user_error("%d:%d BC_FREE_BUFFER u%016llx no match\n",
 					proc->pid, thread->pid, (u64)data_ptr);
@@ -1939,18 +3649,41 @@ static int binder_thread_write(struct binder_proc *proc,
 				buffer->transaction = NULL;
 			}
 			if (buffer->async_transaction && buffer->target_node) {
-				BUG_ON(!buffer->target_node->has_async_transaction);
-				if (list_empty(&buffer->target_node->async_todo))
-					buffer->target_node->has_async_transaction = 0;
-				else
-					list_move_tail(buffer->target_node->async_todo.next, &thread->todo);
+				struct binder_node *buf_node;
+				struct binder_work *w;
+
+				buf_node = buffer->target_node;
+				binder_node_inner_lock(buf_node);
+				BUG_ON(!buf_node->has_async_transaction);
+				BUG_ON(buf_node->proc != proc);
+				w = binder_dequeue_work_head_ilocked(
+						&buf_node->async_todo);
+				if (!w) {
+					buf_node->has_async_transaction = 0;
+				} else {
+					binder_enqueue_work_ilocked(
+							w, &proc->todo);
+					binder_wakeup_proc_ilocked(proc);
+				}
+				binder_node_inner_unlock(buf_node);
 			}
 			trace_binder_transaction_buffer_release(buffer);
 			binder_transaction_buffer_release(proc, buffer, NULL);
-			binder_free_buf(proc, buffer);
+			binder_alloc_free_buf(&proc->alloc, buffer);
 			break;
 		}
 
+		case BC_TRANSACTION_SG:
+		case BC_REPLY_SG: {
+			struct binder_transaction_data_sg tr;
+
+			if (copy_from_user(&tr, ptr, sizeof(tr)))
+				return -EFAULT;
+			ptr += sizeof(tr);
+			binder_transaction(proc, thread, &tr.transaction_data,
+					   cmd == BC_REPLY_SG, tr.buffers_size);
+			break;
+		}
 		case BC_TRANSACTION:
 		case BC_REPLY: {
 			struct binder_transaction_data tr;
@@ -1958,7 +3691,8 @@ static int binder_thread_write(struct binder_proc *proc,
 			if (copy_from_user(&tr, ptr, sizeof(tr)))
 				return -EFAULT;
 			ptr += sizeof(tr);
-			binder_transaction(proc, thread, &tr, cmd == BC_REPLY);
+			binder_transaction(proc, thread, &tr,
+					   cmd == BC_REPLY, 0);
 			break;
 		}
 
@@ -1966,6 +3700,7 @@ static int binder_thread_write(struct binder_proc *proc,
 			binder_debug(BINDER_DEBUG_THREADS,
 				     "%d:%d BC_REGISTER_LOOPER\n",
 				     proc->pid, thread->pid);
+			binder_inner_proc_lock(proc);
 			if (thread->looper & BINDER_LOOPER_STATE_ENTERED) {
 				thread->looper |= BINDER_LOOPER_STATE_INVALID;
 				binder_user_error("%d:%d ERROR: BC_REGISTER_LOOPER called after BC_ENTER_LOOPER\n",
@@ -1979,6 +3714,7 @@ static int binder_thread_write(struct binder_proc *proc,
 				proc->requested_threads_started++;
 			}
 			thread->looper |= BINDER_LOOPER_STATE_REGISTERED;
+			binder_inner_proc_unlock(proc);
 			break;
 		case BC_ENTER_LOOPER:
 			binder_debug(BINDER_DEBUG_THREADS,
@@ -2003,7 +3739,7 @@ static int binder_thread_write(struct binder_proc *proc,
 			uint32_t target;
 			binder_uintptr_t cookie;
 			struct binder_ref *ref;
-			struct binder_ref_death *death;
+			struct binder_ref_death *death = NULL;
 
 			if (get_user(target, (uint32_t __user *)ptr))
 				return -EFAULT;
@@ -2011,7 +3747,28 @@ static int binder_thread_write(struct binder_proc *proc,
 			if (get_user(cookie, (binder_uintptr_t __user *)ptr))
 				return -EFAULT;
 			ptr += sizeof(binder_uintptr_t);
-			ref = binder_get_ref(proc, target, false);
+			if (cmd == BC_REQUEST_DEATH_NOTIFICATION) {
+				/*
+				 * Allocate memory for death notification
+				 * before taking lock
+				 */
+				death = kzalloc(sizeof(*death), GFP_KERNEL);
+				if (death == NULL) {
+					WARN_ON(thread->return_error.cmd !=
+						BR_OK);
+					thread->return_error.cmd = BR_ERROR;
+					binder_enqueue_thread_work(
+						thread,
+						&thread->return_error.work);
+					binder_debug(
+						BINDER_DEBUG_FAILED_TRANSACTION,
+						"%d:%d BC_REQUEST_DEATH_NOTIFICATION failed\n",
+						proc->pid, thread->pid);
+					break;
+				}
+			}
+			binder_proc_lock(proc);
+			ref = binder_get_ref_olocked(proc, target, false);
 			if (ref == NULL) {
 				binder_user_error("%d:%d %s invalid ref %d\n",
 					proc->pid, thread->pid,
@@ -2019,6 +3776,8 @@ static int binder_thread_write(struct binder_proc *proc,
 					"BC_REQUEST_DEATH_NOTIFICATION" :
 					"BC_CLEAR_DEATH_NOTIFICATION",
 					target);
+				binder_proc_unlock(proc);
+				kfree(death);
 				break;
 			}
 
@@ -2028,21 +3787,18 @@ static int binder_thread_write(struct binder_proc *proc,
 				     cmd == BC_REQUEST_DEATH_NOTIFICATION ?
 				     "BC_REQUEST_DEATH_NOTIFICATION" :
 				     "BC_CLEAR_DEATH_NOTIFICATION",
-				     (u64)cookie, ref->debug_id, ref->desc,
-				     ref->strong, ref->weak, ref->node->debug_id);
+				     (u64)cookie, ref->data.debug_id,
+				     ref->data.desc, ref->data.strong,
+				     ref->data.weak, ref->node->debug_id);
 
+			binder_node_lock(ref->node);
 			if (cmd == BC_REQUEST_DEATH_NOTIFICATION) {
 				if (ref->death) {
 					binder_user_error("%d:%d BC_REQUEST_DEATH_NOTIFICATION death notification already set\n",
 						proc->pid, thread->pid);
-					break;
-				}
-				death = kzalloc(sizeof(*death), GFP_KERNEL);
-				if (death == NULL) {
-					thread->return_error = BR_ERROR;
-					binder_debug(BINDER_DEBUG_FAILED_TRANSACTION,
-						     "%d:%d BC_REQUEST_DEATH_NOTIFICATION failed\n",
-						     proc->pid, thread->pid);
+					binder_node_unlock(ref->node);
+					binder_proc_unlock(proc);
+					kfree(death);
 					break;
 				}
 				binder_stats_created(BINDER_STAT_DEATH);
@@ -2051,17 +3807,19 @@ static int binder_thread_write(struct binder_proc *proc,
 				ref->death = death;
 				if (ref->node->proc == NULL) {
 					ref->death->work.type = BINDER_WORK_DEAD_BINDER;
-					if (thread->looper & (BINDER_LOOPER_STATE_REGISTERED | BINDER_LOOPER_STATE_ENTERED)) {
-						list_add_tail(&ref->death->work.entry, &thread->todo);
-					} else {
-						list_add_tail(&ref->death->work.entry, &proc->todo);
-						wake_up_interruptible(&proc->wait);
-					}
+
+					binder_inner_proc_lock(proc);
+					binder_enqueue_work_ilocked(
+						&ref->death->work, &proc->todo);
+					binder_wakeup_proc_ilocked(proc);
+					binder_inner_proc_unlock(proc);
 				}
 			} else {
 				if (ref->death == NULL) {
 					binder_user_error("%d:%d BC_CLEAR_DEATH_NOTIFICATION death notification not active\n",
 						proc->pid, thread->pid);
+					binder_node_unlock(ref->node);
+					binder_proc_unlock(proc);
 					break;
 				}
 				death = ref->death;
@@ -2070,22 +3828,35 @@ static int binder_thread_write(struct binder_proc *proc,
 						proc->pid, thread->pid,
 						(u64)death->cookie,
 						(u64)cookie);
+					binder_node_unlock(ref->node);
+					binder_proc_unlock(proc);
 					break;
 				}
 				ref->death = NULL;
+				binder_inner_proc_lock(proc);
 				if (list_empty(&death->work.entry)) {
 					death->work.type = BINDER_WORK_CLEAR_DEATH_NOTIFICATION;
-					if (thread->looper & (BINDER_LOOPER_STATE_REGISTERED | BINDER_LOOPER_STATE_ENTERED)) {
-						list_add_tail(&death->work.entry, &thread->todo);
-					} else {
-						list_add_tail(&death->work.entry, &proc->todo);
-						wake_up_interruptible(&proc->wait);
+					if (thread->looper &
+					    (BINDER_LOOPER_STATE_REGISTERED |
+					     BINDER_LOOPER_STATE_ENTERED))
+						binder_enqueue_thread_work_ilocked(
+								thread,
+								&death->work);
+					else {
+						binder_enqueue_work_ilocked(
+								&death->work,
+								&proc->todo);
+						binder_wakeup_proc_ilocked(
+								proc);
 					}
 				} else {
 					BUG_ON(death->work.type != BINDER_WORK_DEAD_BINDER);
 					death->work.type = BINDER_WORK_DEAD_BINDER_AND_CLEAR;
 				}
+				binder_inner_proc_unlock(proc);
 			}
+			binder_node_unlock(ref->node);
+			binder_proc_unlock(proc);
 		} break;
 		case BC_DEAD_BINDER_DONE: {
 			struct binder_work *w;
@@ -2096,8 +3867,13 @@ static int binder_thread_write(struct binder_proc *proc,
 				return -EFAULT;
 
 			ptr += sizeof(cookie);
-			list_for_each_entry(w, &proc->delivered_death, entry) {
-				struct binder_ref_death *tmp_death = container_of(w, struct binder_ref_death, work);
+			binder_inner_proc_lock(proc);
+			list_for_each_entry(w, &proc->delivered_death,
+					    entry) {
+				struct binder_ref_death *tmp_death =
+					container_of(w,
+						     struct binder_ref_death,
+						     work);
 
 				if (tmp_death->cookie == cookie) {
 					death = tmp_death;
@@ -2111,19 +3887,25 @@ static int binder_thread_write(struct binder_proc *proc,
 			if (death == NULL) {
 				binder_user_error("%d:%d BC_DEAD_BINDER_DONE %016llx not found\n",
 					proc->pid, thread->pid, (u64)cookie);
+				binder_inner_proc_unlock(proc);
 				break;
 			}
-
-			list_del_init(&death->work.entry);
+			binder_dequeue_work_ilocked(&death->work);
 			if (death->work.type == BINDER_WORK_DEAD_BINDER_AND_CLEAR) {
 				death->work.type = BINDER_WORK_CLEAR_DEATH_NOTIFICATION;
-				if (thread->looper & (BINDER_LOOPER_STATE_REGISTERED | BINDER_LOOPER_STATE_ENTERED)) {
-					list_add_tail(&death->work.entry, &thread->todo);
-				} else {
-					list_add_tail(&death->work.entry, &proc->todo);
-					wake_up_interruptible(&proc->wait);
+				if (thread->looper &
+					(BINDER_LOOPER_STATE_REGISTERED |
+					 BINDER_LOOPER_STATE_ENTERED))
+					binder_enqueue_thread_work_ilocked(
+						thread, &death->work);
+				else {
+					binder_enqueue_work_ilocked(
+							&death->work,
+							&proc->todo);
+					binder_wakeup_proc_ilocked(proc);
 				}
 			}
+			binder_inner_proc_unlock(proc);
 		} break;
 
 		default:
@@ -2141,23 +3923,73 @@ static void binder_stat_br(struct binder_proc *proc,
 {
 	trace_binder_return(cmd);
 	if (_IOC_NR(cmd) < ARRAY_SIZE(binder_stats.br)) {
-		binder_stats.br[_IOC_NR(cmd)]++;
-		proc->stats.br[_IOC_NR(cmd)]++;
-		thread->stats.br[_IOC_NR(cmd)]++;
+		atomic_inc(&binder_stats.br[_IOC_NR(cmd)]);
+		atomic_inc(&proc->stats.br[_IOC_NR(cmd)]);
+		atomic_inc(&thread->stats.br[_IOC_NR(cmd)]);
 	}
 }
 
-static int binder_has_proc_work(struct binder_proc *proc,
-				struct binder_thread *thread)
+static int binder_put_node_cmd(struct binder_proc *proc,
+			       struct binder_thread *thread,
+			       void __user **ptrp,
+			       binder_uintptr_t node_ptr,
+			       binder_uintptr_t node_cookie,
+			       int node_debug_id,
+			       uint32_t cmd, const char *cmd_name)
 {
-	return !list_empty(&proc->todo) ||
-		(thread->looper & BINDER_LOOPER_STATE_NEED_RETURN);
+	void __user *ptr = *ptrp;
+
+	if (put_user(cmd, (uint32_t __user *)ptr))
+		return -EFAULT;
+	ptr += sizeof(uint32_t);
+
+	if (put_user(node_ptr, (binder_uintptr_t __user *)ptr))
+		return -EFAULT;
+	ptr += sizeof(binder_uintptr_t);
+
+	if (put_user(node_cookie, (binder_uintptr_t __user *)ptr))
+		return -EFAULT;
+	ptr += sizeof(binder_uintptr_t);
+
+	binder_stat_br(proc, thread, cmd);
+	binder_debug(BINDER_DEBUG_USER_REFS, "%d:%d %s %d u%016llx c%016llx\n",
+		     proc->pid, thread->pid, cmd_name, node_debug_id,
+		     (u64)node_ptr, (u64)node_cookie);
+
+	*ptrp = ptr;
+	return 0;
 }
 
-static int binder_has_thread_work(struct binder_thread *thread)
+static int binder_wait_for_work(struct binder_thread *thread,
+				bool do_proc_work)
 {
-	return !list_empty(&thread->todo) || thread->return_error != BR_OK ||
-		(thread->looper & BINDER_LOOPER_STATE_NEED_RETURN);
+	DEFINE_WAIT(wait);
+	struct binder_proc *proc = thread->proc;
+	int ret = 0;
+
+	freezer_do_not_count();
+	binder_inner_proc_lock(proc);
+	for (;;) {
+		prepare_to_wait(&thread->wait, &wait, TASK_INTERRUPTIBLE);
+		if (binder_has_work_ilocked(thread, do_proc_work))
+			break;
+		if (do_proc_work)
+			list_add(&thread->waiting_thread_node,
+				 &proc->waiting_threads);
+		binder_inner_proc_unlock(proc);
+		schedule();
+		binder_inner_proc_lock(proc);
+		list_del_init(&thread->waiting_thread_node);
+		if (signal_pending(current)) {
+			ret = -ERESTARTSYS;
+			break;
+		}
+	}
+	finish_wait(&thread->wait, &wait);
+	binder_inner_proc_unlock(proc);
+	freezer_count();
+
+	return ret;
 }
 
 static int binder_thread_read(struct binder_proc *proc,
@@ -2179,37 +4011,15 @@ static int binder_thread_read(struct binder_proc *proc,
 	}
 
 retry:
-	wait_for_proc_work = thread->transaction_stack == NULL &&
-				list_empty(&thread->todo);
-
-	if (thread->return_error != BR_OK && ptr < end) {
-		if (thread->return_error2 != BR_OK) {
-			if (put_user(thread->return_error2, (uint32_t __user *)ptr))
-				return -EFAULT;
-			ptr += sizeof(uint32_t);
-			binder_stat_br(proc, thread, thread->return_error2);
-			if (ptr == end)
-				goto done;
-			thread->return_error2 = BR_OK;
-		}
-		if (put_user(thread->return_error, (uint32_t __user *)ptr))
-			return -EFAULT;
-		ptr += sizeof(uint32_t);
-		binder_stat_br(proc, thread, thread->return_error);
-		thread->return_error = BR_OK;
-		goto done;
-	}
-
+	binder_inner_proc_lock(proc);
+	wait_for_proc_work = binder_available_for_proc_work_ilocked(thread);
+	binder_inner_proc_unlock(proc);
 
 	thread->looper |= BINDER_LOOPER_STATE_WAITING;
-	if (wait_for_proc_work)
-		proc->ready_threads++;
-
-	binder_unlock(__func__);
 
 	trace_binder_wait_for_work(wait_for_proc_work,
 				   !!thread->transaction_stack,
-				   !list_empty(&thread->todo));
+				   !binder_worklist_empty(proc, &thread->todo));
 	if (wait_for_proc_work) {
 		if (!(thread->looper & (BINDER_LOOPER_STATE_REGISTERED |
 					BINDER_LOOPER_STATE_ENTERED))) {
@@ -2218,24 +4028,16 @@ retry:
 			wait_event_interruptible(binder_user_error_wait,
 						 binder_stop_on_user_error < 2);
 		}
-		binder_set_nice(proc->default_priority);
-		if (non_block) {
-			if (!binder_has_proc_work(proc, thread))
-				ret = -EAGAIN;
-		} else
-			ret = wait_event_freezable_exclusive(proc->wait, binder_has_proc_work(proc, thread));
-	} else {
-		if (non_block) {
-			if (!binder_has_thread_work(thread))
-				ret = -EAGAIN;
-		} else
-			ret = wait_event_freezable(thread->wait, binder_has_thread_work(thread));
+		binder_restore_priority(current, proc->default_priority);
 	}
 
-	binder_lock(__func__);
+	if (non_block) {
+		if (!binder_has_work(thread, wait_for_proc_work))
+			ret = -EAGAIN;
+	} else {
+		ret = binder_wait_for_work(thread, wait_for_proc_work);
+	}
 
-	if (wait_for_proc_work)
-		proc->ready_threads--;
 	thread->looper &= ~BINDER_LOOPER_STATE_WAITING;
 
 	if (ret)
@@ -2244,31 +4046,54 @@ retry:
 	while (1) {
 		uint32_t cmd;
 		struct binder_transaction_data tr;
-		struct binder_work *w;
+		struct binder_work *w = NULL;
+		struct list_head *list = NULL;
 		struct binder_transaction *t = NULL;
+		struct binder_thread *t_from;
+
+		binder_inner_proc_lock(proc);
+		if (!binder_worklist_empty_ilocked(&thread->todo))
+			list = &thread->todo;
+		else if (!binder_worklist_empty_ilocked(&proc->todo) &&
+			   wait_for_proc_work)
+			list = &proc->todo;
+		else {
+			binder_inner_proc_unlock(proc);
 
-		if (!list_empty(&thread->todo)) {
-			w = list_first_entry(&thread->todo, struct binder_work,
-					     entry);
-		} else if (!list_empty(&proc->todo) && wait_for_proc_work) {
-			w = list_first_entry(&proc->todo, struct binder_work,
-					     entry);
-		} else {
 			/* no data added */
-			if (ptr - buffer == 4 &&
-			    !(thread->looper & BINDER_LOOPER_STATE_NEED_RETURN))
+			if (ptr - buffer == 4 && !thread->looper_need_return)
 				goto retry;
 			break;
 		}
 
-		if (end - ptr < sizeof(tr) + 4)
+		if (end - ptr < sizeof(tr) + 4) {
+			binder_inner_proc_unlock(proc);
 			break;
+		}
+		w = binder_dequeue_work_head_ilocked(list);
+		if (binder_worklist_empty_ilocked(&thread->todo))
+			thread->process_todo = false;
 
 		switch (w->type) {
 		case BINDER_WORK_TRANSACTION: {
+			binder_inner_proc_unlock(proc);
 			t = container_of(w, struct binder_transaction, work);
 		} break;
+		case BINDER_WORK_RETURN_ERROR: {
+			struct binder_error *e = container_of(
+					w, struct binder_error, work);
+
+			WARN_ON(e->cmd == BR_OK);
+			binder_inner_proc_unlock(proc);
+			if (put_user(e->cmd, (uint32_t __user *)ptr))
+				return -EFAULT;
+			e->cmd = BR_OK;
+			ptr += sizeof(uint32_t);
+
+			binder_stat_br(proc, thread, cmd);
+		} break;
 		case BINDER_WORK_TRANSACTION_COMPLETE: {
+			binder_inner_proc_unlock(proc);
 			cmd = BR_TRANSACTION_COMPLETE;
 			if (put_user(cmd, (uint32_t __user *)ptr))
 				return -EFAULT;
@@ -2278,113 +4103,134 @@ retry:
 			binder_debug(BINDER_DEBUG_TRANSACTION_COMPLETE,
 				     "%d:%d BR_TRANSACTION_COMPLETE\n",
 				     proc->pid, thread->pid);
-
-			list_del(&w->entry);
 			kfree(w);
 			binder_stats_deleted(BINDER_STAT_TRANSACTION_COMPLETE);
 		} break;
 		case BINDER_WORK_NODE: {
 			struct binder_node *node = container_of(w, struct binder_node, work);
-			uint32_t cmd = BR_NOOP;
-			const char *cmd_name;
-			int strong = node->internal_strong_refs || node->local_strong_refs;
-			int weak = !hlist_empty(&node->refs) || node->local_weak_refs || strong;
-
-			if (weak && !node->has_weak_ref) {
-				cmd = BR_INCREFS;
-				cmd_name = "BR_INCREFS";
+			int strong, weak;
+			binder_uintptr_t node_ptr = node->ptr;
+			binder_uintptr_t node_cookie = node->cookie;
+			int node_debug_id = node->debug_id;
+			int has_weak_ref;
+			int has_strong_ref;
+			void __user *orig_ptr = ptr;
+
+			BUG_ON(proc != node->proc);
+			strong = node->internal_strong_refs ||
+					node->local_strong_refs;
+			weak = !hlist_empty(&node->refs) ||
+					node->local_weak_refs ||
+					node->tmp_refs || strong;
+			has_strong_ref = node->has_strong_ref;
+			has_weak_ref = node->has_weak_ref;
+
+			if (weak && !has_weak_ref) {
 				node->has_weak_ref = 1;
 				node->pending_weak_ref = 1;
 				node->local_weak_refs++;
-			} else if (strong && !node->has_strong_ref) {
-				cmd = BR_ACQUIRE;
-				cmd_name = "BR_ACQUIRE";
+			}
+			if (strong && !has_strong_ref) {
 				node->has_strong_ref = 1;
 				node->pending_strong_ref = 1;
 				node->local_strong_refs++;
-			} else if (!strong && node->has_strong_ref) {
-				cmd = BR_RELEASE;
-				cmd_name = "BR_RELEASE";
+			}
+			if (!strong && has_strong_ref)
 				node->has_strong_ref = 0;
-			} else if (!weak && node->has_weak_ref) {
-				cmd = BR_DECREFS;
-				cmd_name = "BR_DECREFS";
+			if (!weak && has_weak_ref)
 				node->has_weak_ref = 0;
-			}
-			if (cmd != BR_NOOP) {
-				if (put_user(cmd, (uint32_t __user *)ptr))
-					return -EFAULT;
-				ptr += sizeof(uint32_t);
-				if (put_user(node->ptr,
-					     (binder_uintptr_t __user *)ptr))
-					return -EFAULT;
-				ptr += sizeof(binder_uintptr_t);
-				if (put_user(node->cookie,
-					     (binder_uintptr_t __user *)ptr))
-					return -EFAULT;
-				ptr += sizeof(binder_uintptr_t);
-
-				binder_stat_br(proc, thread, cmd);
-				binder_debug(BINDER_DEBUG_USER_REFS,
-					     "%d:%d %s %d u%016llx c%016llx\n",
-					     proc->pid, thread->pid, cmd_name,
-					     node->debug_id,
-					     (u64)node->ptr, (u64)node->cookie);
-			} else {
-				list_del_init(&w->entry);
-				if (!weak && !strong) {
-					binder_debug(BINDER_DEBUG_INTERNAL_REFS,
-						     "%d:%d node %d u%016llx c%016llx deleted\n",
-						     proc->pid, thread->pid,
-						     node->debug_id,
-						     (u64)node->ptr,
-						     (u64)node->cookie);
-					rb_erase(&node->rb_node, &proc->nodes);
-					kfree(node);
-					binder_stats_deleted(BINDER_STAT_NODE);
-				} else {
-					binder_debug(BINDER_DEBUG_INTERNAL_REFS,
-						     "%d:%d node %d u%016llx c%016llx state unchanged\n",
-						     proc->pid, thread->pid,
-						     node->debug_id,
-						     (u64)node->ptr,
-						     (u64)node->cookie);
-				}
-			}
+			if (!weak && !strong) {
+				binder_debug(BINDER_DEBUG_INTERNAL_REFS,
+					     "%d:%d node %d u%016llx c%016llx deleted\n",
+					     proc->pid, thread->pid,
+					     node_debug_id,
+					     (u64)node_ptr,
+					     (u64)node_cookie);
+				rb_erase(&node->rb_node, &proc->nodes);
+				binder_inner_proc_unlock(proc);
+				binder_node_lock(node);
+				/*
+				 * Acquire the node lock before freeing the
+				 * node to serialize with other threads that
+				 * may have been holding the node lock while
+				 * decrementing this node (avoids race where
+				 * this thread frees while the other thread
+				 * is unlocking the node after the final
+				 * decrement)
+				 */
+				binder_node_unlock(node);
+				binder_free_node(node);
+			} else
+				binder_inner_proc_unlock(proc);
+
+			if (weak && !has_weak_ref)
+				ret = binder_put_node_cmd(
+						proc, thread, &ptr, node_ptr,
+						node_cookie, node_debug_id,
+						BR_INCREFS, "BR_INCREFS");
+			if (!ret && strong && !has_strong_ref)
+				ret = binder_put_node_cmd(
+						proc, thread, &ptr, node_ptr,
+						node_cookie, node_debug_id,
+						BR_ACQUIRE, "BR_ACQUIRE");
+			if (!ret && !strong && has_strong_ref)
+				ret = binder_put_node_cmd(
+						proc, thread, &ptr, node_ptr,
+						node_cookie, node_debug_id,
+						BR_RELEASE, "BR_RELEASE");
+			if (!ret && !weak && has_weak_ref)
+				ret = binder_put_node_cmd(
+						proc, thread, &ptr, node_ptr,
+						node_cookie, node_debug_id,
+						BR_DECREFS, "BR_DECREFS");
+			if (orig_ptr == ptr)
+				binder_debug(BINDER_DEBUG_INTERNAL_REFS,
+					     "%d:%d node %d u%016llx c%016llx state unchanged\n",
+					     proc->pid, thread->pid,
+					     node_debug_id,
+					     (u64)node_ptr,
+					     (u64)node_cookie);
+			if (ret)
+				return ret;
 		} break;
 		case BINDER_WORK_DEAD_BINDER:
 		case BINDER_WORK_DEAD_BINDER_AND_CLEAR:
 		case BINDER_WORK_CLEAR_DEATH_NOTIFICATION: {
 			struct binder_ref_death *death;
 			uint32_t cmd;
+			binder_uintptr_t cookie;
 
 			death = container_of(w, struct binder_ref_death, work);
 			if (w->type == BINDER_WORK_CLEAR_DEATH_NOTIFICATION)
 				cmd = BR_CLEAR_DEATH_NOTIFICATION_DONE;
 			else
 				cmd = BR_DEAD_BINDER;
-			if (put_user(cmd, (uint32_t __user *)ptr))
-				return -EFAULT;
-			ptr += sizeof(uint32_t);
-			if (put_user(death->cookie,
-				     (binder_uintptr_t __user *)ptr))
-				return -EFAULT;
-			ptr += sizeof(binder_uintptr_t);
-			binder_stat_br(proc, thread, cmd);
+			cookie = death->cookie;
+
 			binder_debug(BINDER_DEBUG_DEATH_NOTIFICATION,
 				     "%d:%d %s %016llx\n",
 				      proc->pid, thread->pid,
 				      cmd == BR_DEAD_BINDER ?
 				      "BR_DEAD_BINDER" :
 				      "BR_CLEAR_DEATH_NOTIFICATION_DONE",
-				      (u64)death->cookie);
-
+				      (u64)cookie);
 			if (w->type == BINDER_WORK_CLEAR_DEATH_NOTIFICATION) {
-				list_del(&w->entry);
+				binder_inner_proc_unlock(proc);
 				kfree(death);
 				binder_stats_deleted(BINDER_STAT_DEATH);
-			} else
-				list_move(&w->entry, &proc->delivered_death);
+			} else {
+				binder_enqueue_work_ilocked(
+						w, &proc->delivered_death);
+				binder_inner_proc_unlock(proc);
+			}
+			if (put_user(cmd, (uint32_t __user *)ptr))
+				return -EFAULT;
+			ptr += sizeof(uint32_t);
+			if (put_user(cookie,
+				     (binder_uintptr_t __user *)ptr))
+				return -EFAULT;
+			ptr += sizeof(binder_uintptr_t);
+			binder_stat_br(proc, thread, cmd);
 			if (cmd == BR_DEAD_BINDER)
 				goto done; /* DEAD_BINDER notifications can cause transactions */
 		} break;
@@ -2396,16 +4242,14 @@ retry:
 		BUG_ON(t->buffer == NULL);
 		if (t->buffer->target_node) {
 			struct binder_node *target_node = t->buffer->target_node;
+			struct binder_priority node_prio;
 
 			tr.target.ptr = target_node->ptr;
 			tr.cookie =  target_node->cookie;
-			t->saved_priority = task_nice(current);
-			if (t->priority < target_node->min_priority &&
-			    !(t->flags & TF_ONE_WAY))
-				binder_set_nice(t->priority);
-			else if (!(t->flags & TF_ONE_WAY) ||
-				 t->saved_priority > target_node->min_priority)
-				binder_set_nice(target_node->min_priority);
+			node_prio.sched_policy = target_node->sched_policy;
+			node_prio.prio = target_node->min_priority;
+			binder_transaction_priority(current, t, node_prio,
+						    target_node->inherit_rt);
 			cmd = BR_TRANSACTION;
 		} else {
 			tr.target.ptr = 0;
@@ -2416,8 +4260,9 @@ retry:
 		tr.flags = t->flags;
 		tr.sender_euid = from_kuid(current_user_ns(), t->sender_euid);
 
-		if (t->from) {
-			struct task_struct *sender = t->from->proc->tsk;
+		t_from = binder_get_txn_from(t);
+		if (t_from) {
+			struct task_struct *sender = t_from->proc->tsk;
 
 			tr.sender_pid = task_tgid_nr_ns(sender,
 							task_active_pid_ns(current));
@@ -2427,18 +4272,32 @@ retry:
 
 		tr.data_size = t->buffer->data_size;
 		tr.offsets_size = t->buffer->offsets_size;
-		tr.data.ptr.buffer = (binder_uintptr_t)(
-					(uintptr_t)t->buffer->data +
-					proc->user_buffer_offset);
+		tr.data.ptr.buffer = (binder_uintptr_t)
+			((uintptr_t)t->buffer->data +
+			binder_alloc_get_user_buffer_offset(&proc->alloc));
 		tr.data.ptr.offsets = tr.data.ptr.buffer +
 					ALIGN(t->buffer->data_size,
 					    sizeof(void *));
 
-		if (put_user(cmd, (uint32_t __user *)ptr))
+		if (put_user(cmd, (uint32_t __user *)ptr)) {
+			if (t_from)
+				binder_thread_dec_tmpref(t_from);
+
+			binder_cleanup_transaction(t, "put_user failed",
+						   BR_FAILED_REPLY);
+
 			return -EFAULT;
+		}
 		ptr += sizeof(uint32_t);
-		if (copy_to_user(ptr, &tr, sizeof(tr)))
+		if (copy_to_user(ptr, &tr, sizeof(tr))) {
+			if (t_from)
+				binder_thread_dec_tmpref(t_from);
+
+			binder_cleanup_transaction(t, "copy_to_user failed",
+						   BR_FAILED_REPLY);
+
 			return -EFAULT;
+		}
 		ptr += sizeof(tr);
 
 		trace_binder_transaction_received(t);
@@ -2448,21 +4307,22 @@ retry:
 			     proc->pid, thread->pid,
 			     (cmd == BR_TRANSACTION) ? "BR_TRANSACTION" :
 			     "BR_REPLY",
-			     t->debug_id, t->from ? t->from->proc->pid : 0,
-			     t->from ? t->from->pid : 0, cmd,
+			     t->debug_id, t_from ? t_from->proc->pid : 0,
+			     t_from ? t_from->pid : 0, cmd,
 			     t->buffer->data_size, t->buffer->offsets_size,
 			     (u64)tr.data.ptr.buffer, (u64)tr.data.ptr.offsets);
 
-		list_del(&t->work.entry);
+		if (t_from)
+			binder_thread_dec_tmpref(t_from);
 		t->buffer->allow_user_free = 1;
 		if (cmd == BR_TRANSACTION && !(t->flags & TF_ONE_WAY)) {
+			binder_inner_proc_lock(thread->proc);
 			t->to_parent = thread->transaction_stack;
 			t->to_thread = thread;
 			thread->transaction_stack = t;
+			binder_inner_proc_unlock(thread->proc);
 		} else {
-			t->buffer->transaction = NULL;
-			kfree(t);
-			binder_stats_deleted(BINDER_STAT_TRANSACTION);
+			binder_free_transaction(t);
 		}
 		break;
 	}
@@ -2470,45 +4330,52 @@ retry:
 done:
 
 	*consumed = ptr - buffer;
-	if (proc->requested_threads + proc->ready_threads == 0 &&
+	binder_inner_proc_lock(proc);
+	if (proc->requested_threads == 0 &&
+	    list_empty(&thread->proc->waiting_threads) &&
 	    proc->requested_threads_started < proc->max_threads &&
 	    (thread->looper & (BINDER_LOOPER_STATE_REGISTERED |
 	     BINDER_LOOPER_STATE_ENTERED)) /* the user-space code fails to */
 	     /*spawn a new thread if we leave this out */) {
 		proc->requested_threads++;
+		binder_inner_proc_unlock(proc);
 		binder_debug(BINDER_DEBUG_THREADS,
 			     "%d:%d BR_SPAWN_LOOPER\n",
 			     proc->pid, thread->pid);
 		if (put_user(BR_SPAWN_LOOPER, (uint32_t __user *)buffer))
 			return -EFAULT;
 		binder_stat_br(proc, thread, BR_SPAWN_LOOPER);
-	}
+	} else
+		binder_inner_proc_unlock(proc);
 	return 0;
 }
 
-static void binder_release_work(struct list_head *list)
+static void binder_release_work(struct binder_proc *proc,
+				struct list_head *list)
 {
 	struct binder_work *w;
 
-	while (!list_empty(list)) {
-		w = list_first_entry(list, struct binder_work, entry);
-		list_del_init(&w->entry);
+	while (1) {
+		w = binder_dequeue_work_head(proc, list);
+		if (!w)
+			return;
+
 		switch (w->type) {
 		case BINDER_WORK_TRANSACTION: {
 			struct binder_transaction *t;
 
 			t = container_of(w, struct binder_transaction, work);
-			if (t->buffer->target_node &&
-			    !(t->flags & TF_ONE_WAY)) {
-				binder_send_failed_reply(t, BR_DEAD_REPLY);
-			} else {
-				binder_debug(BINDER_DEBUG_DEAD_TRANSACTION,
-					"undelivered transaction %d\n",
-					t->debug_id);
-				t->buffer->transaction = NULL;
-				kfree(t);
-				binder_stats_deleted(BINDER_STAT_TRANSACTION);
-			}
+
+			binder_cleanup_transaction(t, "process died.",
+						   BR_DEAD_REPLY);
+		} break;
+		case BINDER_WORK_RETURN_ERROR: {
+			struct binder_error *e = container_of(
+					w, struct binder_error, work);
+
+			binder_debug(BINDER_DEBUG_DEAD_TRANSACTION,
+				"undelivered TRANSACTION_ERROR: %u\n",
+				e->cmd);
 		} break;
 		case BINDER_WORK_TRANSACTION_COMPLETE: {
 			binder_debug(BINDER_DEBUG_DEAD_TRANSACTION,
@@ -2536,7 +4403,8 @@ static void binder_release_work(struct list_head *list)
 
 }
 
-static struct binder_thread *binder_get_thread(struct binder_proc *proc)
+static struct binder_thread *binder_get_thread_ilocked(
+		struct binder_proc *proc, struct binder_thread *new_thread)
 {
 	struct binder_thread *thread = NULL;
 	struct rb_node *parent = NULL;
@@ -2551,38 +4419,102 @@ static struct binder_thread *binder_get_thread(struct binder_proc *proc)
 		else if (current->pid > thread->pid)
 			p = &(*p)->rb_right;
 		else
-			break;
+			return thread;
 	}
-	if (*p == NULL) {
-		thread = kzalloc(sizeof(*thread), GFP_KERNEL);
-		if (thread == NULL)
+	if (!new_thread)
+		return NULL;
+	thread = new_thread;
+	binder_stats_created(BINDER_STAT_THREAD);
+	thread->proc = proc;
+	thread->pid = current->pid;
+	get_task_struct(current);
+	thread->task = current;
+	atomic_set(&thread->tmp_ref, 0);
+	init_waitqueue_head(&thread->wait);
+	INIT_LIST_HEAD(&thread->todo);
+	rb_link_node(&thread->rb_node, parent, p);
+	rb_insert_color(&thread->rb_node, &proc->threads);
+	thread->looper_need_return = true;
+	thread->return_error.work.type = BINDER_WORK_RETURN_ERROR;
+	thread->return_error.cmd = BR_OK;
+	thread->reply_error.work.type = BINDER_WORK_RETURN_ERROR;
+	thread->reply_error.cmd = BR_OK;
+	INIT_LIST_HEAD(&new_thread->waiting_thread_node);
+	return thread;
+}
+
+static struct binder_thread *binder_get_thread(struct binder_proc *proc)
+{
+	struct binder_thread *thread;
+	struct binder_thread *new_thread;
+
+	binder_inner_proc_lock(proc);
+	thread = binder_get_thread_ilocked(proc, NULL);
+	binder_inner_proc_unlock(proc);
+	if (!thread) {
+		new_thread = kzalloc(sizeof(*thread), GFP_KERNEL);
+		if (new_thread == NULL)
 			return NULL;
-		binder_stats_created(BINDER_STAT_THREAD);
-		thread->proc = proc;
-		thread->pid = current->pid;
-		init_waitqueue_head(&thread->wait);
-		INIT_LIST_HEAD(&thread->todo);
-		rb_link_node(&thread->rb_node, parent, p);
-		rb_insert_color(&thread->rb_node, &proc->threads);
-		thread->looper |= BINDER_LOOPER_STATE_NEED_RETURN;
-		thread->return_error = BR_OK;
-		thread->return_error2 = BR_OK;
+		binder_inner_proc_lock(proc);
+		thread = binder_get_thread_ilocked(proc, new_thread);
+		binder_inner_proc_unlock(proc);
+		if (thread != new_thread)
+			kfree(new_thread);
 	}
 	return thread;
 }
 
-static int binder_free_thread(struct binder_proc *proc,
-			      struct binder_thread *thread)
+static void binder_free_proc(struct binder_proc *proc)
+{
+	BUG_ON(!list_empty(&proc->todo));
+	BUG_ON(!list_empty(&proc->delivered_death));
+	binder_alloc_deferred_release(&proc->alloc);
+	put_task_struct(proc->tsk);
+	binder_stats_deleted(BINDER_STAT_PROC);
+	kfree(proc);
+}
+
+static void binder_free_thread(struct binder_thread *thread)
+{
+	BUG_ON(!list_empty(&thread->todo));
+	binder_stats_deleted(BINDER_STAT_THREAD);
+	binder_proc_dec_tmpref(thread->proc);
+	put_task_struct(thread->task);
+	kfree(thread);
+}
+
+static int binder_thread_release(struct binder_proc *proc,
+				 struct binder_thread *thread)
 {
 	struct binder_transaction *t;
 	struct binder_transaction *send_reply = NULL;
 	int active_transactions = 0;
-
+	struct binder_transaction *last_t = NULL;
+
+	binder_inner_proc_lock(thread->proc);
+	/*
+	 * take a ref on the proc so it survives
+	 * after we remove this thread from proc->threads.
+	 * The corresponding dec is when we actually
+	 * free the thread in binder_free_thread()
+	 */
+	proc->tmp_ref++;
+	/*
+	 * take a ref on this thread to ensure it
+	 * survives while we are releasing it
+	 */
+	atomic_inc(&thread->tmp_ref);
 	rb_erase(&thread->rb_node, &proc->threads);
 	t = thread->transaction_stack;
-	if (t && t->to_thread == thread)
-		send_reply = t;
+	if (t) {
+		spin_lock(&t->lock);
+		if (t->to_thread == thread)
+			send_reply = t;
+	}
+	thread->is_dead = true;
+
 	while (t) {
+		last_t = t;
 		active_transactions++;
 		binder_debug(BINDER_DEBUG_DEAD_TRANSACTION,
 			     "release %d:%d transaction %d %s, still active\n",
@@ -2603,12 +4535,16 @@ static int binder_free_thread(struct binder_proc *proc,
 			t = t->from_parent;
 		} else
 			BUG();
+		spin_unlock(&last_t->lock);
+		if (t)
+			spin_lock(&t->lock);
 	}
+	binder_inner_proc_unlock(thread->proc);
+
 	if (send_reply)
 		binder_send_failed_reply(send_reply, BR_DEAD_REPLY);
-	binder_release_work(&thread->todo);
-	kfree(thread);
-	binder_stats_deleted(BINDER_STAT_THREAD);
+	binder_release_work(proc, &thread->todo);
+	binder_thread_dec_tmpref(thread);
 	return active_transactions;
 }
 
@@ -2617,30 +4553,21 @@ static unsigned int binder_poll(struct file *filp,
 {
 	struct binder_proc *proc = filp->private_data;
 	struct binder_thread *thread = NULL;
-	int wait_for_proc_work;
-
-	binder_lock(__func__);
+	bool wait_for_proc_work;
 
 	thread = binder_get_thread(proc);
 
-	wait_for_proc_work = thread->transaction_stack == NULL &&
-		list_empty(&thread->todo) && thread->return_error == BR_OK;
+	binder_inner_proc_lock(thread->proc);
+	thread->looper |= BINDER_LOOPER_STATE_POLL;
+	wait_for_proc_work = binder_available_for_proc_work_ilocked(thread);
 
-	binder_unlock(__func__);
+	binder_inner_proc_unlock(thread->proc);
+
+	poll_wait(filp, &thread->wait, wait);
+
+	if (binder_has_work(thread, wait_for_proc_work))
+		return POLLIN;
 
-	if (wait_for_proc_work) {
-		if (binder_has_proc_work(proc, thread))
-			return POLLIN;
-		poll_wait(filp, &proc->wait, wait);
-		if (binder_has_proc_work(proc, thread))
-			return POLLIN;
-	} else {
-		if (binder_has_thread_work(thread))
-			return POLLIN;
-		poll_wait(filp, &thread->wait, wait);
-		if (binder_has_thread_work(thread))
-			return POLLIN;
-	}
 	return 0;
 }
 
@@ -2687,8 +4614,10 @@ static int binder_ioctl_write_read(struct file *filp,
 					 &bwr.read_consumed,
 					 filp->f_flags & O_NONBLOCK);
 		trace_binder_read_done(ret);
-		if (!list_empty(&proc->todo))
-			wake_up_interruptible(&proc->wait);
+		binder_inner_proc_lock(proc);
+		if (!binder_worklist_empty_ilocked(&proc->todo))
+			binder_wakeup_proc_ilocked(proc);
+		binder_inner_proc_unlock(proc);
 		if (ret < 0) {
 			if (copy_to_user(ubuf, &bwr, sizeof(bwr)))
 				ret = -EFAULT;
@@ -2712,9 +4641,12 @@ static int binder_ioctl_set_ctx_mgr(struct file *filp)
 {
 	int ret = 0;
 	struct binder_proc *proc = filp->private_data;
+	struct binder_context *context = proc->context;
+	struct binder_node *new_node;
 	kuid_t curr_euid = current_euid();
 
-	if (binder_context_mgr_node != NULL) {
+	mutex_lock(&context->context_mgr_node_lock);
+	if (context->binder_context_mgr_node) {
 		pr_err("BINDER_SET_CONTEXT_MGR already set\n");
 		ret = -EBUSY;
 		goto out;
@@ -2722,31 +4654,60 @@ static int binder_ioctl_set_ctx_mgr(struct file *filp)
 	ret = security_binder_set_context_mgr(proc->tsk);
 	if (ret < 0)
 		goto out;
-	if (uid_valid(binder_context_mgr_uid)) {
-		if (!uid_eq(binder_context_mgr_uid, curr_euid)) {
+	if (uid_valid(context->binder_context_mgr_uid)) {
+		if (!uid_eq(context->binder_context_mgr_uid, curr_euid)) {
 			pr_err("BINDER_SET_CONTEXT_MGR bad uid %d != %d\n",
 			       from_kuid(&init_user_ns, curr_euid),
 			       from_kuid(&init_user_ns,
-					binder_context_mgr_uid));
+					 context->binder_context_mgr_uid));
 			ret = -EPERM;
 			goto out;
 		}
 	} else {
-		binder_context_mgr_uid = curr_euid;
+		context->binder_context_mgr_uid = curr_euid;
 	}
-	binder_context_mgr_node = binder_new_node(proc, 0, 0);
-	if (binder_context_mgr_node == NULL) {
+	new_node = binder_new_node(proc, NULL);
+	if (!new_node) {
 		ret = -ENOMEM;
 		goto out;
 	}
-	binder_context_mgr_node->local_weak_refs++;
-	binder_context_mgr_node->local_strong_refs++;
-	binder_context_mgr_node->has_strong_ref = 1;
-	binder_context_mgr_node->has_weak_ref = 1;
+	binder_node_lock(new_node);
+	new_node->local_weak_refs++;
+	new_node->local_strong_refs++;
+	new_node->has_strong_ref = 1;
+	new_node->has_weak_ref = 1;
+	context->binder_context_mgr_node = new_node;
+	binder_node_unlock(new_node);
+	binder_put_node(new_node);
 out:
+	mutex_unlock(&context->context_mgr_node_lock);
 	return ret;
 }
 
+static int binder_ioctl_get_node_debug_info(struct binder_proc *proc,
+				struct binder_node_debug_info *info) {
+	struct rb_node *n;
+	binder_uintptr_t ptr = info->ptr;
+
+	memset(info, 0, sizeof(*info));
+
+	binder_inner_proc_lock(proc);
+	for (n = rb_first(&proc->nodes); n != NULL; n = rb_next(n)) {
+		struct binder_node *node = rb_entry(n, struct binder_node,
+						    rb_node);
+		if (node->ptr > ptr) {
+			info->ptr = node->ptr;
+			info->cookie = node->cookie;
+			info->has_strong_ref = node->has_strong_ref;
+			info->has_weak_ref = node->has_weak_ref;
+			break;
+		}
+	}
+	binder_inner_proc_unlock(proc);
+
+	return 0;
+}
+
 static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	int ret;
@@ -2758,13 +4719,14 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	/*pr_info("binder_ioctl: %d:%d %x %lx\n",
 			proc->pid, current->pid, cmd, arg);*/
 
+	binder_selftest_alloc(&proc->alloc);
+
 	trace_binder_ioctl(cmd, arg);
 
 	ret = wait_event_interruptible(binder_user_error_wait, binder_stop_on_user_error < 2);
 	if (ret)
 		goto err_unlocked;
 
-	binder_lock(__func__);
 	thread = binder_get_thread(proc);
 	if (thread == NULL) {
 		ret = -ENOMEM;
@@ -2777,12 +4739,19 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		if (ret)
 			goto err;
 		break;
-	case BINDER_SET_MAX_THREADS:
-		if (copy_from_user(&proc->max_threads, ubuf, sizeof(proc->max_threads))) {
+	case BINDER_SET_MAX_THREADS: {
+		int max_threads;
+
+		if (copy_from_user(&max_threads, ubuf,
+				   sizeof(max_threads))) {
 			ret = -EINVAL;
 			goto err;
 		}
+		binder_inner_proc_lock(proc);
+		proc->max_threads = max_threads;
+		binder_inner_proc_unlock(proc);
 		break;
+	}
 	case BINDER_SET_CONTEXT_MGR:
 		ret = binder_ioctl_set_ctx_mgr(filp);
 		if (ret)
@@ -2791,7 +4760,7 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	case BINDER_THREAD_EXIT:
 		binder_debug(BINDER_DEBUG_THREADS, "%d:%d exit\n",
 			     proc->pid, thread->pid);
-		binder_free_thread(proc, thread);
+		binder_thread_release(proc, thread);
 		thread = NULL;
 		break;
 	case BINDER_VERSION: {
@@ -2808,6 +4777,24 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		}
 		break;
 	}
+	case BINDER_GET_NODE_DEBUG_INFO: {
+		struct binder_node_debug_info info;
+
+		if (copy_from_user(&info, ubuf, sizeof(info))) {
+			ret = -EFAULT;
+			goto err;
+		}
+
+		ret = binder_ioctl_get_node_debug_info(proc, &info);
+		if (ret < 0)
+			goto err;
+
+		if (copy_to_user(ubuf, &info, sizeof(info))) {
+			ret = -EFAULT;
+			goto err;
+		}
+		break;
+	}
 	default:
 		ret = -EINVAL;
 		goto err;
@@ -2815,8 +4802,7 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 	ret = 0;
 err:
 	if (thread)
-		thread->looper &= ~BINDER_LOOPER_STATE_NEED_RETURN;
-	binder_unlock(__func__);
+		thread->looper_need_return = false;
 	wait_event_interruptible(binder_user_error_wait, binder_stop_on_user_error < 2);
 	if (ret && ret != -ERESTARTSYS)
 		pr_info("%d:%d ioctl %x %lx returned %d\n", proc->pid, current->pid, cmd, arg, ret);
@@ -2845,9 +4831,7 @@ static void binder_vma_close(struct vm_area_struct *vma)
 		     proc->pid, vma->vm_start, vma->vm_end,
 		     (vma->vm_end - vma->vm_start) / SZ_1K, vma->vm_flags,
 		     (unsigned long)pgprot_val(vma->vm_page_prot));
-	proc->vma = NULL;
-	proc->vma_vm_mm = NULL;
-	binder_defer_work(proc, BINDER_DEFERRED_PUT_FILES);
+	binder_alloc_vma_close(&proc->alloc);
 }
 
 static int binder_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
@@ -2864,10 +4848,8 @@ static const struct vm_operations_struct binder_vm_ops = {
 static int binder_mmap(struct file *filp, struct vm_area_struct *vma)
 {
 	int ret;
-	struct vm_struct *area;
 	struct binder_proc *proc = filp->private_data;
 	const char *failure_string;
-	struct binder_buffer *buffer;
 
 	if (proc->tsk != current->group_leader)
 		return -EINVAL;
@@ -2876,8 +4858,8 @@ static int binder_mmap(struct file *filp, struct vm_area_struct *vma)
 		vma->vm_end = vma->vm_start + SZ_4M;
 
 	binder_debug(BINDER_DEBUG_OPEN_CLOSE,
-		     "binder_mmap: %d %lx-%lx (%ld K) vma %lx pagep %lx\n",
-		     proc->pid, vma->vm_start, vma->vm_end,
+		     "%s: %d %lx-%lx (%ld K) vma %lx pagep %lx\n",
+		     __func__, proc->pid, vma->vm_start, vma->vm_end,
 		     (vma->vm_end - vma->vm_start) / SZ_1K, vma->vm_flags,
 		     (unsigned long)pgprot_val(vma->vm_page_prot));
 
@@ -2887,73 +4869,13 @@ static int binder_mmap(struct file *filp, struct vm_area_struct *vma)
 		goto err_bad_arg;
 	}
 	vma->vm_flags = (vma->vm_flags | VM_DONTCOPY) & ~VM_MAYWRITE;
-
-	mutex_lock(&binder_mmap_lock);
-	if (proc->buffer) {
-		ret = -EBUSY;
-		failure_string = "already mapped";
-		goto err_already_mapped;
-	}
-
-	area = get_vm_area(vma->vm_end - vma->vm_start, VM_IOREMAP);
-	if (area == NULL) {
-		ret = -ENOMEM;
-		failure_string = "get_vm_area";
-		goto err_get_vm_area_failed;
-	}
-	proc->buffer = area->addr;
-	proc->user_buffer_offset = vma->vm_start - (uintptr_t)proc->buffer;
-	mutex_unlock(&binder_mmap_lock);
-
-#ifdef CONFIG_CPU_CACHE_VIPT
-	if (cache_is_vipt_aliasing()) {
-		while (CACHE_COLOUR((vma->vm_start ^ (uint32_t)proc->buffer))) {
-			pr_info("binder_mmap: %d %lx-%lx maps %p bad alignment\n", proc->pid, vma->vm_start, vma->vm_end, proc->buffer);
-			vma->vm_start += PAGE_SIZE;
-		}
-	}
-#endif
-	proc->pages = kzalloc(sizeof(proc->pages[0]) * ((vma->vm_end - vma->vm_start) / PAGE_SIZE), GFP_KERNEL);
-	if (proc->pages == NULL) {
-		ret = -ENOMEM;
-		failure_string = "alloc page array";
-		goto err_alloc_pages_failed;
-	}
-	proc->buffer_size = vma->vm_end - vma->vm_start;
-
 	vma->vm_ops = &binder_vm_ops;
 	vma->vm_private_data = proc;
 
-	if (binder_update_page_range(proc, 1, proc->buffer, proc->buffer + PAGE_SIZE, vma)) {
-		ret = -ENOMEM;
-		failure_string = "alloc small buf";
-		goto err_alloc_small_buf_failed;
-	}
-	buffer = proc->buffer;
-	INIT_LIST_HEAD(&proc->buffers);
-	list_add(&buffer->entry, &proc->buffers);
-	buffer->free = 1;
-	binder_insert_free_buffer(proc, buffer);
-	proc->free_async_space = proc->buffer_size / 2;
-	barrier();
-	proc->files = get_files_struct(current);
-	proc->vma = vma;
-	proc->vma_vm_mm = vma->vm_mm;
-
-	/*pr_info("binder_mmap: %d %lx-%lx maps %p\n",
-		 proc->pid, vma->vm_start, vma->vm_end, proc->buffer);*/
-	return 0;
+	ret = binder_alloc_mmap_handler(&proc->alloc, vma);
+
+	return ret;
 
-err_alloc_small_buf_failed:
-	kfree(proc->pages);
-	proc->pages = NULL;
-err_alloc_pages_failed:
-	mutex_lock(&binder_mmap_lock);
-	vfree(proc->buffer);
-	proc->buffer = NULL;
-err_get_vm_area_failed:
-err_already_mapped:
-	mutex_unlock(&binder_mmap_lock);
 err_bad_arg:
 	pr_err("binder_mmap: %d %lx-%lx %s failed %d\n",
 	       proc->pid, vma->vm_start, vma->vm_end, failure_string, ret);
@@ -2963,6 +4885,7 @@ err_bad_arg:
 static int binder_open(struct inode *nodp, struct file *filp)
 {
 	struct binder_proc *proc;
+	struct binder_device *binder_dev;
 
 	binder_debug(BINDER_DEBUG_OPEN_CLOSE, "binder_open: %d:%d\n",
 		     current->group_leader->pid, current->pid);
@@ -2970,28 +4893,49 @@ static int binder_open(struct inode *nodp, struct file *filp)
 	proc = kzalloc(sizeof(*proc), GFP_KERNEL);
 	if (proc == NULL)
 		return -ENOMEM;
+	spin_lock_init(&proc->inner_lock);
+	spin_lock_init(&proc->outer_lock);
 	get_task_struct(current->group_leader);
 	proc->tsk = current->group_leader;
 	INIT_LIST_HEAD(&proc->todo);
-	init_waitqueue_head(&proc->wait);
-	proc->default_priority = task_nice(current);
+	if (binder_supported_policy(current->policy)) {
+		proc->default_priority.sched_policy = current->policy;
+		proc->default_priority.prio = current->normal_prio;
+	} else {
+		proc->default_priority.sched_policy = SCHED_NORMAL;
+		proc->default_priority.prio = NICE_TO_PRIO(0);
+	}
 
-	binder_lock(__func__);
+	binder_dev = container_of(filp->private_data, struct binder_device,
+				  miscdev);
+	proc->context = &binder_dev->context;
+	binder_alloc_init(&proc->alloc);
 
 	binder_stats_created(BINDER_STAT_PROC);
-	hlist_add_head(&proc->proc_node, &binder_procs);
 	proc->pid = current->group_leader->pid;
 	INIT_LIST_HEAD(&proc->delivered_death);
+	INIT_LIST_HEAD(&proc->waiting_threads);
 	filp->private_data = proc;
 
-	binder_unlock(__func__);
+	mutex_lock(&binder_procs_lock);
+	hlist_add_head(&proc->proc_node, &binder_procs);
+	mutex_unlock(&binder_procs_lock);
 
 	if (binder_debugfs_dir_entry_proc) {
 		char strbuf[11];
 
 		snprintf(strbuf, sizeof(strbuf), "%u", proc->pid);
+		/*
+		 * proc debug entries are shared between contexts, so
+		 * this will fail if the process tries to open the driver
+		 * again with a different context. The priting code will
+		 * anyway print all contexts that a given PID has, so this
+		 * is not a problem.
+		 */
 		proc->debugfs_entry = debugfs_create_file(strbuf, S_IRUGO,
-			binder_debugfs_dir_entry_proc, proc, &binder_proc_fops);
+			binder_debugfs_dir_entry_proc,
+			(void *)(unsigned long)proc->pid,
+			&binder_proc_fops);
 	}
 
 	return 0;
@@ -3011,16 +4955,17 @@ static void binder_deferred_flush(struct binder_proc *proc)
 	struct rb_node *n;
 	int wake_count = 0;
 
+	binder_inner_proc_lock(proc);
 	for (n = rb_first(&proc->threads); n != NULL; n = rb_next(n)) {
 		struct binder_thread *thread = rb_entry(n, struct binder_thread, rb_node);
 
-		thread->looper |= BINDER_LOOPER_STATE_NEED_RETURN;
+		thread->looper_need_return = true;
 		if (thread->looper & BINDER_LOOPER_STATE_WAITING) {
 			wake_up_interruptible(&thread->wait);
 			wake_count++;
 		}
 	}
-	wake_up_interruptible_all(&proc->wait);
+	binder_inner_proc_unlock(proc);
 
 	binder_debug(BINDER_DEBUG_OPEN_CLOSE,
 		     "binder_flush: %d woke %d threads\n", proc->pid,
@@ -3041,13 +4986,21 @@ static int binder_node_release(struct binder_node *node, int refs)
 {
 	struct binder_ref *ref;
 	int death = 0;
-
-	list_del_init(&node->work.entry);
-	binder_release_work(&node->async_todo);
-
-	if (hlist_empty(&node->refs)) {
-		kfree(node);
-		binder_stats_deleted(BINDER_STAT_NODE);
+	struct binder_proc *proc = node->proc;
+
+	binder_release_work(proc, &node->async_todo);
+
+	binder_node_lock(node);
+	binder_inner_proc_lock(proc);
+	binder_dequeue_work_ilocked(&node->work);
+	/*
+	 * The caller must have taken a temporary ref on the node,
+	 */
+	BUG_ON(!node->tmp_refs);
+	if (hlist_empty(&node->refs) && node->tmp_refs == 1) {
+		binder_inner_proc_unlock(proc);
+		binder_node_unlock(node);
+		binder_free_node(node);
 
 		return refs;
 	}
@@ -3055,59 +5008,82 @@ static int binder_node_release(struct binder_node *node, int refs)
 	node->proc = NULL;
 	node->local_strong_refs = 0;
 	node->local_weak_refs = 0;
+	binder_inner_proc_unlock(proc);
+
+	spin_lock(&binder_dead_nodes_lock);
 	hlist_add_head(&node->dead_node, &binder_dead_nodes);
+	spin_unlock(&binder_dead_nodes_lock);
 
 	hlist_for_each_entry(ref, &node->refs, node_entry) {
 		refs++;
-
-		if (!ref->death)
+		/*
+		 * Need the node lock to synchronize
+		 * with new notification requests and the
+		 * inner lock to synchronize with queued
+		 * death notifications.
+		 */
+		binder_inner_proc_lock(ref->proc);
+		if (!ref->death) {
+			binder_inner_proc_unlock(ref->proc);
 			continue;
+		}
 
 		death++;
 
-		if (list_empty(&ref->death->work.entry)) {
-			ref->death->work.type = BINDER_WORK_DEAD_BINDER;
-			list_add_tail(&ref->death->work.entry,
-				      &ref->proc->todo);
-			wake_up_interruptible(&ref->proc->wait);
-		} else
-			BUG();
+		BUG_ON(!list_empty(&ref->death->work.entry));
+		ref->death->work.type = BINDER_WORK_DEAD_BINDER;
+		binder_enqueue_work_ilocked(&ref->death->work,
+					    &ref->proc->todo);
+		binder_wakeup_proc_ilocked(ref->proc);
+		binder_inner_proc_unlock(ref->proc);
 	}
 
 	binder_debug(BINDER_DEBUG_DEAD_BINDER,
 		     "node %d now dead, refs %d, death %d\n",
 		     node->debug_id, refs, death);
+	binder_node_unlock(node);
+	binder_put_node(node);
 
 	return refs;
 }
 
 static void binder_deferred_release(struct binder_proc *proc)
 {
-	struct binder_transaction *t;
+	struct binder_context *context = proc->context;
 	struct rb_node *n;
-	int threads, nodes, incoming_refs, outgoing_refs, buffers,
-		active_transactions, page_count;
-
-	BUG_ON(proc->vma);
-	BUG_ON(proc->files);
+	int threads, nodes, incoming_refs, outgoing_refs, active_transactions;
 
+	mutex_lock(&binder_procs_lock);
 	hlist_del(&proc->proc_node);
+	mutex_unlock(&binder_procs_lock);
 
-	if (binder_context_mgr_node && binder_context_mgr_node->proc == proc) {
+	mutex_lock(&context->context_mgr_node_lock);
+	if (context->binder_context_mgr_node &&
+	    context->binder_context_mgr_node->proc == proc) {
 		binder_debug(BINDER_DEBUG_DEAD_BINDER,
 			     "%s: %d context_mgr_node gone\n",
 			     __func__, proc->pid);
-		binder_context_mgr_node = NULL;
+		context->binder_context_mgr_node = NULL;
 	}
-
+	mutex_unlock(&context->context_mgr_node_lock);
+	binder_inner_proc_lock(proc);
+	/*
+	 * Make sure proc stays alive after we
+	 * remove all the threads
+	 */
+	proc->tmp_ref++;
+
+	proc->is_dead = true;
 	threads = 0;
 	active_transactions = 0;
 	while ((n = rb_first(&proc->threads))) {
 		struct binder_thread *thread;
 
 		thread = rb_entry(n, struct binder_thread, rb_node);
+		binder_inner_proc_unlock(proc);
 		threads++;
-		active_transactions += binder_free_thread(proc, thread);
+		active_transactions += binder_thread_release(proc, thread);
+		binder_inner_proc_lock(proc);
 	}
 
 	nodes = 0;
@@ -3117,84 +5093,50 @@ static void binder_deferred_release(struct binder_proc *proc)
 
 		node = rb_entry(n, struct binder_node, rb_node);
 		nodes++;
+		/*
+		 * take a temporary ref on the node before
+		 * calling binder_node_release() which will either
+		 * kfree() the node or call binder_put_node()
+		 */
+		binder_inc_node_tmpref_ilocked(node);
 		rb_erase(&node->rb_node, &proc->nodes);
+		binder_inner_proc_unlock(proc);
 		incoming_refs = binder_node_release(node, incoming_refs);
+		binder_inner_proc_lock(proc);
 	}
+	binder_inner_proc_unlock(proc);
 
 	outgoing_refs = 0;
+	binder_proc_lock(proc);
 	while ((n = rb_first(&proc->refs_by_desc))) {
 		struct binder_ref *ref;
 
 		ref = rb_entry(n, struct binder_ref, rb_node_desc);
 		outgoing_refs++;
-		binder_delete_ref(ref);
+		binder_cleanup_ref_olocked(ref);
+		binder_proc_unlock(proc);
+		binder_free_ref(ref);
+		binder_proc_lock(proc);
 	}
+	binder_proc_unlock(proc);
 
-	binder_release_work(&proc->todo);
-	binder_release_work(&proc->delivered_death);
-
-	buffers = 0;
-	while ((n = rb_first(&proc->allocated_buffers))) {
-		struct binder_buffer *buffer;
-
-		buffer = rb_entry(n, struct binder_buffer, rb_node);
-
-		t = buffer->transaction;
-		if (t) {
-			t->buffer = NULL;
-			buffer->transaction = NULL;
-			pr_err("release proc %d, transaction %d, not freed\n",
-			       proc->pid, t->debug_id);
-			/*BUG();*/
-		}
-
-		binder_free_buf(proc, buffer);
-		buffers++;
-	}
-
-	binder_stats_deleted(BINDER_STAT_PROC);
-
-	page_count = 0;
-	if (proc->pages) {
-		int i;
-
-		for (i = 0; i < proc->buffer_size / PAGE_SIZE; i++) {
-			void *page_addr;
-
-			if (!proc->pages[i])
-				continue;
-
-			page_addr = proc->buffer + i * PAGE_SIZE;
-			binder_debug(BINDER_DEBUG_BUFFER_ALLOC,
-				     "%s: %d: page %d at %p not freed\n",
-				     __func__, proc->pid, i, page_addr);
-			unmap_kernel_range((unsigned long)page_addr, PAGE_SIZE);
-			__free_page(proc->pages[i]);
-			page_count++;
-		}
-		kfree(proc->pages);
-		vfree(proc->buffer);
-	}
-
-	put_task_struct(proc->tsk);
+	binder_release_work(proc, &proc->todo);
+	binder_release_work(proc, &proc->delivered_death);
 
 	binder_debug(BINDER_DEBUG_OPEN_CLOSE,
-		     "%s: %d threads %d, nodes %d (ref %d), refs %d, active transactions %d, buffers %d, pages %d\n",
+		     "%s: %d threads %d, nodes %d (ref %d), refs %d, active transactions %d\n",
 		     __func__, proc->pid, threads, nodes, incoming_refs,
-		     outgoing_refs, active_transactions, buffers, page_count);
+		     outgoing_refs, active_transactions);
 
-	kfree(proc);
+	binder_proc_dec_tmpref(proc);
 }
 
 static void binder_deferred_func(struct work_struct *work)
 {
 	struct binder_proc *proc;
-	struct files_struct *files;
-
 	int defer;
 
 	do {
-		binder_lock(__func__);
 		mutex_lock(&binder_deferred_lock);
 		if (!hlist_empty(&binder_deferred_list)) {
 			proc = hlist_entry(binder_deferred_list.first,
@@ -3208,22 +5150,11 @@ static void binder_deferred_func(struct work_struct *work)
 		}
 		mutex_unlock(&binder_deferred_lock);
 
-		files = NULL;
-		if (defer & BINDER_DEFERRED_PUT_FILES) {
-			files = proc->files;
-			if (files)
-				proc->files = NULL;
-		}
-
 		if (defer & BINDER_DEFERRED_FLUSH)
 			binder_deferred_flush(proc);
 
 		if (defer & BINDER_DEFERRED_RELEASE)
 			binder_deferred_release(proc); /* frees proc */
-
-		binder_unlock(__func__);
-		if (files)
-			put_files_struct(files);
 	} while (proc);
 }
 static DECLARE_WORK(binder_deferred_work, binder_deferred_func);
@@ -3241,41 +5172,52 @@ binder_defer_work(struct binder_proc *proc, enum binder_deferred_state defer)
 	mutex_unlock(&binder_deferred_lock);
 }
 
-static void print_binder_transaction(struct seq_file *m, const char *prefix,
-				     struct binder_transaction *t)
+static void print_binder_transaction_ilocked(struct seq_file *m,
+					     struct binder_proc *proc,
+					     const char *prefix,
+					     struct binder_transaction *t)
 {
+	struct binder_proc *to_proc;
+	struct binder_buffer *buffer = t->buffer;
+
+	spin_lock(&t->lock);
+	to_proc = t->to_proc;
 	seq_printf(m,
-		   "%s %d: %p from %d:%d to %d:%d code %x flags %x pri %ld r%d",
+		   "%s %d: %p from %d:%d to %d:%d code %x flags %x pri %d:%d r%d",
 		   prefix, t->debug_id, t,
 		   t->from ? t->from->proc->pid : 0,
 		   t->from ? t->from->pid : 0,
-		   t->to_proc ? t->to_proc->pid : 0,
+		   to_proc ? to_proc->pid : 0,
 		   t->to_thread ? t->to_thread->pid : 0,
-		   t->code, t->flags, t->priority, t->need_reply);
-	if (t->buffer == NULL) {
+		   t->code, t->flags, t->priority.sched_policy,
+		   t->priority.prio, t->need_reply);
+	spin_unlock(&t->lock);
+
+	if (proc != to_proc) {
+		/*
+		 * Can only safely deref buffer if we are holding the
+		 * correct proc inner lock for this node
+		 */
+		seq_puts(m, "\n");
+		return;
+	}
+
+	if (buffer == NULL) {
 		seq_puts(m, " buffer free\n");
 		return;
 	}
-	if (t->buffer->target_node)
-		seq_printf(m, " node %d",
-			   t->buffer->target_node->debug_id);
+	if (buffer->target_node)
+		seq_printf(m, " node %d", buffer->target_node->debug_id);
 	seq_printf(m, " size %zd:%zd data %p\n",
-		   t->buffer->data_size, t->buffer->offsets_size,
-		   t->buffer->data);
-}
-
-static void print_binder_buffer(struct seq_file *m, const char *prefix,
-				struct binder_buffer *buffer)
-{
-	seq_printf(m, "%s %d: %p size %zd:%zd %s\n",
-		   prefix, buffer->debug_id, buffer->data,
 		   buffer->data_size, buffer->offsets_size,
-		   buffer->transaction ? "active" : "delivered");
+		   buffer->data);
 }
 
-static void print_binder_work(struct seq_file *m, const char *prefix,
-			      const char *transaction_prefix,
-			      struct binder_work *w)
+static void print_binder_work_ilocked(struct seq_file *m,
+				     struct binder_proc *proc,
+				     const char *prefix,
+				     const char *transaction_prefix,
+				     struct binder_work *w)
 {
 	struct binder_node *node;
 	struct binder_transaction *t;
@@ -3283,8 +5225,16 @@ static void print_binder_work(struct seq_file *m, const char *prefix,
 	switch (w->type) {
 	case BINDER_WORK_TRANSACTION:
 		t = container_of(w, struct binder_transaction, work);
-		print_binder_transaction(m, transaction_prefix, t);
+		print_binder_transaction_ilocked(
+				m, proc, transaction_prefix, t);
 		break;
+	case BINDER_WORK_RETURN_ERROR: {
+		struct binder_error *e = container_of(
+				w, struct binder_error, work);
+
+		seq_printf(m, "%stransaction error: %u\n",
+			   prefix, e->cmd);
+	} break;
 	case BINDER_WORK_TRANSACTION_COMPLETE:
 		seq_printf(m, "%stransaction complete\n", prefix);
 		break;
@@ -3309,40 +5259,46 @@ static void print_binder_work(struct seq_file *m, const char *prefix,
 	}
 }
 
-static void print_binder_thread(struct seq_file *m,
-				struct binder_thread *thread,
-				int print_always)
+static void print_binder_thread_ilocked(struct seq_file *m,
+					struct binder_thread *thread,
+					int print_always)
 {
 	struct binder_transaction *t;
 	struct binder_work *w;
 	size_t start_pos = m->count;
 	size_t header_pos;
 
-	seq_printf(m, "  thread %d: l %02x\n", thread->pid, thread->looper);
+	seq_printf(m, "  thread %d: l %02x need_return %d tr %d\n",
+			thread->pid, thread->looper,
+			thread->looper_need_return,
+			atomic_read(&thread->tmp_ref));
 	header_pos = m->count;
 	t = thread->transaction_stack;
 	while (t) {
 		if (t->from == thread) {
-			print_binder_transaction(m,
-						 "    outgoing transaction", t);
+			print_binder_transaction_ilocked(m, thread->proc,
+					"    outgoing transaction", t);
 			t = t->from_parent;
 		} else if (t->to_thread == thread) {
-			print_binder_transaction(m,
+			print_binder_transaction_ilocked(m, thread->proc,
 						 "    incoming transaction", t);
 			t = t->to_parent;
 		} else {
-			print_binder_transaction(m, "    bad transaction", t);
+			print_binder_transaction_ilocked(m, thread->proc,
+					"    bad transaction", t);
 			t = NULL;
 		}
 	}
 	list_for_each_entry(w, &thread->todo, entry) {
-		print_binder_work(m, "    ", "    pending transaction", w);
+		print_binder_work_ilocked(m, thread->proc, "    ",
+					  "    pending transaction", w);
 	}
 	if (!print_always && m->count == header_pos)
 		m->count = start_pos;
 }
 
-static void print_binder_node(struct seq_file *m, struct binder_node *node)
+static void print_binder_node_nilocked(struct seq_file *m,
+				       struct binder_node *node)
 {
 	struct binder_ref *ref;
 	struct binder_work *w;
@@ -3352,27 +5308,35 @@ static void print_binder_node(struct seq_file *m, struct binder_node *node)
 	hlist_for_each_entry(ref, &node->refs, node_entry)
 		count++;
 
-	seq_printf(m, "  node %d: u%016llx c%016llx hs %d hw %d ls %d lw %d is %d iw %d",
+	seq_printf(m, "  node %d: u%016llx c%016llx pri %d:%d hs %d hw %d ls %d lw %d is %d iw %d tr %d",
 		   node->debug_id, (u64)node->ptr, (u64)node->cookie,
+		   node->sched_policy, node->min_priority,
 		   node->has_strong_ref, node->has_weak_ref,
 		   node->local_strong_refs, node->local_weak_refs,
-		   node->internal_strong_refs, count);
+		   node->internal_strong_refs, count, node->tmp_refs);
 	if (count) {
 		seq_puts(m, " proc");
 		hlist_for_each_entry(ref, &node->refs, node_entry)
 			seq_printf(m, " %d", ref->proc->pid);
 	}
 	seq_puts(m, "\n");
-	list_for_each_entry(w, &node->async_todo, entry)
-		print_binder_work(m, "    ",
-				  "    pending async transaction", w);
+	if (node->proc) {
+		list_for_each_entry(w, &node->async_todo, entry)
+			print_binder_work_ilocked(m, node->proc, "    ",
+					  "    pending async transaction", w);
+	}
 }
 
-static void print_binder_ref(struct seq_file *m, struct binder_ref *ref)
+static void print_binder_ref_olocked(struct seq_file *m,
+				     struct binder_ref *ref)
 {
-	seq_printf(m, "  ref %d: desc %d %snode %d s %d w %d d %p\n",
-		   ref->debug_id, ref->desc, ref->node->proc ? "" : "dead ",
-		   ref->node->debug_id, ref->strong, ref->weak, ref->death);
+	binder_node_lock(ref->node);
+	seq_printf(m, "  ref %d: desc %d %snode %d s %d w %d d %pK\n",
+		   ref->data.debug_id, ref->data.desc,
+		   ref->node->proc ? "" : "dead ",
+		   ref->node->debug_id, ref->data.strong,
+		   ref->data.weak, ref->death);
+	binder_node_unlock(ref->node);
 }
 
 static void print_binder_proc(struct seq_file *m,
@@ -3382,35 +5346,60 @@ static void print_binder_proc(struct seq_file *m,
 	struct rb_node *n;
 	size_t start_pos = m->count;
 	size_t header_pos;
+	struct binder_node *last_node = NULL;
 
 	seq_printf(m, "proc %d\n", proc->pid);
+	seq_printf(m, "context %s\n", proc->context->name);
 	header_pos = m->count;
 
+	binder_inner_proc_lock(proc);
 	for (n = rb_first(&proc->threads); n != NULL; n = rb_next(n))
-		print_binder_thread(m, rb_entry(n, struct binder_thread,
+		print_binder_thread_ilocked(m, rb_entry(n, struct binder_thread,
 						rb_node), print_all);
+
 	for (n = rb_first(&proc->nodes); n != NULL; n = rb_next(n)) {
 		struct binder_node *node = rb_entry(n, struct binder_node,
 						    rb_node);
-		if (print_all || node->has_async_transaction)
-			print_binder_node(m, node);
+		/*
+		 * take a temporary reference on the node so it
+		 * survives and isn't removed from the tree
+		 * while we print it.
+		 */
+		binder_inc_node_tmpref_ilocked(node);
+		/* Need to drop inner lock to take node lock */
+		binder_inner_proc_unlock(proc);
+		if (last_node)
+			binder_put_node(last_node);
+		binder_node_inner_lock(node);
+		print_binder_node_nilocked(m, node);
+		binder_node_inner_unlock(node);
+		last_node = node;
+		binder_inner_proc_lock(proc);
 	}
+	binder_inner_proc_unlock(proc);
+	if (last_node)
+		binder_put_node(last_node);
+
 	if (print_all) {
+		binder_proc_lock(proc);
 		for (n = rb_first(&proc->refs_by_desc);
 		     n != NULL;
 		     n = rb_next(n))
-			print_binder_ref(m, rb_entry(n, struct binder_ref,
-						     rb_node_desc));
+			print_binder_ref_olocked(m, rb_entry(n,
+							    struct binder_ref,
+							    rb_node_desc));
+		binder_proc_unlock(proc);
 	}
-	for (n = rb_first(&proc->allocated_buffers); n != NULL; n = rb_next(n))
-		print_binder_buffer(m, "  buffer",
-				    rb_entry(n, struct binder_buffer, rb_node));
+	binder_alloc_print_allocated(m, &proc->alloc);
+	binder_inner_proc_lock(proc);
 	list_for_each_entry(w, &proc->todo, entry)
-		print_binder_work(m, "  ", "  pending transaction", w);
+		print_binder_work_ilocked(m, proc, "  ",
+					  "  pending transaction", w);
 	list_for_each_entry(w, &proc->delivered_death, entry) {
 		seq_puts(m, "  has delivered dead binder\n");
 		break;
 	}
+	binder_inner_proc_unlock(proc);
 	if (!print_all && m->count == header_pos)
 		m->count = start_pos;
 }
@@ -3453,7 +5442,9 @@ static const char * const binder_command_strings[] = {
 	"BC_EXIT_LOOPER",
 	"BC_REQUEST_DEATH_NOTIFICATION",
 	"BC_CLEAR_DEATH_NOTIFICATION",
-	"BC_DEAD_BINDER_DONE"
+	"BC_DEAD_BINDER_DONE",
+	"BC_TRANSACTION_SG",
+	"BC_REPLY_SG",
 };
 
 static const char * const binder_objstat_strings[] = {
@@ -3474,17 +5465,21 @@ static void print_binder_stats(struct seq_file *m, const char *prefix,
 	BUILD_BUG_ON(ARRAY_SIZE(stats->bc) !=
 		     ARRAY_SIZE(binder_command_strings));
 	for (i = 0; i < ARRAY_SIZE(stats->bc); i++) {
-		if (stats->bc[i])
+		int temp = atomic_read(&stats->bc[i]);
+
+		if (temp)
 			seq_printf(m, "%s%s: %d\n", prefix,
-				   binder_command_strings[i], stats->bc[i]);
+				   binder_command_strings[i], temp);
 	}
 
 	BUILD_BUG_ON(ARRAY_SIZE(stats->br) !=
 		     ARRAY_SIZE(binder_return_strings));
 	for (i = 0; i < ARRAY_SIZE(stats->br); i++) {
-		if (stats->br[i])
+		int temp = atomic_read(&stats->br[i]);
+
+		if (temp)
 			seq_printf(m, "%s%s: %d\n", prefix,
-				   binder_return_strings[i], stats->br[i]);
+				   binder_return_strings[i], temp);
 	}
 
 	BUILD_BUG_ON(ARRAY_SIZE(stats->obj_created) !=
@@ -3492,11 +5487,15 @@ static void print_binder_stats(struct seq_file *m, const char *prefix,
 	BUILD_BUG_ON(ARRAY_SIZE(stats->obj_created) !=
 		     ARRAY_SIZE(stats->obj_deleted));
 	for (i = 0; i < ARRAY_SIZE(stats->obj_created); i++) {
-		if (stats->obj_created[i] || stats->obj_deleted[i])
-			seq_printf(m, "%s%s: active %d total %d\n", prefix,
+		int created = atomic_read(&stats->obj_created[i]);
+		int deleted = atomic_read(&stats->obj_deleted[i]);
+
+		if (created || deleted)
+			seq_printf(m, "%s%s: active %d total %d\n",
+				prefix,
 				binder_objstat_strings[i],
-				stats->obj_created[i] - stats->obj_deleted[i],
-				stats->obj_created[i]);
+				created - deleted,
+				created);
 	}
 }
 
@@ -3504,50 +5503,61 @@ static void print_binder_proc_stats(struct seq_file *m,
 				    struct binder_proc *proc)
 {
 	struct binder_work *w;
+	struct binder_thread *thread;
 	struct rb_node *n;
-	int count, strong, weak;
+	int count, strong, weak, ready_threads;
+	size_t free_async_space =
+		binder_alloc_get_free_async_space(&proc->alloc);
 
 	seq_printf(m, "proc %d\n", proc->pid);
+	seq_printf(m, "context %s\n", proc->context->name);
 	count = 0;
+	ready_threads = 0;
+	binder_inner_proc_lock(proc);
 	for (n = rb_first(&proc->threads); n != NULL; n = rb_next(n))
 		count++;
+
+	list_for_each_entry(thread, &proc->waiting_threads, waiting_thread_node)
+		ready_threads++;
+
 	seq_printf(m, "  threads: %d\n", count);
 	seq_printf(m, "  requested threads: %d+%d/%d\n"
 			"  ready threads %d\n"
 			"  free async space %zd\n", proc->requested_threads,
 			proc->requested_threads_started, proc->max_threads,
-			proc->ready_threads, proc->free_async_space);
+			ready_threads,
+			free_async_space);
 	count = 0;
 	for (n = rb_first(&proc->nodes); n != NULL; n = rb_next(n))
 		count++;
+	binder_inner_proc_unlock(proc);
 	seq_printf(m, "  nodes: %d\n", count);
 	count = 0;
 	strong = 0;
 	weak = 0;
+	binder_proc_lock(proc);
 	for (n = rb_first(&proc->refs_by_desc); n != NULL; n = rb_next(n)) {
 		struct binder_ref *ref = rb_entry(n, struct binder_ref,
 						  rb_node_desc);
 		count++;
-		strong += ref->strong;
-		weak += ref->weak;
+		strong += ref->data.strong;
+		weak += ref->data.weak;
 	}
+	binder_proc_unlock(proc);
 	seq_printf(m, "  refs: %d s %d w %d\n", count, strong, weak);
 
-	count = 0;
-	for (n = rb_first(&proc->allocated_buffers); n != NULL; n = rb_next(n))
-		count++;
+	count = binder_alloc_get_allocated_count(&proc->alloc);
 	seq_printf(m, "  buffers: %d\n", count);
 
+	binder_alloc_print_pages(m, &proc->alloc);
+
 	count = 0;
+	binder_inner_proc_lock(proc);
 	list_for_each_entry(w, &proc->todo, entry) {
-		switch (w->type) {
-		case BINDER_WORK_TRANSACTION:
+		if (w->type == BINDER_WORK_TRANSACTION)
 			count++;
-			break;
-		default:
-			break;
-		}
 	}
+	binder_inner_proc_unlock(proc);
 	seq_printf(m, "  pending transactions: %d\n", count);
 
 	print_binder_stats(m, "  ", &proc->stats);
@@ -3558,96 +5568,131 @@ static int binder_state_show(struct seq_file *m, void *unused)
 {
 	struct binder_proc *proc;
 	struct binder_node *node;
-	int do_lock = !binder_debug_no_lock;
-
-	if (do_lock)
-		binder_lock(__func__);
+	struct binder_node *last_node = NULL;
 
 	seq_puts(m, "binder state:\n");
 
+	spin_lock(&binder_dead_nodes_lock);
 	if (!hlist_empty(&binder_dead_nodes))
 		seq_puts(m, "dead nodes:\n");
-	hlist_for_each_entry(node, &binder_dead_nodes, dead_node)
-		print_binder_node(m, node);
+	hlist_for_each_entry(node, &binder_dead_nodes, dead_node) {
+		/*
+		 * take a temporary reference on the node so it
+		 * survives and isn't removed from the list
+		 * while we print it.
+		 */
+		node->tmp_refs++;
+		spin_unlock(&binder_dead_nodes_lock);
+		if (last_node)
+			binder_put_node(last_node);
+		binder_node_lock(node);
+		print_binder_node_nilocked(m, node);
+		binder_node_unlock(node);
+		last_node = node;
+		spin_lock(&binder_dead_nodes_lock);
+	}
+	spin_unlock(&binder_dead_nodes_lock);
+	if (last_node)
+		binder_put_node(last_node);
 
+	mutex_lock(&binder_procs_lock);
 	hlist_for_each_entry(proc, &binder_procs, proc_node)
 		print_binder_proc(m, proc, 1);
-	if (do_lock)
-		binder_unlock(__func__);
+	mutex_unlock(&binder_procs_lock);
+
 	return 0;
 }
 
 static int binder_stats_show(struct seq_file *m, void *unused)
 {
 	struct binder_proc *proc;
-	int do_lock = !binder_debug_no_lock;
-
-	if (do_lock)
-		binder_lock(__func__);
 
 	seq_puts(m, "binder stats:\n");
 
 	print_binder_stats(m, "", &binder_stats);
 
+	mutex_lock(&binder_procs_lock);
 	hlist_for_each_entry(proc, &binder_procs, proc_node)
 		print_binder_proc_stats(m, proc);
-	if (do_lock)
-		binder_unlock(__func__);
+	mutex_unlock(&binder_procs_lock);
+
 	return 0;
 }
 
 static int binder_transactions_show(struct seq_file *m, void *unused)
 {
 	struct binder_proc *proc;
-	int do_lock = !binder_debug_no_lock;
-
-	if (do_lock)
-		binder_lock(__func__);
 
 	seq_puts(m, "binder transactions:\n");
+	mutex_lock(&binder_procs_lock);
 	hlist_for_each_entry(proc, &binder_procs, proc_node)
 		print_binder_proc(m, proc, 0);
-	if (do_lock)
-		binder_unlock(__func__);
+	mutex_unlock(&binder_procs_lock);
+
 	return 0;
 }
 
 static int binder_proc_show(struct seq_file *m, void *unused)
 {
-	struct binder_proc *proc = m->private;
-	int do_lock = !binder_debug_no_lock;
+	struct binder_proc *itr;
+	int pid = (unsigned long)m->private;
+
+	mutex_lock(&binder_procs_lock);
+	hlist_for_each_entry(itr, &binder_procs, proc_node) {
+		if (itr->pid == pid) {
+			seq_puts(m, "binder proc state:\n");
+			print_binder_proc(m, itr, 1);
+		}
+	}
+	mutex_unlock(&binder_procs_lock);
 
-	if (do_lock)
-		binder_lock(__func__);
-	seq_puts(m, "binder proc state:\n");
-	print_binder_proc(m, proc, 1);
-	if (do_lock)
-		binder_unlock(__func__);
 	return 0;
 }
 
 static void print_binder_transaction_log_entry(struct seq_file *m,
 					struct binder_transaction_log_entry *e)
 {
+	int debug_id = READ_ONCE(e->debug_id_done);
+	/*
+	 * read barrier to guarantee debug_id_done read before
+	 * we print the log values
+	 */
+	smp_rmb();
 	seq_printf(m,
-		   "%d: %s from %d:%d to %d:%d node %d handle %d size %d:%d\n",
+		   "%d: %s from %d:%d to %d:%d context %s node %d handle %d size %d:%d ret %d/%d l=%d",
 		   e->debug_id, (e->call_type == 2) ? "reply" :
 		   ((e->call_type == 1) ? "async" : "call "), e->from_proc,
-		   e->from_thread, e->to_proc, e->to_thread, e->to_node,
-		   e->target_handle, e->data_size, e->offsets_size);
+		   e->from_thread, e->to_proc, e->to_thread, e->context_name,
+		   e->to_node, e->target_handle, e->data_size, e->offsets_size,
+		   e->return_error, e->return_error_param,
+		   e->return_error_line);
+	/*
+	 * read-barrier to guarantee read of debug_id_done after
+	 * done printing the fields of the entry
+	 */
+	smp_rmb();
+	seq_printf(m, debug_id && debug_id == READ_ONCE(e->debug_id_done) ?
+			"\n" : " (incomplete)\n");
 }
 
 static int binder_transaction_log_show(struct seq_file *m, void *unused)
 {
 	struct binder_transaction_log *log = m->private;
+	unsigned int log_cur = atomic_read(&log->cur);
+	unsigned int count;
+	unsigned int cur;
 	int i;
 
-	if (log->full) {
-		for (i = log->next; i < ARRAY_SIZE(log->entry); i++)
-			print_binder_transaction_log_entry(m, &log->entry[i]);
+	count = log_cur + 1;
+	cur = count < ARRAY_SIZE(log->entry) && !log->full ?
+		0 : count % ARRAY_SIZE(log->entry);
+	if (count > ARRAY_SIZE(log->entry) || log->full)
+		count = ARRAY_SIZE(log->entry);
+	for (i = 0; i < count; i++) {
+		unsigned int index = cur++ % ARRAY_SIZE(log->entry);
+
+		print_binder_transaction_log_entry(m, &log->entry[index]);
 	}
-	for (i = 0; i < log->next; i++)
-		print_binder_transaction_log_entry(m, &log->entry[i]);
 	return 0;
 }
 
@@ -3662,21 +5707,50 @@ static const struct file_operations binder_fops = {
 	.release = binder_release,
 };
 
-static struct miscdevice binder_miscdev = {
-	.minor = MISC_DYNAMIC_MINOR,
-	.name = "binder",
-	.fops = &binder_fops
-};
-
 BINDER_DEBUG_ENTRY(state);
 BINDER_DEBUG_ENTRY(stats);
 BINDER_DEBUG_ENTRY(transactions);
 BINDER_DEBUG_ENTRY(transaction_log);
 
+static int __init init_binder_device(const char *name)
+{
+	int ret;
+	struct binder_device *binder_device;
+
+	binder_device = kzalloc(sizeof(*binder_device), GFP_KERNEL);
+	if (!binder_device)
+		return -ENOMEM;
+
+	binder_device->miscdev.fops = &binder_fops;
+	binder_device->miscdev.minor = MISC_DYNAMIC_MINOR;
+	binder_device->miscdev.name = name;
+
+	binder_device->context.binder_context_mgr_uid = INVALID_UID;
+	binder_device->context.name = name;
+	mutex_init(&binder_device->context.context_mgr_node_lock);
+
+	ret = misc_register(&binder_device->miscdev);
+	if (ret < 0) {
+		kfree(binder_device);
+		return ret;
+	}
+
+	hlist_add_head(&binder_device->hlist, &binder_devices);
+
+	return ret;
+}
+
 static int __init binder_init(void)
 {
 	int ret;
+	char *device_name, *device_names;
+	struct binder_device *device;
+	struct hlist_node *tmp;
 
+	binder_alloc_shrinker_init();
+
+	atomic_set(&binder_transaction_log.cur, ~0U);
+	atomic_set(&binder_transaction_log_failed.cur, ~0U);
 	binder_deferred_workqueue = create_singlethread_workqueue("binder");
 	if (!binder_deferred_workqueue)
 		return -ENOMEM;
@@ -3685,7 +5759,7 @@ static int __init binder_init(void)
 	if (binder_debugfs_dir_entry_root)
 		binder_debugfs_dir_entry_proc = debugfs_create_dir("proc",
 						 binder_debugfs_dir_entry_root);
-	ret = misc_register(&binder_miscdev);
+
 	if (binder_debugfs_dir_entry_root) {
 		debugfs_create_file("state",
 				    S_IRUGO,
@@ -3713,6 +5787,37 @@ static int __init binder_init(void)
 				    &binder_transaction_log_failed,
 				    &binder_transaction_log_fops);
 	}
+
+	/*
+	 * Copy the module_parameter string, because we don't want to
+	 * tokenize it in-place.
+	 */
+	device_names = kzalloc(strlen(binder_devices_param) + 1, GFP_KERNEL);
+	if (!device_names) {
+		ret = -ENOMEM;
+		goto err_alloc_device_names_failed;
+	}
+	strcpy(device_names, binder_devices_param);
+
+	while ((device_name = strsep(&device_names, ","))) {
+		ret = init_binder_device(device_name);
+		if (ret)
+			goto err_init_binder_device_failed;
+	}
+
+	return ret;
+
+err_init_binder_device_failed:
+	hlist_for_each_entry_safe(device, tmp, &binder_devices, hlist) {
+		misc_deregister(&device->miscdev);
+		hlist_del(&device->hlist);
+		kfree(device);
+	}
+err_alloc_device_names_failed:
+	debugfs_remove_recursive(binder_debugfs_dir_entry_root);
+
+	destroy_workqueue(binder_deferred_workqueue);
+
 	return ret;
 }
 
diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c
new file mode 100644
index 000000000000..3a4279d219f7
--- /dev/null
+++ b/drivers/android/binder_alloc.c
@@ -0,0 +1,1017 @@
+/* binder_alloc.c
+ *
+ * Android IPC Subsystem
+ *
+ * Copyright (C) 2007-2017 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <asm/cacheflush.h>
+#include <linux/list.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/rtmutex.h>
+#include <linux/rbtree.h>
+#include <linux/seq_file.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/list_lru.h>
+#include "binder_alloc.h"
+#include "binder_trace.h"
+
+struct list_lru binder_alloc_lru;
+
+static DEFINE_MUTEX(binder_alloc_mmap_lock);
+
+enum {
+	BINDER_DEBUG_OPEN_CLOSE             = 1U << 1,
+	BINDER_DEBUG_BUFFER_ALLOC           = 1U << 2,
+	BINDER_DEBUG_BUFFER_ALLOC_ASYNC     = 1U << 3,
+};
+static uint32_t binder_alloc_debug_mask;
+
+module_param_named(debug_mask, binder_alloc_debug_mask,
+		   uint, S_IWUSR | S_IRUGO);
+
+#define binder_alloc_debug(mask, x...) \
+	do { \
+		if (binder_alloc_debug_mask & mask) \
+			pr_info(x); \
+	} while (0)
+
+static struct binder_buffer *binder_buffer_next(struct binder_buffer *buffer)
+{
+	return list_entry(buffer->entry.next, struct binder_buffer, entry);
+}
+
+static struct binder_buffer *binder_buffer_prev(struct binder_buffer *buffer)
+{
+	return list_entry(buffer->entry.prev, struct binder_buffer, entry);
+}
+
+static size_t binder_alloc_buffer_size(struct binder_alloc *alloc,
+				       struct binder_buffer *buffer)
+{
+	if (list_is_last(&buffer->entry, &alloc->buffers))
+		return (u8 *)alloc->buffer +
+			alloc->buffer_size - (u8 *)buffer->data;
+	return (u8 *)binder_buffer_next(buffer)->data - (u8 *)buffer->data;
+}
+
+static void binder_insert_free_buffer(struct binder_alloc *alloc,
+				      struct binder_buffer *new_buffer)
+{
+	struct rb_node **p = &alloc->free_buffers.rb_node;
+	struct rb_node *parent = NULL;
+	struct binder_buffer *buffer;
+	size_t buffer_size;
+	size_t new_buffer_size;
+
+	BUG_ON(!new_buffer->free);
+
+	new_buffer_size = binder_alloc_buffer_size(alloc, new_buffer);
+
+	binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+		     "%d: add free buffer, size %zd, at %pK\n",
+		      alloc->pid, new_buffer_size, new_buffer);
+
+	while (*p) {
+		parent = *p;
+		buffer = rb_entry(parent, struct binder_buffer, rb_node);
+		BUG_ON(!buffer->free);
+
+		buffer_size = binder_alloc_buffer_size(alloc, buffer);
+
+		if (new_buffer_size < buffer_size)
+			p = &parent->rb_left;
+		else
+			p = &parent->rb_right;
+	}
+	rb_link_node(&new_buffer->rb_node, parent, p);
+	rb_insert_color(&new_buffer->rb_node, &alloc->free_buffers);
+}
+
+static void binder_insert_allocated_buffer_locked(
+		struct binder_alloc *alloc, struct binder_buffer *new_buffer)
+{
+	struct rb_node **p = &alloc->allocated_buffers.rb_node;
+	struct rb_node *parent = NULL;
+	struct binder_buffer *buffer;
+
+	BUG_ON(new_buffer->free);
+
+	while (*p) {
+		parent = *p;
+		buffer = rb_entry(parent, struct binder_buffer, rb_node);
+		BUG_ON(buffer->free);
+
+		if (new_buffer->data < buffer->data)
+			p = &parent->rb_left;
+		else if (new_buffer->data > buffer->data)
+			p = &parent->rb_right;
+		else
+			BUG();
+	}
+	rb_link_node(&new_buffer->rb_node, parent, p);
+	rb_insert_color(&new_buffer->rb_node, &alloc->allocated_buffers);
+}
+
+static struct binder_buffer *binder_alloc_prepare_to_free_locked(
+		struct binder_alloc *alloc,
+		uintptr_t user_ptr)
+{
+	struct rb_node *n = alloc->allocated_buffers.rb_node;
+	struct binder_buffer *buffer;
+	void *kern_ptr;
+
+	kern_ptr = (void *)(user_ptr - alloc->user_buffer_offset);
+
+	while (n) {
+		buffer = rb_entry(n, struct binder_buffer, rb_node);
+		BUG_ON(buffer->free);
+
+		if (kern_ptr < buffer->data)
+			n = n->rb_left;
+		else if (kern_ptr > buffer->data)
+			n = n->rb_right;
+		else {
+			/*
+			 * Guard against user threads attempting to
+			 * free the buffer twice
+			 */
+			if (buffer->free_in_progress) {
+				pr_err("%d:%d FREE_BUFFER u%016llx user freed buffer twice\n",
+				       alloc->pid, current->pid, (u64)user_ptr);
+				return NULL;
+			}
+			buffer->free_in_progress = 1;
+			return buffer;
+		}
+	}
+	return NULL;
+}
+
+/**
+ * binder_alloc_buffer_lookup() - get buffer given user ptr
+ * @alloc:	binder_alloc for this proc
+ * @user_ptr:	User pointer to buffer data
+ *
+ * Validate userspace pointer to buffer data and return buffer corresponding to
+ * that user pointer. Search the rb tree for buffer that matches user data
+ * pointer.
+ *
+ * Return:	Pointer to buffer or NULL
+ */
+struct binder_buffer *binder_alloc_prepare_to_free(struct binder_alloc *alloc,
+						   uintptr_t user_ptr)
+{
+	struct binder_buffer *buffer;
+
+	mutex_lock(&alloc->mutex);
+	buffer = binder_alloc_prepare_to_free_locked(alloc, user_ptr);
+	mutex_unlock(&alloc->mutex);
+	return buffer;
+}
+
+static int binder_update_page_range(struct binder_alloc *alloc, int allocate,
+				    void *start, void *end)
+{
+	void *page_addr;
+	unsigned long user_page_addr;
+	struct binder_lru_page *page;
+	struct vm_area_struct *vma = NULL;
+	struct mm_struct *mm = NULL;
+	bool need_mm = false;
+
+	binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+		     "%d: %s pages %pK-%pK\n", alloc->pid,
+		     allocate ? "allocate" : "free", start, end);
+
+	if (end <= start)
+		return 0;
+
+	trace_binder_update_page_range(alloc, allocate, start, end);
+
+	if (allocate == 0)
+		goto free_range;
+
+	for (page_addr = start; page_addr < end; page_addr += PAGE_SIZE) {
+		page = &alloc->pages[(page_addr - alloc->buffer) / PAGE_SIZE];
+		if (!page->page_ptr) {
+			need_mm = true;
+			break;
+		}
+	}
+
+	/* Same as mmget_not_zero() in later kernel versions */
+	if (need_mm && atomic_inc_not_zero(&alloc->vma_vm_mm->mm_users))
+		mm = alloc->vma_vm_mm;
+
+	if (mm) {
+		down_write(&mm->mmap_sem);
+		vma = alloc->vma;
+	}
+
+	if (!vma && need_mm) {
+		pr_err("%d: binder_alloc_buf failed to map pages in userspace, no vma\n",
+			alloc->pid);
+		goto err_no_vma;
+	}
+
+	for (page_addr = start; page_addr < end; page_addr += PAGE_SIZE) {
+		int ret;
+		bool on_lru;
+		size_t index;
+
+		index = (page_addr - alloc->buffer) / PAGE_SIZE;
+		page = &alloc->pages[index];
+
+		if (page->page_ptr) {
+			trace_binder_alloc_lru_start(alloc, index);
+
+			on_lru = list_lru_del(&binder_alloc_lru, &page->lru);
+			WARN_ON(!on_lru);
+
+			trace_binder_alloc_lru_end(alloc, index);
+			continue;
+		}
+
+		if (WARN_ON(!vma))
+			goto err_page_ptr_cleared;
+
+		trace_binder_alloc_page_start(alloc, index);
+		page->page_ptr = alloc_page(GFP_KERNEL |
+					    __GFP_HIGHMEM |
+					    __GFP_ZERO);
+		if (!page->page_ptr) {
+			pr_err("%d: binder_alloc_buf failed for page at %pK\n",
+				alloc->pid, page_addr);
+			goto err_alloc_page_failed;
+		}
+		page->alloc = alloc;
+		INIT_LIST_HEAD(&page->lru);
+
+		ret = map_kernel_range_noflush((unsigned long)page_addr,
+					       PAGE_SIZE, PAGE_KERNEL,
+					       &page->page_ptr);
+		flush_cache_vmap((unsigned long)page_addr,
+				(unsigned long)page_addr + PAGE_SIZE);
+		if (ret != 1) {
+			pr_err("%d: binder_alloc_buf failed to map page at %pK in kernel\n",
+			       alloc->pid, page_addr);
+			goto err_map_kernel_failed;
+		}
+		user_page_addr =
+			(uintptr_t)page_addr + alloc->user_buffer_offset;
+		ret = vm_insert_page(vma, user_page_addr, page[0].page_ptr);
+		if (ret) {
+			pr_err("%d: binder_alloc_buf failed to map page at %lx in userspace\n",
+			       alloc->pid, user_page_addr);
+			goto err_vm_insert_page_failed;
+		}
+
+		if (index + 1 > alloc->pages_high)
+			alloc->pages_high = index + 1;
+
+		trace_binder_alloc_page_end(alloc, index);
+		/* vm_insert_page does not seem to increment the refcount */
+	}
+	if (mm) {
+		up_write(&mm->mmap_sem);
+		mmput(mm);
+	}
+	return 0;
+
+free_range:
+	for (page_addr = end - PAGE_SIZE; page_addr >= start;
+	     page_addr -= PAGE_SIZE) {
+		bool ret;
+		size_t index;
+
+		index = (page_addr - alloc->buffer) / PAGE_SIZE;
+		page = &alloc->pages[index];
+
+		trace_binder_free_lru_start(alloc, index);
+
+		ret = list_lru_add(&binder_alloc_lru, &page->lru);
+		WARN_ON(!ret);
+
+		trace_binder_free_lru_end(alloc, index);
+		continue;
+
+err_vm_insert_page_failed:
+		unmap_kernel_range((unsigned long)page_addr, PAGE_SIZE);
+err_map_kernel_failed:
+		__free_page(page->page_ptr);
+		page->page_ptr = NULL;
+err_alloc_page_failed:
+err_page_ptr_cleared:
+		;
+	}
+err_no_vma:
+	if (mm) {
+		up_write(&mm->mmap_sem);
+		mmput(mm);
+	}
+	return vma ? -ENOMEM : -ESRCH;
+}
+
+struct binder_buffer *binder_alloc_new_buf_locked(struct binder_alloc *alloc,
+						  size_t data_size,
+						  size_t offsets_size,
+						  size_t extra_buffers_size,
+						  int is_async)
+{
+	struct rb_node *n = alloc->free_buffers.rb_node;
+	struct binder_buffer *buffer;
+	size_t buffer_size;
+	struct rb_node *best_fit = NULL;
+	void *has_page_addr;
+	void *end_page_addr;
+	size_t size, data_offsets_size;
+	int ret;
+
+	if (alloc->vma == NULL) {
+		pr_err("%d: binder_alloc_buf, no vma\n",
+		       alloc->pid);
+		return ERR_PTR(-ESRCH);
+	}
+
+	data_offsets_size = ALIGN(data_size, sizeof(void *)) +
+		ALIGN(offsets_size, sizeof(void *));
+
+	if (data_offsets_size < data_size || data_offsets_size < offsets_size) {
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+				"%d: got transaction with invalid size %zd-%zd\n",
+				alloc->pid, data_size, offsets_size);
+		return ERR_PTR(-EINVAL);
+	}
+	size = data_offsets_size + ALIGN(extra_buffers_size, sizeof(void *));
+	if (size < data_offsets_size || size < extra_buffers_size) {
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+				"%d: got transaction with invalid extra_buffers_size %zd\n",
+				alloc->pid, extra_buffers_size);
+		return ERR_PTR(-EINVAL);
+	}
+	if (is_async &&
+	    alloc->free_async_space < size + sizeof(struct binder_buffer)) {
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+			     "%d: binder_alloc_buf size %zd failed, no async space left\n",
+			      alloc->pid, size);
+		return ERR_PTR(-ENOSPC);
+	}
+
+	/* Pad 0-size buffers so they get assigned unique addresses */
+	size = max(size, sizeof(void *));
+
+	while (n) {
+		buffer = rb_entry(n, struct binder_buffer, rb_node);
+		BUG_ON(!buffer->free);
+		buffer_size = binder_alloc_buffer_size(alloc, buffer);
+
+		if (size < buffer_size) {
+			best_fit = n;
+			n = n->rb_left;
+		} else if (size > buffer_size)
+			n = n->rb_right;
+		else {
+			best_fit = n;
+			break;
+		}
+	}
+	if (best_fit == NULL) {
+		size_t allocated_buffers = 0;
+		size_t largest_alloc_size = 0;
+		size_t total_alloc_size = 0;
+		size_t free_buffers = 0;
+		size_t largest_free_size = 0;
+		size_t total_free_size = 0;
+
+		for (n = rb_first(&alloc->allocated_buffers); n != NULL;
+		     n = rb_next(n)) {
+			buffer = rb_entry(n, struct binder_buffer, rb_node);
+			buffer_size = binder_alloc_buffer_size(alloc, buffer);
+			allocated_buffers++;
+			total_alloc_size += buffer_size;
+			if (buffer_size > largest_alloc_size)
+				largest_alloc_size = buffer_size;
+		}
+		for (n = rb_first(&alloc->free_buffers); n != NULL;
+		     n = rb_next(n)) {
+			buffer = rb_entry(n, struct binder_buffer, rb_node);
+			buffer_size = binder_alloc_buffer_size(alloc, buffer);
+			free_buffers++;
+			total_free_size += buffer_size;
+			if (buffer_size > largest_free_size)
+				largest_free_size = buffer_size;
+		}
+		pr_err("%d: binder_alloc_buf size %zd failed, no address space\n",
+			alloc->pid, size);
+		pr_err("allocated: %zd (num: %zd largest: %zd), free: %zd (num: %zd largest: %zd)\n",
+		       total_alloc_size, allocated_buffers, largest_alloc_size,
+		       total_free_size, free_buffers, largest_free_size);
+		return ERR_PTR(-ENOSPC);
+	}
+	if (n == NULL) {
+		buffer = rb_entry(best_fit, struct binder_buffer, rb_node);
+		buffer_size = binder_alloc_buffer_size(alloc, buffer);
+	}
+
+	binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+		     "%d: binder_alloc_buf size %zd got buffer %pK size %zd\n",
+		      alloc->pid, size, buffer, buffer_size);
+
+	has_page_addr =
+		(void *)(((uintptr_t)buffer->data + buffer_size) & PAGE_MASK);
+	WARN_ON(n && buffer_size != size);
+	end_page_addr =
+		(void *)PAGE_ALIGN((uintptr_t)buffer->data + size);
+	if (end_page_addr > has_page_addr)
+		end_page_addr = has_page_addr;
+	ret = binder_update_page_range(alloc, 1,
+	    (void *)PAGE_ALIGN((uintptr_t)buffer->data), end_page_addr);
+	if (ret)
+		return ERR_PTR(ret);
+
+	if (buffer_size != size) {
+		struct binder_buffer *new_buffer;
+
+		new_buffer = kzalloc(sizeof(*buffer), GFP_KERNEL);
+		if (!new_buffer) {
+			pr_err("%s: %d failed to alloc new buffer struct\n",
+			       __func__, alloc->pid);
+			goto err_alloc_buf_struct_failed;
+		}
+		new_buffer->data = (u8 *)buffer->data + size;
+		list_add(&new_buffer->entry, &buffer->entry);
+		new_buffer->free = 1;
+		binder_insert_free_buffer(alloc, new_buffer);
+	}
+
+	rb_erase(best_fit, &alloc->free_buffers);
+	buffer->free = 0;
+	buffer->free_in_progress = 0;
+	binder_insert_allocated_buffer_locked(alloc, buffer);
+	binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+		     "%d: binder_alloc_buf size %zd got %pK\n",
+		      alloc->pid, size, buffer);
+	buffer->data_size = data_size;
+	buffer->offsets_size = offsets_size;
+	buffer->async_transaction = is_async;
+	buffer->extra_buffers_size = extra_buffers_size;
+	if (is_async) {
+		alloc->free_async_space -= size + sizeof(struct binder_buffer);
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC_ASYNC,
+			     "%d: binder_alloc_buf size %zd async free %zd\n",
+			      alloc->pid, size, alloc->free_async_space);
+	}
+	return buffer;
+
+err_alloc_buf_struct_failed:
+	binder_update_page_range(alloc, 0,
+				 (void *)PAGE_ALIGN((uintptr_t)buffer->data),
+				 end_page_addr);
+	return ERR_PTR(-ENOMEM);
+}
+
+/**
+ * binder_alloc_new_buf() - Allocate a new binder buffer
+ * @alloc:              binder_alloc for this proc
+ * @data_size:          size of user data buffer
+ * @offsets_size:       user specified buffer offset
+ * @extra_buffers_size: size of extra space for meta-data (eg, security context)
+ * @is_async:           buffer for async transaction
+ *
+ * Allocate a new buffer given the requested sizes. Returns
+ * the kernel version of the buffer pointer. The size allocated
+ * is the sum of the three given sizes (each rounded up to
+ * pointer-sized boundary)
+ *
+ * Return:	The allocated buffer or %NULL if error
+ */
+struct binder_buffer *binder_alloc_new_buf(struct binder_alloc *alloc,
+					   size_t data_size,
+					   size_t offsets_size,
+					   size_t extra_buffers_size,
+					   int is_async)
+{
+	struct binder_buffer *buffer;
+
+	mutex_lock(&alloc->mutex);
+	buffer = binder_alloc_new_buf_locked(alloc, data_size, offsets_size,
+					     extra_buffers_size, is_async);
+	mutex_unlock(&alloc->mutex);
+	return buffer;
+}
+
+static void *buffer_start_page(struct binder_buffer *buffer)
+{
+	return (void *)((uintptr_t)buffer->data & PAGE_MASK);
+}
+
+static void *prev_buffer_end_page(struct binder_buffer *buffer)
+{
+	return (void *)(((uintptr_t)(buffer->data) - 1) & PAGE_MASK);
+}
+
+static void binder_delete_free_buffer(struct binder_alloc *alloc,
+				      struct binder_buffer *buffer)
+{
+	struct binder_buffer *prev, *next = NULL;
+	bool to_free = true;
+	BUG_ON(alloc->buffers.next == &buffer->entry);
+	prev = binder_buffer_prev(buffer);
+	BUG_ON(!prev->free);
+	if (prev_buffer_end_page(prev) == buffer_start_page(buffer)) {
+		to_free = false;
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+				   "%d: merge free, buffer %pK share page with %pK\n",
+				   alloc->pid, buffer->data, prev->data);
+	}
+
+	if (!list_is_last(&buffer->entry, &alloc->buffers)) {
+		next = binder_buffer_next(buffer);
+		if (buffer_start_page(next) == buffer_start_page(buffer)) {
+			to_free = false;
+			binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+					   "%d: merge free, buffer %pK share page with %pK\n",
+					   alloc->pid,
+					   buffer->data,
+					   next->data);
+		}
+	}
+
+	if (PAGE_ALIGNED(buffer->data)) {
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+				   "%d: merge free, buffer start %pK is page aligned\n",
+				   alloc->pid, buffer->data);
+		to_free = false;
+	}
+
+	if (to_free) {
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+				   "%d: merge free, buffer %pK do not share page with %pK or %pK\n",
+				   alloc->pid, buffer->data,
+				   prev->data, next ? next->data : NULL);
+		binder_update_page_range(alloc, 0, buffer_start_page(buffer),
+					 buffer_start_page(buffer) + PAGE_SIZE);
+	}
+	list_del(&buffer->entry);
+	kfree(buffer);
+}
+
+static void binder_free_buf_locked(struct binder_alloc *alloc,
+				   struct binder_buffer *buffer)
+{
+	size_t size, buffer_size;
+
+	buffer_size = binder_alloc_buffer_size(alloc, buffer);
+
+	size = ALIGN(buffer->data_size, sizeof(void *)) +
+		ALIGN(buffer->offsets_size, sizeof(void *)) +
+		ALIGN(buffer->extra_buffers_size, sizeof(void *));
+
+	binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+		     "%d: binder_free_buf %pK size %zd buffer_size %zd\n",
+		      alloc->pid, buffer, size, buffer_size);
+
+	BUG_ON(buffer->free);
+	BUG_ON(size > buffer_size);
+	BUG_ON(buffer->transaction != NULL);
+	BUG_ON(buffer->data < alloc->buffer);
+	BUG_ON(buffer->data > alloc->buffer + alloc->buffer_size);
+
+	if (buffer->async_transaction) {
+		alloc->free_async_space += size + sizeof(struct binder_buffer);
+
+		binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC_ASYNC,
+			     "%d: binder_free_buf size %zd async free %zd\n",
+			      alloc->pid, size, alloc->free_async_space);
+	}
+
+	binder_update_page_range(alloc, 0,
+		(void *)PAGE_ALIGN((uintptr_t)buffer->data),
+		(void *)(((uintptr_t)buffer->data + buffer_size) & PAGE_MASK));
+
+	rb_erase(&buffer->rb_node, &alloc->allocated_buffers);
+	buffer->free = 1;
+	if (!list_is_last(&buffer->entry, &alloc->buffers)) {
+		struct binder_buffer *next = binder_buffer_next(buffer);
+
+		if (next->free) {
+			rb_erase(&next->rb_node, &alloc->free_buffers);
+			binder_delete_free_buffer(alloc, next);
+		}
+	}
+	if (alloc->buffers.next != &buffer->entry) {
+		struct binder_buffer *prev = binder_buffer_prev(buffer);
+
+		if (prev->free) {
+			binder_delete_free_buffer(alloc, buffer);
+			rb_erase(&prev->rb_node, &alloc->free_buffers);
+			buffer = prev;
+		}
+	}
+	binder_insert_free_buffer(alloc, buffer);
+}
+
+/**
+ * binder_alloc_free_buf() - free a binder buffer
+ * @alloc:	binder_alloc for this proc
+ * @buffer:	kernel pointer to buffer
+ *
+ * Free the buffer allocated via binder_alloc_new_buffer()
+ */
+void binder_alloc_free_buf(struct binder_alloc *alloc,
+			    struct binder_buffer *buffer)
+{
+	mutex_lock(&alloc->mutex);
+	binder_free_buf_locked(alloc, buffer);
+	mutex_unlock(&alloc->mutex);
+}
+
+/**
+ * binder_alloc_mmap_handler() - map virtual address space for proc
+ * @alloc:	alloc structure for this proc
+ * @vma:	vma passed to mmap()
+ *
+ * Called by binder_mmap() to initialize the space specified in
+ * vma for allocating binder buffers
+ *
+ * Return:
+ *      0 = success
+ *      -EBUSY = address space already mapped
+ *      -ENOMEM = failed to map memory to given address space
+ */
+int binder_alloc_mmap_handler(struct binder_alloc *alloc,
+			      struct vm_area_struct *vma)
+{
+	int ret;
+	struct vm_struct *area;
+	const char *failure_string;
+	struct binder_buffer *buffer;
+
+	mutex_lock(&binder_alloc_mmap_lock);
+	if (alloc->buffer) {
+		ret = -EBUSY;
+		failure_string = "already mapped";
+		goto err_already_mapped;
+	}
+
+	area = get_vm_area(vma->vm_end - vma->vm_start, VM_IOREMAP);
+	if (area == NULL) {
+		ret = -ENOMEM;
+		failure_string = "get_vm_area";
+		goto err_get_vm_area_failed;
+	}
+	alloc->buffer = area->addr;
+	alloc->user_buffer_offset =
+		vma->vm_start - (uintptr_t)alloc->buffer;
+	mutex_unlock(&binder_alloc_mmap_lock);
+
+#ifdef CONFIG_CPU_CACHE_VIPT
+	if (cache_is_vipt_aliasing()) {
+		while (CACHE_COLOUR(
+				(vma->vm_start ^ (uint32_t)alloc->buffer))) {
+			pr_info("binder_mmap: %d %lx-%lx maps %pK bad alignment\n",
+				alloc->pid, vma->vm_start, vma->vm_end,
+				alloc->buffer);
+			vma->vm_start += PAGE_SIZE;
+		}
+	}
+#endif
+	alloc->pages = kzalloc(sizeof(alloc->pages[0]) *
+				   ((vma->vm_end - vma->vm_start) / PAGE_SIZE),
+			       GFP_KERNEL);
+	if (alloc->pages == NULL) {
+		ret = -ENOMEM;
+		failure_string = "alloc page array";
+		goto err_alloc_pages_failed;
+	}
+	alloc->buffer_size = vma->vm_end - vma->vm_start;
+
+	buffer = kzalloc(sizeof(*buffer), GFP_KERNEL);
+	if (!buffer) {
+		ret = -ENOMEM;
+		failure_string = "alloc buffer struct";
+		goto err_alloc_buf_struct_failed;
+	}
+
+	buffer->data = alloc->buffer;
+	list_add(&buffer->entry, &alloc->buffers);
+	buffer->free = 1;
+	binder_insert_free_buffer(alloc, buffer);
+	alloc->free_async_space = alloc->buffer_size / 2;
+	barrier();
+	alloc->vma = vma;
+	alloc->vma_vm_mm = vma->vm_mm;
+	/* Same as mmgrab() in later kernel versions */
+	atomic_inc(&alloc->vma_vm_mm->mm_count);
+
+	return 0;
+
+err_alloc_buf_struct_failed:
+	kfree(alloc->pages);
+	alloc->pages = NULL;
+err_alloc_pages_failed:
+	mutex_lock(&binder_alloc_mmap_lock);
+	vfree(alloc->buffer);
+	alloc->buffer = NULL;
+err_get_vm_area_failed:
+err_already_mapped:
+	mutex_unlock(&binder_alloc_mmap_lock);
+	pr_err("%s: %d %lx-%lx %s failed %d\n", __func__,
+	       alloc->pid, vma->vm_start, vma->vm_end, failure_string, ret);
+	return ret;
+}
+
+
+void binder_alloc_deferred_release(struct binder_alloc *alloc)
+{
+	struct rb_node *n;
+	int buffers, page_count;
+	struct binder_buffer *buffer;
+
+	BUG_ON(alloc->vma);
+
+	buffers = 0;
+	mutex_lock(&alloc->mutex);
+	while ((n = rb_first(&alloc->allocated_buffers))) {
+		buffer = rb_entry(n, struct binder_buffer, rb_node);
+
+		/* Transaction should already have been freed */
+		BUG_ON(buffer->transaction);
+
+		binder_free_buf_locked(alloc, buffer);
+		buffers++;
+	}
+
+	while (!list_empty(&alloc->buffers)) {
+		buffer = list_first_entry(&alloc->buffers,
+					  struct binder_buffer, entry);
+		WARN_ON(!buffer->free);
+
+		list_del(&buffer->entry);
+		WARN_ON_ONCE(!list_empty(&alloc->buffers));
+		kfree(buffer);
+	}
+
+	page_count = 0;
+	if (alloc->pages) {
+		int i;
+
+		for (i = 0; i < alloc->buffer_size / PAGE_SIZE; i++) {
+			void *page_addr;
+			bool on_lru;
+
+			if (!alloc->pages[i].page_ptr)
+				continue;
+
+			on_lru = list_lru_del(&binder_alloc_lru,
+					      &alloc->pages[i].lru);
+			page_addr = alloc->buffer + i * PAGE_SIZE;
+			binder_alloc_debug(BINDER_DEBUG_BUFFER_ALLOC,
+				     "%s: %d: page %d at %pK %s\n",
+				     __func__, alloc->pid, i, page_addr,
+				     on_lru ? "on lru" : "active");
+			unmap_kernel_range((unsigned long)page_addr, PAGE_SIZE);
+			__free_page(alloc->pages[i].page_ptr);
+			page_count++;
+		}
+		kfree(alloc->pages);
+		vfree(alloc->buffer);
+	}
+	mutex_unlock(&alloc->mutex);
+	if (alloc->vma_vm_mm)
+		mmdrop(alloc->vma_vm_mm);
+
+	binder_alloc_debug(BINDER_DEBUG_OPEN_CLOSE,
+		     "%s: %d buffers %d, pages %d\n",
+		     __func__, alloc->pid, buffers, page_count);
+}
+
+static void print_binder_buffer(struct seq_file *m, const char *prefix,
+				struct binder_buffer *buffer)
+{
+	seq_printf(m, "%s %d: %pK size %zd:%zd:%zd %s\n",
+		   prefix, buffer->debug_id, buffer->data,
+		   buffer->data_size, buffer->offsets_size,
+		   buffer->extra_buffers_size,
+		   buffer->transaction ? "active" : "delivered");
+}
+
+/**
+ * binder_alloc_print_allocated() - print buffer info
+ * @m:     seq_file for output via seq_printf()
+ * @alloc: binder_alloc for this proc
+ *
+ * Prints information about every buffer associated with
+ * the binder_alloc state to the given seq_file
+ */
+void binder_alloc_print_allocated(struct seq_file *m,
+				  struct binder_alloc *alloc)
+{
+	struct rb_node *n;
+
+	mutex_lock(&alloc->mutex);
+	for (n = rb_first(&alloc->allocated_buffers); n != NULL; n = rb_next(n))
+		print_binder_buffer(m, "  buffer",
+				    rb_entry(n, struct binder_buffer, rb_node));
+	mutex_unlock(&alloc->mutex);
+}
+
+/**
+ * binder_alloc_print_pages() - print page usage
+ * @m:     seq_file for output via seq_printf()
+ * @alloc: binder_alloc for this proc
+ */
+void binder_alloc_print_pages(struct seq_file *m,
+			      struct binder_alloc *alloc)
+{
+	struct binder_lru_page *page;
+	int i;
+	int active = 0;
+	int lru = 0;
+	int free = 0;
+
+	mutex_lock(&alloc->mutex);
+	for (i = 0; i < alloc->buffer_size / PAGE_SIZE; i++) {
+		page = &alloc->pages[i];
+		if (!page->page_ptr)
+			free++;
+		else if (list_empty(&page->lru))
+			active++;
+		else
+			lru++;
+	}
+	mutex_unlock(&alloc->mutex);
+	seq_printf(m, "  pages: %d:%d:%d\n", active, lru, free);
+	seq_printf(m, "  pages high watermark: %zu\n", alloc->pages_high);
+}
+
+/**
+ * binder_alloc_get_allocated_count() - return count of buffers
+ * @alloc: binder_alloc for this proc
+ *
+ * Return: count of allocated buffers
+ */
+int binder_alloc_get_allocated_count(struct binder_alloc *alloc)
+{
+	struct rb_node *n;
+	int count = 0;
+
+	mutex_lock(&alloc->mutex);
+	for (n = rb_first(&alloc->allocated_buffers); n != NULL; n = rb_next(n))
+		count++;
+	mutex_unlock(&alloc->mutex);
+	return count;
+}
+
+
+/**
+ * binder_alloc_vma_close() - invalidate address space
+ * @alloc: binder_alloc for this proc
+ *
+ * Called from binder_vma_close() when releasing address space.
+ * Clears alloc->vma to prevent new incoming transactions from
+ * allocating more buffers.
+ */
+void binder_alloc_vma_close(struct binder_alloc *alloc)
+{
+	WRITE_ONCE(alloc->vma, NULL);
+}
+
+/**
+ * binder_alloc_free_page() - shrinker callback to free pages
+ * @item:   item to free
+ * @lock:   lock protecting the item
+ * @cb_arg: callback argument
+ *
+ * Called from list_lru_walk() in binder_shrink_scan() to free
+ * up pages when the system is under memory pressure.
+ */
+enum lru_status binder_alloc_free_page(struct list_head *item,
+				       struct list_lru_one *lru,
+				       spinlock_t *lock,
+				       void *cb_arg)
+{
+	struct mm_struct *mm = NULL;
+	struct binder_lru_page *page = container_of(item,
+						    struct binder_lru_page,
+						    lru);
+	struct binder_alloc *alloc;
+	uintptr_t page_addr;
+	size_t index;
+	struct vm_area_struct *vma;
+
+	alloc = page->alloc;
+	if (!mutex_trylock(&alloc->mutex))
+		goto err_get_alloc_mutex_failed;
+
+	if (!page->page_ptr)
+		goto err_page_already_freed;
+
+	index = page - alloc->pages;
+	page_addr = (uintptr_t)alloc->buffer + index * PAGE_SIZE;
+	vma = alloc->vma;
+	if (vma) {
+		/* Same as mmget_not_zero() in later kernel versions */
+		if (!atomic_inc_not_zero(&alloc->vma_vm_mm->mm_users))
+			goto err_mmget;
+		mm = alloc->vma_vm_mm;
+		if (!down_write_trylock(&mm->mmap_sem))
+			goto err_down_write_mmap_sem_failed;
+	}
+
+	list_lru_isolate(lru, item);
+	spin_unlock(lock);
+
+	if (vma) {
+		trace_binder_unmap_user_start(alloc, index);
+
+		zap_page_range(vma,
+			       page_addr +
+			       alloc->user_buffer_offset,
+			       PAGE_SIZE, NULL);
+
+		trace_binder_unmap_user_end(alloc, index);
+
+		up_write(&mm->mmap_sem);
+		mmput(mm);
+	}
+
+	trace_binder_unmap_kernel_start(alloc, index);
+
+	unmap_kernel_range(page_addr, PAGE_SIZE);
+	__free_page(page->page_ptr);
+	page->page_ptr = NULL;
+
+	trace_binder_unmap_kernel_end(alloc, index);
+
+	spin_lock(lock);
+	mutex_unlock(&alloc->mutex);
+	return LRU_REMOVED_RETRY;
+
+err_down_write_mmap_sem_failed:
+	mmput_async(mm);
+err_mmget:
+err_page_already_freed:
+	mutex_unlock(&alloc->mutex);
+err_get_alloc_mutex_failed:
+	return LRU_SKIP;
+}
+
+static unsigned long
+binder_shrink_count(struct shrinker *shrink, struct shrink_control *sc)
+{
+	unsigned long ret = list_lru_count(&binder_alloc_lru);
+	return ret;
+}
+
+static unsigned long
+binder_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
+{
+	unsigned long ret;
+
+	ret = list_lru_walk(&binder_alloc_lru, binder_alloc_free_page,
+			    NULL, sc->nr_to_scan);
+	return ret;
+}
+
+static struct shrinker binder_shrinker = {
+	.count_objects = binder_shrink_count,
+	.scan_objects = binder_shrink_scan,
+	.seeks = DEFAULT_SEEKS,
+};
+
+/**
+ * binder_alloc_init() - called by binder_open() for per-proc initialization
+ * @alloc: binder_alloc for this proc
+ *
+ * Called from binder_open() to initialize binder_alloc fields for
+ * new binder proc
+ */
+void binder_alloc_init(struct binder_alloc *alloc)
+{
+	alloc->pid = current->group_leader->pid;
+	mutex_init(&alloc->mutex);
+	INIT_LIST_HEAD(&alloc->buffers);
+}
+
+void binder_alloc_shrinker_init(void)
+{
+	list_lru_init(&binder_alloc_lru);
+	register_shrinker(&binder_shrinker);
+}
diff --git a/drivers/android/binder_alloc.h b/drivers/android/binder_alloc.h
new file mode 100644
index 000000000000..0b145307f1fd
--- /dev/null
+++ b/drivers/android/binder_alloc.h
@@ -0,0 +1,188 @@
+/*
+ * Copyright (C) 2017 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _LINUX_BINDER_ALLOC_H
+#define _LINUX_BINDER_ALLOC_H
+
+#include <linux/rbtree.h>
+#include <linux/list.h>
+#include <linux/mm.h>
+#include <linux/rtmutex.h>
+#include <linux/vmalloc.h>
+#include <linux/slab.h>
+#include <linux/list_lru.h>
+
+extern struct list_lru binder_alloc_lru;
+struct binder_transaction;
+
+/**
+ * struct binder_buffer - buffer used for binder transactions
+ * @entry:              entry alloc->buffers
+ * @rb_node:            node for allocated_buffers/free_buffers rb trees
+ * @free:               true if buffer is free
+ * @allow_user_free:    describe the second member of struct blah,
+ * @async_transaction:  describe the second member of struct blah,
+ * @debug_id:           describe the second member of struct blah,
+ * @transaction:        describe the second member of struct blah,
+ * @target_node:        describe the second member of struct blah,
+ * @data_size:          describe the second member of struct blah,
+ * @offsets_size:       describe the second member of struct blah,
+ * @extra_buffers_size: describe the second member of struct blah,
+ * @data:i              describe the second member of struct blah,
+ *
+ * Bookkeeping structure for binder transaction buffers
+ */
+struct binder_buffer {
+	struct list_head entry; /* free and allocated entries by address */
+	struct rb_node rb_node; /* free entry by size or allocated entry */
+				/* by address */
+	unsigned free:1;
+	unsigned allow_user_free:1;
+	unsigned async_transaction:1;
+	unsigned free_in_progress:1;
+	unsigned debug_id:28;
+
+	struct binder_transaction *transaction;
+
+	struct binder_node *target_node;
+	size_t data_size;
+	size_t offsets_size;
+	size_t extra_buffers_size;
+	void *data;
+};
+
+/**
+ * struct binder_lru_page - page object used for binder shrinker
+ * @page_ptr: pointer to physical page in mmap'd space
+ * @lru:      entry in binder_alloc_lru
+ * @alloc:    binder_alloc for a proc
+ */
+struct binder_lru_page {
+	struct list_head lru;
+	struct page *page_ptr;
+	struct binder_alloc *alloc;
+};
+
+/**
+ * struct binder_alloc - per-binder proc state for binder allocator
+ * @vma:                vm_area_struct passed to mmap_handler
+ *                      (invarient after mmap)
+ * @tsk:                tid for task that called init for this proc
+ *                      (invariant after init)
+ * @vma_vm_mm:          copy of vma->vm_mm (invarient after mmap)
+ * @buffer:             base of per-proc address space mapped via mmap
+ * @user_buffer_offset: offset between user and kernel VAs for buffer
+ * @buffers:            list of all buffers for this proc
+ * @free_buffers:       rb tree of buffers available for allocation
+ *                      sorted by size
+ * @allocated_buffers:  rb tree of allocated buffers sorted by address
+ * @free_async_space:   VA space available for async buffers. This is
+ *                      initialized at mmap time to 1/2 the full VA space
+ * @pages:              array of binder_lru_page
+ * @buffer_size:        size of address space specified via mmap
+ * @pid:                pid for associated binder_proc (invariant after init)
+ * @pages_high:         high watermark of offset in @pages
+ *
+ * Bookkeeping structure for per-proc address space management for binder
+ * buffers. It is normally initialized during binder_init() and binder_mmap()
+ * calls. The address space is used for both user-visible buffers and for
+ * struct binder_buffer objects used to track the user buffers
+ */
+struct binder_alloc {
+	struct mutex mutex;
+	struct vm_area_struct *vma;
+	struct mm_struct *vma_vm_mm;
+	void *buffer;
+	ptrdiff_t user_buffer_offset;
+	struct list_head buffers;
+	struct rb_root free_buffers;
+	struct rb_root allocated_buffers;
+	size_t free_async_space;
+	struct binder_lru_page *pages;
+	size_t buffer_size;
+	uint32_t buffer_free;
+	int pid;
+	size_t pages_high;
+};
+
+#ifdef CONFIG_ANDROID_BINDER_IPC_SELFTEST
+void binder_selftest_alloc(struct binder_alloc *alloc);
+#else
+static inline void binder_selftest_alloc(struct binder_alloc *alloc) {}
+#endif
+enum lru_status binder_alloc_free_page(struct list_head *item,
+				       struct list_lru_one *lru,
+				       spinlock_t *lock, void *cb_arg);
+extern struct binder_buffer *binder_alloc_new_buf(struct binder_alloc *alloc,
+						  size_t data_size,
+						  size_t offsets_size,
+						  size_t extra_buffers_size,
+						  int is_async);
+extern void binder_alloc_init(struct binder_alloc *alloc);
+void binder_alloc_shrinker_init(void);
+extern void binder_alloc_vma_close(struct binder_alloc *alloc);
+extern struct binder_buffer *
+binder_alloc_prepare_to_free(struct binder_alloc *alloc,
+			     uintptr_t user_ptr);
+extern void binder_alloc_free_buf(struct binder_alloc *alloc,
+				  struct binder_buffer *buffer);
+extern int binder_alloc_mmap_handler(struct binder_alloc *alloc,
+				     struct vm_area_struct *vma);
+extern void binder_alloc_deferred_release(struct binder_alloc *alloc);
+extern int binder_alloc_get_allocated_count(struct binder_alloc *alloc);
+extern void binder_alloc_print_allocated(struct seq_file *m,
+					 struct binder_alloc *alloc);
+void binder_alloc_print_pages(struct seq_file *m,
+			      struct binder_alloc *alloc);
+
+/**
+ * binder_alloc_get_free_async_space() - get free space available for async
+ * @alloc:	binder_alloc for this proc
+ *
+ * Return:	the bytes remaining in the address-space for async transactions
+ */
+static inline size_t
+binder_alloc_get_free_async_space(struct binder_alloc *alloc)
+{
+	size_t free_async_space;
+
+	mutex_lock(&alloc->mutex);
+	free_async_space = alloc->free_async_space;
+	mutex_unlock(&alloc->mutex);
+	return free_async_space;
+}
+
+/**
+ * binder_alloc_get_user_buffer_offset() - get offset between kernel/user addrs
+ * @alloc:	binder_alloc for this proc
+ *
+ * Return:	the offset between kernel and user-space addresses to use for
+ * virtual address conversion
+ */
+static inline ptrdiff_t
+binder_alloc_get_user_buffer_offset(struct binder_alloc *alloc)
+{
+	/*
+	 * user_buffer_offset is constant if vma is set and
+	 * undefined if vma is not set. It is possible to
+	 * get here with !alloc->vma if the target process
+	 * is dying while a transaction is being initiated.
+	 * Returning the old value is ok in this case and
+	 * the transaction will fail.
+	 */
+	return alloc->user_buffer_offset;
+}
+
+#endif /* _LINUX_BINDER_ALLOC_H */
+
diff --git a/drivers/android/binder_alloc_selftest.c b/drivers/android/binder_alloc_selftest.c
new file mode 100644
index 000000000000..8bd7bcef967d
--- /dev/null
+++ b/drivers/android/binder_alloc_selftest.c
@@ -0,0 +1,310 @@
+/* binder_alloc_selftest.c
+ *
+ * Android IPC Subsystem
+ *
+ * Copyright (C) 2017 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/mm_types.h>
+#include <linux/err.h>
+#include "binder_alloc.h"
+
+#define BUFFER_NUM 5
+#define BUFFER_MIN_SIZE (PAGE_SIZE / 8)
+
+static bool binder_selftest_run = true;
+static int binder_selftest_failures;
+static DEFINE_MUTEX(binder_selftest_lock);
+
+/**
+ * enum buf_end_align_type - Page alignment of a buffer
+ * end with regard to the end of the previous buffer.
+ *
+ * In the pictures below, buf2 refers to the buffer we
+ * are aligning. buf1 refers to previous buffer by addr.
+ * Symbol [ means the start of a buffer, ] means the end
+ * of a buffer, and | means page boundaries.
+ */
+enum buf_end_align_type {
+	/**
+	 * @SAME_PAGE_UNALIGNED: The end of this buffer is on
+	 * the same page as the end of the previous buffer and
+	 * is not page aligned. Examples:
+	 * buf1 ][ buf2 ][ ...
+	 * buf1 ]|[ buf2 ][ ...
+	 */
+	SAME_PAGE_UNALIGNED = 0,
+	/**
+	 * @SAME_PAGE_ALIGNED: When the end of the previous buffer
+	 * is not page aligned, the end of this buffer is on the
+	 * same page as the end of the previous buffer and is page
+	 * aligned. When the previous buffer is page aligned, the
+	 * end of this buffer is aligned to the next page boundary.
+	 * Examples:
+	 * buf1 ][ buf2 ]| ...
+	 * buf1 ]|[ buf2 ]| ...
+	 */
+	SAME_PAGE_ALIGNED,
+	/**
+	 * @NEXT_PAGE_UNALIGNED: The end of this buffer is on
+	 * the page next to the end of the previous buffer and
+	 * is not page aligned. Examples:
+	 * buf1 ][ buf2 | buf2 ][ ...
+	 * buf1 ]|[ buf2 | buf2 ][ ...
+	 */
+	NEXT_PAGE_UNALIGNED,
+	/**
+	 * @NEXT_PAGE_ALIGNED: The end of this buffer is on
+	 * the page next to the end of the previous buffer and
+	 * is page aligned. Examples:
+	 * buf1 ][ buf2 | buf2 ]| ...
+	 * buf1 ]|[ buf2 | buf2 ]| ...
+	 */
+	NEXT_PAGE_ALIGNED,
+	/**
+	 * @NEXT_NEXT_UNALIGNED: The end of this buffer is on
+	 * the page that follows the page after the end of the
+	 * previous buffer and is not page aligned. Examples:
+	 * buf1 ][ buf2 | buf2 | buf2 ][ ...
+	 * buf1 ]|[ buf2 | buf2 | buf2 ][ ...
+	 */
+	NEXT_NEXT_UNALIGNED,
+	LOOP_END,
+};
+
+static void pr_err_size_seq(size_t *sizes, int *seq)
+{
+	int i;
+
+	pr_err("alloc sizes: ");
+	for (i = 0; i < BUFFER_NUM; i++)
+		pr_cont("[%zu]", sizes[i]);
+	pr_cont("\n");
+	pr_err("free seq: ");
+	for (i = 0; i < BUFFER_NUM; i++)
+		pr_cont("[%d]", seq[i]);
+	pr_cont("\n");
+}
+
+static bool check_buffer_pages_allocated(struct binder_alloc *alloc,
+					 struct binder_buffer *buffer,
+					 size_t size)
+{
+	void *page_addr, *end;
+	int page_index;
+
+	end = (void *)PAGE_ALIGN((uintptr_t)buffer->data + size);
+	page_addr = buffer->data;
+	for (; page_addr < end; page_addr += PAGE_SIZE) {
+		page_index = (page_addr - alloc->buffer) / PAGE_SIZE;
+		if (!alloc->pages[page_index].page_ptr ||
+		    !list_empty(&alloc->pages[page_index].lru)) {
+			pr_err("expect alloc but is %s at page index %d\n",
+			       alloc->pages[page_index].page_ptr ?
+			       "lru" : "free", page_index);
+			return false;
+		}
+	}
+	return true;
+}
+
+static void binder_selftest_alloc_buf(struct binder_alloc *alloc,
+				      struct binder_buffer *buffers[],
+				      size_t *sizes, int *seq)
+{
+	int i;
+
+	for (i = 0; i < BUFFER_NUM; i++) {
+		buffers[i] = binder_alloc_new_buf(alloc, sizes[i], 0, 0, 0);
+		if (IS_ERR(buffers[i]) ||
+		    !check_buffer_pages_allocated(alloc, buffers[i],
+						  sizes[i])) {
+			pr_err_size_seq(sizes, seq);
+			binder_selftest_failures++;
+		}
+	}
+}
+
+static void binder_selftest_free_buf(struct binder_alloc *alloc,
+				     struct binder_buffer *buffers[],
+				     size_t *sizes, int *seq, size_t end)
+{
+	int i;
+
+	for (i = 0; i < BUFFER_NUM; i++)
+		binder_alloc_free_buf(alloc, buffers[seq[i]]);
+
+	for (i = 0; i < end / PAGE_SIZE; i++) {
+		/**
+		 * Error message on a free page can be false positive
+		 * if binder shrinker ran during binder_alloc_free_buf
+		 * calls above.
+		 */
+		if (list_empty(&alloc->pages[i].lru)) {
+			pr_err_size_seq(sizes, seq);
+			pr_err("expect lru but is %s at page index %d\n",
+			       alloc->pages[i].page_ptr ? "alloc" : "free", i);
+			binder_selftest_failures++;
+		}
+	}
+}
+
+static void binder_selftest_free_page(struct binder_alloc *alloc)
+{
+	int i;
+	unsigned long count;
+
+	while ((count = list_lru_count(&binder_alloc_lru))) {
+		list_lru_walk(&binder_alloc_lru, binder_alloc_free_page,
+			      NULL, count);
+	}
+
+	for (i = 0; i < (alloc->buffer_size / PAGE_SIZE); i++) {
+		if (alloc->pages[i].page_ptr) {
+			pr_err("expect free but is %s at page index %d\n",
+			       list_empty(&alloc->pages[i].lru) ?
+			       "alloc" : "lru", i);
+			binder_selftest_failures++;
+		}
+	}
+}
+
+static void binder_selftest_alloc_free(struct binder_alloc *alloc,
+				       size_t *sizes, int *seq, size_t end)
+{
+	struct binder_buffer *buffers[BUFFER_NUM];
+
+	binder_selftest_alloc_buf(alloc, buffers, sizes, seq);
+	binder_selftest_free_buf(alloc, buffers, sizes, seq, end);
+
+	/* Allocate from lru. */
+	binder_selftest_alloc_buf(alloc, buffers, sizes, seq);
+	if (list_lru_count(&binder_alloc_lru))
+		pr_err("lru list should be empty but is not\n");
+
+	binder_selftest_free_buf(alloc, buffers, sizes, seq, end);
+	binder_selftest_free_page(alloc);
+}
+
+static bool is_dup(int *seq, int index, int val)
+{
+	int i;
+
+	for (i = 0; i < index; i++) {
+		if (seq[i] == val)
+			return true;
+	}
+	return false;
+}
+
+/* Generate BUFFER_NUM factorial free orders. */
+static void binder_selftest_free_seq(struct binder_alloc *alloc,
+				     size_t *sizes, int *seq,
+				     int index, size_t end)
+{
+	int i;
+
+	if (index == BUFFER_NUM) {
+		binder_selftest_alloc_free(alloc, sizes, seq, end);
+		return;
+	}
+	for (i = 0; i < BUFFER_NUM; i++) {
+		if (is_dup(seq, index, i))
+			continue;
+		seq[index] = i;
+		binder_selftest_free_seq(alloc, sizes, seq, index + 1, end);
+	}
+}
+
+static void binder_selftest_alloc_size(struct binder_alloc *alloc,
+				       size_t *end_offset)
+{
+	int i;
+	int seq[BUFFER_NUM] = {0};
+	size_t front_sizes[BUFFER_NUM];
+	size_t back_sizes[BUFFER_NUM];
+	size_t last_offset, offset = 0;
+
+	for (i = 0; i < BUFFER_NUM; i++) {
+		last_offset = offset;
+		offset = end_offset[i];
+		front_sizes[i] = offset - last_offset;
+		back_sizes[BUFFER_NUM - i - 1] = front_sizes[i];
+	}
+	/*
+	 * Buffers share the first or last few pages.
+	 * Only BUFFER_NUM - 1 buffer sizes are adjustable since
+	 * we need one giant buffer before getting to the last page.
+	 */
+	back_sizes[0] += alloc->buffer_size - end_offset[BUFFER_NUM - 1];
+	binder_selftest_free_seq(alloc, front_sizes, seq, 0,
+				 end_offset[BUFFER_NUM - 1]);
+	binder_selftest_free_seq(alloc, back_sizes, seq, 0, alloc->buffer_size);
+}
+
+static void binder_selftest_alloc_offset(struct binder_alloc *alloc,
+					 size_t *end_offset, int index)
+{
+	int align;
+	size_t end, prev;
+
+	if (index == BUFFER_NUM) {
+		binder_selftest_alloc_size(alloc, end_offset);
+		return;
+	}
+	prev = index == 0 ? 0 : end_offset[index - 1];
+	end = prev;
+
+	BUILD_BUG_ON(BUFFER_MIN_SIZE * BUFFER_NUM >= PAGE_SIZE);
+
+	for (align = SAME_PAGE_UNALIGNED; align < LOOP_END; align++) {
+		if (align % 2)
+			end = ALIGN(end, PAGE_SIZE);
+		else
+			end += BUFFER_MIN_SIZE;
+		end_offset[index] = end;
+		binder_selftest_alloc_offset(alloc, end_offset, index + 1);
+	}
+}
+
+/**
+ * binder_selftest_alloc() - Test alloc and free of buffer pages.
+ * @alloc: Pointer to alloc struct.
+ *
+ * Allocate BUFFER_NUM buffers to cover all page alignment cases,
+ * then free them in all orders possible. Check that pages are
+ * correctly allocated, put onto lru when buffers are freed, and
+ * are freed when binder_alloc_free_page is called.
+ */
+void binder_selftest_alloc(struct binder_alloc *alloc)
+{
+	size_t end_offset[BUFFER_NUM];
+
+	if (!binder_selftest_run)
+		return;
+	mutex_lock(&binder_selftest_lock);
+	if (!binder_selftest_run || !alloc->vma)
+		goto done;
+	pr_info("STARTED\n");
+	binder_selftest_alloc_offset(alloc, end_offset, 0);
+	binder_selftest_run = false;
+	if (binder_selftest_failures > 0)
+		pr_info("%d tests FAILED\n", binder_selftest_failures);
+	else
+		pr_info("PASSED\n");
+
+done:
+	mutex_unlock(&binder_selftest_lock);
+}
diff --git a/drivers/android/binder_trace.h b/drivers/android/binder_trace.h
index 7f20f3dc8369..b11dffc521e8 100644
--- a/drivers/android/binder_trace.h
+++ b/drivers/android/binder_trace.h
@@ -23,7 +23,8 @@
 struct binder_buffer;
 struct binder_node;
 struct binder_proc;
-struct binder_ref;
+struct binder_alloc;
+struct binder_ref_data;
 struct binder_thread;
 struct binder_transaction;
 
@@ -84,6 +85,30 @@ DEFINE_BINDER_FUNCTION_RETURN_EVENT(binder_ioctl_done);
 DEFINE_BINDER_FUNCTION_RETURN_EVENT(binder_write_done);
 DEFINE_BINDER_FUNCTION_RETURN_EVENT(binder_read_done);
 
+TRACE_EVENT(binder_set_priority,
+	TP_PROTO(int proc, int thread, unsigned int old_prio,
+		 unsigned int desired_prio, unsigned int new_prio),
+	TP_ARGS(proc, thread, old_prio, new_prio, desired_prio),
+
+	TP_STRUCT__entry(
+		__field(int, proc)
+		__field(int, thread)
+		__field(unsigned int, old_prio)
+		__field(unsigned int, new_prio)
+		__field(unsigned int, desired_prio)
+	),
+	TP_fast_assign(
+		__entry->proc = proc;
+		__entry->thread = thread;
+		__entry->old_prio = old_prio;
+		__entry->new_prio = new_prio;
+		__entry->desired_prio = desired_prio;
+	),
+	TP_printk("proc=%d thread=%d old=%d => new=%d desired=%d",
+		  __entry->proc, __entry->thread, __entry->old_prio,
+		  __entry->new_prio, __entry->desired_prio)
+);
+
 TRACE_EVENT(binder_wait_for_work,
 	TP_PROTO(bool proc_work, bool transaction_stack, bool thread_todo),
 	TP_ARGS(proc_work, transaction_stack, thread_todo),
@@ -146,8 +171,8 @@ TRACE_EVENT(binder_transaction_received,
 
 TRACE_EVENT(binder_transaction_node_to_ref,
 	TP_PROTO(struct binder_transaction *t, struct binder_node *node,
-		 struct binder_ref *ref),
-	TP_ARGS(t, node, ref),
+		 struct binder_ref_data *rdata),
+	TP_ARGS(t, node, rdata),
 
 	TP_STRUCT__entry(
 		__field(int, debug_id)
@@ -160,8 +185,8 @@ TRACE_EVENT(binder_transaction_node_to_ref,
 		__entry->debug_id = t->debug_id;
 		__entry->node_debug_id = node->debug_id;
 		__entry->node_ptr = node->ptr;
-		__entry->ref_debug_id = ref->debug_id;
-		__entry->ref_desc = ref->desc;
+		__entry->ref_debug_id = rdata->debug_id;
+		__entry->ref_desc = rdata->desc;
 	),
 	TP_printk("transaction=%d node=%d src_ptr=0x%016llx ==> dest_ref=%d dest_desc=%d",
 		  __entry->debug_id, __entry->node_debug_id,
@@ -170,8 +195,9 @@ TRACE_EVENT(binder_transaction_node_to_ref,
 );
 
 TRACE_EVENT(binder_transaction_ref_to_node,
-	TP_PROTO(struct binder_transaction *t, struct binder_ref *ref),
-	TP_ARGS(t, ref),
+	TP_PROTO(struct binder_transaction *t, struct binder_node *node,
+		 struct binder_ref_data *rdata),
+	TP_ARGS(t, node, rdata),
 
 	TP_STRUCT__entry(
 		__field(int, debug_id)
@@ -182,10 +208,10 @@ TRACE_EVENT(binder_transaction_ref_to_node,
 	),
 	TP_fast_assign(
 		__entry->debug_id = t->debug_id;
-		__entry->ref_debug_id = ref->debug_id;
-		__entry->ref_desc = ref->desc;
-		__entry->node_debug_id = ref->node->debug_id;
-		__entry->node_ptr = ref->node->ptr;
+		__entry->ref_debug_id = rdata->debug_id;
+		__entry->ref_desc = rdata->desc;
+		__entry->node_debug_id = node->debug_id;
+		__entry->node_ptr = node->ptr;
 	),
 	TP_printk("transaction=%d node=%d src_ref=%d src_desc=%d ==> dest_ptr=0x%016llx",
 		  __entry->debug_id, __entry->node_debug_id,
@@ -194,9 +220,10 @@ TRACE_EVENT(binder_transaction_ref_to_node,
 );
 
 TRACE_EVENT(binder_transaction_ref_to_ref,
-	TP_PROTO(struct binder_transaction *t, struct binder_ref *src_ref,
-		 struct binder_ref *dest_ref),
-	TP_ARGS(t, src_ref, dest_ref),
+	TP_PROTO(struct binder_transaction *t, struct binder_node *node,
+		 struct binder_ref_data *src_ref,
+		 struct binder_ref_data *dest_ref),
+	TP_ARGS(t, node, src_ref, dest_ref),
 
 	TP_STRUCT__entry(
 		__field(int, debug_id)
@@ -208,7 +235,7 @@ TRACE_EVENT(binder_transaction_ref_to_ref,
 	),
 	TP_fast_assign(
 		__entry->debug_id = t->debug_id;
-		__entry->node_debug_id = src_ref->node->debug_id;
+		__entry->node_debug_id = node->debug_id;
 		__entry->src_ref_debug_id = src_ref->debug_id;
 		__entry->src_ref_desc = src_ref->desc;
 		__entry->dest_ref_debug_id = dest_ref->debug_id;
@@ -268,9 +295,9 @@ DEFINE_EVENT(binder_buffer_class, binder_transaction_failed_buffer_release,
 	TP_ARGS(buffer));
 
 TRACE_EVENT(binder_update_page_range,
-	TP_PROTO(struct binder_proc *proc, bool allocate,
+	TP_PROTO(struct binder_alloc *alloc, bool allocate,
 		 void *start, void *end),
-	TP_ARGS(proc, allocate, start, end),
+	TP_ARGS(alloc, allocate, start, end),
 	TP_STRUCT__entry(
 		__field(int, proc)
 		__field(bool, allocate)
@@ -278,9 +305,9 @@ TRACE_EVENT(binder_update_page_range,
 		__field(size_t, size)
 	),
 	TP_fast_assign(
-		__entry->proc = proc->pid;
+		__entry->proc = alloc->pid;
 		__entry->allocate = allocate;
-		__entry->offset = start - proc->buffer;
+		__entry->offset = start - alloc->buffer;
 		__entry->size = end - start;
 	),
 	TP_printk("proc=%d allocate=%d offset=%zu size=%zu",
@@ -288,6 +315,61 @@ TRACE_EVENT(binder_update_page_range,
 		  __entry->offset, __entry->size)
 );
 
+DECLARE_EVENT_CLASS(binder_lru_page_class,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index),
+	TP_STRUCT__entry(
+		__field(int, proc)
+		__field(size_t, page_index)
+	),
+	TP_fast_assign(
+		__entry->proc = alloc->pid;
+		__entry->page_index = page_index;
+	),
+	TP_printk("proc=%d page_index=%zu",
+		  __entry->proc, __entry->page_index)
+);
+
+DEFINE_EVENT(binder_lru_page_class, binder_alloc_lru_start,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_alloc_lru_end,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_free_lru_start,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_free_lru_end,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_alloc_page_start,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_alloc_page_end,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_unmap_user_start,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_unmap_user_end,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_unmap_kernel_start,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
+DEFINE_EVENT(binder_lru_page_class, binder_unmap_kernel_end,
+	TP_PROTO(const struct binder_alloc *alloc, size_t page_index),
+	TP_ARGS(alloc, page_index));
+
 TRACE_EVENT(binder_command,
 	TP_PROTO(uint32_t cmd),
 	TP_ARGS(cmd),
diff --git a/drivers/base/dd.c b/drivers/base/dd.c
index a641cf3ccad6..9e425fbf83cb 100644
--- a/drivers/base/dd.c
+++ b/drivers/base/dd.c
@@ -205,6 +205,8 @@ static void driver_bound(struct device *dev)
 
 	klist_add_tail(&dev->p->knode_driver, &dev->driver->p->klist_devices);
 
+	device_pm_check_callbacks(dev);
+
 	/*
 	 * Make sure the device is no longer in one of the deferred lists and
 	 * kick off retrying all pending devices
@@ -697,6 +699,7 @@ static void __device_release_driver(struct device *dev)
 			dev->pm_domain->dismiss(dev);
 
 		klist_remove(&dev->p->knode_driver);
+		device_pm_check_callbacks(dev);
 		if (dev->bus)
 			blocking_notifier_call_chain(&dev->bus->p->bus_notifier,
 						     BUS_NOTIFY_UNBOUND_DRIVER,
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index e9b713675c7c..6c5bc3fadfcf 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -33,6 +33,7 @@
 #include <linux/cpufreq.h>
 #include <linux/cpuidle.h>
 #include <linux/timer.h>
+#include <linux/wakeup_reason.h>
 
 #include "../base.h"
 #include "power.h"
@@ -125,6 +126,7 @@ void device_pm_add(struct device *dev)
 {
 	pr_debug("PM: Adding info for %s:%s\n",
 		 dev->bus ? dev->bus->name : "No Bus", dev_name(dev));
+	device_pm_check_callbacks(dev);
 	mutex_lock(&dpm_list_mtx);
 	if (dev->parent && dev->parent->power.is_prepared)
 		dev_warn(dev, "parent %s should not be sleeping\n",
@@ -147,6 +149,7 @@ void device_pm_remove(struct device *dev)
 	mutex_unlock(&dpm_list_mtx);
 	device_wakeup_disable(dev);
 	pm_runtime_remove(dev);
+	device_pm_check_callbacks(dev);
 }
 
 /**
@@ -1348,6 +1351,7 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async)
 	pm_callback_t callback = NULL;
 	char *info = NULL;
 	int error = 0;
+	char suspend_abort[MAX_SUSPEND_ABORT_LEN];
 	DECLARE_DPM_WATCHDOG_ON_STACK(wd);
 
 	TRACE_DEVICE(dev);
@@ -1368,6 +1372,9 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async)
 		pm_wakeup_event(dev, 0);
 
 	if (pm_wakeup_pending()) {
+		pm_get_active_wakeup_sources(suspend_abort,
+			MAX_SUSPEND_ABORT_LEN);
+		log_suspend_abort_reason(suspend_abort);
 		async_error = -EBUSY;
 		goto Complete;
 	}
@@ -1570,6 +1577,11 @@ static int device_prepare(struct device *dev, pm_message_t state)
 
 	dev->power.wakeup_path = device_may_wakeup(dev);
 
+	if (dev->power.no_pm_callbacks) {
+		ret = 1;	/* Let device go direct_complete */
+		goto unlock;
+	}
+
 	if (dev->pm_domain) {
 		info = "preparing power domain ";
 		callback = dev->pm_domain->ops.prepare;
@@ -1592,6 +1604,7 @@ static int device_prepare(struct device *dev, pm_message_t state)
 	if (callback)
 		ret = callback(dev);
 
+unlock:
 	device_unlock(dev);
 
 	if (ret < 0) {
@@ -1720,3 +1733,30 @@ void dpm_for_each_dev(void *data, void (*fn)(struct device *, void *))
 	device_pm_unlock();
 }
 EXPORT_SYMBOL_GPL(dpm_for_each_dev);
+
+static bool pm_ops_is_empty(const struct dev_pm_ops *ops)
+{
+	if (!ops)
+		return true;
+
+	return !ops->prepare &&
+	       !ops->suspend &&
+	       !ops->suspend_late &&
+	       !ops->suspend_noirq &&
+	       !ops->resume_noirq &&
+	       !ops->resume_early &&
+	       !ops->resume &&
+	       !ops->complete;
+}
+
+void device_pm_check_callbacks(struct device *dev)
+{
+	spin_lock_irq(&dev->power.lock);
+	dev->power.no_pm_callbacks =
+		(!dev->bus || pm_ops_is_empty(dev->bus->pm)) &&
+		(!dev->class || pm_ops_is_empty(dev->class->pm)) &&
+		(!dev->type || pm_ops_is_empty(dev->type->pm)) &&
+		(!dev->pm_domain || pm_ops_is_empty(&dev->pm_domain->ops)) &&
+		(!dev->driver || pm_ops_is_empty(dev->driver->pm));
+	spin_unlock_irq(&dev->power.lock);
+}
diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h
index 01ac5b1aeaeb..297aa5cf393e 100644
--- a/drivers/base/power/power.h
+++ b/drivers/base/power/power.h
@@ -140,6 +140,7 @@ extern void device_pm_remove(struct device *);
 extern void device_pm_move_before(struct device *, struct device *);
 extern void device_pm_move_after(struct device *, struct device *);
 extern void device_pm_move_last(struct device *);
+extern void device_pm_check_callbacks(struct device *dev);
 
 #else /* !CONFIG_PM_SLEEP */
 
@@ -158,6 +159,8 @@ static inline void device_pm_move_after(struct device *deva,
 					struct device *devb) {}
 static inline void device_pm_move_last(struct device *dev) {}
 
+static inline void device_pm_check_callbacks(struct device *dev) {}
+
 #endif /* !CONFIG_PM_SLEEP */
 
 static inline void device_pm_init(struct device *dev)
diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c
index e613633ffe9c..7af116e12e53 100644
--- a/drivers/base/power/wakeup.c
+++ b/drivers/base/power/wakeup.c
@@ -15,6 +15,7 @@
 #include <linux/seq_file.h>
 #include <linux/debugfs.h>
 #include <linux/pm_wakeirq.h>
+#include <linux/types.h>
 #include <trace/events/power.h>
 
 #include "power.h"
@@ -808,6 +809,37 @@ void pm_wakeup_event(struct device *dev, unsigned int msec)
 }
 EXPORT_SYMBOL_GPL(pm_wakeup_event);
 
+void pm_get_active_wakeup_sources(char *pending_wakeup_source, size_t max)
+{
+	struct wakeup_source *ws, *last_active_ws = NULL;
+	int len = 0;
+	bool active = false;
+
+	rcu_read_lock();
+	list_for_each_entry_rcu(ws, &wakeup_sources, entry) {
+		if (ws->active && len < max) {
+			if (!active)
+				len += scnprintf(pending_wakeup_source, max,
+						"Pending Wakeup Sources: ");
+			len += scnprintf(pending_wakeup_source + len, max - len,
+				"%s ", ws->name);
+			active = true;
+		} else if (!active &&
+			   (!last_active_ws ||
+			    ktime_to_ns(ws->last_time) >
+			    ktime_to_ns(last_active_ws->last_time))) {
+			last_active_ws = ws;
+		}
+	}
+	if (!active && last_active_ws) {
+		scnprintf(pending_wakeup_source, max,
+				"Last active Wakeup Source: %s",
+				last_active_ws->name);
+	}
+	rcu_read_unlock();
+}
+EXPORT_SYMBOL_GPL(pm_get_active_wakeup_sources);
+
 void pm_print_active_wakeup_sources(void)
 {
 	struct wakeup_source *ws;
@@ -1015,7 +1047,7 @@ static int print_wakeup_source_stats(struct seq_file *m,
 		active_time = ktime_set(0, 0);
 	}
 
-	seq_printf(m, "%-12s\t%lu\t\t%lu\t\t%lu\t\t%lu\t\t%lld\t\t%lld\t\t%lld\t\t%lld\t\t%lld\n",
+	seq_printf(m, "%-32s\t%lu\t\t%lu\t\t%lu\t\t%lu\t\t%lld\t\t%lld\t\t%lld\t\t%lld\t\t%lld\n",
 		   ws->name, active_count, ws->event_count,
 		   ws->wakeup_count, ws->expire_count,
 		   ktime_to_ms(active_time), ktime_to_ms(total_time),
@@ -1036,7 +1068,7 @@ static int wakeup_sources_stats_show(struct seq_file *m, void *unused)
 	struct wakeup_source *ws;
 	int srcuidx;
 
-	seq_puts(m, "name\t\tactive_count\tevent_count\twakeup_count\t"
+	seq_puts(m, "name\t\t\t\t\tactive_count\tevent_count\twakeup_count\t"
 		"expire_count\tactive_since\ttotal_time\tmax_time\t"
 		"last_change\tprevent_suspend_time\n");
 
diff --git a/drivers/base/syscore.c b/drivers/base/syscore.c
index 8d98a329f6ea..96c34a95cc62 100644
--- a/drivers/base/syscore.c
+++ b/drivers/base/syscore.c
@@ -11,6 +11,7 @@
 #include <linux/module.h>
 #include <linux/suspend.h>
 #include <trace/events/power.h>
+#include <linux/wakeup_reason.h>
 
 static LIST_HEAD(syscore_ops_list);
 static DEFINE_MUTEX(syscore_ops_lock);
@@ -75,6 +76,8 @@ int syscore_suspend(void)
 	return 0;
 
  err_out:
+	log_suspend_abort_reason("System core suspend callback %pF failed",
+		ops->suspend);
 	pr_err("PM: System core suspend callback %pF failed.\n", ops->suspend);
 
 	list_for_each_entry_continue(ops, &syscore_ops_list, node)
diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig
index 659879a56dba..91e2743f057b 100644
--- a/drivers/cpufreq/Kconfig
+++ b/drivers/cpufreq/Kconfig
@@ -102,6 +102,33 @@ config CPU_FREQ_DEFAULT_GOV_CONSERVATIVE
 	  Be aware that not all cpufreq drivers support the conservative
 	  governor. If unsure have a look at the help section of the
 	  driver. Fallback governor will be the performance governor.
+
+config CPU_FREQ_DEFAULT_GOV_INTERACTIVE
+	bool "interactive"
+	select CPU_FREQ_GOV_INTERACTIVE
+	help
+	  Use the CPUFreq governor 'interactive' as default. This allows
+	  you to get a full dynamic cpu frequency capable system by simply
+	  loading your cpufreq low-level hardware driver, using the
+	  'interactive' governor for latency-sensitive workloads.
+
+config CPU_FREQ_DEFAULT_GOV_SCHED
+	bool "sched"
+	select CPU_FREQ_GOV_SCHED
+	help
+	  Use the CPUfreq governor 'sched' as default. This scales
+	  cpu frequency using CPU utilization estimates from the
+	  scheduler.
+
+config CPU_FREQ_DEFAULT_GOV_SCHEDUTIL
+	bool "schedutil"
+	depends on SMP
+	select CPU_FREQ_GOV_SCHEDUTIL
+	select CPU_FREQ_GOV_PERFORMANCE
+	help
+	  Use the 'schedutil' CPUFreq governor by default. If unsure,
+	  have a look at the help section of that governor. The fallback
+	  governor will be 'performance'.
 endchoice
 
 config CPU_FREQ_GOV_PERFORMANCE
@@ -159,6 +186,20 @@ config CPU_FREQ_GOV_ONDEMAND
 
 	  If in doubt, say N.
 
+config CPU_FREQ_GOV_INTERACTIVE
+	bool "'interactive' cpufreq policy governor"
+	help
+	  'interactive' - This driver adds a dynamic cpufreq policy governor
+	  designed for latency-sensitive workloads.
+
+	  This governor attempts to reduce the latency of clock
+	  increases so that the system is more responsive to
+	  interactive workloads.
+
+	  For details, take a look at linux/Documentation/cpu-freq.
+
+	  If in doubt, say N.
+
 config CPU_FREQ_GOV_CONSERVATIVE
 	tristate "'conservative' cpufreq governor"
 	depends on CPU_FREQ
@@ -183,6 +224,23 @@ config CPU_FREQ_GOV_CONSERVATIVE
 
 	  If in doubt, say N.
 
+config CPU_FREQ_GOV_SCHEDUTIL
+	bool "'schedutil' cpufreq policy governor"
+	depends on CPU_FREQ && SMP
+	select CPU_FREQ_GOV_ATTR_SET
+	select IRQ_WORK
+	help
+	  This governor makes decisions based on the utilization data provided
+	  by the scheduler.  It sets the CPU frequency to be proportional to
+	  the utilization/capacity ratio coming from the scheduler.  If the
+	  utilization is frequency-invariant, the new frequency is also
+	  proportional to the maximum available frequency.  If that is not the
+	  case, it is proportional to the current frequency of the CPU.  The
+	  frequency tipping point is at utilization/capacity equal to 80% in
+	  both cases.
+
+	  If in doubt, say N.
+
 comment "CPU frequency scaling drivers"
 
 config CPUFREQ_DT
diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile
index c0af1a1281c8..04e6324fd638 100644
--- a/drivers/cpufreq/Makefile
+++ b/drivers/cpufreq/Makefile
@@ -1,5 +1,5 @@
 # CPUfreq core
-obj-$(CONFIG_CPU_FREQ)			+= cpufreq.o freq_table.o
+obj-$(CONFIG_CPU_FREQ)			+= cpufreq.o freq_table.o cpufreq_governor_attr_set.o
 
 # CPUfreq stats
 obj-$(CONFIG_CPU_FREQ_STAT)             += cpufreq_stats.o
@@ -10,6 +10,7 @@ obj-$(CONFIG_CPU_FREQ_GOV_POWERSAVE)	+= cpufreq_powersave.o
 obj-$(CONFIG_CPU_FREQ_GOV_USERSPACE)	+= cpufreq_userspace.o
 obj-$(CONFIG_CPU_FREQ_GOV_ONDEMAND)	+= cpufreq_ondemand.o
 obj-$(CONFIG_CPU_FREQ_GOV_CONSERVATIVE)	+= cpufreq_conservative.o
+obj-$(CONFIG_CPU_FREQ_GOV_INTERACTIVE)	+= cpufreq_interactive.o
 obj-$(CONFIG_CPU_FREQ_GOV_COMMON)		+= cpufreq_governor.o
 
 obj-$(CONFIG_CPUFREQ_DT)		+= cpufreq-dt.o
diff --git a/drivers/cpufreq/cpufreq-dt.c b/drivers/cpufreq/cpufreq-dt.c
index f951f911786e..a72ae98b4838 100644
--- a/drivers/cpufreq/cpufreq-dt.c
+++ b/drivers/cpufreq/cpufreq-dt.c
@@ -279,6 +279,13 @@ static int cpufreq_init(struct cpufreq_policy *policy)
 
 	policy->cpuinfo.transition_latency = transition_latency;
 
+        /*
+         * Android: set default parameters for parity between schedutil and
+         * schedfreq
+         */
+	policy->up_transition_delay_us = transition_latency / NSEC_PER_USEC;
+	policy->down_transition_delay_us = 50000; /* 50ms */
+
 	return 0;
 
 out_free_cpufreq_table:
diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
index ebed319657e7..620ab220feff 100644
--- a/drivers/cpufreq/cpufreq.c
+++ b/drivers/cpufreq/cpufreq.c
@@ -29,6 +29,9 @@
 #include <linux/suspend.h>
 #include <linux/syscore_ops.h>
 #include <linux/tick.h>
+#ifdef CONFIG_SMP
+#include <linux/sched.h>
+#endif
 #include <trace/events/power.h>
 
 static LIST_HEAD(cpufreq_policy_list);
@@ -154,6 +157,12 @@ bool have_governor_per_policy(void)
 }
 EXPORT_SYMBOL_GPL(have_governor_per_policy);
 
+bool cpufreq_driver_is_slow(void)
+{
+	return !(cpufreq_driver->flags & CPUFREQ_DRIVER_FAST);
+}
+EXPORT_SYMBOL_GPL(cpufreq_driver_is_slow);
+
 struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy)
 {
 	if (have_governor_per_policy())
@@ -347,6 +356,50 @@ static void adjust_jiffies(unsigned long val, struct cpufreq_freqs *ci)
 #endif
 }
 
+/*********************************************************************
+ *               FREQUENCY INVARIANT CPU CAPACITY                    *
+ *********************************************************************/
+
+static DEFINE_PER_CPU(unsigned long, freq_scale) = SCHED_CAPACITY_SCALE;
+static DEFINE_PER_CPU(unsigned long, max_freq_scale) = SCHED_CAPACITY_SCALE;
+
+static void
+scale_freq_capacity(struct cpufreq_policy *policy, struct cpufreq_freqs *freqs)
+{
+	unsigned long cur = freqs ? freqs->new : policy->cur;
+	unsigned long scale = (cur << SCHED_CAPACITY_SHIFT) / policy->max;
+	struct cpufreq_cpuinfo *cpuinfo = &policy->cpuinfo;
+	int cpu;
+
+	pr_debug("cpus %*pbl cur/cur max freq %lu/%u kHz freq scale %lu\n",
+		 cpumask_pr_args(policy->cpus), cur, policy->max, scale);
+
+	for_each_cpu(cpu, policy->cpus)
+		per_cpu(freq_scale, cpu) = scale;
+
+	if (freqs)
+		return;
+
+	scale = (policy->max << SCHED_CAPACITY_SHIFT) / cpuinfo->max_freq;
+
+	pr_debug("cpus %*pbl cur max/max freq %u/%u kHz max freq scale %lu\n",
+		 cpumask_pr_args(policy->cpus), policy->max, cpuinfo->max_freq,
+		 scale);
+
+	for_each_cpu(cpu, policy->cpus)
+		per_cpu(max_freq_scale, cpu) = scale;
+}
+
+unsigned long cpufreq_scale_freq_capacity(struct sched_domain *sd, int cpu)
+{
+	return per_cpu(freq_scale, cpu);
+}
+
+unsigned long cpufreq_scale_max_freq_capacity(int cpu)
+{
+	return per_cpu(max_freq_scale, cpu);
+}
+
 static void __cpufreq_notify_transition(struct cpufreq_policy *policy,
 		struct cpufreq_freqs *freqs, unsigned int state)
 {
@@ -423,6 +476,9 @@ static void cpufreq_notify_post_transition(struct cpufreq_policy *policy,
 void cpufreq_freq_transition_begin(struct cpufreq_policy *policy,
 		struct cpufreq_freqs *freqs)
 {
+#ifdef CONFIG_SMP
+	int cpu;
+#endif
 
 	/*
 	 * Catch double invocations of _begin() which lead to self-deadlock.
@@ -450,6 +506,12 @@ wait:
 
 	spin_unlock(&policy->transition_lock);
 
+	scale_freq_capacity(policy, freqs);
+#ifdef CONFIG_SMP
+	for_each_cpu(cpu, policy->cpus)
+		trace_cpu_capacity(capacity_curr_of(cpu), cpu);
+#endif
+
 	cpufreq_notify_transition(policy, freqs, CPUFREQ_PRECHANGE);
 }
 EXPORT_SYMBOL_GPL(cpufreq_freq_transition_begin);
@@ -469,6 +531,38 @@ void cpufreq_freq_transition_end(struct cpufreq_policy *policy,
 }
 EXPORT_SYMBOL_GPL(cpufreq_freq_transition_end);
 
+/**
+ * cpufreq_driver_resolve_freq - Map a target frequency to a driver-supported
+ * one.
+ * @target_freq: target frequency to resolve.
+ *
+ * The target to driver frequency mapping is cached in the policy.
+ *
+ * Return: Lowest driver-supported frequency greater than or equal to the
+ * given target_freq, subject to policy (min/max) and driver limitations.
+ */
+unsigned int cpufreq_driver_resolve_freq(struct cpufreq_policy *policy,
+					 unsigned int target_freq)
+{
+	target_freq = clamp_val(target_freq, policy->min, policy->max);
+	policy->cached_target_freq = target_freq;
+
+	if (cpufreq_driver->target_index) {
+		int idx, rv;
+
+		rv = cpufreq_frequency_table_target(policy, policy->freq_table,
+						    target_freq,
+						    CPUFREQ_RELATION_L,
+						    &idx);
+		if (rv)
+			return target_freq;
+		policy->cached_resolved_idx = idx;
+		return policy->freq_table[idx].frequency;
+        }
+
+	return target_freq;
+}
+EXPORT_SYMBOL_GPL(cpufreq_driver_resolve_freq);
 
 /*********************************************************************
  *                          SYSFS INTERFACE                          *
@@ -2131,8 +2225,11 @@ static int cpufreq_set_policy(struct cpufreq_policy *policy,
 	blocking_notifier_call_chain(&cpufreq_policy_notifier_list,
 			CPUFREQ_NOTIFY, new_policy);
 
+	scale_freq_capacity(new_policy, NULL);
+
 	policy->min = new_policy->min;
 	policy->max = new_policy->max;
+	trace_cpu_frequency_limits(policy->max, policy->min, policy->cpu);
 
 	pr_debug("new min and max freqs are %u - %u kHz\n",
 		 policy->min, policy->max);
diff --git a/drivers/cpufreq/cpufreq_governor_attr_set.c b/drivers/cpufreq/cpufreq_governor_attr_set.c
new file mode 100644
index 000000000000..52841f807a7e
--- /dev/null
+++ b/drivers/cpufreq/cpufreq_governor_attr_set.c
@@ -0,0 +1,84 @@
+/*
+ * Abstract code for CPUFreq governor tunable sysfs attributes.
+ *
+ * Copyright (C) 2016, Intel Corporation
+ * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include "cpufreq_governor.h"
+
+static inline struct gov_attr_set *to_gov_attr_set(struct kobject *kobj)
+{
+	return container_of(kobj, struct gov_attr_set, kobj);
+}
+
+static inline struct governor_attr *to_gov_attr(struct attribute *attr)
+{
+	return container_of(attr, struct governor_attr, attr);
+}
+
+static ssize_t governor_show(struct kobject *kobj, struct attribute *attr,
+			     char *buf)
+{
+	struct governor_attr *gattr = to_gov_attr(attr);
+
+	return gattr->show(to_gov_attr_set(kobj), buf);
+}
+
+static ssize_t governor_store(struct kobject *kobj, struct attribute *attr,
+			      const char *buf, size_t count)
+{
+	struct gov_attr_set *attr_set = to_gov_attr_set(kobj);
+	struct governor_attr *gattr = to_gov_attr(attr);
+	int ret;
+
+	mutex_lock(&attr_set->update_lock);
+	ret = attr_set->usage_count ? gattr->store(attr_set, buf, count) : -EBUSY;
+	mutex_unlock(&attr_set->update_lock);
+	return ret;
+}
+
+const struct sysfs_ops governor_sysfs_ops = {
+	.show	= governor_show,
+	.store	= governor_store,
+};
+EXPORT_SYMBOL_GPL(governor_sysfs_ops);
+
+void gov_attr_set_init(struct gov_attr_set *attr_set, struct list_head *list_node)
+{
+	INIT_LIST_HEAD(&attr_set->policy_list);
+	mutex_init(&attr_set->update_lock);
+	attr_set->usage_count = 1;
+	list_add(list_node, &attr_set->policy_list);
+}
+EXPORT_SYMBOL_GPL(gov_attr_set_init);
+
+void gov_attr_set_get(struct gov_attr_set *attr_set, struct list_head *list_node)
+{
+	mutex_lock(&attr_set->update_lock);
+	attr_set->usage_count++;
+	list_add(list_node, &attr_set->policy_list);
+	mutex_unlock(&attr_set->update_lock);
+}
+EXPORT_SYMBOL_GPL(gov_attr_set_get);
+
+unsigned int gov_attr_set_put(struct gov_attr_set *attr_set, struct list_head *list_node)
+{
+	unsigned int count;
+
+	mutex_lock(&attr_set->update_lock);
+	list_del(list_node);
+	count = --attr_set->usage_count;
+	mutex_unlock(&attr_set->update_lock);
+	if (count)
+		return count;
+
+	kobject_put(&attr_set->kobj);
+	mutex_destroy(&attr_set->update_lock);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gov_attr_set_put);
diff --git a/drivers/cpufreq/cpufreq_interactive.c b/drivers/cpufreq/cpufreq_interactive.c
new file mode 100644
index 000000000000..e0a586895136
--- /dev/null
+++ b/drivers/cpufreq/cpufreq_interactive.c
@@ -0,0 +1,1374 @@
+/*
+ * drivers/cpufreq/cpufreq_interactive.c
+ *
+ * Copyright (C) 2010 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * Author: Mike Chan (mike@android.com)
+ *
+ */
+
+#include <linux/cpu.h>
+#include <linux/cpumask.h>
+#include <linux/cpufreq.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#include <linux/rwsem.h>
+#include <linux/sched.h>
+#include <linux/sched/rt.h>
+#include <linux/tick.h>
+#include <linux/time.h>
+#include <linux/timer.h>
+#include <linux/workqueue.h>
+#include <linux/kthread.h>
+#include <linux/slab.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/cpufreq_interactive.h>
+
+struct cpufreq_interactive_cpuinfo {
+	struct timer_list cpu_timer;
+	struct timer_list cpu_slack_timer;
+	spinlock_t load_lock; /* protects the next 4 fields */
+	u64 time_in_idle;
+	u64 time_in_idle_timestamp;
+	u64 cputime_speedadj;
+	u64 cputime_speedadj_timestamp;
+	struct cpufreq_policy *policy;
+	struct cpufreq_frequency_table *freq_table;
+	spinlock_t target_freq_lock; /*protects target freq */
+	unsigned int target_freq;
+	unsigned int floor_freq;
+	u64 pol_floor_val_time; /* policy floor_validate_time */
+	u64 loc_floor_val_time; /* per-cpu floor_validate_time */
+	u64 pol_hispeed_val_time; /* policy hispeed_validate_time */
+	u64 loc_hispeed_val_time; /* per-cpu hispeed_validate_time */
+	struct rw_semaphore enable_sem;
+	int governor_enabled;
+};
+
+static DEFINE_PER_CPU(struct cpufreq_interactive_cpuinfo, cpuinfo);
+
+/* realtime thread handles frequency scaling */
+static struct task_struct *speedchange_task;
+static cpumask_t speedchange_cpumask;
+static spinlock_t speedchange_cpumask_lock;
+static struct mutex gov_lock;
+
+/* Target load.  Lower values result in higher CPU speeds. */
+#define DEFAULT_TARGET_LOAD 90
+static unsigned int default_target_loads[] = {DEFAULT_TARGET_LOAD};
+
+#define DEFAULT_TIMER_RATE (20 * USEC_PER_MSEC)
+#define DEFAULT_ABOVE_HISPEED_DELAY DEFAULT_TIMER_RATE
+static unsigned int default_above_hispeed_delay[] = {
+	DEFAULT_ABOVE_HISPEED_DELAY };
+
+struct cpufreq_interactive_tunables {
+	int usage_count;
+	/* Hi speed to bump to from lo speed when load burst (default max) */
+	unsigned int hispeed_freq;
+	/* Go to hi speed when CPU load at or above this value. */
+#define DEFAULT_GO_HISPEED_LOAD 99
+	unsigned long go_hispeed_load;
+	/* Target load. Lower values result in higher CPU speeds. */
+	spinlock_t target_loads_lock;
+	unsigned int *target_loads;
+	int ntarget_loads;
+	/*
+	 * The minimum amount of time to spend at a frequency before we can ramp
+	 * down.
+	 */
+#define DEFAULT_MIN_SAMPLE_TIME (80 * USEC_PER_MSEC)
+	unsigned long min_sample_time;
+	/*
+	 * The sample rate of the timer used to increase frequency
+	 */
+	unsigned long timer_rate;
+	/*
+	 * Wait this long before raising speed above hispeed, by default a
+	 * single timer interval.
+	 */
+	spinlock_t above_hispeed_delay_lock;
+	unsigned int *above_hispeed_delay;
+	int nabove_hispeed_delay;
+	/* Non-zero means indefinite speed boost active */
+	int boost_val;
+	/* Duration of a boot pulse in usecs */
+	int boostpulse_duration_val;
+	/* End time of boost pulse in ktime converted to usecs */
+	u64 boostpulse_endtime;
+	bool boosted;
+	/*
+	 * Max additional time to wait in idle, beyond timer_rate, at speeds
+	 * above minimum before wakeup to reduce speed, or -1 if unnecessary.
+	 */
+#define DEFAULT_TIMER_SLACK (4 * DEFAULT_TIMER_RATE)
+	int timer_slack_val;
+	bool io_is_busy;
+};
+
+/* For cases where we have single governor instance for system */
+static struct cpufreq_interactive_tunables *common_tunables;
+
+static struct attribute_group *get_sysfs_attr(void);
+
+static void cpufreq_interactive_timer_resched(
+	struct cpufreq_interactive_cpuinfo *pcpu)
+{
+	struct cpufreq_interactive_tunables *tunables =
+		pcpu->policy->governor_data;
+	unsigned long expires;
+	unsigned long flags;
+
+	spin_lock_irqsave(&pcpu->load_lock, flags);
+	pcpu->time_in_idle =
+		get_cpu_idle_time(smp_processor_id(),
+				  &pcpu->time_in_idle_timestamp,
+				  tunables->io_is_busy);
+	pcpu->cputime_speedadj = 0;
+	pcpu->cputime_speedadj_timestamp = pcpu->time_in_idle_timestamp;
+	expires = jiffies + usecs_to_jiffies(tunables->timer_rate);
+	mod_timer_pinned(&pcpu->cpu_timer, expires);
+
+	if (tunables->timer_slack_val >= 0 &&
+	    pcpu->target_freq > pcpu->policy->min) {
+		expires += usecs_to_jiffies(tunables->timer_slack_val);
+		mod_timer_pinned(&pcpu->cpu_slack_timer, expires);
+	}
+
+	spin_unlock_irqrestore(&pcpu->load_lock, flags);
+}
+
+/* The caller shall take enable_sem write semaphore to avoid any timer race.
+ * The cpu_timer and cpu_slack_timer must be deactivated when calling this
+ * function.
+ */
+static void cpufreq_interactive_timer_start(
+	struct cpufreq_interactive_tunables *tunables, int cpu)
+{
+	struct cpufreq_interactive_cpuinfo *pcpu = &per_cpu(cpuinfo, cpu);
+	unsigned long expires = jiffies +
+		usecs_to_jiffies(tunables->timer_rate);
+	unsigned long flags;
+
+	pcpu->cpu_timer.expires = expires;
+	add_timer_on(&pcpu->cpu_timer, cpu);
+	if (tunables->timer_slack_val >= 0 &&
+	    pcpu->target_freq > pcpu->policy->min) {
+		expires += usecs_to_jiffies(tunables->timer_slack_val);
+		pcpu->cpu_slack_timer.expires = expires;
+		add_timer_on(&pcpu->cpu_slack_timer, cpu);
+	}
+
+	spin_lock_irqsave(&pcpu->load_lock, flags);
+	pcpu->time_in_idle =
+		get_cpu_idle_time(cpu, &pcpu->time_in_idle_timestamp,
+				  tunables->io_is_busy);
+	pcpu->cputime_speedadj = 0;
+	pcpu->cputime_speedadj_timestamp = pcpu->time_in_idle_timestamp;
+	spin_unlock_irqrestore(&pcpu->load_lock, flags);
+}
+
+static unsigned int freq_to_above_hispeed_delay(
+	struct cpufreq_interactive_tunables *tunables,
+	unsigned int freq)
+{
+	int i;
+	unsigned int ret;
+	unsigned long flags;
+
+	spin_lock_irqsave(&tunables->above_hispeed_delay_lock, flags);
+
+	for (i = 0; i < tunables->nabove_hispeed_delay - 1 &&
+			freq >= tunables->above_hispeed_delay[i+1]; i += 2)
+		;
+
+	ret = tunables->above_hispeed_delay[i];
+	spin_unlock_irqrestore(&tunables->above_hispeed_delay_lock, flags);
+	return ret;
+}
+
+static unsigned int freq_to_targetload(
+	struct cpufreq_interactive_tunables *tunables, unsigned int freq)
+{
+	int i;
+	unsigned int ret;
+	unsigned long flags;
+
+	spin_lock_irqsave(&tunables->target_loads_lock, flags);
+
+	for (i = 0; i < tunables->ntarget_loads - 1 &&
+		    freq >= tunables->target_loads[i+1]; i += 2)
+		;
+
+	ret = tunables->target_loads[i];
+	spin_unlock_irqrestore(&tunables->target_loads_lock, flags);
+	return ret;
+}
+
+/*
+ * If increasing frequencies never map to a lower target load then
+ * choose_freq() will find the minimum frequency that does not exceed its
+ * target load given the current load.
+ */
+static unsigned int choose_freq(struct cpufreq_interactive_cpuinfo *pcpu,
+		unsigned int loadadjfreq)
+{
+	unsigned int freq = pcpu->policy->cur;
+	unsigned int prevfreq, freqmin, freqmax;
+	unsigned int tl;
+	int index;
+
+	freqmin = 0;
+	freqmax = UINT_MAX;
+
+	do {
+		prevfreq = freq;
+		tl = freq_to_targetload(pcpu->policy->governor_data, freq);
+
+		/*
+		 * Find the lowest frequency where the computed load is less
+		 * than or equal to the target load.
+		 */
+
+		if (cpufreq_frequency_table_target(
+			    pcpu->policy, pcpu->freq_table, loadadjfreq / tl,
+			    CPUFREQ_RELATION_L, &index))
+			break;
+		freq = pcpu->freq_table[index].frequency;
+
+		if (freq > prevfreq) {
+			/* The previous frequency is too low. */
+			freqmin = prevfreq;
+
+			if (freq >= freqmax) {
+				/*
+				 * Find the highest frequency that is less
+				 * than freqmax.
+				 */
+				if (cpufreq_frequency_table_target(
+					    pcpu->policy, pcpu->freq_table,
+					    freqmax - 1, CPUFREQ_RELATION_H,
+					    &index))
+					break;
+				freq = pcpu->freq_table[index].frequency;
+
+				if (freq == freqmin) {
+					/*
+					 * The first frequency below freqmax
+					 * has already been found to be too
+					 * low.  freqmax is the lowest speed
+					 * we found that is fast enough.
+					 */
+					freq = freqmax;
+					break;
+				}
+			}
+		} else if (freq < prevfreq) {
+			/* The previous frequency is high enough. */
+			freqmax = prevfreq;
+
+			if (freq <= freqmin) {
+				/*
+				 * Find the lowest frequency that is higher
+				 * than freqmin.
+				 */
+				if (cpufreq_frequency_table_target(
+					    pcpu->policy, pcpu->freq_table,
+					    freqmin + 1, CPUFREQ_RELATION_L,
+					    &index))
+					break;
+				freq = pcpu->freq_table[index].frequency;
+
+				/*
+				 * If freqmax is the first frequency above
+				 * freqmin then we have already found that
+				 * this speed is fast enough.
+				 */
+				if (freq == freqmax)
+					break;
+			}
+		}
+
+		/* If same frequency chosen as previous then done. */
+	} while (freq != prevfreq);
+
+	return freq;
+}
+
+static u64 update_load(int cpu)
+{
+	struct cpufreq_interactive_cpuinfo *pcpu = &per_cpu(cpuinfo, cpu);
+	struct cpufreq_interactive_tunables *tunables =
+		pcpu->policy->governor_data;
+	u64 now;
+	u64 now_idle;
+	u64 delta_idle;
+	u64 delta_time;
+	u64 active_time;
+
+	now_idle = get_cpu_idle_time(cpu, &now, tunables->io_is_busy);
+	delta_idle = (now_idle - pcpu->time_in_idle);
+	delta_time = (now - pcpu->time_in_idle_timestamp);
+
+	if (delta_time <= delta_idle)
+		active_time = 0;
+	else
+		active_time = delta_time - delta_idle;
+
+	pcpu->cputime_speedadj += active_time * pcpu->policy->cur;
+
+	pcpu->time_in_idle = now_idle;
+	pcpu->time_in_idle_timestamp = now;
+	return now;
+}
+
+static void cpufreq_interactive_timer(unsigned long data)
+{
+	u64 now;
+	unsigned int delta_time;
+	u64 cputime_speedadj;
+	int cpu_load;
+	struct cpufreq_interactive_cpuinfo *pcpu =
+		&per_cpu(cpuinfo, data);
+	struct cpufreq_interactive_tunables *tunables =
+		pcpu->policy->governor_data;
+	unsigned int new_freq;
+	unsigned int loadadjfreq;
+	unsigned int index;
+	unsigned long flags;
+	u64 max_fvtime;
+
+	if (!down_read_trylock(&pcpu->enable_sem))
+		return;
+	if (!pcpu->governor_enabled)
+		goto exit;
+
+	spin_lock_irqsave(&pcpu->load_lock, flags);
+	now = update_load(data);
+	delta_time = (unsigned int)(now - pcpu->cputime_speedadj_timestamp);
+	cputime_speedadj = pcpu->cputime_speedadj;
+	spin_unlock_irqrestore(&pcpu->load_lock, flags);
+
+	if (WARN_ON_ONCE(!delta_time))
+		goto rearm;
+
+	spin_lock_irqsave(&pcpu->target_freq_lock, flags);
+	do_div(cputime_speedadj, delta_time);
+	loadadjfreq = (unsigned int)cputime_speedadj * 100;
+	cpu_load = loadadjfreq / pcpu->policy->cur;
+	tunables->boosted = tunables->boost_val || now < tunables->boostpulse_endtime;
+
+	if (cpu_load >= tunables->go_hispeed_load || tunables->boosted) {
+		if (pcpu->policy->cur < tunables->hispeed_freq) {
+			new_freq = tunables->hispeed_freq;
+		} else {
+			new_freq = choose_freq(pcpu, loadadjfreq);
+
+			if (new_freq < tunables->hispeed_freq)
+				new_freq = tunables->hispeed_freq;
+		}
+	} else {
+		new_freq = choose_freq(pcpu, loadadjfreq);
+		if (new_freq > tunables->hispeed_freq &&
+				pcpu->policy->cur < tunables->hispeed_freq)
+			new_freq = tunables->hispeed_freq;
+	}
+
+	if (pcpu->policy->cur >= tunables->hispeed_freq &&
+	    new_freq > pcpu->policy->cur &&
+	    now - pcpu->pol_hispeed_val_time <
+	    freq_to_above_hispeed_delay(tunables, pcpu->policy->cur)) {
+		trace_cpufreq_interactive_notyet(
+			data, cpu_load, pcpu->target_freq,
+			pcpu->policy->cur, new_freq);
+		spin_unlock_irqrestore(&pcpu->target_freq_lock, flags);
+		goto rearm;
+	}
+
+	pcpu->loc_hispeed_val_time = now;
+
+	if (cpufreq_frequency_table_target(pcpu->policy, pcpu->freq_table,
+					   new_freq, CPUFREQ_RELATION_L,
+					   &index)) {
+		spin_unlock_irqrestore(&pcpu->target_freq_lock, flags);
+		goto rearm;
+	}
+
+	new_freq = pcpu->freq_table[index].frequency;
+
+	/*
+	 * Do not scale below floor_freq unless we have been at or above the
+	 * floor frequency for the minimum sample time since last validated.
+	 */
+	max_fvtime = max(pcpu->pol_floor_val_time, pcpu->loc_floor_val_time);
+	if (new_freq < pcpu->floor_freq &&
+	    pcpu->target_freq >= pcpu->policy->cur) {
+		if (now - max_fvtime < tunables->min_sample_time) {
+			trace_cpufreq_interactive_notyet(
+				data, cpu_load, pcpu->target_freq,
+				pcpu->policy->cur, new_freq);
+			spin_unlock_irqrestore(&pcpu->target_freq_lock, flags);
+			goto rearm;
+		}
+	}
+
+	/*
+	 * Update the timestamp for checking whether speed has been held at
+	 * or above the selected frequency for a minimum of min_sample_time,
+	 * if not boosted to hispeed_freq.  If boosted to hispeed_freq then we
+	 * allow the speed to drop as soon as the boostpulse duration expires
+	 * (or the indefinite boost is turned off).
+	 */
+
+	if (!tunables->boosted || new_freq > tunables->hispeed_freq) {
+		pcpu->floor_freq = new_freq;
+		if (pcpu->target_freq >= pcpu->policy->cur ||
+		    new_freq >= pcpu->policy->cur)
+			pcpu->loc_floor_val_time = now;
+	}
+
+	if (pcpu->target_freq == new_freq &&
+			pcpu->target_freq <= pcpu->policy->cur) {
+		trace_cpufreq_interactive_already(
+			data, cpu_load, pcpu->target_freq,
+			pcpu->policy->cur, new_freq);
+		spin_unlock_irqrestore(&pcpu->target_freq_lock, flags);
+		goto rearm;
+	}
+
+	trace_cpufreq_interactive_target(data, cpu_load, pcpu->target_freq,
+					 pcpu->policy->cur, new_freq);
+
+	pcpu->target_freq = new_freq;
+	spin_unlock_irqrestore(&pcpu->target_freq_lock, flags);
+	spin_lock_irqsave(&speedchange_cpumask_lock, flags);
+	cpumask_set_cpu(data, &speedchange_cpumask);
+	spin_unlock_irqrestore(&speedchange_cpumask_lock, flags);
+	wake_up_process(speedchange_task);
+
+rearm:
+	if (!timer_pending(&pcpu->cpu_timer))
+		cpufreq_interactive_timer_resched(pcpu);
+
+exit:
+	up_read(&pcpu->enable_sem);
+	return;
+}
+
+static void cpufreq_interactive_idle_end(void)
+{
+	struct cpufreq_interactive_cpuinfo *pcpu =
+		&per_cpu(cpuinfo, smp_processor_id());
+
+	if (!down_read_trylock(&pcpu->enable_sem))
+		return;
+	if (!pcpu->governor_enabled) {
+		up_read(&pcpu->enable_sem);
+		return;
+	}
+
+	/* Arm the timer for 1-2 ticks later if not already. */
+	if (!timer_pending(&pcpu->cpu_timer)) {
+		cpufreq_interactive_timer_resched(pcpu);
+	} else if (time_after_eq(jiffies, pcpu->cpu_timer.expires)) {
+		del_timer(&pcpu->cpu_timer);
+		del_timer(&pcpu->cpu_slack_timer);
+		cpufreq_interactive_timer(smp_processor_id());
+	}
+
+	up_read(&pcpu->enable_sem);
+}
+
+static void cpufreq_interactive_get_policy_info(struct cpufreq_policy *policy,
+						unsigned int *pmax_freq,
+						u64 *phvt, u64 *pfvt)
+{
+	struct cpufreq_interactive_cpuinfo *pcpu;
+	unsigned int max_freq = 0;
+	u64 hvt = ~0ULL, fvt = 0;
+	unsigned int i;
+
+	for_each_cpu(i, policy->cpus) {
+		pcpu = &per_cpu(cpuinfo, i);
+
+		fvt = max(fvt, pcpu->loc_floor_val_time);
+		if (pcpu->target_freq > max_freq) {
+			max_freq = pcpu->target_freq;
+			hvt = pcpu->loc_hispeed_val_time;
+		} else if (pcpu->target_freq == max_freq) {
+			hvt = min(hvt, pcpu->loc_hispeed_val_time);
+		}
+	}
+
+	*pmax_freq = max_freq;
+	*phvt = hvt;
+	*pfvt = fvt;
+}
+
+static void cpufreq_interactive_adjust_cpu(unsigned int cpu,
+					   struct cpufreq_policy *policy)
+{
+	struct cpufreq_interactive_cpuinfo *pcpu;
+	u64 hvt, fvt;
+	unsigned int max_freq;
+	int i;
+
+	cpufreq_interactive_get_policy_info(policy, &max_freq, &hvt, &fvt);
+
+	for_each_cpu(i, policy->cpus) {
+		pcpu = &per_cpu(cpuinfo, i);
+		pcpu->pol_floor_val_time = fvt;
+	}
+
+	if (max_freq != policy->cur) {
+		__cpufreq_driver_target(policy, max_freq, CPUFREQ_RELATION_H);
+		for_each_cpu(i, policy->cpus) {
+			pcpu = &per_cpu(cpuinfo, i);
+			pcpu->pol_hispeed_val_time = hvt;
+		}
+	}
+
+	trace_cpufreq_interactive_setspeed(cpu, max_freq, policy->cur);
+}
+
+static int cpufreq_interactive_speedchange_task(void *data)
+{
+	unsigned int cpu;
+	cpumask_t tmp_mask;
+	unsigned long flags;
+	struct cpufreq_interactive_cpuinfo *pcpu;
+
+	while (1) {
+		set_current_state(TASK_INTERRUPTIBLE);
+		spin_lock_irqsave(&speedchange_cpumask_lock, flags);
+
+		if (cpumask_empty(&speedchange_cpumask)) {
+			spin_unlock_irqrestore(&speedchange_cpumask_lock,
+					       flags);
+			schedule();
+
+			if (kthread_should_stop())
+				break;
+
+			spin_lock_irqsave(&speedchange_cpumask_lock, flags);
+		}
+
+		set_current_state(TASK_RUNNING);
+		tmp_mask = speedchange_cpumask;
+		cpumask_clear(&speedchange_cpumask);
+		spin_unlock_irqrestore(&speedchange_cpumask_lock, flags);
+
+		for_each_cpu(cpu, &tmp_mask) {
+			pcpu = &per_cpu(cpuinfo, cpu);
+
+			down_write(&pcpu->policy->rwsem);
+
+			if (likely(down_read_trylock(&pcpu->enable_sem))) {
+				if (likely(pcpu->governor_enabled))
+					cpufreq_interactive_adjust_cpu(cpu,
+							pcpu->policy);
+				up_read(&pcpu->enable_sem);
+			}
+
+			up_write(&pcpu->policy->rwsem);
+		}
+	}
+
+	return 0;
+}
+
+static void cpufreq_interactive_boost(struct cpufreq_interactive_tunables *tunables)
+{
+	int i;
+	int anyboost = 0;
+	unsigned long flags[2];
+	struct cpufreq_interactive_cpuinfo *pcpu;
+
+	tunables->boosted = true;
+
+	spin_lock_irqsave(&speedchange_cpumask_lock, flags[0]);
+
+	for_each_online_cpu(i) {
+		pcpu = &per_cpu(cpuinfo, i);
+
+		if (!down_read_trylock(&pcpu->enable_sem))
+			continue;
+
+		if (!pcpu->governor_enabled) {
+			up_read(&pcpu->enable_sem);
+			continue;
+		}
+
+		if (tunables != pcpu->policy->governor_data) {
+			up_read(&pcpu->enable_sem);
+			continue;
+		}
+
+		spin_lock_irqsave(&pcpu->target_freq_lock, flags[1]);
+		if (pcpu->target_freq < tunables->hispeed_freq) {
+			pcpu->target_freq = tunables->hispeed_freq;
+			cpumask_set_cpu(i, &speedchange_cpumask);
+			pcpu->pol_hispeed_val_time =
+				ktime_to_us(ktime_get());
+			anyboost = 1;
+		}
+		spin_unlock_irqrestore(&pcpu->target_freq_lock, flags[1]);
+
+		up_read(&pcpu->enable_sem);
+	}
+
+	spin_unlock_irqrestore(&speedchange_cpumask_lock, flags[0]);
+
+	if (anyboost)
+		wake_up_process(speedchange_task);
+}
+
+static int cpufreq_interactive_notifier(
+	struct notifier_block *nb, unsigned long val, void *data)
+{
+	struct cpufreq_freqs *freq = data;
+	struct cpufreq_interactive_cpuinfo *pcpu;
+	int cpu;
+	unsigned long flags;
+
+	if (val == CPUFREQ_POSTCHANGE) {
+		pcpu = &per_cpu(cpuinfo, freq->cpu);
+		if (!down_read_trylock(&pcpu->enable_sem))
+			return 0;
+		if (!pcpu->governor_enabled) {
+			up_read(&pcpu->enable_sem);
+			return 0;
+		}
+
+		for_each_cpu(cpu, pcpu->policy->cpus) {
+			struct cpufreq_interactive_cpuinfo *pjcpu =
+				&per_cpu(cpuinfo, cpu);
+			if (cpu != freq->cpu) {
+				if (!down_read_trylock(&pjcpu->enable_sem))
+					continue;
+				if (!pjcpu->governor_enabled) {
+					up_read(&pjcpu->enable_sem);
+					continue;
+				}
+			}
+			spin_lock_irqsave(&pjcpu->load_lock, flags);
+			update_load(cpu);
+			spin_unlock_irqrestore(&pjcpu->load_lock, flags);
+			if (cpu != freq->cpu)
+				up_read(&pjcpu->enable_sem);
+		}
+
+		up_read(&pcpu->enable_sem);
+	}
+	return 0;
+}
+
+static struct notifier_block cpufreq_notifier_block = {
+	.notifier_call = cpufreq_interactive_notifier,
+};
+
+static unsigned int *get_tokenized_data(const char *buf, int *num_tokens)
+{
+	const char *cp;
+	int i;
+	int ntokens = 1;
+	unsigned int *tokenized_data;
+	int err = -EINVAL;
+
+	cp = buf;
+	while ((cp = strpbrk(cp + 1, " :")))
+		ntokens++;
+
+	if (!(ntokens & 0x1))
+		goto err;
+
+	tokenized_data = kmalloc(ntokens * sizeof(unsigned int), GFP_KERNEL);
+	if (!tokenized_data) {
+		err = -ENOMEM;
+		goto err;
+	}
+
+	cp = buf;
+	i = 0;
+	while (i < ntokens) {
+		if (sscanf(cp, "%u", &tokenized_data[i++]) != 1)
+			goto err_kfree;
+
+		cp = strpbrk(cp, " :");
+		if (!cp)
+			break;
+		cp++;
+	}
+
+	if (i != ntokens)
+		goto err_kfree;
+
+	*num_tokens = ntokens;
+	return tokenized_data;
+
+err_kfree:
+	kfree(tokenized_data);
+err:
+	return ERR_PTR(err);
+}
+
+static ssize_t show_target_loads(
+	struct cpufreq_interactive_tunables *tunables,
+	char *buf)
+{
+	int i;
+	ssize_t ret = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&tunables->target_loads_lock, flags);
+
+	for (i = 0; i < tunables->ntarget_loads; i++)
+		ret += sprintf(buf + ret, "%u%s", tunables->target_loads[i],
+			       i & 0x1 ? ":" : " ");
+
+	sprintf(buf + ret - 1, "\n");
+	spin_unlock_irqrestore(&tunables->target_loads_lock, flags);
+	return ret;
+}
+
+static ssize_t store_target_loads(
+	struct cpufreq_interactive_tunables *tunables,
+	const char *buf, size_t count)
+{
+	int ntokens;
+	unsigned int *new_target_loads = NULL;
+	unsigned long flags;
+
+	new_target_loads = get_tokenized_data(buf, &ntokens);
+	if (IS_ERR(new_target_loads))
+		return PTR_RET(new_target_loads);
+
+	spin_lock_irqsave(&tunables->target_loads_lock, flags);
+	if (tunables->target_loads != default_target_loads)
+		kfree(tunables->target_loads);
+	tunables->target_loads = new_target_loads;
+	tunables->ntarget_loads = ntokens;
+	spin_unlock_irqrestore(&tunables->target_loads_lock, flags);
+	return count;
+}
+
+static ssize_t show_above_hispeed_delay(
+	struct cpufreq_interactive_tunables *tunables, char *buf)
+{
+	int i;
+	ssize_t ret = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&tunables->above_hispeed_delay_lock, flags);
+
+	for (i = 0; i < tunables->nabove_hispeed_delay; i++)
+		ret += sprintf(buf + ret, "%u%s",
+			       tunables->above_hispeed_delay[i],
+			       i & 0x1 ? ":" : " ");
+
+	sprintf(buf + ret - 1, "\n");
+	spin_unlock_irqrestore(&tunables->above_hispeed_delay_lock, flags);
+	return ret;
+}
+
+static ssize_t store_above_hispeed_delay(
+	struct cpufreq_interactive_tunables *tunables,
+	const char *buf, size_t count)
+{
+	int ntokens;
+	unsigned int *new_above_hispeed_delay = NULL;
+	unsigned long flags;
+
+	new_above_hispeed_delay = get_tokenized_data(buf, &ntokens);
+	if (IS_ERR(new_above_hispeed_delay))
+		return PTR_RET(new_above_hispeed_delay);
+
+	spin_lock_irqsave(&tunables->above_hispeed_delay_lock, flags);
+	if (tunables->above_hispeed_delay != default_above_hispeed_delay)
+		kfree(tunables->above_hispeed_delay);
+	tunables->above_hispeed_delay = new_above_hispeed_delay;
+	tunables->nabove_hispeed_delay = ntokens;
+	spin_unlock_irqrestore(&tunables->above_hispeed_delay_lock, flags);
+	return count;
+
+}
+
+static ssize_t show_hispeed_freq(struct cpufreq_interactive_tunables *tunables,
+		char *buf)
+{
+	return sprintf(buf, "%u\n", tunables->hispeed_freq);
+}
+
+static ssize_t store_hispeed_freq(struct cpufreq_interactive_tunables *tunables,
+		const char *buf, size_t count)
+{
+	int ret;
+	long unsigned int val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+	tunables->hispeed_freq = val;
+	return count;
+}
+
+static ssize_t show_go_hispeed_load(struct cpufreq_interactive_tunables
+		*tunables, char *buf)
+{
+	return sprintf(buf, "%lu\n", tunables->go_hispeed_load);
+}
+
+static ssize_t store_go_hispeed_load(struct cpufreq_interactive_tunables
+		*tunables, const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+	tunables->go_hispeed_load = val;
+	return count;
+}
+
+static ssize_t show_min_sample_time(struct cpufreq_interactive_tunables
+		*tunables, char *buf)
+{
+	return sprintf(buf, "%lu\n", tunables->min_sample_time);
+}
+
+static ssize_t store_min_sample_time(struct cpufreq_interactive_tunables
+		*tunables, const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+	tunables->min_sample_time = val;
+	return count;
+}
+
+static ssize_t show_timer_rate(struct cpufreq_interactive_tunables *tunables,
+		char *buf)
+{
+	return sprintf(buf, "%lu\n", tunables->timer_rate);
+}
+
+static ssize_t store_timer_rate(struct cpufreq_interactive_tunables *tunables,
+		const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val, val_round;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+
+	val_round = jiffies_to_usecs(usecs_to_jiffies(val));
+	if (val != val_round)
+		pr_warn("timer_rate not aligned to jiffy. Rounded up to %lu\n",
+			val_round);
+
+	tunables->timer_rate = val_round;
+	return count;
+}
+
+static ssize_t show_timer_slack(struct cpufreq_interactive_tunables *tunables,
+		char *buf)
+{
+	return sprintf(buf, "%d\n", tunables->timer_slack_val);
+}
+
+static ssize_t store_timer_slack(struct cpufreq_interactive_tunables *tunables,
+		const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val;
+
+	ret = kstrtol(buf, 10, &val);
+	if (ret < 0)
+		return ret;
+
+	tunables->timer_slack_val = val;
+	return count;
+}
+
+static ssize_t show_boost(struct cpufreq_interactive_tunables *tunables,
+			  char *buf)
+{
+	return sprintf(buf, "%d\n", tunables->boost_val);
+}
+
+static ssize_t store_boost(struct cpufreq_interactive_tunables *tunables,
+			   const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+
+	tunables->boost_val = val;
+
+	if (tunables->boost_val) {
+		trace_cpufreq_interactive_boost("on");
+		if (!tunables->boosted)
+			cpufreq_interactive_boost(tunables);
+	} else {
+		tunables->boostpulse_endtime = ktime_to_us(ktime_get());
+		trace_cpufreq_interactive_unboost("off");
+	}
+
+	return count;
+}
+
+static ssize_t store_boostpulse(struct cpufreq_interactive_tunables *tunables,
+				const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+
+	tunables->boostpulse_endtime = ktime_to_us(ktime_get()) +
+		tunables->boostpulse_duration_val;
+	trace_cpufreq_interactive_boost("pulse");
+	if (!tunables->boosted)
+		cpufreq_interactive_boost(tunables);
+	return count;
+}
+
+static ssize_t show_boostpulse_duration(struct cpufreq_interactive_tunables
+		*tunables, char *buf)
+{
+	return sprintf(buf, "%d\n", tunables->boostpulse_duration_val);
+}
+
+static ssize_t store_boostpulse_duration(struct cpufreq_interactive_tunables
+		*tunables, const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+
+	tunables->boostpulse_duration_val = val;
+	return count;
+}
+
+static ssize_t show_io_is_busy(struct cpufreq_interactive_tunables *tunables,
+		char *buf)
+{
+	return sprintf(buf, "%u\n", tunables->io_is_busy);
+}
+
+static ssize_t store_io_is_busy(struct cpufreq_interactive_tunables *tunables,
+		const char *buf, size_t count)
+{
+	int ret;
+	unsigned long val;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret < 0)
+		return ret;
+	tunables->io_is_busy = val;
+	return count;
+}
+
+/*
+ * Create show/store routines
+ * - sys: One governor instance for complete SYSTEM
+ * - pol: One governor instance per struct cpufreq_policy
+ */
+#define show_gov_pol_sys(file_name)					\
+static ssize_t show_##file_name##_gov_sys				\
+(struct kobject *kobj, struct attribute *attr, char *buf)		\
+{									\
+	return show_##file_name(common_tunables, buf);			\
+}									\
+									\
+static ssize_t show_##file_name##_gov_pol				\
+(struct cpufreq_policy *policy, char *buf)				\
+{									\
+	return show_##file_name(policy->governor_data, buf);		\
+}
+
+#define store_gov_pol_sys(file_name)					\
+static ssize_t store_##file_name##_gov_sys				\
+(struct kobject *kobj, struct attribute *attr, const char *buf,		\
+	size_t count)							\
+{									\
+	return store_##file_name(common_tunables, buf, count);		\
+}									\
+									\
+static ssize_t store_##file_name##_gov_pol				\
+(struct cpufreq_policy *policy, const char *buf, size_t count)		\
+{									\
+	return store_##file_name(policy->governor_data, buf, count);	\
+}
+
+#define show_store_gov_pol_sys(file_name)				\
+show_gov_pol_sys(file_name);						\
+store_gov_pol_sys(file_name)
+
+show_store_gov_pol_sys(target_loads);
+show_store_gov_pol_sys(above_hispeed_delay);
+show_store_gov_pol_sys(hispeed_freq);
+show_store_gov_pol_sys(go_hispeed_load);
+show_store_gov_pol_sys(min_sample_time);
+show_store_gov_pol_sys(timer_rate);
+show_store_gov_pol_sys(timer_slack);
+show_store_gov_pol_sys(boost);
+store_gov_pol_sys(boostpulse);
+show_store_gov_pol_sys(boostpulse_duration);
+show_store_gov_pol_sys(io_is_busy);
+
+#define gov_sys_attr_rw(_name)						\
+static struct global_attr _name##_gov_sys =				\
+__ATTR(_name, 0644, show_##_name##_gov_sys, store_##_name##_gov_sys)
+
+#define gov_pol_attr_rw(_name)						\
+static struct freq_attr _name##_gov_pol =				\
+__ATTR(_name, 0644, show_##_name##_gov_pol, store_##_name##_gov_pol)
+
+#define gov_sys_pol_attr_rw(_name)					\
+	gov_sys_attr_rw(_name);						\
+	gov_pol_attr_rw(_name)
+
+gov_sys_pol_attr_rw(target_loads);
+gov_sys_pol_attr_rw(above_hispeed_delay);
+gov_sys_pol_attr_rw(hispeed_freq);
+gov_sys_pol_attr_rw(go_hispeed_load);
+gov_sys_pol_attr_rw(min_sample_time);
+gov_sys_pol_attr_rw(timer_rate);
+gov_sys_pol_attr_rw(timer_slack);
+gov_sys_pol_attr_rw(boost);
+gov_sys_pol_attr_rw(boostpulse_duration);
+gov_sys_pol_attr_rw(io_is_busy);
+
+static struct global_attr boostpulse_gov_sys =
+	__ATTR(boostpulse, 0200, NULL, store_boostpulse_gov_sys);
+
+static struct freq_attr boostpulse_gov_pol =
+	__ATTR(boostpulse, 0200, NULL, store_boostpulse_gov_pol);
+
+/* One Governor instance for entire system */
+static struct attribute *interactive_attributes_gov_sys[] = {
+	&target_loads_gov_sys.attr,
+	&above_hispeed_delay_gov_sys.attr,
+	&hispeed_freq_gov_sys.attr,
+	&go_hispeed_load_gov_sys.attr,
+	&min_sample_time_gov_sys.attr,
+	&timer_rate_gov_sys.attr,
+	&timer_slack_gov_sys.attr,
+	&boost_gov_sys.attr,
+	&boostpulse_gov_sys.attr,
+	&boostpulse_duration_gov_sys.attr,
+	&io_is_busy_gov_sys.attr,
+	NULL,
+};
+
+static struct attribute_group interactive_attr_group_gov_sys = {
+	.attrs = interactive_attributes_gov_sys,
+	.name = "interactive",
+};
+
+/* Per policy governor instance */
+static struct attribute *interactive_attributes_gov_pol[] = {
+	&target_loads_gov_pol.attr,
+	&above_hispeed_delay_gov_pol.attr,
+	&hispeed_freq_gov_pol.attr,
+	&go_hispeed_load_gov_pol.attr,
+	&min_sample_time_gov_pol.attr,
+	&timer_rate_gov_pol.attr,
+	&timer_slack_gov_pol.attr,
+	&boost_gov_pol.attr,
+	&boostpulse_gov_pol.attr,
+	&boostpulse_duration_gov_pol.attr,
+	&io_is_busy_gov_pol.attr,
+	NULL,
+};
+
+static struct attribute_group interactive_attr_group_gov_pol = {
+	.attrs = interactive_attributes_gov_pol,
+	.name = "interactive",
+};
+
+static struct attribute_group *get_sysfs_attr(void)
+{
+	if (have_governor_per_policy())
+		return &interactive_attr_group_gov_pol;
+	else
+		return &interactive_attr_group_gov_sys;
+}
+
+static int cpufreq_interactive_idle_notifier(struct notifier_block *nb,
+					     unsigned long val,
+					     void *data)
+{
+	if (val == IDLE_END)
+		cpufreq_interactive_idle_end();
+
+	return 0;
+}
+
+static struct notifier_block cpufreq_interactive_idle_nb = {
+	.notifier_call = cpufreq_interactive_idle_notifier,
+};
+
+static int cpufreq_governor_interactive(struct cpufreq_policy *policy,
+		unsigned int event)
+{
+	int rc;
+	unsigned int j;
+	struct cpufreq_interactive_cpuinfo *pcpu;
+	struct cpufreq_frequency_table *freq_table;
+	struct cpufreq_interactive_tunables *tunables;
+	unsigned long flags;
+
+	if (have_governor_per_policy())
+		tunables = policy->governor_data;
+	else
+		tunables = common_tunables;
+
+	if (WARN_ON(!tunables && (event != CPUFREQ_GOV_POLICY_INIT)))
+		return -EINVAL;
+
+	switch (event) {
+	case CPUFREQ_GOV_POLICY_INIT:
+		if (have_governor_per_policy()) {
+			WARN_ON(tunables);
+		} else if (tunables) {
+			tunables->usage_count++;
+			policy->governor_data = tunables;
+			return 0;
+		}
+
+		tunables = kzalloc(sizeof(*tunables), GFP_KERNEL);
+		if (!tunables) {
+			pr_err("%s: POLICY_INIT: kzalloc failed\n", __func__);
+			return -ENOMEM;
+		}
+
+		tunables->usage_count = 1;
+		tunables->above_hispeed_delay = default_above_hispeed_delay;
+		tunables->nabove_hispeed_delay =
+			ARRAY_SIZE(default_above_hispeed_delay);
+		tunables->go_hispeed_load = DEFAULT_GO_HISPEED_LOAD;
+		tunables->target_loads = default_target_loads;
+		tunables->ntarget_loads = ARRAY_SIZE(default_target_loads);
+		tunables->min_sample_time = DEFAULT_MIN_SAMPLE_TIME;
+		tunables->timer_rate = DEFAULT_TIMER_RATE;
+		tunables->boostpulse_duration_val = DEFAULT_MIN_SAMPLE_TIME;
+		tunables->timer_slack_val = DEFAULT_TIMER_SLACK;
+
+		spin_lock_init(&tunables->target_loads_lock);
+		spin_lock_init(&tunables->above_hispeed_delay_lock);
+
+		policy->governor_data = tunables;
+		if (!have_governor_per_policy()) {
+			common_tunables = tunables;
+		}
+
+		rc = sysfs_create_group(get_governor_parent_kobj(policy),
+				get_sysfs_attr());
+		if (rc) {
+			kfree(tunables);
+			policy->governor_data = NULL;
+			if (!have_governor_per_policy()) {
+				common_tunables = NULL;
+			}
+			return rc;
+		}
+
+		if (!policy->governor->initialized) {
+			idle_notifier_register(&cpufreq_interactive_idle_nb);
+			cpufreq_register_notifier(&cpufreq_notifier_block,
+					CPUFREQ_TRANSITION_NOTIFIER);
+		}
+
+		break;
+
+	case CPUFREQ_GOV_POLICY_EXIT:
+		if (!--tunables->usage_count) {
+			if (policy->governor->initialized == 1) {
+				cpufreq_unregister_notifier(&cpufreq_notifier_block,
+						CPUFREQ_TRANSITION_NOTIFIER);
+				idle_notifier_unregister(&cpufreq_interactive_idle_nb);
+			}
+
+			sysfs_remove_group(get_governor_parent_kobj(policy),
+					get_sysfs_attr());
+
+			kfree(tunables);
+			common_tunables = NULL;
+		}
+
+		policy->governor_data = NULL;
+		break;
+
+	case CPUFREQ_GOV_START:
+		mutex_lock(&gov_lock);
+
+		freq_table = cpufreq_frequency_get_table(policy->cpu);
+		if (!tunables->hispeed_freq)
+			tunables->hispeed_freq = policy->max;
+
+		for_each_cpu(j, policy->cpus) {
+			pcpu = &per_cpu(cpuinfo, j);
+			pcpu->policy = policy;
+			pcpu->target_freq = policy->cur;
+			pcpu->freq_table = freq_table;
+			pcpu->floor_freq = pcpu->target_freq;
+			pcpu->pol_floor_val_time =
+				ktime_to_us(ktime_get());
+			pcpu->loc_floor_val_time = pcpu->pol_floor_val_time;
+			pcpu->pol_hispeed_val_time = pcpu->pol_floor_val_time;
+			pcpu->loc_hispeed_val_time = pcpu->pol_floor_val_time;
+			down_write(&pcpu->enable_sem);
+			del_timer_sync(&pcpu->cpu_timer);
+			del_timer_sync(&pcpu->cpu_slack_timer);
+			cpufreq_interactive_timer_start(tunables, j);
+			pcpu->governor_enabled = 1;
+			up_write(&pcpu->enable_sem);
+		}
+
+		mutex_unlock(&gov_lock);
+		break;
+
+	case CPUFREQ_GOV_STOP:
+		mutex_lock(&gov_lock);
+		for_each_cpu(j, policy->cpus) {
+			pcpu = &per_cpu(cpuinfo, j);
+			down_write(&pcpu->enable_sem);
+			pcpu->governor_enabled = 0;
+			del_timer_sync(&pcpu->cpu_timer);
+			del_timer_sync(&pcpu->cpu_slack_timer);
+			up_write(&pcpu->enable_sem);
+		}
+
+		mutex_unlock(&gov_lock);
+		break;
+
+	case CPUFREQ_GOV_LIMITS:
+		if (policy->max < policy->cur)
+			__cpufreq_driver_target(policy,
+					policy->max, CPUFREQ_RELATION_H);
+		else if (policy->min > policy->cur)
+			__cpufreq_driver_target(policy,
+					policy->min, CPUFREQ_RELATION_L);
+		for_each_cpu(j, policy->cpus) {
+			pcpu = &per_cpu(cpuinfo, j);
+
+			down_read(&pcpu->enable_sem);
+			if (pcpu->governor_enabled == 0) {
+				up_read(&pcpu->enable_sem);
+				continue;
+			}
+
+			spin_lock_irqsave(&pcpu->target_freq_lock, flags);
+			if (policy->max < pcpu->target_freq)
+				pcpu->target_freq = policy->max;
+			else if (policy->min > pcpu->target_freq)
+				pcpu->target_freq = policy->min;
+
+			spin_unlock_irqrestore(&pcpu->target_freq_lock, flags);
+			up_read(&pcpu->enable_sem);
+		}
+		break;
+	}
+	return 0;
+}
+
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_INTERACTIVE
+static
+#endif
+struct cpufreq_governor cpufreq_gov_interactive = {
+	.name = "interactive",
+	.governor = cpufreq_governor_interactive,
+	.max_transition_latency = 10000000,
+	.owner = THIS_MODULE,
+};
+
+static void cpufreq_interactive_nop_timer(unsigned long data)
+{
+}
+
+static int __init cpufreq_interactive_init(void)
+{
+	unsigned int i;
+	struct cpufreq_interactive_cpuinfo *pcpu;
+	struct sched_param param = { .sched_priority = MAX_RT_PRIO-1 };
+	int ret = 0;
+
+	/* Initalize per-cpu timers */
+	for_each_possible_cpu(i) {
+		pcpu = &per_cpu(cpuinfo, i);
+		init_timer_deferrable(&pcpu->cpu_timer);
+		pcpu->cpu_timer.function = cpufreq_interactive_timer;
+		pcpu->cpu_timer.data = i;
+		init_timer(&pcpu->cpu_slack_timer);
+		pcpu->cpu_slack_timer.function = cpufreq_interactive_nop_timer;
+		spin_lock_init(&pcpu->load_lock);
+		spin_lock_init(&pcpu->target_freq_lock);
+		init_rwsem(&pcpu->enable_sem);
+	}
+
+	spin_lock_init(&speedchange_cpumask_lock);
+	mutex_init(&gov_lock);
+	speedchange_task =
+		kthread_create(cpufreq_interactive_speedchange_task, NULL,
+			       "cfinteractive");
+	if (IS_ERR(speedchange_task))
+		return PTR_ERR(speedchange_task);
+
+	sched_setscheduler_nocheck(speedchange_task, SCHED_FIFO, &param);
+	get_task_struct(speedchange_task);
+
+	/* NB: wake up so the thread does not look hung to the freezer */
+	wake_up_process(speedchange_task);
+
+	ret = cpufreq_register_governor(&cpufreq_gov_interactive);
+	if (ret) {
+		kthread_stop(speedchange_task);
+		put_task_struct(speedchange_task);
+	}
+	return ret;
+}
+
+#ifdef CONFIG_CPU_FREQ_DEFAULT_GOV_INTERACTIVE
+fs_initcall(cpufreq_interactive_init);
+#else
+module_init(cpufreq_interactive_init);
+#endif
+
+static void __exit cpufreq_interactive_exit(void)
+{
+	cpufreq_unregister_governor(&cpufreq_gov_interactive);
+	kthread_stop(speedchange_task);
+	put_task_struct(speedchange_task);
+}
+
+module_exit(cpufreq_interactive_exit);
+
+MODULE_AUTHOR("Mike Chan <mike@android.com>");
+MODULE_DESCRIPTION("'cpufreq_interactive' - A cpufreq governor for "
+	"Latency sensitive workloads");
+MODULE_LICENSE("GPL");
diff --git a/drivers/cpuidle/cpuidle.c b/drivers/cpuidle/cpuidle.c
index d40b2c077746..151971627757 100644
--- a/drivers/cpuidle/cpuidle.c
+++ b/drivers/cpuidle/cpuidle.c
@@ -192,7 +192,7 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv,
 	}
 
 	/* Take note of the planned idle state. */
-	sched_idle_set_state(target_state);
+	sched_idle_set_state(target_state, index);
 
 	trace_cpu_idle_rcuidle(index, dev->cpu);
 	time_start = ktime_get();
@@ -205,7 +205,7 @@ int cpuidle_enter_state(struct cpuidle_device *dev, struct cpuidle_driver *drv,
 	trace_cpu_idle_rcuidle(PWR_EVENT_EXIT, dev->cpu);
 
 	/* The cpu is no longer idle or about to enter idle. */
-	sched_idle_set_state(NULL);
+	sched_idle_set_state(NULL, -1);
 
 	if (broadcast) {
 		if (WARN_ON_ONCE(!irqs_disabled()))
diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c
index 22e4463d1787..0846b39b4ca6 100644
--- a/drivers/cpuidle/governors/menu.c
+++ b/drivers/cpuidle/governors/menu.c
@@ -178,7 +178,12 @@ static inline int performance_multiplier(unsigned long nr_iowaiters, unsigned lo
 
 	/* for higher loadavg, we are more reluctant */
 
-	mult += 2 * get_loadavg(load);
+	/*
+	 * this doesn't work as intended - it is almost always 0, but can
+	 * sometimes, depending on workload, spike very high into the hundreds
+	 * even when the average cpu load is under 10%.
+	 */
+	/* mult += 2 * get_loadavg(); */
 
 	/* for IO wait tasks (per cpu!) we add 5x each */
 	mult += 10 * nr_iowaiters;
diff --git a/drivers/dma-buf/fence.c b/drivers/dma-buf/fence.c
index 7b05dbe9b296..0ed73ad1b876 100644
--- a/drivers/dma-buf/fence.c
+++ b/drivers/dma-buf/fence.c
@@ -304,8 +304,12 @@ fence_remove_callback(struct fence *fence, struct fence_cb *cb)
 	spin_lock_irqsave(fence->lock, flags);
 
 	ret = !list_empty(&cb->node);
-	if (ret)
+	if (ret) {
 		list_del_init(&cb->node);
+		if (list_empty(&fence->cb_list))
+			if (fence->ops->disable_signaling)
+				fence->ops->disable_signaling(fence);
+	}
 
 	spin_unlock_irqrestore(fence->lock, flags);
 
diff --git a/drivers/firmware/efi/Makefile b/drivers/firmware/efi/Makefile
index ec379a4164cc..f292917b00e7 100644
--- a/drivers/firmware/efi/Makefile
+++ b/drivers/firmware/efi/Makefile
@@ -18,3 +18,6 @@ obj-$(CONFIG_EFI_RUNTIME_MAP)		+= runtime-map.o
 obj-$(CONFIG_EFI_RUNTIME_WRAPPERS)	+= runtime-wrappers.o
 obj-$(CONFIG_EFI_STUB)			+= libstub/
 obj-$(CONFIG_EFI_FAKE_MEMMAP)		+= fake_mem.o
+
+arm-obj-$(CONFIG_EFI)			:= arm-init.o arm-runtime.o
+obj-$(CONFIG_ARM64)			+= $(arm-obj-y)
diff --git a/drivers/firmware/efi/arm-init.c b/drivers/firmware/efi/arm-init.c
new file mode 100644
index 000000000000..a76c35fc0b92
--- /dev/null
+++ b/drivers/firmware/efi/arm-init.c
@@ -0,0 +1,221 @@
+/*
+ * Extensible Firmware Interface
+ *
+ * Based on Extensible Firmware Interface Specification version 2.4
+ *
+ * Copyright (C) 2013 - 2015 Linaro Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/efi.h>
+#include <linux/init.h>
+#include <linux/memblock.h>
+#include <linux/mm_types.h>
+#include <linux/of.h>
+#include <linux/of_fdt.h>
+
+#include <asm/efi.h>
+
+struct efi_memory_map memmap;
+
+u64 efi_system_table;
+
+static int __init is_normal_ram(efi_memory_desc_t *md)
+{
+	if (md->attribute & EFI_MEMORY_WB)
+		return 1;
+	return 0;
+}
+
+/*
+ * Translate a EFI virtual address into a physical address: this is necessary,
+ * as some data members of the EFI system table are virtually remapped after
+ * SetVirtualAddressMap() has been called.
+ */
+static phys_addr_t efi_to_phys(unsigned long addr)
+{
+	efi_memory_desc_t *md;
+
+	for_each_efi_memory_desc(&memmap, md) {
+		if (!(md->attribute & EFI_MEMORY_RUNTIME))
+			continue;
+		if (md->virt_addr == 0)
+			/* no virtual mapping has been installed by the stub */
+			break;
+		if (md->virt_addr <= addr &&
+		    (addr - md->virt_addr) < (md->num_pages << EFI_PAGE_SHIFT))
+			return md->phys_addr + addr - md->virt_addr;
+	}
+	return addr;
+}
+
+static int __init uefi_init(void)
+{
+	efi_char16_t *c16;
+	void *config_tables;
+	size_t table_size;
+	char vendor[100] = "unknown";
+	int i, retval;
+
+	efi.systab = early_memremap(efi_system_table,
+				    sizeof(efi_system_table_t));
+	if (efi.systab == NULL) {
+		pr_warn("Unable to map EFI system table.\n");
+		return -ENOMEM;
+	}
+
+	set_bit(EFI_BOOT, &efi.flags);
+	if (IS_ENABLED(CONFIG_64BIT))
+		set_bit(EFI_64BIT, &efi.flags);
+
+	/*
+	 * Verify the EFI Table
+	 */
+	if (efi.systab->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE) {
+		pr_err("System table signature incorrect\n");
+		retval = -EINVAL;
+		goto out;
+	}
+	if ((efi.systab->hdr.revision >> 16) < 2)
+		pr_warn("Warning: EFI system table version %d.%02d, expected 2.00 or greater\n",
+			efi.systab->hdr.revision >> 16,
+			efi.systab->hdr.revision & 0xffff);
+
+	/* Show what we know for posterity */
+	c16 = early_memremap(efi_to_phys(efi.systab->fw_vendor),
+			     sizeof(vendor) * sizeof(efi_char16_t));
+	if (c16) {
+		for (i = 0; i < (int) sizeof(vendor) - 1 && *c16; ++i)
+			vendor[i] = c16[i];
+		vendor[i] = '\0';
+		early_memunmap(c16, sizeof(vendor) * sizeof(efi_char16_t));
+	}
+
+	pr_info("EFI v%u.%.02u by %s\n",
+		efi.systab->hdr.revision >> 16,
+		efi.systab->hdr.revision & 0xffff, vendor);
+
+	table_size = sizeof(efi_config_table_64_t) * efi.systab->nr_tables;
+	config_tables = early_memremap(efi_to_phys(efi.systab->tables),
+				       table_size);
+	if (config_tables == NULL) {
+		pr_warn("Unable to map EFI config table array.\n");
+		retval = -ENOMEM;
+		goto out;
+	}
+	retval = efi_config_parse_tables(config_tables, efi.systab->nr_tables,
+					 sizeof(efi_config_table_t), NULL);
+
+	early_memunmap(config_tables, table_size);
+out:
+	early_memunmap(efi.systab,  sizeof(efi_system_table_t));
+	return retval;
+}
+
+/*
+ * Return true for RAM regions we want to permanently reserve.
+ */
+static __init int is_reserve_region(efi_memory_desc_t *md)
+{
+	switch (md->type) {
+	case EFI_LOADER_CODE:
+	case EFI_LOADER_DATA:
+	case EFI_BOOT_SERVICES_CODE:
+	case EFI_BOOT_SERVICES_DATA:
+	case EFI_CONVENTIONAL_MEMORY:
+	case EFI_PERSISTENT_MEMORY:
+		return 0;
+	default:
+		break;
+	}
+	return is_normal_ram(md);
+}
+
+static __init void reserve_regions(void)
+{
+	efi_memory_desc_t *md;
+	u64 paddr, npages, size;
+
+	if (efi_enabled(EFI_DBG))
+		pr_info("Processing EFI memory map:\n");
+
+	for_each_efi_memory_desc(&memmap, md) {
+		paddr = md->phys_addr;
+		npages = md->num_pages;
+
+		if (efi_enabled(EFI_DBG)) {
+			char buf[64];
+
+			pr_info("  0x%012llx-0x%012llx %s",
+				paddr, paddr + (npages << EFI_PAGE_SHIFT) - 1,
+				efi_md_typeattr_format(buf, sizeof(buf), md));
+		}
+
+		memrange_efi_to_native(&paddr, &npages);
+		size = npages << PAGE_SHIFT;
+
+		if (is_normal_ram(md))
+			early_init_dt_add_memory_arch(paddr, size);
+
+		if (is_reserve_region(md)) {
+			memblock_mark_nomap(paddr, size);
+			if (efi_enabled(EFI_DBG))
+				pr_cont("*");
+		}
+
+		if (efi_enabled(EFI_DBG))
+			pr_cont("\n");
+	}
+
+	set_bit(EFI_MEMMAP, &efi.flags);
+}
+
+void __init efi_init(void)
+{
+	struct efi_fdt_params params;
+
+	/* Grab UEFI information placed in FDT by stub */
+	if (!efi_get_fdt_params(&params))
+		return;
+
+	efi_system_table = params.system_table;
+
+	memmap.phys_map = params.mmap;
+	memmap.map = early_memremap(params.mmap, params.mmap_size);
+	if (memmap.map == NULL) {
+		/*
+		* If we are booting via UEFI, the UEFI memory map is the only
+		* description of memory we have, so there is little point in
+		* proceeding if we cannot access it.
+		*/
+		panic("Unable to map EFI memory map.\n");
+	}
+	memmap.map_end = memmap.map + params.mmap_size;
+	memmap.desc_size = params.desc_size;
+	memmap.desc_version = params.desc_ver;
+
+	if (uefi_init() < 0)
+		return;
+
+	reserve_regions();
+	early_memunmap(memmap.map, params.mmap_size);
+
+	if (IS_ENABLED(CONFIG_ARM)) {
+		/*
+		 * ARM currently does not allow ioremap_cache() to be called on
+		 * memory regions that are covered by struct page. So remove the
+		 * UEFI memory map from the linear mapping.
+		 */
+		memblock_mark_nomap(params.mmap & PAGE_MASK,
+				    PAGE_ALIGN(params.mmap_size +
+					       (params.mmap & ~PAGE_MASK)));
+	} else {
+		memblock_reserve(params.mmap & PAGE_MASK,
+				 PAGE_ALIGN(params.mmap_size +
+					    (params.mmap & ~PAGE_MASK)));
+	}
+}
diff --git a/drivers/firmware/efi/arm-runtime.c b/drivers/firmware/efi/arm-runtime.c
new file mode 100644
index 000000000000..6ae21e41a429
--- /dev/null
+++ b/drivers/firmware/efi/arm-runtime.c
@@ -0,0 +1,135 @@
+/*
+ * Extensible Firmware Interface
+ *
+ * Based on Extensible Firmware Interface Specification version 2.4
+ *
+ * Copyright (C) 2013, 2014 Linaro Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ */
+
+#include <linux/efi.h>
+#include <linux/io.h>
+#include <linux/memblock.h>
+#include <linux/mm_types.h>
+#include <linux/preempt.h>
+#include <linux/rbtree.h>
+#include <linux/rwsem.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/spinlock.h>
+
+#include <asm/cacheflush.h>
+#include <asm/efi.h>
+#include <asm/mmu.h>
+#include <asm/pgalloc.h>
+#include <asm/pgtable.h>
+
+extern u64 efi_system_table;
+
+static struct mm_struct efi_mm = {
+	.mm_rb			= RB_ROOT,
+	.mm_users		= ATOMIC_INIT(2),
+	.mm_count		= ATOMIC_INIT(1),
+	.mmap_sem		= __RWSEM_INITIALIZER(efi_mm.mmap_sem),
+	.page_table_lock	= __SPIN_LOCK_UNLOCKED(efi_mm.page_table_lock),
+	.mmlist			= LIST_HEAD_INIT(efi_mm.mmlist),
+};
+
+static bool __init efi_virtmap_init(void)
+{
+	efi_memory_desc_t *md;
+
+	efi_mm.pgd = pgd_alloc(&efi_mm);
+	init_new_context(NULL, &efi_mm);
+
+	for_each_efi_memory_desc(&memmap, md) {
+		phys_addr_t phys = md->phys_addr;
+		int ret;
+
+		if (!(md->attribute & EFI_MEMORY_RUNTIME))
+			continue;
+		if (md->virt_addr == 0)
+			return false;
+
+		ret = efi_create_mapping(&efi_mm, md);
+		if  (!ret) {
+			pr_info("  EFI remap %pa => %p\n",
+				&phys, (void *)(unsigned long)md->virt_addr);
+		} else {
+			pr_warn("  EFI remap %pa: failed to create mapping (%d)\n",
+				&phys, ret);
+			return false;
+		}
+	}
+	return true;
+}
+
+/*
+ * Enable the UEFI Runtime Services if all prerequisites are in place, i.e.,
+ * non-early mapping of the UEFI system table and virtual mappings for all
+ * EFI_MEMORY_RUNTIME regions.
+ */
+static int __init arm_enable_runtime_services(void)
+{
+	u64 mapsize;
+
+	if (!efi_enabled(EFI_BOOT)) {
+		pr_info("EFI services will not be available.\n");
+		return 0;
+	}
+
+	if (efi_runtime_disabled()) {
+		pr_info("EFI runtime services will be disabled.\n");
+		return 0;
+	}
+
+	pr_info("Remapping and enabling EFI services.\n");
+
+	mapsize = memmap.map_end - memmap.map;
+	memmap.map = (__force void *)ioremap_cache(memmap.phys_map,
+						   mapsize);
+	if (!memmap.map) {
+		pr_err("Failed to remap EFI memory map\n");
+		return -ENOMEM;
+	}
+	memmap.map_end = memmap.map + mapsize;
+	efi.memmap = &memmap;
+
+	efi.systab = (__force void *)ioremap_cache(efi_system_table,
+						   sizeof(efi_system_table_t));
+	if (!efi.systab) {
+		pr_err("Failed to remap EFI System Table\n");
+		return -ENOMEM;
+	}
+	set_bit(EFI_SYSTEM_TABLES, &efi.flags);
+
+	if (!efi_virtmap_init()) {
+		pr_err("No UEFI virtual mapping was installed -- runtime services will not be available\n");
+		return -ENOMEM;
+	}
+
+	/* Set up runtime services function pointers */
+	efi_native_runtime_setup();
+	set_bit(EFI_RUNTIME_SERVICES, &efi.flags);
+
+	efi.runtime_version = efi.systab->hdr.revision;
+
+	return 0;
+}
+early_initcall(arm_enable_runtime_services);
+
+void efi_virtmap_load(void)
+{
+	preempt_disable();
+	efi_set_pgd(&efi_mm);
+}
+
+void efi_virtmap_unload(void)
+{
+	efi_set_pgd(current->active_mm);
+	preempt_enable();
+}
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index 3b52677f459a..c51f3b2fe3c0 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -25,6 +25,8 @@
 #include <linux/io.h>
 #include <linux/platform_device.h>
 
+#include <asm/early_ioremap.h>
+
 struct efi __read_mostly efi = {
 	.mps			= EFI_INVALID_TABLE_ADDR,
 	.acpi			= EFI_INVALID_TABLE_ADDR,
diff --git a/drivers/firmware/efi/libstub/Makefile b/drivers/firmware/efi/libstub/Makefile
index c4098748e1fe..0e4b85e31b69 100644
--- a/drivers/firmware/efi/libstub/Makefile
+++ b/drivers/firmware/efi/libstub/Makefile
@@ -10,7 +10,7 @@ cflags-$(CONFIG_X86)		+= -m$(BITS) -D__KERNEL__ $(LINUX_INCLUDE) -O2 \
 				   -fPIC -fno-strict-aliasing -mno-red-zone \
 				   -mno-mmx -mno-sse
 
-cflags-$(CONFIG_ARM64)		:= $(subst -pg,,$(KBUILD_CFLAGS))
+cflags-$(CONFIG_ARM64)		:= $(subst -pg,,$(KBUILD_CFLAGS)) -fpie
 cflags-$(CONFIG_ARM)		:= $(subst -pg,,$(KBUILD_CFLAGS)) \
 				   -fno-builtin -fpic -mno-single-pic-base
 
diff --git a/drivers/firmware/efi/libstub/arm-stub.c b/drivers/firmware/efi/libstub/arm-stub.c
index d5aa1d16154f..832df3c58e2f 100644
--- a/drivers/firmware/efi/libstub/arm-stub.c
+++ b/drivers/firmware/efi/libstub/arm-stub.c
@@ -18,8 +18,6 @@
 
 #include "efistub.h"
 
-bool __nokaslr;
-
 static int efi_secureboot_enabled(efi_system_table_t *sys_table_arg)
 {
 	static efi_guid_t const var_guid = EFI_GLOBAL_VARIABLE_GUID;
@@ -221,18 +219,6 @@ unsigned long efi_entry(void *handle, efi_system_table_t *sys_table,
 		goto fail;
 	}
 
-	/* check whether 'nokaslr' was passed on the command line */
-	if (IS_ENABLED(CONFIG_RANDOMIZE_BASE)) {
-		static const u8 default_cmdline[] = CONFIG_CMDLINE;
-		const u8 *str, *cmdline = cmdline_ptr;
-
-		if (IS_ENABLED(CONFIG_CMDLINE_FORCE))
-			cmdline = default_cmdline;
-		str = strstr(cmdline, "nokaslr");
-		if (str == cmdline || (str > cmdline && *(str - 1) == ' '))
-			__nokaslr = true;
-	}
-
 	status = handle_kernel_image(sys_table, image_addr, &image_size,
 				     &reserve_addr,
 				     &reserve_size,
@@ -242,9 +228,13 @@ unsigned long efi_entry(void *handle, efi_system_table_t *sys_table,
 		goto fail_free_cmdline;
 	}
 
-	status = efi_parse_options(cmdline_ptr);
-	if (status != EFI_SUCCESS)
-		pr_efi_err(sys_table, "Failed to parse EFI cmdline options\n");
+	if (IS_ENABLED(CONFIG_CMDLINE_EXTEND) ||
+	    IS_ENABLED(CONFIG_CMDLINE_FORCE) ||
+	    cmdline_size == 0)
+		efi_parse_options(CONFIG_CMDLINE);
+
+	if (!IS_ENABLED(CONFIG_CMDLINE_FORCE) && cmdline_size > 0)
+		efi_parse_options(cmdline_ptr);
 
 	/*
 	 * Unauthenticated device tree data is a security hazard, so
diff --git a/drivers/firmware/efi/libstub/arm64-stub.c b/drivers/firmware/efi/libstub/arm64-stub.c
index 377d935a3380..e33593ed8f84 100644
--- a/drivers/firmware/efi/libstub/arm64-stub.c
+++ b/drivers/firmware/efi/libstub/arm64-stub.c
@@ -9,14 +9,20 @@
  * published by the Free Software Foundation.
  *
  */
+
+/*
+ * To prevent the compiler from emitting GOT-indirected (and thus absolute)
+ * references to the section markers, override their visibility as 'hidden'
+ */
+#pragma GCC visibility push(hidden)
+#include <asm/sections.h>
+#pragma GCC visibility pop
+
 #include <linux/efi.h>
 #include <asm/efi.h>
-#include <asm/sections.h>
 
 #include "efistub.h"
 
-extern bool __nokaslr;
-
 efi_status_t __init handle_kernel_image(efi_system_table_t *sys_table_arg,
 					unsigned long *image_addr,
 					unsigned long *image_size,
@@ -32,7 +38,7 @@ efi_status_t __init handle_kernel_image(efi_system_table_t *sys_table_arg,
 	u64 phys_seed = 0;
 
 	if (IS_ENABLED(CONFIG_RANDOMIZE_BASE)) {
-		if (!__nokaslr) {
+		if (!nokaslr()) {
 			status = efi_get_random_bytes(sys_table_arg,
 						      sizeof(phys_seed),
 						      (u8 *)&phys_seed);
diff --git a/drivers/firmware/efi/libstub/efi-stub-helper.c b/drivers/firmware/efi/libstub/efi-stub-helper.c
index 29ed2f9b218c..58539c1280a5 100644
--- a/drivers/firmware/efi/libstub/efi-stub-helper.c
+++ b/drivers/firmware/efi/libstub/efi-stub-helper.c
@@ -41,6 +41,13 @@ static unsigned long __chunk_size = EFI_READ_CHUNK_SIZE;
 #define EFI_ALLOC_ALIGN		EFI_PAGE_SIZE
 #endif
 
+static int __section(.data) __nokaslr;
+
+int __pure nokaslr(void)
+{
+	return __nokaslr;
+}
+
 struct file_info {
 	efi_file_handle_t *handle;
 	u64 size;
@@ -313,10 +320,14 @@ void efi_free(efi_system_table_t *sys_table_arg, unsigned long size,
  * environments, first in the early boot environment of the EFI boot
  * stub, and subsequently during the kernel boot.
  */
-efi_status_t efi_parse_options(char *cmdline)
+efi_status_t efi_parse_options(char const *cmdline)
 {
 	char *str;
 
+	str = strstr(cmdline, "nokaslr");
+	if (str == cmdline || (str && str > cmdline && *(str - 1) == ' '))
+		__nokaslr = 1;
+
 	/*
 	 * If no EFI parameters were specified on the cmdline we've got
 	 * nothing to do.
diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h
index 5ed3d3f38166..a5eaa3ac0a5d 100644
--- a/drivers/firmware/efi/libstub/efistub.h
+++ b/drivers/firmware/efi/libstub/efistub.h
@@ -5,6 +5,8 @@
 /* error code which can't be mistaken for valid address */
 #define EFI_ERROR	(~0UL)
 
+extern int __pure nokaslr(void);
+
 void efi_char16_printk(efi_system_table_t *, efi_char16_t *);
 
 efi_status_t efi_open_volume(efi_system_table_t *sys_table_arg, void *__image,
diff --git a/drivers/hid/hid-sensor-hub.c b/drivers/hid/hid-sensor-hub.c
index 92870cdb52d9..8efaa88329aa 100644
--- a/drivers/hid/hid-sensor-hub.c
+++ b/drivers/hid/hid-sensor-hub.c
@@ -218,7 +218,8 @@ int sensor_hub_set_feature(struct hid_sensor_hub_device *hsdev, u32 report_id,
 		goto done_proc;
 	}
 
-	remaining_bytes = do_div(buffer_size, sizeof(__s32));
+	remaining_bytes = buffer_size % sizeof(__s32);
+	buffer_size = buffer_size / sizeof(__s32);
 	if (buffer_size) {
 		for (i = 0; i < buffer_size; ++i) {
 			hid_set_field(report->field[field_index], i,
diff --git a/drivers/hid/uhid.c b/drivers/hid/uhid.c
index 1a2032c2c1fb..690a9f0fa042 100644
--- a/drivers/hid/uhid.c
+++ b/drivers/hid/uhid.c
@@ -28,6 +28,8 @@
 #define UHID_NAME	"uhid"
 #define UHID_BUFSIZE	32
 
+static DEFINE_MUTEX(uhid_open_mutex);
+
 struct uhid_device {
 	struct mutex devlock;
 	bool running;
@@ -142,15 +144,26 @@ static void uhid_hid_stop(struct hid_device *hid)
 static int uhid_hid_open(struct hid_device *hid)
 {
 	struct uhid_device *uhid = hid->driver_data;
+	int retval = 0;
 
-	return uhid_queue_event(uhid, UHID_OPEN);
+	mutex_lock(&uhid_open_mutex);
+	if (!hid->open++) {
+		retval = uhid_queue_event(uhid, UHID_OPEN);
+		if (retval)
+			hid->open--;
+	}
+	mutex_unlock(&uhid_open_mutex);
+	return retval;
 }
 
 static void uhid_hid_close(struct hid_device *hid)
 {
 	struct uhid_device *uhid = hid->driver_data;
 
-	uhid_queue_event(uhid, UHID_CLOSE);
+	mutex_lock(&uhid_open_mutex);
+	if (!--hid->open)
+		uhid_queue_event(uhid, UHID_CLOSE);
+	mutex_unlock(&uhid_open_mutex);
 }
 
 static int uhid_hid_parse(struct hid_device *hid)
diff --git a/drivers/input/Kconfig b/drivers/input/Kconfig
index a35532ec00e4..2557dcda7621 100644
--- a/drivers/input/Kconfig
+++ b/drivers/input/Kconfig
@@ -187,6 +187,19 @@ config INPUT_APMPOWER
 	  To compile this driver as a module, choose M here: the
 	  module will be called apm-power.
 
+config INPUT_KEYRESET
+	bool "Reset key"
+	depends on INPUT
+	select INPUT_KEYCOMBO
+	---help---
+	  Say Y here if you want to reboot when some keys are pressed;
+
+config INPUT_KEYCOMBO
+	bool "Key combo"
+	depends on INPUT
+	---help---
+	  Say Y here if you want to take action when some keys are pressed;
+
 comment "Input Device Drivers"
 
 source "drivers/input/keyboard/Kconfig"
diff --git a/drivers/input/Makefile b/drivers/input/Makefile
index 0c9302ca9954..2a6d05ab9170 100644
--- a/drivers/input/Makefile
+++ b/drivers/input/Makefile
@@ -26,3 +26,6 @@ obj-$(CONFIG_INPUT_TOUCHSCREEN)	+= touchscreen/
 obj-$(CONFIG_INPUT_MISC)	+= misc/
 
 obj-$(CONFIG_INPUT_APMPOWER)	+= apm-power.o
+obj-$(CONFIG_INPUT_KEYRESET)	+= keyreset.o
+obj-$(CONFIG_INPUT_KEYCOMBO)	+= keycombo.o
+
diff --git a/drivers/input/keyboard/goldfish_events.c b/drivers/input/keyboard/goldfish_events.c
index 907e4e278fce..c877e56a9bd5 100644
--- a/drivers/input/keyboard/goldfish_events.c
+++ b/drivers/input/keyboard/goldfish_events.c
@@ -17,11 +17,15 @@
 #include <linux/interrupt.h>
 #include <linux/types.h>
 #include <linux/input.h>
+#include <linux/input/mt.h>
 #include <linux/kernel.h>
 #include <linux/platform_device.h>
 #include <linux/slab.h>
 #include <linux/irq.h>
 #include <linux/io.h>
+#include <linux/acpi.h>
+
+#define GOLDFISH_MAX_FINGERS 5
 
 enum {
 	REG_READ        = 0x00,
@@ -51,7 +55,21 @@ static irqreturn_t events_interrupt(int irq, void *dev_id)
 	value = __raw_readl(edev->addr + REG_READ);
 
 	input_event(edev->input, type, code, value);
-	input_sync(edev->input);
+	// Send an extra (EV_SYN, SYN_REPORT, 0x0) event
+	// if a key was pressed. Some keyboard device
+        // drivers may only send the EV_KEY event and
+        // not EV_SYN.
+        // Note that sending an extra SYN_REPORT is not
+        // necessary nor correct protocol with other
+        // devices such as touchscreens, which will send
+        // their own SYN_REPORT's when sufficient event
+        // information has been collected (e.g., for
+        // touchscreens, when pressure and X/Y coordinates
+	// have been received). Hence, we will only send
+	// this extra SYN_REPORT if type == EV_KEY.
+	if (type == EV_KEY) {
+		input_sync(edev->input);
+	}
 	return IRQ_HANDLED;
 }
 
@@ -153,6 +171,15 @@ static int events_probe(struct platform_device *pdev)
 
 	input_dev->name = edev->name;
 	input_dev->id.bustype = BUS_HOST;
+	// Set the Goldfish Device to be multi-touch.
+	// In the Ranchu kernel, there is multi-touch-specific
+	// code for handling ABS_MT_SLOT events.
+	// See drivers/input/input.c:input_handle_abs_event.
+	// If we do not issue input_mt_init_slots,
+        // the kernel will filter out needed ABS_MT_SLOT
+        // events when we touch the screen in more than one place,
+        // preventing multi-touch with more than one finger from working.
+	input_mt_init_slots(input_dev, GOLDFISH_MAX_FINGERS, 0);
 
 	events_import_bits(edev, input_dev->evbit, EV_SYN, EV_MAX);
 	events_import_bits(edev, input_dev->keybit, EV_KEY, KEY_MAX);
@@ -178,10 +205,26 @@ static int events_probe(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id goldfish_events_of_match[] = {
+	{ .compatible = "google,goldfish-events-keypad", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, goldfish_events_of_match);
+
+#ifdef CONFIG_ACPI
+static const struct acpi_device_id goldfish_events_acpi_match[] = {
+	{ "GFSH0002", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, goldfish_events_acpi_match);
+#endif
+
 static struct platform_driver events_driver = {
 	.probe	= events_probe,
 	.driver	= {
 		.name	= "goldfish_events",
+		.of_match_table = goldfish_events_of_match,
+		.acpi_match_table = ACPI_PTR(goldfish_events_acpi_match),
 	},
 };
 
diff --git a/drivers/input/keycombo.c b/drivers/input/keycombo.c
new file mode 100644
index 000000000000..2fba451b91d5
--- /dev/null
+++ b/drivers/input/keycombo.c
@@ -0,0 +1,261 @@
+/* drivers/input/keycombo.c
+ *
+ * Copyright (C) 2014 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/input.h>
+#include <linux/keycombo.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/reboot.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+
+struct keycombo_state {
+	struct input_handler input_handler;
+	unsigned long keybit[BITS_TO_LONGS(KEY_CNT)];
+	unsigned long upbit[BITS_TO_LONGS(KEY_CNT)];
+	unsigned long key[BITS_TO_LONGS(KEY_CNT)];
+	spinlock_t lock;
+	struct  workqueue_struct *wq;
+	int key_down_target;
+	int key_down;
+	int key_up;
+	struct delayed_work key_down_work;
+	int delay;
+	struct work_struct key_up_work;
+	void (*key_up_fn)(void *);
+	void (*key_down_fn)(void *);
+	void *priv;
+	int key_is_down;
+	struct wakeup_source combo_held_wake_source;
+	struct wakeup_source combo_up_wake_source;
+};
+
+static void do_key_down(struct work_struct *work)
+{
+	struct delayed_work *dwork = container_of(work, struct delayed_work,
+									work);
+	struct keycombo_state *state = container_of(dwork,
+					struct keycombo_state, key_down_work);
+	if (state->key_down_fn)
+		state->key_down_fn(state->priv);
+}
+
+static void do_key_up(struct work_struct *work)
+{
+	struct keycombo_state *state = container_of(work, struct keycombo_state,
+								key_up_work);
+	if (state->key_up_fn)
+		state->key_up_fn(state->priv);
+	__pm_relax(&state->combo_up_wake_source);
+}
+
+static void keycombo_event(struct input_handle *handle, unsigned int type,
+		unsigned int code, int value)
+{
+	unsigned long flags;
+	struct keycombo_state *state = handle->private;
+
+	if (type != EV_KEY)
+		return;
+
+	if (code >= KEY_MAX)
+		return;
+
+	if (!test_bit(code, state->keybit))
+		return;
+
+	spin_lock_irqsave(&state->lock, flags);
+	if (!test_bit(code, state->key) == !value)
+		goto done;
+	__change_bit(code, state->key);
+	if (test_bit(code, state->upbit)) {
+		if (value)
+			state->key_up++;
+		else
+			state->key_up--;
+	} else {
+		if (value)
+			state->key_down++;
+		else
+			state->key_down--;
+	}
+	if (state->key_down == state->key_down_target && state->key_up == 0) {
+		__pm_stay_awake(&state->combo_held_wake_source);
+		state->key_is_down = 1;
+		if (queue_delayed_work(state->wq, &state->key_down_work,
+								state->delay))
+			pr_debug("Key down work already queued!");
+	} else if (state->key_is_down) {
+		if (!cancel_delayed_work(&state->key_down_work)) {
+			__pm_stay_awake(&state->combo_up_wake_source);
+			queue_work(state->wq, &state->key_up_work);
+		}
+		__pm_relax(&state->combo_held_wake_source);
+		state->key_is_down = 0;
+	}
+done:
+	spin_unlock_irqrestore(&state->lock, flags);
+}
+
+static int keycombo_connect(struct input_handler *handler,
+		struct input_dev *dev,
+		const struct input_device_id *id)
+{
+	int i;
+	int ret;
+	struct input_handle *handle;
+	struct keycombo_state *state =
+		container_of(handler, struct keycombo_state, input_handler);
+	for (i = 0; i < KEY_MAX; i++) {
+		if (test_bit(i, state->keybit) && test_bit(i, dev->keybit))
+			break;
+	}
+	if (i == KEY_MAX)
+		return -ENODEV;
+
+	handle = kzalloc(sizeof(*handle), GFP_KERNEL);
+	if (!handle)
+		return -ENOMEM;
+
+	handle->dev = dev;
+	handle->handler = handler;
+	handle->name = KEYCOMBO_NAME;
+	handle->private = state;
+
+	ret = input_register_handle(handle);
+	if (ret)
+		goto err_input_register_handle;
+
+	ret = input_open_device(handle);
+	if (ret)
+		goto err_input_open_device;
+
+	return 0;
+
+err_input_open_device:
+	input_unregister_handle(handle);
+err_input_register_handle:
+	kfree(handle);
+	return ret;
+}
+
+static void keycombo_disconnect(struct input_handle *handle)
+{
+	input_close_device(handle);
+	input_unregister_handle(handle);
+	kfree(handle);
+}
+
+static const struct input_device_id keycombo_ids[] = {
+		{
+				.flags = INPUT_DEVICE_ID_MATCH_EVBIT,
+				.evbit = { BIT_MASK(EV_KEY) },
+		},
+		{ },
+};
+MODULE_DEVICE_TABLE(input, keycombo_ids);
+
+static int keycombo_probe(struct platform_device *pdev)
+{
+	int ret;
+	int key, *keyp;
+	struct keycombo_state *state;
+	struct keycombo_platform_data *pdata = pdev->dev.platform_data;
+
+	if (!pdata)
+		return -EINVAL;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return -ENOMEM;
+
+	spin_lock_init(&state->lock);
+	keyp = pdata->keys_down;
+	while ((key = *keyp++)) {
+		if (key >= KEY_MAX)
+			continue;
+		state->key_down_target++;
+		__set_bit(key, state->keybit);
+	}
+	if (pdata->keys_up) {
+		keyp = pdata->keys_up;
+		while ((key = *keyp++)) {
+			if (key >= KEY_MAX)
+				continue;
+			__set_bit(key, state->keybit);
+			__set_bit(key, state->upbit);
+		}
+	}
+
+	state->wq = alloc_ordered_workqueue("keycombo", 0);
+	if (!state->wq)
+		return -ENOMEM;
+
+	state->priv = pdata->priv;
+
+	if (pdata->key_down_fn)
+		state->key_down_fn = pdata->key_down_fn;
+	INIT_DELAYED_WORK(&state->key_down_work, do_key_down);
+
+	if (pdata->key_up_fn)
+		state->key_up_fn = pdata->key_up_fn;
+	INIT_WORK(&state->key_up_work, do_key_up);
+
+	wakeup_source_init(&state->combo_held_wake_source, "key combo");
+	wakeup_source_init(&state->combo_up_wake_source, "key combo up");
+	state->delay = msecs_to_jiffies(pdata->key_down_delay);
+
+	state->input_handler.event = keycombo_event;
+	state->input_handler.connect = keycombo_connect;
+	state->input_handler.disconnect = keycombo_disconnect;
+	state->input_handler.name = KEYCOMBO_NAME;
+	state->input_handler.id_table = keycombo_ids;
+	ret = input_register_handler(&state->input_handler);
+	if (ret) {
+		kfree(state);
+		return ret;
+	}
+	platform_set_drvdata(pdev, state);
+	return 0;
+}
+
+int keycombo_remove(struct platform_device *pdev)
+{
+	struct keycombo_state *state = platform_get_drvdata(pdev);
+	input_unregister_handler(&state->input_handler);
+	destroy_workqueue(state->wq);
+	kfree(state);
+	return 0;
+}
+
+
+struct platform_driver keycombo_driver = {
+		.driver.name = KEYCOMBO_NAME,
+		.probe = keycombo_probe,
+		.remove = keycombo_remove,
+};
+
+static int __init keycombo_init(void)
+{
+	return platform_driver_register(&keycombo_driver);
+}
+
+static void __exit keycombo_exit(void)
+{
+	return platform_driver_unregister(&keycombo_driver);
+}
+
+module_init(keycombo_init);
+module_exit(keycombo_exit);
diff --git a/drivers/input/keyreset.c b/drivers/input/keyreset.c
new file mode 100644
index 000000000000..7e5222aec7c1
--- /dev/null
+++ b/drivers/input/keyreset.c
@@ -0,0 +1,144 @@
+/* drivers/input/keyreset.c
+ *
+ * Copyright (C) 2014 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/input.h>
+#include <linux/keyreset.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/reboot.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/syscalls.h>
+#include <linux/keycombo.h>
+
+struct keyreset_state {
+	int restart_requested;
+	int (*reset_fn)(void);
+	struct platform_device *pdev_child;
+	struct work_struct restart_work;
+};
+
+static void do_restart(struct work_struct *unused)
+{
+	orderly_reboot();
+}
+
+static void do_reset_fn(void *priv)
+{
+	struct keyreset_state *state = priv;
+	if (state->restart_requested)
+		panic("keyboard reset failed, %d", state->restart_requested);
+	if (state->reset_fn) {
+		state->restart_requested = state->reset_fn();
+	} else {
+		pr_info("keyboard reset\n");
+		schedule_work(&state->restart_work);
+		state->restart_requested = 1;
+	}
+}
+
+static int keyreset_probe(struct platform_device *pdev)
+{
+	int ret = -ENOMEM;
+	struct keycombo_platform_data *pdata_child;
+	struct keyreset_platform_data *pdata = pdev->dev.platform_data;
+	int up_size = 0, down_size = 0, size;
+	int key, *keyp;
+	struct keyreset_state *state;
+
+	if (!pdata)
+		return -EINVAL;
+	state = devm_kzalloc(&pdev->dev, sizeof(*state), GFP_KERNEL);
+	if (!state)
+		return -ENOMEM;
+
+	state->pdev_child = platform_device_alloc(KEYCOMBO_NAME,
+							PLATFORM_DEVID_AUTO);
+	if (!state->pdev_child)
+		return -ENOMEM;
+	state->pdev_child->dev.parent = &pdev->dev;
+	INIT_WORK(&state->restart_work, do_restart);
+
+	keyp = pdata->keys_down;
+	while ((key = *keyp++)) {
+		if (key >= KEY_MAX)
+			continue;
+		down_size++;
+	}
+	if (pdata->keys_up) {
+		keyp = pdata->keys_up;
+		while ((key = *keyp++)) {
+			if (key >= KEY_MAX)
+				continue;
+			up_size++;
+		}
+	}
+	size = sizeof(struct keycombo_platform_data)
+			+ sizeof(int) * (down_size + 1);
+	pdata_child = devm_kzalloc(&pdev->dev, size, GFP_KERNEL);
+	if (!pdata_child)
+		goto error;
+	memcpy(pdata_child->keys_down, pdata->keys_down,
+						sizeof(int) * down_size);
+	if (up_size > 0) {
+		pdata_child->keys_up = devm_kzalloc(&pdev->dev, up_size + 1,
+								GFP_KERNEL);
+		if (!pdata_child->keys_up)
+			goto error;
+		memcpy(pdata_child->keys_up, pdata->keys_up,
+							sizeof(int) * up_size);
+		if (!pdata_child->keys_up)
+			goto error;
+	}
+	state->reset_fn = pdata->reset_fn;
+	pdata_child->key_down_fn = do_reset_fn;
+	pdata_child->priv = state;
+	pdata_child->key_down_delay = pdata->key_down_delay;
+	ret = platform_device_add_data(state->pdev_child, pdata_child, size);
+	if (ret)
+		goto error;
+	platform_set_drvdata(pdev, state);
+	return platform_device_add(state->pdev_child);
+error:
+	platform_device_put(state->pdev_child);
+	return ret;
+}
+
+int keyreset_remove(struct platform_device *pdev)
+{
+	struct keyreset_state *state = platform_get_drvdata(pdev);
+	platform_device_put(state->pdev_child);
+	return 0;
+}
+
+
+struct platform_driver keyreset_driver = {
+	.driver.name = KEYRESET_NAME,
+	.probe = keyreset_probe,
+	.remove = keyreset_remove,
+};
+
+static int __init keyreset_init(void)
+{
+	return platform_driver_register(&keyreset_driver);
+}
+
+static void __exit keyreset_exit(void)
+{
+	return platform_driver_unregister(&keyreset_driver);
+}
+
+module_init(keyreset_init);
+module_exit(keyreset_exit);
diff --git a/drivers/input/misc/Kconfig b/drivers/input/misc/Kconfig
index d6d16fa78281..25ac47b9a180 100644
--- a/drivers/input/misc/Kconfig
+++ b/drivers/input/misc/Kconfig
@@ -341,6 +341,17 @@ config INPUT_ATI_REMOTE2
 	  To compile this driver as a module, choose M here: the module will be
 	  called ati_remote2.
 
+config INPUT_KEYCHORD
+	tristate "Key chord input driver support"
+	help
+	  Say Y here if you want to enable the key chord driver
+	  accessible at /dev/keychord.  This driver can be used
+	  for receiving notifications when client specified key
+	  combinations are pressed.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called keychord.
+
 config INPUT_KEYSPAN_REMOTE
 	tristate "Keyspan DMR USB remote control"
 	depends on USB_ARCH_HAS_HCD
@@ -509,6 +520,11 @@ config INPUT_SGI_BTNS
 	  To compile this driver as a module, choose M here: the
 	  module will be called sgi_btns.
 
+config INPUT_GPIO
+	tristate "GPIO driver support"
+	help
+	  Say Y here if you want to support gpio based keys, wheels etc...
+
 config HP_SDC_RTC
 	tristate "HP SDC Real Time Clock"
 	depends on (GSC || HP300) && SERIO
diff --git a/drivers/input/misc/Makefile b/drivers/input/misc/Makefile
index 0357a088c6a9..66c3cc9f181c 100644
--- a/drivers/input/misc/Makefile
+++ b/drivers/input/misc/Makefile
@@ -34,9 +34,11 @@ obj-$(CONFIG_INPUT_DRV2667_HAPTICS)	+= drv2667.o
 obj-$(CONFIG_INPUT_GP2A)		+= gp2ap002a00f.o
 obj-$(CONFIG_INPUT_GPIO_BEEPER)		+= gpio-beeper.o
 obj-$(CONFIG_INPUT_GPIO_TILT_POLLED)	+= gpio_tilt_polled.o
+obj-$(CONFIG_INPUT_GPIO)		+= gpio_event.o gpio_matrix.o gpio_input.o gpio_output.o gpio_axis.o
 obj-$(CONFIG_HP_SDC_RTC)		+= hp_sdc_rtc.o
 obj-$(CONFIG_INPUT_IMS_PCU)		+= ims-pcu.o
 obj-$(CONFIG_INPUT_IXP4XX_BEEPER)	+= ixp4xx-beeper.o
+obj-$(CONFIG_INPUT_KEYCHORD)		+= keychord.o
 obj-$(CONFIG_INPUT_KEYSPAN_REMOTE)	+= keyspan_remote.o
 obj-$(CONFIG_INPUT_KXTJ9)		+= kxtj9.o
 obj-$(CONFIG_INPUT_M68K_BEEP)		+= m68kspkr.o
diff --git a/drivers/input/misc/gpio_axis.c b/drivers/input/misc/gpio_axis.c
new file mode 100644
index 000000000000..0acf4a576f53
--- /dev/null
+++ b/drivers/input/misc/gpio_axis.c
@@ -0,0 +1,192 @@
+/* drivers/input/misc/gpio_axis.c
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/gpio.h>
+#include <linux/gpio_event.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+
+struct gpio_axis_state {
+	struct gpio_event_input_devs *input_devs;
+	struct gpio_event_axis_info *info;
+	uint32_t pos;
+};
+
+uint16_t gpio_axis_4bit_gray_map_table[] = {
+	[0x0] = 0x0, [0x1] = 0x1, /* 0000 0001 */
+	[0x3] = 0x2, [0x2] = 0x3, /* 0011 0010 */
+	[0x6] = 0x4, [0x7] = 0x5, /* 0110 0111 */
+	[0x5] = 0x6, [0x4] = 0x7, /* 0101 0100 */
+	[0xc] = 0x8, [0xd] = 0x9, /* 1100 1101 */
+	[0xf] = 0xa, [0xe] = 0xb, /* 1111 1110 */
+	[0xa] = 0xc, [0xb] = 0xd, /* 1010 1011 */
+	[0x9] = 0xe, [0x8] = 0xf, /* 1001 1000 */
+};
+uint16_t gpio_axis_4bit_gray_map(struct gpio_event_axis_info *info, uint16_t in)
+{
+	return gpio_axis_4bit_gray_map_table[in];
+}
+
+uint16_t gpio_axis_5bit_singletrack_map_table[] = {
+	[0x10] = 0x00, [0x14] = 0x01, [0x1c] = 0x02, /*     10000 10100 11100 */
+	[0x1e] = 0x03, [0x1a] = 0x04, [0x18] = 0x05, /*     11110 11010 11000 */
+	[0x08] = 0x06, [0x0a] = 0x07, [0x0e] = 0x08, /*    01000 01010 01110  */
+	[0x0f] = 0x09, [0x0d] = 0x0a, [0x0c] = 0x0b, /*    01111 01101 01100  */
+	[0x04] = 0x0c, [0x05] = 0x0d, [0x07] = 0x0e, /*   00100 00101 00111   */
+	[0x17] = 0x0f, [0x16] = 0x10, [0x06] = 0x11, /*   10111 10110 00110   */
+	[0x02] = 0x12, [0x12] = 0x13, [0x13] = 0x14, /*  00010 10010 10011    */
+	[0x1b] = 0x15, [0x0b] = 0x16, [0x03] = 0x17, /*  11011 01011 00011    */
+	[0x01] = 0x18, [0x09] = 0x19, [0x19] = 0x1a, /* 00001 01001 11001     */
+	[0x1d] = 0x1b, [0x15] = 0x1c, [0x11] = 0x1d, /* 11101 10101 10001     */
+};
+uint16_t gpio_axis_5bit_singletrack_map(
+	struct gpio_event_axis_info *info, uint16_t in)
+{
+	return gpio_axis_5bit_singletrack_map_table[in];
+}
+
+static void gpio_event_update_axis(struct gpio_axis_state *as, int report)
+{
+	struct gpio_event_axis_info *ai = as->info;
+	int i;
+	int change;
+	uint16_t state = 0;
+	uint16_t pos;
+	uint16_t old_pos = as->pos;
+	for (i = ai->count - 1; i >= 0; i--)
+		state = (state << 1) | gpio_get_value(ai->gpio[i]);
+	pos = ai->map(ai, state);
+	if (ai->flags & GPIOEAF_PRINT_RAW)
+		pr_info("axis %d-%d raw %x, pos %d -> %d\n",
+			ai->type, ai->code, state, old_pos, pos);
+	if (report && pos != old_pos) {
+		if (ai->type == EV_REL) {
+			change = (ai->decoded_size + pos - old_pos) %
+				  ai->decoded_size;
+			if (change > ai->decoded_size / 2)
+				change -= ai->decoded_size;
+			if (change == ai->decoded_size / 2) {
+				if (ai->flags & GPIOEAF_PRINT_EVENT)
+					pr_info("axis %d-%d unknown direction, "
+						"pos %d -> %d\n", ai->type,
+						ai->code, old_pos, pos);
+				change = 0; /* no closest direction */
+			}
+			if (ai->flags & GPIOEAF_PRINT_EVENT)
+				pr_info("axis %d-%d change %d\n",
+					ai->type, ai->code, change);
+			input_report_rel(as->input_devs->dev[ai->dev],
+						ai->code, change);
+		} else {
+			if (ai->flags & GPIOEAF_PRINT_EVENT)
+				pr_info("axis %d-%d now %d\n",
+					ai->type, ai->code, pos);
+			input_event(as->input_devs->dev[ai->dev],
+					ai->type, ai->code, pos);
+		}
+		input_sync(as->input_devs->dev[ai->dev]);
+	}
+	as->pos = pos;
+}
+
+static irqreturn_t gpio_axis_irq_handler(int irq, void *dev_id)
+{
+	struct gpio_axis_state *as = dev_id;
+	gpio_event_update_axis(as, 1);
+	return IRQ_HANDLED;
+}
+
+int gpio_event_axis_func(struct gpio_event_input_devs *input_devs,
+			 struct gpio_event_info *info, void **data, int func)
+{
+	int ret;
+	int i;
+	int irq;
+	struct gpio_event_axis_info *ai;
+	struct gpio_axis_state *as;
+
+	ai = container_of(info, struct gpio_event_axis_info, info);
+	if (func == GPIO_EVENT_FUNC_SUSPEND) {
+		for (i = 0; i < ai->count; i++)
+			disable_irq(gpio_to_irq(ai->gpio[i]));
+		return 0;
+	}
+	if (func == GPIO_EVENT_FUNC_RESUME) {
+		for (i = 0; i < ai->count; i++)
+			enable_irq(gpio_to_irq(ai->gpio[i]));
+		return 0;
+	}
+
+	if (func == GPIO_EVENT_FUNC_INIT) {
+		*data = as = kmalloc(sizeof(*as), GFP_KERNEL);
+		if (as == NULL) {
+			ret = -ENOMEM;
+			goto err_alloc_axis_state_failed;
+		}
+		as->input_devs = input_devs;
+		as->info = ai;
+		if (ai->dev >= input_devs->count) {
+			pr_err("gpio_event_axis: bad device index %d >= %d "
+				"for %d:%d\n", ai->dev, input_devs->count,
+				ai->type, ai->code);
+			ret = -EINVAL;
+			goto err_bad_device_index;
+		}
+
+		input_set_capability(input_devs->dev[ai->dev],
+				     ai->type, ai->code);
+		if (ai->type == EV_ABS) {
+			input_set_abs_params(input_devs->dev[ai->dev], ai->code,
+					     0, ai->decoded_size - 1, 0, 0);
+		}
+		for (i = 0; i < ai->count; i++) {
+			ret = gpio_request(ai->gpio[i], "gpio_event_axis");
+			if (ret < 0)
+				goto err_request_gpio_failed;
+			ret = gpio_direction_input(ai->gpio[i]);
+			if (ret < 0)
+				goto err_gpio_direction_input_failed;
+			ret = irq = gpio_to_irq(ai->gpio[i]);
+			if (ret < 0)
+				goto err_get_irq_num_failed;
+			ret = request_irq(irq, gpio_axis_irq_handler,
+					  IRQF_TRIGGER_RISING |
+					  IRQF_TRIGGER_FALLING,
+					  "gpio_event_axis", as);
+			if (ret < 0)
+				goto err_request_irq_failed;
+		}
+		gpio_event_update_axis(as, 0);
+		return 0;
+	}
+
+	ret = 0;
+	as = *data;
+	for (i = ai->count - 1; i >= 0; i--) {
+		free_irq(gpio_to_irq(ai->gpio[i]), as);
+err_request_irq_failed:
+err_get_irq_num_failed:
+err_gpio_direction_input_failed:
+		gpio_free(ai->gpio[i]);
+err_request_gpio_failed:
+		;
+	}
+err_bad_device_index:
+	kfree(as);
+	*data = NULL;
+err_alloc_axis_state_failed:
+	return ret;
+}
diff --git a/drivers/input/misc/gpio_event.c b/drivers/input/misc/gpio_event.c
new file mode 100644
index 000000000000..90f07eba3ce9
--- /dev/null
+++ b/drivers/input/misc/gpio_event.c
@@ -0,0 +1,228 @@
+/* drivers/input/misc/gpio_event.c
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/input.h>
+#include <linux/gpio_event.h>
+#include <linux/hrtimer.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+struct gpio_event {
+	struct gpio_event_input_devs *input_devs;
+	const struct gpio_event_platform_data *info;
+	void *state[0];
+};
+
+static int gpio_input_event(
+	struct input_dev *dev, unsigned int type, unsigned int code, int value)
+{
+	int i;
+	int devnr;
+	int ret = 0;
+	int tmp_ret;
+	struct gpio_event_info **ii;
+	struct gpio_event *ip = input_get_drvdata(dev);
+
+	for (devnr = 0; devnr < ip->input_devs->count; devnr++)
+		if (ip->input_devs->dev[devnr] == dev)
+			break;
+	if (devnr == ip->input_devs->count) {
+		pr_err("gpio_input_event: unknown device %p\n", dev);
+		return -EIO;
+	}
+
+	for (i = 0, ii = ip->info->info; i < ip->info->info_count; i++, ii++) {
+		if ((*ii)->event) {
+			tmp_ret = (*ii)->event(ip->input_devs, *ii,
+						&ip->state[i],
+						devnr, type, code, value);
+			if (tmp_ret)
+				ret = tmp_ret;
+		}
+	}
+	return ret;
+}
+
+static int gpio_event_call_all_func(struct gpio_event *ip, int func)
+{
+	int i;
+	int ret;
+	struct gpio_event_info **ii;
+
+	if (func == GPIO_EVENT_FUNC_INIT || func == GPIO_EVENT_FUNC_RESUME) {
+		ii = ip->info->info;
+		for (i = 0; i < ip->info->info_count; i++, ii++) {
+			if ((*ii)->func == NULL) {
+				ret = -ENODEV;
+				pr_err("gpio_event_probe: Incomplete pdata, "
+					"no function\n");
+				goto err_no_func;
+			}
+			if (func == GPIO_EVENT_FUNC_RESUME && (*ii)->no_suspend)
+				continue;
+			ret = (*ii)->func(ip->input_devs, *ii, &ip->state[i],
+					  func);
+			if (ret) {
+				pr_err("gpio_event_probe: function failed\n");
+				goto err_func_failed;
+			}
+		}
+		return 0;
+	}
+
+	ret = 0;
+	i = ip->info->info_count;
+	ii = ip->info->info + i;
+	while (i > 0) {
+		i--;
+		ii--;
+		if ((func & ~1) == GPIO_EVENT_FUNC_SUSPEND && (*ii)->no_suspend)
+			continue;
+		(*ii)->func(ip->input_devs, *ii, &ip->state[i], func & ~1);
+err_func_failed:
+err_no_func:
+		;
+	}
+	return ret;
+}
+
+static void __maybe_unused gpio_event_suspend(struct gpio_event *ip)
+{
+	gpio_event_call_all_func(ip, GPIO_EVENT_FUNC_SUSPEND);
+	if (ip->info->power)
+		ip->info->power(ip->info, 0);
+}
+
+static void __maybe_unused gpio_event_resume(struct gpio_event *ip)
+{
+	if (ip->info->power)
+		ip->info->power(ip->info, 1);
+	gpio_event_call_all_func(ip, GPIO_EVENT_FUNC_RESUME);
+}
+
+static int gpio_event_probe(struct platform_device *pdev)
+{
+	int err;
+	struct gpio_event *ip;
+	struct gpio_event_platform_data *event_info;
+	int dev_count = 1;
+	int i;
+	int registered = 0;
+
+	event_info = pdev->dev.platform_data;
+	if (event_info == NULL) {
+		pr_err("gpio_event_probe: No pdata\n");
+		return -ENODEV;
+	}
+	if ((!event_info->name && !event_info->names[0]) ||
+	    !event_info->info || !event_info->info_count) {
+		pr_err("gpio_event_probe: Incomplete pdata\n");
+		return -ENODEV;
+	}
+	if (!event_info->name)
+		while (event_info->names[dev_count])
+			dev_count++;
+	ip = kzalloc(sizeof(*ip) +
+		     sizeof(ip->state[0]) * event_info->info_count +
+		     sizeof(*ip->input_devs) +
+		     sizeof(ip->input_devs->dev[0]) * dev_count, GFP_KERNEL);
+	if (ip == NULL) {
+		err = -ENOMEM;
+		pr_err("gpio_event_probe: Failed to allocate private data\n");
+		goto err_kp_alloc_failed;
+	}
+	ip->input_devs = (void*)&ip->state[event_info->info_count];
+	platform_set_drvdata(pdev, ip);
+
+	for (i = 0; i < dev_count; i++) {
+		struct input_dev *input_dev = input_allocate_device();
+		if (input_dev == NULL) {
+			err = -ENOMEM;
+			pr_err("gpio_event_probe: "
+				"Failed to allocate input device\n");
+			goto err_input_dev_alloc_failed;
+		}
+		input_set_drvdata(input_dev, ip);
+		input_dev->name = event_info->name ?
+					event_info->name : event_info->names[i];
+		input_dev->event = gpio_input_event;
+		ip->input_devs->dev[i] = input_dev;
+	}
+	ip->input_devs->count = dev_count;
+	ip->info = event_info;
+	if (event_info->power)
+		ip->info->power(ip->info, 1);
+
+	err = gpio_event_call_all_func(ip, GPIO_EVENT_FUNC_INIT);
+	if (err)
+		goto err_call_all_func_failed;
+
+	for (i = 0; i < dev_count; i++) {
+		err = input_register_device(ip->input_devs->dev[i]);
+		if (err) {
+			pr_err("gpio_event_probe: Unable to register %s "
+				"input device\n", ip->input_devs->dev[i]->name);
+			goto err_input_register_device_failed;
+		}
+		registered++;
+	}
+
+	return 0;
+
+err_input_register_device_failed:
+	gpio_event_call_all_func(ip, GPIO_EVENT_FUNC_UNINIT);
+err_call_all_func_failed:
+	if (event_info->power)
+		ip->info->power(ip->info, 0);
+	for (i = 0; i < registered; i++)
+		input_unregister_device(ip->input_devs->dev[i]);
+	for (i = dev_count - 1; i >= registered; i--) {
+		input_free_device(ip->input_devs->dev[i]);
+err_input_dev_alloc_failed:
+		;
+	}
+	kfree(ip);
+err_kp_alloc_failed:
+	return err;
+}
+
+static int gpio_event_remove(struct platform_device *pdev)
+{
+	struct gpio_event *ip = platform_get_drvdata(pdev);
+	int i;
+
+	gpio_event_call_all_func(ip, GPIO_EVENT_FUNC_UNINIT);
+	if (ip->info->power)
+		ip->info->power(ip->info, 0);
+	for (i = 0; i < ip->input_devs->count; i++)
+		input_unregister_device(ip->input_devs->dev[i]);
+	kfree(ip);
+	return 0;
+}
+
+static struct platform_driver gpio_event_driver = {
+	.probe		= gpio_event_probe,
+	.remove		= gpio_event_remove,
+	.driver		= {
+		.name	= GPIO_EVENT_DEV_NAME,
+	},
+};
+
+module_platform_driver(gpio_event_driver);
+
+MODULE_DESCRIPTION("GPIO Event Driver");
+MODULE_LICENSE("GPL");
+
diff --git a/drivers/input/misc/gpio_input.c b/drivers/input/misc/gpio_input.c
new file mode 100644
index 000000000000..eefd02725aff
--- /dev/null
+++ b/drivers/input/misc/gpio_input.c
@@ -0,0 +1,390 @@
+/* drivers/input/misc/gpio_input.c
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/gpio.h>
+#include <linux/gpio_event.h>
+#include <linux/hrtimer.h>
+#include <linux/input.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+#include <linux/pm_wakeup.h>
+
+enum {
+	DEBOUNCE_UNSTABLE     = BIT(0),	/* Got irq, while debouncing */
+	DEBOUNCE_PRESSED      = BIT(1),
+	DEBOUNCE_NOTPRESSED   = BIT(2),
+	DEBOUNCE_WAIT_IRQ     = BIT(3),	/* Stable irq state */
+	DEBOUNCE_POLL         = BIT(4),	/* Stable polling state */
+
+	DEBOUNCE_UNKNOWN =
+		DEBOUNCE_PRESSED | DEBOUNCE_NOTPRESSED,
+};
+
+struct gpio_key_state {
+	struct gpio_input_state *ds;
+	uint8_t debounce;
+};
+
+struct gpio_input_state {
+	struct gpio_event_input_devs *input_devs;
+	const struct gpio_event_input_info *info;
+	struct hrtimer timer;
+	int use_irq;
+	int debounce_count;
+	spinlock_t irq_lock;
+	struct wakeup_source *ws;
+	struct gpio_key_state key_state[0];
+};
+
+static enum hrtimer_restart gpio_event_input_timer_func(struct hrtimer *timer)
+{
+	int i;
+	int pressed;
+	struct gpio_input_state *ds =
+		container_of(timer, struct gpio_input_state, timer);
+	unsigned gpio_flags = ds->info->flags;
+	unsigned npolarity;
+	int nkeys = ds->info->keymap_size;
+	const struct gpio_event_direct_entry *key_entry;
+	struct gpio_key_state *key_state;
+	unsigned long irqflags;
+	uint8_t debounce;
+	bool sync_needed;
+
+#if 0
+	key_entry = kp->keys_info->keymap;
+	key_state = kp->key_state;
+	for (i = 0; i < nkeys; i++, key_entry++, key_state++)
+		pr_info("gpio_read_detect_status %d %d\n", key_entry->gpio,
+			gpio_read_detect_status(key_entry->gpio));
+#endif
+	key_entry = ds->info->keymap;
+	key_state = ds->key_state;
+	sync_needed = false;
+	spin_lock_irqsave(&ds->irq_lock, irqflags);
+	for (i = 0; i < nkeys; i++, key_entry++, key_state++) {
+		debounce = key_state->debounce;
+		if (debounce & DEBOUNCE_WAIT_IRQ)
+			continue;
+		if (key_state->debounce & DEBOUNCE_UNSTABLE) {
+			debounce = key_state->debounce = DEBOUNCE_UNKNOWN;
+			enable_irq(gpio_to_irq(key_entry->gpio));
+			if (gpio_flags & GPIOEDF_PRINT_KEY_UNSTABLE)
+				pr_info("gpio_keys_scan_keys: key %x-%x, %d "
+					"(%d) continue debounce\n",
+					ds->info->type, key_entry->code,
+					i, key_entry->gpio);
+		}
+		npolarity = !(gpio_flags & GPIOEDF_ACTIVE_HIGH);
+		pressed = gpio_get_value(key_entry->gpio) ^ npolarity;
+		if (debounce & DEBOUNCE_POLL) {
+			if (pressed == !(debounce & DEBOUNCE_PRESSED)) {
+				ds->debounce_count++;
+				key_state->debounce = DEBOUNCE_UNKNOWN;
+				if (gpio_flags & GPIOEDF_PRINT_KEY_DEBOUNCE)
+					pr_info("gpio_keys_scan_keys: key %x-"
+						"%x, %d (%d) start debounce\n",
+						ds->info->type, key_entry->code,
+						i, key_entry->gpio);
+			}
+			continue;
+		}
+		if (pressed && (debounce & DEBOUNCE_NOTPRESSED)) {
+			if (gpio_flags & GPIOEDF_PRINT_KEY_DEBOUNCE)
+				pr_info("gpio_keys_scan_keys: key %x-%x, %d "
+					"(%d) debounce pressed 1\n",
+					ds->info->type, key_entry->code,
+					i, key_entry->gpio);
+			key_state->debounce = DEBOUNCE_PRESSED;
+			continue;
+		}
+		if (!pressed && (debounce & DEBOUNCE_PRESSED)) {
+			if (gpio_flags & GPIOEDF_PRINT_KEY_DEBOUNCE)
+				pr_info("gpio_keys_scan_keys: key %x-%x, %d "
+					"(%d) debounce pressed 0\n",
+					ds->info->type, key_entry->code,
+					i, key_entry->gpio);
+			key_state->debounce = DEBOUNCE_NOTPRESSED;
+			continue;
+		}
+		/* key is stable */
+		ds->debounce_count--;
+		if (ds->use_irq)
+			key_state->debounce |= DEBOUNCE_WAIT_IRQ;
+		else
+			key_state->debounce |= DEBOUNCE_POLL;
+		if (gpio_flags & GPIOEDF_PRINT_KEYS)
+			pr_info("gpio_keys_scan_keys: key %x-%x, %d (%d) "
+				"changed to %d\n", ds->info->type,
+				key_entry->code, i, key_entry->gpio, pressed);
+		input_event(ds->input_devs->dev[key_entry->dev], ds->info->type,
+			    key_entry->code, pressed);
+		sync_needed = true;
+	}
+	if (sync_needed) {
+		for (i = 0; i < ds->input_devs->count; i++)
+			input_sync(ds->input_devs->dev[i]);
+	}
+
+#if 0
+	key_entry = kp->keys_info->keymap;
+	key_state = kp->key_state;
+	for (i = 0; i < nkeys; i++, key_entry++, key_state++) {
+		pr_info("gpio_read_detect_status %d %d\n", key_entry->gpio,
+			gpio_read_detect_status(key_entry->gpio));
+	}
+#endif
+
+	if (ds->debounce_count)
+		hrtimer_start(timer, ds->info->debounce_time, HRTIMER_MODE_REL);
+	else if (!ds->use_irq)
+		hrtimer_start(timer, ds->info->poll_time, HRTIMER_MODE_REL);
+	else
+		__pm_relax(ds->ws);
+
+	spin_unlock_irqrestore(&ds->irq_lock, irqflags);
+
+	return HRTIMER_NORESTART;
+}
+
+static irqreturn_t gpio_event_input_irq_handler(int irq, void *dev_id)
+{
+	struct gpio_key_state *ks = dev_id;
+	struct gpio_input_state *ds = ks->ds;
+	int keymap_index = ks - ds->key_state;
+	const struct gpio_event_direct_entry *key_entry;
+	unsigned long irqflags;
+	int pressed;
+
+	if (!ds->use_irq)
+		return IRQ_HANDLED;
+
+	key_entry = &ds->info->keymap[keymap_index];
+
+	if (ds->info->debounce_time.tv64) {
+		spin_lock_irqsave(&ds->irq_lock, irqflags);
+		if (ks->debounce & DEBOUNCE_WAIT_IRQ) {
+			ks->debounce = DEBOUNCE_UNKNOWN;
+			if (ds->debounce_count++ == 0) {
+				__pm_stay_awake(ds->ws);
+				hrtimer_start(
+					&ds->timer, ds->info->debounce_time,
+					HRTIMER_MODE_REL);
+			}
+			if (ds->info->flags & GPIOEDF_PRINT_KEY_DEBOUNCE)
+				pr_info("gpio_event_input_irq_handler: "
+					"key %x-%x, %d (%d) start debounce\n",
+					ds->info->type, key_entry->code,
+					keymap_index, key_entry->gpio);
+		} else {
+			disable_irq_nosync(irq);
+			ks->debounce = DEBOUNCE_UNSTABLE;
+		}
+		spin_unlock_irqrestore(&ds->irq_lock, irqflags);
+	} else {
+		pressed = gpio_get_value(key_entry->gpio) ^
+			!(ds->info->flags & GPIOEDF_ACTIVE_HIGH);
+		if (ds->info->flags & GPIOEDF_PRINT_KEYS)
+			pr_info("gpio_event_input_irq_handler: key %x-%x, %d "
+				"(%d) changed to %d\n",
+				ds->info->type, key_entry->code, keymap_index,
+				key_entry->gpio, pressed);
+		input_event(ds->input_devs->dev[key_entry->dev], ds->info->type,
+			    key_entry->code, pressed);
+		input_sync(ds->input_devs->dev[key_entry->dev]);
+	}
+	return IRQ_HANDLED;
+}
+
+static int gpio_event_input_request_irqs(struct gpio_input_state *ds)
+{
+	int i;
+	int err;
+	unsigned int irq;
+	unsigned long req_flags = IRQF_TRIGGER_RISING | IRQF_TRIGGER_FALLING;
+
+	for (i = 0; i < ds->info->keymap_size; i++) {
+		err = irq = gpio_to_irq(ds->info->keymap[i].gpio);
+		if (err < 0)
+			goto err_gpio_get_irq_num_failed;
+		err = request_irq(irq, gpio_event_input_irq_handler,
+				  req_flags, "gpio_keys", &ds->key_state[i]);
+		if (err) {
+			pr_err("gpio_event_input_request_irqs: request_irq "
+				"failed for input %d, irq %d\n",
+				ds->info->keymap[i].gpio, irq);
+			goto err_request_irq_failed;
+		}
+		if (ds->info->info.no_suspend) {
+			err = enable_irq_wake(irq);
+			if (err) {
+				pr_err("gpio_event_input_request_irqs: "
+					"enable_irq_wake failed for input %d, "
+					"irq %d\n",
+					ds->info->keymap[i].gpio, irq);
+				goto err_enable_irq_wake_failed;
+			}
+		}
+	}
+	return 0;
+
+	for (i = ds->info->keymap_size - 1; i >= 0; i--) {
+		irq = gpio_to_irq(ds->info->keymap[i].gpio);
+		if (ds->info->info.no_suspend)
+			disable_irq_wake(irq);
+err_enable_irq_wake_failed:
+		free_irq(irq, &ds->key_state[i]);
+err_request_irq_failed:
+err_gpio_get_irq_num_failed:
+		;
+	}
+	return err;
+}
+
+int gpio_event_input_func(struct gpio_event_input_devs *input_devs,
+			struct gpio_event_info *info, void **data, int func)
+{
+	int ret;
+	int i;
+	unsigned long irqflags;
+	struct gpio_event_input_info *di;
+	struct gpio_input_state *ds = *data;
+	char *wlname;
+
+	di = container_of(info, struct gpio_event_input_info, info);
+
+	if (func == GPIO_EVENT_FUNC_SUSPEND) {
+		if (ds->use_irq)
+			for (i = 0; i < di->keymap_size; i++)
+				disable_irq(gpio_to_irq(di->keymap[i].gpio));
+		hrtimer_cancel(&ds->timer);
+		return 0;
+	}
+	if (func == GPIO_EVENT_FUNC_RESUME) {
+		spin_lock_irqsave(&ds->irq_lock, irqflags);
+		if (ds->use_irq)
+			for (i = 0; i < di->keymap_size; i++)
+				enable_irq(gpio_to_irq(di->keymap[i].gpio));
+		hrtimer_start(&ds->timer, ktime_set(0, 0), HRTIMER_MODE_REL);
+		spin_unlock_irqrestore(&ds->irq_lock, irqflags);
+		return 0;
+	}
+
+	if (func == GPIO_EVENT_FUNC_INIT) {
+		if (ktime_to_ns(di->poll_time) <= 0)
+			di->poll_time = ktime_set(0, 20 * NSEC_PER_MSEC);
+
+		*data = ds = kzalloc(sizeof(*ds) + sizeof(ds->key_state[0]) *
+					di->keymap_size, GFP_KERNEL);
+		if (ds == NULL) {
+			ret = -ENOMEM;
+			pr_err("gpio_event_input_func: "
+				"Failed to allocate private data\n");
+			goto err_ds_alloc_failed;
+		}
+		ds->debounce_count = di->keymap_size;
+		ds->input_devs = input_devs;
+		ds->info = di;
+		wlname = kasprintf(GFP_KERNEL, "gpio_input:%s%s",
+				   input_devs->dev[0]->name,
+				   (input_devs->count > 1) ? "..." : "");
+
+		ds->ws = wakeup_source_register(wlname);
+		kfree(wlname);
+		if (!ds->ws) {
+			ret = -ENOMEM;
+			pr_err("gpio_event_input_func: "
+				"Failed to allocate wakeup source\n");
+			goto err_ws_failed;
+		}
+
+		spin_lock_init(&ds->irq_lock);
+
+		for (i = 0; i < di->keymap_size; i++) {
+			int dev = di->keymap[i].dev;
+			if (dev >= input_devs->count) {
+				pr_err("gpio_event_input_func: bad device "
+					"index %d >= %d for key code %d\n",
+					dev, input_devs->count,
+					di->keymap[i].code);
+				ret = -EINVAL;
+				goto err_bad_keymap;
+			}
+			input_set_capability(input_devs->dev[dev], di->type,
+					     di->keymap[i].code);
+			ds->key_state[i].ds = ds;
+			ds->key_state[i].debounce = DEBOUNCE_UNKNOWN;
+		}
+
+		for (i = 0; i < di->keymap_size; i++) {
+			ret = gpio_request(di->keymap[i].gpio, "gpio_kp_in");
+			if (ret) {
+				pr_err("gpio_event_input_func: gpio_request "
+					"failed for %d\n", di->keymap[i].gpio);
+				goto err_gpio_request_failed;
+			}
+			ret = gpio_direction_input(di->keymap[i].gpio);
+			if (ret) {
+				pr_err("gpio_event_input_func: "
+					"gpio_direction_input failed for %d\n",
+					di->keymap[i].gpio);
+				goto err_gpio_configure_failed;
+			}
+		}
+
+		ret = gpio_event_input_request_irqs(ds);
+
+		spin_lock_irqsave(&ds->irq_lock, irqflags);
+		ds->use_irq = ret == 0;
+
+		pr_info("GPIO Input Driver: Start gpio inputs for %s%s in %s "
+			"mode\n", input_devs->dev[0]->name,
+			(input_devs->count > 1) ? "..." : "",
+			ret == 0 ? "interrupt" : "polling");
+
+		hrtimer_init(&ds->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+		ds->timer.function = gpio_event_input_timer_func;
+		hrtimer_start(&ds->timer, ktime_set(0, 0), HRTIMER_MODE_REL);
+		spin_unlock_irqrestore(&ds->irq_lock, irqflags);
+		return 0;
+	}
+
+	ret = 0;
+	spin_lock_irqsave(&ds->irq_lock, irqflags);
+	hrtimer_cancel(&ds->timer);
+	if (ds->use_irq) {
+		for (i = di->keymap_size - 1; i >= 0; i--) {
+			int irq = gpio_to_irq(di->keymap[i].gpio);
+			if (ds->info->info.no_suspend)
+				disable_irq_wake(irq);
+			free_irq(irq, &ds->key_state[i]);
+		}
+	}
+	spin_unlock_irqrestore(&ds->irq_lock, irqflags);
+
+	for (i = di->keymap_size - 1; i >= 0; i--) {
+err_gpio_configure_failed:
+		gpio_free(di->keymap[i].gpio);
+err_gpio_request_failed:
+		;
+	}
+err_bad_keymap:
+	wakeup_source_unregister(ds->ws);
+err_ws_failed:
+	kfree(ds);
+err_ds_alloc_failed:
+	return ret;
+}
diff --git a/drivers/input/misc/gpio_matrix.c b/drivers/input/misc/gpio_matrix.c
new file mode 100644
index 000000000000..eaa9e89d473a
--- /dev/null
+++ b/drivers/input/misc/gpio_matrix.c
@@ -0,0 +1,441 @@
+/* drivers/input/misc/gpio_matrix.c
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/gpio.h>
+#include <linux/gpio_event.h>
+#include <linux/hrtimer.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+#include <linux/wakelock.h>
+
+struct gpio_kp {
+	struct gpio_event_input_devs *input_devs;
+	struct gpio_event_matrix_info *keypad_info;
+	struct hrtimer timer;
+	struct wake_lock wake_lock;
+	int current_output;
+	unsigned int use_irq:1;
+	unsigned int key_state_changed:1;
+	unsigned int last_key_state_changed:1;
+	unsigned int some_keys_pressed:2;
+	unsigned int disabled_irq:1;
+	unsigned long keys_pressed[0];
+};
+
+static void clear_phantom_key(struct gpio_kp *kp, int out, int in)
+{
+	struct gpio_event_matrix_info *mi = kp->keypad_info;
+	int key_index = out * mi->ninputs + in;
+	unsigned short keyentry = mi->keymap[key_index];
+	unsigned short keycode = keyentry & MATRIX_KEY_MASK;
+	unsigned short dev = keyentry >> MATRIX_CODE_BITS;
+
+	if (!test_bit(keycode, kp->input_devs->dev[dev]->key)) {
+		if (mi->flags & GPIOKPF_PRINT_PHANTOM_KEYS)
+			pr_info("gpiomatrix: phantom key %x, %d-%d (%d-%d) "
+				"cleared\n", keycode, out, in,
+				mi->output_gpios[out], mi->input_gpios[in]);
+		__clear_bit(key_index, kp->keys_pressed);
+	} else {
+		if (mi->flags & GPIOKPF_PRINT_PHANTOM_KEYS)
+			pr_info("gpiomatrix: phantom key %x, %d-%d (%d-%d) "
+				"not cleared\n", keycode, out, in,
+				mi->output_gpios[out], mi->input_gpios[in]);
+	}
+}
+
+static int restore_keys_for_input(struct gpio_kp *kp, int out, int in)
+{
+	int rv = 0;
+	int key_index;
+
+	key_index = out * kp->keypad_info->ninputs + in;
+	while (out < kp->keypad_info->noutputs) {
+		if (test_bit(key_index, kp->keys_pressed)) {
+			rv = 1;
+			clear_phantom_key(kp, out, in);
+		}
+		key_index += kp->keypad_info->ninputs;
+		out++;
+	}
+	return rv;
+}
+
+static void remove_phantom_keys(struct gpio_kp *kp)
+{
+	int out, in, inp;
+	int key_index;
+
+	if (kp->some_keys_pressed < 3)
+		return;
+
+	for (out = 0; out < kp->keypad_info->noutputs; out++) {
+		inp = -1;
+		key_index = out * kp->keypad_info->ninputs;
+		for (in = 0; in < kp->keypad_info->ninputs; in++, key_index++) {
+			if (test_bit(key_index, kp->keys_pressed)) {
+				if (inp == -1) {
+					inp = in;
+					continue;
+				}
+				if (inp >= 0) {
+					if (!restore_keys_for_input(kp, out + 1,
+									inp))
+						break;
+					clear_phantom_key(kp, out, inp);
+					inp = -2;
+				}
+				restore_keys_for_input(kp, out, in);
+			}
+		}
+	}
+}
+
+static void report_key(struct gpio_kp *kp, int key_index, int out, int in)
+{
+	struct gpio_event_matrix_info *mi = kp->keypad_info;
+	int pressed = test_bit(key_index, kp->keys_pressed);
+	unsigned short keyentry = mi->keymap[key_index];
+	unsigned short keycode = keyentry & MATRIX_KEY_MASK;
+	unsigned short dev = keyentry >> MATRIX_CODE_BITS;
+
+	if (pressed != test_bit(keycode, kp->input_devs->dev[dev]->key)) {
+		if (keycode == KEY_RESERVED) {
+			if (mi->flags & GPIOKPF_PRINT_UNMAPPED_KEYS)
+				pr_info("gpiomatrix: unmapped key, %d-%d "
+					"(%d-%d) changed to %d\n",
+					out, in, mi->output_gpios[out],
+					mi->input_gpios[in], pressed);
+		} else {
+			if (mi->flags & GPIOKPF_PRINT_MAPPED_KEYS)
+				pr_info("gpiomatrix: key %x, %d-%d (%d-%d) "
+					"changed to %d\n", keycode,
+					out, in, mi->output_gpios[out],
+					mi->input_gpios[in], pressed);
+			input_report_key(kp->input_devs->dev[dev], keycode, pressed);
+		}
+	}
+}
+
+static void report_sync(struct gpio_kp *kp)
+{
+	int i;
+
+	for (i = 0; i < kp->input_devs->count; i++)
+		input_sync(kp->input_devs->dev[i]);
+}
+
+static enum hrtimer_restart gpio_keypad_timer_func(struct hrtimer *timer)
+{
+	int out, in;
+	int key_index;
+	int gpio;
+	struct gpio_kp *kp = container_of(timer, struct gpio_kp, timer);
+	struct gpio_event_matrix_info *mi = kp->keypad_info;
+	unsigned gpio_keypad_flags = mi->flags;
+	unsigned polarity = !!(gpio_keypad_flags & GPIOKPF_ACTIVE_HIGH);
+
+	out = kp->current_output;
+	if (out == mi->noutputs) {
+		out = 0;
+		kp->last_key_state_changed = kp->key_state_changed;
+		kp->key_state_changed = 0;
+		kp->some_keys_pressed = 0;
+	} else {
+		key_index = out * mi->ninputs;
+		for (in = 0; in < mi->ninputs; in++, key_index++) {
+			gpio = mi->input_gpios[in];
+			if (gpio_get_value(gpio) ^ !polarity) {
+				if (kp->some_keys_pressed < 3)
+					kp->some_keys_pressed++;
+				kp->key_state_changed |= !__test_and_set_bit(
+						key_index, kp->keys_pressed);
+			} else
+				kp->key_state_changed |= __test_and_clear_bit(
+						key_index, kp->keys_pressed);
+		}
+		gpio = mi->output_gpios[out];
+		if (gpio_keypad_flags & GPIOKPF_DRIVE_INACTIVE)
+			gpio_set_value(gpio, !polarity);
+		else
+			gpio_direction_input(gpio);
+		out++;
+	}
+	kp->current_output = out;
+	if (out < mi->noutputs) {
+		gpio = mi->output_gpios[out];
+		if (gpio_keypad_flags & GPIOKPF_DRIVE_INACTIVE)
+			gpio_set_value(gpio, polarity);
+		else
+			gpio_direction_output(gpio, polarity);
+		hrtimer_start(timer, mi->settle_time, HRTIMER_MODE_REL);
+		return HRTIMER_NORESTART;
+	}
+	if (gpio_keypad_flags & GPIOKPF_DEBOUNCE) {
+		if (kp->key_state_changed) {
+			hrtimer_start(&kp->timer, mi->debounce_delay,
+				      HRTIMER_MODE_REL);
+			return HRTIMER_NORESTART;
+		}
+		kp->key_state_changed = kp->last_key_state_changed;
+	}
+	if (kp->key_state_changed) {
+		if (gpio_keypad_flags & GPIOKPF_REMOVE_SOME_PHANTOM_KEYS)
+			remove_phantom_keys(kp);
+		key_index = 0;
+		for (out = 0; out < mi->noutputs; out++)
+			for (in = 0; in < mi->ninputs; in++, key_index++)
+				report_key(kp, key_index, out, in);
+		report_sync(kp);
+	}
+	if (!kp->use_irq || kp->some_keys_pressed) {
+		hrtimer_start(timer, mi->poll_time, HRTIMER_MODE_REL);
+		return HRTIMER_NORESTART;
+	}
+
+	/* No keys are pressed, reenable interrupt */
+	for (out = 0; out < mi->noutputs; out++) {
+		if (gpio_keypad_flags & GPIOKPF_DRIVE_INACTIVE)
+			gpio_set_value(mi->output_gpios[out], polarity);
+		else
+			gpio_direction_output(mi->output_gpios[out], polarity);
+	}
+	for (in = 0; in < mi->ninputs; in++)
+		enable_irq(gpio_to_irq(mi->input_gpios[in]));
+	wake_unlock(&kp->wake_lock);
+	return HRTIMER_NORESTART;
+}
+
+static irqreturn_t gpio_keypad_irq_handler(int irq_in, void *dev_id)
+{
+	int i;
+	struct gpio_kp *kp = dev_id;
+	struct gpio_event_matrix_info *mi = kp->keypad_info;
+	unsigned gpio_keypad_flags = mi->flags;
+
+	if (!kp->use_irq) {
+		/* ignore interrupt while registering the handler */
+		kp->disabled_irq = 1;
+		disable_irq_nosync(irq_in);
+		return IRQ_HANDLED;
+	}
+
+	for (i = 0; i < mi->ninputs; i++)
+		disable_irq_nosync(gpio_to_irq(mi->input_gpios[i]));
+	for (i = 0; i < mi->noutputs; i++) {
+		if (gpio_keypad_flags & GPIOKPF_DRIVE_INACTIVE)
+			gpio_set_value(mi->output_gpios[i],
+				!(gpio_keypad_flags & GPIOKPF_ACTIVE_HIGH));
+		else
+			gpio_direction_input(mi->output_gpios[i]);
+	}
+	wake_lock(&kp->wake_lock);
+	hrtimer_start(&kp->timer, ktime_set(0, 0), HRTIMER_MODE_REL);
+	return IRQ_HANDLED;
+}
+
+static int gpio_keypad_request_irqs(struct gpio_kp *kp)
+{
+	int i;
+	int err;
+	unsigned int irq;
+	unsigned long request_flags;
+	struct gpio_event_matrix_info *mi = kp->keypad_info;
+
+	switch (mi->flags & (GPIOKPF_ACTIVE_HIGH|GPIOKPF_LEVEL_TRIGGERED_IRQ)) {
+	default:
+		request_flags = IRQF_TRIGGER_FALLING;
+		break;
+	case GPIOKPF_ACTIVE_HIGH:
+		request_flags = IRQF_TRIGGER_RISING;
+		break;
+	case GPIOKPF_LEVEL_TRIGGERED_IRQ:
+		request_flags = IRQF_TRIGGER_LOW;
+		break;
+	case GPIOKPF_LEVEL_TRIGGERED_IRQ | GPIOKPF_ACTIVE_HIGH:
+		request_flags = IRQF_TRIGGER_HIGH;
+		break;
+	}
+
+	for (i = 0; i < mi->ninputs; i++) {
+		err = irq = gpio_to_irq(mi->input_gpios[i]);
+		if (err < 0)
+			goto err_gpio_get_irq_num_failed;
+		err = request_irq(irq, gpio_keypad_irq_handler, request_flags,
+				  "gpio_kp", kp);
+		if (err) {
+			pr_err("gpiomatrix: request_irq failed for input %d, "
+				"irq %d\n", mi->input_gpios[i], irq);
+			goto err_request_irq_failed;
+		}
+		err = enable_irq_wake(irq);
+		if (err) {
+			pr_err("gpiomatrix: set_irq_wake failed for input %d, "
+				"irq %d\n", mi->input_gpios[i], irq);
+		}
+		disable_irq(irq);
+		if (kp->disabled_irq) {
+			kp->disabled_irq = 0;
+			enable_irq(irq);
+		}
+	}
+	return 0;
+
+	for (i = mi->noutputs - 1; i >= 0; i--) {
+		free_irq(gpio_to_irq(mi->input_gpios[i]), kp);
+err_request_irq_failed:
+err_gpio_get_irq_num_failed:
+		;
+	}
+	return err;
+}
+
+int gpio_event_matrix_func(struct gpio_event_input_devs *input_devs,
+	struct gpio_event_info *info, void **data, int func)
+{
+	int i;
+	int err;
+	int key_count;
+	struct gpio_kp *kp;
+	struct gpio_event_matrix_info *mi;
+
+	mi = container_of(info, struct gpio_event_matrix_info, info);
+	if (func == GPIO_EVENT_FUNC_SUSPEND || func == GPIO_EVENT_FUNC_RESUME) {
+		/* TODO: disable scanning */
+		return 0;
+	}
+
+	if (func == GPIO_EVENT_FUNC_INIT) {
+		if (mi->keymap == NULL ||
+		   mi->input_gpios == NULL ||
+		   mi->output_gpios == NULL) {
+			err = -ENODEV;
+			pr_err("gpiomatrix: Incomplete pdata\n");
+			goto err_invalid_platform_data;
+		}
+		key_count = mi->ninputs * mi->noutputs;
+
+		*data = kp = kzalloc(sizeof(*kp) + sizeof(kp->keys_pressed[0]) *
+				     BITS_TO_LONGS(key_count), GFP_KERNEL);
+		if (kp == NULL) {
+			err = -ENOMEM;
+			pr_err("gpiomatrix: Failed to allocate private data\n");
+			goto err_kp_alloc_failed;
+		}
+		kp->input_devs = input_devs;
+		kp->keypad_info = mi;
+		for (i = 0; i < key_count; i++) {
+			unsigned short keyentry = mi->keymap[i];
+			unsigned short keycode = keyentry & MATRIX_KEY_MASK;
+			unsigned short dev = keyentry >> MATRIX_CODE_BITS;
+			if (dev >= input_devs->count) {
+				pr_err("gpiomatrix: bad device index %d >= "
+					"%d for key code %d\n",
+					dev, input_devs->count, keycode);
+				err = -EINVAL;
+				goto err_bad_keymap;
+			}
+			if (keycode && keycode <= KEY_MAX)
+				input_set_capability(input_devs->dev[dev],
+							EV_KEY, keycode);
+		}
+
+		for (i = 0; i < mi->noutputs; i++) {
+			err = gpio_request(mi->output_gpios[i], "gpio_kp_out");
+			if (err) {
+				pr_err("gpiomatrix: gpio_request failed for "
+					"output %d\n", mi->output_gpios[i]);
+				goto err_request_output_gpio_failed;
+			}
+			if (gpio_cansleep(mi->output_gpios[i])) {
+				pr_err("gpiomatrix: unsupported output gpio %d,"
+					" can sleep\n", mi->output_gpios[i]);
+				err = -EINVAL;
+				goto err_output_gpio_configure_failed;
+			}
+			if (mi->flags & GPIOKPF_DRIVE_INACTIVE)
+				err = gpio_direction_output(mi->output_gpios[i],
+					!(mi->flags & GPIOKPF_ACTIVE_HIGH));
+			else
+				err = gpio_direction_input(mi->output_gpios[i]);
+			if (err) {
+				pr_err("gpiomatrix: gpio_configure failed for "
+					"output %d\n", mi->output_gpios[i]);
+				goto err_output_gpio_configure_failed;
+			}
+		}
+		for (i = 0; i < mi->ninputs; i++) {
+			err = gpio_request(mi->input_gpios[i], "gpio_kp_in");
+			if (err) {
+				pr_err("gpiomatrix: gpio_request failed for "
+					"input %d\n", mi->input_gpios[i]);
+				goto err_request_input_gpio_failed;
+			}
+			err = gpio_direction_input(mi->input_gpios[i]);
+			if (err) {
+				pr_err("gpiomatrix: gpio_direction_input failed"
+					" for input %d\n", mi->input_gpios[i]);
+				goto err_gpio_direction_input_failed;
+			}
+		}
+		kp->current_output = mi->noutputs;
+		kp->key_state_changed = 1;
+
+		hrtimer_init(&kp->timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
+		kp->timer.function = gpio_keypad_timer_func;
+		wake_lock_init(&kp->wake_lock, WAKE_LOCK_SUSPEND, "gpio_kp");
+		err = gpio_keypad_request_irqs(kp);
+		kp->use_irq = err == 0;
+
+		pr_info("GPIO Matrix Keypad Driver: Start keypad matrix for "
+			"%s%s in %s mode\n", input_devs->dev[0]->name,
+			(input_devs->count > 1) ? "..." : "",
+			kp->use_irq ? "interrupt" : "polling");
+
+		if (kp->use_irq)
+			wake_lock(&kp->wake_lock);
+		hrtimer_start(&kp->timer, ktime_set(0, 0), HRTIMER_MODE_REL);
+
+		return 0;
+	}
+
+	err = 0;
+	kp = *data;
+
+	if (kp->use_irq)
+		for (i = mi->noutputs - 1; i >= 0; i--)
+			free_irq(gpio_to_irq(mi->input_gpios[i]), kp);
+
+	hrtimer_cancel(&kp->timer);
+	wake_lock_destroy(&kp->wake_lock);
+	for (i = mi->noutputs - 1; i >= 0; i--) {
+err_gpio_direction_input_failed:
+		gpio_free(mi->input_gpios[i]);
+err_request_input_gpio_failed:
+		;
+	}
+	for (i = mi->noutputs - 1; i >= 0; i--) {
+err_output_gpio_configure_failed:
+		gpio_free(mi->output_gpios[i]);
+err_request_output_gpio_failed:
+		;
+	}
+err_bad_keymap:
+	kfree(kp);
+err_kp_alloc_failed:
+err_invalid_platform_data:
+	return err;
+}
diff --git a/drivers/input/misc/gpio_output.c b/drivers/input/misc/gpio_output.c
new file mode 100644
index 000000000000..2aac2fad0a17
--- /dev/null
+++ b/drivers/input/misc/gpio_output.c
@@ -0,0 +1,97 @@
+/* drivers/input/misc/gpio_output.c
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/gpio.h>
+#include <linux/gpio_event.h>
+
+int gpio_event_output_event(
+	struct gpio_event_input_devs *input_devs, struct gpio_event_info *info,
+	void **data, unsigned int dev, unsigned int type,
+	unsigned int code, int value)
+{
+	int i;
+	struct gpio_event_output_info *oi;
+	oi = container_of(info, struct gpio_event_output_info, info);
+	if (type != oi->type)
+		return 0;
+	if (!(oi->flags & GPIOEDF_ACTIVE_HIGH))
+		value = !value;
+	for (i = 0; i < oi->keymap_size; i++)
+		if (dev == oi->keymap[i].dev && code == oi->keymap[i].code)
+			gpio_set_value(oi->keymap[i].gpio, value);
+	return 0;
+}
+
+int gpio_event_output_func(
+	struct gpio_event_input_devs *input_devs, struct gpio_event_info *info,
+	void **data, int func)
+{
+	int ret;
+	int i;
+	struct gpio_event_output_info *oi;
+	oi = container_of(info, struct gpio_event_output_info, info);
+
+	if (func == GPIO_EVENT_FUNC_SUSPEND || func == GPIO_EVENT_FUNC_RESUME)
+		return 0;
+
+	if (func == GPIO_EVENT_FUNC_INIT) {
+		int output_level = !(oi->flags & GPIOEDF_ACTIVE_HIGH);
+
+		for (i = 0; i < oi->keymap_size; i++) {
+			int dev = oi->keymap[i].dev;
+			if (dev >= input_devs->count) {
+				pr_err("gpio_event_output_func: bad device "
+					"index %d >= %d for key code %d\n",
+					dev, input_devs->count,
+					oi->keymap[i].code);
+				ret = -EINVAL;
+				goto err_bad_keymap;
+			}
+			input_set_capability(input_devs->dev[dev], oi->type,
+					     oi->keymap[i].code);
+		}
+
+		for (i = 0; i < oi->keymap_size; i++) {
+			ret = gpio_request(oi->keymap[i].gpio,
+					   "gpio_event_output");
+			if (ret) {
+				pr_err("gpio_event_output_func: gpio_request "
+					"failed for %d\n", oi->keymap[i].gpio);
+				goto err_gpio_request_failed;
+			}
+			ret = gpio_direction_output(oi->keymap[i].gpio,
+						    output_level);
+			if (ret) {
+				pr_err("gpio_event_output_func: "
+					"gpio_direction_output failed for %d\n",
+					oi->keymap[i].gpio);
+				goto err_gpio_direction_output_failed;
+			}
+		}
+		return 0;
+	}
+
+	ret = 0;
+	for (i = oi->keymap_size - 1; i >= 0; i--) {
+err_gpio_direction_output_failed:
+		gpio_free(oi->keymap[i].gpio);
+err_gpio_request_failed:
+		;
+	}
+err_bad_keymap:
+	return ret;
+}
+
diff --git a/drivers/input/misc/keychord.c b/drivers/input/misc/keychord.c
new file mode 100644
index 000000000000..fdcc14653b64
--- /dev/null
+++ b/drivers/input/misc/keychord.c
@@ -0,0 +1,467 @@
+/*
+ *  drivers/input/misc/keychord.c
+ *
+ * Copyright (C) 2008 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+*/
+
+#include <linux/poll.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/keychord.h>
+#include <linux/sched.h>
+
+#define KEYCHORD_NAME		"keychord"
+#define BUFFER_SIZE			16
+
+MODULE_AUTHOR("Mike Lockwood <lockwood@android.com>");
+MODULE_DESCRIPTION("Key chord input driver");
+MODULE_SUPPORTED_DEVICE("keychord");
+MODULE_LICENSE("GPL");
+
+#define NEXT_KEYCHORD(kc) ((struct input_keychord *) \
+		((char *)kc + sizeof(struct input_keychord) + \
+		kc->count * sizeof(kc->keycodes[0])))
+
+struct keychord_device {
+	struct input_handler	input_handler;
+	int			registered;
+
+	/* list of keychords to monitor */
+	struct input_keychord	*keychords;
+	int			keychord_count;
+
+	/* bitmask of keys contained in our keychords */
+	unsigned long keybit[BITS_TO_LONGS(KEY_CNT)];
+	/* current state of the keys */
+	unsigned long keystate[BITS_TO_LONGS(KEY_CNT)];
+	/* number of keys that are currently pressed */
+	int key_down;
+
+	/* second input_device_id is needed for null termination */
+	struct input_device_id  device_ids[2];
+
+	spinlock_t		lock;
+	wait_queue_head_t	waitq;
+	unsigned char		head;
+	unsigned char		tail;
+	__u16			buff[BUFFER_SIZE];
+	/* Bit to serialize writes to this device */
+#define KEYCHORD_BUSY			0x01
+	unsigned long		flags;
+	wait_queue_head_t	write_waitq;
+};
+
+static int check_keychord(struct keychord_device *kdev,
+		struct input_keychord *keychord)
+{
+	int i;
+
+	if (keychord->count != kdev->key_down)
+		return 0;
+
+	for (i = 0; i < keychord->count; i++) {
+		if (!test_bit(keychord->keycodes[i], kdev->keystate))
+			return 0;
+	}
+
+	/* we have a match */
+	return 1;
+}
+
+static void keychord_event(struct input_handle *handle, unsigned int type,
+			   unsigned int code, int value)
+{
+	struct keychord_device *kdev = handle->private;
+	struct input_keychord *keychord;
+	unsigned long flags;
+	int i, got_chord = 0;
+
+	if (type != EV_KEY || code >= KEY_MAX)
+		return;
+
+	spin_lock_irqsave(&kdev->lock, flags);
+	/* do nothing if key state did not change */
+	if (!test_bit(code, kdev->keystate) == !value)
+		goto done;
+	__change_bit(code, kdev->keystate);
+	if (value)
+		kdev->key_down++;
+	else
+		kdev->key_down--;
+
+	/* don't notify on key up */
+	if (!value)
+		goto done;
+	/* ignore this event if it is not one of the keys we are monitoring */
+	if (!test_bit(code, kdev->keybit))
+		goto done;
+
+	keychord = kdev->keychords;
+	if (!keychord)
+		goto done;
+
+	/* check to see if the keyboard state matches any keychords */
+	for (i = 0; i < kdev->keychord_count; i++) {
+		if (check_keychord(kdev, keychord)) {
+			kdev->buff[kdev->head] = keychord->id;
+			kdev->head = (kdev->head + 1) % BUFFER_SIZE;
+			got_chord = 1;
+			break;
+		}
+		/* skip to next keychord */
+		keychord = NEXT_KEYCHORD(keychord);
+	}
+
+done:
+	spin_unlock_irqrestore(&kdev->lock, flags);
+
+	if (got_chord) {
+		pr_info("keychord: got keychord id %d. Any tasks: %d\n",
+			keychord->id,
+			!list_empty_careful(&kdev->waitq.task_list));
+		wake_up_interruptible(&kdev->waitq);
+	}
+}
+
+static int keychord_connect(struct input_handler *handler,
+					  struct input_dev *dev,
+					  const struct input_device_id *id)
+{
+	int i, ret;
+	struct input_handle *handle;
+	struct keychord_device *kdev =
+		container_of(handler, struct keychord_device, input_handler);
+
+	/*
+	 * ignore this input device if it does not contain any keycodes
+	 * that we are monitoring
+	 */
+	for (i = 0; i < KEY_MAX; i++) {
+		if (test_bit(i, kdev->keybit) && test_bit(i, dev->keybit))
+			break;
+	}
+	if (i == KEY_MAX)
+		return -ENODEV;
+
+	handle = kzalloc(sizeof(*handle), GFP_KERNEL);
+	if (!handle)
+		return -ENOMEM;
+
+	handle->dev = dev;
+	handle->handler = handler;
+	handle->name = KEYCHORD_NAME;
+	handle->private = kdev;
+
+	ret = input_register_handle(handle);
+	if (ret)
+		goto err_input_register_handle;
+
+	ret = input_open_device(handle);
+	if (ret)
+		goto err_input_open_device;
+
+	pr_info("keychord: using input dev %s for fevent\n", dev->name);
+	return 0;
+
+err_input_open_device:
+	input_unregister_handle(handle);
+err_input_register_handle:
+	kfree(handle);
+	return ret;
+}
+
+static void keychord_disconnect(struct input_handle *handle)
+{
+	input_close_device(handle);
+	input_unregister_handle(handle);
+	kfree(handle);
+}
+
+/*
+ * keychord_read is used to read keychord events from the driver
+ */
+static ssize_t keychord_read(struct file *file, char __user *buffer,
+		size_t count, loff_t *ppos)
+{
+	struct keychord_device *kdev = file->private_data;
+	__u16   id;
+	int retval;
+	unsigned long flags;
+
+	if (count < sizeof(id))
+		return -EINVAL;
+	count = sizeof(id);
+
+	if (kdev->head == kdev->tail && (file->f_flags & O_NONBLOCK))
+		return -EAGAIN;
+
+	retval = wait_event_interruptible(kdev->waitq,
+			kdev->head != kdev->tail);
+	if (retval)
+		return retval;
+
+	spin_lock_irqsave(&kdev->lock, flags);
+	/* pop a keychord ID off the queue */
+	id = kdev->buff[kdev->tail];
+	kdev->tail = (kdev->tail + 1) % BUFFER_SIZE;
+	spin_unlock_irqrestore(&kdev->lock, flags);
+
+	if (copy_to_user(buffer, &id, count))
+		return -EFAULT;
+
+	return count;
+}
+
+/*
+ * serializes writes on a device. can use mutex_lock_interruptible()
+ * for this particular use case as well - a matter of preference.
+ */
+static int
+keychord_write_lock(struct keychord_device *kdev)
+{
+	int ret;
+	unsigned long flags;
+
+	spin_lock_irqsave(&kdev->lock, flags);
+	while (kdev->flags & KEYCHORD_BUSY) {
+		spin_unlock_irqrestore(&kdev->lock, flags);
+		ret = wait_event_interruptible(kdev->write_waitq,
+			       ((kdev->flags & KEYCHORD_BUSY) == 0));
+		if (ret)
+			return ret;
+		spin_lock_irqsave(&kdev->lock, flags);
+	}
+	kdev->flags |= KEYCHORD_BUSY;
+	spin_unlock_irqrestore(&kdev->lock, flags);
+	return 0;
+}
+
+static void
+keychord_write_unlock(struct keychord_device *kdev)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&kdev->lock, flags);
+	kdev->flags &= ~KEYCHORD_BUSY;
+	spin_unlock_irqrestore(&kdev->lock, flags);
+	wake_up_interruptible(&kdev->write_waitq);
+}
+
+/*
+ * keychord_write is used to configure the driver
+ */
+static ssize_t keychord_write(struct file *file, const char __user *buffer,
+		size_t count, loff_t *ppos)
+{
+	struct keychord_device *kdev = file->private_data;
+	struct input_keychord *keychords = 0;
+	struct input_keychord *keychord;
+	int ret, i, key;
+	unsigned long flags;
+	size_t resid = count;
+	size_t key_bytes;
+
+	if (count < sizeof(struct input_keychord))
+		return -EINVAL;
+	keychords = kzalloc(count, GFP_KERNEL);
+	if (!keychords)
+		return -ENOMEM;
+
+	/* read list of keychords from userspace */
+	if (copy_from_user(keychords, buffer, count)) {
+		kfree(keychords);
+		return -EFAULT;
+	}
+
+	/*
+	 * Serialize writes to this device to prevent various races.
+	 * 1) writers racing here could do duplicate input_unregister_handler()
+	 *    calls, resulting in attempting to unlink a node from a list that
+	 *    does not exist.
+	 * 2) writers racing here could do duplicate input_register_handler() calls
+	 *    below, resulting in a duplicate insertion of a node into the list.
+	 * 3) a double kfree of keychords can occur (in the event that
+	 *    input_register_handler() fails below.
+	 */
+	ret = keychord_write_lock(kdev);
+	if (ret) {
+		kfree(keychords);
+		return ret;
+	}
+
+	/* unregister handler before changing configuration */
+	if (kdev->registered) {
+		input_unregister_handler(&kdev->input_handler);
+		kdev->registered = 0;
+	}
+
+	spin_lock_irqsave(&kdev->lock, flags);
+	/* clear any existing configuration */
+	kfree(kdev->keychords);
+	kdev->keychords = 0;
+	kdev->keychord_count = 0;
+	kdev->key_down = 0;
+	memset(kdev->keybit, 0, sizeof(kdev->keybit));
+	memset(kdev->keystate, 0, sizeof(kdev->keystate));
+	kdev->head = kdev->tail = 0;
+
+	keychord = keychords;
+
+	while (resid > 0) {
+		/* Is the entire keychord entry header present ? */
+		if (resid < sizeof(struct input_keychord)) {
+			pr_err("keychord: Insufficient bytes present for header %zu\n",
+			       resid);
+			goto err_unlock_return;
+		}
+		resid -= sizeof(struct input_keychord);
+		if (keychord->count <= 0) {
+			pr_err("keychord: invalid keycode count %d\n",
+				keychord->count);
+			goto err_unlock_return;
+		}
+		key_bytes = keychord->count * sizeof(keychord->keycodes[0]);
+		/* Do we have all the expected keycodes ? */
+		if (resid < key_bytes) {
+			pr_err("keychord: Insufficient bytes present for keycount %zu\n",
+			       resid);
+			goto err_unlock_return;
+		}
+		resid -= key_bytes;
+
+		if (keychord->version != KEYCHORD_VERSION) {
+			pr_err("keychord: unsupported version %d\n",
+				keychord->version);
+			goto err_unlock_return;
+		}
+
+		/* keep track of the keys we are monitoring in keybit */
+		for (i = 0; i < keychord->count; i++) {
+			key = keychord->keycodes[i];
+			if (key < 0 || key >= KEY_CNT) {
+				pr_err("keychord: keycode %d out of range\n",
+					key);
+				goto err_unlock_return;
+			}
+			__set_bit(key, kdev->keybit);
+		}
+
+		kdev->keychord_count++;
+		keychord = NEXT_KEYCHORD(keychord);
+	}
+
+	kdev->keychords = keychords;
+	spin_unlock_irqrestore(&kdev->lock, flags);
+
+	ret = input_register_handler(&kdev->input_handler);
+	if (ret) {
+		kfree(keychords);
+		kdev->keychords = 0;
+		keychord_write_unlock(kdev);
+		return ret;
+	}
+	kdev->registered = 1;
+
+	keychord_write_unlock(kdev);
+
+	return count;
+
+err_unlock_return:
+	spin_unlock_irqrestore(&kdev->lock, flags);
+	kfree(keychords);
+	keychord_write_unlock(kdev);
+	return -EINVAL;
+}
+
+static unsigned int keychord_poll(struct file *file, poll_table *wait)
+{
+	struct keychord_device *kdev = file->private_data;
+
+	poll_wait(file, &kdev->waitq, wait);
+
+	if (kdev->head != kdev->tail)
+		return POLLIN | POLLRDNORM;
+
+	return 0;
+}
+
+static int keychord_open(struct inode *inode, struct file *file)
+{
+	struct keychord_device *kdev;
+
+	kdev = kzalloc(sizeof(struct keychord_device), GFP_KERNEL);
+	if (!kdev)
+		return -ENOMEM;
+
+	spin_lock_init(&kdev->lock);
+	init_waitqueue_head(&kdev->waitq);
+	init_waitqueue_head(&kdev->write_waitq);
+
+	kdev->input_handler.event = keychord_event;
+	kdev->input_handler.connect = keychord_connect;
+	kdev->input_handler.disconnect = keychord_disconnect;
+	kdev->input_handler.name = KEYCHORD_NAME;
+	kdev->input_handler.id_table = kdev->device_ids;
+
+	kdev->device_ids[0].flags = INPUT_DEVICE_ID_MATCH_EVBIT;
+	__set_bit(EV_KEY, kdev->device_ids[0].evbit);
+
+	file->private_data = kdev;
+
+	return 0;
+}
+
+static int keychord_release(struct inode *inode, struct file *file)
+{
+	struct keychord_device *kdev = file->private_data;
+
+	if (kdev->registered)
+		input_unregister_handler(&kdev->input_handler);
+	kfree(kdev->keychords);
+	kfree(kdev);
+
+	return 0;
+}
+
+static const struct file_operations keychord_fops = {
+	.owner		= THIS_MODULE,
+	.open		= keychord_open,
+	.release	= keychord_release,
+	.read		= keychord_read,
+	.write		= keychord_write,
+	.poll		= keychord_poll,
+};
+
+static struct miscdevice keychord_misc = {
+	.fops		= &keychord_fops,
+	.name		= KEYCHORD_NAME,
+	.minor		= MISC_DYNAMIC_MINOR,
+};
+
+static int __init keychord_init(void)
+{
+	return misc_register(&keychord_misc);
+}
+
+static void __exit keychord_exit(void)
+{
+	misc_deregister(&keychord_misc);
+}
+
+module_init(keychord_init);
+module_exit(keychord_exit);
diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig
index 108249f52fa9..5649d1ab0083 100644
--- a/drivers/md/Kconfig
+++ b/drivers/md/Kconfig
@@ -459,6 +459,33 @@ config DM_VERITY
 
 	  If unsure, say N.
 
+config DM_VERITY_HASH_PREFETCH_MIN_SIZE_128
+	bool "Prefetch size 128"
+
+config DM_VERITY_HASH_PREFETCH_MIN_SIZE
+	int "Verity hash prefetch minimum size"
+	depends on DM_VERITY
+	range 1 4096
+	default 128 if DM_VERITY_HASH_PREFETCH_MIN_SIZE_128
+	default 1
+	---help---
+	  This sets minimum number of hash blocks to prefetch for dm-verity.
+	  For devices like eMMC, having larger prefetch size like 128 can improve
+	  performance with increased memory consumption for keeping more hashes
+	  in RAM.
+
+config DM_VERITY_FEC
+	bool "Verity forward error correction support"
+	depends on DM_VERITY
+	select REED_SOLOMON
+	select REED_SOLOMON_DEC8
+	---help---
+	  Add forward error correction support to dm-verity. This option
+	  makes it possible to use pre-generated error correction data to
+	  recover from corrupted blocks.
+
+	  If unsure, say N.
+
 config DM_SWITCH
 	tristate "Switch target support (EXPERIMENTAL)"
 	depends on BLK_DEV_DM
@@ -489,4 +516,33 @@ config DM_LOG_WRITES
 
 	  If unsure, say N.
 
+config DM_VERITY_AVB
+	tristate "Support AVB specific verity error behavior"
+	depends on DM_VERITY
+	---help---
+	  Enables Android Verified Boot platform-specific error
+	  behavior. In particular, it will modify the vbmeta partition
+	  specified on the kernel command-line when non-transient error
+	  occurs (followed by a panic).
+
+	  If unsure, say N.
+
+config DM_ANDROID_VERITY
+	bool "Android verity target support"
+	depends on DM_VERITY=y
+	depends on X509_CERTIFICATE_PARSER
+	depends on SYSTEM_TRUSTED_KEYRING
+	depends on PUBLIC_KEY_ALGO_RSA
+	depends on KEYS
+	depends on ASYMMETRIC_KEY_TYPE
+	depends on ASYMMETRIC_PUBLIC_KEY_SUBTYPE
+	depends on MD_LINEAR=y
+	select DM_VERITY_HASH_PREFETCH_MIN_SIZE_128
+	---help---
+	  This device-mapper target is virtually a VERITY target. This
+	  target is setup by reading the metadata contents piggybacked
+	  to the actual data blocks in the block device. The signature
+	  of the metadata contents are verified against the key included
+	  in the system keyring. Upon success, the underlying verity
+	  target is setup.
 endif # MD
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index f34979cd141a..c22cc74c9fa8 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -16,6 +16,7 @@ dm-cache-mq-y   += dm-cache-policy-mq.o
 dm-cache-smq-y   += dm-cache-policy-smq.o
 dm-cache-cleaner-y += dm-cache-policy-cleaner.o
 dm-era-y	+= dm-era-target.o
+dm-verity-y	+= dm-verity-target.o
 md-mod-y	+= md.o bitmap.o
 raid456-y	+= raid5.o raid5-cache.o
 
@@ -59,7 +60,16 @@ obj-$(CONFIG_DM_CACHE_SMQ)	+= dm-cache-smq.o
 obj-$(CONFIG_DM_CACHE_CLEANER)	+= dm-cache-cleaner.o
 obj-$(CONFIG_DM_ERA)		+= dm-era.o
 obj-$(CONFIG_DM_LOG_WRITES)	+= dm-log-writes.o
+obj-$(CONFIG_DM_ANDROID_VERITY) += dm-android-verity.o
 
 ifeq ($(CONFIG_DM_UEVENT),y)
 dm-mod-objs			+= dm-uevent.o
 endif
+
+ifeq ($(CONFIG_DM_VERITY_FEC),y)
+dm-verity-objs			+= dm-verity-fec.o
+endif
+
+ifeq ($(CONFIG_DM_VERITY_AVB),y)
+dm-verity-objs			+= dm-verity-avb.o
+endif
diff --git a/drivers/md/dm-android-verity.c b/drivers/md/dm-android-verity.c
new file mode 100644
index 000000000000..c521df010ee3
--- /dev/null
+++ b/drivers/md/dm-android-verity.c
@@ -0,0 +1,946 @@
+/*
+ * Copyright (C) 2015 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/buffer_head.h>
+#include <linux/debugfs.h>
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/device-mapper.h>
+#include <linux/errno.h>
+#include <linux/fs.h>
+#include <linux/fcntl.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/key.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/of.h>
+#include <linux/reboot.h>
+#include <linux/string.h>
+#include <linux/vmalloc.h>
+
+#include <asm/setup.h>
+#include <crypto/hash.h>
+#include <crypto/public_key.h>
+#include <crypto/sha.h>
+#include <keys/asymmetric-type.h>
+#include <keys/system_keyring.h>
+
+#include "dm-verity.h"
+#include "dm-android-verity.h"
+
+static char verifiedbootstate[VERITY_COMMANDLINE_PARAM_LENGTH];
+static char veritymode[VERITY_COMMANDLINE_PARAM_LENGTH];
+static char veritykeyid[VERITY_DEFAULT_KEY_ID_LENGTH];
+static char buildvariant[BUILD_VARIANT];
+
+static bool target_added;
+static bool verity_enabled = true;
+struct dentry *debug_dir;
+static int android_verity_ctr(struct dm_target *ti, unsigned argc, char **argv);
+
+static struct target_type android_verity_target = {
+	.name                   = "android-verity",
+	.version                = {1, 0, 0},
+	.module                 = THIS_MODULE,
+	.ctr                    = android_verity_ctr,
+	.dtr                    = verity_dtr,
+	.map                    = verity_map,
+	.status                 = verity_status,
+	.prepare_ioctl          = verity_prepare_ioctl,
+	.iterate_devices        = verity_iterate_devices,
+	.io_hints               = verity_io_hints,
+};
+
+static int __init verified_boot_state_param(char *line)
+{
+	strlcpy(verifiedbootstate, line, sizeof(verifiedbootstate));
+	return 1;
+}
+
+__setup("androidboot.verifiedbootstate=", verified_boot_state_param);
+
+static int __init verity_mode_param(char *line)
+{
+	strlcpy(veritymode, line, sizeof(veritymode));
+	return 1;
+}
+
+__setup("androidboot.veritymode=", verity_mode_param);
+
+static int __init verity_keyid_param(char *line)
+{
+	strlcpy(veritykeyid, line, sizeof(veritykeyid));
+	return 1;
+}
+
+__setup("veritykeyid=", verity_keyid_param);
+
+static int __init verity_buildvariant(char *line)
+{
+	strlcpy(buildvariant, line, sizeof(buildvariant));
+	return 1;
+}
+
+__setup("buildvariant=", verity_buildvariant);
+
+static inline bool default_verity_key_id(void)
+{
+	return veritykeyid[0] != '\0';
+}
+
+static inline bool is_eng(void)
+{
+	static const char typeeng[]  = "eng";
+
+	return !strncmp(buildvariant, typeeng, sizeof(typeeng));
+}
+
+static inline bool is_userdebug(void)
+{
+	static const char typeuserdebug[]  = "userdebug";
+
+	return !strncmp(buildvariant, typeuserdebug, sizeof(typeuserdebug));
+}
+
+static inline bool is_unlocked(void)
+{
+	static const char unlocked[] = "orange";
+
+	return !strncmp(verifiedbootstate, unlocked, sizeof(unlocked));
+}
+
+static int table_extract_mpi_array(struct public_key_signature *pks,
+				const void *data, size_t len)
+{
+	MPI mpi = mpi_read_raw_data(data, len);
+
+	if (!mpi) {
+		DMERR("Error while allocating mpi array");
+		return -ENOMEM;
+	}
+
+	pks->mpi[0] = mpi;
+	pks->nr_mpi = 1;
+	return 0;
+}
+
+static struct public_key_signature *table_make_digest(
+						enum hash_algo hash,
+						const void *table,
+						unsigned long table_len)
+{
+	struct public_key_signature *pks = NULL;
+	struct crypto_shash *tfm;
+	struct shash_desc *desc;
+	size_t digest_size, desc_size;
+	int ret;
+
+	/* Allocate the hashing algorithm we're going to need and find out how
+	 * big the hash operational data will be.
+	 */
+	tfm = crypto_alloc_shash(hash_algo_name[hash], 0, 0);
+	if (IS_ERR(tfm))
+		return ERR_CAST(tfm);
+
+	desc_size = crypto_shash_descsize(tfm) + sizeof(*desc);
+	digest_size = crypto_shash_digestsize(tfm);
+
+	/* We allocate the hash operational data storage on the end of out
+	 * context data and the digest output buffer on the end of that.
+	 */
+	ret = -ENOMEM;
+	pks = kzalloc(digest_size + sizeof(*pks) + desc_size, GFP_KERNEL);
+	if (!pks)
+		goto error;
+
+	pks->pkey_hash_algo = hash;
+	pks->digest = (u8 *)pks + sizeof(*pks) + desc_size;
+	pks->digest_size = digest_size;
+
+	desc = (struct shash_desc *)(pks + 1);
+	desc->tfm = tfm;
+	desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	ret = crypto_shash_init(desc);
+	if (ret < 0)
+		goto error;
+
+	ret = crypto_shash_finup(desc, table, table_len, pks->digest);
+	if (ret < 0)
+		goto error;
+
+	crypto_free_shash(tfm);
+	return pks;
+
+error:
+	kfree(pks);
+	crypto_free_shash(tfm);
+	return ERR_PTR(ret);
+}
+
+static int read_block_dev(struct bio_read *payload, struct block_device *bdev,
+		sector_t offset, int length)
+{
+	struct bio *bio;
+	int err = 0, i;
+
+	payload->number_of_pages = DIV_ROUND_UP(length, PAGE_SIZE);
+
+	bio = bio_alloc(GFP_KERNEL, payload->number_of_pages);
+	if (!bio) {
+		DMERR("Error while allocating bio");
+		return -ENOMEM;
+	}
+
+	bio->bi_bdev = bdev;
+	bio->bi_iter.bi_sector = offset;
+
+	payload->page_io = kzalloc(sizeof(struct page *) *
+		payload->number_of_pages, GFP_KERNEL);
+	if (!payload->page_io) {
+		DMERR("page_io array alloc failed");
+		err = -ENOMEM;
+		goto free_bio;
+	}
+
+	for (i = 0; i < payload->number_of_pages; i++) {
+		payload->page_io[i] = alloc_page(GFP_KERNEL);
+		if (!payload->page_io[i]) {
+			DMERR("alloc_page failed");
+			err = -ENOMEM;
+			goto free_pages;
+		}
+		if (!bio_add_page(bio, payload->page_io[i], PAGE_SIZE, 0)) {
+			DMERR("bio_add_page error");
+			err = -EIO;
+			goto free_pages;
+		}
+	}
+
+	if (!submit_bio_wait(READ, bio))
+		/* success */
+		goto free_bio;
+	DMERR("bio read failed");
+	err = -EIO;
+
+free_pages:
+	for (i = 0; i < payload->number_of_pages; i++)
+		if (payload->page_io[i])
+			__free_page(payload->page_io[i]);
+	kfree(payload->page_io);
+free_bio:
+	bio_put(bio);
+	return err;
+}
+
+static inline u64 fec_div_round_up(u64 x, u64 y)
+{
+	u64 remainder;
+
+	return div64_u64_rem(x, y, &remainder) +
+		(remainder > 0 ? 1 : 0);
+}
+
+static inline void populate_fec_metadata(struct fec_header *header,
+				struct fec_ecc_metadata *ecc)
+{
+	ecc->blocks = fec_div_round_up(le64_to_cpu(header->inp_size),
+			FEC_BLOCK_SIZE);
+	ecc->roots = le32_to_cpu(header->roots);
+	ecc->start = le64_to_cpu(header->inp_size);
+}
+
+static inline int validate_fec_header(struct fec_header *header, u64 offset)
+{
+	/* move offset to make the sanity check work for backup header
+	 * as well. */
+	offset -= offset % FEC_BLOCK_SIZE;
+	if (le32_to_cpu(header->magic) != FEC_MAGIC ||
+		le32_to_cpu(header->version) != FEC_VERSION ||
+		le32_to_cpu(header->size) != sizeof(struct fec_header) ||
+		le32_to_cpu(header->roots) == 0 ||
+		le32_to_cpu(header->roots) >= FEC_RSM)
+		return -EINVAL;
+
+	return 0;
+}
+
+static int extract_fec_header(dev_t dev, struct fec_header *fec,
+				struct fec_ecc_metadata *ecc)
+{
+	u64 device_size;
+	struct bio_read payload;
+	int i, err = 0;
+	struct block_device *bdev;
+
+	bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL);
+
+	if (IS_ERR_OR_NULL(bdev)) {
+		DMERR("bdev get error");
+		return PTR_ERR(bdev);
+	}
+
+	device_size = i_size_read(bdev->bd_inode);
+
+	/* fec metadata size is a power of 2 and PAGE_SIZE
+	 * is a power of 2 as well.
+	 */
+	BUG_ON(FEC_BLOCK_SIZE > PAGE_SIZE);
+	/* 512 byte sector alignment */
+	BUG_ON(((device_size - FEC_BLOCK_SIZE) % (1 << SECTOR_SHIFT)) != 0);
+
+	err = read_block_dev(&payload, bdev, (device_size -
+		FEC_BLOCK_SIZE) / (1 << SECTOR_SHIFT), FEC_BLOCK_SIZE);
+	if (err) {
+		DMERR("Error while reading verity metadata");
+		goto error;
+	}
+
+	BUG_ON(sizeof(struct fec_header) > PAGE_SIZE);
+	memcpy(fec, page_address(payload.page_io[0]),
+			sizeof(*fec));
+
+	ecc->valid = true;
+	if (validate_fec_header(fec, device_size - FEC_BLOCK_SIZE)) {
+		/* Try the backup header */
+		memcpy(fec, page_address(payload.page_io[0]) + FEC_BLOCK_SIZE
+			- sizeof(*fec) ,
+			sizeof(*fec));
+		if (validate_fec_header(fec, device_size -
+			sizeof(struct fec_header)))
+			ecc->valid = false;
+	}
+
+	if (ecc->valid)
+		populate_fec_metadata(fec, ecc);
+
+	for (i = 0; i < payload.number_of_pages; i++)
+		__free_page(payload.page_io[i]);
+	kfree(payload.page_io);
+
+error:
+	blkdev_put(bdev, FMODE_READ);
+	return err;
+}
+static void find_metadata_offset(struct fec_header *fec,
+		struct block_device *bdev, u64 *metadata_offset)
+{
+	u64 device_size;
+
+	device_size = i_size_read(bdev->bd_inode);
+
+	if (le32_to_cpu(fec->magic) == FEC_MAGIC)
+		*metadata_offset = le64_to_cpu(fec->inp_size) -
+					VERITY_METADATA_SIZE;
+	else
+		*metadata_offset = device_size - VERITY_METADATA_SIZE;
+}
+
+static int find_size(dev_t dev, u64 *device_size)
+{
+	struct block_device *bdev;
+
+	bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL);
+	if (IS_ERR_OR_NULL(bdev)) {
+		DMERR("blkdev_get_by_dev failed");
+		return PTR_ERR(bdev);
+	}
+
+	*device_size = i_size_read(bdev->bd_inode);
+	*device_size >>= SECTOR_SHIFT;
+
+	DMINFO("blkdev size in sectors: %llu", *device_size);
+	blkdev_put(bdev, FMODE_READ);
+	return 0;
+}
+
+static int verify_header(struct android_metadata_header *header)
+{
+	int retval = -EINVAL;
+
+	if (is_userdebug() && le32_to_cpu(header->magic_number) ==
+			VERITY_METADATA_MAGIC_DISABLE)
+		return VERITY_STATE_DISABLE;
+
+	if (!(le32_to_cpu(header->magic_number) ==
+			VERITY_METADATA_MAGIC_NUMBER) ||
+			(le32_to_cpu(header->magic_number) ==
+			VERITY_METADATA_MAGIC_DISABLE)) {
+		DMERR("Incorrect magic number");
+		return retval;
+	}
+
+	if (le32_to_cpu(header->protocol_version) !=
+			VERITY_METADATA_VERSION) {
+		DMERR("Unsupported version %u",
+			le32_to_cpu(header->protocol_version));
+		return retval;
+	}
+
+	return 0;
+}
+
+static int extract_metadata(dev_t dev, struct fec_header *fec,
+				struct android_metadata **metadata,
+				bool *verity_enabled)
+{
+	struct block_device *bdev;
+	struct android_metadata_header *header;
+	int i;
+	u32 table_length, copy_length, offset;
+	u64 metadata_offset;
+	struct bio_read payload;
+	int err = 0;
+
+	bdev = blkdev_get_by_dev(dev, FMODE_READ, NULL);
+
+	if (IS_ERR_OR_NULL(bdev)) {
+		DMERR("blkdev_get_by_dev failed");
+		return -ENODEV;
+	}
+
+	find_metadata_offset(fec, bdev, &metadata_offset);
+
+	/* Verity metadata size is a power of 2 and PAGE_SIZE
+	 * is a power of 2 as well.
+	 * PAGE_SIZE is also a multiple of 512 bytes.
+	*/
+	if (VERITY_METADATA_SIZE > PAGE_SIZE)
+		BUG_ON(VERITY_METADATA_SIZE % PAGE_SIZE != 0);
+	/* 512 byte sector alignment */
+	BUG_ON(metadata_offset % (1 << SECTOR_SHIFT) != 0);
+
+	err = read_block_dev(&payload, bdev, metadata_offset /
+		(1 << SECTOR_SHIFT), VERITY_METADATA_SIZE);
+	if (err) {
+		DMERR("Error while reading verity metadata");
+		goto blkdev_release;
+	}
+
+	header = kzalloc(sizeof(*header), GFP_KERNEL);
+	if (!header) {
+		DMERR("kzalloc failed for header");
+		err = -ENOMEM;
+		goto free_payload;
+	}
+
+	memcpy(header, page_address(payload.page_io[0]),
+		sizeof(*header));
+
+	DMINFO("bio magic_number:%u protocol_version:%d table_length:%u",
+		le32_to_cpu(header->magic_number),
+		le32_to_cpu(header->protocol_version),
+		le32_to_cpu(header->table_length));
+
+	err = verify_header(header);
+
+	if (err == VERITY_STATE_DISABLE) {
+		DMERR("Mounting root with verity disabled");
+		*verity_enabled = false;
+		/* we would still have to read the metadata to figure out
+		 * the data blocks size. Or may be could map the entire
+		 * partition similar to mounting the device.
+		 *
+		 * Reset error as well as the verity_enabled flag is changed.
+		 */
+		err = 0;
+	} else if (err)
+		goto free_header;
+
+	*metadata = kzalloc(sizeof(**metadata), GFP_KERNEL);
+	if (!*metadata) {
+		DMERR("kzalloc for metadata failed");
+		err = -ENOMEM;
+		goto free_header;
+	}
+
+	(*metadata)->header = header;
+	table_length = le32_to_cpu(header->table_length);
+
+	if (table_length == 0 ||
+		table_length > (VERITY_METADATA_SIZE -
+			sizeof(struct android_metadata_header))) {
+		DMERR("table_length too long");
+		err = -EINVAL;
+		goto free_metadata;
+	}
+
+	(*metadata)->verity_table = kzalloc(table_length + 1, GFP_KERNEL);
+
+	if (!(*metadata)->verity_table) {
+		DMERR("kzalloc verity_table failed");
+		err = -ENOMEM;
+		goto free_metadata;
+	}
+
+	if (sizeof(struct android_metadata_header) +
+			table_length <= PAGE_SIZE) {
+		memcpy((*metadata)->verity_table,
+			page_address(payload.page_io[0])
+			+ sizeof(struct android_metadata_header),
+			table_length);
+	} else {
+		copy_length = PAGE_SIZE -
+			sizeof(struct android_metadata_header);
+		memcpy((*metadata)->verity_table,
+			page_address(payload.page_io[0])
+			+ sizeof(struct android_metadata_header),
+			copy_length);
+		table_length -= copy_length;
+		offset = copy_length;
+		i = 1;
+		while (table_length != 0) {
+			if (table_length > PAGE_SIZE) {
+				memcpy((*metadata)->verity_table + offset,
+					page_address(payload.page_io[i]),
+					PAGE_SIZE);
+				offset += PAGE_SIZE;
+				table_length -= PAGE_SIZE;
+			} else {
+				memcpy((*metadata)->verity_table + offset,
+					page_address(payload.page_io[i]),
+					table_length);
+				table_length = 0;
+			}
+			i++;
+		}
+	}
+	(*metadata)->verity_table[table_length] = '\0';
+
+	DMINFO("verity_table: %s", (*metadata)->verity_table);
+	goto free_payload;
+
+free_metadata:
+	kfree(*metadata);
+free_header:
+	kfree(header);
+free_payload:
+	for (i = 0; i < payload.number_of_pages; i++)
+		if (payload.page_io[i])
+			__free_page(payload.page_io[i]);
+	kfree(payload.page_io);
+blkdev_release:
+	blkdev_put(bdev, FMODE_READ);
+	return err;
+}
+
+/* helper functions to extract properties from dts */
+const char *find_dt_value(const char *name)
+{
+	struct device_node *firmware;
+	const char *value;
+
+	firmware = of_find_node_by_path("/firmware/android");
+	if (!firmware)
+		return NULL;
+	value = of_get_property(firmware, name, NULL);
+	of_node_put(firmware);
+
+	return value;
+}
+
+static int verity_mode(void)
+{
+	static const char enforcing[] = "enforcing";
+	static const char verified_mode_prop[] = "veritymode";
+	const char *value;
+
+	value = find_dt_value(verified_mode_prop);
+	if (!value)
+		value = veritymode;
+	if (!strncmp(value, enforcing, sizeof(enforcing) - 1))
+		return DM_VERITY_MODE_RESTART;
+
+	return DM_VERITY_MODE_EIO;
+}
+
+static int verify_verity_signature(char *key_id,
+		struct android_metadata *metadata)
+{
+	key_ref_t key_ref;
+	struct key *key;
+	struct public_key_signature *pks = NULL;
+	int retval = -EINVAL;
+
+	key_ref = keyring_search(make_key_ref(system_trusted_keyring, 1),
+		&key_type_asymmetric, key_id);
+
+	if (IS_ERR(key_ref)) {
+		DMERR("keyring: key not found");
+		return -ENOKEY;
+	}
+
+	key = key_ref_to_ptr(key_ref);
+
+	pks = table_make_digest(HASH_ALGO_SHA256,
+			(const void *)metadata->verity_table,
+			le32_to_cpu(metadata->header->table_length));
+
+	if (IS_ERR(pks)) {
+		DMERR("hashing failed");
+		retval = PTR_ERR(pks);
+		pks = NULL;
+		goto error;
+	}
+
+	retval = table_extract_mpi_array(pks, &metadata->header->signature[0],
+				RSANUMBYTES);
+	if (retval < 0) {
+		DMERR("Error extracting mpi %d", retval);
+		goto error;
+	}
+
+	retval = verify_signature(key, pks);
+	mpi_free(pks->rsa.s);
+error:
+	kfree(pks);
+	key_put(key);
+
+	return retval;
+}
+
+static void handle_error(void)
+{
+	int mode = verity_mode();
+	if (mode == DM_VERITY_MODE_RESTART) {
+		DMERR("triggering restart");
+		kernel_restart("dm-verity device corrupted");
+	} else {
+		DMERR("Mounting verity root failed");
+	}
+}
+
+static inline bool test_mult_overflow(sector_t a, u32 b)
+{
+	sector_t r = (sector_t)~0ULL;
+
+	sector_div(r, b);
+	return a > r;
+}
+
+static int add_as_linear_device(struct dm_target *ti, char *dev)
+{
+	/*Move to linear mapping defines*/
+	char *linear_table_args[DM_LINEAR_ARGS] = {dev,
+					DM_LINEAR_TARGET_OFFSET};
+	int err = 0;
+
+	android_verity_target.dtr = dm_linear_dtr,
+	android_verity_target.map = dm_linear_map,
+	android_verity_target.status = dm_linear_status,
+	android_verity_target.prepare_ioctl = dm_linear_prepare_ioctl,
+	android_verity_target.iterate_devices = dm_linear_iterate_devices,
+	android_verity_target.io_hints = NULL;
+
+	set_disk_ro(dm_disk(dm_table_get_md(ti->table)), 0);
+
+	err = dm_linear_ctr(ti, DM_LINEAR_ARGS, linear_table_args);
+
+	if (!err) {
+		DMINFO("Added android-verity as a linear target");
+		target_added = true;
+	} else
+		DMERR("Failed to add android-verity as linear target");
+
+	return err;
+}
+
+static int create_linear_device(struct dm_target *ti, dev_t dev,
+				char *target_device)
+{
+	u64 device_size = 0;
+	int err = find_size(dev, &device_size);
+
+	if (err) {
+		DMERR("error finding bdev size");
+		handle_error();
+		return err;
+	}
+
+	ti->len = device_size;
+	err = add_as_linear_device(ti, target_device);
+	if (err) {
+		handle_error();
+		return err;
+	}
+	verity_enabled = false;
+	return 0;
+}
+
+/*
+ * Target parameters:
+ *	<key id>	Key id of the public key in the system keyring.
+ *			Verity metadata's signature would be verified against
+ *			this. If the key id contains spaces, replace them
+ *			with '#'.
+ *	<block device>	The block device for which dm-verity is being setup.
+ */
+static int android_verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
+{
+	dev_t uninitialized_var(dev);
+	struct android_metadata *metadata = NULL;
+	int err = 0, i, mode;
+	char *key_id, *table_ptr, dummy, *target_device,
+	*verity_table_args[VERITY_TABLE_ARGS + 2 + VERITY_TABLE_OPT_FEC_ARGS];
+	/* One for specifying number of opt args and one for mode */
+	sector_t data_sectors;
+	u32 data_block_size;
+	unsigned int no_of_args = VERITY_TABLE_ARGS + 2 + VERITY_TABLE_OPT_FEC_ARGS;
+	struct fec_header uninitialized_var(fec);
+	struct fec_ecc_metadata uninitialized_var(ecc);
+	char buf[FEC_ARG_LENGTH], *buf_ptr;
+	unsigned long long tmpll;
+
+	if (argc == 1) {
+		/* Use the default keyid */
+		if (default_verity_key_id())
+			key_id = veritykeyid;
+		else if (!is_eng()) {
+			DMERR("veritykeyid= is not set");
+			handle_error();
+			return -EINVAL;
+		}
+	} else if (argc == 2)
+		key_id = argv[1];
+	else {
+		DMERR("Incorrect number of arguments");
+		handle_error();
+		return -EINVAL;
+	}
+
+	target_device = argv[0];
+
+	dev = name_to_dev_t(target_device);
+	if (!dev) {
+		DMERR("no dev found for %s", target_device);
+		handle_error();
+		return -EINVAL;
+	}
+
+	if (is_eng())
+		return create_linear_device(ti, dev, target_device);
+
+	strreplace(key_id, '#', ' ');
+
+	DMINFO("key:%s dev:%s", key_id, target_device);
+
+	if (extract_fec_header(dev, &fec, &ecc)) {
+		DMERR("Error while extracting fec header");
+		handle_error();
+		return -EINVAL;
+	}
+
+	err = extract_metadata(dev, &fec, &metadata, &verity_enabled);
+
+	if (err) {
+		/* Allow invalid metadata when the device is unlocked */
+		if (is_unlocked()) {
+			DMWARN("Allow invalid metadata when unlocked");
+			return create_linear_device(ti, dev, target_device);
+		}
+		DMERR("Error while extracting metadata");
+		handle_error();
+		goto free_metadata;
+	}
+
+	if (verity_enabled) {
+		err = verify_verity_signature(key_id, metadata);
+
+		if (err) {
+			DMERR("Signature verification failed");
+			handle_error();
+			goto free_metadata;
+		} else
+			DMINFO("Signature verification success");
+	}
+
+	table_ptr = metadata->verity_table;
+
+	for (i = 0; i < VERITY_TABLE_ARGS; i++) {
+		verity_table_args[i] = strsep(&table_ptr, " ");
+		if (verity_table_args[i] == NULL)
+			break;
+	}
+
+	if (i != VERITY_TABLE_ARGS) {
+		DMERR("Verity table not in the expected format");
+		err = -EINVAL;
+		handle_error();
+		goto free_metadata;
+	}
+
+	if (sscanf(verity_table_args[5], "%llu%c", &tmpll, &dummy)
+							!= 1) {
+		DMERR("Verity table not in the expected format");
+		handle_error();
+		err = -EINVAL;
+		goto free_metadata;
+	}
+
+	if (tmpll > ULONG_MAX) {
+		DMERR("<num_data_blocks> too large. Forgot to turn on CONFIG_LBDAF?");
+		handle_error();
+		err = -EINVAL;
+		goto free_metadata;
+	}
+
+	data_sectors = tmpll;
+
+	if (sscanf(verity_table_args[3], "%u%c", &data_block_size, &dummy)
+								!= 1) {
+		DMERR("Verity table not in the expected format");
+		handle_error();
+		err = -EINVAL;
+		goto free_metadata;
+	}
+
+	if (test_mult_overflow(data_sectors, data_block_size >>
+							SECTOR_SHIFT)) {
+		DMERR("data_sectors too large");
+		handle_error();
+		err = -EOVERFLOW;
+		goto free_metadata;
+	}
+
+	data_sectors *= data_block_size >> SECTOR_SHIFT;
+	DMINFO("Data sectors %llu", (unsigned long long)data_sectors);
+
+	/* update target length */
+	ti->len = data_sectors;
+
+	/* Setup linear target and free */
+	if (!verity_enabled) {
+		err = add_as_linear_device(ti, target_device);
+		goto free_metadata;
+	}
+
+	/*substitute data_dev and hash_dev*/
+	verity_table_args[1] = target_device;
+	verity_table_args[2] = target_device;
+
+	mode = verity_mode();
+
+	if (ecc.valid && IS_BUILTIN(CONFIG_DM_VERITY_FEC)) {
+		if (mode) {
+			err = snprintf(buf, FEC_ARG_LENGTH,
+				"%u %s " VERITY_TABLE_OPT_FEC_FORMAT,
+				1 + VERITY_TABLE_OPT_FEC_ARGS,
+				mode == DM_VERITY_MODE_RESTART ?
+					VERITY_TABLE_OPT_RESTART :
+					VERITY_TABLE_OPT_LOGGING,
+				target_device,
+				ecc.start / FEC_BLOCK_SIZE, ecc.blocks,
+				ecc.roots);
+		} else {
+			err = snprintf(buf, FEC_ARG_LENGTH,
+				"%u " VERITY_TABLE_OPT_FEC_FORMAT,
+				VERITY_TABLE_OPT_FEC_ARGS, target_device,
+				ecc.start / FEC_BLOCK_SIZE, ecc.blocks,
+				ecc.roots);
+		}
+	} else if (mode) {
+		err = snprintf(buf, FEC_ARG_LENGTH,
+			"2 " VERITY_TABLE_OPT_IGNZERO " %s",
+			mode == DM_VERITY_MODE_RESTART ?
+			VERITY_TABLE_OPT_RESTART : VERITY_TABLE_OPT_LOGGING);
+	} else {
+		err = snprintf(buf, FEC_ARG_LENGTH, "1 %s",
+				 "ignore_zero_blocks");
+	}
+
+	if (err < 0 || err >= FEC_ARG_LENGTH)
+		goto free_metadata;
+
+	buf_ptr = buf;
+
+	for (i = VERITY_TABLE_ARGS; i < (VERITY_TABLE_ARGS +
+		VERITY_TABLE_OPT_FEC_ARGS + 2); i++) {
+		verity_table_args[i] = strsep(&buf_ptr, " ");
+		if (verity_table_args[i] == NULL) {
+			no_of_args = i;
+			break;
+		}
+	}
+
+	err = verity_ctr(ti, no_of_args, verity_table_args);
+
+	if (err)
+		DMERR("android-verity failed to mount as verity target");
+	else {
+		target_added = true;
+		DMINFO("android-verity mounted as verity target");
+	}
+
+free_metadata:
+	if (metadata) {
+		kfree(metadata->header);
+		kfree(metadata->verity_table);
+	}
+	kfree(metadata);
+	return err;
+}
+
+static int __init dm_android_verity_init(void)
+{
+	int r;
+	struct dentry *file;
+
+	r = dm_register_target(&android_verity_target);
+	if (r < 0)
+		DMERR("register failed %d", r);
+
+	/* Tracks the status of the last added target */
+	debug_dir = debugfs_create_dir("android_verity", NULL);
+
+	if (IS_ERR_OR_NULL(debug_dir)) {
+		DMERR("Cannot create android_verity debugfs directory: %ld",
+			PTR_ERR(debug_dir));
+		goto end;
+	}
+
+	file = debugfs_create_bool("target_added", S_IRUGO, debug_dir,
+				&target_added);
+
+	if (IS_ERR_OR_NULL(file)) {
+		DMERR("Cannot create android_verity debugfs directory: %ld",
+			PTR_ERR(debug_dir));
+		debugfs_remove_recursive(debug_dir);
+		goto end;
+	}
+
+	file = debugfs_create_bool("verity_enabled", S_IRUGO, debug_dir,
+				&verity_enabled);
+
+	if (IS_ERR_OR_NULL(file)) {
+		DMERR("Cannot create android_verity debugfs directory: %ld",
+			PTR_ERR(debug_dir));
+		debugfs_remove_recursive(debug_dir);
+	}
+
+end:
+	return r;
+}
+
+static void __exit dm_android_verity_exit(void)
+{
+	if (!IS_ERR_OR_NULL(debug_dir))
+		debugfs_remove_recursive(debug_dir);
+
+	dm_unregister_target(&android_verity_target);
+}
+
+module_init(dm_android_verity_init);
+module_exit(dm_android_verity_exit);
diff --git a/drivers/md/dm-android-verity.h b/drivers/md/dm-android-verity.h
new file mode 100644
index 000000000000..0c7ff6afec69
--- /dev/null
+++ b/drivers/md/dm-android-verity.h
@@ -0,0 +1,121 @@
+/*
+ * Copyright (C) 2015 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef DM_ANDROID_VERITY_H
+#define DM_ANDROID_VERITY_H
+
+#include <crypto/sha.h>
+
+#define RSANUMBYTES 256
+#define VERITY_METADATA_MAGIC_NUMBER 0xb001b001
+#define VERITY_METADATA_MAGIC_DISABLE 0x46464f56
+#define VERITY_METADATA_VERSION 0
+#define VERITY_STATE_DISABLE 1
+#define DATA_BLOCK_SIZE (4 * 1024)
+#define VERITY_METADATA_SIZE (8 * DATA_BLOCK_SIZE)
+#define VERITY_TABLE_ARGS 10
+#define VERITY_COMMANDLINE_PARAM_LENGTH 20
+#define BUILD_VARIANT 20
+
+/*
+ * <subject>:<sha1-id> is the format for the identifier.
+ * subject can either be the Common Name(CN) + Organization Name(O) or
+ * just the CN if the it is prefixed with O
+ * From https://tools.ietf.org/html/rfc5280#appendix-A
+ * ub-organization-name-length INTEGER ::= 64
+ * ub-common-name-length INTEGER ::= 64
+ *
+ * http://lxr.free-electrons.com/source/crypto/asymmetric_keys/x509_cert_parser.c?v=3.9#L278
+ * ctx->o_size + 2 + ctx->cn_size + 1
+ * + 41 characters for ":" and sha1 id
+ * 64 + 2 + 64 + 1 + 1 + 40 (172)
+ * setting VERITY_DEFAULT_KEY_ID_LENGTH to 200 characters.
+ */
+#define VERITY_DEFAULT_KEY_ID_LENGTH 200
+
+#define FEC_MAGIC 0xFECFECFE
+#define FEC_BLOCK_SIZE (4 * 1024)
+#define FEC_VERSION 0
+#define FEC_RSM 255
+#define FEC_ARG_LENGTH 300
+
+#define VERITY_TABLE_OPT_RESTART "restart_on_corruption"
+#define VERITY_TABLE_OPT_LOGGING "ignore_corruption"
+#define VERITY_TABLE_OPT_IGNZERO "ignore_zero_blocks"
+
+#define VERITY_TABLE_OPT_FEC_FORMAT \
+	"use_fec_from_device %s fec_start %llu fec_blocks %llu fec_roots %u ignore_zero_blocks"
+#define VERITY_TABLE_OPT_FEC_ARGS 9
+
+#define VERITY_DEBUG 0
+
+#define DM_MSG_PREFIX                   "android-verity"
+
+#define DM_LINEAR_ARGS 2
+#define DM_LINEAR_TARGET_OFFSET "0"
+
+/*
+ * There can be two formats.
+ * if fec is present
+ * <data_blocks> <verity_tree> <verity_metdata_32K><fec_data><fec_data_4K>
+ * if fec is not present
+ * <data_blocks> <verity_tree> <verity_metdata_32K>
+ */
+struct fec_header {
+	__le32 magic;
+	__le32 version;
+	__le32 size;
+	__le32 roots;
+	__le32 fec_size;
+	__le64 inp_size;
+	u8 hash[SHA256_DIGEST_SIZE];
+} __attribute__((packed));
+
+struct android_metadata_header {
+	__le32 magic_number;
+	__le32 protocol_version;
+	char signature[RSANUMBYTES];
+	__le32 table_length;
+};
+
+struct android_metadata {
+	struct android_metadata_header *header;
+	char *verity_table;
+};
+
+struct fec_ecc_metadata {
+	bool valid;
+	u32 roots;
+	u64 blocks;
+	u64 rounds;
+	u64 start;
+};
+
+struct bio_read {
+	struct page **page_io;
+	int number_of_pages;
+};
+
+extern struct target_type linear_target;
+
+extern void dm_linear_dtr(struct dm_target *ti);
+extern int dm_linear_map(struct dm_target *ti, struct bio *bio);
+extern void dm_linear_status(struct dm_target *ti, status_type_t type,
+			unsigned status_flags, char *result, unsigned maxlen);
+extern int dm_linear_prepare_ioctl(struct dm_target *ti,
+                struct block_device **bdev, fmode_t *mode);
+extern int dm_linear_iterate_devices(struct dm_target *ti,
+			iterate_devices_callout_fn fn, void *data);
+extern int dm_linear_ctr(struct dm_target *ti, unsigned int argc, char **argv);
+#endif /* DM_ANDROID_VERITY_H */
diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c
index b59615ddf6ba..0da5efaad85c 100644
--- a/drivers/md/dm-cache-target.c
+++ b/drivers/md/dm-cache-target.c
@@ -118,14 +118,12 @@ static void iot_io_end(struct io_tracker *iot, sector_t len)
  */
 struct dm_hook_info {
 	bio_end_io_t *bi_end_io;
-	void *bi_private;
 };
 
 static void dm_hook_bio(struct dm_hook_info *h, struct bio *bio,
 			bio_end_io_t *bi_end_io, void *bi_private)
 {
 	h->bi_end_io = bio->bi_end_io;
-	h->bi_private = bio->bi_private;
 
 	bio->bi_end_io = bi_end_io;
 	bio->bi_private = bi_private;
@@ -134,7 +132,6 @@ static void dm_hook_bio(struct dm_hook_info *h, struct bio *bio,
 static void dm_unhook_bio(struct dm_hook_info *h, struct bio *bio)
 {
 	bio->bi_end_io = h->bi_end_io;
-	bio->bi_private = h->bi_private;
 }
 
 /*----------------------------------------------------------------*/
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index de628883ee3d..799b9a5ad4f5 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -1863,16 +1863,24 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	}
 
 	ret = -ENOMEM;
-	cc->io_queue = alloc_workqueue("kcryptd_io", WQ_MEM_RECLAIM, 1);
+	cc->io_queue = alloc_workqueue("kcryptd_io",
+				       WQ_HIGHPRI |
+				       WQ_MEM_RECLAIM,
+				       1);
 	if (!cc->io_queue) {
 		ti->error = "Couldn't create kcryptd io queue";
 		goto bad;
 	}
 
 	if (test_bit(DM_CRYPT_SAME_CPU, &cc->flags))
-		cc->crypt_queue = alloc_workqueue("kcryptd", WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM, 1);
+		cc->crypt_queue = alloc_workqueue("kcryptd",
+						  WQ_HIGHPRI |
+						  WQ_MEM_RECLAIM, 1);
 	else
-		cc->crypt_queue = alloc_workqueue("kcryptd", WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM | WQ_UNBOUND,
+		cc->crypt_queue = alloc_workqueue("kcryptd",
+						  WQ_HIGHPRI |
+						  WQ_MEM_RECLAIM |
+						  WQ_UNBOUND,
 						  num_online_cpus());
 	if (!cc->crypt_queue) {
 		ti->error = "Couldn't create kcryptd queue";
diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index e503279c34fc..7baeeafa059d 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -1923,6 +1923,45 @@ void dm_interface_exit(void)
 	dm_hash_exit();
 }
 
+
+/**
+ * dm_ioctl_export - Permanently export a mapped device via the ioctl interface
+ * @md: Pointer to mapped_device
+ * @name: Buffer (size DM_NAME_LEN) for name
+ * @uuid: Buffer (size DM_UUID_LEN) for uuid or NULL if not desired
+ */
+int dm_ioctl_export(struct mapped_device *md, const char *name,
+		    const char *uuid)
+{
+	int r = 0;
+	struct hash_cell *hc;
+
+	if (!md) {
+		r = -ENXIO;
+		goto out;
+	}
+
+	/* The name and uuid can only be set once. */
+	mutex_lock(&dm_hash_cells_mutex);
+	hc = dm_get_mdptr(md);
+	mutex_unlock(&dm_hash_cells_mutex);
+	if (hc) {
+		DMERR("%s: already exported", dm_device_name(md));
+		r = -ENXIO;
+		goto out;
+	}
+
+	r = dm_hash_insert(name, uuid, md);
+	if (r) {
+		DMERR("%s: could not bind to '%s'", dm_device_name(md), name);
+		goto out;
+	}
+
+	/* Let udev know we've changed. */
+	dm_kobject_uevent(md, KOBJ_CHANGE, dm_get_event_nr(md));
+out:
+	return r;
+}
 /**
  * dm_copy_name_and_uuid - Copy mapped device name & uuid into supplied buffers
  * @md: Pointer to mapped_device
diff --git a/drivers/md/dm-linear.c b/drivers/md/dm-linear.c
index 05c35aacb3aa..2ff5f32a4b99 100644
--- a/drivers/md/dm-linear.c
+++ b/drivers/md/dm-linear.c
@@ -25,7 +25,7 @@ struct linear_c {
 /*
  * Construct a linear mapping: <dev_path> <offset>
  */
-static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv)
+int dm_linear_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 {
 	struct linear_c *lc;
 	unsigned long long tmp;
@@ -66,14 +66,16 @@ static int linear_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	kfree(lc);
 	return ret;
 }
+EXPORT_SYMBOL_GPL(dm_linear_ctr);
 
-static void linear_dtr(struct dm_target *ti)
+void dm_linear_dtr(struct dm_target *ti)
 {
 	struct linear_c *lc = (struct linear_c *) ti->private;
 
 	dm_put_device(ti, lc->dev);
 	kfree(lc);
 }
+EXPORT_SYMBOL_GPL(dm_linear_dtr);
 
 static sector_t linear_map_sector(struct dm_target *ti, sector_t bi_sector)
 {
@@ -92,14 +94,15 @@ static void linear_map_bio(struct dm_target *ti, struct bio *bio)
 			linear_map_sector(ti, bio->bi_iter.bi_sector);
 }
 
-static int linear_map(struct dm_target *ti, struct bio *bio)
+int dm_linear_map(struct dm_target *ti, struct bio *bio)
 {
 	linear_map_bio(ti, bio);
 
 	return DM_MAPIO_REMAPPED;
 }
+EXPORT_SYMBOL_GPL(dm_linear_map);
 
-static void linear_status(struct dm_target *ti, status_type_t type,
+void dm_linear_status(struct dm_target *ti, status_type_t type,
 			  unsigned status_flags, char *result, unsigned maxlen)
 {
 	struct linear_c *lc = (struct linear_c *) ti->private;
@@ -115,8 +118,9 @@ static void linear_status(struct dm_target *ti, status_type_t type,
 		break;
 	}
 }
+EXPORT_SYMBOL_GPL(dm_linear_status);
 
-static int linear_prepare_ioctl(struct dm_target *ti,
+int dm_linear_prepare_ioctl(struct dm_target *ti,
 		struct block_device **bdev, fmode_t *mode)
 {
 	struct linear_c *lc = (struct linear_c *) ti->private;
@@ -132,25 +136,27 @@ static int linear_prepare_ioctl(struct dm_target *ti,
 		return 1;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(dm_linear_prepare_ioctl);
 
-static int linear_iterate_devices(struct dm_target *ti,
+int dm_linear_iterate_devices(struct dm_target *ti,
 				  iterate_devices_callout_fn fn, void *data)
 {
 	struct linear_c *lc = ti->private;
 
 	return fn(ti, lc->dev, lc->start, ti->len, data);
 }
+EXPORT_SYMBOL_GPL(dm_linear_iterate_devices);
 
 static struct target_type linear_target = {
 	.name   = "linear",
 	.version = {1, 2, 1},
 	.module = THIS_MODULE,
-	.ctr    = linear_ctr,
-	.dtr    = linear_dtr,
-	.map    = linear_map,
-	.status = linear_status,
-	.prepare_ioctl = linear_prepare_ioctl,
-	.iterate_devices = linear_iterate_devices,
+	.ctr    = dm_linear_ctr,
+	.dtr    = dm_linear_dtr,
+	.map    = dm_linear_map,
+	.status = dm_linear_status,
+	.prepare_ioctl  = dm_linear_prepare_ioctl,
+	.iterate_devices = dm_linear_iterate_devices,
 };
 
 int __init dm_linear_init(void)
diff --git a/drivers/md/dm-snap.c b/drivers/md/dm-snap.c
index e108deebbaaa..e4d1bafe78c1 100644
--- a/drivers/md/dm-snap.c
+++ b/drivers/md/dm-snap.c
@@ -207,7 +207,6 @@ struct dm_snap_pending_exception {
 	 */
 	struct bio *full_bio;
 	bio_end_io_t *full_bio_end_io;
-	void *full_bio_private;
 };
 
 /*
@@ -1495,10 +1494,8 @@ out:
 	snapshot_bios = bio_list_get(&pe->snapshot_bios);
 	origin_bios = bio_list_get(&pe->origin_bios);
 	full_bio = pe->full_bio;
-	if (full_bio) {
+	if (full_bio)
 		full_bio->bi_end_io = pe->full_bio_end_io;
-		full_bio->bi_private = pe->full_bio_private;
-	}
 	increment_pending_exceptions_done_count();
 
 	up_write(&s->lock);
@@ -1604,7 +1601,6 @@ static void start_full_bio(struct dm_snap_pending_exception *pe,
 
 	pe->full_bio = bio;
 	pe->full_bio_end_io = bio->bi_end_io;
-	pe->full_bio_private = bio->bi_private;
 
 	callback_data = dm_kcopyd_prepare_callback(s->kcopyd_client,
 						   copy_callback, pe);
diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index cb5d0daf53bb..b3d78bba3a79 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -11,6 +11,7 @@
 #include <linux/vmalloc.h>
 #include <linux/blkdev.h>
 #include <linux/namei.h>
+#include <linux/mount.h>
 #include <linux/ctype.h>
 #include <linux/string.h>
 #include <linux/slab.h>
diff --git a/drivers/md/dm-verity-avb.c b/drivers/md/dm-verity-avb.c
new file mode 100644
index 000000000000..727aacbb1480
--- /dev/null
+++ b/drivers/md/dm-verity-avb.c
@@ -0,0 +1,229 @@
+/*
+ * Copyright (C) 2017 Google.
+ *
+ * This file is released under the GPLv2.
+ *
+ * Based on drivers/md/dm-verity-chromeos.c
+ */
+
+#include <linux/device-mapper.h>
+#include <linux/module.h>
+#include <linux/mount.h>
+
+#define DM_MSG_PREFIX "verity-avb"
+
+/* Set via module parameters. */
+static char avb_vbmeta_device[64];
+static char avb_invalidate_on_error[4];
+
+static void invalidate_vbmeta_endio(struct bio *bio)
+{
+	if (bio->bi_error)
+		DMERR("invalidate_vbmeta_endio: error %d", bio->bi_error);
+	complete(bio->bi_private);
+}
+
+static int invalidate_vbmeta_submit(struct bio *bio,
+				    struct block_device *bdev,
+				    int rw, int access_last_sector,
+				    struct page *page)
+{
+	DECLARE_COMPLETION_ONSTACK(wait);
+
+	bio->bi_private = &wait;
+	bio->bi_end_io = invalidate_vbmeta_endio;
+	bio->bi_bdev = bdev;
+	bio->bi_rw = rw;
+
+	bio->bi_iter.bi_sector = 0;
+	if (access_last_sector) {
+		sector_t last_sector;
+
+		last_sector = (i_size_read(bdev->bd_inode)>>SECTOR_SHIFT) - 1;
+		bio->bi_iter.bi_sector = last_sector;
+	}
+	if (!bio_add_page(bio, page, PAGE_SIZE, 0)) {
+		DMERR("invalidate_vbmeta_submit: bio_add_page error");
+		return -EIO;
+	}
+
+	submit_bio(rw, bio);
+	/* Wait up to 2 seconds for completion or fail. */
+	if (!wait_for_completion_timeout(&wait, msecs_to_jiffies(2000)))
+		return -EIO;
+	return 0;
+}
+
+static int invalidate_vbmeta(dev_t vbmeta_devt)
+{
+	int ret = 0;
+	struct block_device *bdev;
+	struct bio *bio;
+	struct page *page;
+	fmode_t dev_mode;
+	/* Ensure we do synchronous unblocked I/O. We may also need
+	 * sync_bdev() on completion, but it really shouldn't.
+	 */
+	int rw = REQ_SYNC | REQ_SOFTBARRIER | REQ_NOIDLE;
+	int access_last_sector = 0;
+
+	DMINFO("invalidate_vbmeta: acting on device %d:%d",
+	       MAJOR(vbmeta_devt), MINOR(vbmeta_devt));
+
+	/* First we open the device for reading. */
+	dev_mode = FMODE_READ | FMODE_EXCL;
+	bdev = blkdev_get_by_dev(vbmeta_devt, dev_mode,
+				 invalidate_vbmeta);
+	if (IS_ERR(bdev)) {
+		DMERR("invalidate_kernel: could not open device for reading");
+		dev_mode = 0;
+		ret = -ENOENT;
+		goto failed_to_read;
+	}
+
+	bio = bio_alloc(GFP_NOIO, 1);
+	if (!bio) {
+		ret = -ENOMEM;
+		goto failed_bio_alloc;
+	}
+
+	page = alloc_page(GFP_NOIO);
+	if (!page) {
+		ret = -ENOMEM;
+		goto failed_to_alloc_page;
+	}
+
+	access_last_sector = 0;
+	ret = invalidate_vbmeta_submit(bio, bdev, rw, access_last_sector, page);
+	if (ret) {
+		DMERR("invalidate_vbmeta: error reading");
+		goto failed_to_submit_read;
+	}
+
+	/* We have a page. Let's make sure it looks right. */
+	if (memcmp("AVB0", page_address(page), 4) == 0) {
+		/* Stamp it. */
+		memcpy(page_address(page), "AVE0", 4);
+		DMINFO("invalidate_vbmeta: found vbmeta partition");
+	} else {
+		/* Could be this is on a AVB footer, check. Also, since the
+		 * AVB footer is in the last 64 bytes, adjust for the fact that
+		 * we're dealing with 512-byte sectors.
+		 */
+		size_t offset = (1<<SECTOR_SHIFT) - 64;
+
+		access_last_sector = 1;
+		ret = invalidate_vbmeta_submit(bio, bdev, rw,
+					       access_last_sector, page);
+		if (ret) {
+			DMERR("invalidate_vbmeta: error reading");
+			goto failed_to_submit_read;
+		}
+		if (memcmp("AVBf", page_address(page) + offset, 4) != 0) {
+			DMERR("invalidate_vbmeta on non-vbmeta partition");
+			ret = -EINVAL;
+			goto invalid_header;
+		}
+		/* Stamp it. */
+		memcpy(page_address(page) + offset, "AVE0", 4);
+		DMINFO("invalidate_vbmeta: found vbmeta footer partition");
+	}
+
+	/* Now rewrite the changed page - the block dev was being
+	 * changed on read. Let's reopen here.
+	 */
+	blkdev_put(bdev, dev_mode);
+	dev_mode = FMODE_WRITE | FMODE_EXCL;
+	bdev = blkdev_get_by_dev(vbmeta_devt, dev_mode,
+				 invalidate_vbmeta);
+	if (IS_ERR(bdev)) {
+		DMERR("invalidate_vbmeta: could not open device for writing");
+		dev_mode = 0;
+		ret = -ENOENT;
+		goto failed_to_write;
+	}
+
+	/* We re-use the same bio to do the write after the read. Need to reset
+	 * it to initialize bio->bi_remaining.
+	 */
+	bio_reset(bio);
+
+	rw |= REQ_WRITE;
+	ret = invalidate_vbmeta_submit(bio, bdev, rw, access_last_sector, page);
+	if (ret) {
+		DMERR("invalidate_vbmeta: error writing");
+		goto failed_to_submit_write;
+	}
+
+	DMERR("invalidate_vbmeta: completed.");
+	ret = 0;
+failed_to_submit_write:
+failed_to_write:
+invalid_header:
+	__free_page(page);
+failed_to_submit_read:
+	/* Technically, we'll leak a page with the pending bio, but
+	 * we're about to reboot anyway.
+	 */
+failed_to_alloc_page:
+	bio_put(bio);
+failed_bio_alloc:
+	if (dev_mode)
+		blkdev_put(bdev, dev_mode);
+failed_to_read:
+	return ret;
+}
+
+void dm_verity_avb_error_handler(void)
+{
+	dev_t dev;
+
+	DMINFO("AVB error handler called for %s", avb_vbmeta_device);
+
+	if (strcmp(avb_invalidate_on_error, "yes") != 0) {
+		DMINFO("Not configured to invalidate");
+		return;
+	}
+
+	if (avb_vbmeta_device[0] == '\0') {
+		DMERR("avb_vbmeta_device parameter not set");
+		goto fail_no_dev;
+	}
+
+	dev = name_to_dev_t(avb_vbmeta_device);
+	if (!dev) {
+		DMERR("No matching partition for device: %s",
+		      avb_vbmeta_device);
+		goto fail_no_dev;
+	}
+
+	invalidate_vbmeta(dev);
+
+fail_no_dev:
+	;
+}
+
+static int __init dm_verity_avb_init(void)
+{
+	DMINFO("AVB error handler initialized with vbmeta device: %s",
+	       avb_vbmeta_device);
+	return 0;
+}
+
+static void __exit dm_verity_avb_exit(void)
+{
+}
+
+module_init(dm_verity_avb_init);
+module_exit(dm_verity_avb_exit);
+
+MODULE_AUTHOR("David Zeuthen <zeuthen@google.com>");
+MODULE_DESCRIPTION("AVB-specific error handler for dm-verity");
+MODULE_LICENSE("GPL");
+
+/* Declare parameter with no module prefix */
+#undef MODULE_PARAM_PREFIX
+#define MODULE_PARAM_PREFIX	"androidboot.vbmeta."
+module_param_string(device, avb_vbmeta_device, sizeof(avb_vbmeta_device), 0);
+module_param_string(invalidate_on_error, avb_invalidate_on_error,
+		    sizeof(avb_invalidate_on_error), 0);
diff --git a/drivers/md/dm-verity-fec.c b/drivers/md/dm-verity-fec.c
new file mode 100644
index 000000000000..1dd667b97530
--- /dev/null
+++ b/drivers/md/dm-verity-fec.c
@@ -0,0 +1,870 @@
+/*
+ * Copyright (C) 2015 Google, Inc.
+ *
+ * Author: Sami Tolvanen <samitolvanen@google.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ */
+
+#include "dm-verity-fec.h"
+#include <linux/math64.h>
+#include <linux/sysfs.h>
+
+#define DM_MSG_PREFIX	"verity-fec"
+
+/*
+ * If error correction has been configured, returns true.
+ */
+bool verity_fec_is_enabled(struct dm_verity *v)
+{
+	return v->fec && v->fec->dev;
+}
+
+/*
+ * Return a pointer to dm_verity_fec_io after dm_verity_io and its variable
+ * length fields.
+ */
+static inline struct dm_verity_fec_io *fec_io(struct dm_verity_io *io)
+{
+	return (struct dm_verity_fec_io *) verity_io_digest_end(io->v, io);
+}
+
+/*
+ * Return an interleaved offset for a byte in RS block.
+ */
+static inline u64 fec_interleave(struct dm_verity *v, u64 offset)
+{
+	u32 mod;
+
+	mod = do_div(offset, v->fec->rsn);
+	return offset + mod * (v->fec->rounds << v->data_dev_block_bits);
+}
+
+/*
+ * Decode an RS block using Reed-Solomon.
+ */
+static int fec_decode_rs8(struct dm_verity *v, struct dm_verity_fec_io *fio,
+			  u8 *data, u8 *fec, int neras)
+{
+	int i;
+	uint16_t par[DM_VERITY_FEC_RSM - DM_VERITY_FEC_MIN_RSN];
+
+	for (i = 0; i < v->fec->roots; i++)
+		par[i] = fec[i];
+
+	return decode_rs8(fio->rs, data, par, v->fec->rsn, NULL, neras,
+			  fio->erasures, 0, NULL);
+}
+
+/*
+ * Read error-correcting codes for the requested RS block. Returns a pointer
+ * to the data block. Caller is responsible for releasing buf.
+ */
+static u8 *fec_read_parity(struct dm_verity *v, u64 rsb, int index,
+			   unsigned *offset, struct dm_buffer **buf)
+{
+	u64 position, block;
+	u8 *res;
+
+	position = (index + rsb) * v->fec->roots;
+	block = position >> v->data_dev_block_bits;
+	*offset = (unsigned)(position - (block << v->data_dev_block_bits));
+
+	res = dm_bufio_read(v->fec->bufio, v->fec->start + block, buf);
+	if (unlikely(IS_ERR(res))) {
+		DMERR("%s: FEC %llu: parity read failed (block %llu): %ld",
+		      v->data_dev->name, (unsigned long long)rsb,
+		      (unsigned long long)(v->fec->start + block),
+		      PTR_ERR(res));
+		*buf = NULL;
+	}
+
+	return res;
+}
+
+/* Loop over each preallocated buffer slot. */
+#define fec_for_each_prealloc_buffer(__i) \
+	for (__i = 0; __i < DM_VERITY_FEC_BUF_PREALLOC; __i++)
+
+/* Loop over each extra buffer slot. */
+#define fec_for_each_extra_buffer(io, __i) \
+	for (__i = DM_VERITY_FEC_BUF_PREALLOC; __i < DM_VERITY_FEC_BUF_MAX; __i++)
+
+/* Loop over each allocated buffer. */
+#define fec_for_each_buffer(io, __i) \
+	for (__i = 0; __i < (io)->nbufs; __i++)
+
+/* Loop over each RS block in each allocated buffer. */
+#define fec_for_each_buffer_rs_block(io, __i, __j) \
+	fec_for_each_buffer(io, __i) \
+		for (__j = 0; __j < 1 << DM_VERITY_FEC_BUF_RS_BITS; __j++)
+
+/*
+ * Return a pointer to the current RS block when called inside
+ * fec_for_each_buffer_rs_block.
+ */
+static inline u8 *fec_buffer_rs_block(struct dm_verity *v,
+				      struct dm_verity_fec_io *fio,
+				      unsigned i, unsigned j)
+{
+	return &fio->bufs[i][j * v->fec->rsn];
+}
+
+/*
+ * Return an index to the current RS block when called inside
+ * fec_for_each_buffer_rs_block.
+ */
+static inline unsigned fec_buffer_rs_index(unsigned i, unsigned j)
+{
+	return (i << DM_VERITY_FEC_BUF_RS_BITS) + j;
+}
+
+/*
+ * Decode all RS blocks from buffers and copy corrected bytes into fio->output
+ * starting from block_offset.
+ */
+static int fec_decode_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio,
+			   u64 rsb, int byte_index, unsigned block_offset,
+			   int neras)
+{
+	int r, corrected = 0, res;
+	struct dm_buffer *buf;
+	unsigned n, i, offset;
+	u8 *par, *block;
+
+	par = fec_read_parity(v, rsb, block_offset, &offset, &buf);
+	if (IS_ERR(par))
+		return PTR_ERR(par);
+
+	/*
+	 * Decode the RS blocks we have in bufs. Each RS block results in
+	 * one corrected target byte and consumes fec->roots parity bytes.
+	 */
+	fec_for_each_buffer_rs_block(fio, n, i) {
+		block = fec_buffer_rs_block(v, fio, n, i);
+		res = fec_decode_rs8(v, fio, block, &par[offset], neras);
+		if (res < 0) {
+			dm_bufio_release(buf);
+
+			r = res;
+			goto error;
+		}
+
+		corrected += res;
+		fio->output[block_offset] = block[byte_index];
+
+		block_offset++;
+		if (block_offset >= 1 << v->data_dev_block_bits)
+			goto done;
+
+		/* read the next block when we run out of parity bytes */
+		offset += v->fec->roots;
+		if (offset >= 1 << v->data_dev_block_bits) {
+			dm_bufio_release(buf);
+
+			par = fec_read_parity(v, rsb, block_offset, &offset, &buf);
+			if (unlikely(IS_ERR(par)))
+				return PTR_ERR(par);
+		}
+	}
+done:
+	r = corrected;
+error:
+	if (r < 0 && neras)
+		DMERR_LIMIT("%s: FEC %llu: failed to correct: %d",
+			    v->data_dev->name, (unsigned long long)rsb, r);
+	else if (r > 0) {
+		DMWARN_LIMIT("%s: FEC %llu: corrected %d errors",
+			     v->data_dev->name, (unsigned long long)rsb, r);
+		atomic_add_unless(&v->fec->corrected, 1, INT_MAX);
+	}
+
+	return r;
+}
+
+/*
+ * Locate data block erasures using verity hashes.
+ */
+static int fec_is_erasure(struct dm_verity *v, struct dm_verity_io *io,
+			  u8 *want_digest, u8 *data)
+{
+	if (unlikely(verity_hash(v, verity_io_hash_desc(v, io),
+				 data, 1 << v->data_dev_block_bits,
+				 verity_io_real_digest(v, io))))
+		return 0;
+
+	return memcmp(verity_io_real_digest(v, io), want_digest,
+		      v->digest_size) != 0;
+}
+
+/*
+ * Read data blocks that are part of the RS block and deinterleave as much as
+ * fits into buffers. Check for erasure locations if @neras is non-NULL.
+ */
+static int fec_read_bufs(struct dm_verity *v, struct dm_verity_io *io,
+			 u64 rsb, u64 target, unsigned block_offset,
+			 int *neras)
+{
+	bool is_zero;
+	int i, j, target_index = -1;
+	struct dm_buffer *buf;
+	struct dm_bufio_client *bufio;
+	struct dm_verity_fec_io *fio = fec_io(io);
+	u64 block, ileaved;
+	u8 *bbuf, *rs_block;
+	u8 want_digest[v->digest_size];
+	unsigned n, k;
+
+	if (neras)
+		*neras = 0;
+
+	/*
+	 * read each of the rsn data blocks that are part of the RS block, and
+	 * interleave contents to available bufs
+	 */
+	for (i = 0; i < v->fec->rsn; i++) {
+		ileaved = fec_interleave(v, rsb * v->fec->rsn + i);
+
+		/*
+		 * target is the data block we want to correct, target_index is
+		 * the index of this block within the rsn RS blocks
+		 */
+		if (ileaved == target)
+			target_index = i;
+
+		block = ileaved >> v->data_dev_block_bits;
+		bufio = v->fec->data_bufio;
+
+		if (block >= v->data_blocks) {
+			block -= v->data_blocks;
+
+			/*
+			 * blocks outside the area were assumed to contain
+			 * zeros when encoding data was generated
+			 */
+			if (unlikely(block >= v->fec->hash_blocks))
+				continue;
+
+			block += v->hash_start;
+			bufio = v->bufio;
+		}
+
+		bbuf = dm_bufio_read(bufio, block, &buf);
+		if (unlikely(IS_ERR(bbuf))) {
+			DMWARN_LIMIT("%s: FEC %llu: read failed (%llu): %ld",
+				     v->data_dev->name,
+				     (unsigned long long)rsb,
+				     (unsigned long long)block, PTR_ERR(bbuf));
+
+			/* assume the block is corrupted */
+			if (neras && *neras <= v->fec->roots)
+				fio->erasures[(*neras)++] = i;
+
+			continue;
+		}
+
+		/* locate erasures if the block is on the data device */
+		if (bufio == v->fec->data_bufio &&
+		    verity_hash_for_block(v, io, block, want_digest,
+					  &is_zero) == 0) {
+			/* skip known zero blocks entirely */
+			if (is_zero)
+				continue;
+
+			/*
+			 * skip if we have already found the theoretical
+			 * maximum number (i.e. fec->roots) of erasures
+			 */
+			if (neras && *neras <= v->fec->roots &&
+			    fec_is_erasure(v, io, want_digest, bbuf))
+				fio->erasures[(*neras)++] = i;
+		}
+
+		/*
+		 * deinterleave and copy the bytes that fit into bufs,
+		 * starting from block_offset
+		 */
+		fec_for_each_buffer_rs_block(fio, n, j) {
+			k = fec_buffer_rs_index(n, j) + block_offset;
+
+			if (k >= 1 << v->data_dev_block_bits)
+				goto done;
+
+			rs_block = fec_buffer_rs_block(v, fio, n, j);
+			rs_block[i] = bbuf[k];
+		}
+done:
+		dm_bufio_release(buf);
+	}
+
+	return target_index;
+}
+
+/*
+ * Allocate RS control structure and FEC buffers from preallocated mempools,
+ * and attempt to allocate as many extra buffers as available.
+ */
+static int fec_alloc_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio)
+{
+	unsigned n;
+
+	if (!fio->rs) {
+		fio->rs = mempool_alloc(v->fec->rs_pool, 0);
+		if (unlikely(!fio->rs)) {
+			DMERR("failed to allocate RS");
+			return -ENOMEM;
+		}
+	}
+
+	fec_for_each_prealloc_buffer(n) {
+		if (fio->bufs[n])
+			continue;
+
+		fio->bufs[n] = mempool_alloc(v->fec->prealloc_pool, GFP_NOIO);
+		if (unlikely(!fio->bufs[n])) {
+			DMERR("failed to allocate FEC buffer");
+			return -ENOMEM;
+		}
+	}
+
+	/* try to allocate the maximum number of buffers */
+	fec_for_each_extra_buffer(fio, n) {
+		if (fio->bufs[n])
+			continue;
+
+		fio->bufs[n] = mempool_alloc(v->fec->extra_pool, GFP_NOIO);
+		/* we can manage with even one buffer if necessary */
+		if (unlikely(!fio->bufs[n]))
+			break;
+	}
+	fio->nbufs = n;
+
+	if (!fio->output) {
+		fio->output = mempool_alloc(v->fec->output_pool, GFP_NOIO);
+
+		if (!fio->output) {
+			DMERR("failed to allocate FEC page");
+			return -ENOMEM;
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * Initialize buffers and clear erasures. fec_read_bufs() assumes buffers are
+ * zeroed before deinterleaving.
+ */
+static void fec_init_bufs(struct dm_verity *v, struct dm_verity_fec_io *fio)
+{
+	unsigned n;
+
+	fec_for_each_buffer(fio, n)
+		memset(fio->bufs[n], 0, v->fec->rsn << DM_VERITY_FEC_BUF_RS_BITS);
+
+	memset(fio->erasures, 0, sizeof(fio->erasures));
+}
+
+/*
+ * Decode all RS blocks in a single data block and return the target block
+ * (indicated by @offset) in fio->output. If @use_erasures is non-zero, uses
+ * hashes to locate erasures.
+ */
+static int fec_decode_rsb(struct dm_verity *v, struct dm_verity_io *io,
+			  struct dm_verity_fec_io *fio, u64 rsb, u64 offset,
+			  bool use_erasures)
+{
+	int r, neras = 0;
+	unsigned pos;
+
+	r = fec_alloc_bufs(v, fio);
+	if (unlikely(r < 0))
+		return r;
+
+	for (pos = 0; pos < 1 << v->data_dev_block_bits; ) {
+		fec_init_bufs(v, fio);
+
+		r = fec_read_bufs(v, io, rsb, offset, pos,
+				  use_erasures ? &neras : NULL);
+		if (unlikely(r < 0))
+			return r;
+
+		r = fec_decode_bufs(v, fio, rsb, r, pos, neras);
+		if (r < 0)
+			return r;
+
+		pos += fio->nbufs << DM_VERITY_FEC_BUF_RS_BITS;
+	}
+
+	/* Always re-validate the corrected block against the expected hash */
+	r = verity_hash(v, verity_io_hash_desc(v, io), fio->output,
+			1 << v->data_dev_block_bits,
+			verity_io_real_digest(v, io));
+	if (unlikely(r < 0))
+		return r;
+
+	if (memcmp(verity_io_real_digest(v, io), verity_io_want_digest(v, io),
+		   v->digest_size)) {
+		DMERR_LIMIT("%s: FEC %llu: failed to correct (%d erasures)",
+			    v->data_dev->name, (unsigned long long)rsb, neras);
+		return -EILSEQ;
+	}
+
+	return 0;
+}
+
+static int fec_bv_copy(struct dm_verity *v, struct dm_verity_io *io, u8 *data,
+		       size_t len)
+{
+	struct dm_verity_fec_io *fio = fec_io(io);
+
+	memcpy(data, &fio->output[fio->output_pos], len);
+	fio->output_pos += len;
+
+	return 0;
+}
+
+/*
+ * Correct errors in a block. Copies corrected block to dest if non-NULL,
+ * otherwise to a bio_vec starting from iter.
+ */
+int verity_fec_decode(struct dm_verity *v, struct dm_verity_io *io,
+		      enum verity_block_type type, sector_t block, u8 *dest,
+		      struct bvec_iter *iter)
+{
+	int r;
+	struct dm_verity_fec_io *fio = fec_io(io);
+	u64 offset, res, rsb;
+
+	if (!verity_fec_is_enabled(v))
+		return -EOPNOTSUPP;
+
+	if (fio->level >= DM_VERITY_FEC_MAX_RECURSION) {
+		DMWARN_LIMIT("%s: FEC: recursion too deep", v->data_dev->name);
+		return -EIO;
+	}
+
+	fio->level++;
+
+	if (type == DM_VERITY_BLOCK_TYPE_METADATA)
+		block += v->data_blocks;
+
+	/*
+	 * For RS(M, N), the continuous FEC data is divided into blocks of N
+	 * bytes. Since block size may not be divisible by N, the last block
+	 * is zero padded when decoding.
+	 *
+	 * Each byte of the block is covered by a different RS(M, N) code,
+	 * and each code is interleaved over N blocks to make it less likely
+	 * that bursty corruption will leave us in unrecoverable state.
+	 */
+
+	offset = block << v->data_dev_block_bits;
+	res = div64_u64(offset, v->fec->rounds << v->data_dev_block_bits);
+
+	/*
+	 * The base RS block we can feed to the interleaver to find out all
+	 * blocks required for decoding.
+	 */
+	rsb = offset - res * (v->fec->rounds << v->data_dev_block_bits);
+
+	/*
+	 * Locating erasures is slow, so attempt to recover the block without
+	 * them first. Do a second attempt with erasures if the corruption is
+	 * bad enough.
+	 */
+	r = fec_decode_rsb(v, io, fio, rsb, offset, false);
+	if (r < 0) {
+		r = fec_decode_rsb(v, io, fio, rsb, offset, true);
+		if (r < 0)
+			goto done;
+	}
+
+	if (dest)
+		memcpy(dest, fio->output, 1 << v->data_dev_block_bits);
+	else if (iter) {
+		fio->output_pos = 0;
+		r = verity_for_bv_block(v, io, iter, fec_bv_copy);
+	}
+
+done:
+	fio->level--;
+	return r;
+}
+
+/*
+ * Clean up per-bio data.
+ */
+void verity_fec_finish_io(struct dm_verity_io *io)
+{
+	unsigned n;
+	struct dm_verity_fec *f = io->v->fec;
+	struct dm_verity_fec_io *fio = fec_io(io);
+
+	if (!verity_fec_is_enabled(io->v))
+		return;
+
+	mempool_free(fio->rs, f->rs_pool);
+
+	fec_for_each_prealloc_buffer(n)
+		mempool_free(fio->bufs[n], f->prealloc_pool);
+
+	fec_for_each_extra_buffer(fio, n)
+		mempool_free(fio->bufs[n], f->extra_pool);
+
+	mempool_free(fio->output, f->output_pool);
+}
+
+/*
+ * Initialize per-bio data.
+ */
+void verity_fec_init_io(struct dm_verity_io *io)
+{
+	struct dm_verity_fec_io *fio = fec_io(io);
+
+	if (!verity_fec_is_enabled(io->v))
+		return;
+
+	fio->rs = NULL;
+	memset(fio->bufs, 0, sizeof(fio->bufs));
+	fio->nbufs = 0;
+	fio->output = NULL;
+	fio->level = 0;
+}
+
+/*
+ * Append feature arguments and values to the status table.
+ */
+unsigned verity_fec_status_table(struct dm_verity *v, unsigned sz,
+				 char *result, unsigned maxlen)
+{
+	if (!verity_fec_is_enabled(v))
+		return sz;
+
+	DMEMIT(" " DM_VERITY_OPT_FEC_DEV " %s "
+	       DM_VERITY_OPT_FEC_BLOCKS " %llu "
+	       DM_VERITY_OPT_FEC_START " %llu "
+	       DM_VERITY_OPT_FEC_ROOTS " %d",
+	       v->fec->dev->name,
+	       (unsigned long long)v->fec->blocks,
+	       (unsigned long long)v->fec->start,
+	       v->fec->roots);
+
+	return sz;
+}
+
+void verity_fec_dtr(struct dm_verity *v)
+{
+	struct dm_verity_fec *f = v->fec;
+	struct kobject *kobj = &f->kobj_holder.kobj;
+
+	if (!verity_fec_is_enabled(v))
+		goto out;
+
+	mempool_destroy(f->rs_pool);
+	mempool_destroy(f->prealloc_pool);
+	mempool_destroy(f->extra_pool);
+	kmem_cache_destroy(f->cache);
+
+	if (f->data_bufio)
+		dm_bufio_client_destroy(f->data_bufio);
+	if (f->bufio)
+		dm_bufio_client_destroy(f->bufio);
+
+	if (f->dev)
+		dm_put_device(v->ti, f->dev);
+
+	if (kobj->state_initialized) {
+		kobject_put(kobj);
+		wait_for_completion(dm_get_completion_from_kobject(kobj));
+	}
+
+out:
+	kfree(f);
+	v->fec = NULL;
+}
+
+static void *fec_rs_alloc(gfp_t gfp_mask, void *pool_data)
+{
+	struct dm_verity *v = (struct dm_verity *)pool_data;
+
+	return init_rs(8, 0x11d, 0, 1, v->fec->roots);
+}
+
+static void fec_rs_free(void *element, void *pool_data)
+{
+	struct rs_control *rs = (struct rs_control *)element;
+
+	if (rs)
+		free_rs(rs);
+}
+
+bool verity_is_fec_opt_arg(const char *arg_name)
+{
+	return (!strcasecmp(arg_name, DM_VERITY_OPT_FEC_DEV) ||
+		!strcasecmp(arg_name, DM_VERITY_OPT_FEC_BLOCKS) ||
+		!strcasecmp(arg_name, DM_VERITY_OPT_FEC_START) ||
+		!strcasecmp(arg_name, DM_VERITY_OPT_FEC_ROOTS));
+}
+
+int verity_fec_parse_opt_args(struct dm_arg_set *as, struct dm_verity *v,
+			      unsigned *argc, const char *arg_name)
+{
+	int r;
+	struct dm_target *ti = v->ti;
+	const char *arg_value;
+	unsigned long long num_ll;
+	unsigned char num_c;
+	char dummy;
+
+	if (!*argc) {
+		ti->error = "FEC feature arguments require a value";
+		return -EINVAL;
+	}
+
+	arg_value = dm_shift_arg(as);
+	(*argc)--;
+
+	if (!strcasecmp(arg_name, DM_VERITY_OPT_FEC_DEV)) {
+		r = dm_get_device(ti, arg_value, FMODE_READ, &v->fec->dev);
+		if (r) {
+			ti->error = "FEC device lookup failed";
+			return r;
+		}
+
+	} else if (!strcasecmp(arg_name, DM_VERITY_OPT_FEC_BLOCKS)) {
+		if (sscanf(arg_value, "%llu%c", &num_ll, &dummy) != 1 ||
+		    ((sector_t)(num_ll << (v->data_dev_block_bits - SECTOR_SHIFT))
+		     >> (v->data_dev_block_bits - SECTOR_SHIFT) != num_ll)) {
+			ti->error = "Invalid " DM_VERITY_OPT_FEC_BLOCKS;
+			return -EINVAL;
+		}
+		v->fec->blocks = num_ll;
+
+	} else if (!strcasecmp(arg_name, DM_VERITY_OPT_FEC_START)) {
+		if (sscanf(arg_value, "%llu%c", &num_ll, &dummy) != 1 ||
+		    ((sector_t)(num_ll << (v->data_dev_block_bits - SECTOR_SHIFT)) >>
+		     (v->data_dev_block_bits - SECTOR_SHIFT) != num_ll)) {
+			ti->error = "Invalid " DM_VERITY_OPT_FEC_START;
+			return -EINVAL;
+		}
+		v->fec->start = num_ll;
+
+	} else if (!strcasecmp(arg_name, DM_VERITY_OPT_FEC_ROOTS)) {
+		if (sscanf(arg_value, "%hhu%c", &num_c, &dummy) != 1 || !num_c ||
+		    num_c < (DM_VERITY_FEC_RSM - DM_VERITY_FEC_MAX_RSN) ||
+		    num_c > (DM_VERITY_FEC_RSM - DM_VERITY_FEC_MIN_RSN)) {
+			ti->error = "Invalid " DM_VERITY_OPT_FEC_ROOTS;
+			return -EINVAL;
+		}
+		v->fec->roots = num_c;
+
+	} else {
+		ti->error = "Unrecognized verity FEC feature request";
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static ssize_t corrected_show(struct kobject *kobj, struct kobj_attribute *attr,
+			      char *buf)
+{
+	struct dm_verity_fec *f = container_of(kobj, struct dm_verity_fec,
+					       kobj_holder.kobj);
+
+	return sprintf(buf, "%d\n", atomic_read(&f->corrected));
+}
+
+static struct kobj_attribute attr_corrected = __ATTR_RO(corrected);
+
+static struct attribute *fec_attrs[] = {
+	&attr_corrected.attr,
+	NULL
+};
+
+static struct kobj_type fec_ktype = {
+	.sysfs_ops = &kobj_sysfs_ops,
+	.default_attrs = fec_attrs,
+	.release = dm_kobject_release
+};
+
+/*
+ * Allocate dm_verity_fec for v->fec. Must be called before verity_fec_ctr.
+ */
+int verity_fec_ctr_alloc(struct dm_verity *v)
+{
+	struct dm_verity_fec *f;
+
+	f = kzalloc(sizeof(struct dm_verity_fec), GFP_KERNEL);
+	if (!f) {
+		v->ti->error = "Cannot allocate FEC structure";
+		return -ENOMEM;
+	}
+	v->fec = f;
+
+	return 0;
+}
+
+/*
+ * Validate arguments and preallocate memory. Must be called after arguments
+ * have been parsed using verity_fec_parse_opt_args.
+ */
+int verity_fec_ctr(struct dm_verity *v)
+{
+	int r;
+	struct dm_verity_fec *f = v->fec;
+	struct dm_target *ti = v->ti;
+	struct mapped_device *md = dm_table_get_md(ti->table);
+	u64 hash_blocks;
+
+	if (!verity_fec_is_enabled(v)) {
+		verity_fec_dtr(v);
+		return 0;
+	}
+
+	/* Create a kobject and sysfs attributes */
+	init_completion(&f->kobj_holder.completion);
+
+	r = kobject_init_and_add(&f->kobj_holder.kobj, &fec_ktype,
+				 &disk_to_dev(dm_disk(md))->kobj, "%s", "fec");
+	if (r) {
+		ti->error = "Cannot create kobject";
+		return r;
+	}
+
+	/*
+	 * FEC is computed over data blocks, possible metadata, and
+	 * hash blocks. In other words, FEC covers total of fec_blocks
+	 * blocks consisting of the following:
+	 *
+	 *  data blocks | hash blocks | metadata (optional)
+	 *
+	 * We allow metadata after hash blocks to support a use case
+	 * where all data is stored on the same device and FEC covers
+	 * the entire area.
+	 *
+	 * If metadata is included, we require it to be available on the
+	 * hash device after the hash blocks.
+	 */
+
+	hash_blocks = v->hash_blocks - v->hash_start;
+
+	/*
+	 * Require matching block sizes for data and hash devices for
+	 * simplicity.
+	 */
+	if (v->data_dev_block_bits != v->hash_dev_block_bits) {
+		ti->error = "Block sizes must match to use FEC";
+		return -EINVAL;
+	}
+
+	if (!f->roots) {
+		ti->error = "Missing " DM_VERITY_OPT_FEC_ROOTS;
+		return -EINVAL;
+	}
+	f->rsn = DM_VERITY_FEC_RSM - f->roots;
+
+	if (!f->blocks) {
+		ti->error = "Missing " DM_VERITY_OPT_FEC_BLOCKS;
+		return -EINVAL;
+	}
+
+	f->rounds = f->blocks;
+	if (sector_div(f->rounds, f->rsn))
+		f->rounds++;
+
+	/*
+	 * Due to optional metadata, f->blocks can be larger than
+	 * data_blocks and hash_blocks combined.
+	 */
+	if (f->blocks < v->data_blocks + hash_blocks || !f->rounds) {
+		ti->error = "Invalid " DM_VERITY_OPT_FEC_BLOCKS;
+		return -EINVAL;
+	}
+
+	/*
+	 * Metadata is accessed through the hash device, so we require
+	 * it to be large enough.
+	 */
+	f->hash_blocks = f->blocks - v->data_blocks;
+	if (dm_bufio_get_device_size(v->bufio) < f->hash_blocks) {
+		ti->error = "Hash device is too small for "
+			DM_VERITY_OPT_FEC_BLOCKS;
+		return -E2BIG;
+	}
+
+	f->bufio = dm_bufio_client_create(f->dev->bdev,
+					  1 << v->data_dev_block_bits,
+					  1, 0, NULL, NULL);
+	if (IS_ERR(f->bufio)) {
+		ti->error = "Cannot initialize FEC bufio client";
+		return PTR_ERR(f->bufio);
+	}
+
+	if (dm_bufio_get_device_size(f->bufio) <
+	    ((f->start + f->rounds * f->roots) >> v->data_dev_block_bits)) {
+		ti->error = "FEC device is too small";
+		return -E2BIG;
+	}
+
+	f->data_bufio = dm_bufio_client_create(v->data_dev->bdev,
+					       1 << v->data_dev_block_bits,
+					       1, 0, NULL, NULL);
+	if (IS_ERR(f->data_bufio)) {
+		ti->error = "Cannot initialize FEC data bufio client";
+		return PTR_ERR(f->data_bufio);
+	}
+
+	if (dm_bufio_get_device_size(f->data_bufio) < v->data_blocks) {
+		ti->error = "Data device is too small";
+		return -E2BIG;
+	}
+
+	/* Preallocate an rs_control structure for each worker thread */
+	f->rs_pool = mempool_create(num_online_cpus(), fec_rs_alloc,
+				    fec_rs_free, (void *) v);
+	if (!f->rs_pool) {
+		ti->error = "Cannot allocate RS pool";
+		return -ENOMEM;
+	}
+
+	f->cache = kmem_cache_create("dm_verity_fec_buffers",
+				     f->rsn << DM_VERITY_FEC_BUF_RS_BITS,
+				     0, 0, NULL);
+	if (!f->cache) {
+		ti->error = "Cannot create FEC buffer cache";
+		return -ENOMEM;
+	}
+
+	/* Preallocate DM_VERITY_FEC_BUF_PREALLOC buffers for each thread */
+	f->prealloc_pool = mempool_create_slab_pool(num_online_cpus() *
+						    DM_VERITY_FEC_BUF_PREALLOC,
+						    f->cache);
+	if (!f->prealloc_pool) {
+		ti->error = "Cannot allocate FEC buffer prealloc pool";
+		return -ENOMEM;
+	}
+
+	f->extra_pool = mempool_create_slab_pool(0, f->cache);
+	if (!f->extra_pool) {
+		ti->error = "Cannot allocate FEC buffer extra pool";
+		return -ENOMEM;
+	}
+
+	/* Preallocate an output buffer for each thread */
+	f->output_pool = mempool_create_kmalloc_pool(num_online_cpus(),
+						     1 << v->data_dev_block_bits);
+	if (!f->output_pool) {
+		ti->error = "Cannot allocate FEC output pool";
+		return -ENOMEM;
+	}
+
+	/* Reserve space for our per-bio data */
+	ti->per_bio_data_size += sizeof(struct dm_verity_fec_io);
+
+	return 0;
+}
diff --git a/drivers/md/dm-verity-fec.h b/drivers/md/dm-verity-fec.h
new file mode 100644
index 000000000000..b8e21cef3ad1
--- /dev/null
+++ b/drivers/md/dm-verity-fec.h
@@ -0,0 +1,159 @@
+/*
+ * Copyright (C) 2015 Google, Inc.
+ *
+ * Author: Sami Tolvanen <samitolvanen@google.com>
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ */
+
+#ifndef DM_VERITY_FEC_H
+#define DM_VERITY_FEC_H
+
+#include "dm.h"
+#include "dm-verity.h"
+#include <linux/rslib.h>
+
+/* Reed-Solomon(M, N) parameters */
+#define DM_VERITY_FEC_RSM		255
+#define DM_VERITY_FEC_MAX_RSN		253
+#define DM_VERITY_FEC_MIN_RSN		231	/* ~10% space overhead */
+
+/* buffers for deinterleaving and decoding */
+#define DM_VERITY_FEC_BUF_PREALLOC	1	/* buffers to preallocate */
+#define DM_VERITY_FEC_BUF_RS_BITS	4	/* 1 << RS blocks per buffer */
+/* we need buffers for at most 1 << block size RS blocks */
+#define DM_VERITY_FEC_BUF_MAX \
+	(1 << (PAGE_SHIFT - DM_VERITY_FEC_BUF_RS_BITS))
+
+/* maximum recursion level for verity_fec_decode */
+#define DM_VERITY_FEC_MAX_RECURSION	4
+
+#define DM_VERITY_OPT_FEC_DEV		"use_fec_from_device"
+#define DM_VERITY_OPT_FEC_BLOCKS	"fec_blocks"
+#define DM_VERITY_OPT_FEC_START		"fec_start"
+#define DM_VERITY_OPT_FEC_ROOTS		"fec_roots"
+
+/* configuration */
+struct dm_verity_fec {
+	struct dm_dev *dev;	/* parity data device */
+	struct dm_bufio_client *data_bufio;	/* for data dev access */
+	struct dm_bufio_client *bufio;		/* for parity data access */
+	sector_t start;		/* parity data start in blocks */
+	sector_t blocks;	/* number of blocks covered */
+	sector_t rounds;	/* number of interleaving rounds */
+	sector_t hash_blocks;	/* blocks covered after v->hash_start */
+	unsigned char roots;	/* number of parity bytes, M-N of RS(M, N) */
+	unsigned char rsn;	/* N of RS(M, N) */
+	mempool_t *rs_pool;	/* mempool for fio->rs */
+	mempool_t *prealloc_pool;	/* mempool for preallocated buffers */
+	mempool_t *extra_pool;	/* mempool for extra buffers */
+	mempool_t *output_pool;	/* mempool for output */
+	struct kmem_cache *cache;	/* cache for buffers */
+	atomic_t corrected;		/* corrected errors */
+	struct dm_kobject_holder kobj_holder;	/* for sysfs attributes */
+};
+
+/* per-bio data */
+struct dm_verity_fec_io {
+	struct rs_control *rs;	/* Reed-Solomon state */
+	int erasures[DM_VERITY_FEC_MAX_RSN];	/* erasures for decode_rs8 */
+	u8 *bufs[DM_VERITY_FEC_BUF_MAX];	/* bufs for deinterleaving */
+	unsigned nbufs;		/* number of buffers allocated */
+	u8 *output;		/* buffer for corrected output */
+	size_t output_pos;
+	unsigned level;		/* recursion level */
+};
+
+#ifdef CONFIG_DM_VERITY_FEC
+
+/* each feature parameter requires a value */
+#define DM_VERITY_OPTS_FEC	8
+
+extern bool verity_fec_is_enabled(struct dm_verity *v);
+
+extern int verity_fec_decode(struct dm_verity *v, struct dm_verity_io *io,
+			     enum verity_block_type type, sector_t block,
+			     u8 *dest, struct bvec_iter *iter);
+
+extern unsigned verity_fec_status_table(struct dm_verity *v, unsigned sz,
+					char *result, unsigned maxlen);
+
+extern void verity_fec_finish_io(struct dm_verity_io *io);
+extern void verity_fec_init_io(struct dm_verity_io *io);
+
+extern bool verity_is_fec_opt_arg(const char *arg_name);
+extern int verity_fec_parse_opt_args(struct dm_arg_set *as,
+				     struct dm_verity *v, unsigned *argc,
+				     const char *arg_name);
+
+extern void verity_fec_dtr(struct dm_verity *v);
+
+extern int verity_fec_ctr_alloc(struct dm_verity *v);
+extern int verity_fec_ctr(struct dm_verity *v);
+
+#else /* !CONFIG_DM_VERITY_FEC */
+
+#define DM_VERITY_OPTS_FEC	0
+
+static inline bool verity_fec_is_enabled(struct dm_verity *v)
+{
+	return false;
+}
+
+static inline int verity_fec_decode(struct dm_verity *v,
+				    struct dm_verity_io *io,
+				    enum verity_block_type type,
+				    sector_t block, u8 *dest,
+				    struct bvec_iter *iter)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline unsigned verity_fec_status_table(struct dm_verity *v,
+					       unsigned sz, char *result,
+					       unsigned maxlen)
+{
+	return sz;
+}
+
+static inline void verity_fec_finish_io(struct dm_verity_io *io)
+{
+}
+
+static inline void verity_fec_init_io(struct dm_verity_io *io)
+{
+}
+
+static inline bool verity_is_fec_opt_arg(const char *arg_name)
+{
+	return false;
+}
+
+static inline int verity_fec_parse_opt_args(struct dm_arg_set *as,
+					    struct dm_verity *v,
+					    unsigned *argc,
+					    const char *arg_name)
+{
+	return -EINVAL;
+}
+
+static inline void verity_fec_dtr(struct dm_verity *v)
+{
+}
+
+static inline int verity_fec_ctr_alloc(struct dm_verity *v)
+{
+	return 0;
+}
+
+static inline int verity_fec_ctr(struct dm_verity *v)
+{
+	return 0;
+}
+
+#endif /* CONFIG_DM_VERITY_FEC */
+
+#endif /* DM_VERITY_FEC_H */
diff --git a/drivers/md/dm-verity.c b/drivers/md/dm-verity-target.c
index ccf41886ebcf..e34cf53bd068 100644
--- a/drivers/md/dm-verity.c
+++ b/drivers/md/dm-verity-target.c
@@ -14,12 +14,11 @@
  * access behavior.
  */
 
-#include "dm-bufio.h"
+#include "dm-verity.h"
+#include "dm-verity-fec.h"
 
 #include <linux/module.h>
-#include <linux/device-mapper.h>
 #include <linux/reboot.h>
-#include <crypto/hash.h>
 
 #define DM_MSG_PREFIX			"verity"
 
@@ -28,83 +27,18 @@
 
 #define DM_VERITY_DEFAULT_PREFETCH_SIZE	262144
 
-#define DM_VERITY_MAX_LEVELS		63
 #define DM_VERITY_MAX_CORRUPTED_ERRS	100
 
 #define DM_VERITY_OPT_LOGGING		"ignore_corruption"
 #define DM_VERITY_OPT_RESTART		"restart_on_corruption"
+#define DM_VERITY_OPT_IGN_ZEROES	"ignore_zero_blocks"
+
+#define DM_VERITY_OPTS_MAX		(2 + DM_VERITY_OPTS_FEC)
 
 static unsigned dm_verity_prefetch_cluster = DM_VERITY_DEFAULT_PREFETCH_SIZE;
 
 module_param_named(prefetch_cluster, dm_verity_prefetch_cluster, uint, S_IRUGO | S_IWUSR);
 
-enum verity_mode {
-	DM_VERITY_MODE_EIO,
-	DM_VERITY_MODE_LOGGING,
-	DM_VERITY_MODE_RESTART
-};
-
-enum verity_block_type {
-	DM_VERITY_BLOCK_TYPE_DATA,
-	DM_VERITY_BLOCK_TYPE_METADATA
-};
-
-struct dm_verity {
-	struct dm_dev *data_dev;
-	struct dm_dev *hash_dev;
-	struct dm_target *ti;
-	struct dm_bufio_client *bufio;
-	char *alg_name;
-	struct crypto_shash *tfm;
-	u8 *root_digest;	/* digest of the root block */
-	u8 *salt;		/* salt: its size is salt_size */
-	unsigned salt_size;
-	sector_t data_start;	/* data offset in 512-byte sectors */
-	sector_t hash_start;	/* hash start in blocks */
-	sector_t data_blocks;	/* the number of data blocks */
-	sector_t hash_blocks;	/* the number of hash blocks */
-	unsigned char data_dev_block_bits;	/* log2(data blocksize) */
-	unsigned char hash_dev_block_bits;	/* log2(hash blocksize) */
-	unsigned char hash_per_block_bits;	/* log2(hashes in hash block) */
-	unsigned char levels;	/* the number of tree levels */
-	unsigned char version;
-	unsigned digest_size;	/* digest size for the current hash algorithm */
-	unsigned shash_descsize;/* the size of temporary space for crypto */
-	int hash_failed;	/* set to 1 if hash of any block failed */
-	enum verity_mode mode;	/* mode for handling verification errors */
-	unsigned corrupted_errs;/* Number of errors for corrupted blocks */
-
-	struct workqueue_struct *verify_wq;
-
-	/* starting blocks for each tree level. 0 is the lowest level. */
-	sector_t hash_level_block[DM_VERITY_MAX_LEVELS];
-};
-
-struct dm_verity_io {
-	struct dm_verity *v;
-
-	/* original values of bio->bi_end_io and bio->bi_private */
-	bio_end_io_t *orig_bi_end_io;
-	void *orig_bi_private;
-
-	sector_t block;
-	unsigned n_blocks;
-
-	struct bvec_iter iter;
-
-	struct work_struct work;
-
-	/*
-	 * Three variably-size fields follow this struct:
-	 *
-	 * u8 hash_desc[v->shash_descsize];
-	 * u8 real_digest[v->digest_size];
-	 * u8 want_digest[v->digest_size];
-	 *
-	 * To access them use: io_hash_desc(), io_real_digest() and io_want_digest().
-	 */
-};
-
 struct dm_verity_prefetch_work {
 	struct work_struct work;
 	struct dm_verity *v;
@@ -112,21 +46,6 @@ struct dm_verity_prefetch_work {
 	unsigned n_blocks;
 };
 
-static struct shash_desc *io_hash_desc(struct dm_verity *v, struct dm_verity_io *io)
-{
-	return (struct shash_desc *)(io + 1);
-}
-
-static u8 *io_real_digest(struct dm_verity *v, struct dm_verity_io *io)
-{
-	return (u8 *)(io + 1) + v->shash_descsize;
-}
-
-static u8 *io_want_digest(struct dm_verity *v, struct dm_verity_io *io)
-{
-	return (u8 *)(io + 1) + v->shash_descsize + v->digest_size;
-}
-
 /*
  * Auxiliary structure appended to each dm-bufio buffer. If the value
  * hash_verified is nonzero, hash of the block has been verified.
@@ -173,6 +92,84 @@ static sector_t verity_position_at_level(struct dm_verity *v, sector_t block,
 	return block >> (level * v->hash_per_block_bits);
 }
 
+/*
+ * Wrapper for crypto_shash_init, which handles verity salting.
+ */
+static int verity_hash_init(struct dm_verity *v, struct shash_desc *desc)
+{
+	int r;
+
+	desc->tfm = v->tfm;
+	desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
+
+	r = crypto_shash_init(desc);
+
+	if (unlikely(r < 0)) {
+		DMERR("crypto_shash_init failed: %d", r);
+		return r;
+	}
+
+	if (likely(v->version >= 1)) {
+		r = crypto_shash_update(desc, v->salt, v->salt_size);
+
+		if (unlikely(r < 0)) {
+			DMERR("crypto_shash_update failed: %d", r);
+			return r;
+		}
+	}
+
+	return 0;
+}
+
+static int verity_hash_update(struct dm_verity *v, struct shash_desc *desc,
+			      const u8 *data, size_t len)
+{
+	int r = crypto_shash_update(desc, data, len);
+
+	if (unlikely(r < 0))
+		DMERR("crypto_shash_update failed: %d", r);
+
+	return r;
+}
+
+static int verity_hash_final(struct dm_verity *v, struct shash_desc *desc,
+			     u8 *digest)
+{
+	int r;
+
+	if (unlikely(!v->version)) {
+		r = crypto_shash_update(desc, v->salt, v->salt_size);
+
+		if (r < 0) {
+			DMERR("crypto_shash_update failed: %d", r);
+			return r;
+		}
+	}
+
+	r = crypto_shash_final(desc, digest);
+
+	if (unlikely(r < 0))
+		DMERR("crypto_shash_final failed: %d", r);
+
+	return r;
+}
+
+int verity_hash(struct dm_verity *v, struct shash_desc *desc,
+		const u8 *data, size_t len, u8 *digest)
+{
+	int r;
+
+	r = verity_hash_init(v, desc);
+	if (unlikely(r < 0))
+		return r;
+
+	r = verity_hash_update(v, desc, data, len);
+	if (unlikely(r < 0))
+		return r;
+
+	return verity_hash_final(v, desc, digest);
+}
+
 static void verity_hash_at_level(struct dm_verity *v, sector_t block, int level,
 				 sector_t *hash_block, unsigned *offset)
 {
@@ -236,8 +233,12 @@ out:
 	if (v->mode == DM_VERITY_MODE_LOGGING)
 		return 0;
 
-	if (v->mode == DM_VERITY_MODE_RESTART)
+	if (v->mode == DM_VERITY_MODE_RESTART) {
+#ifdef CONFIG_DM_VERITY_AVB
+		dm_verity_avb_error_handler();
+#endif
 		kernel_restart("dm-verity device corrupted");
+	}
 
 	return 1;
 }
@@ -246,17 +247,17 @@ out:
  * Verify hash of a metadata block pertaining to the specified data block
  * ("block" argument) at a specified level ("level" argument).
  *
- * On successful return, io_want_digest(v, io) contains the hash value for
- * a lower tree level or for the data block (if we're at the lowest leve).
+ * On successful return, verity_io_want_digest(v, io) contains the hash value
+ * for a lower tree level or for the data block (if we're at the lowest level).
  *
  * If "skip_unverified" is true, unverified buffer is skipped and 1 is returned.
  * If "skip_unverified" is false, unverified buffer is hashed and verified
- * against current value of io_want_digest(v, io).
+ * against current value of verity_io_want_digest(v, io).
  */
-static int verity_verify_level(struct dm_verity_io *io, sector_t block,
-			       int level, bool skip_unverified)
+static int verity_verify_level(struct dm_verity *v, struct dm_verity_io *io,
+			       sector_t block, int level, bool skip_unverified,
+			       u8 *want_digest)
 {
-	struct dm_verity *v = io->v;
 	struct dm_buffer *buf;
 	struct buffer_aux *aux;
 	u8 *data;
@@ -273,72 +274,128 @@ static int verity_verify_level(struct dm_verity_io *io, sector_t block,
 	aux = dm_bufio_get_aux_data(buf);
 
 	if (!aux->hash_verified) {
-		struct shash_desc *desc;
-		u8 *result;
-
 		if (skip_unverified) {
 			r = 1;
 			goto release_ret_r;
 		}
 
-		desc = io_hash_desc(v, io);
-		desc->tfm = v->tfm;
-		desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
-		r = crypto_shash_init(desc);
-		if (r < 0) {
-			DMERR("crypto_shash_init failed: %d", r);
+		r = verity_hash(v, verity_io_hash_desc(v, io),
+				data, 1 << v->hash_dev_block_bits,
+				verity_io_real_digest(v, io));
+		if (unlikely(r < 0))
 			goto release_ret_r;
-		}
 
-		if (likely(v->version >= 1)) {
-			r = crypto_shash_update(desc, v->salt, v->salt_size);
-			if (r < 0) {
-				DMERR("crypto_shash_update failed: %d", r);
-				goto release_ret_r;
-			}
-		}
-
-		r = crypto_shash_update(desc, data, 1 << v->hash_dev_block_bits);
-		if (r < 0) {
-			DMERR("crypto_shash_update failed: %d", r);
+		if (likely(memcmp(verity_io_real_digest(v, io), want_digest,
+				  v->digest_size) == 0))
+			aux->hash_verified = 1;
+		else if (verity_fec_decode(v, io,
+					   DM_VERITY_BLOCK_TYPE_METADATA,
+					   hash_block, data, NULL) == 0)
+			aux->hash_verified = 1;
+		else if (verity_handle_err(v,
+					   DM_VERITY_BLOCK_TYPE_METADATA,
+					   hash_block)) {
+			r = -EIO;
 			goto release_ret_r;
 		}
+	}
 
-		if (!v->version) {
-			r = crypto_shash_update(desc, v->salt, v->salt_size);
-			if (r < 0) {
-				DMERR("crypto_shash_update failed: %d", r);
-				goto release_ret_r;
-			}
-		}
+	data += offset;
+	memcpy(want_digest, data, v->digest_size);
+	r = 0;
 
-		result = io_real_digest(v, io);
-		r = crypto_shash_final(desc, result);
-		if (r < 0) {
-			DMERR("crypto_shash_final failed: %d", r);
-			goto release_ret_r;
-		}
-		if (unlikely(memcmp(result, io_want_digest(v, io), v->digest_size))) {
-			if (verity_handle_err(v, DM_VERITY_BLOCK_TYPE_METADATA,
-					      hash_block)) {
-				r = -EIO;
-				goto release_ret_r;
-			}
-		} else
-			aux->hash_verified = 1;
+release_ret_r:
+	dm_bufio_release(buf);
+	return r;
+}
+
+/*
+ * Find a hash for a given block, write it to digest and verify the integrity
+ * of the hash tree if necessary.
+ */
+int verity_hash_for_block(struct dm_verity *v, struct dm_verity_io *io,
+			  sector_t block, u8 *digest, bool *is_zero)
+{
+	int r = 0, i;
+
+	if (likely(v->levels)) {
+		/*
+		 * First, we try to get the requested hash for
+		 * the current block. If the hash block itself is
+		 * verified, zero is returned. If it isn't, this
+		 * function returns 1 and we fall back to whole
+		 * chain verification.
+		 */
+		r = verity_verify_level(v, io, block, 0, true, digest);
+		if (likely(r <= 0))
+			goto out;
 	}
 
-	data += offset;
+	memcpy(digest, v->root_digest, v->digest_size);
 
-	memcpy(io_want_digest(v, io), data, v->digest_size);
+	for (i = v->levels - 1; i >= 0; i--) {
+		r = verity_verify_level(v, io, block, i, false, digest);
+		if (unlikely(r))
+			goto out;
+	}
+out:
+	if (!r && v->zero_digest)
+		*is_zero = !memcmp(v->zero_digest, digest, v->digest_size);
+	else
+		*is_zero = false;
+
+	return r;
+}
+
+/*
+ * Calls function process for 1 << v->data_dev_block_bits bytes in the bio_vec
+ * starting from iter.
+ */
+int verity_for_bv_block(struct dm_verity *v, struct dm_verity_io *io,
+			struct bvec_iter *iter,
+			int (*process)(struct dm_verity *v,
+				       struct dm_verity_io *io, u8 *data,
+				       size_t len))
+{
+	unsigned todo = 1 << v->data_dev_block_bits;
+	struct bio *bio = dm_bio_from_per_bio_data(io, v->ti->per_bio_data_size);
+
+	do {
+		int r;
+		u8 *page;
+		unsigned len;
+		struct bio_vec bv = bio_iter_iovec(bio, *iter);
+
+		page = kmap_atomic(bv.bv_page);
+		len = bv.bv_len;
+
+		if (likely(len >= todo))
+			len = todo;
+
+		r = process(v, io, page + bv.bv_offset, len);
+		kunmap_atomic(page);
+
+		if (r < 0)
+			return r;
+
+		bio_advance_iter(bio, iter, len);
+		todo -= len;
+	} while (todo);
 
-	dm_bufio_release(buf);
 	return 0;
+}
 
-release_ret_r:
-	dm_bufio_release(buf);
+static int verity_bv_hash_update(struct dm_verity *v, struct dm_verity_io *io,
+				 u8 *data, size_t len)
+{
+	return verity_hash_update(v, verity_io_hash_desc(v, io), data, len);
+}
 
-	return r;
+static int verity_bv_zero(struct dm_verity *v, struct dm_verity_io *io,
+			  u8 *data, size_t len)
+{
+	memset(data, 0, len);
+	return 0;
 }
 
 /*
@@ -346,99 +403,56 @@ release_ret_r:
  */
 static int verity_verify_io(struct dm_verity_io *io)
 {
+	bool is_zero;
 	struct dm_verity *v = io->v;
-	struct bio *bio = dm_bio_from_per_bio_data(io,
-						   v->ti->per_bio_data_size);
+	struct bvec_iter start;
 	unsigned b;
-	int i;
 
 	for (b = 0; b < io->n_blocks; b++) {
-		struct shash_desc *desc;
-		u8 *result;
 		int r;
-		unsigned todo;
+		struct shash_desc *desc = verity_io_hash_desc(v, io);
 
-		if (likely(v->levels)) {
+		r = verity_hash_for_block(v, io, io->block + b,
+					  verity_io_want_digest(v, io),
+					  &is_zero);
+		if (unlikely(r < 0))
+			return r;
+
+		if (is_zero) {
 			/*
-			 * First, we try to get the requested hash for
-			 * the current block. If the hash block itself is
-			 * verified, zero is returned. If it isn't, this
-			 * function returns 0 and we fall back to whole
-			 * chain verification.
+			 * If we expect a zero block, don't validate, just
+			 * return zeros.
 			 */
-			int r = verity_verify_level(io, io->block + b, 0, true);
-			if (likely(!r))
-				goto test_block_hash;
-			if (r < 0)
+			r = verity_for_bv_block(v, io, &io->iter,
+						verity_bv_zero);
+			if (unlikely(r < 0))
 				return r;
-		}
-
-		memcpy(io_want_digest(v, io), v->root_digest, v->digest_size);
 
-		for (i = v->levels - 1; i >= 0; i--) {
-			int r = verity_verify_level(io, io->block + b, i, false);
-			if (unlikely(r))
-				return r;
+			continue;
 		}
 
-test_block_hash:
-		desc = io_hash_desc(v, io);
-		desc->tfm = v->tfm;
-		desc->flags = CRYPTO_TFM_REQ_MAY_SLEEP;
-		r = crypto_shash_init(desc);
-		if (r < 0) {
-			DMERR("crypto_shash_init failed: %d", r);
+		r = verity_hash_init(v, desc);
+		if (unlikely(r < 0))
 			return r;
-		}
-
-		if (likely(v->version >= 1)) {
-			r = crypto_shash_update(desc, v->salt, v->salt_size);
-			if (r < 0) {
-				DMERR("crypto_shash_update failed: %d", r);
-				return r;
-			}
-		}
-		todo = 1 << v->data_dev_block_bits;
-		do {
-			u8 *page;
-			unsigned len;
-			struct bio_vec bv = bio_iter_iovec(bio, io->iter);
-
-			page = kmap_atomic(bv.bv_page);
-			len = bv.bv_len;
-			if (likely(len >= todo))
-				len = todo;
-			r = crypto_shash_update(desc, page + bv.bv_offset, len);
-			kunmap_atomic(page);
-
-			if (r < 0) {
-				DMERR("crypto_shash_update failed: %d", r);
-				return r;
-			}
 
-			bio_advance_iter(bio, &io->iter, len);
-			todo -= len;
-		} while (todo);
-
-		if (!v->version) {
-			r = crypto_shash_update(desc, v->salt, v->salt_size);
-			if (r < 0) {
-				DMERR("crypto_shash_update failed: %d", r);
-				return r;
-			}
-		}
+		start = io->iter;
+		r = verity_for_bv_block(v, io, &io->iter, verity_bv_hash_update);
+		if (unlikely(r < 0))
+			return r;
 
-		result = io_real_digest(v, io);
-		r = crypto_shash_final(desc, result);
-		if (r < 0) {
-			DMERR("crypto_shash_final failed: %d", r);
+		r = verity_hash_final(v, desc, verity_io_real_digest(v, io));
+		if (unlikely(r < 0))
 			return r;
-		}
-		if (unlikely(memcmp(result, io_want_digest(v, io), v->digest_size))) {
-			if (verity_handle_err(v, DM_VERITY_BLOCK_TYPE_DATA,
-					      io->block + b))
-				return -EIO;
-		}
+
+		if (likely(memcmp(verity_io_real_digest(v, io),
+				  verity_io_want_digest(v, io), v->digest_size) == 0))
+			continue;
+		else if (verity_fec_decode(v, io, DM_VERITY_BLOCK_TYPE_DATA,
+					   io->block + b, NULL, &start) == 0)
+			continue;
+		else if (verity_handle_err(v, DM_VERITY_BLOCK_TYPE_DATA,
+					   io->block + b))
+			return -EIO;
 	}
 
 	return 0;
@@ -453,9 +467,10 @@ static void verity_finish_io(struct dm_verity_io *io, int error)
 	struct bio *bio = dm_bio_from_per_bio_data(io, v->ti->per_bio_data_size);
 
 	bio->bi_end_io = io->orig_bi_end_io;
-	bio->bi_private = io->orig_bi_private;
 	bio->bi_error = error;
 
+	verity_fec_finish_io(io);
+
 	bio_endio(bio);
 }
 
@@ -470,7 +485,7 @@ static void verity_end_io(struct bio *bio)
 {
 	struct dm_verity_io *io = bio->bi_private;
 
-	if (bio->bi_error) {
+	if (bio->bi_error && !verity_fec_is_enabled(io->v)) {
 		verity_finish_io(io, bio->bi_error);
 		return;
 	}
@@ -490,6 +505,7 @@ static void verity_prefetch_io(struct work_struct *work)
 		container_of(work, struct dm_verity_prefetch_work, work);
 	struct dm_verity *v = pw->v;
 	int i;
+	sector_t prefetch_size;
 
 	for (i = v->levels - 2; i >= 0; i--) {
 		sector_t hash_block_start;
@@ -512,8 +528,14 @@ static void verity_prefetch_io(struct work_struct *work)
 				hash_block_end = v->hash_blocks - 1;
 		}
 no_prefetch_cluster:
+		// for emmc, it is more efficient to send bigger read
+		prefetch_size = max((sector_t)CONFIG_DM_VERITY_HASH_PREFETCH_MIN_SIZE,
+			hash_block_end - hash_block_start + 1);
+		if ((hash_block_start + prefetch_size) >= (v->hash_start + v->hash_blocks)) {
+			prefetch_size = hash_block_end - hash_block_start + 1;
+		}
 		dm_bufio_prefetch(v->bufio, hash_block_start,
-				  hash_block_end - hash_block_start + 1);
+				  prefetch_size);
 	}
 
 	kfree(pw);
@@ -540,7 +562,7 @@ static void verity_submit_prefetch(struct dm_verity *v, struct dm_verity_io *io)
  * Bio map function. It allocates dm_verity_io structure and bio vector and
  * fills them. Then it issues prefetches and the I/O.
  */
-static int verity_map(struct dm_target *ti, struct bio *bio)
+int verity_map(struct dm_target *ti, struct bio *bio)
 {
 	struct dm_verity *v = ti->private;
 	struct dm_verity_io *io;
@@ -566,7 +588,6 @@ static int verity_map(struct dm_target *ti, struct bio *bio)
 	io = dm_per_bio_data(bio, ti->per_bio_data_size);
 	io->v = v;
 	io->orig_bi_end_io = bio->bi_end_io;
-	io->orig_bi_private = bio->bi_private;
 	io->block = bio->bi_iter.bi_sector >> (v->data_dev_block_bits - SECTOR_SHIFT);
 	io->n_blocks = bio->bi_iter.bi_size >> v->data_dev_block_bits;
 
@@ -574,20 +595,24 @@ static int verity_map(struct dm_target *ti, struct bio *bio)
 	bio->bi_private = io;
 	io->iter = bio->bi_iter;
 
+	verity_fec_init_io(io);
+
 	verity_submit_prefetch(v, io);
 
 	generic_make_request(bio);
 
 	return DM_MAPIO_SUBMITTED;
 }
+EXPORT_SYMBOL_GPL(verity_map);
 
 /*
  * Status: V (valid) or C (corruption found)
  */
-static void verity_status(struct dm_target *ti, status_type_t type,
+void verity_status(struct dm_target *ti, status_type_t type,
 			  unsigned status_flags, char *result, unsigned maxlen)
 {
 	struct dm_verity *v = ti->private;
+	unsigned args = 0;
 	unsigned sz = 0;
 	unsigned x;
 
@@ -614,8 +639,17 @@ static void verity_status(struct dm_target *ti, status_type_t type,
 		else
 			for (x = 0; x < v->salt_size; x++)
 				DMEMIT("%02x", v->salt[x]);
+		if (v->mode != DM_VERITY_MODE_EIO)
+			args++;
+		if (verity_fec_is_enabled(v))
+			args += DM_VERITY_OPTS_FEC;
+		if (v->zero_digest)
+			args++;
+		if (!args)
+			return;
+		DMEMIT(" %u", args);
 		if (v->mode != DM_VERITY_MODE_EIO) {
-			DMEMIT(" 1 ");
+			DMEMIT(" ");
 			switch (v->mode) {
 			case DM_VERITY_MODE_LOGGING:
 				DMEMIT(DM_VERITY_OPT_LOGGING);
@@ -627,11 +661,15 @@ static void verity_status(struct dm_target *ti, status_type_t type,
 				BUG();
 			}
 		}
+		if (v->zero_digest)
+			DMEMIT(" " DM_VERITY_OPT_IGN_ZEROES);
+		sz = verity_fec_status_table(v, sz, result, maxlen);
 		break;
 	}
 }
+EXPORT_SYMBOL_GPL(verity_status);
 
-static int verity_prepare_ioctl(struct dm_target *ti,
+int verity_prepare_ioctl(struct dm_target *ti,
 		struct block_device **bdev, fmode_t *mode)
 {
 	struct dm_verity *v = ti->private;
@@ -643,16 +681,18 @@ static int verity_prepare_ioctl(struct dm_target *ti,
 		return 1;
 	return 0;
 }
+EXPORT_SYMBOL_GPL(verity_prepare_ioctl);
 
-static int verity_iterate_devices(struct dm_target *ti,
+int verity_iterate_devices(struct dm_target *ti,
 				  iterate_devices_callout_fn fn, void *data)
 {
 	struct dm_verity *v = ti->private;
 
 	return fn(ti, v->data_dev, v->data_start, ti->len, data);
 }
+EXPORT_SYMBOL_GPL(verity_iterate_devices);
 
-static void verity_io_hints(struct dm_target *ti, struct queue_limits *limits)
+void verity_io_hints(struct dm_target *ti, struct queue_limits *limits)
 {
 	struct dm_verity *v = ti->private;
 
@@ -664,8 +704,9 @@ static void verity_io_hints(struct dm_target *ti, struct queue_limits *limits)
 
 	blk_limits_io_min(limits, limits->logical_block_size);
 }
+EXPORT_SYMBOL_GPL(verity_io_hints);
 
-static void verity_dtr(struct dm_target *ti)
+void verity_dtr(struct dm_target *ti)
 {
 	struct dm_verity *v = ti->private;
 
@@ -677,6 +718,7 @@ static void verity_dtr(struct dm_target *ti)
 
 	kfree(v->salt);
 	kfree(v->root_digest);
+	kfree(v->zero_digest);
 
 	if (v->tfm)
 		crypto_free_shash(v->tfm);
@@ -689,8 +731,94 @@ static void verity_dtr(struct dm_target *ti)
 	if (v->data_dev)
 		dm_put_device(ti, v->data_dev);
 
+	verity_fec_dtr(v);
+
 	kfree(v);
 }
+EXPORT_SYMBOL_GPL(verity_dtr);
+
+static int verity_alloc_zero_digest(struct dm_verity *v)
+{
+	int r = -ENOMEM;
+	struct shash_desc *desc;
+	u8 *zero_data;
+
+	v->zero_digest = kmalloc(v->digest_size, GFP_KERNEL);
+
+	if (!v->zero_digest)
+		return r;
+
+	desc = kmalloc(v->shash_descsize, GFP_KERNEL);
+
+	if (!desc)
+		return r; /* verity_dtr will free zero_digest */
+
+	zero_data = kzalloc(1 << v->data_dev_block_bits, GFP_KERNEL);
+
+	if (!zero_data)
+		goto out;
+
+	r = verity_hash(v, desc, zero_data, 1 << v->data_dev_block_bits,
+			v->zero_digest);
+
+out:
+	kfree(desc);
+	kfree(zero_data);
+
+	return r;
+}
+
+static int verity_parse_opt_args(struct dm_arg_set *as, struct dm_verity *v)
+{
+	int r;
+	unsigned argc;
+	struct dm_target *ti = v->ti;
+	const char *arg_name;
+
+	static struct dm_arg _args[] = {
+		{0, DM_VERITY_OPTS_MAX, "Invalid number of feature args"},
+	};
+
+	r = dm_read_arg_group(_args, as, &argc, &ti->error);
+	if (r)
+		return -EINVAL;
+
+	if (!argc)
+		return 0;
+
+	do {
+		arg_name = dm_shift_arg(as);
+		argc--;
+
+		if (!strcasecmp(arg_name, DM_VERITY_OPT_LOGGING)) {
+			v->mode = DM_VERITY_MODE_LOGGING;
+			continue;
+
+		} else if (!strcasecmp(arg_name, DM_VERITY_OPT_RESTART)) {
+			v->mode = DM_VERITY_MODE_RESTART;
+			continue;
+
+		} else if (!strcasecmp(arg_name, DM_VERITY_OPT_IGN_ZEROES)) {
+			r = verity_alloc_zero_digest(v);
+			if (r) {
+				ti->error = "Cannot allocate zero digest";
+				return r;
+			}
+			continue;
+
+		} else if (verity_is_fec_opt_arg(arg_name)) {
+			r = verity_fec_parse_opt_args(as, v, &argc, arg_name);
+			if (r)
+				return r;
+			continue;
+		}
+
+		ti->error = "Unrecognized verity feature request";
+		return -EINVAL;
+	} while (argc && !r);
+
+	return r;
+}
 
 /*
  * Target parameters:
@@ -706,22 +834,17 @@ static void verity_dtr(struct dm_target *ti)
  *	<digest>
  *	<salt>		Hex string or "-" if no salt.
  */
-static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
+int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 {
 	struct dm_verity *v;
 	struct dm_arg_set as;
-	const char *opt_string;
-	unsigned int num, opt_params;
+	unsigned int num;
 	unsigned long long num_ll;
 	int r;
 	int i;
 	sector_t hash_position;
 	char dummy;
 
-	static struct dm_arg _args[] = {
-		{0, 1, "Invalid number of feature args"},
-	};
-
 	v = kzalloc(sizeof(struct dm_verity), GFP_KERNEL);
 	if (!v) {
 		ti->error = "Cannot allocate verity structure";
@@ -730,6 +853,10 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	ti->private = v;
 	v->ti = ti;
 
+	r = verity_fec_ctr_alloc(v);
+	if (r)
+		goto bad;
+
 	if ((dm_table_get_mode(ti->table) & ~FMODE_READ)) {
 		ti->error = "Device must be readonly";
 		r = -EINVAL;
@@ -866,29 +993,9 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 		as.argc = argc;
 		as.argv = argv;
 
-		r = dm_read_arg_group(_args, &as, &opt_params, &ti->error);
-		if (r)
+		r = verity_parse_opt_args(&as, v);
+		if (r < 0)
 			goto bad;
-
-		while (opt_params) {
-			opt_params--;
-			opt_string = dm_shift_arg(&as);
-			if (!opt_string) {
-				ti->error = "Not enough feature arguments";
-				r = -EINVAL;
-				goto bad;
-			}
-
-			if (!strcasecmp(opt_string, DM_VERITY_OPT_LOGGING))
-				v->mode = DM_VERITY_MODE_LOGGING;
-			else if (!strcasecmp(opt_string, DM_VERITY_OPT_RESTART))
-				v->mode = DM_VERITY_MODE_RESTART;
-			else {
-				ti->error = "Invalid feature arguments";
-				r = -EINVAL;
-				goto bad;
-			}
-		}
 	}
 
 	v->hash_per_block_bits =
@@ -938,8 +1045,6 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 		goto bad;
 	}
 
-	ti->per_bio_data_size = roundup(sizeof(struct dm_verity_io) + v->shash_descsize + v->digest_size * 2, __alignof__(struct dm_verity_io));
-
 	/* WQ_UNBOUND greatly improves performance when running on ramdisk */
 	v->verify_wq = alloc_workqueue("kverityd", WQ_CPU_INTENSIVE | WQ_MEM_RECLAIM | WQ_UNBOUND, num_online_cpus());
 	if (!v->verify_wq) {
@@ -948,6 +1053,16 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv)
 		goto bad;
 	}
 
+	ti->per_bio_data_size = sizeof(struct dm_verity_io) +
+				v->shash_descsize + v->digest_size * 2;
+
+	r = verity_fec_ctr(v);
+	if (r)
+		goto bad;
+
+	ti->per_bio_data_size = roundup(ti->per_bio_data_size,
+					__alignof__(struct dm_verity_io));
+
 	return 0;
 
 bad:
@@ -955,10 +1070,11 @@ bad:
 
 	return r;
 }
+EXPORT_SYMBOL_GPL(verity_ctr);
 
 static struct target_type verity_target = {
 	.name		= "verity",
-	.version	= {1, 2, 0},
+	.version	= {1, 3, 0},
 	.module		= THIS_MODULE,
 	.ctr		= verity_ctr,
 	.dtr		= verity_dtr,
diff --git a/drivers/md/dm-verity.h b/drivers/md/dm-verity.h
new file mode 100644
index 000000000000..a90d1d416107
--- /dev/null
+++ b/drivers/md/dm-verity.h
@@ -0,0 +1,140 @@
+/*
+ * Copyright (C) 2012 Red Hat, Inc.
+ * Copyright (C) 2015 Google, Inc.
+ *
+ * Author: Mikulas Patocka <mpatocka@redhat.com>
+ *
+ * Based on Chromium dm-verity driver (C) 2011 The Chromium OS Authors
+ *
+ * This file is released under the GPLv2.
+ */
+
+#ifndef DM_VERITY_H
+#define DM_VERITY_H
+
+#include "dm-bufio.h"
+#include <linux/device-mapper.h>
+#include <crypto/hash.h>
+
+#define DM_VERITY_MAX_LEVELS		63
+
+enum verity_mode {
+	DM_VERITY_MODE_EIO,
+	DM_VERITY_MODE_LOGGING,
+	DM_VERITY_MODE_RESTART
+};
+
+enum verity_block_type {
+	DM_VERITY_BLOCK_TYPE_DATA,
+	DM_VERITY_BLOCK_TYPE_METADATA
+};
+
+struct dm_verity_fec;
+
+struct dm_verity {
+	struct dm_dev *data_dev;
+	struct dm_dev *hash_dev;
+	struct dm_target *ti;
+	struct dm_bufio_client *bufio;
+	char *alg_name;
+	struct crypto_shash *tfm;
+	u8 *root_digest;	/* digest of the root block */
+	u8 *salt;		/* salt: its size is salt_size */
+	u8 *zero_digest;	/* digest for a zero block */
+	unsigned salt_size;
+	sector_t data_start;	/* data offset in 512-byte sectors */
+	sector_t hash_start;	/* hash start in blocks */
+	sector_t data_blocks;	/* the number of data blocks */
+	sector_t hash_blocks;	/* the number of hash blocks */
+	unsigned char data_dev_block_bits;	/* log2(data blocksize) */
+	unsigned char hash_dev_block_bits;	/* log2(hash blocksize) */
+	unsigned char hash_per_block_bits;	/* log2(hashes in hash block) */
+	unsigned char levels;	/* the number of tree levels */
+	unsigned char version;
+	unsigned digest_size;	/* digest size for the current hash algorithm */
+	unsigned shash_descsize;/* the size of temporary space for crypto */
+	int hash_failed;	/* set to 1 if hash of any block failed */
+	enum verity_mode mode;	/* mode for handling verification errors */
+	unsigned corrupted_errs;/* Number of errors for corrupted blocks */
+
+	struct workqueue_struct *verify_wq;
+
+	/* starting blocks for each tree level. 0 is the lowest level. */
+	sector_t hash_level_block[DM_VERITY_MAX_LEVELS];
+
+	struct dm_verity_fec *fec;	/* forward error correction */
+};
+
+struct dm_verity_io {
+	struct dm_verity *v;
+
+	/* original value of bio->bi_end_io */
+	bio_end_io_t *orig_bi_end_io;
+
+	sector_t block;
+	unsigned n_blocks;
+
+	struct bvec_iter iter;
+
+	struct work_struct work;
+
+	/*
+	 * Three variably-size fields follow this struct:
+	 *
+	 * u8 hash_desc[v->shash_descsize];
+	 * u8 real_digest[v->digest_size];
+	 * u8 want_digest[v->digest_size];
+	 *
+	 * To access them use: verity_io_hash_desc(), verity_io_real_digest()
+	 * and verity_io_want_digest().
+	 */
+};
+
+static inline struct shash_desc *verity_io_hash_desc(struct dm_verity *v,
+						     struct dm_verity_io *io)
+{
+	return (struct shash_desc *)(io + 1);
+}
+
+static inline u8 *verity_io_real_digest(struct dm_verity *v,
+					struct dm_verity_io *io)
+{
+	return (u8 *)(io + 1) + v->shash_descsize;
+}
+
+static inline u8 *verity_io_want_digest(struct dm_verity *v,
+					struct dm_verity_io *io)
+{
+	return (u8 *)(io + 1) + v->shash_descsize + v->digest_size;
+}
+
+static inline u8 *verity_io_digest_end(struct dm_verity *v,
+				       struct dm_verity_io *io)
+{
+	return verity_io_want_digest(v, io) + v->digest_size;
+}
+
+extern int verity_for_bv_block(struct dm_verity *v, struct dm_verity_io *io,
+			       struct bvec_iter *iter,
+			       int (*process)(struct dm_verity *v,
+					      struct dm_verity_io *io,
+					      u8 *data, size_t len));
+
+extern int verity_hash(struct dm_verity *v, struct shash_desc *desc,
+		       const u8 *data, size_t len, u8 *digest);
+
+extern int verity_hash_for_block(struct dm_verity *v, struct dm_verity_io *io,
+				 sector_t block, u8 *digest, bool *is_zero);
+
+extern void verity_status(struct dm_target *ti, status_type_t type,
+			unsigned status_flags, char *result, unsigned maxlen);
+extern int verity_prepare_ioctl(struct dm_target *ti,
+                struct block_device **bdev, fmode_t *mode);
+extern int verity_iterate_devices(struct dm_target *ti,
+				iterate_devices_callout_fn fn, void *data);
+extern void verity_io_hints(struct dm_target *ti, struct queue_limits *limits);
+extern void verity_dtr(struct dm_target *ti);
+extern int verity_ctr(struct dm_target *ti, unsigned argc, char **argv);
+extern int verity_map(struct dm_target *ti, struct bio *bio);
+extern void dm_verity_avb_error_handler(void);
+#endif /* DM_VERITY_H */
diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
index 4bf7d50b1bc7..88056d1e8feb 100644
--- a/drivers/misc/Kconfig
+++ b/drivers/misc/Kconfig
@@ -525,6 +525,27 @@ config VEXPRESS_SYSCFG
 	  bus. System Configuration interface is one of the possible means
 	  of generating transactions on this bus.
 
+config UID_SYS_STATS
+	bool "Per-UID statistics"
+	depends on PROFILING && TASK_XACCT && TASK_IO_ACCOUNTING
+	help
+	  Per UID based cpu time statistics exported to /proc/uid_cputime
+	  Per UID based io statistics exported to /proc/uid_io
+	  Per UID based procstat control in /proc/uid_procstat
+
+config UID_SYS_STATS_DEBUG
+	bool "Per-TASK statistics"
+	depends on UID_SYS_STATS
+	default n
+	help
+	  Per TASK based io statistics exported to /proc/uid_io
+
+config MEMORY_STATE_TIME
+	tristate "Memory freq/bandwidth time statistics"
+	depends on PROFILING
+	help
+	  Memory time statistics exported to /sys/kernel/memory_state_time
+
 source "drivers/misc/c2port/Kconfig"
 source "drivers/misc/eeprom/Kconfig"
 source "drivers/misc/cb710/Kconfig"
diff --git a/drivers/misc/Makefile b/drivers/misc/Makefile
index 537d7f3b78da..9a3b402921b2 100644
--- a/drivers/misc/Makefile
+++ b/drivers/misc/Makefile
@@ -56,3 +56,5 @@ obj-$(CONFIG_GENWQE)		+= genwqe/
 obj-$(CONFIG_ECHO)		+= echo/
 obj-$(CONFIG_VEXPRESS_SYSCFG)	+= vexpress-syscfg.o
 obj-$(CONFIG_CXL_BASE)		+= cxl/
+obj-$(CONFIG_UID_SYS_STATS) += uid_sys_stats.o
+obj-$(CONFIG_MEMORY_STATE_TIME) += memory_state_time.o
diff --git a/drivers/misc/memory_state_time.c b/drivers/misc/memory_state_time.c
new file mode 100644
index 000000000000..ba94dcf09169
--- /dev/null
+++ b/drivers/misc/memory_state_time.c
@@ -0,0 +1,462 @@
+/* drivers/misc/memory_state_time.c
+ *
+ * Copyright (C) 2016 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/errno.h>
+#include <linux/hashtable.h>
+#include <linux/kconfig.h>
+#include <linux/kernel.h>
+#include <linux/kobject.h>
+#include <linux/memory-state-time.h>
+#include <linux/module.h>
+#include <linux/mutex.h>
+#include <linux/of_platform.h>
+#include <linux/slab.h>
+#include <linux/sysfs.h>
+#include <linux/time.h>
+#include <linux/timekeeping.h>
+#include <linux/workqueue.h>
+
+#define KERNEL_ATTR_RO(_name) \
+static struct kobj_attribute _name##_attr = __ATTR_RO(_name)
+
+#define KERNEL_ATTR_RW(_name) \
+static struct kobj_attribute _name##_attr = \
+	__ATTR(_name, 0644, _name##_show, _name##_store)
+
+#define FREQ_HASH_BITS 4
+DECLARE_HASHTABLE(freq_hash_table, FREQ_HASH_BITS);
+
+static DEFINE_MUTEX(mem_lock);
+
+#define TAG "memory_state_time"
+#define BW_NODE "/soc/memory-state-time"
+#define FREQ_TBL "freq-tbl"
+#define BW_TBL "bw-buckets"
+#define NUM_SOURCES "num-sources"
+
+#define LOWEST_FREQ 2
+
+static int curr_bw;
+static int curr_freq;
+static u32 *bw_buckets;
+static u32 *freq_buckets;
+static int num_freqs;
+static int num_buckets;
+static int registered_bw_sources;
+static u64 last_update;
+static bool init_success;
+static struct workqueue_struct *memory_wq;
+static u32 num_sources = 10;
+static int *bandwidths;
+
+struct freq_entry {
+	int freq;
+	u64 *buckets; /* Bandwidth buckets. */
+	struct hlist_node hash;
+};
+
+struct queue_container {
+	struct work_struct update_state;
+	int value;
+	u64 time_now;
+	int id;
+	struct mutex *lock;
+};
+
+static int find_bucket(int bw)
+{
+	int i;
+
+	if (bw_buckets != NULL) {
+		for (i = 0; i < num_buckets; i++) {
+			if (bw_buckets[i] > bw) {
+				pr_debug("Found bucket %d for bandwidth %d\n",
+					i, bw);
+				return i;
+			}
+		}
+		return num_buckets - 1;
+	}
+	return 0;
+}
+
+static u64 get_time_diff(u64 time_now)
+{
+	u64 ms;
+
+	ms = time_now - last_update;
+	last_update = time_now;
+	return ms;
+}
+
+static ssize_t show_stat_show(struct kobject *kobj,
+		struct kobj_attribute *attr, char *buf)
+{
+	int i, j;
+	int len = 0;
+	struct freq_entry *freq_entry;
+
+	for (i = 0; i < num_freqs; i++) {
+		hash_for_each_possible(freq_hash_table, freq_entry, hash,
+				freq_buckets[i]) {
+			if (freq_entry->freq == freq_buckets[i]) {
+				len += scnprintf(buf + len, PAGE_SIZE - len,
+						"%d ", freq_buckets[i]);
+				if (len >= PAGE_SIZE)
+					break;
+				for (j = 0; j < num_buckets; j++) {
+					len += scnprintf(buf + len,
+							PAGE_SIZE - len,
+							"%llu ",
+							freq_entry->buckets[j]);
+				}
+				len += scnprintf(buf + len, PAGE_SIZE - len,
+						"\n");
+			}
+		}
+	}
+	pr_debug("Current Time: %llu\n", ktime_get_boot_ns());
+	return len;
+}
+KERNEL_ATTR_RO(show_stat);
+
+static void update_table(u64 time_now)
+{
+	struct freq_entry *freq_entry;
+
+	pr_debug("Last known bw %d freq %d\n", curr_bw, curr_freq);
+	hash_for_each_possible(freq_hash_table, freq_entry, hash, curr_freq) {
+		if (curr_freq == freq_entry->freq) {
+			freq_entry->buckets[find_bucket(curr_bw)]
+					+= get_time_diff(time_now);
+			break;
+		}
+	}
+}
+
+static bool freq_exists(int freq)
+{
+	int i;
+
+	for (i = 0; i < num_freqs; i++) {
+		if (freq == freq_buckets[i])
+			return true;
+	}
+	return false;
+}
+
+static int calculate_total_bw(int bw, int index)
+{
+	int i;
+	int total_bw = 0;
+
+	pr_debug("memory_state_time New bw %d for id %d\n", bw, index);
+	bandwidths[index] = bw;
+	for (i = 0; i < registered_bw_sources; i++)
+		total_bw += bandwidths[i];
+	return total_bw;
+}
+
+static void freq_update_do_work(struct work_struct *work)
+{
+	struct queue_container *freq_state_update
+			= container_of(work, struct queue_container,
+			update_state);
+	if (freq_state_update) {
+		mutex_lock(&mem_lock);
+		update_table(freq_state_update->time_now);
+		curr_freq = freq_state_update->value;
+		mutex_unlock(&mem_lock);
+		kfree(freq_state_update);
+	}
+}
+
+static void bw_update_do_work(struct work_struct *work)
+{
+	struct queue_container *bw_state_update
+			= container_of(work, struct queue_container,
+			update_state);
+	if (bw_state_update) {
+		mutex_lock(&mem_lock);
+		update_table(bw_state_update->time_now);
+		curr_bw = calculate_total_bw(bw_state_update->value,
+				bw_state_update->id);
+		mutex_unlock(&mem_lock);
+		kfree(bw_state_update);
+	}
+}
+
+static void memory_state_freq_update(struct memory_state_update_block *ub,
+		int value)
+{
+	if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) {
+		if (freq_exists(value) && init_success) {
+			struct queue_container *freq_container
+				= kmalloc(sizeof(struct queue_container),
+				GFP_KERNEL);
+			if (!freq_container)
+				return;
+			INIT_WORK(&freq_container->update_state,
+					freq_update_do_work);
+			freq_container->time_now = ktime_get_boot_ns();
+			freq_container->value = value;
+			pr_debug("Scheduling freq update in work queue\n");
+			queue_work(memory_wq, &freq_container->update_state);
+		} else {
+			pr_debug("Freq does not exist.\n");
+		}
+	}
+}
+
+static void memory_state_bw_update(struct memory_state_update_block *ub,
+		int value)
+{
+	if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) {
+		if (init_success) {
+			struct queue_container *bw_container
+				= kmalloc(sizeof(struct queue_container),
+				GFP_KERNEL);
+			if (!bw_container)
+				return;
+			INIT_WORK(&bw_container->update_state,
+					bw_update_do_work);
+			bw_container->time_now = ktime_get_boot_ns();
+			bw_container->value = value;
+			bw_container->id = ub->id;
+			pr_debug("Scheduling bandwidth update in work queue\n");
+			queue_work(memory_wq, &bw_container->update_state);
+		}
+	}
+}
+
+struct memory_state_update_block *memory_state_register_frequency_source(void)
+{
+	struct memory_state_update_block *block;
+
+	if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) {
+		pr_debug("Allocating frequency source\n");
+		block = kmalloc(sizeof(struct memory_state_update_block),
+					GFP_KERNEL);
+		if (!block)
+			return NULL;
+		block->update_call = memory_state_freq_update;
+		return block;
+	}
+	pr_err("Config option disabled.\n");
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(memory_state_register_frequency_source);
+
+struct memory_state_update_block *memory_state_register_bandwidth_source(void)
+{
+	struct memory_state_update_block *block;
+
+	if (IS_ENABLED(CONFIG_MEMORY_STATE_TIME)) {
+		pr_debug("Allocating bandwidth source %d\n",
+				registered_bw_sources);
+		block = kmalloc(sizeof(struct memory_state_update_block),
+					GFP_KERNEL);
+		if (!block)
+			return NULL;
+		block->update_call = memory_state_bw_update;
+		if (registered_bw_sources < num_sources) {
+			block->id = registered_bw_sources++;
+		} else {
+			pr_err("Unable to allocate source; max number reached\n");
+			kfree(block);
+			return NULL;
+		}
+		return block;
+	}
+	pr_err("Config option disabled.\n");
+	return NULL;
+}
+EXPORT_SYMBOL_GPL(memory_state_register_bandwidth_source);
+
+/* Buckets are designated by their maximum.
+ * Returns the buckets decided by the capability of the device.
+ */
+static int get_bw_buckets(struct device *dev)
+{
+	int ret, lenb;
+	struct device_node *node = dev->of_node;
+
+	of_property_read_u32(node, NUM_SOURCES, &num_sources);
+	if (!of_find_property(node, BW_TBL, &lenb)) {
+		pr_err("Missing %s property\n", BW_TBL);
+		return -ENODATA;
+	}
+
+	bandwidths = devm_kzalloc(dev,
+			sizeof(*bandwidths) * num_sources, GFP_KERNEL);
+	if (!bandwidths)
+		return -ENOMEM;
+	lenb /= sizeof(*bw_buckets);
+	bw_buckets = devm_kzalloc(dev, lenb * sizeof(*bw_buckets),
+			GFP_KERNEL);
+	if (!bw_buckets) {
+		devm_kfree(dev, bandwidths);
+		return -ENOMEM;
+	}
+	ret = of_property_read_u32_array(node, BW_TBL, bw_buckets,
+			lenb);
+	if (ret < 0) {
+		devm_kfree(dev, bandwidths);
+		devm_kfree(dev, bw_buckets);
+		pr_err("Unable to read bandwidth table from device tree.\n");
+		return ret;
+	}
+
+	curr_bw = 0;
+	num_buckets = lenb;
+	return 0;
+}
+
+/* Adds struct freq_entry nodes to the hashtable for each compatible frequency.
+ * Returns the supported number of frequencies.
+ */
+static int freq_buckets_init(struct device *dev)
+{
+	struct freq_entry *freq_entry;
+	int i;
+	int ret, lenf;
+	struct device_node *node = dev->of_node;
+
+	if (!of_find_property(node, FREQ_TBL, &lenf)) {
+		pr_err("Missing %s property\n", FREQ_TBL);
+		return -ENODATA;
+	}
+
+	lenf /= sizeof(*freq_buckets);
+	freq_buckets = devm_kzalloc(dev, lenf * sizeof(*freq_buckets),
+			GFP_KERNEL);
+	if (!freq_buckets)
+		return -ENOMEM;
+	pr_debug("freqs found len %d\n", lenf);
+	ret = of_property_read_u32_array(node, FREQ_TBL, freq_buckets,
+			lenf);
+	if (ret < 0) {
+		devm_kfree(dev, freq_buckets);
+		pr_err("Unable to read frequency table from device tree.\n");
+		return ret;
+	}
+	pr_debug("ret freq %d\n", ret);
+
+	num_freqs = lenf;
+	curr_freq = freq_buckets[LOWEST_FREQ];
+
+	for (i = 0; i < num_freqs; i++) {
+		freq_entry = devm_kzalloc(dev, sizeof(struct freq_entry),
+				GFP_KERNEL);
+		if (!freq_entry)
+			return -ENOMEM;
+		freq_entry->buckets = devm_kzalloc(dev, sizeof(u64)*num_buckets,
+				GFP_KERNEL);
+		if (!freq_entry->buckets) {
+			devm_kfree(dev, freq_entry);
+			return -ENOMEM;
+		}
+		pr_debug("memory_state_time Adding freq to ht %d\n",
+				freq_buckets[i]);
+		freq_entry->freq = freq_buckets[i];
+		hash_add(freq_hash_table, &freq_entry->hash, freq_buckets[i]);
+	}
+	return 0;
+}
+
+struct kobject *memory_kobj;
+EXPORT_SYMBOL_GPL(memory_kobj);
+
+static struct attribute *memory_attrs[] = {
+	&show_stat_attr.attr,
+	NULL
+};
+
+static struct attribute_group memory_attr_group = {
+	.attrs = memory_attrs,
+};
+
+static int memory_state_time_probe(struct platform_device *pdev)
+{
+	int error;
+
+	error = get_bw_buckets(&pdev->dev);
+	if (error)
+		return error;
+	error = freq_buckets_init(&pdev->dev);
+	if (error)
+		return error;
+	last_update = ktime_get_boot_ns();
+	init_success = true;
+
+	pr_debug("memory_state_time initialized with num_freqs %d\n",
+			num_freqs);
+	return 0;
+}
+
+static const struct of_device_id match_table[] = {
+	{ .compatible = "memory-state-time" },
+	{}
+};
+
+static struct platform_driver memory_state_time_driver = {
+	.probe = memory_state_time_probe,
+	.driver = {
+		.name = "memory-state-time",
+		.of_match_table = match_table,
+		.owner = THIS_MODULE,
+	},
+};
+
+static int __init memory_state_time_init(void)
+{
+	int error;
+
+	hash_init(freq_hash_table);
+	memory_wq = create_singlethread_workqueue("memory_wq");
+	if (!memory_wq) {
+		pr_err("Unable to create workqueue.\n");
+		return -EINVAL;
+	}
+	/*
+	 * Create sys/kernel directory for memory_state_time.
+	 */
+	memory_kobj = kobject_create_and_add(TAG, kernel_kobj);
+	if (!memory_kobj) {
+		pr_err("Unable to allocate memory_kobj for sysfs directory.\n");
+		error = -ENOMEM;
+		goto wq;
+	}
+	error = sysfs_create_group(memory_kobj, &memory_attr_group);
+	if (error) {
+		pr_err("Unable to create sysfs folder.\n");
+		goto kobj;
+	}
+
+	error = platform_driver_register(&memory_state_time_driver);
+	if (error) {
+		pr_err("Unable to register memory_state_time platform driver.\n");
+		goto group;
+	}
+	return 0;
+
+group:	sysfs_remove_group(memory_kobj, &memory_attr_group);
+kobj:	kobject_put(memory_kobj);
+wq:	destroy_workqueue(memory_wq);
+	return error;
+}
+module_init(memory_state_time_init);
diff --git a/drivers/misc/uid_sys_stats.c b/drivers/misc/uid_sys_stats.c
new file mode 100644
index 000000000000..031320e51522
--- /dev/null
+++ b/drivers/misc/uid_sys_stats.c
@@ -0,0 +1,700 @@
+/* drivers/misc/uid_cputime.c
+ *
+ * Copyright (C) 2014 - 2015 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/atomic.h>
+#include <linux/err.h>
+#include <linux/hashtable.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/list.h>
+#include <linux/mm.h>
+#include <linux/proc_fs.h>
+#include <linux/profile.h>
+#include <linux/rtmutex.h>
+#include <linux/sched.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+
+#define UID_HASH_BITS	10
+DECLARE_HASHTABLE(hash_table, UID_HASH_BITS);
+
+static DEFINE_RT_MUTEX(uid_lock);
+static struct proc_dir_entry *cpu_parent;
+static struct proc_dir_entry *io_parent;
+static struct proc_dir_entry *proc_parent;
+
+struct io_stats {
+	u64 read_bytes;
+	u64 write_bytes;
+	u64 rchar;
+	u64 wchar;
+	u64 fsync;
+};
+
+#define UID_STATE_FOREGROUND	0
+#define UID_STATE_BACKGROUND	1
+#define UID_STATE_BUCKET_SIZE	2
+
+#define UID_STATE_TOTAL_CURR	2
+#define UID_STATE_TOTAL_LAST	3
+#define UID_STATE_DEAD_TASKS	4
+#define UID_STATE_SIZE		5
+
+#define MAX_TASK_COMM_LEN 256
+
+struct task_entry {
+	char comm[MAX_TASK_COMM_LEN];
+	pid_t pid;
+	struct io_stats io[UID_STATE_SIZE];
+	struct hlist_node hash;
+};
+
+struct uid_entry {
+	uid_t uid;
+	cputime_t utime;
+	cputime_t stime;
+	cputime_t active_utime;
+	cputime_t active_stime;
+	int state;
+	struct io_stats io[UID_STATE_SIZE];
+	struct hlist_node hash;
+#ifdef CONFIG_UID_SYS_STATS_DEBUG
+	DECLARE_HASHTABLE(task_entries, UID_HASH_BITS);
+#endif
+};
+
+static u64 compute_write_bytes(struct task_struct *task)
+{
+	if (task->ioac.write_bytes <= task->ioac.cancelled_write_bytes)
+		return 0;
+
+	return task->ioac.write_bytes - task->ioac.cancelled_write_bytes;
+}
+
+static void compute_io_bucket_stats(struct io_stats *io_bucket,
+					struct io_stats *io_curr,
+					struct io_stats *io_last,
+					struct io_stats *io_dead)
+{
+	/* tasks could switch to another uid group, but its io_last in the
+	 * previous uid group could still be positive.
+	 * therefore before each update, do an overflow check first
+	 */
+	int64_t delta;
+
+	delta = io_curr->read_bytes + io_dead->read_bytes -
+		io_last->read_bytes;
+	io_bucket->read_bytes += delta > 0 ? delta : 0;
+	delta = io_curr->write_bytes + io_dead->write_bytes -
+		io_last->write_bytes;
+	io_bucket->write_bytes += delta > 0 ? delta : 0;
+	delta = io_curr->rchar + io_dead->rchar - io_last->rchar;
+	io_bucket->rchar += delta > 0 ? delta : 0;
+	delta = io_curr->wchar + io_dead->wchar - io_last->wchar;
+	io_bucket->wchar += delta > 0 ? delta : 0;
+	delta = io_curr->fsync + io_dead->fsync - io_last->fsync;
+	io_bucket->fsync += delta > 0 ? delta : 0;
+
+	io_last->read_bytes = io_curr->read_bytes;
+	io_last->write_bytes = io_curr->write_bytes;
+	io_last->rchar = io_curr->rchar;
+	io_last->wchar = io_curr->wchar;
+	io_last->fsync = io_curr->fsync;
+
+	memset(io_dead, 0, sizeof(struct io_stats));
+}
+
+#ifdef CONFIG_UID_SYS_STATS_DEBUG
+static void get_full_task_comm(struct task_entry *task_entry,
+		struct task_struct *task)
+{
+	int i = 0, offset = 0, len = 0;
+	/* save one byte for terminating null character */
+	int unused_len = MAX_TASK_COMM_LEN - TASK_COMM_LEN - 1;
+	char buf[unused_len];
+	struct mm_struct *mm = task->mm;
+
+	/* fill the first TASK_COMM_LEN bytes with thread name */
+	get_task_comm(task_entry->comm, task);
+	i = strlen(task_entry->comm);
+	while (i < TASK_COMM_LEN)
+		task_entry->comm[i++] = ' ';
+
+	/* next the executable file name */
+	if (mm) {
+		down_read(&mm->mmap_sem);
+		if (mm->exe_file) {
+			char *pathname = d_path(&mm->exe_file->f_path, buf,
+					unused_len);
+
+			if (!IS_ERR(pathname)) {
+				len = strlcpy(task_entry->comm + i, pathname,
+						unused_len);
+				i += len;
+				task_entry->comm[i++] = ' ';
+				unused_len--;
+			}
+		}
+		up_read(&mm->mmap_sem);
+	}
+	unused_len -= len;
+
+	/* fill the rest with command line argument
+	 * replace each null or new line character
+	 * between args in argv with whitespace */
+	len = get_cmdline(task, buf, unused_len);
+	while (offset < len) {
+		if (buf[offset] != '\0' && buf[offset] != '\n')
+			task_entry->comm[i++] = buf[offset];
+		else
+			task_entry->comm[i++] = ' ';
+		offset++;
+	}
+
+	/* get rid of trailing whitespaces in case when arg is memset to
+	 * zero before being reset in userspace
+	 */
+	while (task_entry->comm[i-1] == ' ')
+		i--;
+	task_entry->comm[i] = '\0';
+}
+
+static struct task_entry *find_task_entry(struct uid_entry *uid_entry,
+		struct task_struct *task)
+{
+	struct task_entry *task_entry;
+
+	hash_for_each_possible(uid_entry->task_entries, task_entry, hash,
+			task->pid) {
+		if (task->pid == task_entry->pid) {
+			/* if thread name changed, update the entire command */
+			int len = strnchr(task_entry->comm, ' ', TASK_COMM_LEN)
+				- task_entry->comm;
+
+			if (strncmp(task_entry->comm, task->comm, len))
+				get_full_task_comm(task_entry, task);
+			return task_entry;
+		}
+	}
+	return NULL;
+}
+
+static struct task_entry *find_or_register_task(struct uid_entry *uid_entry,
+		struct task_struct *task)
+{
+	struct task_entry *task_entry;
+	pid_t pid = task->pid;
+
+	task_entry = find_task_entry(uid_entry, task);
+	if (task_entry)
+		return task_entry;
+
+	task_entry = kzalloc(sizeof(struct task_entry), GFP_ATOMIC);
+	if (!task_entry)
+		return NULL;
+
+	get_full_task_comm(task_entry, task);
+
+	task_entry->pid = pid;
+	hash_add(uid_entry->task_entries, &task_entry->hash, (unsigned int)pid);
+
+	return task_entry;
+}
+
+static void remove_uid_tasks(struct uid_entry *uid_entry)
+{
+	struct task_entry *task_entry;
+	unsigned long bkt_task;
+	struct hlist_node *tmp_task;
+
+	hash_for_each_safe(uid_entry->task_entries, bkt_task,
+			tmp_task, task_entry, hash) {
+		hash_del(&task_entry->hash);
+		kfree(task_entry);
+	}
+}
+
+static void set_io_uid_tasks_zero(struct uid_entry *uid_entry)
+{
+	struct task_entry *task_entry;
+	unsigned long bkt_task;
+
+	hash_for_each(uid_entry->task_entries, bkt_task, task_entry, hash) {
+		memset(&task_entry->io[UID_STATE_TOTAL_CURR], 0,
+			sizeof(struct io_stats));
+	}
+}
+
+static void add_uid_tasks_io_stats(struct uid_entry *uid_entry,
+		struct task_struct *task, int slot)
+{
+	struct task_entry *task_entry = find_or_register_task(uid_entry, task);
+	struct io_stats *task_io_slot = &task_entry->io[slot];
+
+	task_io_slot->read_bytes += task->ioac.read_bytes;
+	task_io_slot->write_bytes += compute_write_bytes(task);
+	task_io_slot->rchar += task->ioac.rchar;
+	task_io_slot->wchar += task->ioac.wchar;
+	task_io_slot->fsync += task->ioac.syscfs;
+}
+
+static void compute_io_uid_tasks(struct uid_entry *uid_entry)
+{
+	struct task_entry *task_entry;
+	unsigned long bkt_task;
+
+	hash_for_each(uid_entry->task_entries, bkt_task, task_entry, hash) {
+		compute_io_bucket_stats(&task_entry->io[uid_entry->state],
+					&task_entry->io[UID_STATE_TOTAL_CURR],
+					&task_entry->io[UID_STATE_TOTAL_LAST],
+					&task_entry->io[UID_STATE_DEAD_TASKS]);
+	}
+}
+
+static void show_io_uid_tasks(struct seq_file *m, struct uid_entry *uid_entry)
+{
+	struct task_entry *task_entry;
+	unsigned long bkt_task;
+
+	hash_for_each(uid_entry->task_entries, bkt_task, task_entry, hash) {
+		/* Separated by comma because space exists in task comm */
+		seq_printf(m, "task,%s,%lu,%llu,%llu,%llu,%llu,%llu,%llu,%llu,%llu,%llu,%llu\n",
+				task_entry->comm,
+				(unsigned long)task_entry->pid,
+				task_entry->io[UID_STATE_FOREGROUND].rchar,
+				task_entry->io[UID_STATE_FOREGROUND].wchar,
+				task_entry->io[UID_STATE_FOREGROUND].read_bytes,
+				task_entry->io[UID_STATE_FOREGROUND].write_bytes,
+				task_entry->io[UID_STATE_BACKGROUND].rchar,
+				task_entry->io[UID_STATE_BACKGROUND].wchar,
+				task_entry->io[UID_STATE_BACKGROUND].read_bytes,
+				task_entry->io[UID_STATE_BACKGROUND].write_bytes,
+				task_entry->io[UID_STATE_FOREGROUND].fsync,
+				task_entry->io[UID_STATE_BACKGROUND].fsync);
+	}
+}
+#else
+static void remove_uid_tasks(struct uid_entry *uid_entry) {};
+static void set_io_uid_tasks_zero(struct uid_entry *uid_entry) {};
+static void add_uid_tasks_io_stats(struct uid_entry *uid_entry,
+		struct task_struct *task, int slot) {};
+static void compute_io_uid_tasks(struct uid_entry *uid_entry) {};
+static void show_io_uid_tasks(struct seq_file *m,
+		struct uid_entry *uid_entry) {}
+#endif
+
+static struct uid_entry *find_uid_entry(uid_t uid)
+{
+	struct uid_entry *uid_entry;
+	hash_for_each_possible(hash_table, uid_entry, hash, uid) {
+		if (uid_entry->uid == uid)
+			return uid_entry;
+	}
+	return NULL;
+}
+
+static struct uid_entry *find_or_register_uid(uid_t uid)
+{
+	struct uid_entry *uid_entry;
+
+	uid_entry = find_uid_entry(uid);
+	if (uid_entry)
+		return uid_entry;
+
+	uid_entry = kzalloc(sizeof(struct uid_entry), GFP_ATOMIC);
+	if (!uid_entry)
+		return NULL;
+
+	uid_entry->uid = uid;
+#ifdef CONFIG_UID_SYS_STATS_DEBUG
+	hash_init(uid_entry->task_entries);
+#endif
+	hash_add(hash_table, &uid_entry->hash, uid);
+
+	return uid_entry;
+}
+
+static int uid_cputime_show(struct seq_file *m, void *v)
+{
+	struct uid_entry *uid_entry = NULL;
+	struct task_struct *task, *temp;
+	struct user_namespace *user_ns = current_user_ns();
+	cputime_t utime;
+	cputime_t stime;
+	unsigned long bkt;
+	uid_t uid;
+
+	rt_mutex_lock(&uid_lock);
+
+	hash_for_each(hash_table, bkt, uid_entry, hash) {
+		uid_entry->active_stime = 0;
+		uid_entry->active_utime = 0;
+	}
+
+	read_lock(&tasklist_lock);
+	do_each_thread(temp, task) {
+		uid = from_kuid_munged(user_ns, task_uid(task));
+		if (!uid_entry || uid_entry->uid != uid)
+			uid_entry = find_or_register_uid(uid);
+		if (!uid_entry) {
+			read_unlock(&tasklist_lock);
+			rt_mutex_unlock(&uid_lock);
+			pr_err("%s: failed to find the uid_entry for uid %d\n",
+				__func__, uid);
+			return -ENOMEM;
+		}
+		task_cputime_adjusted(task, &utime, &stime);
+		uid_entry->active_utime += utime;
+		uid_entry->active_stime += stime;
+	} while_each_thread(temp, task);
+	read_unlock(&tasklist_lock);
+
+	hash_for_each(hash_table, bkt, uid_entry, hash) {
+		cputime_t total_utime = uid_entry->utime +
+							uid_entry->active_utime;
+		cputime_t total_stime = uid_entry->stime +
+							uid_entry->active_stime;
+		seq_printf(m, "%d: %llu %llu\n", uid_entry->uid,
+			(unsigned long long)jiffies_to_msecs(
+				cputime_to_jiffies(total_utime)) * USEC_PER_MSEC,
+			(unsigned long long)jiffies_to_msecs(
+				cputime_to_jiffies(total_stime)) * USEC_PER_MSEC);
+	}
+
+	rt_mutex_unlock(&uid_lock);
+	return 0;
+}
+
+static int uid_cputime_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, uid_cputime_show, PDE_DATA(inode));
+}
+
+static const struct file_operations uid_cputime_fops = {
+	.open		= uid_cputime_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int uid_remove_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, NULL, NULL);
+}
+
+static ssize_t uid_remove_write(struct file *file,
+			const char __user *buffer, size_t count, loff_t *ppos)
+{
+	struct uid_entry *uid_entry;
+	struct hlist_node *tmp;
+	char uids[128];
+	char *start_uid, *end_uid = NULL;
+	long int uid_start = 0, uid_end = 0;
+
+	if (count >= sizeof(uids))
+		count = sizeof(uids) - 1;
+
+	if (copy_from_user(uids, buffer, count))
+		return -EFAULT;
+
+	uids[count] = '\0';
+	end_uid = uids;
+	start_uid = strsep(&end_uid, "-");
+
+	if (!start_uid || !end_uid)
+		return -EINVAL;
+
+	if (kstrtol(start_uid, 10, &uid_start) != 0 ||
+		kstrtol(end_uid, 10, &uid_end) != 0) {
+		return -EINVAL;
+	}
+	rt_mutex_lock(&uid_lock);
+
+	for (; uid_start <= uid_end; uid_start++) {
+		hash_for_each_possible_safe(hash_table, uid_entry, tmp,
+							hash, (uid_t)uid_start) {
+			if (uid_start == uid_entry->uid) {
+				remove_uid_tasks(uid_entry);
+				hash_del(&uid_entry->hash);
+				kfree(uid_entry);
+			}
+		}
+	}
+
+	rt_mutex_unlock(&uid_lock);
+	return count;
+}
+
+static const struct file_operations uid_remove_fops = {
+	.open		= uid_remove_open,
+	.release	= single_release,
+	.write		= uid_remove_write,
+};
+
+
+static void add_uid_io_stats(struct uid_entry *uid_entry,
+			struct task_struct *task, int slot)
+{
+	struct io_stats *io_slot = &uid_entry->io[slot];
+
+	io_slot->read_bytes += task->ioac.read_bytes;
+	io_slot->write_bytes += compute_write_bytes(task);
+	io_slot->rchar += task->ioac.rchar;
+	io_slot->wchar += task->ioac.wchar;
+	io_slot->fsync += task->ioac.syscfs;
+
+	add_uid_tasks_io_stats(uid_entry, task, slot);
+}
+
+static void update_io_stats_all_locked(void)
+{
+	struct uid_entry *uid_entry = NULL;
+	struct task_struct *task, *temp;
+	struct user_namespace *user_ns = current_user_ns();
+	unsigned long bkt;
+	uid_t uid;
+
+	hash_for_each(hash_table, bkt, uid_entry, hash) {
+		memset(&uid_entry->io[UID_STATE_TOTAL_CURR], 0,
+			sizeof(struct io_stats));
+		set_io_uid_tasks_zero(uid_entry);
+	}
+
+	rcu_read_lock();
+	do_each_thread(temp, task) {
+		uid = from_kuid_munged(user_ns, task_uid(task));
+		if (!uid_entry || uid_entry->uid != uid)
+			uid_entry = find_or_register_uid(uid);
+		if (!uid_entry)
+			continue;
+		add_uid_io_stats(uid_entry, task, UID_STATE_TOTAL_CURR);
+	} while_each_thread(temp, task);
+	rcu_read_unlock();
+
+	hash_for_each(hash_table, bkt, uid_entry, hash) {
+		compute_io_bucket_stats(&uid_entry->io[uid_entry->state],
+					&uid_entry->io[UID_STATE_TOTAL_CURR],
+					&uid_entry->io[UID_STATE_TOTAL_LAST],
+					&uid_entry->io[UID_STATE_DEAD_TASKS]);
+		compute_io_uid_tasks(uid_entry);
+	}
+}
+
+static void update_io_stats_uid_locked(struct uid_entry *uid_entry)
+{
+	struct task_struct *task, *temp;
+	struct user_namespace *user_ns = current_user_ns();
+
+	memset(&uid_entry->io[UID_STATE_TOTAL_CURR], 0,
+		sizeof(struct io_stats));
+	set_io_uid_tasks_zero(uid_entry);
+
+	rcu_read_lock();
+	do_each_thread(temp, task) {
+		if (from_kuid_munged(user_ns, task_uid(task)) != uid_entry->uid)
+			continue;
+		add_uid_io_stats(uid_entry, task, UID_STATE_TOTAL_CURR);
+	} while_each_thread(temp, task);
+	rcu_read_unlock();
+
+	compute_io_bucket_stats(&uid_entry->io[uid_entry->state],
+				&uid_entry->io[UID_STATE_TOTAL_CURR],
+				&uid_entry->io[UID_STATE_TOTAL_LAST],
+				&uid_entry->io[UID_STATE_DEAD_TASKS]);
+	compute_io_uid_tasks(uid_entry);
+}
+
+
+static int uid_io_show(struct seq_file *m, void *v)
+{
+	struct uid_entry *uid_entry;
+	unsigned long bkt;
+
+	rt_mutex_lock(&uid_lock);
+
+	update_io_stats_all_locked();
+
+	hash_for_each(hash_table, bkt, uid_entry, hash) {
+		seq_printf(m, "%d %llu %llu %llu %llu %llu %llu %llu %llu %llu %llu\n",
+				uid_entry->uid,
+				uid_entry->io[UID_STATE_FOREGROUND].rchar,
+				uid_entry->io[UID_STATE_FOREGROUND].wchar,
+				uid_entry->io[UID_STATE_FOREGROUND].read_bytes,
+				uid_entry->io[UID_STATE_FOREGROUND].write_bytes,
+				uid_entry->io[UID_STATE_BACKGROUND].rchar,
+				uid_entry->io[UID_STATE_BACKGROUND].wchar,
+				uid_entry->io[UID_STATE_BACKGROUND].read_bytes,
+				uid_entry->io[UID_STATE_BACKGROUND].write_bytes,
+				uid_entry->io[UID_STATE_FOREGROUND].fsync,
+				uid_entry->io[UID_STATE_BACKGROUND].fsync);
+
+		show_io_uid_tasks(m, uid_entry);
+	}
+
+	rt_mutex_unlock(&uid_lock);
+	return 0;
+}
+
+static int uid_io_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, uid_io_show, PDE_DATA(inode));
+}
+
+static const struct file_operations uid_io_fops = {
+	.open		= uid_io_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
+static int uid_procstat_open(struct inode *inode, struct file *file)
+{
+	return single_open(file, NULL, NULL);
+}
+
+static ssize_t uid_procstat_write(struct file *file,
+			const char __user *buffer, size_t count, loff_t *ppos)
+{
+	struct uid_entry *uid_entry;
+	uid_t uid;
+	int argc, state;
+	char input[128];
+
+	if (count >= sizeof(input))
+		return -EINVAL;
+
+	if (copy_from_user(input, buffer, count))
+		return -EFAULT;
+
+	input[count] = '\0';
+
+	argc = sscanf(input, "%u %d", &uid, &state);
+	if (argc != 2)
+		return -EINVAL;
+
+	if (state != UID_STATE_BACKGROUND && state != UID_STATE_FOREGROUND)
+		return -EINVAL;
+
+	rt_mutex_lock(&uid_lock);
+
+	uid_entry = find_or_register_uid(uid);
+	if (!uid_entry) {
+		rt_mutex_unlock(&uid_lock);
+		return -EINVAL;
+	}
+
+	if (uid_entry->state == state) {
+		rt_mutex_unlock(&uid_lock);
+		return count;
+	}
+
+	update_io_stats_uid_locked(uid_entry);
+
+	uid_entry->state = state;
+
+	rt_mutex_unlock(&uid_lock);
+
+	return count;
+}
+
+static const struct file_operations uid_procstat_fops = {
+	.open		= uid_procstat_open,
+	.release	= single_release,
+	.write		= uid_procstat_write,
+};
+
+static int process_notifier(struct notifier_block *self,
+			unsigned long cmd, void *v)
+{
+	struct task_struct *task = v;
+	struct uid_entry *uid_entry;
+	cputime_t utime, stime;
+	uid_t uid;
+
+	if (!task)
+		return NOTIFY_OK;
+
+	rt_mutex_lock(&uid_lock);
+	uid = from_kuid_munged(current_user_ns(), task_uid(task));
+	uid_entry = find_or_register_uid(uid);
+	if (!uid_entry) {
+		pr_err("%s: failed to find uid %d\n", __func__, uid);
+		goto exit;
+	}
+
+	task_cputime_adjusted(task, &utime, &stime);
+	uid_entry->utime += utime;
+	uid_entry->stime += stime;
+
+	add_uid_io_stats(uid_entry, task, UID_STATE_DEAD_TASKS);
+
+exit:
+	rt_mutex_unlock(&uid_lock);
+	return NOTIFY_OK;
+}
+
+static struct notifier_block process_notifier_block = {
+	.notifier_call	= process_notifier,
+};
+
+static int __init proc_uid_sys_stats_init(void)
+{
+	hash_init(hash_table);
+
+	cpu_parent = proc_mkdir("uid_cputime", NULL);
+	if (!cpu_parent) {
+		pr_err("%s: failed to create uid_cputime proc entry\n",
+			__func__);
+		goto err;
+	}
+
+	proc_create_data("remove_uid_range", 0222, cpu_parent,
+		&uid_remove_fops, NULL);
+	proc_create_data("show_uid_stat", 0444, cpu_parent,
+		&uid_cputime_fops, NULL);
+
+	io_parent = proc_mkdir("uid_io", NULL);
+	if (!io_parent) {
+		pr_err("%s: failed to create uid_io proc entry\n",
+			__func__);
+		goto err;
+	}
+
+	proc_create_data("stats", 0444, io_parent,
+		&uid_io_fops, NULL);
+
+	proc_parent = proc_mkdir("uid_procstat", NULL);
+	if (!proc_parent) {
+		pr_err("%s: failed to create uid_procstat proc entry\n",
+			__func__);
+		goto err;
+	}
+
+	proc_create_data("set", 0222, proc_parent,
+		&uid_procstat_fops, NULL);
+
+	profile_event_register(PROFILE_TASK_EXIT, &process_notifier_block);
+
+	return 0;
+
+err:
+	remove_proc_subtree("uid_cputime", NULL);
+	remove_proc_subtree("uid_io", NULL);
+	remove_proc_subtree("uid_procstat", NULL);
+	return -ENOMEM;
+}
+
+early_initcall(proc_uid_sys_stats_init);
diff --git a/drivers/mmc/card/Kconfig b/drivers/mmc/card/Kconfig
index 5562308699bc..6142ec1b9dfb 100644
--- a/drivers/mmc/card/Kconfig
+++ b/drivers/mmc/card/Kconfig
@@ -68,3 +68,15 @@ config MMC_TEST
 
 	  This driver is only of interest to those developing or
 	  testing a host driver. Most people should say N here.
+
+config MMC_SIMULATE_MAX_SPEED
+	bool "Turn on maximum speed control per block device"
+	depends on MMC_BLOCK
+	help
+	  Say Y here to enable MMC device speed limiting. Used to test and
+	  simulate the behavior of the system when confronted with a slow MMC.
+
+	  Enables max_read_speed, max_write_speed and cache_size attributes to
+	  control the write or read maximum KB/second speed behaviors.
+
+	  If unsure, say N here.
diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c
index f2b733275a0a..8d169d5a8b01 100644
--- a/drivers/mmc/card/block.c
+++ b/drivers/mmc/card/block.c
@@ -36,6 +36,8 @@
 #include <linux/compat.h>
 #include <linux/pm_runtime.h>
 
+#include <trace/events/mmc.h>
+
 #include <linux/mmc/ioctl.h>
 #include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
@@ -47,13 +49,10 @@
 #include "queue.h"
 
 MODULE_ALIAS("mmc:block");
-
-#ifdef KERNEL
 #ifdef MODULE_PARAM_PREFIX
 #undef MODULE_PARAM_PREFIX
 #endif
 #define MODULE_PARAM_PREFIX "mmcblk."
-#endif
 
 #define INAND_CMD38_ARG_EXT_CSD  113
 #define INAND_CMD38_ARG_ERASE    0x00
@@ -171,11 +170,7 @@ static struct mmc_blk_data *mmc_blk_get(struct gendisk *disk)
 
 static inline int mmc_get_devidx(struct gendisk *disk)
 {
-	int devmaj = MAJOR(disk_devt(disk));
-	int devidx = MINOR(disk_devt(disk)) / perdev_minors;
-
-	if (!devmaj)
-		devidx = disk->first_minor / perdev_minors;
+	int devidx = disk->first_minor / perdev_minors;
 	return devidx;
 }
 
@@ -293,6 +288,250 @@ out:
 	return ret;
 }
 
+#ifdef CONFIG_MMC_SIMULATE_MAX_SPEED
+
+static int max_read_speed, max_write_speed, cache_size = 4;
+
+module_param(max_read_speed, int, S_IRUSR | S_IRGRP);
+MODULE_PARM_DESC(max_read_speed, "maximum KB/s read speed 0=off");
+module_param(max_write_speed, int, S_IRUSR | S_IRGRP);
+MODULE_PARM_DESC(max_write_speed, "maximum KB/s write speed 0=off");
+module_param(cache_size, int, S_IRUSR | S_IRGRP);
+MODULE_PARM_DESC(cache_size, "MB high speed memory or SLC cache");
+
+/*
+ * helper macros and expectations:
+ *  size    - unsigned long number of bytes
+ *  jiffies - unsigned long HZ timestamp difference
+ *  speed   - unsigned KB/s transfer rate
+ */
+#define size_and_speed_to_jiffies(size, speed) \
+		((size) * HZ / (speed) / 1024UL)
+#define jiffies_and_speed_to_size(jiffies, speed) \
+		(((speed) * (jiffies) * 1024UL) / HZ)
+#define jiffies_and_size_to_speed(jiffies, size) \
+		((size) * HZ / (jiffies) / 1024UL)
+
+/* Limits to report warning */
+/* jiffies_and_size_to_speed(10*HZ, queue_max_hw_sectors(q) * 512UL) ~ 25 */
+#define MIN_SPEED(q) 250 /* 10 times faster than a floppy disk */
+#define MAX_SPEED(q) jiffies_and_size_to_speed(1, queue_max_sectors(q) * 512UL)
+
+#define speed_valid(speed) ((speed) > 0)
+
+static const char off[] = "off\n";
+
+static int max_speed_show(int speed, char *buf)
+{
+	if (speed)
+		return scnprintf(buf, PAGE_SIZE, "%uKB/s\n", speed);
+	else
+		return scnprintf(buf, PAGE_SIZE, off);
+}
+
+static int max_speed_store(const char *buf, struct request_queue *q)
+{
+	unsigned int limit, set = 0;
+
+	if (!strncasecmp(off, buf, sizeof(off) - 2))
+		return set;
+	if (kstrtouint(buf, 0, &set) || (set > INT_MAX))
+		return -EINVAL;
+	if (set == 0)
+		return set;
+	limit = MAX_SPEED(q);
+	if (set > limit)
+		pr_warn("max speed %u ineffective above %u\n", set, limit);
+	limit = MIN_SPEED(q);
+	if (set < limit)
+		pr_warn("max speed %u painful below %u\n", set, limit);
+	return set;
+}
+
+static ssize_t max_write_speed_show(struct device *dev,
+				 struct device_attribute *attr, char *buf)
+{
+	struct mmc_blk_data *md = mmc_blk_get(dev_to_disk(dev));
+	int ret = max_speed_show(atomic_read(&md->queue.max_write_speed), buf);
+
+	mmc_blk_put(md);
+	return ret;
+}
+
+static ssize_t max_write_speed_store(struct device *dev,
+				  struct device_attribute *attr,
+				  const char *buf, size_t count)
+{
+	struct mmc_blk_data *md = mmc_blk_get(dev_to_disk(dev));
+	int set = max_speed_store(buf, md->queue.queue);
+
+	if (set < 0) {
+		mmc_blk_put(md);
+		return set;
+	}
+
+	atomic_set(&md->queue.max_write_speed, set);
+	mmc_blk_put(md);
+	return count;
+}
+
+static const DEVICE_ATTR(max_write_speed, S_IRUGO | S_IWUSR,
+	max_write_speed_show, max_write_speed_store);
+
+static ssize_t max_read_speed_show(struct device *dev,
+				 struct device_attribute *attr, char *buf)
+{
+	struct mmc_blk_data *md = mmc_blk_get(dev_to_disk(dev));
+	int ret = max_speed_show(atomic_read(&md->queue.max_read_speed), buf);
+
+	mmc_blk_put(md);
+	return ret;
+}
+
+static ssize_t max_read_speed_store(struct device *dev,
+				  struct device_attribute *attr,
+				  const char *buf, size_t count)
+{
+	struct mmc_blk_data *md = mmc_blk_get(dev_to_disk(dev));
+	int set = max_speed_store(buf, md->queue.queue);
+
+	if (set < 0) {
+		mmc_blk_put(md);
+		return set;
+	}
+
+	atomic_set(&md->queue.max_read_speed, set);
+	mmc_blk_put(md);
+	return count;
+}
+
+static const DEVICE_ATTR(max_read_speed, S_IRUGO | S_IWUSR,
+	max_read_speed_show, max_read_speed_store);
+
+static ssize_t cache_size_show(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	struct mmc_blk_data *md = mmc_blk_get(dev_to_disk(dev));
+	struct mmc_queue *mq = &md->queue;
+	int cache_size = atomic_read(&mq->cache_size);
+	int ret;
+
+	if (!cache_size)
+		ret = scnprintf(buf, PAGE_SIZE, off);
+	else {
+		int speed = atomic_read(&mq->max_write_speed);
+
+		if (!speed_valid(speed))
+			ret = scnprintf(buf, PAGE_SIZE, "%uMB\n", cache_size);
+		else { /* We accept race between cache_jiffies and cache_used */
+			unsigned long size = jiffies_and_speed_to_size(
+				jiffies - mq->cache_jiffies, speed);
+			long used = atomic_long_read(&mq->cache_used);
+
+			if (size >= used)
+				size = 0;
+			else
+				size = (used - size) * 100 / cache_size
+					/ 1024UL / 1024UL;
+
+			ret = scnprintf(buf, PAGE_SIZE, "%uMB %lu%% used\n",
+				cache_size, size);
+		}
+	}
+
+	mmc_blk_put(md);
+	return ret;
+}
+
+static ssize_t cache_size_store(struct device *dev,
+				  struct device_attribute *attr,
+				  const char *buf, size_t count)
+{
+	struct mmc_blk_data *md;
+	unsigned int set = 0;
+
+	if (strncasecmp(off, buf, sizeof(off) - 2)
+	 && (kstrtouint(buf, 0, &set) || (set > INT_MAX)))
+		return -EINVAL;
+
+	md = mmc_blk_get(dev_to_disk(dev));
+	atomic_set(&md->queue.cache_size, set);
+	mmc_blk_put(md);
+	return count;
+}
+
+static const DEVICE_ATTR(cache_size, S_IRUGO | S_IWUSR,
+	cache_size_show, cache_size_store);
+
+/* correct for write-back */
+static long mmc_blk_cache_used(struct mmc_queue *mq, unsigned long waitfor)
+{
+	long used = 0;
+	int speed = atomic_read(&mq->max_write_speed);
+
+	if (speed_valid(speed)) {
+		unsigned long size = jiffies_and_speed_to_size(
+					waitfor - mq->cache_jiffies, speed);
+		used = atomic_long_read(&mq->cache_used);
+
+		if (size >= used)
+			used = 0;
+		else
+			used -= size;
+	}
+
+	atomic_long_set(&mq->cache_used, used);
+	mq->cache_jiffies = waitfor;
+
+	return used;
+}
+
+static void mmc_blk_simulate_delay(
+	struct mmc_queue *mq,
+	struct request *req,
+	unsigned long waitfor)
+{
+	int max_speed;
+
+	if (!req)
+		return;
+
+	max_speed = (rq_data_dir(req) == READ)
+		? atomic_read(&mq->max_read_speed)
+		: atomic_read(&mq->max_write_speed);
+	if (speed_valid(max_speed)) {
+		unsigned long bytes = blk_rq_bytes(req);
+
+		if (rq_data_dir(req) != READ) {
+			int cache_size = atomic_read(&mq->cache_size);
+
+			if (cache_size) {
+				unsigned long size = cache_size * 1024L * 1024L;
+				long used = mmc_blk_cache_used(mq, waitfor);
+
+				used += bytes;
+				atomic_long_set(&mq->cache_used, used);
+				bytes = 0;
+				if (used > size)
+					bytes = used - size;
+			}
+		}
+		waitfor += size_and_speed_to_jiffies(bytes, max_speed);
+		if (time_is_after_jiffies(waitfor)) {
+			long msecs = jiffies_to_msecs(waitfor - jiffies);
+
+			if (likely(msecs > 0))
+				msleep(msecs);
+		}
+	}
+}
+
+#else
+
+#define mmc_blk_simulate_delay(mq, req, waitfor)
+
+#endif
+
 static int mmc_blk_open(struct block_device *bdev, fmode_t mode)
 {
 	struct mmc_blk_data *md = mmc_blk_get(bdev->bd_disk);
@@ -450,9 +689,11 @@ static int ioctl_do_sanitize(struct mmc_card *card)
 	pr_debug("%s: %s - SANITIZE IN PROGRESS...\n",
 		mmc_hostname(card->host), __func__);
 
+	trace_mmc_blk_erase_start(EXT_CSD_SANITIZE_START, 0, 0);
 	err = mmc_switch(card, EXT_CSD_CMD_SET_NORMAL,
 					EXT_CSD_SANITIZE_START, 1,
 					MMC_SANITIZE_REQ_TIMEOUT);
+	trace_mmc_blk_erase_end(EXT_CSD_SANITIZE_START, 0, 0);
 
 	if (err)
 		pr_err("%s: %s - EXT_CSD_SANITIZE_START failed. err=%d\n",
@@ -952,18 +1193,22 @@ static int mmc_blk_cmd_error(struct request *req, const char *name, int error,
 			req->rq_disk->disk_name, "timed out", name, status);
 
 		/* If the status cmd initially failed, retry the r/w cmd */
-		if (!status_valid)
+		if (!status_valid) {
+			pr_err("%s: status not valid, retrying timeout\n", req->rq_disk->disk_name);
 			return ERR_RETRY;
-
+		}
 		/*
 		 * If it was a r/w cmd crc error, or illegal command
 		 * (eg, issued in wrong state) then retry - we should
 		 * have corrected the state problem above.
 		 */
-		if (status & (R1_COM_CRC_ERROR | R1_ILLEGAL_COMMAND))
+		if (status & (R1_COM_CRC_ERROR | R1_ILLEGAL_COMMAND)) {
+			pr_err("%s: command error, retrying timeout\n", req->rq_disk->disk_name);
 			return ERR_RETRY;
+		}
 
 		/* Otherwise abort the command */
+		pr_err("%s: not retrying timeout\n", req->rq_disk->disk_name);
 		return ERR_ABORT;
 
 	default:
@@ -1263,6 +1508,23 @@ static int mmc_blk_issue_flush(struct mmc_queue *mq, struct request *req)
 	if (ret)
 		ret = -EIO;
 
+#ifdef CONFIG_MMC_SIMULATE_MAX_SPEED
+	else if (atomic_read(&mq->cache_size)) {
+		long used = mmc_blk_cache_used(mq, jiffies);
+
+		if (used) {
+			int speed = atomic_read(&mq->max_write_speed);
+
+			if (speed_valid(speed)) {
+				unsigned long msecs = jiffies_to_msecs(
+					size_and_speed_to_jiffies(
+						used, speed));
+				if (msecs)
+					msleep(msecs);
+			}
+		}
+	}
+#endif
 	blk_end_request_all(req, ret);
 
 	return ret ? 0 : 1;
@@ -1942,6 +2204,9 @@ static int mmc_blk_issue_rw_rq(struct mmc_queue *mq, struct request *rqc)
 	struct mmc_async_req *areq;
 	const u8 packed_nr = 2;
 	u8 reqs = 0;
+#ifdef CONFIG_MMC_SIMULATE_MAX_SPEED
+	unsigned long waitfor = jiffies;
+#endif
 
 	if (!rqc && !mq->mqrq_prev->req)
 		return 0;
@@ -1992,6 +2257,8 @@ static int mmc_blk_issue_rw_rq(struct mmc_queue *mq, struct request *rqc)
 			 */
 			mmc_blk_reset_success(md, type);
 
+			mmc_blk_simulate_delay(mq, rqc, waitfor);
+
 			if (mmc_packed_cmd(mq_rq->cmd_type)) {
 				ret = mmc_blk_end_packed_req(mq_rq);
 				break;
@@ -2252,6 +2519,7 @@ static struct mmc_blk_data *mmc_blk_alloc_req(struct mmc_card *card,
 	md->disk->queue = md->queue.queue;
 	md->disk->driverfs_dev = parent;
 	set_disk_ro(md->disk, md->read_only || default_ro);
+	md->disk->flags = GENHD_FL_EXT_DEVT;
 	if (area_type & (MMC_BLK_DATA_AREA_RPMB | MMC_BLK_DATA_AREA_BOOT))
 		md->disk->flags |= GENHD_FL_NO_PART_SCAN;
 
@@ -2410,6 +2678,14 @@ static void mmc_blk_remove_req(struct mmc_blk_data *md)
 					card->ext_csd.boot_ro_lockable)
 				device_remove_file(disk_to_dev(md->disk),
 					&md->power_ro_lock);
+#ifdef CONFIG_MMC_SIMULATE_MAX_SPEED
+			device_remove_file(disk_to_dev(md->disk),
+						&dev_attr_max_write_speed);
+			device_remove_file(disk_to_dev(md->disk),
+						&dev_attr_max_read_speed);
+			device_remove_file(disk_to_dev(md->disk),
+						&dev_attr_cache_size);
+#endif
 
 			del_gendisk(md->disk);
 		}
@@ -2445,6 +2721,24 @@ static int mmc_add_disk(struct mmc_blk_data *md)
 	ret = device_create_file(disk_to_dev(md->disk), &md->force_ro);
 	if (ret)
 		goto force_ro_fail;
+#ifdef CONFIG_MMC_SIMULATE_MAX_SPEED
+	atomic_set(&md->queue.max_write_speed, max_write_speed);
+	ret = device_create_file(disk_to_dev(md->disk),
+			&dev_attr_max_write_speed);
+	if (ret)
+		goto max_write_speed_fail;
+	atomic_set(&md->queue.max_read_speed, max_read_speed);
+	ret = device_create_file(disk_to_dev(md->disk),
+			&dev_attr_max_read_speed);
+	if (ret)
+		goto max_read_speed_fail;
+	atomic_set(&md->queue.cache_size, cache_size);
+	atomic_long_set(&md->queue.cache_used, 0);
+	md->queue.cache_jiffies = jiffies;
+	ret = device_create_file(disk_to_dev(md->disk), &dev_attr_cache_size);
+	if (ret)
+		goto cache_size_fail;
+#endif
 
 	if ((md->area_type & MMC_BLK_DATA_AREA_BOOT) &&
 	     card->ext_csd.boot_ro_lockable) {
@@ -2469,6 +2763,14 @@ static int mmc_add_disk(struct mmc_blk_data *md)
 	return ret;
 
 power_ro_lock_fail:
+#ifdef CONFIG_MMC_SIMULATE_MAX_SPEED
+	device_remove_file(disk_to_dev(md->disk), &dev_attr_cache_size);
+cache_size_fail:
+	device_remove_file(disk_to_dev(md->disk), &dev_attr_max_read_speed);
+max_read_speed_fail:
+	device_remove_file(disk_to_dev(md->disk), &dev_attr_max_write_speed);
+max_write_speed_fail:
+#endif
 	device_remove_file(disk_to_dev(md->disk), &md->force_ro);
 force_ro_fail:
 	del_gendisk(md->disk);
diff --git a/drivers/mmc/card/queue.c b/drivers/mmc/card/queue.c
index 6f4323c6d653..6a4cd2bb4629 100644
--- a/drivers/mmc/card/queue.c
+++ b/drivers/mmc/card/queue.c
@@ -19,6 +19,7 @@
 
 #include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
+#include <linux/sched/rt.h>
 #include "queue.h"
 
 #define MMC_QUEUE_BOUNCESZ	65536
@@ -50,6 +51,11 @@ static int mmc_queue_thread(void *d)
 {
 	struct mmc_queue *mq = d;
 	struct request_queue *q = mq->queue;
+	struct sched_param scheduler_params = {0};
+
+	scheduler_params.sched_priority = 1;
+
+	sched_setscheduler(current, SCHED_FIFO, &scheduler_params);
 
 	current->flags |= PF_MEMALLOC;
 
diff --git a/drivers/mmc/card/queue.h b/drivers/mmc/card/queue.h
index cf30b3712cb2..1dc4c99f52a1 100644
--- a/drivers/mmc/card/queue.h
+++ b/drivers/mmc/card/queue.h
@@ -58,6 +58,14 @@ struct mmc_queue {
 	struct mmc_queue_req	mqrq[2];
 	struct mmc_queue_req	*mqrq_cur;
 	struct mmc_queue_req	*mqrq_prev;
+#ifdef CONFIG_MMC_SIMULATE_MAX_SPEED
+	atomic_t max_write_speed;
+	atomic_t max_read_speed;
+	atomic_t cache_size;
+	/* i/o tracking */
+	atomic_long_t cache_used;
+	unsigned long cache_jiffies;
+#endif
 };
 
 extern int mmc_init_queue(struct mmc_queue *, struct mmc_card *, spinlock_t *,
diff --git a/drivers/mmc/core/Kconfig b/drivers/mmc/core/Kconfig
index 4c33d7690f2f..87cc07dedd9f 100644
--- a/drivers/mmc/core/Kconfig
+++ b/drivers/mmc/core/Kconfig
@@ -1,3 +1,18 @@
 #
 # MMC core configuration
 #
+
+config MMC_EMBEDDED_SDIO
+	boolean "MMC embedded SDIO device support (EXPERIMENTAL)"
+	help
+	  If you say Y here, support will be added for embedded SDIO
+	  devices which do not contain the necessary enumeration
+	  support in hardware to be properly detected.
+
+config MMC_PARANOID_SD_INIT
+	bool "Enable paranoid SD card initialization (EXPERIMENTAL)"
+	help
+	  If you say Y here, the MMC layer will be extra paranoid
+	  about re-trying SD init requests. This can be a useful
+	  work-around for buggy controllers and hardware. Enable
+	  if you are experiencing issues with SD detection.
diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
index 5f7d10ba498a..2986e270d19a 100644
--- a/drivers/mmc/core/core.c
+++ b/drivers/mmc/core/core.c
@@ -30,6 +30,9 @@
 #include <linux/slab.h>
 #include <linux/of.h>
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/mmc.h>
+
 #include <linux/mmc/card.h>
 #include <linux/mmc/host.h>
 #include <linux/mmc/mmc.h>
@@ -46,6 +49,11 @@
 #include "sd_ops.h"
 #include "sdio_ops.h"
 
+EXPORT_TRACEPOINT_SYMBOL_GPL(mmc_blk_erase_start);
+EXPORT_TRACEPOINT_SYMBOL_GPL(mmc_blk_erase_end);
+EXPORT_TRACEPOINT_SYMBOL_GPL(mmc_blk_rw_start);
+EXPORT_TRACEPOINT_SYMBOL_GPL(mmc_blk_rw_end);
+
 /* If the device is not responding */
 #define MMC_CORE_TIMEOUT_MS	(10 * 60 * 1000) /* 10 minute timeout */
 
@@ -175,6 +183,20 @@ void mmc_request_done(struct mmc_host *host, struct mmc_request *mrq)
 			pr_debug("%s:     %d bytes transferred: %d\n",
 				mmc_hostname(host),
 				mrq->data->bytes_xfered, mrq->data->error);
+#ifdef CONFIG_BLOCK
+			if (mrq->lat_hist_enabled) {
+				ktime_t completion;
+				u_int64_t delta_us;
+
+				completion = ktime_get();
+				delta_us = ktime_us_delta(completion,
+							  mrq->io_start);
+				blk_update_latency_hist(&host->io_lat_s,
+					(mrq->data->flags & MMC_DATA_READ),
+					delta_us);
+			}
+#endif
+			trace_mmc_blk_rw_end(cmd->opcode, cmd->arg, mrq->data);
 		}
 
 		if (mrq->stop) {
@@ -617,8 +639,19 @@ struct mmc_async_req *mmc_start_req(struct mmc_host *host,
 		}
 	}
 
-	if (!err && areq)
+	if (!err && areq) {
+#ifdef CONFIG_BLOCK
+		if (host->latency_hist_enabled) {
+			areq->mrq->io_start = ktime_get();
+			areq->mrq->lat_hist_enabled = 1;
+		} else
+			areq->mrq->lat_hist_enabled = 0;
+#endif
+		trace_mmc_blk_rw_start(areq->mrq->cmd->opcode,
+				       areq->mrq->cmd->arg,
+				       areq->mrq->data);
 		start_err = __mmc_start_data_req(host, areq->mrq);
+	}
 
 	if (host->areq)
 		mmc_post_req(host, host->areq->mrq, 0);
@@ -1951,7 +1984,7 @@ void mmc_init_erase(struct mmc_card *card)
 }
 
 static unsigned int mmc_mmc_erase_timeout(struct mmc_card *card,
-				          unsigned int arg, unsigned int qty)
+					  unsigned int arg, unsigned int qty)
 {
 	unsigned int erase_timeout;
 
@@ -2055,8 +2088,13 @@ static int mmc_do_erase(struct mmc_card *card, unsigned int from,
 	struct mmc_command cmd = {0};
 	unsigned int qty = 0;
 	unsigned long timeout;
+	unsigned int fr, nr;
 	int err;
 
+	fr = from;
+	nr = to - from + 1;
+	trace_mmc_blk_erase_start(arg, fr, nr);
+
 	mmc_retune_hold(card->host);
 
 	/*
@@ -2163,6 +2201,7 @@ static int mmc_do_erase(struct mmc_card *card, unsigned int from,
 		 (R1_CURRENT_STATE(cmd.resp[0]) == R1_STATE_PRG));
 out:
 	mmc_retune_release(card->host);
+	trace_mmc_blk_erase_end(arg, fr, nr);
 	return err;
 }
 
@@ -2832,6 +2871,22 @@ void mmc_init_context_info(struct mmc_host *host)
 	init_waitqueue_head(&host->context_info.wait);
 }
 
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+void mmc_set_embedded_sdio_data(struct mmc_host *host,
+				struct sdio_cis *cis,
+				struct sdio_cccr *cccr,
+				struct sdio_embedded_func *funcs,
+				int num_funcs)
+{
+	host->embedded_sdio_data.cis = cis;
+	host->embedded_sdio_data.cccr = cccr;
+	host->embedded_sdio_data.funcs = funcs;
+	host->embedded_sdio_data.num_funcs = num_funcs;
+}
+
+EXPORT_SYMBOL(mmc_set_embedded_sdio_data);
+#endif
+
 static int __init mmc_init(void)
 {
 	int ret;
@@ -2872,6 +2927,56 @@ static void __exit mmc_exit(void)
 	destroy_workqueue(workqueue);
 }
 
+#ifdef CONFIG_BLOCK
+static ssize_t
+latency_hist_show(struct device *dev, struct device_attribute *attr, char *buf)
+{
+	struct mmc_host *host = cls_dev_to_mmc_host(dev);
+
+	return blk_latency_hist_show(&host->io_lat_s, buf);
+}
+
+/*
+ * Values permitted 0, 1, 2.
+ * 0 -> Disable IO latency histograms (default)
+ * 1 -> Enable IO latency histograms
+ * 2 -> Zero out IO latency histograms
+ */
+static ssize_t
+latency_hist_store(struct device *dev, struct device_attribute *attr,
+		   const char *buf, size_t count)
+{
+	struct mmc_host *host = cls_dev_to_mmc_host(dev);
+	long value;
+
+	if (kstrtol(buf, 0, &value))
+		return -EINVAL;
+	if (value == BLK_IO_LAT_HIST_ZERO)
+		blk_zero_latency_hist(&host->io_lat_s);
+	else if (value == BLK_IO_LAT_HIST_ENABLE ||
+		 value == BLK_IO_LAT_HIST_DISABLE)
+		host->latency_hist_enabled = value;
+	return count;
+}
+
+static DEVICE_ATTR(latency_hist, S_IRUGO | S_IWUSR,
+		   latency_hist_show, latency_hist_store);
+
+void
+mmc_latency_hist_sysfs_init(struct mmc_host *host)
+{
+	if (device_create_file(&host->class_dev, &dev_attr_latency_hist))
+		dev_err(&host->class_dev,
+			"Failed to create latency_hist sysfs entry\n");
+}
+
+void
+mmc_latency_hist_sysfs_exit(struct mmc_host *host)
+{
+	device_remove_file(&host->class_dev, &dev_attr_latency_hist);
+}
+#endif
+
 subsys_initcall(mmc_init);
 module_exit(mmc_exit);
 
diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c
index da950c44204d..443fdfc22d8a 100644
--- a/drivers/mmc/core/host.c
+++ b/drivers/mmc/core/host.c
@@ -32,8 +32,6 @@
 #include "slot-gpio.h"
 #include "pwrseq.h"
 
-#define cls_dev_to_mmc_host(d)	container_of(d, struct mmc_host, class_dev)
-
 static DEFINE_IDR(mmc_host_idr);
 static DEFINE_SPINLOCK(mmc_host_lock);
 
@@ -394,8 +392,13 @@ int mmc_add_host(struct mmc_host *host)
 	mmc_add_host_debugfs(host);
 #endif
 
+#ifdef CONFIG_BLOCK
+	mmc_latency_hist_sysfs_init(host);
+#endif
+
 	mmc_start_host(host);
-	register_pm_notifier(&host->pm_notify);
+	if (!(host->pm_flags & MMC_PM_IGNORE_PM_NOTIFY))
+		register_pm_notifier(&host->pm_notify);
 
 	return 0;
 }
@@ -412,13 +415,19 @@ EXPORT_SYMBOL(mmc_add_host);
  */
 void mmc_remove_host(struct mmc_host *host)
 {
-	unregister_pm_notifier(&host->pm_notify);
+	if (!(host->pm_flags & MMC_PM_IGNORE_PM_NOTIFY))
+		unregister_pm_notifier(&host->pm_notify);
+
 	mmc_stop_host(host);
 
 #ifdef CONFIG_DEBUG_FS
 	mmc_remove_host_debugfs(host);
 #endif
 
+#ifdef CONFIG_BLOCK
+	mmc_latency_hist_sysfs_exit(host);
+#endif
+
 	device_del(&host->class_dev);
 
 	led_trigger_unregister_simple(host->led);
diff --git a/drivers/mmc/core/host.h b/drivers/mmc/core/host.h
index 992bf5397633..bf38533406fd 100644
--- a/drivers/mmc/core/host.h
+++ b/drivers/mmc/core/host.h
@@ -12,6 +12,8 @@
 #define _MMC_CORE_HOST_H
 #include <linux/mmc/host.h>
 
+#define cls_dev_to_mmc_host(d)	container_of(d, struct mmc_host, class_dev)
+
 int mmc_register_host_class(void);
 void mmc_unregister_host_class(void);
 
@@ -21,5 +23,8 @@ void mmc_retune_hold(struct mmc_host *host);
 void mmc_retune_release(struct mmc_host *host);
 int mmc_retune(struct mmc_host *host);
 
+void mmc_latency_hist_sysfs_init(struct mmc_host *host);
+void mmc_latency_hist_sysfs_exit(struct mmc_host *host);
+
 #endif
 
diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c
index 79a0c26e1419..83b9bf880834 100644
--- a/drivers/mmc/core/mmc.c
+++ b/drivers/mmc/core/mmc.c
@@ -592,6 +592,12 @@ static int mmc_decode_ext_csd(struct mmc_card *card, u8 *ext_csd)
 		card->ext_csd.ffu_capable =
 			(ext_csd[EXT_CSD_SUPPORTED_MODE] & 0x1) &&
 			!(ext_csd[EXT_CSD_FW_CONFIG] & 0x1);
+
+		card->ext_csd.pre_eol_info = ext_csd[EXT_CSD_PRE_EOL_INFO];
+		card->ext_csd.device_life_time_est_typ_a =
+			ext_csd[EXT_CSD_DEVICE_LIFE_TIME_EST_TYP_A];
+		card->ext_csd.device_life_time_est_typ_b =
+			ext_csd[EXT_CSD_DEVICE_LIFE_TIME_EST_TYP_B];
 	}
 out:
 	return err;
@@ -721,6 +727,11 @@ MMC_DEV_ATTR(manfid, "0x%06x\n", card->cid.manfid);
 MMC_DEV_ATTR(name, "%s\n", card->cid.prod_name);
 MMC_DEV_ATTR(oemid, "0x%04x\n", card->cid.oemid);
 MMC_DEV_ATTR(prv, "0x%x\n", card->cid.prv);
+MMC_DEV_ATTR(rev, "0x%x\n", card->ext_csd.rev);
+MMC_DEV_ATTR(pre_eol_info, "%02x\n", card->ext_csd.pre_eol_info);
+MMC_DEV_ATTR(life_time, "0x%02x 0x%02x\n",
+	card->ext_csd.device_life_time_est_typ_a,
+	card->ext_csd.device_life_time_est_typ_b);
 MMC_DEV_ATTR(serial, "0x%08x\n", card->cid.serial);
 MMC_DEV_ATTR(enhanced_area_offset, "%llu\n",
 		card->ext_csd.enhanced_area_offset);
@@ -757,6 +768,9 @@ static struct attribute *mmc_std_attrs[] = {
 	&dev_attr_name.attr,
 	&dev_attr_oemid.attr,
 	&dev_attr_prv.attr,
+	&dev_attr_rev.attr,
+	&dev_attr_pre_eol_info.attr,
+	&dev_attr_life_time.attr,
 	&dev_attr_serial.attr,
 	&dev_attr_enhanced_area_offset.attr,
 	&dev_attr_enhanced_area_size.attr,
diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c
index 967535d76e34..ee145d4cc541 100644
--- a/drivers/mmc/core/sd.c
+++ b/drivers/mmc/core/sd.c
@@ -809,6 +809,9 @@ int mmc_sd_setup_card(struct mmc_host *host, struct mmc_card *card,
 	bool reinit)
 {
 	int err;
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+	int retries;
+#endif
 
 	if (!reinit) {
 		/*
@@ -835,7 +838,26 @@ int mmc_sd_setup_card(struct mmc_host *host, struct mmc_card *card,
 		/*
 		 * Fetch switch information from card.
 		 */
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+		for (retries = 1; retries <= 3; retries++) {
+			err = mmc_read_switch(card);
+			if (!err) {
+				if (retries > 1) {
+					printk(KERN_WARNING
+					       "%s: recovered\n",
+					       mmc_hostname(host));
+				}
+				break;
+			} else {
+				printk(KERN_WARNING
+				       "%s: read switch failed (attempt %d)\n",
+				       mmc_hostname(host), retries);
+			}
+		}
+#else
 		err = mmc_read_switch(card);
+#endif
+
 		if (err)
 			return err;
 	}
@@ -1033,7 +1055,10 @@ static int mmc_sd_alive(struct mmc_host *host)
  */
 static void mmc_sd_detect(struct mmc_host *host)
 {
-	int err;
+	int err = 0;
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+	int retries = 5;
+#endif
 
 	BUG_ON(!host);
 	BUG_ON(!host->card);
@@ -1043,7 +1068,23 @@ static void mmc_sd_detect(struct mmc_host *host)
 	/*
 	 * Just check if our card has been removed.
 	 */
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+	while(retries) {
+		err = mmc_send_status(host->card, NULL);
+		if (err) {
+			retries--;
+			udelay(5);
+			continue;
+		}
+		break;
+	}
+	if (!retries) {
+		printk(KERN_ERR "%s(%s): Unable to re-detect card (%d)\n",
+		       __func__, mmc_hostname(host), err);
+	}
+#else
 	err = _mmc_detect_card_removed(host);
+#endif
 
 	mmc_put_card(host->card);
 
@@ -1105,6 +1146,9 @@ static int mmc_sd_suspend(struct mmc_host *host)
 static int _mmc_sd_resume(struct mmc_host *host)
 {
 	int err = 0;
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+	int retries;
+#endif
 
 	BUG_ON(!host);
 	BUG_ON(!host->card);
@@ -1115,7 +1159,23 @@ static int _mmc_sd_resume(struct mmc_host *host)
 		goto out;
 
 	mmc_power_up(host, host->card->ocr);
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+	retries = 5;
+	while (retries) {
+		err = mmc_sd_init_card(host, host->card->ocr, host->card);
+
+		if (err) {
+			printk(KERN_ERR "%s: Re-init card rc = %d (retries = %d)\n",
+			       mmc_hostname(host), err, retries);
+			mdelay(5);
+			retries--;
+			continue;
+		}
+		break;
+	}
+#else
 	err = mmc_sd_init_card(host, host->card->ocr, host->card);
+#endif
 	mmc_card_clr_suspended(host->card);
 
 out:
@@ -1201,6 +1261,9 @@ int mmc_attach_sd(struct mmc_host *host)
 {
 	int err;
 	u32 ocr, rocr;
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+	int retries;
+#endif
 
 	BUG_ON(!host);
 	WARN_ON(!host->claimed);
@@ -1237,9 +1300,27 @@ int mmc_attach_sd(struct mmc_host *host)
 	/*
 	 * Detect and init the card.
 	 */
+#ifdef CONFIG_MMC_PARANOID_SD_INIT
+	retries = 5;
+	while (retries) {
+		err = mmc_sd_init_card(host, rocr, NULL);
+		if (err) {
+			retries--;
+			continue;
+		}
+		break;
+	}
+
+	if (!retries) {
+		printk(KERN_ERR "%s: mmc_sd_init_card() failure (err = %d)\n",
+		       mmc_hostname(host), err);
+		goto err;
+	}
+#else
 	err = mmc_sd_init_card(host, rocr, NULL);
 	if (err)
 		goto err;
+#endif
 
 	mmc_release_host(host);
 	err = mmc_add_card(host->card);
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c
index 467b3cf80c44..b47957122fd7 100644
--- a/drivers/mmc/core/sdio.c
+++ b/drivers/mmc/core/sdio.c
@@ -10,6 +10,7 @@
  */
 
 #include <linux/err.h>
+#include <linux/module.h>
 #include <linux/pm_runtime.h>
 
 #include <linux/mmc/host.h>
@@ -21,6 +22,7 @@
 
 #include "core.h"
 #include "bus.h"
+#include "host.h"
 #include "sd.h"
 #include "sdio_bus.h"
 #include "mmc_ops.h"
@@ -28,6 +30,10 @@
 #include "sdio_ops.h"
 #include "sdio_cis.h"
 
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+#include <linux/mmc/sdio_ids.h>
+#endif
+
 static int sdio_read_fbr(struct sdio_func *func)
 {
 	int ret;
@@ -699,19 +705,35 @@ try_again:
 		goto finish;
 	}
 
-	/*
-	 * Read the common registers.
-	 */
-	err = sdio_read_cccr(card, ocr);
-	if (err)
-		goto remove;
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+	if (host->embedded_sdio_data.cccr)
+		memcpy(&card->cccr, host->embedded_sdio_data.cccr, sizeof(struct sdio_cccr));
+	else {
+#endif
+		/*
+		 * Read the common registers.
+		 */
+		err = sdio_read_cccr(card,  ocr);
+		if (err)
+			goto remove;
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+	}
+#endif
 
-	/*
-	 * Read the common CIS tuples.
-	 */
-	err = sdio_read_common_cis(card);
-	if (err)
-		goto remove;
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+	if (host->embedded_sdio_data.cis)
+		memcpy(&card->cis, host->embedded_sdio_data.cis, sizeof(struct sdio_cis));
+	else {
+#endif
+		/*
+		 * Read the common CIS tuples.
+		 */
+		err = sdio_read_common_cis(card);
+		if (err)
+			goto remove;
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+	}
+#endif
 
 	if (oldcard) {
 		int same = (card->cis.vendor == oldcard->cis.vendor &&
@@ -1120,14 +1142,36 @@ int mmc_attach_sdio(struct mmc_host *host)
 	funcs = (ocr & 0x70000000) >> 28;
 	card->sdio_funcs = 0;
 
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+	if (host->embedded_sdio_data.funcs)
+		card->sdio_funcs = funcs = host->embedded_sdio_data.num_funcs;
+#endif
+
 	/*
 	 * Initialize (but don't add) all present functions.
 	 */
 	for (i = 0; i < funcs; i++, card->sdio_funcs++) {
-		err = sdio_init_func(host->card, i + 1);
-		if (err)
-			goto remove;
-
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+		if (host->embedded_sdio_data.funcs) {
+			struct sdio_func *tmp;
+
+			tmp = sdio_alloc_func(host->card);
+			if (IS_ERR(tmp))
+				goto remove;
+			tmp->num = (i + 1);
+			card->sdio_func[i] = tmp;
+			tmp->class = host->embedded_sdio_data.funcs[i].f_class;
+			tmp->max_blksize = host->embedded_sdio_data.funcs[i].f_maxblksize;
+			tmp->vendor = card->cis.vendor;
+			tmp->device = card->cis.device;
+		} else {
+#endif
+			err = sdio_init_func(host->card, i + 1);
+			if (err)
+				goto remove;
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+		}
+#endif
 		/*
 		 * Enable Runtime PM for this func (if supported)
 		 */
@@ -1175,3 +1219,42 @@ err:
 	return err;
 }
 
+int sdio_reset_comm(struct mmc_card *card)
+{
+	struct mmc_host *host = card->host;
+	u32 ocr;
+	u32 rocr;
+	int err;
+
+	printk("%s():\n", __func__);
+	mmc_claim_host(host);
+
+	mmc_retune_disable(host);
+
+	mmc_go_idle(host);
+
+	mmc_set_clock(host, host->f_min);
+
+	err = mmc_send_io_op_cond(host, 0, &ocr);
+	if (err)
+		goto err;
+
+	rocr = mmc_select_voltage(host, ocr);
+	if (!rocr) {
+		err = -EINVAL;
+		goto err;
+	}
+
+	err = mmc_sdio_init_card(host, rocr, card, 0);
+	if (err)
+		goto err;
+
+	mmc_release_host(host);
+	return 0;
+err:
+	printk("%s: Error resetting SDIO communications (%d)\n",
+	       mmc_hostname(host), err);
+	mmc_release_host(host);
+	return err;
+}
+EXPORT_SYMBOL(sdio_reset_comm);
diff --git a/drivers/mmc/core/sdio_bus.c b/drivers/mmc/core/sdio_bus.c
index c23bc4f331bd..60984899b135 100644
--- a/drivers/mmc/core/sdio_bus.c
+++ b/drivers/mmc/core/sdio_bus.c
@@ -28,6 +28,10 @@
 #include "sdio_cis.h"
 #include "sdio_bus.h"
 
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+#include <linux/mmc/host.h>
+#endif
+
 #define to_sdio_driver(d)	container_of(d, struct sdio_driver, drv)
 
 /* show configuration fields */
@@ -263,7 +267,14 @@ static void sdio_release_func(struct device *dev)
 {
 	struct sdio_func *func = dev_to_sdio_func(dev);
 
-	sdio_free_func_cis(func);
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+	/*
+	 * If this device is embedded then we never allocated
+	 * cis tables for this func
+	 */
+	if (!func->card->host->embedded_sdio_data.funcs)
+#endif
+		sdio_free_func_cis(func);
 
 	kfree(func->info);
 	kfree(func->tmpbuf);
diff --git a/drivers/mmc/core/sdio_io.c b/drivers/mmc/core/sdio_io.c
index 78cb4d5d9d58..8fdeb07723a6 100644
--- a/drivers/mmc/core/sdio_io.c
+++ b/drivers/mmc/core/sdio_io.c
@@ -384,6 +384,39 @@ u8 sdio_readb(struct sdio_func *func, unsigned int addr, int *err_ret)
 EXPORT_SYMBOL_GPL(sdio_readb);
 
 /**
+ *	sdio_readb_ext - read a single byte from a SDIO function
+ *	@func: SDIO function to access
+ *	@addr: address to read
+ *	@err_ret: optional status value from transfer
+ *	@in: value to add to argument
+ *
+ *	Reads a single byte from the address space of a given SDIO
+ *	function. If there is a problem reading the address, 0xff
+ *	is returned and @err_ret will contain the error code.
+ */
+unsigned char sdio_readb_ext(struct sdio_func *func, unsigned int addr,
+	int *err_ret, unsigned in)
+{
+	int ret;
+	unsigned char val;
+
+	BUG_ON(!func);
+
+	if (err_ret)
+		*err_ret = 0;
+
+	ret = mmc_io_rw_direct(func->card, 0, func->num, addr, (u8)in, &val);
+	if (ret) {
+		if (err_ret)
+			*err_ret = ret;
+		return 0xFF;
+	}
+
+	return val;
+}
+EXPORT_SYMBOL_GPL(sdio_readb_ext);
+
+/**
  *	sdio_writeb - write a single byte to a SDIO function
  *	@func: SDIO function to access
  *	@b: byte to write
diff --git a/drivers/mtd/nand/Kconfig b/drivers/mtd/nand/Kconfig
index 8f49f8aeff1a..132d0d04726c 100644
--- a/drivers/mtd/nand/Kconfig
+++ b/drivers/mtd/nand/Kconfig
@@ -1,3 +1,10 @@
+config MTD_NAND_IDS
+	tristate "Include chip ids for known NAND devices."
+	depends on MTD
+	help
+	  Useful for NAND drivers that do not use the NAND subsystem but
+	  still like to take advantage of the known chip information.
+
 config MTD_NAND_ECC
 	tristate
 
@@ -108,9 +115,6 @@ config MTD_NAND_OMAP_BCH
 config MTD_NAND_OMAP_BCH_BUILD
 	def_tristate MTD_NAND_OMAP2 && MTD_NAND_OMAP_BCH
 
-config MTD_NAND_IDS
-	tristate
-
 config MTD_NAND_RICOH
 	tristate "Ricoh xD card reader"
 	default n
diff --git a/drivers/net/ppp/Kconfig b/drivers/net/ppp/Kconfig
index 1373c6d7278d..282aec4860eb 100644
--- a/drivers/net/ppp/Kconfig
+++ b/drivers/net/ppp/Kconfig
@@ -149,6 +149,23 @@ config PPPOL2TP
 	  tunnels. L2TP is replacing PPTP for VPN uses.
 if TTY
 
+config PPPOLAC
+	tristate "PPP on L2TP Access Concentrator"
+	depends on PPP && INET
+	help
+	  L2TP (RFC 2661) is a tunneling protocol widely used in virtual private
+	  networks. This driver handles L2TP data packets between a UDP socket
+	  and a PPP channel, but only permits one session per socket. Thus it is
+	  fairly simple and suited for clients.
+
+config PPPOPNS
+	tristate "PPP on PPTP Network Server"
+	depends on PPP && INET
+	help
+	  PPTP (RFC 2637) is a tunneling protocol widely used in virtual private
+	  networks. This driver handles PPTP data packets between a RAW socket
+	  and a PPP channel. It is fairly simple and easy to use.
+
 config PPP_ASYNC
 	tristate "PPP support for async serial ports"
 	depends on PPP
diff --git a/drivers/net/ppp/Makefile b/drivers/net/ppp/Makefile
index a6b6297b0066..d283d03c4683 100644
--- a/drivers/net/ppp/Makefile
+++ b/drivers/net/ppp/Makefile
@@ -11,3 +11,5 @@ obj-$(CONFIG_PPP_SYNC_TTY) += ppp_synctty.o
 obj-$(CONFIG_PPPOE) += pppox.o pppoe.o
 obj-$(CONFIG_PPPOL2TP) += pppox.o
 obj-$(CONFIG_PPTP) += pppox.o pptp.o
+obj-$(CONFIG_PPPOLAC) += pppox.o pppolac.o
+obj-$(CONFIG_PPPOPNS) += pppox.o pppopns.o
diff --git a/drivers/net/ppp/pppolac.c b/drivers/net/ppp/pppolac.c
new file mode 100644
index 000000000000..3a45cf805288
--- /dev/null
+++ b/drivers/net/ppp/pppolac.c
@@ -0,0 +1,450 @@
+/* drivers/net/pppolac.c
+ *
+ * Driver for PPP on L2TP Access Concentrator / PPPoLAC Socket (RFC 2661)
+ *
+ * Copyright (C) 2009 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+/* This driver handles L2TP data packets between a UDP socket and a PPP channel.
+ * The socket must keep connected, and only one session per socket is permitted.
+ * Sequencing of outgoing packets is controlled by LNS. Incoming packets with
+ * sequences are reordered within a sliding window of one second. Currently
+ * reordering only happens when a packet is received. It is done for simplicity
+ * since no additional locks or threads are required. This driver only works on
+ * IPv4 due to the lack of UDP encapsulation support in IPv6. */
+
+#include <linux/module.h>
+#include <linux/jiffies.h>
+#include <linux/workqueue.h>
+#include <linux/skbuff.h>
+#include <linux/file.h>
+#include <linux/netdevice.h>
+#include <linux/net.h>
+#include <linux/udp.h>
+#include <linux/ppp_defs.h>
+#include <linux/if_ppp.h>
+#include <linux/if_pppox.h>
+#include <linux/ppp_channel.h>
+#include <net/tcp_states.h>
+#include <asm/uaccess.h>
+
+#define L2TP_CONTROL_BIT	0x80
+#define L2TP_LENGTH_BIT		0x40
+#define L2TP_SEQUENCE_BIT	0x08
+#define L2TP_OFFSET_BIT		0x02
+#define L2TP_VERSION		0x02
+#define L2TP_VERSION_MASK	0x0F
+
+#define PPP_ADDR	0xFF
+#define PPP_CTRL	0x03
+
+union unaligned {
+	__u32 u32;
+} __attribute__((packed));
+
+static inline union unaligned *unaligned(void *ptr)
+{
+	return (union unaligned *)ptr;
+}
+
+struct meta {
+	__u32 sequence;
+	__u32 timestamp;
+};
+
+static inline struct meta *skb_meta(struct sk_buff *skb)
+{
+	return (struct meta *)skb->cb;
+}
+
+/******************************************************************************/
+
+static int pppolac_recv_core(struct sock *sk_udp, struct sk_buff *skb)
+{
+	struct sock *sk = (struct sock *)sk_udp->sk_user_data;
+	struct pppolac_opt *opt = &pppox_sk(sk)->proto.lac;
+	struct meta *meta = skb_meta(skb);
+	__u32 now = jiffies;
+	__u8 bits;
+	__u8 *ptr;
+
+	/* Drop the packet if L2TP header is missing. */
+	if (skb->len < sizeof(struct udphdr) + 6)
+		goto drop;
+
+	/* Put it back if it is a control packet. */
+	if (skb->data[sizeof(struct udphdr)] & L2TP_CONTROL_BIT)
+		return opt->backlog_rcv(sk_udp, skb);
+
+	/* Skip UDP header. */
+	skb_pull(skb, sizeof(struct udphdr));
+
+	/* Check the version. */
+	if ((skb->data[1] & L2TP_VERSION_MASK) != L2TP_VERSION)
+		goto drop;
+	bits = skb->data[0];
+	ptr = &skb->data[2];
+
+	/* Check the length if it is present. */
+	if (bits & L2TP_LENGTH_BIT) {
+		if ((ptr[0] << 8 | ptr[1]) != skb->len)
+			goto drop;
+		ptr += 2;
+	}
+
+	/* Skip all fields including optional ones. */
+	if (!skb_pull(skb, 6 + (bits & L2TP_SEQUENCE_BIT ? 4 : 0) +
+			(bits & L2TP_LENGTH_BIT ? 2 : 0) +
+			(bits & L2TP_OFFSET_BIT ? 2 : 0)))
+		goto drop;
+
+	/* Skip the offset padding if it is present. */
+	if (bits & L2TP_OFFSET_BIT &&
+			!skb_pull(skb, skb->data[-2] << 8 | skb->data[-1]))
+		goto drop;
+
+	/* Check the tunnel and the session. */
+	if (unaligned(ptr)->u32 != opt->local)
+		goto drop;
+
+	/* Check the sequence if it is present. */
+	if (bits & L2TP_SEQUENCE_BIT) {
+		meta->sequence = ptr[4] << 8 | ptr[5];
+		if ((__s16)(meta->sequence - opt->recv_sequence) < 0)
+			goto drop;
+	}
+
+	/* Skip PPP address and control if they are present. */
+	if (skb->len >= 2 && skb->data[0] == PPP_ADDR &&
+			skb->data[1] == PPP_CTRL)
+		skb_pull(skb, 2);
+
+	/* Fix PPP protocol if it is compressed. */
+	if (skb->len >= 1 && skb->data[0] & 1)
+		skb_push(skb, 1)[0] = 0;
+
+	/* Drop the packet if PPP protocol is missing. */
+	if (skb->len < 2)
+		goto drop;
+
+	/* Perform reordering if sequencing is enabled. */
+	atomic_set(&opt->sequencing, bits & L2TP_SEQUENCE_BIT);
+	if (bits & L2TP_SEQUENCE_BIT) {
+		struct sk_buff *skb1;
+
+		/* Insert the packet into receive queue in order. */
+		skb_set_owner_r(skb, sk);
+		skb_queue_walk(&sk->sk_receive_queue, skb1) {
+			struct meta *meta1 = skb_meta(skb1);
+			__s16 order = meta->sequence - meta1->sequence;
+			if (order == 0)
+				goto drop;
+			if (order < 0) {
+				meta->timestamp = meta1->timestamp;
+				skb_insert(skb1, skb, &sk->sk_receive_queue);
+				skb = NULL;
+				break;
+			}
+		}
+		if (skb) {
+			meta->timestamp = now;
+			skb_queue_tail(&sk->sk_receive_queue, skb);
+		}
+
+		/* Remove packets from receive queue as long as
+		 * 1. the receive buffer is full,
+		 * 2. they are queued longer than one second, or
+		 * 3. there are no missing packets before them. */
+		skb_queue_walk_safe(&sk->sk_receive_queue, skb, skb1) {
+			meta = skb_meta(skb);
+			if (atomic_read(&sk->sk_rmem_alloc) < sk->sk_rcvbuf &&
+					now - meta->timestamp < HZ &&
+					meta->sequence != opt->recv_sequence)
+				break;
+			skb_unlink(skb, &sk->sk_receive_queue);
+			opt->recv_sequence = (__u16)(meta->sequence + 1);
+			skb_orphan(skb);
+			ppp_input(&pppox_sk(sk)->chan, skb);
+		}
+		return NET_RX_SUCCESS;
+	}
+
+	/* Flush receive queue if sequencing is disabled. */
+	skb_queue_purge(&sk->sk_receive_queue);
+	skb_orphan(skb);
+	ppp_input(&pppox_sk(sk)->chan, skb);
+	return NET_RX_SUCCESS;
+drop:
+	kfree_skb(skb);
+	return NET_RX_DROP;
+}
+
+static int pppolac_recv(struct sock *sk_udp, struct sk_buff *skb)
+{
+	sock_hold(sk_udp);
+	sk_receive_skb(sk_udp, skb, 0);
+	return 0;
+}
+
+static struct sk_buff_head delivery_queue;
+
+static void pppolac_xmit_core(struct work_struct *delivery_work)
+{
+	mm_segment_t old_fs = get_fs();
+	struct sk_buff *skb;
+
+	set_fs(KERNEL_DS);
+	while ((skb = skb_dequeue(&delivery_queue))) {
+		struct sock *sk_udp = skb->sk;
+		struct kvec iov = {.iov_base = skb->data, .iov_len = skb->len};
+		struct msghdr msg = {
+			.msg_flags = MSG_NOSIGNAL | MSG_DONTWAIT,
+		};
+
+		iov_iter_kvec(&msg.msg_iter, WRITE | ITER_KVEC, &iov, 1,
+			      skb->len);
+		sk_udp->sk_prot->sendmsg(sk_udp, &msg, skb->len);
+		kfree_skb(skb);
+	}
+	set_fs(old_fs);
+}
+
+static DECLARE_WORK(delivery_work, pppolac_xmit_core);
+
+static int pppolac_xmit(struct ppp_channel *chan, struct sk_buff *skb)
+{
+	struct sock *sk_udp = (struct sock *)chan->private;
+	struct pppolac_opt *opt = &pppox_sk(sk_udp->sk_user_data)->proto.lac;
+
+	/* Install PPP address and control. */
+	skb_push(skb, 2);
+	skb->data[0] = PPP_ADDR;
+	skb->data[1] = PPP_CTRL;
+
+	/* Install L2TP header. */
+	if (atomic_read(&opt->sequencing)) {
+		skb_push(skb, 10);
+		skb->data[0] = L2TP_SEQUENCE_BIT;
+		skb->data[6] = opt->xmit_sequence >> 8;
+		skb->data[7] = opt->xmit_sequence;
+		skb->data[8] = 0;
+		skb->data[9] = 0;
+		opt->xmit_sequence++;
+	} else {
+		skb_push(skb, 6);
+		skb->data[0] = 0;
+	}
+	skb->data[1] = L2TP_VERSION;
+	unaligned(&skb->data[2])->u32 = opt->remote;
+
+	/* Now send the packet via the delivery queue. */
+	skb_set_owner_w(skb, sk_udp);
+	skb_queue_tail(&delivery_queue, skb);
+	schedule_work(&delivery_work);
+	return 1;
+}
+
+/******************************************************************************/
+
+static struct ppp_channel_ops pppolac_channel_ops = {
+	.start_xmit = pppolac_xmit,
+};
+
+static int pppolac_connect(struct socket *sock, struct sockaddr *useraddr,
+	int addrlen, int flags)
+{
+	struct sock *sk = sock->sk;
+	struct pppox_sock *po = pppox_sk(sk);
+	struct sockaddr_pppolac *addr = (struct sockaddr_pppolac *)useraddr;
+	struct socket *sock_udp = NULL;
+	struct sock *sk_udp;
+	int error;
+
+	if (addrlen != sizeof(struct sockaddr_pppolac) ||
+			!addr->local.tunnel || !addr->local.session ||
+			!addr->remote.tunnel || !addr->remote.session) {
+		return -EINVAL;
+	}
+
+	lock_sock(sk);
+	error = -EALREADY;
+	if (sk->sk_state != PPPOX_NONE)
+		goto out;
+
+	sock_udp = sockfd_lookup(addr->udp_socket, &error);
+	if (!sock_udp)
+		goto out;
+	sk_udp = sock_udp->sk;
+	lock_sock(sk_udp);
+
+	/* Remove this check when IPv6 supports UDP encapsulation. */
+	error = -EAFNOSUPPORT;
+	if (sk_udp->sk_family != AF_INET)
+		goto out;
+	error = -EPROTONOSUPPORT;
+	if (sk_udp->sk_protocol != IPPROTO_UDP)
+		goto out;
+	error = -EDESTADDRREQ;
+	if (sk_udp->sk_state != TCP_ESTABLISHED)
+		goto out;
+	error = -EBUSY;
+	if (udp_sk(sk_udp)->encap_type || sk_udp->sk_user_data)
+		goto out;
+	if (!sk_udp->sk_bound_dev_if) {
+		struct dst_entry *dst = sk_dst_get(sk_udp);
+		error = -ENODEV;
+		if (!dst)
+			goto out;
+		sk_udp->sk_bound_dev_if = dst->dev->ifindex;
+		dst_release(dst);
+	}
+
+	po->chan.hdrlen = 12;
+	po->chan.private = sk_udp;
+	po->chan.ops = &pppolac_channel_ops;
+	po->chan.mtu = PPP_MRU - 80;
+	po->proto.lac.local = unaligned(&addr->local)->u32;
+	po->proto.lac.remote = unaligned(&addr->remote)->u32;
+	atomic_set(&po->proto.lac.sequencing, 1);
+	po->proto.lac.backlog_rcv = sk_udp->sk_backlog_rcv;
+
+	error = ppp_register_channel(&po->chan);
+	if (error)
+		goto out;
+
+	sk->sk_state = PPPOX_CONNECTED;
+	udp_sk(sk_udp)->encap_type = UDP_ENCAP_L2TPINUDP;
+	udp_sk(sk_udp)->encap_rcv = pppolac_recv;
+	sk_udp->sk_backlog_rcv = pppolac_recv_core;
+	sk_udp->sk_user_data = sk;
+out:
+	if (sock_udp) {
+		release_sock(sk_udp);
+		if (error)
+			sockfd_put(sock_udp);
+	}
+	release_sock(sk);
+	return error;
+}
+
+static int pppolac_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+
+	if (!sk)
+		return 0;
+
+	lock_sock(sk);
+	if (sock_flag(sk, SOCK_DEAD)) {
+		release_sock(sk);
+		return -EBADF;
+	}
+
+	if (sk->sk_state != PPPOX_NONE) {
+		struct sock *sk_udp = (struct sock *)pppox_sk(sk)->chan.private;
+		lock_sock(sk_udp);
+		skb_queue_purge(&sk->sk_receive_queue);
+		pppox_unbind_sock(sk);
+		udp_sk(sk_udp)->encap_type = 0;
+		udp_sk(sk_udp)->encap_rcv = NULL;
+		sk_udp->sk_backlog_rcv = pppox_sk(sk)->proto.lac.backlog_rcv;
+		sk_udp->sk_user_data = NULL;
+		release_sock(sk_udp);
+		sockfd_put(sk_udp->sk_socket);
+	}
+
+	sock_orphan(sk);
+	sock->sk = NULL;
+	release_sock(sk);
+	sock_put(sk);
+	return 0;
+}
+
+/******************************************************************************/
+
+static struct proto pppolac_proto = {
+	.name = "PPPOLAC",
+	.owner = THIS_MODULE,
+	.obj_size = sizeof(struct pppox_sock),
+};
+
+static struct proto_ops pppolac_proto_ops = {
+	.family = PF_PPPOX,
+	.owner = THIS_MODULE,
+	.release = pppolac_release,
+	.bind = sock_no_bind,
+	.connect = pppolac_connect,
+	.socketpair = sock_no_socketpair,
+	.accept = sock_no_accept,
+	.getname = sock_no_getname,
+	.poll = sock_no_poll,
+	.ioctl = pppox_ioctl,
+	.listen = sock_no_listen,
+	.shutdown = sock_no_shutdown,
+	.setsockopt = sock_no_setsockopt,
+	.getsockopt = sock_no_getsockopt,
+	.sendmsg = sock_no_sendmsg,
+	.recvmsg = sock_no_recvmsg,
+	.mmap = sock_no_mmap,
+};
+
+static int pppolac_create(struct net *net, struct socket *sock, int kern)
+{
+	struct sock *sk;
+
+	sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pppolac_proto, kern);
+	if (!sk)
+		return -ENOMEM;
+
+	sock_init_data(sock, sk);
+	sock->state = SS_UNCONNECTED;
+	sock->ops = &pppolac_proto_ops;
+	sk->sk_protocol = PX_PROTO_OLAC;
+	sk->sk_state = PPPOX_NONE;
+	return 0;
+}
+
+/******************************************************************************/
+
+static struct pppox_proto pppolac_pppox_proto = {
+	.create = pppolac_create,
+	.owner = THIS_MODULE,
+};
+
+static int __init pppolac_init(void)
+{
+	int error;
+
+	error = proto_register(&pppolac_proto, 0);
+	if (error)
+		return error;
+
+	error = register_pppox_proto(PX_PROTO_OLAC, &pppolac_pppox_proto);
+	if (error)
+		proto_unregister(&pppolac_proto);
+	else
+		skb_queue_head_init(&delivery_queue);
+	return error;
+}
+
+static void __exit pppolac_exit(void)
+{
+	unregister_pppox_proto(PX_PROTO_OLAC);
+	proto_unregister(&pppolac_proto);
+}
+
+module_init(pppolac_init);
+module_exit(pppolac_exit);
+
+MODULE_DESCRIPTION("PPP on L2TP Access Concentrator (PPPoLAC)");
+MODULE_AUTHOR("Chia-chi Yeh <chiachi@android.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/ppp/pppopns.c b/drivers/net/ppp/pppopns.c
new file mode 100644
index 000000000000..cdb4fa1af734
--- /dev/null
+++ b/drivers/net/ppp/pppopns.c
@@ -0,0 +1,429 @@
+/* drivers/net/pppopns.c
+ *
+ * Driver for PPP on PPTP Network Server / PPPoPNS Socket (RFC 2637)
+ *
+ * Copyright (C) 2009 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+/* This driver handles PPTP data packets between a RAW socket and a PPP channel.
+ * The socket is created in the kernel space and connected to the same address
+ * of the control socket. Outgoing packets are always sent with sequences but
+ * without acknowledgements. Incoming packets with sequences are reordered
+ * within a sliding window of one second. Currently reordering only happens when
+ * a packet is received. It is done for simplicity since no additional locks or
+ * threads are required. This driver should work on both IPv4 and IPv6. */
+
+#include <linux/module.h>
+#include <linux/jiffies.h>
+#include <linux/workqueue.h>
+#include <linux/skbuff.h>
+#include <linux/file.h>
+#include <linux/netdevice.h>
+#include <linux/net.h>
+#include <linux/ppp_defs.h>
+#include <linux/if.h>
+#include <linux/if_ppp.h>
+#include <linux/if_pppox.h>
+#include <linux/ppp_channel.h>
+#include <asm/uaccess.h>
+
+#define GRE_HEADER_SIZE		8
+
+#define PPTP_GRE_BITS		htons(0x2001)
+#define PPTP_GRE_BITS_MASK	htons(0xEF7F)
+#define PPTP_GRE_SEQ_BIT	htons(0x1000)
+#define PPTP_GRE_ACK_BIT	htons(0x0080)
+#define PPTP_GRE_TYPE		htons(0x880B)
+
+#define PPP_ADDR	0xFF
+#define PPP_CTRL	0x03
+
+struct header {
+	__u16	bits;
+	__u16	type;
+	__u16	length;
+	__u16	call;
+	__u32	sequence;
+} __attribute__((packed));
+
+struct meta {
+	__u32 sequence;
+	__u32 timestamp;
+};
+
+static inline struct meta *skb_meta(struct sk_buff *skb)
+{
+	return (struct meta *)skb->cb;
+}
+
+/******************************************************************************/
+
+static int pppopns_recv_core(struct sock *sk_raw, struct sk_buff *skb)
+{
+	struct sock *sk = (struct sock *)sk_raw->sk_user_data;
+	struct pppopns_opt *opt = &pppox_sk(sk)->proto.pns;
+	struct meta *meta = skb_meta(skb);
+	__u32 now = jiffies;
+	struct header *hdr;
+
+	/* Skip transport header */
+	skb_pull(skb, skb_transport_header(skb) - skb->data);
+
+	/* Drop the packet if GRE header is missing. */
+	if (skb->len < GRE_HEADER_SIZE)
+		goto drop;
+	hdr = (struct header *)skb->data;
+
+	/* Check the header. */
+	if (hdr->type != PPTP_GRE_TYPE || hdr->call != opt->local ||
+			(hdr->bits & PPTP_GRE_BITS_MASK) != PPTP_GRE_BITS)
+		goto drop;
+
+	/* Skip all fields including optional ones. */
+	if (!skb_pull(skb, GRE_HEADER_SIZE +
+			(hdr->bits & PPTP_GRE_SEQ_BIT ? 4 : 0) +
+			(hdr->bits & PPTP_GRE_ACK_BIT ? 4 : 0)))
+		goto drop;
+
+	/* Check the length. */
+	if (skb->len != ntohs(hdr->length))
+		goto drop;
+
+	/* Check the sequence if it is present. */
+	if (hdr->bits & PPTP_GRE_SEQ_BIT) {
+		meta->sequence = ntohl(hdr->sequence);
+		if ((__s32)(meta->sequence - opt->recv_sequence) < 0)
+			goto drop;
+	}
+
+	/* Skip PPP address and control if they are present. */
+	if (skb->len >= 2 && skb->data[0] == PPP_ADDR &&
+			skb->data[1] == PPP_CTRL)
+		skb_pull(skb, 2);
+
+	/* Fix PPP protocol if it is compressed. */
+	if (skb->len >= 1 && skb->data[0] & 1)
+		skb_push(skb, 1)[0] = 0;
+
+	/* Drop the packet if PPP protocol is missing. */
+	if (skb->len < 2)
+		goto drop;
+
+	/* Perform reordering if sequencing is enabled. */
+	if (hdr->bits & PPTP_GRE_SEQ_BIT) {
+		struct sk_buff *skb1;
+
+		/* Insert the packet into receive queue in order. */
+		skb_set_owner_r(skb, sk);
+		skb_queue_walk(&sk->sk_receive_queue, skb1) {
+			struct meta *meta1 = skb_meta(skb1);
+			__s32 order = meta->sequence - meta1->sequence;
+			if (order == 0)
+				goto drop;
+			if (order < 0) {
+				meta->timestamp = meta1->timestamp;
+				skb_insert(skb1, skb, &sk->sk_receive_queue);
+				skb = NULL;
+				break;
+			}
+		}
+		if (skb) {
+			meta->timestamp = now;
+			skb_queue_tail(&sk->sk_receive_queue, skb);
+		}
+
+		/* Remove packets from receive queue as long as
+		 * 1. the receive buffer is full,
+		 * 2. they are queued longer than one second, or
+		 * 3. there are no missing packets before them. */
+		skb_queue_walk_safe(&sk->sk_receive_queue, skb, skb1) {
+			meta = skb_meta(skb);
+			if (atomic_read(&sk->sk_rmem_alloc) < sk->sk_rcvbuf &&
+					now - meta->timestamp < HZ &&
+					meta->sequence != opt->recv_sequence)
+				break;
+			skb_unlink(skb, &sk->sk_receive_queue);
+			opt->recv_sequence = meta->sequence + 1;
+			skb_orphan(skb);
+			ppp_input(&pppox_sk(sk)->chan, skb);
+		}
+		return NET_RX_SUCCESS;
+	}
+
+	/* Flush receive queue if sequencing is disabled. */
+	skb_queue_purge(&sk->sk_receive_queue);
+	skb_orphan(skb);
+	ppp_input(&pppox_sk(sk)->chan, skb);
+	return NET_RX_SUCCESS;
+drop:
+	kfree_skb(skb);
+	return NET_RX_DROP;
+}
+
+static void pppopns_recv(struct sock *sk_raw)
+{
+	struct sk_buff *skb;
+	while ((skb = skb_dequeue(&sk_raw->sk_receive_queue))) {
+		sock_hold(sk_raw);
+		sk_receive_skb(sk_raw, skb, 0);
+	}
+}
+
+static struct sk_buff_head delivery_queue;
+
+static void pppopns_xmit_core(struct work_struct *delivery_work)
+{
+	mm_segment_t old_fs = get_fs();
+	struct sk_buff *skb;
+
+	set_fs(KERNEL_DS);
+	while ((skb = skb_dequeue(&delivery_queue))) {
+		struct sock *sk_raw = skb->sk;
+		struct kvec iov = {.iov_base = skb->data, .iov_len = skb->len};
+		struct msghdr msg = {
+			.msg_flags = MSG_NOSIGNAL | MSG_DONTWAIT,
+		};
+
+		iov_iter_kvec(&msg.msg_iter, WRITE | ITER_KVEC, &iov, 1,
+			      skb->len);
+		sk_raw->sk_prot->sendmsg(sk_raw, &msg, skb->len);
+		kfree_skb(skb);
+	}
+	set_fs(old_fs);
+}
+
+static DECLARE_WORK(delivery_work, pppopns_xmit_core);
+
+static int pppopns_xmit(struct ppp_channel *chan, struct sk_buff *skb)
+{
+	struct sock *sk_raw = (struct sock *)chan->private;
+	struct pppopns_opt *opt = &pppox_sk(sk_raw->sk_user_data)->proto.pns;
+	struct header *hdr;
+	__u16 length;
+
+	/* Install PPP address and control. */
+	skb_push(skb, 2);
+	skb->data[0] = PPP_ADDR;
+	skb->data[1] = PPP_CTRL;
+	length = skb->len;
+
+	/* Install PPTP GRE header. */
+	hdr = (struct header *)skb_push(skb, 12);
+	hdr->bits = PPTP_GRE_BITS | PPTP_GRE_SEQ_BIT;
+	hdr->type = PPTP_GRE_TYPE;
+	hdr->length = htons(length);
+	hdr->call = opt->remote;
+	hdr->sequence = htonl(opt->xmit_sequence);
+	opt->xmit_sequence++;
+
+	/* Now send the packet via the delivery queue. */
+	skb_set_owner_w(skb, sk_raw);
+	skb_queue_tail(&delivery_queue, skb);
+	schedule_work(&delivery_work);
+	return 1;
+}
+
+/******************************************************************************/
+
+static struct ppp_channel_ops pppopns_channel_ops = {
+	.start_xmit = pppopns_xmit,
+};
+
+static int pppopns_connect(struct socket *sock, struct sockaddr *useraddr,
+	int addrlen, int flags)
+{
+	struct sock *sk = sock->sk;
+	struct pppox_sock *po = pppox_sk(sk);
+	struct sockaddr_pppopns *addr = (struct sockaddr_pppopns *)useraddr;
+	struct sockaddr_storage ss;
+	struct socket *sock_tcp = NULL;
+	struct socket *sock_raw = NULL;
+	struct sock *sk_tcp;
+	struct sock *sk_raw;
+	int error;
+
+	if (addrlen != sizeof(struct sockaddr_pppopns))
+		return -EINVAL;
+
+	lock_sock(sk);
+	error = -EALREADY;
+	if (sk->sk_state != PPPOX_NONE)
+		goto out;
+
+	sock_tcp = sockfd_lookup(addr->tcp_socket, &error);
+	if (!sock_tcp)
+		goto out;
+	sk_tcp = sock_tcp->sk;
+	error = -EPROTONOSUPPORT;
+	if (sk_tcp->sk_protocol != IPPROTO_TCP)
+		goto out;
+	addrlen = sizeof(struct sockaddr_storage);
+	error = kernel_getpeername(sock_tcp, (struct sockaddr *)&ss, &addrlen);
+	if (error)
+		goto out;
+	if (!sk_tcp->sk_bound_dev_if) {
+		struct dst_entry *dst = sk_dst_get(sk_tcp);
+		error = -ENODEV;
+		if (!dst)
+			goto out;
+		sk_tcp->sk_bound_dev_if = dst->dev->ifindex;
+		dst_release(dst);
+	}
+
+	error = sock_create(ss.ss_family, SOCK_RAW, IPPROTO_GRE, &sock_raw);
+	if (error)
+		goto out;
+	sk_raw = sock_raw->sk;
+	sk_raw->sk_bound_dev_if = sk_tcp->sk_bound_dev_if;
+	error = kernel_connect(sock_raw, (struct sockaddr *)&ss, addrlen, 0);
+	if (error)
+		goto out;
+
+	po->chan.hdrlen = 14;
+	po->chan.private = sk_raw;
+	po->chan.ops = &pppopns_channel_ops;
+	po->chan.mtu = PPP_MRU - 80;
+	po->proto.pns.local = addr->local;
+	po->proto.pns.remote = addr->remote;
+	po->proto.pns.data_ready = sk_raw->sk_data_ready;
+	po->proto.pns.backlog_rcv = sk_raw->sk_backlog_rcv;
+
+	error = ppp_register_channel(&po->chan);
+	if (error)
+		goto out;
+
+	sk->sk_state = PPPOX_CONNECTED;
+	lock_sock(sk_raw);
+	sk_raw->sk_data_ready = pppopns_recv;
+	sk_raw->sk_backlog_rcv = pppopns_recv_core;
+	sk_raw->sk_user_data = sk;
+	release_sock(sk_raw);
+out:
+	if (sock_tcp)
+		sockfd_put(sock_tcp);
+	if (error && sock_raw)
+		sock_release(sock_raw);
+	release_sock(sk);
+	return error;
+}
+
+static int pppopns_release(struct socket *sock)
+{
+	struct sock *sk = sock->sk;
+
+	if (!sk)
+		return 0;
+
+	lock_sock(sk);
+	if (sock_flag(sk, SOCK_DEAD)) {
+		release_sock(sk);
+		return -EBADF;
+	}
+
+	if (sk->sk_state != PPPOX_NONE) {
+		struct sock *sk_raw = (struct sock *)pppox_sk(sk)->chan.private;
+		lock_sock(sk_raw);
+		skb_queue_purge(&sk->sk_receive_queue);
+		pppox_unbind_sock(sk);
+		sk_raw->sk_data_ready = pppox_sk(sk)->proto.pns.data_ready;
+		sk_raw->sk_backlog_rcv = pppox_sk(sk)->proto.pns.backlog_rcv;
+		sk_raw->sk_user_data = NULL;
+		release_sock(sk_raw);
+		sock_release(sk_raw->sk_socket);
+	}
+
+	sock_orphan(sk);
+	sock->sk = NULL;
+	release_sock(sk);
+	sock_put(sk);
+	return 0;
+}
+
+/******************************************************************************/
+
+static struct proto pppopns_proto = {
+	.name = "PPPOPNS",
+	.owner = THIS_MODULE,
+	.obj_size = sizeof(struct pppox_sock),
+};
+
+static struct proto_ops pppopns_proto_ops = {
+	.family = PF_PPPOX,
+	.owner = THIS_MODULE,
+	.release = pppopns_release,
+	.bind = sock_no_bind,
+	.connect = pppopns_connect,
+	.socketpair = sock_no_socketpair,
+	.accept = sock_no_accept,
+	.getname = sock_no_getname,
+	.poll = sock_no_poll,
+	.ioctl = pppox_ioctl,
+	.listen = sock_no_listen,
+	.shutdown = sock_no_shutdown,
+	.setsockopt = sock_no_setsockopt,
+	.getsockopt = sock_no_getsockopt,
+	.sendmsg = sock_no_sendmsg,
+	.recvmsg = sock_no_recvmsg,
+	.mmap = sock_no_mmap,
+};
+
+static int pppopns_create(struct net *net, struct socket *sock, int kern)
+{
+	struct sock *sk;
+
+	sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pppopns_proto, kern);
+	if (!sk)
+		return -ENOMEM;
+
+	sock_init_data(sock, sk);
+	sock->state = SS_UNCONNECTED;
+	sock->ops = &pppopns_proto_ops;
+	sk->sk_protocol = PX_PROTO_OPNS;
+	sk->sk_state = PPPOX_NONE;
+	return 0;
+}
+
+/******************************************************************************/
+
+static struct pppox_proto pppopns_pppox_proto = {
+	.create = pppopns_create,
+	.owner = THIS_MODULE,
+};
+
+static int __init pppopns_init(void)
+{
+	int error;
+
+	error = proto_register(&pppopns_proto, 0);
+	if (error)
+		return error;
+
+	error = register_pppox_proto(PX_PROTO_OPNS, &pppopns_pppox_proto);
+	if (error)
+		proto_unregister(&pppopns_proto);
+	else
+		skb_queue_head_init(&delivery_queue);
+	return error;
+}
+
+static void __exit pppopns_exit(void)
+{
+	unregister_pppox_proto(PX_PROTO_OPNS);
+	proto_unregister(&pppopns_proto);
+}
+
+module_init(pppopns_init);
+module_exit(pppopns_exit);
+
+MODULE_DESCRIPTION("PPP on PPTP Network Server (PPPoPNS)");
+MODULE_AUTHOR("Chia-chi Yeh <chiachi@android.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/net/tun.c b/drivers/net/tun.c
index 50bfded6d7ef..e33ec3bee2e5 100644
--- a/drivers/net/tun.c
+++ b/drivers/net/tun.c
@@ -862,10 +862,7 @@ static netdev_tx_t tun_net_xmit(struct sk_buff *skb, struct net_device *dev)
 	if (unlikely(skb_orphan_frags(skb, GFP_ATOMIC)))
 		goto drop;
 
-	if (skb->sk && sk_fullsock(skb->sk)) {
-		sock_tx_timestamp(skb->sk, &skb_shinfo(skb)->tx_flags);
-		sw_tx_timestamp(skb);
-	}
+	skb_tx_timestamp(skb);
 
 	/* Orphan the skb - required as we might hang on to it
 	 * for indefinite time.
@@ -1895,6 +1892,12 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd,
 	int le;
 	int ret;
 
+#ifdef CONFIG_ANDROID_PARANOID_NETWORK
+	if (cmd != TUNGETIFF && !capable(CAP_NET_ADMIN)) {
+		return -EPERM;
+	}
+#endif
+
 	if (cmd == TUNSETIFF || cmd == TUNSETQUEUE || _IOC_TYPE(cmd) == 0x89) {
 		if (copy_from_user(&ifr, argp, ifreq_len))
 			return -EFAULT;
diff --git a/drivers/net/wireless/brcm80211/brcmfmac/sdio.c b/drivers/net/wireless/brcm80211/brcmfmac/sdio.c
index 7e74ac3ad815..bcf29bf6f727 100644
--- a/drivers/net/wireless/brcm80211/brcmfmac/sdio.c
+++ b/drivers/net/wireless/brcm80211/brcmfmac/sdio.c
@@ -3401,10 +3401,6 @@ static int brcmf_sdio_download_firmware(struct brcmf_sdio *bus,
 		goto err;
 	}
 
-	/* Allow full data communication using DPC from now on. */
-	brcmf_sdiod_change_state(bus->sdiodev, BRCMF_SDIOD_DATA);
-	bcmerror = 0;
-
 err:
 	brcmf_sdio_clkctl(bus, CLK_SDONLY, false);
 	sdio_release_host(bus->sdiodev->func[1]);
@@ -4112,6 +4108,9 @@ static void brcmf_sdio_firmware_callback(struct device *dev,
 	}
 
 	if (err == 0) {
+		/* Allow full data communication using DPC from now on. */
+		brcmf_sdiod_change_state(bus->sdiodev, BRCMF_SDIOD_DATA);
+
 		err = brcmf_sdiod_intr_register(sdiodev);
 		if (err != 0)
 			brcmf_err("intr register failed:%d\n", err);
diff --git a/drivers/net/wireless/ti/wlcore/init.c b/drivers/net/wireless/ti/wlcore/init.c
index e92f2639af2c..9fd3c6af0a61 100644
--- a/drivers/net/wireless/ti/wlcore/init.c
+++ b/drivers/net/wireless/ti/wlcore/init.c
@@ -549,6 +549,11 @@ static int wl12xx_init_ap_role(struct wl1271 *wl, struct wl12xx_vif *wlvif)
 {
 	int ret;
 
+	/* Disable filtering */
+	ret = wl1271_acx_group_address_tbl(wl, wlvif, false, NULL, 0);
+	if (ret < 0)
+		return ret;
+
 	ret = wl1271_acx_ap_max_tx_retry(wl, wlvif);
 	if (ret < 0)
 		return ret;
diff --git a/drivers/nfc/fdp/i2c.c b/drivers/nfc/fdp/i2c.c
index a5d7332dfce5..2d043415f326 100644
--- a/drivers/nfc/fdp/i2c.c
+++ b/drivers/nfc/fdp/i2c.c
@@ -177,6 +177,16 @@ static int fdp_nci_i2c_read(struct fdp_i2c_phy *phy, struct sk_buff **skb)
 		/* Packet that contains a length */
 		if (tmp[0] == 0 && tmp[1] == 0) {
 			phy->next_read_size = (tmp[2] << 8) + tmp[3] + 3;
+			/*
+			 * Ensure next_read_size does not exceed sizeof(tmp)
+			 * for reading that many bytes during next iteration
+			 */
+			if (phy->next_read_size > FDP_NCI_I2C_MAX_PAYLOAD) {
+				dev_dbg(&client->dev, "%s: corrupted packet\n",
+					__func__);
+				phy->next_read_size = 5;
+				goto flush;
+			}
 		} else {
 			phy->next_read_size = FDP_NCI_I2C_MIN_PAYLOAD;
 
diff --git a/drivers/nfc/st21nfca/dep.c b/drivers/nfc/st21nfca/dep.c
index 798a32bbac5d..206285210ab5 100644
--- a/drivers/nfc/st21nfca/dep.c
+++ b/drivers/nfc/st21nfca/dep.c
@@ -217,7 +217,8 @@ static int st21nfca_tm_recv_atr_req(struct nfc_hci_dev *hdev,
 
 	atr_req = (struct st21nfca_atr_req *)skb->data;
 
-	if (atr_req->length < sizeof(struct st21nfca_atr_req)) {
+	if (atr_req->length < sizeof(struct st21nfca_atr_req) ||
+	    atr_req->length > skb->len) {
 		r = -EPROTO;
 		goto exit;
 	}
diff --git a/drivers/nfc/st21nfca/se.c b/drivers/nfc/st21nfca/se.c
index c79d99b24c96..2d4b6e910b87 100644
--- a/drivers/nfc/st21nfca/se.c
+++ b/drivers/nfc/st21nfca/se.c
@@ -321,23 +321,33 @@ int st21nfca_connectivity_event_received(struct nfc_hci_dev *hdev, u8 host,
 		 * AID		81	5 to 16
 		 * PARAMETERS	82	0 to 255
 		 */
-		if (skb->len < NFC_MIN_AID_LENGTH + 2 &&
+		if (skb->len < NFC_MIN_AID_LENGTH + 2 ||
 		    skb->data[0] != NFC_EVT_TRANSACTION_AID_TAG)
 			return -EPROTO;
 
+		/*
+		 * Buffer should have enough space for at least
+		 * two tag fields + two length fields + aid_len (skb->data[1])
+		 */
+		if (skb->len < skb->data[1] + 4)
+			return -EPROTO;
+
 		transaction = (struct nfc_evt_transaction *)devm_kzalloc(dev,
 						   skb->len - 2, GFP_KERNEL);
 
 		transaction->aid_len = skb->data[1];
 		memcpy(transaction->aid, &skb->data[2],
 		       transaction->aid_len);
+		transaction->params_len = skb->data[transaction->aid_len + 3];
 
-		/* Check next byte is PARAMETERS tag (82) */
+		/* Check next byte is PARAMETERS tag (82) and the length field */
 		if (skb->data[transaction->aid_len + 2] !=
-		    NFC_EVT_TRANSACTION_PARAMS_TAG)
+		    NFC_EVT_TRANSACTION_PARAMS_TAG ||
+		    skb->len < transaction->aid_len + transaction->params_len + 4) {
+			devm_kfree(dev, transaction);
 			return -EPROTO;
+		}
 
-		transaction->params_len = skb->data[transaction->aid_len + 3];
 		memcpy(transaction->params, skb->data +
 		       transaction->aid_len + 4, transaction->params_len);
 
diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c
index 3319df172f6a..901dd1ca2781 100644
--- a/drivers/of/fdt.c
+++ b/drivers/of/fdt.c
@@ -949,36 +949,66 @@ int __init early_init_dt_scan_memory(unsigned long node, const char *uname,
 	return 0;
 }
 
+/*
+ * Convert configs to something easy to use in C code
+ */
+#if defined(CONFIG_CMDLINE_FORCE)
+static const int overwrite_incoming_cmdline = 1;
+static const int read_dt_cmdline;
+static const int concat_cmdline;
+#elif defined(CONFIG_CMDLINE_EXTEND)
+static const int overwrite_incoming_cmdline;
+static const int read_dt_cmdline = 1;
+static const int concat_cmdline = 1;
+#else /* CMDLINE_FROM_BOOTLOADER */
+static const int overwrite_incoming_cmdline;
+static const int read_dt_cmdline = 1;
+static const int concat_cmdline;
+#endif
+
+#ifdef CONFIG_CMDLINE
+static const char *config_cmdline = CONFIG_CMDLINE;
+#else
+static const char *config_cmdline = "";
+#endif
+
 int __init early_init_dt_scan_chosen(unsigned long node, const char *uname,
 				     int depth, void *data)
 {
-	int l;
-	const char *p;
+	int l = 0;
+	const char *p = NULL;
+	char *cmdline = data;
 
 	pr_debug("search \"chosen\", depth: %d, uname: %s\n", depth, uname);
 
-	if (depth != 1 || !data ||
+	if (depth != 1 || !cmdline ||
 	    (strcmp(uname, "chosen") != 0 && strcmp(uname, "chosen@0") != 0))
 		return 0;
 
 	early_init_dt_check_for_initrd(node);
 
-	/* Retrieve command line */
-	p = of_get_flat_dt_prop(node, "bootargs", &l);
-	if (p != NULL && l > 0)
-		strlcpy(data, p, min((int)l, COMMAND_LINE_SIZE));
-
-	/*
-	 * CONFIG_CMDLINE is meant to be a default in case nothing else
-	 * managed to set the command line, unless CONFIG_CMDLINE_FORCE
-	 * is set in which case we override whatever was found earlier.
-	 */
-#ifdef CONFIG_CMDLINE
-#ifndef CONFIG_CMDLINE_FORCE
-	if (!((char *)data)[0])
-#endif
-		strlcpy(data, CONFIG_CMDLINE, COMMAND_LINE_SIZE);
-#endif /* CONFIG_CMDLINE */
+	/* Put CONFIG_CMDLINE in if forced or if data had nothing in it to start */
+	if (overwrite_incoming_cmdline || !cmdline[0])
+		strlcpy(cmdline, config_cmdline, COMMAND_LINE_SIZE);
+
+	/* Retrieve command line unless forcing */
+	if (read_dt_cmdline)
+		p = of_get_flat_dt_prop(node, "bootargs", &l);
+
+	if (p != NULL && l > 0) {
+		if (concat_cmdline) {
+			int cmdline_len;
+			int copy_len;
+			strlcat(cmdline, " ", COMMAND_LINE_SIZE);
+			cmdline_len = strlen(cmdline);
+			copy_len = COMMAND_LINE_SIZE - cmdline_len - 1;
+			copy_len = min((int)l, copy_len);
+			strncpy(cmdline + cmdline_len, p, copy_len);
+			cmdline[cmdline_len + copy_len] = '\0';
+		} else {
+			strlcpy(cmdline, p, min((int)l, COMMAND_LINE_SIZE));
+		}
+	}
 
 	pr_debug("Command line is: %s\n", (char*)data);
 
diff --git a/drivers/perf/arm_pmu.c b/drivers/perf/arm_pmu.c
index 8af1f900ea65..64b07b2d47f8 100644
--- a/drivers/perf/arm_pmu.c
+++ b/drivers/perf/arm_pmu.c
@@ -13,6 +13,7 @@
 
 #include <linux/bitmap.h>
 #include <linux/cpumask.h>
+#include <linux/cpu_pm.h>
 #include <linux/export.h>
 #include <linux/kernel.h>
 #include <linux/of_device.h>
@@ -718,6 +719,104 @@ static int cpu_pmu_notify(struct notifier_block *b, unsigned long action,
 	return NOTIFY_OK;
 }
 
+#ifdef CONFIG_CPU_PM
+static void cpu_pm_pmu_setup(struct arm_pmu *armpmu, unsigned long cmd)
+{
+	struct pmu_hw_events *hw_events = this_cpu_ptr(armpmu->hw_events);
+	struct perf_event *event;
+	int idx;
+
+	for (idx = 0; idx < armpmu->num_events; idx++) {
+		/*
+		 * If the counter is not used skip it, there is no
+		 * need of stopping/restarting it.
+		 */
+		if (!test_bit(idx, hw_events->used_mask))
+			continue;
+
+		event = hw_events->events[idx];
+
+		switch (cmd) {
+		case CPU_PM_ENTER:
+			/*
+			 * Stop and update the counter
+			 */
+			armpmu_stop(event, PERF_EF_UPDATE);
+			break;
+		case CPU_PM_EXIT:
+		case CPU_PM_ENTER_FAILED:
+			 /*
+			  * Restore and enable the counter.
+			  * armpmu_start() indirectly calls
+			  *
+			  * perf_event_update_userpage()
+			  *
+			  * that requires RCU read locking to be functional,
+			  * wrap the call within RCU_NONIDLE to make the
+			  * RCU subsystem aware this cpu is not idle from
+			  * an RCU perspective for the armpmu_start() call
+			  * duration.
+			  */
+			RCU_NONIDLE(armpmu_start(event, PERF_EF_RELOAD));
+			break;
+		default:
+			break;
+		}
+	}
+}
+
+static int cpu_pm_pmu_notify(struct notifier_block *b, unsigned long cmd,
+			     void *v)
+{
+	struct arm_pmu *armpmu = container_of(b, struct arm_pmu, cpu_pm_nb);
+	struct pmu_hw_events *hw_events = this_cpu_ptr(armpmu->hw_events);
+	int enabled = bitmap_weight(hw_events->used_mask, armpmu->num_events);
+
+	if (!cpumask_test_cpu(smp_processor_id(), &armpmu->supported_cpus))
+		return NOTIFY_DONE;
+
+	/*
+	 * Always reset the PMU registers on power-up even if
+	 * there are no events running.
+	 */
+	if (cmd == CPU_PM_EXIT && armpmu->reset)
+		armpmu->reset(armpmu);
+
+	if (!enabled)
+		return NOTIFY_OK;
+
+	switch (cmd) {
+	case CPU_PM_ENTER:
+		armpmu->stop(armpmu);
+		cpu_pm_pmu_setup(armpmu, cmd);
+		break;
+	case CPU_PM_EXIT:
+		cpu_pm_pmu_setup(armpmu, cmd);
+	case CPU_PM_ENTER_FAILED:
+		armpmu->start(armpmu);
+		break;
+	default:
+		return NOTIFY_DONE;
+	}
+
+	return NOTIFY_OK;
+}
+
+static int cpu_pm_pmu_register(struct arm_pmu *cpu_pmu)
+{
+	cpu_pmu->cpu_pm_nb.notifier_call = cpu_pm_pmu_notify;
+	return cpu_pm_register_notifier(&cpu_pmu->cpu_pm_nb);
+}
+
+static void cpu_pm_pmu_unregister(struct arm_pmu *cpu_pmu)
+{
+	cpu_pm_unregister_notifier(&cpu_pmu->cpu_pm_nb);
+}
+#else
+static inline int cpu_pm_pmu_register(struct arm_pmu *cpu_pmu) { return 0; }
+static inline void cpu_pm_pmu_unregister(struct arm_pmu *cpu_pmu) { }
+#endif
+
 static int cpu_pmu_init(struct arm_pmu *cpu_pmu)
 {
 	int err;
@@ -733,6 +832,10 @@ static int cpu_pmu_init(struct arm_pmu *cpu_pmu)
 	if (err)
 		goto out_hw_events;
 
+	err = cpu_pm_pmu_register(cpu_pmu);
+	if (err)
+		goto out_unregister;
+
 	for_each_possible_cpu(cpu) {
 		struct pmu_hw_events *events = per_cpu_ptr(cpu_hw_events, cpu);
 		raw_spin_lock_init(&events->pmu_lock);
@@ -754,6 +857,8 @@ static int cpu_pmu_init(struct arm_pmu *cpu_pmu)
 
 	return 0;
 
+out_unregister:
+	unregister_cpu_notifier(&cpu_pmu->hotplug_nb);
 out_hw_events:
 	free_percpu(cpu_hw_events);
 	return err;
@@ -761,6 +866,7 @@ out_hw_events:
 
 static void cpu_pmu_destroy(struct arm_pmu *cpu_pmu)
 {
+	cpu_pm_pmu_unregister(cpu_pmu);
 	unregister_cpu_notifier(&cpu_pmu->hotplug_nb);
 	free_percpu(cpu_pmu->hw_events);
 }
diff --git a/drivers/platform/Kconfig b/drivers/platform/Kconfig
index 0adccbf5c83f..5bd93a25d2ce 100644
--- a/drivers/platform/Kconfig
+++ b/drivers/platform/Kconfig
@@ -4,8 +4,6 @@ endif
 if MIPS
 source "drivers/platform/mips/Kconfig"
 endif
-if GOLDFISH
 source "drivers/platform/goldfish/Kconfig"
-endif
 
 source "drivers/platform/chrome/Kconfig"
diff --git a/drivers/platform/goldfish/Kconfig b/drivers/platform/goldfish/Kconfig
index 635ef25cc722..50331e3e54f3 100644
--- a/drivers/platform/goldfish/Kconfig
+++ b/drivers/platform/goldfish/Kconfig
@@ -1,5 +1,23 @@
+menuconfig GOLDFISH
+	bool "Platform support for Goldfish virtual devices"
+	depends on X86_32 || X86_64 || ARM || ARM64 || MIPS
+	---help---
+	  Say Y here to get to see options for the Goldfish virtual platform.
+	  This option alone does not add any kernel code.
+
+	  Unless you are building for the Android Goldfish emulator say N here.
+
+if GOLDFISH
+
+config GOLDFISH_BUS
+	bool "Goldfish platform bus"
+	---help---
+	  This is a virtual bus to host Goldfish Android Virtual Devices.
+
 config GOLDFISH_PIPE
 	tristate "Goldfish virtual device for QEMU pipes"
 	---help---
 	  This is a virtual device to drive the QEMU pipe interface used by
 	  the Goldfish Android Virtual Device.
+
+endif # GOLDFISH
diff --git a/drivers/platform/goldfish/Makefile b/drivers/platform/goldfish/Makefile
index a0022395eee9..277a820ee4e1 100644
--- a/drivers/platform/goldfish/Makefile
+++ b/drivers/platform/goldfish/Makefile
@@ -1,5 +1,6 @@
 #
 # Makefile for Goldfish platform specific drivers
 #
-obj-$(CONFIG_GOLDFISH)	+=	pdev_bus.o
-obj-$(CONFIG_GOLDFISH_PIPE)	+= goldfish_pipe.o
+obj-$(CONFIG_GOLDFISH_BUS)	+= pdev_bus.o
+obj-$(CONFIG_GOLDFISH_PIPE)	+= goldfish_pipe_all.o
+goldfish_pipe_all-objs := goldfish_pipe.o goldfish_pipe_v2.o
diff --git a/drivers/platform/goldfish/goldfish_pipe.c b/drivers/platform/goldfish/goldfish_pipe.c
index e7a29e2750c6..fd1452e28352 100644
--- a/drivers/platform/goldfish/goldfish_pipe.c
+++ b/drivers/platform/goldfish/goldfish_pipe.c
@@ -2,6 +2,7 @@
  * Copyright (C) 2011 Google, Inc.
  * Copyright (C) 2012 Intel, Inc.
  * Copyright (C) 2013 Intel, Inc.
+ * Copyright (C) 2014 Linaro Limited
  *
  * This software is licensed under the terms of the GNU General Public
  * License version 2, as published by the Free Software Foundation, and
@@ -14,49 +15,11 @@
  *
  */
 
-/* This source file contains the implementation of a special device driver
- * that intends to provide a *very* fast communication channel between the
- * guest system and the QEMU emulator.
- *
- * Usage from the guest is simply the following (error handling simplified):
- *
- *    int  fd = open("/dev/qemu_pipe",O_RDWR);
- *    .... write() or read() through the pipe.
- *
- * This driver doesn't deal with the exact protocol used during the session.
- * It is intended to be as simple as something like:
- *
- *    // do this _just_ after opening the fd to connect to a specific
- *    // emulator service.
- *    const char*  msg = "<pipename>";
- *    if (write(fd, msg, strlen(msg)+1) < 0) {
- *       ... could not connect to <pipename> service
- *       close(fd);
- *    }
- *
- *    // after this, simply read() and write() to communicate with the
- *    // service. Exact protocol details left as an exercise to the reader.
- *
- * This driver is very fast because it doesn't copy any data through
- * intermediate buffers, since the emulator is capable of translating
- * guest user addresses into host ones.
- *
- * Note that we must however ensure that each user page involved in the
- * exchange is properly mapped during a transfer.
+/* This source file contains the implementation of the legacy version of
+ * a goldfish pipe device driver. See goldfish_pipe_v2.c for the current
+ * version.
  */
-
-#include <linux/module.h>
-#include <linux/interrupt.h>
-#include <linux/kernel.h>
-#include <linux/spinlock.h>
-#include <linux/miscdevice.h>
-#include <linux/platform_device.h>
-#include <linux/poll.h>
-#include <linux/sched.h>
-#include <linux/bitops.h>
-#include <linux/slab.h>
-#include <linux/io.h>
-#include <linux/goldfish.h>
+#include "goldfish_pipe.h"
 
 /*
  * IMPORTANT: The following constants must match the ones used and defined
@@ -75,6 +38,7 @@
 #define PIPE_REG_PARAMS_ADDR_LOW	0x18  /* read/write: batch data address */
 #define PIPE_REG_PARAMS_ADDR_HIGH	0x1c  /* read/write: batch data address */
 #define PIPE_REG_ACCESS_PARAMS		0x20  /* write: batch access */
+#define PIPE_REG_VERSION		0x24  /* read: device version */
 
 /* list of commands for PIPE_REG_COMMAND */
 #define CMD_OPEN			1  /* open new channel */
@@ -90,12 +54,6 @@
 #define CMD_WRITE_BUFFER	4  /* send a user buffer to the emulator */
 #define CMD_WAKE_ON_WRITE	5  /* tell the emulator to wake us when writing
 				     is possible */
-
-/* The following commands are related to read operations, they must be
- * listed in the same order than the corresponding write ones, since we
- * will use (CMD_READ_BUFFER - CMD_WRITE_BUFFER) as a special offset
- * in goldfish_pipe_read_write() below.
- */
 #define CMD_READ_BUFFER        6  /* receive a user buffer from the emulator */
 #define CMD_WAKE_ON_READ       7  /* tell the emulator to wake us when reading
 				   * is possible */
@@ -111,28 +69,15 @@
 #define PIPE_WAKE_READ         (1 << 1)  /* pipe can now be read from */
 #define PIPE_WAKE_WRITE        (1 << 2)  /* pipe can now be written to */
 
-struct access_params {
-	unsigned long channel;
-	u32 size;
-	unsigned long address;
-	u32 cmd;
-	u32 result;
-	/* reserved for future extension */
-	u32 flags;
-};
+#define MAX_PAGES_TO_GRAB 32
 
-/* The global driver data. Holds a reference to the i/o page used to
- * communicate with the emulator, and a wake queue for blocked tasks
- * waiting to be awoken.
- */
-struct goldfish_pipe_dev {
-	spinlock_t lock;
-	unsigned char __iomem *base;
-	struct access_params *aps;
-	int irq;
-};
+#define DEBUG 0
 
-static struct goldfish_pipe_dev   pipe_dev[1];
+#if DEBUG
+#define DPRINT(...) { printk(KERN_ERR __VA_ARGS__); }
+#else
+#define DPRINT(...)
+#endif
 
 /* This data type models a given pipe instance */
 struct goldfish_pipe {
@@ -142,6 +87,15 @@ struct goldfish_pipe {
 	wait_queue_head_t wake_queue;
 };
 
+struct access_params {
+	unsigned long channel;
+	u32 size;
+	unsigned long address;
+	u32 cmd;
+	u32 result;
+	/* reserved for future extension */
+	u32 flags;
+};
 
 /* Bit flags for the 'flags' field */
 enum {
@@ -232,8 +186,10 @@ static int setup_access_params_addr(struct platform_device *pdev,
 	if (valid_batchbuffer_addr(dev, aps)) {
 		dev->aps = aps;
 		return 0;
-	} else
+	} else {
+		devm_kfree(&pdev->dev, aps);
 		return -1;
+	}
 }
 
 /* A value that will not be set by qemu emulator */
@@ -263,19 +219,15 @@ static int access_with_param(struct goldfish_pipe_dev *dev, const int cmd,
 	return 0;
 }
 
-/* This function is used for both reading from and writing to a given
- * pipe.
- */
 static ssize_t goldfish_pipe_read_write(struct file *filp, char __user *buffer,
 				    size_t bufflen, int is_write)
 {
 	unsigned long irq_flags;
 	struct goldfish_pipe *pipe = filp->private_data;
 	struct goldfish_pipe_dev *dev = pipe->dev;
-	const int cmd_offset = is_write ? 0
-					: (CMD_READ_BUFFER - CMD_WRITE_BUFFER);
 	unsigned long address, address_end;
-	int ret = 0;
+	struct page* pages[MAX_PAGES_TO_GRAB] = {};
+	int count = 0, ret = -EINVAL;
 
 	/* If the emulator already closed the pipe, no need to go further */
 	if (test_bit(BIT_CLOSED_ON_HOST, &pipe->flags))
@@ -298,79 +250,127 @@ static ssize_t goldfish_pipe_read_write(struct file *filp, char __user *buffer,
 	address_end = address + bufflen;
 
 	while (address < address_end) {
-		unsigned long  page_end = (address & PAGE_MASK) + PAGE_SIZE;
-		unsigned long  next     = page_end < address_end ? page_end
-								 : address_end;
-		unsigned long  avail    = next - address;
-		int status, wakeBit;
-
-		/* Ensure that the corresponding page is properly mapped */
-		/* FIXME: this isn't safe or sufficient - use get_user_pages */
-		if (is_write) {
-			char c;
-			/* Ensure that the page is mapped and readable */
-			if (__get_user(c, (char __user *)address)) {
-				if (!ret)
-					ret = -EFAULT;
-				break;
-			}
-		} else {
-			/* Ensure that the page is mapped and writable */
-			if (__put_user(0, (char __user *)address)) {
-				if (!ret)
-					ret = -EFAULT;
+		unsigned long page_end = (address & PAGE_MASK) + PAGE_SIZE;
+		unsigned long next, avail;
+		int status, wakeBit, page_i, num_contiguous_pages;
+		long first_page, last_page, requested_pages;
+		unsigned long xaddr, xaddr_prev, xaddr_i;
+
+		/*
+		 * Attempt to grab multiple physically contiguous pages.
+		 */
+		first_page = address & PAGE_MASK;
+		last_page = (address_end - 1) & PAGE_MASK;
+		requested_pages = ((last_page - first_page) >> PAGE_SHIFT) + 1;
+		if (requested_pages > MAX_PAGES_TO_GRAB) {
+			requested_pages = MAX_PAGES_TO_GRAB;
+		}
+		ret = get_user_pages_fast(first_page, requested_pages,
+				!is_write, pages);
+
+		DPRINT("%s: requested pages: %d %d %p\n", __FUNCTION__,
+			ret, requested_pages, first_page);
+		if (ret == 0) {
+			DPRINT("%s: error: (requested pages == 0) (wanted %d)\n",
+					__FUNCTION__, requested_pages);
+			mutex_unlock(&pipe->lock);
+			return ret;
+		}
+		if (ret < 0) {
+			DPRINT("%s: (requested pages < 0) %d \n",
+				 	__FUNCTION__, requested_pages);
+			mutex_unlock(&pipe->lock);
+			return ret;
+		}
+
+		xaddr = page_to_phys(pages[0]) | (address & ~PAGE_MASK);
+		xaddr_prev = xaddr;
+		num_contiguous_pages = ret == 0 ? 0 : 1;
+		for (page_i = 1; page_i < ret; page_i++) {
+			xaddr_i = page_to_phys(pages[page_i]) | (address & ~PAGE_MASK);
+			if (xaddr_i == xaddr_prev + PAGE_SIZE) {
+				page_end += PAGE_SIZE;
+				xaddr_prev = xaddr_i;
+				num_contiguous_pages++;
+			} else {
+				DPRINT("%s: discontinuous page boundary: %d pages instead\n",
+						__FUNCTION__, page_i);
 				break;
 			}
 		}
+		next = page_end < address_end ? page_end : address_end;
+		avail = next - address;
 
 		/* Now, try to transfer the bytes in the current page */
 		spin_lock_irqsave(&dev->lock, irq_flags);
-		if (access_with_param(dev, CMD_WRITE_BUFFER + cmd_offset,
-				address, avail, pipe, &status)) {
+		if (access_with_param(dev,
+					is_write ? CMD_WRITE_BUFFER : CMD_READ_BUFFER,
+					xaddr, avail, pipe, &status)) {
 			gf_write_ptr(pipe, dev->base + PIPE_REG_CHANNEL,
 				     dev->base + PIPE_REG_CHANNEL_HIGH);
 			writel(avail, dev->base + PIPE_REG_SIZE);
-			gf_write_ptr((void *)address,
+			gf_write_ptr((void *)xaddr,
 				     dev->base + PIPE_REG_ADDRESS,
 				     dev->base + PIPE_REG_ADDRESS_HIGH);
-			writel(CMD_WRITE_BUFFER + cmd_offset,
-					dev->base + PIPE_REG_COMMAND);
+			writel(is_write ? CMD_WRITE_BUFFER : CMD_READ_BUFFER,
+			       dev->base + PIPE_REG_COMMAND);
 			status = readl(dev->base + PIPE_REG_STATUS);
 		}
 		spin_unlock_irqrestore(&dev->lock, irq_flags);
 
+		for (page_i = 0; page_i < ret; page_i++) {
+			if (status > 0 && !is_write &&
+				page_i < num_contiguous_pages) {
+				set_page_dirty(pages[page_i]);
+			}
+			put_page(pages[page_i]);
+		}
+
 		if (status > 0) { /* Correct transfer */
-			ret += status;
+			count += status;
 			address += status;
 			continue;
-		}
-
-		if (status == 0)  /* EOF */
+		} else if (status == 0) { /* EOF */
+			ret = 0;
 			break;
-
-		/* An error occured. If we already transfered stuff, just
-		* return with its count. We expect the next call to return
-		* an error code */
-		if (ret > 0)
+		} else if (status < 0 && count > 0) {
+			/*
+			 * An error occured and we already transfered
+			 * something on one of the previous pages.
+			 * Just return what we already copied and log this
+			 * err.
+			 *
+			 * Note: This seems like an incorrect approach but
+			 * cannot change it until we check if any user space
+			 * ABI relies on this behavior.
+			 */
+			if (status != PIPE_ERROR_AGAIN)
+				pr_info_ratelimited("goldfish_pipe: backend returned error %d on %s\n",
+						status, is_write ? "write" : "read");
+			ret = 0;
 			break;
+		}
 
-		/* If the error is not PIPE_ERROR_AGAIN, or if we are not in
-		* non-blocking mode, just return the error code.
-		*/
+		/*
+		 * If the error is not PIPE_ERROR_AGAIN, or if we are not in
+		 * non-blocking mode, just return the error code.
+		 */
 		if (status != PIPE_ERROR_AGAIN ||
-			(filp->f_flags & O_NONBLOCK) != 0) {
+				(filp->f_flags & O_NONBLOCK) != 0) {
 			ret = goldfish_pipe_error_convert(status);
 			break;
 		}
 
-		/* We will have to wait until more data/space is available.
-		* First, mark the pipe as waiting for a specific wake signal.
-		*/
+		/*
+		 * The backend blocked the read/write, wait until the backend
+		 * tells us it's ready to process more data.
+		 */
 		wakeBit = is_write ? BIT_WAKE_ON_WRITE : BIT_WAKE_ON_READ;
 		set_bit(wakeBit, &pipe->flags);
 
 		/* Tell the emulator we're going to wait for a wake event */
-		goldfish_cmd(pipe, CMD_WAKE_ON_WRITE + cmd_offset);
+		goldfish_cmd(pipe,
+				is_write ? CMD_WAKE_ON_WRITE : CMD_WAKE_ON_READ);
 
 		/* Unlock the pipe, then wait for the wake signal */
 		mutex_unlock(&pipe->lock);
@@ -388,12 +388,13 @@ static ssize_t goldfish_pipe_read_write(struct file *filp, char __user *buffer,
 		/* Try to re-acquire the lock */
 		if (mutex_lock_interruptible(&pipe->lock))
 			return -ERESTARTSYS;
-
-		/* Try the transfer again */
-		continue;
 	}
 	mutex_unlock(&pipe->lock);
-	return ret;
+
+	if (ret < 0)
+		return ret;
+	else
+		return count;
 }
 
 static ssize_t goldfish_pipe_read(struct file *filp, char __user *buffer,
@@ -446,10 +447,11 @@ static irqreturn_t goldfish_pipe_interrupt(int irq, void *dev_id)
 	unsigned long irq_flags;
 	int count = 0;
 
-	/* We're going to read from the emulator a list of (channel,flags)
-	* pairs corresponding to the wake events that occured on each
-	* blocked pipe (i.e. channel).
-	*/
+	/*
+	 * We're going to read from the emulator a list of (channel,flags)
+	 * pairs corresponding to the wake events that occured on each
+	 * blocked pipe (i.e. channel).
+	 */
 	spin_lock_irqsave(&dev->lock, irq_flags);
 	for (;;) {
 		/* First read the channel, 0 means the end of the list */
@@ -514,6 +516,8 @@ static int goldfish_pipe_open(struct inode *inode, struct file *file)
 
 	pipe->dev = dev;
 	mutex_init(&pipe->lock);
+	DPRINT("%s: call. pipe_dev pipe_dev=0x%lx new_pipe_addr=0x%lx file=0x%lx\n", __FUNCTION__, pipe_dev, pipe, file);
+	// spin lock init, write head of list, i guess
 	init_waitqueue_head(&pipe->wake_queue);
 
 	/*
@@ -536,6 +540,7 @@ static int goldfish_pipe_release(struct inode *inode, struct file *filp)
 {
 	struct goldfish_pipe *pipe = filp->private_data;
 
+	DPRINT("%s: call. pipe=0x%lx file=0x%lx\n", __FUNCTION__, pipe, filp);
 	/* The guest is closing the channel, so tell the emulator right now */
 	goldfish_cmd(pipe, CMD_CLOSE);
 	kfree(pipe);
@@ -552,77 +557,33 @@ static const struct file_operations goldfish_pipe_fops = {
 	.release = goldfish_pipe_release,
 };
 
-static struct miscdevice goldfish_pipe_device = {
+static struct miscdevice goldfish_pipe_dev = {
 	.minor = MISC_DYNAMIC_MINOR,
 	.name = "goldfish_pipe",
 	.fops = &goldfish_pipe_fops,
 };
 
-static int goldfish_pipe_probe(struct platform_device *pdev)
+int goldfish_pipe_device_init_v1(struct platform_device *pdev)
 {
-	int err;
-	struct resource *r;
 	struct goldfish_pipe_dev *dev = pipe_dev;
-
-	/* not thread safe, but this should not happen */
-	WARN_ON(dev->base != NULL);
-
-	spin_lock_init(&dev->lock);
-
-	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (r == NULL || resource_size(r) < PAGE_SIZE) {
-		dev_err(&pdev->dev, "can't allocate i/o page\n");
-		return -EINVAL;
-	}
-	dev->base = devm_ioremap(&pdev->dev, r->start, PAGE_SIZE);
-	if (dev->base == NULL) {
-		dev_err(&pdev->dev, "ioremap failed\n");
-		return -EINVAL;
-	}
-
-	r = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
-	if (r == NULL) {
-		err = -EINVAL;
-		goto error;
-	}
-	dev->irq = r->start;
-
-	err = devm_request_irq(&pdev->dev, dev->irq, goldfish_pipe_interrupt,
+	int err = devm_request_irq(&pdev->dev, dev->irq, goldfish_pipe_interrupt,
 				IRQF_SHARED, "goldfish_pipe", dev);
 	if (err) {
-		dev_err(&pdev->dev, "unable to allocate IRQ\n");
-		goto error;
+		dev_err(&pdev->dev, "unable to allocate IRQ for v1\n");
+		return err;
 	}
 
-	err = misc_register(&goldfish_pipe_device);
+	err = misc_register(&goldfish_pipe_dev);
 	if (err) {
-		dev_err(&pdev->dev, "unable to register device\n");
-		goto error;
+		dev_err(&pdev->dev, "unable to register v1 device\n");
+		return err;
 	}
+
 	setup_access_params_addr(pdev, dev);
 	return 0;
-
-error:
-	dev->base = NULL;
-	return err;
 }
 
-static int goldfish_pipe_remove(struct platform_device *pdev)
+void goldfish_pipe_device_deinit_v1(struct platform_device *pdev)
 {
-	struct goldfish_pipe_dev *dev = pipe_dev;
-	misc_deregister(&goldfish_pipe_device);
-	dev->base = NULL;
-	return 0;
+    misc_deregister(&goldfish_pipe_dev);
 }
-
-static struct platform_driver goldfish_pipe = {
-	.probe = goldfish_pipe_probe,
-	.remove = goldfish_pipe_remove,
-	.driver = {
-		.name = "goldfish_pipe"
-	}
-};
-
-module_platform_driver(goldfish_pipe);
-MODULE_AUTHOR("David Turner <digit@google.com>");
-MODULE_LICENSE("GPL");
diff --git a/drivers/platform/goldfish/goldfish_pipe.h b/drivers/platform/goldfish/goldfish_pipe.h
new file mode 100644
index 000000000000..9b75a51dba24
--- /dev/null
+++ b/drivers/platform/goldfish/goldfish_pipe.h
@@ -0,0 +1,91 @@
+/*
+ * Copyright (C) 2016 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+#ifndef GOLDFISH_PIPE_H
+#define GOLDFISH_PIPE_H
+
+#include <linux/module.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/spinlock.h>
+#include <linux/miscdevice.h>
+#include <linux/platform_device.h>
+#include <linux/poll.h>
+#include <linux/sched.h>
+#include <linux/bitops.h>
+#include <linux/slab.h>
+#include <linux/io.h>
+#include <linux/goldfish.h>
+#include <linux/mm.h>
+#include <linux/acpi.h>
+
+
+/* Initialize the legacy version of the pipe device driver */
+int goldfish_pipe_device_init_v1(struct platform_device *pdev);
+
+/* Deinitialize the legacy version of the pipe device driver */
+void goldfish_pipe_device_deinit_v1(struct platform_device *pdev);
+
+/* Forward declarations for the device struct */
+struct goldfish_pipe;
+struct goldfish_pipe_device_buffers;
+
+/* The global driver data. Holds a reference to the i/o page used to
+ * communicate with the emulator, and a wake queue for blocked tasks
+ * waiting to be awoken.
+ */
+struct goldfish_pipe_dev {
+	/*
+	 * Global device spinlock. Protects the following members:
+	 *  - pipes, pipes_capacity
+	 *  - [*pipes, *pipes + pipes_capacity) - array data
+	 *  - first_signalled_pipe,
+	 *      goldfish_pipe::prev_signalled,
+	 *      goldfish_pipe::next_signalled,
+	 *      goldfish_pipe::signalled_flags - all singnalled-related fields,
+	 *                                       in all allocated pipes
+	 *  - open_command_params - PIPE_CMD_OPEN-related buffers
+	 *
+	 * It looks like a lot of different fields, but the trick is that the only
+	 * operation that happens often is the signalled pipes array manipulation.
+	 * That's why it's OK for now to keep the rest of the fields under the same
+	 * lock. If we notice too much contention because of PIPE_CMD_OPEN,
+	 * then we should add a separate lock there.
+	 */
+	spinlock_t lock;
+
+	/*
+	 * Array of the pipes of |pipes_capacity| elements,
+	 * indexed by goldfish_pipe::id
+	 */
+	struct goldfish_pipe **pipes;
+	u32 pipes_capacity;
+
+	/* Pointers to the buffers host uses for interaction with this driver */
+	struct goldfish_pipe_dev_buffers *buffers;
+
+	/* Head of a doubly linked list of signalled pipes */
+	struct goldfish_pipe *first_signalled_pipe;
+
+	/* Some device-specific data */
+	int irq;
+	int version;
+	unsigned char __iomem *base;
+
+	/* v1-specific access parameters */
+	struct access_params *aps;
+};
+
+extern struct goldfish_pipe_dev pipe_dev[1];
+
+#endif /* GOLDFISH_PIPE_H */
diff --git a/drivers/platform/goldfish/goldfish_pipe_v2.c b/drivers/platform/goldfish/goldfish_pipe_v2.c
new file mode 100644
index 000000000000..ad373ed36555
--- /dev/null
+++ b/drivers/platform/goldfish/goldfish_pipe_v2.c
@@ -0,0 +1,889 @@
+/*
+ * Copyright (C) 2012 Intel, Inc.
+ * Copyright (C) 2013 Intel, Inc.
+ * Copyright (C) 2014 Linaro Limited
+ * Copyright (C) 2011-2016 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+/* This source file contains the implementation of a special device driver
+ * that intends to provide a *very* fast communication channel between the
+ * guest system and the QEMU emulator.
+ *
+ * Usage from the guest is simply the following (error handling simplified):
+ *
+ *    int  fd = open("/dev/qemu_pipe",O_RDWR);
+ *    .... write() or read() through the pipe.
+ *
+ * This driver doesn't deal with the exact protocol used during the session.
+ * It is intended to be as simple as something like:
+ *
+ *    // do this _just_ after opening the fd to connect to a specific
+ *    // emulator service.
+ *    const char*  msg = "<pipename>";
+ *    if (write(fd, msg, strlen(msg)+1) < 0) {
+ *       ... could not connect to <pipename> service
+ *       close(fd);
+ *    }
+ *
+ *    // after this, simply read() and write() to communicate with the
+ *    // service. Exact protocol details left as an exercise to the reader.
+ *
+ * This driver is very fast because it doesn't copy any data through
+ * intermediate buffers, since the emulator is capable of translating
+ * guest user addresses into host ones.
+ *
+ * Note that we must however ensure that each user page involved in the
+ * exchange is properly mapped during a transfer.
+ */
+
+#include "goldfish_pipe.h"
+
+
+/*
+ * Update this when something changes in the driver's behavior so the host
+ * can benefit from knowing it
+ */
+enum {
+	PIPE_DRIVER_VERSION = 2,
+	PIPE_CURRENT_DEVICE_VERSION = 2
+};
+
+/*
+ * IMPORTANT: The following constants must match the ones used and defined
+ * in external/qemu/hw/goldfish_pipe.c in the Android source tree.
+ */
+
+/* List of bitflags returned in status of CMD_POLL command */
+enum PipePollFlags {
+	PIPE_POLL_IN	= 1 << 0,
+	PIPE_POLL_OUT	= 1 << 1,
+	PIPE_POLL_HUP	= 1 << 2
+};
+
+/* Possible status values used to signal errors - see goldfish_pipe_error_convert */
+enum PipeErrors {
+	PIPE_ERROR_INVAL  = -1,
+	PIPE_ERROR_AGAIN  = -2,
+	PIPE_ERROR_NOMEM  = -3,
+	PIPE_ERROR_IO     = -4
+};
+
+/* Bit-flags used to signal events from the emulator */
+enum PipeWakeFlags {
+	PIPE_WAKE_CLOSED = 1 << 0,  /* emulator closed pipe */
+	PIPE_WAKE_READ   = 1 << 1,  /* pipe can now be read from */
+	PIPE_WAKE_WRITE  = 1 << 2  /* pipe can now be written to */
+};
+
+/* Bit flags for the 'flags' field */
+enum PipeFlagsBits {
+	BIT_CLOSED_ON_HOST = 0,  /* pipe closed by host */
+	BIT_WAKE_ON_WRITE  = 1,  /* want to be woken on writes */
+	BIT_WAKE_ON_READ   = 2,  /* want to be woken on reads */
+};
+
+enum PipeRegs {
+	PIPE_REG_CMD = 0,
+
+	PIPE_REG_SIGNAL_BUFFER_HIGH = 4,
+	PIPE_REG_SIGNAL_BUFFER = 8,
+	PIPE_REG_SIGNAL_BUFFER_COUNT = 12,
+
+	PIPE_REG_OPEN_BUFFER_HIGH = 20,
+	PIPE_REG_OPEN_BUFFER = 24,
+
+	PIPE_REG_VERSION = 36,
+
+	PIPE_REG_GET_SIGNALLED = 48,
+};
+
+enum PipeCmdCode {
+	PIPE_CMD_OPEN = 1,	/* to be used by the pipe device itself */
+	PIPE_CMD_CLOSE,
+	PIPE_CMD_POLL,
+	PIPE_CMD_WRITE,
+	PIPE_CMD_WAKE_ON_WRITE,
+	PIPE_CMD_READ,
+	PIPE_CMD_WAKE_ON_READ,
+
+	/*
+	 * TODO(zyy): implement a deferred read/write execution to allow parallel
+	 *  processing of pipe operations on the host.
+	*/
+	PIPE_CMD_WAKE_ON_DONE_IO,
+};
+
+enum {
+	MAX_BUFFERS_PER_COMMAND = 336,
+	MAX_SIGNALLED_PIPES = 64,
+	INITIAL_PIPES_CAPACITY = 64
+};
+
+struct goldfish_pipe_dev;
+struct goldfish_pipe;
+struct goldfish_pipe_command;
+
+/* A per-pipe command structure, shared with the host */
+struct goldfish_pipe_command {
+	s32 cmd;		/* PipeCmdCode, guest -> host */
+	s32 id;			/* pipe id, guest -> host */
+	s32 status;		/* command execution status, host -> guest */
+	s32 reserved;	/* to pad to 64-bit boundary */
+	union {
+		/* Parameters for PIPE_CMD_{READ,WRITE} */
+		struct {
+			u32 buffers_count;					/* number of buffers, guest -> host */
+			s32 consumed_size;					/* number of consumed bytes, host -> guest */
+			u64 ptrs[MAX_BUFFERS_PER_COMMAND]; 	/* buffer pointers, guest -> host */
+			u32 sizes[MAX_BUFFERS_PER_COMMAND];	/* buffer sizes, guest -> host */
+		} rw_params;
+	};
+};
+
+/* A single signalled pipe information */
+struct signalled_pipe_buffer {
+	u32 id;
+	u32 flags;
+};
+
+/* Parameters for the PIPE_CMD_OPEN command */
+struct open_command_param {
+	u64 command_buffer_ptr;
+	u32 rw_params_max_count;
+};
+
+/* Device-level set of buffers shared with the host */
+struct goldfish_pipe_dev_buffers {
+	struct open_command_param open_command_params;
+	struct signalled_pipe_buffer signalled_pipe_buffers[MAX_SIGNALLED_PIPES];
+};
+
+/* This data type models a given pipe instance */
+struct goldfish_pipe {
+	u32 id;							/* pipe ID - index into goldfish_pipe_dev::pipes array */
+	unsigned long flags;			/* The wake flags pipe is waiting for
+									 * Note: not protected with any lock, uses atomic operations
+									 *  and barriers to make it thread-safe.
+									 */
+	unsigned long signalled_flags;	/* wake flags host have signalled,
+									 *  - protected by goldfish_pipe_dev::lock */
+
+	struct goldfish_pipe_command *command_buffer;	/* A pointer to command buffer */
+
+	/* doubly linked list of signalled pipes, protected by goldfish_pipe_dev::lock */
+	struct goldfish_pipe *prev_signalled;
+	struct goldfish_pipe *next_signalled;
+
+	/*
+	 * A pipe's own lock. Protects the following:
+	 *  - *command_buffer - makes sure a command can safely write its parameters
+	 *    to the host and read the results back.
+	 */
+	struct mutex lock;
+
+	wait_queue_head_t wake_queue;	/* A wake queue for sleeping until host signals an event */
+	struct goldfish_pipe_dev *dev;	/* Pointer to the parent goldfish_pipe_dev instance */
+};
+
+struct goldfish_pipe_dev pipe_dev[1] = {};
+
+static int goldfish_cmd_locked(struct goldfish_pipe *pipe, enum PipeCmdCode cmd)
+{
+	pipe->command_buffer->cmd = cmd;
+	pipe->command_buffer->status = PIPE_ERROR_INVAL;	/* failure by default */
+	writel(pipe->id, pipe->dev->base + PIPE_REG_CMD);
+	return pipe->command_buffer->status;
+}
+
+static int goldfish_cmd(struct goldfish_pipe *pipe, enum PipeCmdCode cmd)
+{
+	int status;
+	if (mutex_lock_interruptible(&pipe->lock))
+		return PIPE_ERROR_IO;
+	status = goldfish_cmd_locked(pipe, cmd);
+	mutex_unlock(&pipe->lock);
+	return status;
+}
+
+/*
+ * This function converts an error code returned by the emulator through
+ * the PIPE_REG_STATUS i/o register into a valid negative errno value.
+ */
+static int goldfish_pipe_error_convert(int status)
+{
+	switch (status) {
+	case PIPE_ERROR_AGAIN:
+		return -EAGAIN;
+	case PIPE_ERROR_NOMEM:
+		return -ENOMEM;
+	case PIPE_ERROR_IO:
+		return -EIO;
+	default:
+		return -EINVAL;
+	}
+}
+
+static int pin_user_pages(unsigned long first_page, unsigned long last_page,
+	unsigned last_page_size, int is_write,
+	struct page *pages[MAX_BUFFERS_PER_COMMAND], unsigned *iter_last_page_size)
+{
+	int ret;
+	int requested_pages = ((last_page - first_page) >> PAGE_SHIFT) + 1;
+	if (requested_pages > MAX_BUFFERS_PER_COMMAND) {
+		requested_pages = MAX_BUFFERS_PER_COMMAND;
+		*iter_last_page_size = PAGE_SIZE;
+	} else {
+		*iter_last_page_size = last_page_size;
+	}
+
+	ret = get_user_pages_fast(
+			first_page, requested_pages, !is_write, pages);
+	if (ret <= 0)
+		return -EFAULT;
+	if (ret < requested_pages)
+		*iter_last_page_size = PAGE_SIZE;
+	return ret;
+
+}
+
+static void release_user_pages(struct page **pages, int pages_count,
+	int is_write, s32 consumed_size)
+{
+	int i;
+	for (i = 0; i < pages_count; i++) {
+		if (!is_write && consumed_size > 0) {
+			set_page_dirty(pages[i]);
+		}
+		put_page(pages[i]);
+	}
+}
+
+/* Populate the call parameters, merging adjacent pages together */
+static void populate_rw_params(
+	struct page **pages, int pages_count,
+	unsigned long address, unsigned long address_end,
+	unsigned long first_page, unsigned long last_page,
+	unsigned iter_last_page_size, int is_write,
+	struct goldfish_pipe_command *command)
+{
+	/*
+	 * Process the first page separately - it's the only page that
+	 * needs special handling for its start address.
+	 */
+	unsigned long xaddr = page_to_phys(pages[0]);
+	unsigned long xaddr_prev = xaddr;
+	int buffer_idx = 0;
+	int i = 1;
+	int size_on_page = first_page == last_page
+			? (int)(address_end - address)
+			: (PAGE_SIZE - (address & ~PAGE_MASK));
+	command->rw_params.ptrs[0] = (u64)(xaddr | (address & ~PAGE_MASK));
+	command->rw_params.sizes[0] = size_on_page;
+	for (; i < pages_count; ++i) {
+		xaddr = page_to_phys(pages[i]);
+		size_on_page = (i == pages_count - 1) ? iter_last_page_size : PAGE_SIZE;
+		if (xaddr == xaddr_prev + PAGE_SIZE) {
+			command->rw_params.sizes[buffer_idx] += size_on_page;
+		} else {
+			++buffer_idx;
+			command->rw_params.ptrs[buffer_idx] = (u64)xaddr;
+			command->rw_params.sizes[buffer_idx] = size_on_page;
+		}
+		xaddr_prev = xaddr;
+	}
+	command->rw_params.buffers_count = buffer_idx + 1;
+}
+
+static int transfer_max_buffers(struct goldfish_pipe* pipe,
+	unsigned long address, unsigned long address_end, int is_write,
+	unsigned long last_page, unsigned int last_page_size,
+	s32* consumed_size, int* status)
+{
+	struct page *pages[MAX_BUFFERS_PER_COMMAND];
+	unsigned long first_page = address & PAGE_MASK;
+	unsigned int iter_last_page_size;
+	int pages_count = pin_user_pages(first_page, last_page,
+			last_page_size, is_write,
+			pages, &iter_last_page_size);
+	if (pages_count < 0)
+		return pages_count;
+
+	/* Serialize access to the pipe command buffers */
+	if (mutex_lock_interruptible(&pipe->lock))
+		return -ERESTARTSYS;
+
+	populate_rw_params(pages, pages_count, address, address_end,
+		first_page, last_page, iter_last_page_size, is_write,
+		pipe->command_buffer);
+
+	/* Transfer the data */
+	*status = goldfish_cmd_locked(pipe,
+						is_write ? PIPE_CMD_WRITE : PIPE_CMD_READ);
+
+	*consumed_size = pipe->command_buffer->rw_params.consumed_size;
+
+	mutex_unlock(&pipe->lock);
+
+	release_user_pages(pages, pages_count, is_write, *consumed_size);
+
+	return 0;
+}
+
+static int wait_for_host_signal(struct goldfish_pipe *pipe, int is_write)
+{
+	u32 wakeBit = is_write ? BIT_WAKE_ON_WRITE : BIT_WAKE_ON_READ;
+	set_bit(wakeBit, &pipe->flags);
+
+	/* Tell the emulator we're going to wait for a wake event */
+	(void)goldfish_cmd(pipe,
+			is_write ? PIPE_CMD_WAKE_ON_WRITE : PIPE_CMD_WAKE_ON_READ);
+
+	while (test_bit(wakeBit, &pipe->flags)) {
+		if (wait_event_interruptible(
+				pipe->wake_queue,
+				!test_bit(wakeBit, &pipe->flags)))
+			return -ERESTARTSYS;
+
+		if (test_bit(BIT_CLOSED_ON_HOST, &pipe->flags))
+			return -EIO;
+	}
+
+	return 0;
+}
+
+static ssize_t goldfish_pipe_read_write(struct file *filp,
+	char __user *buffer, size_t bufflen, int is_write)
+{
+	struct goldfish_pipe *pipe = filp->private_data;
+	int count = 0, ret = -EINVAL;
+	unsigned long address, address_end, last_page;
+	unsigned int last_page_size;
+
+	/* If the emulator already closed the pipe, no need to go further */
+	if (unlikely(test_bit(BIT_CLOSED_ON_HOST, &pipe->flags)))
+		return -EIO;
+	/* Null reads or writes succeeds */
+	if (unlikely(bufflen == 0))
+		return 0;
+	/* Check the buffer range for access */
+	if (unlikely(!access_ok(is_write ? VERIFY_WRITE : VERIFY_READ,
+			buffer, bufflen)))
+		return -EFAULT;
+
+	address = (unsigned long)buffer;
+	address_end = address + bufflen;
+	last_page = (address_end - 1) & PAGE_MASK;
+	last_page_size = ((address_end - 1) & ~PAGE_MASK) + 1;
+
+	while (address < address_end) {
+		s32 consumed_size;
+		int status;
+		ret = transfer_max_buffers(pipe, address, address_end, is_write,
+				last_page, last_page_size, &consumed_size, &status);
+		if (ret < 0)
+			break;
+
+		if (consumed_size > 0) {
+			/* No matter what's the status, we've transfered something */
+			count += consumed_size;
+			address += consumed_size;
+		}
+		if (status > 0)
+			continue;
+		if (status == 0) {
+			/* EOF */
+			ret = 0;
+			break;
+		}
+		if (count > 0) {
+			/*
+			 * An error occured, but we already transfered
+			 * something on one of the previous iterations.
+			 * Just return what we already copied and log this
+			 * err.
+			 */
+			if (status != PIPE_ERROR_AGAIN)
+				pr_info_ratelimited("goldfish_pipe: backend error %d on %s\n",
+									status, is_write ? "write" : "read");
+			break;
+		}
+
+		/*
+		 * If the error is not PIPE_ERROR_AGAIN, or if we are in
+		 * non-blocking mode, just return the error code.
+		 */
+		if (status != PIPE_ERROR_AGAIN || (filp->f_flags & O_NONBLOCK) != 0) {
+			ret = goldfish_pipe_error_convert(status);
+			break;
+		}
+
+		status = wait_for_host_signal(pipe, is_write);
+		if (status < 0)
+			return status;
+	}
+
+	if (count > 0)
+		return count;
+	return ret;
+}
+
+static ssize_t goldfish_pipe_read(struct file *filp, char __user *buffer,
+				size_t bufflen, loff_t *ppos)
+{
+	return goldfish_pipe_read_write(filp, buffer, bufflen, /* is_write */ 0);
+}
+
+static ssize_t goldfish_pipe_write(struct file *filp,
+				const char __user *buffer, size_t bufflen,
+				loff_t *ppos)
+{
+	return goldfish_pipe_read_write(filp,
+			/* cast away the const */(char __user *)buffer, bufflen,
+			/* is_write */ 1);
+}
+
+static unsigned int goldfish_pipe_poll(struct file *filp, poll_table *wait)
+{
+	struct goldfish_pipe *pipe = filp->private_data;
+	unsigned int mask = 0;
+	int status;
+
+	poll_wait(filp, &pipe->wake_queue, wait);
+
+	status = goldfish_cmd(pipe, PIPE_CMD_POLL);
+	if (status < 0) {
+		return -ERESTARTSYS;
+	}
+
+	if (status & PIPE_POLL_IN)
+		mask |= POLLIN | POLLRDNORM;
+	if (status & PIPE_POLL_OUT)
+		mask |= POLLOUT | POLLWRNORM;
+	if (status & PIPE_POLL_HUP)
+		mask |= POLLHUP;
+	if (test_bit(BIT_CLOSED_ON_HOST, &pipe->flags))
+		mask |= POLLERR;
+
+	return mask;
+}
+
+static void signalled_pipes_add_locked(struct goldfish_pipe_dev *dev,
+	u32 id, u32 flags)
+{
+	struct goldfish_pipe *pipe;
+
+	BUG_ON(id >= dev->pipes_capacity);
+
+	pipe = dev->pipes[id];
+	if (!pipe)
+		return;
+	pipe->signalled_flags |= flags;
+
+	if (pipe->prev_signalled || pipe->next_signalled
+		|| dev->first_signalled_pipe == pipe)
+		return;	/* already in the list */
+	pipe->next_signalled = dev->first_signalled_pipe;
+	if (dev->first_signalled_pipe) {
+		dev->first_signalled_pipe->prev_signalled = pipe;
+	}
+	dev->first_signalled_pipe = pipe;
+}
+
+static void signalled_pipes_remove_locked(struct goldfish_pipe_dev *dev,
+	struct goldfish_pipe *pipe) {
+	if (pipe->prev_signalled)
+		pipe->prev_signalled->next_signalled = pipe->next_signalled;
+	if (pipe->next_signalled)
+		pipe->next_signalled->prev_signalled = pipe->prev_signalled;
+	if (pipe == dev->first_signalled_pipe)
+		dev->first_signalled_pipe = pipe->next_signalled;
+	pipe->prev_signalled = NULL;
+	pipe->next_signalled = NULL;
+}
+
+static struct goldfish_pipe *signalled_pipes_pop_front(struct goldfish_pipe_dev *dev,
+		int *wakes)
+{
+	struct goldfish_pipe *pipe;
+	unsigned long flags;
+	spin_lock_irqsave(&dev->lock, flags);
+
+	pipe = dev->first_signalled_pipe;
+	if (pipe) {
+		*wakes = pipe->signalled_flags;
+		pipe->signalled_flags = 0;
+		/*
+		 * This is an optimized version of signalled_pipes_remove_locked() -
+		 * we want to make it as fast as possible to wake the sleeping pipe
+		 * operations faster
+		 */
+		dev->first_signalled_pipe = pipe->next_signalled;
+		if (dev->first_signalled_pipe)
+			dev->first_signalled_pipe->prev_signalled = NULL;
+		pipe->next_signalled = NULL;
+	}
+
+	spin_unlock_irqrestore(&dev->lock, flags);
+	return pipe;
+}
+
+static void goldfish_interrupt_task(unsigned long unused)
+{
+	struct goldfish_pipe_dev *dev = pipe_dev;
+	/* Iterate over the signalled pipes and wake them one by one */
+	struct goldfish_pipe *pipe;
+	int wakes;
+	while ((pipe = signalled_pipes_pop_front(dev, &wakes)) != NULL) {
+		if (wakes & PIPE_WAKE_CLOSED) {
+			pipe->flags = 1 << BIT_CLOSED_ON_HOST;
+		} else {
+			if (wakes & PIPE_WAKE_READ)
+				clear_bit(BIT_WAKE_ON_READ, &pipe->flags);
+			if (wakes & PIPE_WAKE_WRITE)
+				clear_bit(BIT_WAKE_ON_WRITE, &pipe->flags);
+		}
+		/*
+		 * wake_up_interruptible() implies a write barrier, so don't explicitly
+		 * add another one here.
+		 */
+		wake_up_interruptible(&pipe->wake_queue);
+	}
+}
+DECLARE_TASKLET(goldfish_interrupt_tasklet, goldfish_interrupt_task, 0);
+
+/*
+ * The general idea of the interrupt handling:
+ *
+ *  1. device raises an interrupt if there's at least one signalled pipe
+ *  2. IRQ handler reads the signalled pipes and their count from the device
+ *  3. device writes them into a shared buffer and returns the count
+ *      it only resets the IRQ if it has returned all signalled pipes,
+ *      otherwise it leaves it raised, so IRQ handler will be called
+ *      again for the next chunk
+ *  4. IRQ handler adds all returned pipes to the device's signalled pipes list
+ *  5. IRQ handler launches a tasklet to process the signalled pipes from the
+ *      list in a separate context
+ */
+static irqreturn_t goldfish_pipe_interrupt(int irq, void *dev_id)
+{
+	u32 count;
+	u32 i;
+	unsigned long flags;
+	struct goldfish_pipe_dev *dev = dev_id;
+	if (dev != pipe_dev)
+		return IRQ_NONE;
+
+	/* Request the signalled pipes from the device */
+	spin_lock_irqsave(&dev->lock, flags);
+
+	count = readl(dev->base + PIPE_REG_GET_SIGNALLED);
+	if (count == 0) {
+		spin_unlock_irqrestore(&dev->lock, flags);
+		return IRQ_NONE;
+	}
+	if (count > MAX_SIGNALLED_PIPES)
+		count = MAX_SIGNALLED_PIPES;
+
+	for (i = 0; i < count; ++i)
+		signalled_pipes_add_locked(dev,
+			dev->buffers->signalled_pipe_buffers[i].id,
+			dev->buffers->signalled_pipe_buffers[i].flags);
+
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	tasklet_schedule(&goldfish_interrupt_tasklet);
+	return IRQ_HANDLED;
+}
+
+static int get_free_pipe_id_locked(struct goldfish_pipe_dev *dev)
+{
+	int id;
+	for (id = 0; id < dev->pipes_capacity; ++id)
+		if (!dev->pipes[id])
+			return id;
+
+	{
+		/* Reallocate the array */
+		u32 new_capacity = 2 * dev->pipes_capacity;
+		struct goldfish_pipe **pipes =
+				kcalloc(new_capacity, sizeof(*pipes),
+					GFP_ATOMIC);
+		if (!pipes)
+			return -ENOMEM;
+		memcpy(pipes, dev->pipes, sizeof(*pipes) * dev->pipes_capacity);
+		kfree(dev->pipes);
+		dev->pipes = pipes;
+		id = dev->pipes_capacity;
+		dev->pipes_capacity = new_capacity;
+	}
+	return id;
+}
+
+/**
+ *	goldfish_pipe_open - open a channel to the AVD
+ *	@inode: inode of device
+ *	@file: file struct of opener
+ *
+ *	Create a new pipe link between the emulator and the use application.
+ *	Each new request produces a new pipe.
+ *
+ *	Note: we use the pipe ID as a mux. All goldfish emulations are 32bit
+ *	right now so this is fine. A move to 64bit will need this addressing
+ */
+static int goldfish_pipe_open(struct inode *inode, struct file *file)
+{
+	struct goldfish_pipe_dev *dev = pipe_dev;
+	unsigned long flags;
+	int id;
+	int status;
+
+	/* Allocate new pipe kernel object */
+	struct goldfish_pipe *pipe = kzalloc(sizeof(*pipe), GFP_KERNEL);
+	if (pipe == NULL)
+		return -ENOMEM;
+
+	pipe->dev = dev;
+	mutex_init(&pipe->lock);
+	init_waitqueue_head(&pipe->wake_queue);
+
+	/*
+	 * Command buffer needs to be allocated on its own page to make sure it is
+	 * physically contiguous in host's address space.
+	 */
+	pipe->command_buffer =
+			(struct goldfish_pipe_command*)__get_free_page(GFP_KERNEL);
+	if (!pipe->command_buffer) {
+		status = -ENOMEM;
+		goto err_pipe;
+	}
+
+	spin_lock_irqsave(&dev->lock, flags);
+
+	id = get_free_pipe_id_locked(dev);
+	if (id < 0) {
+		status = id;
+		goto err_id_locked;
+	}
+
+	dev->pipes[id] = pipe;
+	pipe->id = id;
+	pipe->command_buffer->id = id;
+
+	/* Now tell the emulator we're opening a new pipe. */
+	dev->buffers->open_command_params.rw_params_max_count =
+			MAX_BUFFERS_PER_COMMAND;
+	dev->buffers->open_command_params.command_buffer_ptr =
+			(u64)(unsigned long)__pa(pipe->command_buffer);
+	status = goldfish_cmd_locked(pipe, PIPE_CMD_OPEN);
+	spin_unlock_irqrestore(&dev->lock, flags);
+	if (status < 0)
+		goto err_cmd;
+	/* All is done, save the pipe into the file's private data field */
+	file->private_data = pipe;
+	return 0;
+
+err_cmd:
+	spin_lock_irqsave(&dev->lock, flags);
+	dev->pipes[id] = NULL;
+err_id_locked:
+	spin_unlock_irqrestore(&dev->lock, flags);
+	free_page((unsigned long)pipe->command_buffer);
+err_pipe:
+	kfree(pipe);
+	return status;
+}
+
+static int goldfish_pipe_release(struct inode *inode, struct file *filp)
+{
+	unsigned long flags;
+	struct goldfish_pipe *pipe = filp->private_data;
+	struct goldfish_pipe_dev *dev = pipe->dev;
+
+	/* The guest is closing the channel, so tell the emulator right now */
+	(void)goldfish_cmd(pipe, PIPE_CMD_CLOSE);
+
+	spin_lock_irqsave(&dev->lock, flags);
+	dev->pipes[pipe->id] = NULL;
+	signalled_pipes_remove_locked(dev, pipe);
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	filp->private_data = NULL;
+	free_page((unsigned long)pipe->command_buffer);
+	kfree(pipe);
+	return 0;
+}
+
+static const struct file_operations goldfish_pipe_fops = {
+	.owner = THIS_MODULE,
+	.read = goldfish_pipe_read,
+	.write = goldfish_pipe_write,
+	.poll = goldfish_pipe_poll,
+	.open = goldfish_pipe_open,
+	.release = goldfish_pipe_release,
+};
+
+static struct miscdevice goldfish_pipe_dev = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "goldfish_pipe",
+	.fops = &goldfish_pipe_fops,
+};
+
+static int goldfish_pipe_device_init_v2(struct platform_device *pdev)
+{
+	char *page;
+	struct goldfish_pipe_dev *dev = pipe_dev;
+	int err = devm_request_irq(&pdev->dev, dev->irq, goldfish_pipe_interrupt,
+				IRQF_SHARED, "goldfish_pipe", dev);
+	if (err) {
+		dev_err(&pdev->dev, "unable to allocate IRQ for v2\n");
+		return err;
+	}
+
+	err = misc_register(&goldfish_pipe_dev);
+	if (err) {
+		dev_err(&pdev->dev, "unable to register v2 device\n");
+		return err;
+	}
+
+	dev->first_signalled_pipe = NULL;
+	dev->pipes_capacity = INITIAL_PIPES_CAPACITY;
+	dev->pipes = kcalloc(dev->pipes_capacity, sizeof(*dev->pipes), GFP_KERNEL);
+	if (!dev->pipes)
+		return -ENOMEM;
+
+	/*
+	 * We're going to pass two buffers, open_command_params and
+	 * signalled_pipe_buffers, to the host. This means each of those buffers
+	 * needs to be contained in a single physical page. The easiest choice is
+	 * to just allocate a page and place the buffers in it.
+	 */
+	BUG_ON(sizeof(*dev->buffers) > PAGE_SIZE);
+	page = (char*)__get_free_page(GFP_KERNEL);
+	if (!page) {
+		kfree(dev->pipes);
+		return -ENOMEM;
+	}
+	dev->buffers = (struct goldfish_pipe_dev_buffers*)page;
+
+	/* Send the buffer addresses to the host */
+	{
+		u64 paddr = __pa(&dev->buffers->signalled_pipe_buffers);
+		writel((u32)(unsigned long)(paddr >> 32), dev->base + PIPE_REG_SIGNAL_BUFFER_HIGH);
+		writel((u32)(unsigned long)paddr, dev->base + PIPE_REG_SIGNAL_BUFFER);
+		writel((u32)MAX_SIGNALLED_PIPES, dev->base + PIPE_REG_SIGNAL_BUFFER_COUNT);
+
+		paddr = __pa(&dev->buffers->open_command_params);
+		writel((u32)(unsigned long)(paddr >> 32), dev->base + PIPE_REG_OPEN_BUFFER_HIGH);
+		writel((u32)(unsigned long)paddr, dev->base + PIPE_REG_OPEN_BUFFER);
+	}
+	return 0;
+}
+
+static void goldfish_pipe_device_deinit_v2(struct platform_device *pdev) {
+	struct goldfish_pipe_dev *dev = pipe_dev;
+	misc_deregister(&goldfish_pipe_dev);
+	kfree(dev->pipes);
+	free_page((unsigned long)dev->buffers);
+}
+
+static int goldfish_pipe_probe(struct platform_device *pdev)
+{
+	int err;
+	struct resource *r;
+	struct goldfish_pipe_dev *dev = pipe_dev;
+
+	BUG_ON(sizeof(struct goldfish_pipe_command) > PAGE_SIZE);
+
+	/* not thread safe, but this should not happen */
+	WARN_ON(dev->base != NULL);
+
+	spin_lock_init(&dev->lock);
+
+	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (r == NULL || resource_size(r) < PAGE_SIZE) {
+		dev_err(&pdev->dev, "can't allocate i/o page\n");
+		return -EINVAL;
+	}
+	dev->base = devm_ioremap(&pdev->dev, r->start, PAGE_SIZE);
+	if (dev->base == NULL) {
+		dev_err(&pdev->dev, "ioremap failed\n");
+		return -EINVAL;
+	}
+
+	r = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (r == NULL) {
+		err = -EINVAL;
+		goto error;
+	}
+	dev->irq = r->start;
+
+	/*
+	 * Exchange the versions with the host device
+	 *
+	 * Note: v1 driver used to not report its version, so we write it before
+	 *  reading device version back: this allows the host implementation to
+	 *  detect the old driver (if there was no version write before read).
+	 */
+	writel((u32)PIPE_DRIVER_VERSION, dev->base + PIPE_REG_VERSION);
+	dev->version = readl(dev->base + PIPE_REG_VERSION);
+	if (dev->version < PIPE_CURRENT_DEVICE_VERSION) {
+		/* initialize the old device version */
+		err = goldfish_pipe_device_init_v1(pdev);
+	} else {
+		/* Host device supports the new interface */
+		err = goldfish_pipe_device_init_v2(pdev);
+	}
+	if (!err)
+		return 0;
+
+error:
+	dev->base = NULL;
+	return err;
+}
+
+static int goldfish_pipe_remove(struct platform_device *pdev)
+{
+	struct goldfish_pipe_dev *dev = pipe_dev;
+	if (dev->version < PIPE_CURRENT_DEVICE_VERSION)
+		goldfish_pipe_device_deinit_v1(pdev);
+	else
+		goldfish_pipe_device_deinit_v2(pdev);
+	dev->base = NULL;
+	return 0;
+}
+
+static const struct acpi_device_id goldfish_pipe_acpi_match[] = {
+	{ "GFSH0003", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, goldfish_pipe_acpi_match);
+
+static const struct of_device_id goldfish_pipe_of_match[] = {
+	{ .compatible = "google,android-pipe", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, goldfish_pipe_of_match);
+
+static struct platform_driver goldfish_pipe_driver = {
+	.probe = goldfish_pipe_probe,
+	.remove = goldfish_pipe_remove,
+	.driver = {
+		.name = "goldfish_pipe",
+		.of_match_table = goldfish_pipe_of_match,
+		.acpi_match_table = ACPI_PTR(goldfish_pipe_acpi_match),
+	}
+};
+
+module_platform_driver(goldfish_pipe_driver);
+MODULE_AUTHOR("David Turner <digit@google.com>");
+MODULE_LICENSE("GPL");
diff --git a/drivers/power/goldfish_battery.c b/drivers/power/goldfish_battery.c
index a50bb988c69a..f5c525e4482a 100644
--- a/drivers/power/goldfish_battery.c
+++ b/drivers/power/goldfish_battery.c
@@ -24,6 +24,7 @@
 #include <linux/pci.h>
 #include <linux/interrupt.h>
 #include <linux/io.h>
+#include <linux/acpi.h>
 
 struct goldfish_battery_data {
 	void __iomem *reg_base;
@@ -227,11 +228,25 @@ static int goldfish_battery_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id goldfish_battery_of_match[] = {
+	{ .compatible = "google,goldfish-battery", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, goldfish_battery_of_match);
+
+static const struct acpi_device_id goldfish_battery_acpi_match[] = {
+	{ "GFSH0001", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, goldfish_battery_acpi_match);
+
 static struct platform_driver goldfish_battery_device = {
 	.probe		= goldfish_battery_probe,
 	.remove		= goldfish_battery_remove,
 	.driver = {
-		.name = "goldfish-battery"
+		.name = "goldfish-battery",
+		.of_match_table = goldfish_battery_of_match,
+		.acpi_match_table = ACPI_PTR(goldfish_battery_acpi_match),
 	}
 };
 module_platform_driver(goldfish_battery_device);
diff --git a/drivers/power/power_supply_sysfs.c b/drivers/power/power_supply_sysfs.c
index ed2d7fd0c734..280018d59d5a 100644
--- a/drivers/power/power_supply_sysfs.c
+++ b/drivers/power/power_supply_sysfs.c
@@ -106,7 +106,10 @@ static ssize_t power_supply_show_property(struct device *dev,
 	else if (off >= POWER_SUPPLY_PROP_MODEL_NAME)
 		return sprintf(buf, "%s\n", value.strval);
 
-	return sprintf(buf, "%d\n", value.intval);
+	if (off == POWER_SUPPLY_PROP_CHARGE_COUNTER_EXT)
+		return sprintf(buf, "%lld\n", value.int64val);
+	else
+		return sprintf(buf, "%d\n", value.intval);
 }
 
 static ssize_t power_supply_store_property(struct device *dev,
@@ -197,6 +200,12 @@ static struct device_attribute power_supply_attrs[] = {
 	POWER_SUPPLY_ATTR(scope),
 	POWER_SUPPLY_ATTR(charge_term_current),
 	POWER_SUPPLY_ATTR(calibrate),
+	/* Local extensions */
+	POWER_SUPPLY_ATTR(usb_hc),
+	POWER_SUPPLY_ATTR(usb_otg),
+	POWER_SUPPLY_ATTR(charge_enabled),
+	/* Local extensions of type int64_t */
+	POWER_SUPPLY_ATTR(charge_counter_ext),
 	/* Properties of type `const char *' */
 	POWER_SUPPLY_ATTR(model_name),
 	POWER_SUPPLY_ATTR(manufacturer),
diff --git a/drivers/rtc/rtc-palmas.c b/drivers/rtc/rtc-palmas.c
index 7ea2c471feca..3e9663d0752b 100644
--- a/drivers/rtc/rtc-palmas.c
+++ b/drivers/rtc/rtc-palmas.c
@@ -45,6 +45,42 @@ struct palmas_rtc {
 /* Total number of RTC registers needed to set time*/
 #define PALMAS_NUM_TIME_REGS	(PALMAS_YEARS_REG - PALMAS_SECONDS_REG + 1)
 
+/*
+ * Special bin2bcd mapping to deal with bcd storage of year.
+ *
+ *   0-69                -> 0xD0
+ *  70-99  (1970 - 1999) -> 0xD0 - 0xF9 (correctly rolls to 0x00)
+ * 100-199 (2000 - 2099) -> 0x00 - 0x99 (does not roll to 0xA0 :-( )
+ * 200-229 (2100 - 2129) -> 0xA0 - 0xC9 (really for completeness)
+ * 230-                  -> 0xC9
+ *
+ * Confirmed: the only transition that does not work correctly for this rtc
+ * clock is the transition from 2099 to 2100, it proceeds to 2000. We will
+ * accept this issue since the clock retains and transitions the year correctly
+ * in all other conditions.
+ */
+static unsigned char year_bin2bcd(int val)
+{
+	if (val < 70)
+		return 0xD0;
+	if (val < 100)
+		return bin2bcd(val - 20) | 0x80; /* KISS leverage of bin2bcd */
+	if (val >= 230)
+		return 0xC9;
+	if (val >= 200)
+		return bin2bcd(val - 180) | 0x80;
+	return bin2bcd(val - 100);
+}
+
+static int year_bcd2bin(unsigned char val)
+{
+	if (val >= 0xD0)
+		return bcd2bin(val & 0x7F) + 20;
+	if (val >= 0xA0)
+		return bcd2bin(val & 0x7F) + 180;
+	return bcd2bin(val) + 100;
+}
+
 static int palmas_rtc_read_time(struct device *dev, struct rtc_time *tm)
 {
 	unsigned char rtc_data[PALMAS_NUM_TIME_REGS];
@@ -71,7 +107,7 @@ static int palmas_rtc_read_time(struct device *dev, struct rtc_time *tm)
 	tm->tm_hour = bcd2bin(rtc_data[2]);
 	tm->tm_mday = bcd2bin(rtc_data[3]);
 	tm->tm_mon = bcd2bin(rtc_data[4]) - 1;
-	tm->tm_year = bcd2bin(rtc_data[5]) + 100;
+	tm->tm_year = year_bcd2bin(rtc_data[5]);
 
 	return ret;
 }
@@ -87,7 +123,7 @@ static int palmas_rtc_set_time(struct device *dev, struct rtc_time *tm)
 	rtc_data[2] = bin2bcd(tm->tm_hour);
 	rtc_data[3] = bin2bcd(tm->tm_mday);
 	rtc_data[4] = bin2bcd(tm->tm_mon + 1);
-	rtc_data[5] = bin2bcd(tm->tm_year - 100);
+	rtc_data[5] = year_bin2bcd(tm->tm_year);
 
 	/* Stop RTC while updating the RTC time registers */
 	ret = palmas_update_bits(palmas, PALMAS_RTC_BASE, PALMAS_RTC_CTRL_REG,
@@ -142,7 +178,7 @@ static int palmas_rtc_read_alarm(struct device *dev, struct rtc_wkalrm *alm)
 	alm->time.tm_hour = bcd2bin(alarm_data[2]);
 	alm->time.tm_mday = bcd2bin(alarm_data[3]);
 	alm->time.tm_mon = bcd2bin(alarm_data[4]) - 1;
-	alm->time.tm_year = bcd2bin(alarm_data[5]) + 100;
+	alm->time.tm_year = year_bcd2bin(alarm_data[5]);
 
 	ret = palmas_read(palmas, PALMAS_RTC_BASE, PALMAS_RTC_INTERRUPTS_REG,
 			&int_val);
@@ -173,7 +209,7 @@ static int palmas_rtc_set_alarm(struct device *dev, struct rtc_wkalrm *alm)
 	alarm_data[2] = bin2bcd(alm->time.tm_hour);
 	alarm_data[3] = bin2bcd(alm->time.tm_mday);
 	alarm_data[4] = bin2bcd(alm->time.tm_mon + 1);
-	alarm_data[5] = bin2bcd(alm->time.tm_year - 100);
+	alarm_data[5] = year_bin2bcd(alm->time.tm_year);
 
 	ret = palmas_bulk_write(palmas, PALMAS_RTC_BASE,
 		PALMAS_ALARM_SECONDS_REG, alarm_data, PALMAS_NUM_TIME_REGS);
diff --git a/drivers/scsi/ufs/ufshcd.c b/drivers/scsi/ufs/ufshcd.c
index 0c2482ec7d21..1b9008cab6eb 100644
--- a/drivers/scsi/ufs/ufshcd.c
+++ b/drivers/scsi/ufs/ufshcd.c
@@ -39,6 +39,7 @@
 
 #include <linux/async.h>
 #include <linux/devfreq.h>
+#include <linux/blkdev.h>
 
 #include "ufshcd.h"
 #include "unipro.h"
@@ -1332,6 +1333,17 @@ static int ufshcd_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *cmd)
 		clear_bit_unlock(tag, &hba->lrb_in_use);
 		goto out;
 	}
+
+	/* IO svc time latency histogram */
+	if (hba != NULL && cmd->request != NULL) {
+		if (hba->latency_hist_enabled &&
+		    (cmd->request->cmd_type == REQ_TYPE_FS)) {
+			cmd->request->lat_hist_io_start = ktime_get();
+			cmd->request->lat_hist_enabled = 1;
+		} else
+			cmd->request->lat_hist_enabled = 0;
+	}
+
 	WARN_ON(hba->clk_gating.state != CLKS_ON);
 
 	lrbp = &hba->lrb[tag];
@@ -3160,6 +3172,7 @@ static void ufshcd_transfer_req_compl(struct ufs_hba *hba)
 	u32 tr_doorbell;
 	int result;
 	int index;
+	struct request *req;
 
 	/* Resetting interrupt aggregation counters first and reading the
 	 * DOOR_BELL afterward allows us to handle all the completed requests.
@@ -3184,6 +3197,22 @@ static void ufshcd_transfer_req_compl(struct ufs_hba *hba)
 			/* Mark completed command as NULL in LRB */
 			lrbp->cmd = NULL;
 			clear_bit_unlock(index, &hba->lrb_in_use);
+			req = cmd->request;
+			if (req) {
+				/* Update IO svc time latency histogram */
+				if (req->lat_hist_enabled) {
+					ktime_t completion;
+					u_int64_t delta_us;
+
+					completion = ktime_get();
+					delta_us = ktime_us_delta(completion,
+						  req->lat_hist_io_start);
+					/* rq_data_dir() => true if WRITE */
+					blk_update_latency_hist(&hba->io_lat_s,
+						(rq_data_dir(req) == READ),
+						delta_us);
+				}
+			}
 			/* Do not touch lrbp after scsi done */
 			cmd->scsi_done(cmd);
 			__ufshcd_release(hba);
@@ -5338,6 +5367,54 @@ out:
 }
 EXPORT_SYMBOL(ufshcd_shutdown);
 
+/*
+ * Values permitted 0, 1, 2.
+ * 0 -> Disable IO latency histograms (default)
+ * 1 -> Enable IO latency histograms
+ * 2 -> Zero out IO latency histograms
+ */
+static ssize_t
+latency_hist_store(struct device *dev, struct device_attribute *attr,
+		   const char *buf, size_t count)
+{
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+	long value;
+
+	if (kstrtol(buf, 0, &value))
+		return -EINVAL;
+	if (value == BLK_IO_LAT_HIST_ZERO)
+		blk_zero_latency_hist(&hba->io_lat_s);
+	else if (value == BLK_IO_LAT_HIST_ENABLE ||
+		 value == BLK_IO_LAT_HIST_DISABLE)
+		hba->latency_hist_enabled = value;
+	return count;
+}
+
+ssize_t
+latency_hist_show(struct device *dev, struct device_attribute *attr,
+		  char *buf)
+{
+	struct ufs_hba *hba = dev_get_drvdata(dev);
+
+	return blk_latency_hist_show(&hba->io_lat_s, buf);
+}
+
+static DEVICE_ATTR(latency_hist, S_IRUGO | S_IWUSR,
+		   latency_hist_show, latency_hist_store);
+
+static void
+ufshcd_init_latency_hist(struct ufs_hba *hba)
+{
+	if (device_create_file(hba->dev, &dev_attr_latency_hist))
+		dev_err(hba->dev, "Failed to create latency_hist sysfs entry\n");
+}
+
+static void
+ufshcd_exit_latency_hist(struct ufs_hba *hba)
+{
+	device_create_file(hba->dev, &dev_attr_latency_hist);
+}
+
 /**
  * ufshcd_remove - de-allocate SCSI host and host memory space
  *		data structure memory
@@ -5353,6 +5430,7 @@ void ufshcd_remove(struct ufs_hba *hba)
 	scsi_host_put(hba->host);
 
 	ufshcd_exit_clk_gating(hba);
+	ufshcd_exit_latency_hist(hba);
 	if (ufshcd_is_clkscaling_enabled(hba))
 		devfreq_remove_device(hba->devfreq);
 	ufshcd_hba_exit(hba);
@@ -5650,6 +5728,8 @@ int ufshcd_init(struct ufs_hba *hba, void __iomem *mmio_base, unsigned int irq)
 	/* Hold auto suspend until async scan completes */
 	pm_runtime_get_sync(dev);
 
+	ufshcd_init_latency_hist(hba);
+
 	/*
 	 * The device-initialize-sequence hasn't been invoked yet.
 	 * Set the device to power-off state
@@ -5664,6 +5744,7 @@ out_remove_scsi_host:
 	scsi_remove_host(hba->host);
 exit_gating:
 	ufshcd_exit_clk_gating(hba);
+	ufshcd_exit_latency_hist(hba);
 out_disable:
 	hba->is_irq_enabled = false;
 	scsi_host_put(host);
diff --git a/drivers/scsi/ufs/ufshcd.h b/drivers/scsi/ufs/ufshcd.h
index bb02100ab2dc..b74eedbd831f 100644
--- a/drivers/scsi/ufs/ufshcd.h
+++ b/drivers/scsi/ufs/ufshcd.h
@@ -540,6 +540,9 @@ struct ufs_hba {
 	struct devfreq *devfreq;
 	struct ufs_clk_scaling clk_scaling;
 	bool is_sys_suspended;
+
+	int			latency_hist_enabled;
+	struct io_latency_state io_lat_s;
 };
 
 /* Returns true if clocks can be gated. Otherwise false */
diff --git a/drivers/staging/android/Kconfig b/drivers/staging/android/Kconfig
index 42b15126aa06..0c0c092ce354 100644
--- a/drivers/staging/android/Kconfig
+++ b/drivers/staging/android/Kconfig
@@ -38,6 +38,15 @@ config ANDROID_LOW_MEMORY_KILLER
 	  scripts (/init.rc), and it defines priority values with minimum free memory size
 	  for each priority.
 
+config ANDROID_LOW_MEMORY_KILLER_AUTODETECT_OOM_ADJ_VALUES
+	bool "Android Low Memory Killer: detect oom_adj values"
+	depends on ANDROID_LOW_MEMORY_KILLER
+	default y
+	---help---
+	  Detect oom_adj values written to
+	  /sys/module/lowmemorykiller/parameters/adj and convert them
+	  to oom_score_adj values.
+
 config SYNC
 	bool "Synchronization framework"
 	default n
@@ -68,6 +77,8 @@ config SW_SYNC_USER
 
 source "drivers/staging/android/ion/Kconfig"
 
+source "drivers/staging/android/fiq_debugger/Kconfig"
+
 endif # if ANDROID
 
 endmenu
diff --git a/drivers/staging/android/Makefile b/drivers/staging/android/Makefile
index c7b6c99cc5ce..fcb7edca24c9 100644
--- a/drivers/staging/android/Makefile
+++ b/drivers/staging/android/Makefile
@@ -1,6 +1,7 @@
 ccflags-y += -I$(src)			# needed for trace events
 
 obj-y					+= ion/
+obj-$(CONFIG_FIQ_DEBUGGER)		+= fiq_debugger/
 
 obj-$(CONFIG_ASHMEM)			+= ashmem.o
 obj-$(CONFIG_ANDROID_TIMED_OUTPUT)	+= timed_output.o
diff --git a/drivers/staging/android/ashmem.c b/drivers/staging/android/ashmem.c
index 9c6357c03905..e4530ac6d5d4 100644
--- a/drivers/staging/android/ashmem.c
+++ b/drivers/staging/android/ashmem.c
@@ -397,22 +397,14 @@ static int ashmem_mmap(struct file *file, struct vm_area_struct *vma)
 	}
 	get_file(asma->file);
 
-	/*
-	 * XXX - Reworked to use shmem_zero_setup() instead of
-	 * shmem_set_file while we're in staging. -jstultz
-	 */
-	if (vma->vm_flags & VM_SHARED) {
-		ret = shmem_zero_setup(vma);
-		if (ret) {
-			fput(asma->file);
-			goto out;
-		}
+	if (vma->vm_flags & VM_SHARED)
+		shmem_set_file(vma, asma->file);
+	else {
+		if (vma->vm_file)
+			fput(vma->vm_file);
+		vma->vm_file = asma->file;
 	}
 
-	if (vma->vm_file)
-		fput(vma->vm_file);
-	vma->vm_file = asma->file;
-
 out:
 	mutex_unlock(&ashmem_mutex);
 	return ret;
@@ -442,14 +434,16 @@ ashmem_shrink_scan(struct shrinker *shrink, struct shrink_control *sc)
 	if (!(sc->gfp_mask & __GFP_FS))
 		return SHRINK_STOP;
 
-	mutex_lock(&ashmem_mutex);
+	if (!mutex_trylock(&ashmem_mutex))
+		return -1;
+
 	list_for_each_entry_safe(range, next, &ashmem_lru_list, lru) {
 		loff_t start = range->pgstart * PAGE_SIZE;
 		loff_t end = (range->pgend + 1) * PAGE_SIZE;
 
-		vfs_fallocate(range->asma->file,
-			      FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
-			      start, end - start);
+		range->asma->file->f_op->fallocate(range->asma->file,
+				FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE,
+				start, end - start);
 		range->purged = ASHMEM_WAS_PURGED;
 		lru_del(range);
 
diff --git a/drivers/staging/android/fiq_debugger/Kconfig b/drivers/staging/android/fiq_debugger/Kconfig
new file mode 100644
index 000000000000..60fc224d4efc
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/Kconfig
@@ -0,0 +1,58 @@
+config FIQ_DEBUGGER
+	bool "FIQ Mode Serial Debugger"
+	default n
+	depends on ARM || ARM64
+	help
+	  The FIQ serial debugger can accept commands even when the
+	  kernel is unresponsive due to being stuck with interrupts
+	  disabled.
+
+config FIQ_DEBUGGER_NO_SLEEP
+	bool "Keep serial debugger active"
+	depends on FIQ_DEBUGGER
+	default n
+	help
+	  Enables the serial debugger at boot. Passing
+	  fiq_debugger.no_sleep on the kernel commandline will
+	  override this config option.
+
+config FIQ_DEBUGGER_WAKEUP_IRQ_ALWAYS_ON
+	bool "Don't disable wakeup IRQ when debugger is active"
+	depends on FIQ_DEBUGGER
+	default n
+	help
+	  Don't disable the wakeup irq when enabling the uart clock.  This will
+	  cause extra interrupts, but it makes the serial debugger usable with
+	  on some MSM radio builds that ignore the uart clock request in power
+	  collapse.
+
+config FIQ_DEBUGGER_CONSOLE
+	bool "Console on FIQ Serial Debugger port"
+	depends on FIQ_DEBUGGER
+	default n
+	help
+	  Enables a console so that printk messages are displayed on
+	  the debugger serial port as the occur.
+
+config FIQ_DEBUGGER_CONSOLE_DEFAULT_ENABLE
+	bool "Put the FIQ debugger into console mode by default"
+	depends on FIQ_DEBUGGER_CONSOLE
+	default n
+	help
+	  If enabled, this puts the fiq debugger into console mode by default.
+	  Otherwise, the fiq debugger will start out in debug mode.
+
+config FIQ_DEBUGGER_UART_OVERLAY
+	bool "Install uart DT overlay"
+	depends on FIQ_DEBUGGER
+	select OF_OVERLAY
+	default n
+	help
+	  If enabled, fiq debugger is calling fiq_debugger_uart_overlay()
+	  that will apply overlay uart_overlay@0 to disable proper uart.
+
+config FIQ_WATCHDOG
+	bool
+	select FIQ_DEBUGGER
+	select PSTORE_RAM
+	default n
diff --git a/drivers/staging/android/fiq_debugger/Makefile b/drivers/staging/android/fiq_debugger/Makefile
new file mode 100644
index 000000000000..a7ca4871cad3
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/Makefile
@@ -0,0 +1,4 @@
+obj-y			+= fiq_debugger.o
+obj-$(CONFIG_ARM)	+= fiq_debugger_arm.o
+obj-$(CONFIG_ARM64)	+= fiq_debugger_arm64.o
+obj-$(CONFIG_FIQ_WATCHDOG)	+= fiq_watchdog.o
diff --git a/drivers/staging/android/fiq_debugger/fiq_debugger.c b/drivers/staging/android/fiq_debugger/fiq_debugger.c
new file mode 100644
index 000000000000..ef9fa483ac00
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_debugger.c
@@ -0,0 +1,1248 @@
+/*
+ * drivers/staging/android/fiq_debugger.c
+ *
+ * Serial Debugger Interface accessed through an FIQ interrupt.
+ *
+ * Copyright (C) 2008 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <stdarg.h>
+#include <linux/module.h>
+#include <linux/io.h>
+#include <linux/console.h>
+#include <linux/interrupt.h>
+#include <linux/clk.h>
+#include <linux/platform_device.h>
+#include <linux/kernel_stat.h>
+#include <linux/kmsg_dump.h>
+#include <linux/irq.h>
+#include <linux/delay.h>
+#include <linux/reboot.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/smp.h>
+#include <linux/timer.h>
+#include <linux/tty.h>
+#include <linux/tty_flip.h>
+#include <linux/wakelock.h>
+
+#ifdef CONFIG_FIQ_GLUE
+#include <asm/fiq_glue.h>
+#endif
+
+#ifdef CONFIG_FIQ_DEBUGGER_UART_OVERLAY
+#include <linux/of.h>
+#endif
+
+#include <linux/uaccess.h>
+
+#include "fiq_debugger.h"
+#include "fiq_debugger_priv.h"
+#include "fiq_debugger_ringbuf.h"
+
+#define DEBUG_MAX 64
+#define MAX_UNHANDLED_FIQ_COUNT 1000000
+
+#define MAX_FIQ_DEBUGGER_PORTS 4
+
+struct fiq_debugger_state {
+#ifdef CONFIG_FIQ_GLUE
+	struct fiq_glue_handler handler;
+#endif
+	struct fiq_debugger_output output;
+
+	int fiq;
+	int uart_irq;
+	int signal_irq;
+	int wakeup_irq;
+	bool wakeup_irq_no_set_wake;
+	struct clk *clk;
+	struct fiq_debugger_pdata *pdata;
+	struct platform_device *pdev;
+
+	char debug_cmd[DEBUG_MAX];
+	int debug_busy;
+	int debug_abort;
+
+	char debug_buf[DEBUG_MAX];
+	int debug_count;
+
+	bool no_sleep;
+	bool debug_enable;
+	bool ignore_next_wakeup_irq;
+	struct timer_list sleep_timer;
+	spinlock_t sleep_timer_lock;
+	bool uart_enabled;
+	struct wake_lock debugger_wake_lock;
+	bool console_enable;
+	int current_cpu;
+	atomic_t unhandled_fiq_count;
+	bool in_fiq;
+
+	struct work_struct work;
+	spinlock_t work_lock;
+	char work_cmd[DEBUG_MAX];
+
+#ifdef CONFIG_FIQ_DEBUGGER_CONSOLE
+	spinlock_t console_lock;
+	struct console console;
+	struct tty_port tty_port;
+	struct fiq_debugger_ringbuf *tty_rbuf;
+	bool syslog_dumping;
+#endif
+
+	unsigned int last_irqs[NR_IRQS];
+	unsigned int last_local_timer_irqs[NR_CPUS];
+};
+
+#ifdef CONFIG_FIQ_DEBUGGER_CONSOLE
+struct tty_driver *fiq_tty_driver;
+#endif
+
+#ifdef CONFIG_FIQ_DEBUGGER_NO_SLEEP
+static bool initial_no_sleep = true;
+#else
+static bool initial_no_sleep;
+#endif
+
+#ifdef CONFIG_FIQ_DEBUGGER_CONSOLE_DEFAULT_ENABLE
+static bool initial_debug_enable = true;
+static bool initial_console_enable = true;
+#else
+static bool initial_debug_enable;
+static bool initial_console_enable;
+#endif
+
+static bool fiq_kgdb_enable;
+static bool fiq_debugger_disable;
+
+module_param_named(no_sleep, initial_no_sleep, bool, 0644);
+module_param_named(debug_enable, initial_debug_enable, bool, 0644);
+module_param_named(console_enable, initial_console_enable, bool, 0644);
+module_param_named(kgdb_enable, fiq_kgdb_enable, bool, 0644);
+module_param_named(disable, fiq_debugger_disable, bool, 0644);
+
+#ifdef CONFIG_FIQ_DEBUGGER_WAKEUP_IRQ_ALWAYS_ON
+static inline
+void fiq_debugger_enable_wakeup_irq(struct fiq_debugger_state *state) {}
+static inline
+void fiq_debugger_disable_wakeup_irq(struct fiq_debugger_state *state) {}
+#else
+static inline
+void fiq_debugger_enable_wakeup_irq(struct fiq_debugger_state *state)
+{
+	if (state->wakeup_irq < 0)
+		return;
+	enable_irq(state->wakeup_irq);
+	if (!state->wakeup_irq_no_set_wake)
+		enable_irq_wake(state->wakeup_irq);
+}
+static inline
+void fiq_debugger_disable_wakeup_irq(struct fiq_debugger_state *state)
+{
+	if (state->wakeup_irq < 0)
+		return;
+	disable_irq_nosync(state->wakeup_irq);
+	if (!state->wakeup_irq_no_set_wake)
+		disable_irq_wake(state->wakeup_irq);
+}
+#endif
+
+static inline bool fiq_debugger_have_fiq(struct fiq_debugger_state *state)
+{
+	return (state->fiq >= 0);
+}
+
+#ifdef CONFIG_FIQ_GLUE
+static void fiq_debugger_force_irq(struct fiq_debugger_state *state)
+{
+	unsigned int irq = state->signal_irq;
+
+	if (WARN_ON(!fiq_debugger_have_fiq(state)))
+		return;
+	if (state->pdata->force_irq) {
+		state->pdata->force_irq(state->pdev, irq);
+	} else {
+		struct irq_chip *chip = irq_get_chip(irq);
+		if (chip && chip->irq_retrigger)
+			chip->irq_retrigger(irq_get_irq_data(irq));
+	}
+}
+#endif
+
+static void fiq_debugger_uart_enable(struct fiq_debugger_state *state)
+{
+	if (state->clk)
+		clk_enable(state->clk);
+	if (state->pdata->uart_enable)
+		state->pdata->uart_enable(state->pdev);
+}
+
+static void fiq_debugger_uart_disable(struct fiq_debugger_state *state)
+{
+	if (state->pdata->uart_disable)
+		state->pdata->uart_disable(state->pdev);
+	if (state->clk)
+		clk_disable(state->clk);
+}
+
+static void fiq_debugger_uart_flush(struct fiq_debugger_state *state)
+{
+	if (state->pdata->uart_flush)
+		state->pdata->uart_flush(state->pdev);
+}
+
+static void fiq_debugger_putc(struct fiq_debugger_state *state, char c)
+{
+	state->pdata->uart_putc(state->pdev, c);
+}
+
+static void fiq_debugger_puts(struct fiq_debugger_state *state, char *s)
+{
+	unsigned c;
+	while ((c = *s++)) {
+		if (c == '\n')
+			fiq_debugger_putc(state, '\r');
+		fiq_debugger_putc(state, c);
+	}
+}
+
+static void fiq_debugger_prompt(struct fiq_debugger_state *state)
+{
+	fiq_debugger_puts(state, "debug> ");
+}
+
+static void fiq_debugger_dump_kernel_log(struct fiq_debugger_state *state)
+{
+	char buf[512];
+	size_t len;
+	struct kmsg_dumper dumper = { .active = true };
+
+
+	kmsg_dump_rewind_nolock(&dumper);
+	while (kmsg_dump_get_line_nolock(&dumper, true, buf,
+					 sizeof(buf) - 1, &len)) {
+		buf[len] = 0;
+		fiq_debugger_puts(state, buf);
+	}
+}
+
+static void fiq_debugger_printf(struct fiq_debugger_output *output,
+			       const char *fmt, ...)
+{
+	struct fiq_debugger_state *state;
+	char buf[256];
+	va_list ap;
+
+	state = container_of(output, struct fiq_debugger_state, output);
+	va_start(ap, fmt);
+	vsnprintf(buf, sizeof(buf), fmt, ap);
+	va_end(ap);
+
+	fiq_debugger_puts(state, buf);
+}
+
+/* Safe outside fiq context */
+static int fiq_debugger_printf_nfiq(void *cookie, const char *fmt, ...)
+{
+	struct fiq_debugger_state *state = cookie;
+	char buf[256];
+	va_list ap;
+	unsigned long irq_flags;
+
+	va_start(ap, fmt);
+	vsnprintf(buf, 128, fmt, ap);
+	va_end(ap);
+
+	local_irq_save(irq_flags);
+	fiq_debugger_puts(state, buf);
+	fiq_debugger_uart_flush(state);
+	local_irq_restore(irq_flags);
+	return state->debug_abort;
+}
+
+static void fiq_debugger_dump_irqs(struct fiq_debugger_state *state)
+{
+	int n;
+	struct irq_desc *desc;
+
+	fiq_debugger_printf(&state->output,
+			"irqnr       total  since-last   status  name\n");
+	for_each_irq_desc(n, desc) {
+		struct irqaction *act = desc->action;
+		if (!act && !kstat_irqs(n))
+			continue;
+		fiq_debugger_printf(&state->output, "%5d: %10u %11u %8x  %s\n", n,
+			kstat_irqs(n),
+			kstat_irqs(n) - state->last_irqs[n],
+			desc->status_use_accessors,
+			(act && act->name) ? act->name : "???");
+		state->last_irqs[n] = kstat_irqs(n);
+	}
+}
+
+static void fiq_debugger_do_ps(struct fiq_debugger_state *state)
+{
+	struct task_struct *g;
+	struct task_struct *p;
+	unsigned task_state;
+	static const char stat_nam[] = "RSDTtZX";
+
+	fiq_debugger_printf(&state->output, "pid   ppid  prio task            pc\n");
+	read_lock(&tasklist_lock);
+	do_each_thread(g, p) {
+		task_state = p->state ? __ffs(p->state) + 1 : 0;
+		fiq_debugger_printf(&state->output,
+			     "%5d %5d %4d ", p->pid, p->parent->pid, p->prio);
+		fiq_debugger_printf(&state->output, "%-13.13s %c", p->comm,
+			     task_state >= sizeof(stat_nam) ? '?' : stat_nam[task_state]);
+		if (task_state == TASK_RUNNING)
+			fiq_debugger_printf(&state->output, " running\n");
+		else
+			fiq_debugger_printf(&state->output, " %08lx\n",
+					thread_saved_pc(p));
+	} while_each_thread(g, p);
+	read_unlock(&tasklist_lock);
+}
+
+#ifdef CONFIG_FIQ_DEBUGGER_CONSOLE
+static void fiq_debugger_begin_syslog_dump(struct fiq_debugger_state *state)
+{
+	state->syslog_dumping = true;
+}
+
+static void fiq_debugger_end_syslog_dump(struct fiq_debugger_state *state)
+{
+	state->syslog_dumping = false;
+}
+#else
+extern int do_syslog(int type, char __user *bug, int count);
+static void fiq_debugger_begin_syslog_dump(struct fiq_debugger_state *state)
+{
+	do_syslog(5 /* clear */, NULL, 0);
+}
+
+static void fiq_debugger_end_syslog_dump(struct fiq_debugger_state *state)
+{
+	fiq_debugger_dump_kernel_log(state);
+}
+#endif
+
+static void fiq_debugger_do_sysrq(struct fiq_debugger_state *state, char rq)
+{
+	if ((rq == 'g' || rq == 'G') && !fiq_kgdb_enable) {
+		fiq_debugger_printf(&state->output, "sysrq-g blocked\n");
+		return;
+	}
+	fiq_debugger_begin_syslog_dump(state);
+	handle_sysrq(rq);
+	fiq_debugger_end_syslog_dump(state);
+}
+
+#ifdef CONFIG_KGDB
+static void fiq_debugger_do_kgdb(struct fiq_debugger_state *state)
+{
+	if (!fiq_kgdb_enable) {
+		fiq_debugger_printf(&state->output, "kgdb through fiq debugger not enabled\n");
+		return;
+	}
+
+	fiq_debugger_printf(&state->output, "enabling console and triggering kgdb\n");
+	state->console_enable = true;
+	handle_sysrq('g');
+}
+#endif
+
+static void fiq_debugger_schedule_work(struct fiq_debugger_state *state,
+		char *cmd)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&state->work_lock, flags);
+	if (state->work_cmd[0] != '\0') {
+		fiq_debugger_printf(&state->output, "work command processor busy\n");
+		spin_unlock_irqrestore(&state->work_lock, flags);
+		return;
+	}
+
+	strlcpy(state->work_cmd, cmd, sizeof(state->work_cmd));
+	spin_unlock_irqrestore(&state->work_lock, flags);
+
+	schedule_work(&state->work);
+}
+
+static void fiq_debugger_work(struct work_struct *work)
+{
+	struct fiq_debugger_state *state;
+	char work_cmd[DEBUG_MAX];
+	char *cmd;
+	unsigned long flags;
+
+	state = container_of(work, struct fiq_debugger_state, work);
+
+	spin_lock_irqsave(&state->work_lock, flags);
+
+	strlcpy(work_cmd, state->work_cmd, sizeof(work_cmd));
+	state->work_cmd[0] = '\0';
+
+	spin_unlock_irqrestore(&state->work_lock, flags);
+
+	cmd = work_cmd;
+	if (!strncmp(cmd, "reboot", 6)) {
+		cmd += 6;
+		while (*cmd == ' ')
+			cmd++;
+		if (*cmd != '\0')
+			kernel_restart(cmd);
+		else
+			kernel_restart(NULL);
+	} else {
+		fiq_debugger_printf(&state->output, "unknown work command '%s'\n",
+				work_cmd);
+	}
+}
+
+/* This function CANNOT be called in FIQ context */
+static void fiq_debugger_irq_exec(struct fiq_debugger_state *state, char *cmd)
+{
+	if (!strcmp(cmd, "ps"))
+		fiq_debugger_do_ps(state);
+	if (!strcmp(cmd, "sysrq"))
+		fiq_debugger_do_sysrq(state, 'h');
+	if (!strncmp(cmd, "sysrq ", 6))
+		fiq_debugger_do_sysrq(state, cmd[6]);
+#ifdef CONFIG_KGDB
+	if (!strcmp(cmd, "kgdb"))
+		fiq_debugger_do_kgdb(state);
+#endif
+	if (!strncmp(cmd, "reboot", 6))
+		fiq_debugger_schedule_work(state, cmd);
+}
+
+static void fiq_debugger_help(struct fiq_debugger_state *state)
+{
+	fiq_debugger_printf(&state->output,
+				"FIQ Debugger commands:\n"
+				" pc            PC status\n"
+				" regs          Register dump\n"
+				" allregs       Extended Register dump\n"
+				" bt            Stack trace\n"
+				" reboot [<c>]  Reboot with command <c>\n"
+				" reset [<c>]   Hard reset with command <c>\n"
+				" irqs          Interupt status\n"
+				" kmsg          Kernel log\n"
+				" version       Kernel version\n");
+	fiq_debugger_printf(&state->output,
+				" sleep         Allow sleep while in FIQ\n"
+				" nosleep       Disable sleep while in FIQ\n"
+				" console       Switch terminal to console\n"
+				" cpu           Current CPU\n"
+				" cpu <number>  Switch to CPU<number>\n");
+	fiq_debugger_printf(&state->output,
+				" ps            Process list\n"
+				" sysrq         sysrq options\n"
+				" sysrq <param> Execute sysrq with <param>\n");
+#ifdef CONFIG_KGDB
+	fiq_debugger_printf(&state->output,
+				" kgdb          Enter kernel debugger\n");
+#endif
+}
+
+static void fiq_debugger_take_affinity(void *info)
+{
+	struct fiq_debugger_state *state = info;
+	struct cpumask cpumask;
+
+	cpumask_clear(&cpumask);
+	cpumask_set_cpu(get_cpu(), &cpumask);
+
+	irq_set_affinity(state->uart_irq, &cpumask);
+}
+
+static void fiq_debugger_switch_cpu(struct fiq_debugger_state *state, int cpu)
+{
+	if (!fiq_debugger_have_fiq(state))
+		smp_call_function_single(cpu, fiq_debugger_take_affinity, state,
+				false);
+	state->current_cpu = cpu;
+}
+
+static bool fiq_debugger_fiq_exec(struct fiq_debugger_state *state,
+			const char *cmd, const struct pt_regs *regs,
+			void *svc_sp)
+{
+	bool signal_helper = false;
+
+	if (!strcmp(cmd, "help") || !strcmp(cmd, "?")) {
+		fiq_debugger_help(state);
+	} else if (!strcmp(cmd, "pc")) {
+		fiq_debugger_dump_pc(&state->output, regs);
+	} else if (!strcmp(cmd, "regs")) {
+		fiq_debugger_dump_regs(&state->output, regs);
+	} else if (!strcmp(cmd, "allregs")) {
+		fiq_debugger_dump_allregs(&state->output, regs);
+	} else if (!strcmp(cmd, "bt")) {
+		fiq_debugger_dump_stacktrace(&state->output, regs, 100, svc_sp);
+	} else if (!strncmp(cmd, "reset", 5)) {
+		cmd += 5;
+		while (*cmd == ' ')
+			cmd++;
+		if (*cmd) {
+			char tmp_cmd[32];
+			strlcpy(tmp_cmd, cmd, sizeof(tmp_cmd));
+			machine_restart(tmp_cmd);
+		} else {
+			machine_restart(NULL);
+		}
+	} else if (!strcmp(cmd, "irqs")) {
+		fiq_debugger_dump_irqs(state);
+	} else if (!strcmp(cmd, "kmsg")) {
+		fiq_debugger_dump_kernel_log(state);
+	} else if (!strcmp(cmd, "version")) {
+		fiq_debugger_printf(&state->output, "%s\n", linux_banner);
+	} else if (!strcmp(cmd, "sleep")) {
+		state->no_sleep = false;
+		fiq_debugger_printf(&state->output, "enabling sleep\n");
+	} else if (!strcmp(cmd, "nosleep")) {
+		state->no_sleep = true;
+		fiq_debugger_printf(&state->output, "disabling sleep\n");
+	} else if (!strcmp(cmd, "console")) {
+		fiq_debugger_printf(&state->output, "console mode\n");
+		fiq_debugger_uart_flush(state);
+		state->console_enable = true;
+	} else if (!strcmp(cmd, "cpu")) {
+		fiq_debugger_printf(&state->output, "cpu %d\n", state->current_cpu);
+	} else if (!strncmp(cmd, "cpu ", 4)) {
+		unsigned long cpu = 0;
+		if (kstrtoul(cmd + 4, 10, &cpu) == 0)
+			fiq_debugger_switch_cpu(state, cpu);
+		else
+			fiq_debugger_printf(&state->output, "invalid cpu\n");
+		fiq_debugger_printf(&state->output, "cpu %d\n", state->current_cpu);
+	} else {
+		if (state->debug_busy) {
+			fiq_debugger_printf(&state->output,
+				"command processor busy. trying to abort.\n");
+			state->debug_abort = -1;
+		} else {
+			strcpy(state->debug_cmd, cmd);
+			state->debug_busy = 1;
+		}
+
+		return true;
+	}
+	if (!state->console_enable)
+		fiq_debugger_prompt(state);
+
+	return signal_helper;
+}
+
+static void fiq_debugger_sleep_timer_expired(unsigned long data)
+{
+	struct fiq_debugger_state *state = (struct fiq_debugger_state *)data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&state->sleep_timer_lock, flags);
+	if (state->uart_enabled && !state->no_sleep) {
+		if (state->debug_enable && !state->console_enable) {
+			state->debug_enable = false;
+			fiq_debugger_printf_nfiq(state,
+					"suspending fiq debugger\n");
+		}
+		state->ignore_next_wakeup_irq = true;
+		fiq_debugger_uart_disable(state);
+		state->uart_enabled = false;
+		fiq_debugger_enable_wakeup_irq(state);
+	}
+	wake_unlock(&state->debugger_wake_lock);
+	spin_unlock_irqrestore(&state->sleep_timer_lock, flags);
+}
+
+static void fiq_debugger_handle_wakeup(struct fiq_debugger_state *state)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&state->sleep_timer_lock, flags);
+	if (state->wakeup_irq >= 0 && state->ignore_next_wakeup_irq) {
+		state->ignore_next_wakeup_irq = false;
+	} else if (!state->uart_enabled) {
+		wake_lock(&state->debugger_wake_lock);
+		fiq_debugger_uart_enable(state);
+		state->uart_enabled = true;
+		fiq_debugger_disable_wakeup_irq(state);
+		mod_timer(&state->sleep_timer, jiffies + HZ / 2);
+	}
+	spin_unlock_irqrestore(&state->sleep_timer_lock, flags);
+}
+
+static irqreturn_t fiq_debugger_wakeup_irq_handler(int irq, void *dev)
+{
+	struct fiq_debugger_state *state = dev;
+
+	if (!state->no_sleep)
+		fiq_debugger_puts(state, "WAKEUP\n");
+	fiq_debugger_handle_wakeup(state);
+
+	return IRQ_HANDLED;
+}
+
+static
+void fiq_debugger_handle_console_irq_context(struct fiq_debugger_state *state)
+{
+#if defined(CONFIG_FIQ_DEBUGGER_CONSOLE)
+	if (state->tty_port.ops) {
+		int i;
+		int count = fiq_debugger_ringbuf_level(state->tty_rbuf);
+		for (i = 0; i < count; i++) {
+			int c = fiq_debugger_ringbuf_peek(state->tty_rbuf, 0);
+			tty_insert_flip_char(&state->tty_port, c, TTY_NORMAL);
+			if (!fiq_debugger_ringbuf_consume(state->tty_rbuf, 1))
+				pr_warn("fiq tty failed to consume byte\n");
+		}
+		tty_flip_buffer_push(&state->tty_port);
+	}
+#endif
+}
+
+static void fiq_debugger_handle_irq_context(struct fiq_debugger_state *state)
+{
+	if (!state->no_sleep) {
+		unsigned long flags;
+
+		spin_lock_irqsave(&state->sleep_timer_lock, flags);
+		wake_lock(&state->debugger_wake_lock);
+		mod_timer(&state->sleep_timer, jiffies + HZ * 5);
+		spin_unlock_irqrestore(&state->sleep_timer_lock, flags);
+	}
+	fiq_debugger_handle_console_irq_context(state);
+	if (state->debug_busy) {
+		fiq_debugger_irq_exec(state, state->debug_cmd);
+		if (!state->console_enable)
+			fiq_debugger_prompt(state);
+		state->debug_busy = 0;
+	}
+}
+
+static int fiq_debugger_getc(struct fiq_debugger_state *state)
+{
+	return state->pdata->uart_getc(state->pdev);
+}
+
+static bool fiq_debugger_handle_uart_interrupt(struct fiq_debugger_state *state,
+			int this_cpu, const struct pt_regs *regs, void *svc_sp)
+{
+	int c;
+	static int last_c;
+	int count = 0;
+	bool signal_helper = false;
+
+	if (this_cpu != state->current_cpu) {
+		if (state->in_fiq)
+			return false;
+
+		if (atomic_inc_return(&state->unhandled_fiq_count) !=
+					MAX_UNHANDLED_FIQ_COUNT)
+			return false;
+
+		fiq_debugger_printf(&state->output,
+			"fiq_debugger: cpu %d not responding, "
+			"reverting to cpu %d\n", state->current_cpu,
+			this_cpu);
+
+		atomic_set(&state->unhandled_fiq_count, 0);
+		fiq_debugger_switch_cpu(state, this_cpu);
+		return false;
+	}
+
+	state->in_fiq = true;
+
+	while ((c = fiq_debugger_getc(state)) != FIQ_DEBUGGER_NO_CHAR) {
+		count++;
+		if (!state->debug_enable) {
+			if ((c == 13) || (c == 10)) {
+				state->debug_enable = true;
+				state->debug_count = 0;
+				fiq_debugger_prompt(state);
+			}
+		} else if (c == FIQ_DEBUGGER_BREAK) {
+			state->console_enable = false;
+			fiq_debugger_puts(state, "fiq debugger mode\n");
+			state->debug_count = 0;
+			fiq_debugger_prompt(state);
+#ifdef CONFIG_FIQ_DEBUGGER_CONSOLE
+		} else if (state->console_enable && state->tty_rbuf) {
+			fiq_debugger_ringbuf_push(state->tty_rbuf, c);
+			signal_helper = true;
+#endif
+		} else if ((c >= ' ') && (c < 127)) {
+			if (state->debug_count < (DEBUG_MAX - 1)) {
+				state->debug_buf[state->debug_count++] = c;
+				fiq_debugger_putc(state, c);
+			}
+		} else if ((c == 8) || (c == 127)) {
+			if (state->debug_count > 0) {
+				state->debug_count--;
+				fiq_debugger_putc(state, 8);
+				fiq_debugger_putc(state, ' ');
+				fiq_debugger_putc(state, 8);
+			}
+		} else if ((c == 13) || (c == 10)) {
+			if (c == '\r' || (c == '\n' && last_c != '\r')) {
+				fiq_debugger_putc(state, '\r');
+				fiq_debugger_putc(state, '\n');
+			}
+			if (state->debug_count) {
+				state->debug_buf[state->debug_count] = 0;
+				state->debug_count = 0;
+				signal_helper |=
+					fiq_debugger_fiq_exec(state,
+							state->debug_buf,
+							regs, svc_sp);
+			} else {
+				fiq_debugger_prompt(state);
+			}
+		}
+		last_c = c;
+	}
+	if (!state->console_enable)
+		fiq_debugger_uart_flush(state);
+	if (state->pdata->fiq_ack)
+		state->pdata->fiq_ack(state->pdev, state->fiq);
+
+	/* poke sleep timer if necessary */
+	if (state->debug_enable && !state->no_sleep)
+		signal_helper = true;
+
+	atomic_set(&state->unhandled_fiq_count, 0);
+	state->in_fiq = false;
+
+	return signal_helper;
+}
+
+#ifdef CONFIG_FIQ_GLUE
+static void fiq_debugger_fiq(struct fiq_glue_handler *h,
+		const struct pt_regs *regs, void *svc_sp)
+{
+	struct fiq_debugger_state *state =
+		container_of(h, struct fiq_debugger_state, handler);
+	unsigned int this_cpu = THREAD_INFO(svc_sp)->cpu;
+	bool need_irq;
+
+	need_irq = fiq_debugger_handle_uart_interrupt(state, this_cpu, regs,
+			svc_sp);
+	if (need_irq)
+		fiq_debugger_force_irq(state);
+}
+#endif
+
+/*
+ * When not using FIQs, we only use this single interrupt as an entry point.
+ * This just effectively takes over the UART interrupt and does all the work
+ * in this context.
+ */
+static irqreturn_t fiq_debugger_uart_irq(int irq, void *dev)
+{
+	struct fiq_debugger_state *state = dev;
+	bool not_done;
+
+	fiq_debugger_handle_wakeup(state);
+
+	/* handle the debugger irq in regular context */
+	not_done = fiq_debugger_handle_uart_interrupt(state, smp_processor_id(),
+					      get_irq_regs(),
+					      current_thread_info());
+	if (not_done)
+		fiq_debugger_handle_irq_context(state);
+
+	return IRQ_HANDLED;
+}
+
+/*
+ * If FIQs are used, not everything can happen in fiq context.
+ * FIQ handler does what it can and then signals this interrupt to finish the
+ * job in irq context.
+ */
+static irqreturn_t fiq_debugger_signal_irq(int irq, void *dev)
+{
+	struct fiq_debugger_state *state = dev;
+
+	if (state->pdata->force_irq_ack)
+		state->pdata->force_irq_ack(state->pdev, state->signal_irq);
+
+	fiq_debugger_handle_irq_context(state);
+
+	return IRQ_HANDLED;
+}
+
+#ifdef CONFIG_FIQ_GLUE
+static void fiq_debugger_resume(struct fiq_glue_handler *h)
+{
+	struct fiq_debugger_state *state =
+		container_of(h, struct fiq_debugger_state, handler);
+	if (state->pdata->uart_resume)
+		state->pdata->uart_resume(state->pdev);
+}
+#endif
+
+#if defined(CONFIG_FIQ_DEBUGGER_CONSOLE)
+struct tty_driver *fiq_debugger_console_device(struct console *co, int *index)
+{
+	*index = co->index;
+	return fiq_tty_driver;
+}
+
+static void fiq_debugger_console_write(struct console *co,
+				const char *s, unsigned int count)
+{
+	struct fiq_debugger_state *state;
+	unsigned long flags;
+
+	state = container_of(co, struct fiq_debugger_state, console);
+
+	if (!state->console_enable && !state->syslog_dumping)
+		return;
+
+	fiq_debugger_uart_enable(state);
+	spin_lock_irqsave(&state->console_lock, flags);
+	while (count--) {
+		if (*s == '\n')
+			fiq_debugger_putc(state, '\r');
+		fiq_debugger_putc(state, *s++);
+	}
+	fiq_debugger_uart_flush(state);
+	spin_unlock_irqrestore(&state->console_lock, flags);
+	fiq_debugger_uart_disable(state);
+}
+
+static struct console fiq_debugger_console = {
+	.name = "ttyFIQ",
+	.device = fiq_debugger_console_device,
+	.write = fiq_debugger_console_write,
+	.flags = CON_PRINTBUFFER | CON_ANYTIME | CON_ENABLED,
+};
+
+int fiq_tty_open(struct tty_struct *tty, struct file *filp)
+{
+	int line = tty->index;
+	struct fiq_debugger_state **states = tty->driver->driver_state;
+	struct fiq_debugger_state *state = states[line];
+
+	return tty_port_open(&state->tty_port, tty, filp);
+}
+
+void fiq_tty_close(struct tty_struct *tty, struct file *filp)
+{
+	tty_port_close(tty->port, tty, filp);
+}
+
+int  fiq_tty_write(struct tty_struct *tty, const unsigned char *buf, int count)
+{
+	int i;
+	int line = tty->index;
+	struct fiq_debugger_state **states = tty->driver->driver_state;
+	struct fiq_debugger_state *state = states[line];
+
+	if (!state->console_enable)
+		return count;
+
+	fiq_debugger_uart_enable(state);
+	spin_lock_irq(&state->console_lock);
+	for (i = 0; i < count; i++)
+		fiq_debugger_putc(state, *buf++);
+	spin_unlock_irq(&state->console_lock);
+	fiq_debugger_uart_disable(state);
+
+	return count;
+}
+
+int  fiq_tty_write_room(struct tty_struct *tty)
+{
+	return 16;
+}
+
+#ifdef CONFIG_CONSOLE_POLL
+static int fiq_tty_poll_init(struct tty_driver *driver, int line, char *options)
+{
+	return 0;
+}
+
+static int fiq_tty_poll_get_char(struct tty_driver *driver, int line)
+{
+	struct fiq_debugger_state **states = driver->driver_state;
+	struct fiq_debugger_state *state = states[line];
+	int c = NO_POLL_CHAR;
+
+	fiq_debugger_uart_enable(state);
+	if (fiq_debugger_have_fiq(state)) {
+		int count = fiq_debugger_ringbuf_level(state->tty_rbuf);
+		if (count > 0) {
+			c = fiq_debugger_ringbuf_peek(state->tty_rbuf, 0);
+			fiq_debugger_ringbuf_consume(state->tty_rbuf, 1);
+		}
+	} else {
+		c = fiq_debugger_getc(state);
+		if (c == FIQ_DEBUGGER_NO_CHAR)
+			c = NO_POLL_CHAR;
+	}
+	fiq_debugger_uart_disable(state);
+
+	return c;
+}
+
+static void fiq_tty_poll_put_char(struct tty_driver *driver, int line, char ch)
+{
+	struct fiq_debugger_state **states = driver->driver_state;
+	struct fiq_debugger_state *state = states[line];
+	fiq_debugger_uart_enable(state);
+	fiq_debugger_putc(state, ch);
+	fiq_debugger_uart_disable(state);
+}
+#endif
+
+static const struct tty_port_operations fiq_tty_port_ops;
+
+static const struct tty_operations fiq_tty_driver_ops = {
+	.write = fiq_tty_write,
+	.write_room = fiq_tty_write_room,
+	.open = fiq_tty_open,
+	.close = fiq_tty_close,
+#ifdef CONFIG_CONSOLE_POLL
+	.poll_init = fiq_tty_poll_init,
+	.poll_get_char = fiq_tty_poll_get_char,
+	.poll_put_char = fiq_tty_poll_put_char,
+#endif
+};
+
+static int fiq_debugger_tty_init(void)
+{
+	int ret;
+	struct fiq_debugger_state **states = NULL;
+
+	states = kzalloc(sizeof(*states) * MAX_FIQ_DEBUGGER_PORTS, GFP_KERNEL);
+	if (!states) {
+		pr_err("Failed to allocate fiq debugger state structres\n");
+		return -ENOMEM;
+	}
+
+	fiq_tty_driver = alloc_tty_driver(MAX_FIQ_DEBUGGER_PORTS);
+	if (!fiq_tty_driver) {
+		pr_err("Failed to allocate fiq debugger tty\n");
+		ret = -ENOMEM;
+		goto err_free_state;
+	}
+
+	fiq_tty_driver->owner		= THIS_MODULE;
+	fiq_tty_driver->driver_name	= "fiq-debugger";
+	fiq_tty_driver->name		= "ttyFIQ";
+	fiq_tty_driver->type		= TTY_DRIVER_TYPE_SERIAL;
+	fiq_tty_driver->subtype		= SERIAL_TYPE_NORMAL;
+	fiq_tty_driver->init_termios	= tty_std_termios;
+	fiq_tty_driver->flags		= TTY_DRIVER_REAL_RAW |
+					  TTY_DRIVER_DYNAMIC_DEV;
+	fiq_tty_driver->driver_state	= states;
+
+	fiq_tty_driver->init_termios.c_cflag =
+					B115200 | CS8 | CREAD | HUPCL | CLOCAL;
+	fiq_tty_driver->init_termios.c_ispeed = 115200;
+	fiq_tty_driver->init_termios.c_ospeed = 115200;
+
+	tty_set_operations(fiq_tty_driver, &fiq_tty_driver_ops);
+
+	ret = tty_register_driver(fiq_tty_driver);
+	if (ret) {
+		pr_err("Failed to register fiq tty: %d\n", ret);
+		goto err_free_tty;
+	}
+
+	pr_info("Registered FIQ tty driver\n");
+	return 0;
+
+err_free_tty:
+	put_tty_driver(fiq_tty_driver);
+	fiq_tty_driver = NULL;
+err_free_state:
+	kfree(states);
+	return ret;
+}
+
+static int fiq_debugger_tty_init_one(struct fiq_debugger_state *state)
+{
+	int ret;
+	struct device *tty_dev;
+	struct fiq_debugger_state **states = fiq_tty_driver->driver_state;
+
+	states[state->pdev->id] = state;
+
+	state->tty_rbuf = fiq_debugger_ringbuf_alloc(1024);
+	if (!state->tty_rbuf) {
+		pr_err("Failed to allocate fiq debugger ringbuf\n");
+		ret = -ENOMEM;
+		goto err;
+	}
+
+	tty_port_init(&state->tty_port);
+	state->tty_port.ops = &fiq_tty_port_ops;
+
+	tty_dev = tty_port_register_device(&state->tty_port, fiq_tty_driver,
+					   state->pdev->id, &state->pdev->dev);
+	if (IS_ERR(tty_dev)) {
+		pr_err("Failed to register fiq debugger tty device\n");
+		ret = PTR_ERR(tty_dev);
+		goto err;
+	}
+
+	device_set_wakeup_capable(tty_dev, 1);
+
+	pr_info("Registered fiq debugger ttyFIQ%d\n", state->pdev->id);
+
+	return 0;
+
+err:
+	fiq_debugger_ringbuf_free(state->tty_rbuf);
+	state->tty_rbuf = NULL;
+	return ret;
+}
+#endif
+
+static int fiq_debugger_dev_suspend(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct fiq_debugger_state *state = platform_get_drvdata(pdev);
+
+	if (state->pdata->uart_dev_suspend)
+		return state->pdata->uart_dev_suspend(pdev);
+	return 0;
+}
+
+static int fiq_debugger_dev_resume(struct device *dev)
+{
+	struct platform_device *pdev = to_platform_device(dev);
+	struct fiq_debugger_state *state = platform_get_drvdata(pdev);
+
+	if (state->pdata->uart_dev_resume)
+		return state->pdata->uart_dev_resume(pdev);
+	return 0;
+}
+
+static int fiq_debugger_probe(struct platform_device *pdev)
+{
+	int ret;
+	struct fiq_debugger_pdata *pdata = dev_get_platdata(&pdev->dev);
+	struct fiq_debugger_state *state;
+	int fiq;
+	int uart_irq;
+
+	if (pdev->id >= MAX_FIQ_DEBUGGER_PORTS)
+		return -EINVAL;
+
+	if (!pdata->uart_getc || !pdata->uart_putc)
+		return -EINVAL;
+	if ((pdata->uart_enable && !pdata->uart_disable) ||
+	    (!pdata->uart_enable && pdata->uart_disable))
+		return -EINVAL;
+
+	fiq = platform_get_irq_byname(pdev, "fiq");
+	uart_irq = platform_get_irq_byname(pdev, "uart_irq");
+
+	/* uart_irq mode and fiq mode are mutually exclusive, but one of them
+	 * is required */
+	if ((uart_irq < 0 && fiq < 0) || (uart_irq >= 0 && fiq >= 0))
+		return -EINVAL;
+	if (fiq >= 0 && !pdata->fiq_enable)
+		return -EINVAL;
+
+	state = kzalloc(sizeof(*state), GFP_KERNEL);
+	state->output.printf = fiq_debugger_printf;
+	setup_timer(&state->sleep_timer, fiq_debugger_sleep_timer_expired,
+		    (unsigned long)state);
+	state->pdata = pdata;
+	state->pdev = pdev;
+	state->no_sleep = initial_no_sleep;
+	state->debug_enable = initial_debug_enable;
+	state->console_enable = initial_console_enable;
+
+	state->fiq = fiq;
+	state->uart_irq = uart_irq;
+	state->signal_irq = platform_get_irq_byname(pdev, "signal");
+	state->wakeup_irq = platform_get_irq_byname(pdev, "wakeup");
+
+	INIT_WORK(&state->work, fiq_debugger_work);
+	spin_lock_init(&state->work_lock);
+
+	platform_set_drvdata(pdev, state);
+
+	spin_lock_init(&state->sleep_timer_lock);
+
+	if (state->wakeup_irq < 0 && fiq_debugger_have_fiq(state))
+		state->no_sleep = true;
+	state->ignore_next_wakeup_irq = !state->no_sleep;
+
+	wake_lock_init(&state->debugger_wake_lock,
+			WAKE_LOCK_SUSPEND, "serial-debug");
+
+	state->clk = clk_get(&pdev->dev, NULL);
+	if (IS_ERR(state->clk))
+		state->clk = NULL;
+
+	/* do not call pdata->uart_enable here since uart_init may still
+	 * need to do some initialization before uart_enable can work.
+	 * So, only try to manage the clock during init.
+	 */
+	if (state->clk)
+		clk_enable(state->clk);
+
+	if (pdata->uart_init) {
+		ret = pdata->uart_init(pdev);
+		if (ret)
+			goto err_uart_init;
+	}
+
+	fiq_debugger_printf_nfiq(state,
+				"<hit enter %sto activate fiq debugger>\n",
+				state->no_sleep ? "" : "twice ");
+
+#ifdef CONFIG_FIQ_GLUE
+	if (fiq_debugger_have_fiq(state)) {
+		state->handler.fiq = fiq_debugger_fiq;
+		state->handler.resume = fiq_debugger_resume;
+		ret = fiq_glue_register_handler(&state->handler);
+		if (ret) {
+			pr_err("%s: could not install fiq handler\n", __func__);
+			goto err_register_irq;
+		}
+
+		pdata->fiq_enable(pdev, state->fiq, 1);
+	} else
+#endif
+	{
+		ret = request_irq(state->uart_irq, fiq_debugger_uart_irq,
+				  IRQF_NO_SUSPEND, "debug", state);
+		if (ret) {
+			pr_err("%s: could not install irq handler\n", __func__);
+			goto err_register_irq;
+		}
+
+		/* for irq-only mode, we want this irq to wake us up, if it
+		 * can.
+		 */
+		enable_irq_wake(state->uart_irq);
+	}
+
+	if (state->clk)
+		clk_disable(state->clk);
+
+	if (state->signal_irq >= 0) {
+		ret = request_irq(state->signal_irq, fiq_debugger_signal_irq,
+			  IRQF_TRIGGER_RISING, "debug-signal", state);
+		if (ret)
+			pr_err("serial_debugger: could not install signal_irq");
+	}
+
+	if (state->wakeup_irq >= 0) {
+		ret = request_irq(state->wakeup_irq,
+				  fiq_debugger_wakeup_irq_handler,
+				  IRQF_TRIGGER_FALLING,
+				  "debug-wakeup", state);
+		if (ret) {
+			pr_err("serial_debugger: "
+				"could not install wakeup irq\n");
+			state->wakeup_irq = -1;
+		} else {
+			ret = enable_irq_wake(state->wakeup_irq);
+			if (ret) {
+				pr_err("serial_debugger: "
+					"could not enable wakeup\n");
+				state->wakeup_irq_no_set_wake = true;
+			}
+		}
+	}
+	if (state->no_sleep)
+		fiq_debugger_handle_wakeup(state);
+
+#if defined(CONFIG_FIQ_DEBUGGER_CONSOLE)
+	spin_lock_init(&state->console_lock);
+	state->console = fiq_debugger_console;
+	state->console.index = pdev->id;
+	if (!console_set_on_cmdline)
+		add_preferred_console(state->console.name,
+			state->console.index, NULL);
+	register_console(&state->console);
+	fiq_debugger_tty_init_one(state);
+#endif
+	return 0;
+
+err_register_irq:
+	if (pdata->uart_free)
+		pdata->uart_free(pdev);
+err_uart_init:
+	if (state->clk)
+		clk_disable(state->clk);
+	if (state->clk)
+		clk_put(state->clk);
+	wake_lock_destroy(&state->debugger_wake_lock);
+	platform_set_drvdata(pdev, NULL);
+	kfree(state);
+	return ret;
+}
+
+static const struct dev_pm_ops fiq_debugger_dev_pm_ops = {
+	.suspend	= fiq_debugger_dev_suspend,
+	.resume		= fiq_debugger_dev_resume,
+};
+
+static struct platform_driver fiq_debugger_driver = {
+	.probe	= fiq_debugger_probe,
+	.driver	= {
+		.name	= "fiq_debugger",
+		.pm	= &fiq_debugger_dev_pm_ops,
+	},
+};
+
+#if defined(CONFIG_FIQ_DEBUGGER_UART_OVERLAY)
+int fiq_debugger_uart_overlay(void)
+{
+	struct device_node *onp = of_find_node_by_path("/uart_overlay@0");
+	int ret;
+
+	if (!onp) {
+		pr_err("serial_debugger: uart overlay not found\n");
+		return -ENODEV;
+	}
+
+	ret = of_overlay_create(onp);
+	if (ret < 0) {
+		pr_err("serial_debugger: fail to create overlay: %d\n", ret);
+		of_node_put(onp);
+		return ret;
+	}
+
+	pr_info("serial_debugger: uart overlay applied\n");
+	return 0;
+}
+#endif
+
+static int __init fiq_debugger_init(void)
+{
+	if (fiq_debugger_disable) {
+		pr_err("serial_debugger: disabled\n");
+		return -ENODEV;
+	}
+#if defined(CONFIG_FIQ_DEBUGGER_CONSOLE)
+	fiq_debugger_tty_init();
+#endif
+#if defined(CONFIG_FIQ_DEBUGGER_UART_OVERLAY)
+	fiq_debugger_uart_overlay();
+#endif
+	return platform_driver_register(&fiq_debugger_driver);
+}
+
+postcore_initcall(fiq_debugger_init);
diff --git a/drivers/staging/android/fiq_debugger/fiq_debugger.h b/drivers/staging/android/fiq_debugger/fiq_debugger.h
new file mode 100644
index 000000000000..c9ec4f8db086
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_debugger.h
@@ -0,0 +1,64 @@
+/*
+ * drivers/staging/android/fiq_debugger/fiq_debugger.h
+ *
+ * Copyright (C) 2010 Google, Inc.
+ * Author: Colin Cross <ccross@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _ARCH_ARM_MACH_TEGRA_FIQ_DEBUGGER_H_
+#define _ARCH_ARM_MACH_TEGRA_FIQ_DEBUGGER_H_
+
+#include <linux/serial_core.h>
+
+#define FIQ_DEBUGGER_NO_CHAR NO_POLL_CHAR
+#define FIQ_DEBUGGER_BREAK 0x00ff0100
+
+#define FIQ_DEBUGGER_FIQ_IRQ_NAME	"fiq"
+#define FIQ_DEBUGGER_SIGNAL_IRQ_NAME	"signal"
+#define FIQ_DEBUGGER_WAKEUP_IRQ_NAME	"wakeup"
+
+/**
+ * struct fiq_debugger_pdata - fiq debugger platform data
+ * @uart_resume:	used to restore uart state right before enabling
+ *			the fiq.
+ * @uart_enable:	Do the work necessary to communicate with the uart
+ *			hw (enable clocks, etc.). This must be ref-counted.
+ * @uart_disable:	Do the work necessary to disable the uart hw
+ *			(disable clocks, etc.). This must be ref-counted.
+ * @uart_dev_suspend:	called during PM suspend, generally not needed
+ *			for real fiq mode debugger.
+ * @uart_dev_resume:	called during PM resume, generally not needed
+ *			for real fiq mode debugger.
+ */
+struct fiq_debugger_pdata {
+	int (*uart_init)(struct platform_device *pdev);
+	void (*uart_free)(struct platform_device *pdev);
+	int (*uart_resume)(struct platform_device *pdev);
+	int (*uart_getc)(struct platform_device *pdev);
+	void (*uart_putc)(struct platform_device *pdev, unsigned int c);
+	void (*uart_flush)(struct platform_device *pdev);
+	void (*uart_enable)(struct platform_device *pdev);
+	void (*uart_disable)(struct platform_device *pdev);
+
+	int (*uart_dev_suspend)(struct platform_device *pdev);
+	int (*uart_dev_resume)(struct platform_device *pdev);
+
+	void (*fiq_enable)(struct platform_device *pdev, unsigned int fiq,
+								bool enable);
+	void (*fiq_ack)(struct platform_device *pdev, unsigned int fiq);
+
+	void (*force_irq)(struct platform_device *pdev, unsigned int irq);
+	void (*force_irq_ack)(struct platform_device *pdev, unsigned int irq);
+};
+
+#endif
diff --git a/drivers/staging/android/fiq_debugger/fiq_debugger_arm.c b/drivers/staging/android/fiq_debugger/fiq_debugger_arm.c
new file mode 100644
index 000000000000..8b3e0137be1a
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_debugger_arm.c
@@ -0,0 +1,240 @@
+/*
+ * Copyright (C) 2014 Google, Inc.
+ * Author: Colin Cross <ccross@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/ptrace.h>
+#include <linux/uaccess.h>
+
+#include <asm/stacktrace.h>
+
+#include "fiq_debugger_priv.h"
+
+static char *mode_name(unsigned cpsr)
+{
+	switch (cpsr & MODE_MASK) {
+	case USR_MODE: return "USR";
+	case FIQ_MODE: return "FIQ";
+	case IRQ_MODE: return "IRQ";
+	case SVC_MODE: return "SVC";
+	case ABT_MODE: return "ABT";
+	case UND_MODE: return "UND";
+	case SYSTEM_MODE: return "SYS";
+	default: return "???";
+	}
+}
+
+void fiq_debugger_dump_pc(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+	output->printf(output, " pc %08x cpsr %08x mode %s\n",
+		regs->ARM_pc, regs->ARM_cpsr, mode_name(regs->ARM_cpsr));
+}
+
+void fiq_debugger_dump_regs(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+	output->printf(output,
+			" r0 %08x  r1 %08x  r2 %08x  r3 %08x\n",
+			regs->ARM_r0, regs->ARM_r1, regs->ARM_r2, regs->ARM_r3);
+	output->printf(output,
+			" r4 %08x  r5 %08x  r6 %08x  r7 %08x\n",
+			regs->ARM_r4, regs->ARM_r5, regs->ARM_r6, regs->ARM_r7);
+	output->printf(output,
+			" r8 %08x  r9 %08x r10 %08x r11 %08x  mode %s\n",
+			regs->ARM_r8, regs->ARM_r9, regs->ARM_r10, regs->ARM_fp,
+			mode_name(regs->ARM_cpsr));
+	output->printf(output,
+			" ip %08x  sp %08x  lr %08x  pc %08x cpsr %08x\n",
+			regs->ARM_ip, regs->ARM_sp, regs->ARM_lr, regs->ARM_pc,
+			regs->ARM_cpsr);
+}
+
+struct mode_regs {
+	unsigned long sp_svc;
+	unsigned long lr_svc;
+	unsigned long spsr_svc;
+
+	unsigned long sp_abt;
+	unsigned long lr_abt;
+	unsigned long spsr_abt;
+
+	unsigned long sp_und;
+	unsigned long lr_und;
+	unsigned long spsr_und;
+
+	unsigned long sp_irq;
+	unsigned long lr_irq;
+	unsigned long spsr_irq;
+
+	unsigned long r8_fiq;
+	unsigned long r9_fiq;
+	unsigned long r10_fiq;
+	unsigned long r11_fiq;
+	unsigned long r12_fiq;
+	unsigned long sp_fiq;
+	unsigned long lr_fiq;
+	unsigned long spsr_fiq;
+};
+
+static void __naked get_mode_regs(struct mode_regs *regs)
+{
+	asm volatile (
+	"mrs	r1, cpsr\n"
+	"msr	cpsr_c, #0xd3 @(SVC_MODE | PSR_I_BIT | PSR_F_BIT)\n"
+	"stmia	r0!, {r13 - r14}\n"
+	"mrs	r2, spsr\n"
+	"msr	cpsr_c, #0xd7 @(ABT_MODE | PSR_I_BIT | PSR_F_BIT)\n"
+	"stmia	r0!, {r2, r13 - r14}\n"
+	"mrs	r2, spsr\n"
+	"msr	cpsr_c, #0xdb @(UND_MODE | PSR_I_BIT | PSR_F_BIT)\n"
+	"stmia	r0!, {r2, r13 - r14}\n"
+	"mrs	r2, spsr\n"
+	"msr	cpsr_c, #0xd2 @(IRQ_MODE | PSR_I_BIT | PSR_F_BIT)\n"
+	"stmia	r0!, {r2, r13 - r14}\n"
+	"mrs	r2, spsr\n"
+	"msr	cpsr_c, #0xd1 @(FIQ_MODE | PSR_I_BIT | PSR_F_BIT)\n"
+	"stmia	r0!, {r2, r8 - r14}\n"
+	"mrs	r2, spsr\n"
+	"stmia	r0!, {r2}\n"
+	"msr	cpsr_c, r1\n"
+	"bx	lr\n");
+}
+
+
+void fiq_debugger_dump_allregs(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+	struct mode_regs mode_regs;
+	unsigned long mode = regs->ARM_cpsr & MODE_MASK;
+
+	fiq_debugger_dump_regs(output, regs);
+	get_mode_regs(&mode_regs);
+
+	output->printf(output,
+			"%csvc: sp %08x  lr %08x  spsr %08x\n",
+			mode == SVC_MODE ? '*' : ' ',
+			mode_regs.sp_svc, mode_regs.lr_svc, mode_regs.spsr_svc);
+	output->printf(output,
+			"%cabt: sp %08x  lr %08x  spsr %08x\n",
+			mode == ABT_MODE ? '*' : ' ',
+			mode_regs.sp_abt, mode_regs.lr_abt, mode_regs.spsr_abt);
+	output->printf(output,
+			"%cund: sp %08x  lr %08x  spsr %08x\n",
+			mode == UND_MODE ? '*' : ' ',
+			mode_regs.sp_und, mode_regs.lr_und, mode_regs.spsr_und);
+	output->printf(output,
+			"%cirq: sp %08x  lr %08x  spsr %08x\n",
+			mode == IRQ_MODE ? '*' : ' ',
+			mode_regs.sp_irq, mode_regs.lr_irq, mode_regs.spsr_irq);
+	output->printf(output,
+			"%cfiq: r8 %08x  r9 %08x  r10 %08x  r11 %08x  r12 %08x\n",
+			mode == FIQ_MODE ? '*' : ' ',
+			mode_regs.r8_fiq, mode_regs.r9_fiq, mode_regs.r10_fiq,
+			mode_regs.r11_fiq, mode_regs.r12_fiq);
+	output->printf(output,
+			" fiq: sp %08x  lr %08x  spsr %08x\n",
+			mode_regs.sp_fiq, mode_regs.lr_fiq, mode_regs.spsr_fiq);
+}
+
+struct stacktrace_state {
+	struct fiq_debugger_output *output;
+	unsigned int depth;
+};
+
+static int report_trace(struct stackframe *frame, void *d)
+{
+	struct stacktrace_state *sts = d;
+
+	if (sts->depth) {
+		sts->output->printf(sts->output,
+			"  pc: %p (%pF), lr %p (%pF), sp %p, fp %p\n",
+			frame->pc, frame->pc, frame->lr, frame->lr,
+			frame->sp, frame->fp);
+		sts->depth--;
+		return 0;
+	}
+	sts->output->printf(sts->output, "  ...\n");
+
+	return sts->depth == 0;
+}
+
+struct frame_tail {
+	struct frame_tail *fp;
+	unsigned long sp;
+	unsigned long lr;
+} __attribute__((packed));
+
+static struct frame_tail *user_backtrace(struct fiq_debugger_output *output,
+					struct frame_tail *tail)
+{
+	struct frame_tail buftail[2];
+
+	/* Also check accessibility of one struct frame_tail beyond */
+	if (!access_ok(VERIFY_READ, tail, sizeof(buftail))) {
+		output->printf(output, "  invalid frame pointer %p\n",
+				tail);
+		return NULL;
+	}
+	if (__copy_from_user_inatomic(buftail, tail, sizeof(buftail))) {
+		output->printf(output,
+			"  failed to copy frame pointer %p\n", tail);
+		return NULL;
+	}
+
+	output->printf(output, "  %p\n", buftail[0].lr);
+
+	/* frame pointers should strictly progress back up the stack
+	 * (towards higher addresses) */
+	if (tail >= buftail[0].fp)
+		return NULL;
+
+	return buftail[0].fp-1;
+}
+
+void fiq_debugger_dump_stacktrace(struct fiq_debugger_output *output,
+		const struct pt_regs *regs, unsigned int depth, void *ssp)
+{
+	struct frame_tail *tail;
+	struct thread_info *real_thread_info = THREAD_INFO(ssp);
+	struct stacktrace_state sts;
+
+	sts.depth = depth;
+	sts.output = output;
+	*current_thread_info() = *real_thread_info;
+
+	if (!current)
+		output->printf(output, "current NULL\n");
+	else
+		output->printf(output, "pid: %d  comm: %s\n",
+			current->pid, current->comm);
+	fiq_debugger_dump_regs(output, regs);
+
+	if (!user_mode(regs)) {
+		struct stackframe frame;
+		frame.fp = regs->ARM_fp;
+		frame.sp = regs->ARM_sp;
+		frame.lr = regs->ARM_lr;
+		frame.pc = regs->ARM_pc;
+		output->printf(output,
+			"  pc: %p (%pF), lr %p (%pF), sp %p, fp %p\n",
+			regs->ARM_pc, regs->ARM_pc, regs->ARM_lr, regs->ARM_lr,
+			regs->ARM_sp, regs->ARM_fp);
+		walk_stackframe(&frame, report_trace, &sts);
+		return;
+	}
+
+	tail = ((struct frame_tail *) regs->ARM_fp) - 1;
+	while (depth-- && tail && !((unsigned long) tail & 3))
+		tail = user_backtrace(output, tail);
+}
diff --git a/drivers/staging/android/fiq_debugger/fiq_debugger_arm64.c b/drivers/staging/android/fiq_debugger/fiq_debugger_arm64.c
new file mode 100644
index 000000000000..97246bcbcd62
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_debugger_arm64.c
@@ -0,0 +1,202 @@
+/*
+ * Copyright (C) 2014 Google, Inc.
+ * Author: Colin Cross <ccross@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/ptrace.h>
+#include <asm/stacktrace.h>
+
+#include "fiq_debugger_priv.h"
+
+static char *mode_name(const struct pt_regs *regs)
+{
+	if (compat_user_mode(regs)) {
+		return "USR";
+	} else {
+		switch (processor_mode(regs)) {
+		case PSR_MODE_EL0t: return "EL0t";
+		case PSR_MODE_EL1t: return "EL1t";
+		case PSR_MODE_EL1h: return "EL1h";
+		case PSR_MODE_EL2t: return "EL2t";
+		case PSR_MODE_EL2h: return "EL2h";
+		default: return "???";
+		}
+	}
+}
+
+void fiq_debugger_dump_pc(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+	output->printf(output, " pc %016lx cpsr %08lx mode %s\n",
+		regs->pc, regs->pstate, mode_name(regs));
+}
+
+void fiq_debugger_dump_regs_aarch32(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+	output->printf(output, " r0 %08x  r1 %08x  r2 %08x  r3 %08x\n",
+			regs->compat_usr(0), regs->compat_usr(1),
+			regs->compat_usr(2), regs->compat_usr(3));
+	output->printf(output, " r4 %08x  r5 %08x  r6 %08x  r7 %08x\n",
+			regs->compat_usr(4), regs->compat_usr(5),
+			regs->compat_usr(6), regs->compat_usr(7));
+	output->printf(output, " r8 %08x  r9 %08x r10 %08x r11 %08x\n",
+			regs->compat_usr(8), regs->compat_usr(9),
+			regs->compat_usr(10), regs->compat_usr(11));
+	output->printf(output, " ip %08x  sp %08x  lr %08x  pc %08x\n",
+			regs->compat_usr(12), regs->compat_sp,
+			regs->compat_lr, regs->pc);
+	output->printf(output, " cpsr %08x (%s)\n",
+			regs->pstate, mode_name(regs));
+}
+
+void fiq_debugger_dump_regs_aarch64(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+
+	output->printf(output, "  x0 %016lx   x1 %016lx\n",
+			regs->regs[0], regs->regs[1]);
+	output->printf(output, "  x2 %016lx   x3 %016lx\n",
+			regs->regs[2], regs->regs[3]);
+	output->printf(output, "  x4 %016lx   x5 %016lx\n",
+			regs->regs[4], regs->regs[5]);
+	output->printf(output, "  x6 %016lx   x7 %016lx\n",
+			regs->regs[6], regs->regs[7]);
+	output->printf(output, "  x8 %016lx   x9 %016lx\n",
+			regs->regs[8], regs->regs[9]);
+	output->printf(output, " x10 %016lx  x11 %016lx\n",
+			regs->regs[10], regs->regs[11]);
+	output->printf(output, " x12 %016lx  x13 %016lx\n",
+			regs->regs[12], regs->regs[13]);
+	output->printf(output, " x14 %016lx  x15 %016lx\n",
+			regs->regs[14], regs->regs[15]);
+	output->printf(output, " x16 %016lx  x17 %016lx\n",
+			regs->regs[16], regs->regs[17]);
+	output->printf(output, " x18 %016lx  x19 %016lx\n",
+			regs->regs[18], regs->regs[19]);
+	output->printf(output, " x20 %016lx  x21 %016lx\n",
+			regs->regs[20], regs->regs[21]);
+	output->printf(output, " x22 %016lx  x23 %016lx\n",
+			regs->regs[22], regs->regs[23]);
+	output->printf(output, " x24 %016lx  x25 %016lx\n",
+			regs->regs[24], regs->regs[25]);
+	output->printf(output, " x26 %016lx  x27 %016lx\n",
+			regs->regs[26], regs->regs[27]);
+	output->printf(output, " x28 %016lx  x29 %016lx\n",
+			regs->regs[28], regs->regs[29]);
+	output->printf(output, " x30 %016lx   sp %016lx\n",
+			regs->regs[30], regs->sp);
+	output->printf(output, "  pc %016lx cpsr %08x (%s)\n",
+			regs->pc, regs->pstate, mode_name(regs));
+}
+
+void fiq_debugger_dump_regs(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+	if (compat_user_mode(regs))
+		fiq_debugger_dump_regs_aarch32(output, regs);
+	else
+		fiq_debugger_dump_regs_aarch64(output, regs);
+}
+
+#define READ_SPECIAL_REG(x) ({ \
+	u64 val; \
+	asm volatile ("mrs %0, " # x : "=r"(val)); \
+	val; \
+})
+
+void fiq_debugger_dump_allregs(struct fiq_debugger_output *output,
+		const struct pt_regs *regs)
+{
+	u32 pstate = READ_SPECIAL_REG(CurrentEl);
+	bool in_el2 = (pstate & PSR_MODE_MASK) >= PSR_MODE_EL2t;
+
+	fiq_debugger_dump_regs(output, regs);
+
+	output->printf(output, " sp_el0   %016lx\n",
+			READ_SPECIAL_REG(sp_el0));
+
+	if (in_el2)
+		output->printf(output, " sp_el1   %016lx\n",
+				READ_SPECIAL_REG(sp_el1));
+
+	output->printf(output, " elr_el1  %016lx\n",
+			READ_SPECIAL_REG(elr_el1));
+
+	output->printf(output, " spsr_el1 %08lx\n",
+			READ_SPECIAL_REG(spsr_el1));
+
+	if (in_el2) {
+		output->printf(output, " spsr_irq %08lx\n",
+				READ_SPECIAL_REG(spsr_irq));
+		output->printf(output, " spsr_abt %08lx\n",
+				READ_SPECIAL_REG(spsr_abt));
+		output->printf(output, " spsr_und %08lx\n",
+				READ_SPECIAL_REG(spsr_und));
+		output->printf(output, " spsr_fiq %08lx\n",
+				READ_SPECIAL_REG(spsr_fiq));
+		output->printf(output, " spsr_el2 %08lx\n",
+				READ_SPECIAL_REG(elr_el2));
+		output->printf(output, " spsr_el2 %08lx\n",
+				READ_SPECIAL_REG(spsr_el2));
+	}
+}
+
+struct stacktrace_state {
+	struct fiq_debugger_output *output;
+	unsigned int depth;
+};
+
+static int report_trace(struct stackframe *frame, void *d)
+{
+	struct stacktrace_state *sts = d;
+
+	if (sts->depth) {
+		sts->output->printf(sts->output, "%pF:\n", frame->pc);
+		sts->output->printf(sts->output,
+				"  pc %016lx   sp %016lx   fp %016lx\n",
+				frame->pc, frame->sp, frame->fp);
+		sts->depth--;
+		return 0;
+	}
+	sts->output->printf(sts->output, "  ...\n");
+
+	return sts->depth == 0;
+}
+
+void fiq_debugger_dump_stacktrace(struct fiq_debugger_output *output,
+		const struct pt_regs *regs, unsigned int depth, void *ssp)
+{
+	struct thread_info *real_thread_info = THREAD_INFO(ssp);
+	struct stacktrace_state sts;
+
+	sts.depth = depth;
+	sts.output = output;
+	*current_thread_info() = *real_thread_info;
+
+	if (!current)
+		output->printf(output, "current NULL\n");
+	else
+		output->printf(output, "pid: %d  comm: %s\n",
+			current->pid, current->comm);
+	fiq_debugger_dump_regs(output, regs);
+
+	if (!user_mode(regs)) {
+		struct stackframe frame;
+		frame.fp = regs->regs[29];
+		frame.sp = regs->sp;
+		frame.pc = regs->pc;
+		output->printf(output, "\n");
+		walk_stackframe(current, &frame, report_trace, &sts);
+	}
+}
diff --git a/drivers/staging/android/fiq_debugger/fiq_debugger_priv.h b/drivers/staging/android/fiq_debugger/fiq_debugger_priv.h
new file mode 100644
index 000000000000..d5d051f727a8
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_debugger_priv.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright (C) 2014 Google, Inc.
+ * Author: Colin Cross <ccross@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _FIQ_DEBUGGER_PRIV_H_
+#define _FIQ_DEBUGGER_PRIV_H_
+
+#define THREAD_INFO(sp) ((struct thread_info *) \
+		((unsigned long)(sp) & ~(THREAD_SIZE - 1)))
+
+struct fiq_debugger_output {
+	void (*printf)(struct fiq_debugger_output *output, const char *fmt, ...);
+};
+
+struct pt_regs;
+
+void fiq_debugger_dump_pc(struct fiq_debugger_output *output,
+		const struct pt_regs *regs);
+void fiq_debugger_dump_regs(struct fiq_debugger_output *output,
+		const struct pt_regs *regs);
+void fiq_debugger_dump_allregs(struct fiq_debugger_output *output,
+		const struct pt_regs *regs);
+void fiq_debugger_dump_stacktrace(struct fiq_debugger_output *output,
+		const struct pt_regs *regs, unsigned int depth, void *ssp);
+
+#endif
diff --git a/drivers/staging/android/fiq_debugger/fiq_debugger_ringbuf.h b/drivers/staging/android/fiq_debugger/fiq_debugger_ringbuf.h
new file mode 100644
index 000000000000..10c3c5d09098
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_debugger_ringbuf.h
@@ -0,0 +1,94 @@
+/*
+ * drivers/staging/android/fiq_debugger/fiq_debugger_ringbuf.h
+ *
+ * simple lockless ringbuffer
+ *
+ * Copyright (C) 2010 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+
+struct fiq_debugger_ringbuf {
+	int len;
+	int head;
+	int tail;
+	u8 buf[];
+};
+
+
+static inline struct fiq_debugger_ringbuf *fiq_debugger_ringbuf_alloc(int len)
+{
+	struct fiq_debugger_ringbuf *rbuf;
+
+	rbuf = kzalloc(sizeof(*rbuf) + len, GFP_KERNEL);
+	if (rbuf == NULL)
+		return NULL;
+
+	rbuf->len = len;
+	rbuf->head = 0;
+	rbuf->tail = 0;
+	smp_mb();
+
+	return rbuf;
+}
+
+static inline void fiq_debugger_ringbuf_free(struct fiq_debugger_ringbuf *rbuf)
+{
+	kfree(rbuf);
+}
+
+static inline int fiq_debugger_ringbuf_level(struct fiq_debugger_ringbuf *rbuf)
+{
+	int level = rbuf->head - rbuf->tail;
+
+	if (level < 0)
+		level = rbuf->len + level;
+
+	return level;
+}
+
+static inline int fiq_debugger_ringbuf_room(struct fiq_debugger_ringbuf *rbuf)
+{
+	return rbuf->len - fiq_debugger_ringbuf_level(rbuf) - 1;
+}
+
+static inline u8
+fiq_debugger_ringbuf_peek(struct fiq_debugger_ringbuf *rbuf, int i)
+{
+	return rbuf->buf[(rbuf->tail + i) % rbuf->len];
+}
+
+static inline int
+fiq_debugger_ringbuf_consume(struct fiq_debugger_ringbuf *rbuf, int count)
+{
+	count = min(count, fiq_debugger_ringbuf_level(rbuf));
+
+	rbuf->tail = (rbuf->tail + count) % rbuf->len;
+	smp_mb();
+
+	return count;
+}
+
+static inline int
+fiq_debugger_ringbuf_push(struct fiq_debugger_ringbuf *rbuf, u8 datum)
+{
+	if (fiq_debugger_ringbuf_room(rbuf) == 0)
+		return 0;
+
+	rbuf->buf[rbuf->head] = datum;
+	smp_mb();
+	rbuf->head = (rbuf->head + 1) % rbuf->len;
+	smp_mb();
+
+	return 1;
+}
diff --git a/drivers/staging/android/fiq_debugger/fiq_watchdog.c b/drivers/staging/android/fiq_debugger/fiq_watchdog.c
new file mode 100644
index 000000000000..194b54138417
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_watchdog.c
@@ -0,0 +1,56 @@
+/*
+ * Copyright (C) 2014 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/spinlock.h>
+#include <linux/pstore_ram.h>
+
+#include "fiq_watchdog.h"
+#include "fiq_debugger_priv.h"
+
+static DEFINE_RAW_SPINLOCK(fiq_watchdog_lock);
+
+static void fiq_watchdog_printf(struct fiq_debugger_output *output,
+				const char *fmt, ...)
+{
+	char buf[256];
+	va_list ap;
+	int len;
+
+	va_start(ap, fmt);
+	len = vscnprintf(buf, sizeof(buf), fmt, ap);
+	va_end(ap);
+
+	ramoops_console_write_buf(buf, len);
+}
+
+struct fiq_debugger_output fiq_watchdog_output = {
+	.printf = fiq_watchdog_printf,
+};
+
+void fiq_watchdog_triggered(const struct pt_regs *regs, void *svc_sp)
+{
+	char msg[24];
+	int len;
+
+	raw_spin_lock(&fiq_watchdog_lock);
+
+	len = scnprintf(msg, sizeof(msg), "watchdog fiq cpu %d\n",
+			THREAD_INFO(svc_sp)->cpu);
+	ramoops_console_write_buf(msg, len);
+
+	fiq_debugger_dump_stacktrace(&fiq_watchdog_output, regs, 100, svc_sp);
+
+	raw_spin_unlock(&fiq_watchdog_lock);
+}
diff --git a/drivers/staging/android/fiq_debugger/fiq_watchdog.h b/drivers/staging/android/fiq_debugger/fiq_watchdog.h
new file mode 100644
index 000000000000..c6b507f8d976
--- /dev/null
+++ b/drivers/staging/android/fiq_debugger/fiq_watchdog.h
@@ -0,0 +1,20 @@
+/*
+ * Copyright (C) 2014 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _FIQ_WATCHDOG_H_
+#define _FIQ_WATCHDOG_H_
+
+void fiq_watchdog_triggered(const struct pt_regs *regs, void *svc_sp);
+
+#endif
diff --git a/drivers/staging/android/ion/Kconfig b/drivers/staging/android/ion/Kconfig
index 345234624492..356e10969272 100644
--- a/drivers/staging/android/ion/Kconfig
+++ b/drivers/staging/android/ion/Kconfig
@@ -33,3 +33,10 @@ config ION_TEGRA
 	help
 	  Choose this option if you wish to use ion on an nVidia Tegra.
 
+config ION_POOL_CACHE_POLICY
+	bool "Ion set page pool cache policy"
+	depends on ION && X86
+	default y if X86
+	help
+	  Choose this option if need to explicity set cache policy of the
+	  pages in the page pool.
diff --git a/drivers/staging/android/ion/ion_carveout_heap.c b/drivers/staging/android/ion/ion_carveout_heap.c
index 9156d8238c97..e702ce6461fc 100644
--- a/drivers/staging/android/ion/ion_carveout_heap.c
+++ b/drivers/staging/android/ion/ion_carveout_heap.c
@@ -167,7 +167,7 @@ struct ion_heap *ion_carveout_heap_create(struct ion_platform_heap *heap_data)
 	if (!carveout_heap)
 		return ERR_PTR(-ENOMEM);
 
-	carveout_heap->pool = gen_pool_create(12, -1);
+	carveout_heap->pool = gen_pool_create(PAGE_SHIFT, -1);
 	if (!carveout_heap->pool) {
 		kfree(carveout_heap);
 		return ERR_PTR(-ENOMEM);
diff --git a/drivers/staging/android/ion/ion_page_pool.c b/drivers/staging/android/ion/ion_page_pool.c
index fd7e23e0c06e..59ee2f8f6761 100644
--- a/drivers/staging/android/ion/ion_page_pool.c
+++ b/drivers/staging/android/ion/ion_page_pool.c
@@ -30,6 +30,8 @@ static void *ion_page_pool_alloc_pages(struct ion_page_pool *pool)
 
 	if (!page)
 		return NULL;
+	ion_page_pool_alloc_set_cache_policy(pool, page);
+
 	ion_pages_sync_for_device(NULL, page, PAGE_SIZE << pool->order,
 						DMA_BIDIRECTIONAL);
 	return page;
@@ -38,6 +40,7 @@ static void *ion_page_pool_alloc_pages(struct ion_page_pool *pool)
 static void ion_page_pool_free_pages(struct ion_page_pool *pool,
 				     struct page *page)
 {
+	ion_page_pool_free_set_cache_policy(pool, page);
 	__free_pages(page, pool->order);
 }
 
@@ -103,6 +106,11 @@ void ion_page_pool_free(struct ion_page_pool *pool, struct page *page)
 		ion_page_pool_free_pages(pool, page);
 }
 
+void ion_page_pool_free_immediate(struct ion_page_pool *pool, struct page *page)
+{
+	ion_page_pool_free_pages(pool, page);
+}
+
 static int ion_page_pool_total(struct ion_page_pool *pool, bool high)
 {
 	int count = pool->low_count;
diff --git a/drivers/staging/android/ion/ion_priv.h b/drivers/staging/android/ion/ion_priv.h
index 0239883bffb7..6f59a2d36567 100644
--- a/drivers/staging/android/ion/ion_priv.h
+++ b/drivers/staging/android/ion/ion_priv.h
@@ -26,6 +26,9 @@
 #include <linux/sched.h>
 #include <linux/shrinker.h>
 #include <linux/types.h>
+#ifdef CONFIG_ION_POOL_CACHE_POLICY
+#include <asm/cacheflush.h>
+#endif
 
 #include "ion.h"
 
@@ -381,6 +384,37 @@ struct ion_page_pool *ion_page_pool_create(gfp_t gfp_mask, unsigned int order);
 void ion_page_pool_destroy(struct ion_page_pool *);
 struct page *ion_page_pool_alloc(struct ion_page_pool *);
 void ion_page_pool_free(struct ion_page_pool *, struct page *);
+void ion_page_pool_free_immediate(struct ion_page_pool *, struct page *);
+
+#ifdef CONFIG_ION_POOL_CACHE_POLICY
+static inline void ion_page_pool_alloc_set_cache_policy
+				(struct ion_page_pool *pool,
+				struct page *page){
+	void *va = page_address(page);
+
+	if (va)
+		set_memory_wc((unsigned long)va, 1 << pool->order);
+}
+
+static inline void ion_page_pool_free_set_cache_policy
+				(struct ion_page_pool *pool,
+				struct page *page){
+	void *va = page_address(page);
+
+	if (va)
+		set_memory_wb((unsigned long)va, 1 << pool->order);
+
+}
+#else
+static inline void ion_page_pool_alloc_set_cache_policy
+				(struct ion_page_pool *pool,
+				struct page *page){ }
+
+static inline void ion_page_pool_free_set_cache_policy
+				(struct ion_page_pool *pool,
+				struct page *page){ }
+#endif
+
 
 /** ion_page_pool_shrink - shrinks the size of the memory cached in the pool
  * @pool:		the pool
diff --git a/drivers/staging/android/ion/ion_system_heap.c b/drivers/staging/android/ion/ion_system_heap.c
index d4c3e5512dd5..57d115d0f179 100644
--- a/drivers/staging/android/ion/ion_system_heap.c
+++ b/drivers/staging/android/ion/ion_system_heap.c
@@ -83,10 +83,12 @@ static void free_buffer_page(struct ion_system_heap *heap,
 	unsigned int order = compound_order(page);
 	bool cached = ion_buffer_cached(buffer);
 
-	if (!cached && !(buffer->private_flags & ION_PRIV_FLAG_SHRINKER_FREE)) {
+	if (!cached) {
 		struct ion_page_pool *pool = heap->pools[order_to_index(order)];
-
-		ion_page_pool_free(pool, page);
+		if (buffer->private_flags & ION_PRIV_FLAG_SHRINKER_FREE)
+			ion_page_pool_free_immediate(pool, page);
+		else
+			ion_page_pool_free(pool, page);
 	} else {
 		__free_pages(page, order);
 	}
diff --git a/drivers/staging/android/lowmemorykiller.c b/drivers/staging/android/lowmemorykiller.c
index e679d8432810..2f4ef2120d31 100644
--- a/drivers/staging/android/lowmemorykiller.c
+++ b/drivers/staging/android/lowmemorykiller.c
@@ -43,6 +43,9 @@
 #include <linux/profile.h>
 #include <linux/notifier.h>
 
+#define CREATE_TRACE_POINTS
+#include "trace/lowmemorykiller.h"
+
 static uint32_t lowmem_debug_level = 1;
 static short lowmem_adj[6] = {
 	0,
@@ -84,12 +87,14 @@ static unsigned long lowmem_scan(struct shrinker *s, struct shrink_control *sc)
 	int tasksize;
 	int i;
 	short min_score_adj = OOM_SCORE_ADJ_MAX + 1;
+	int minfree = 0;
 	int selected_tasksize = 0;
 	short selected_oom_score_adj;
 	int array_size = ARRAY_SIZE(lowmem_adj);
 	int other_free = global_page_state(NR_FREE_PAGES) - totalreserve_pages;
 	int other_file = global_page_state(NR_FILE_PAGES) -
 						global_page_state(NR_SHMEM) -
+						global_page_state(NR_UNEVICTABLE) -
 						total_swapcache_pages();
 
 	if (lowmem_adj_size < array_size)
@@ -97,8 +102,8 @@ static unsigned long lowmem_scan(struct shrinker *s, struct shrink_control *sc)
 	if (lowmem_minfree_size < array_size)
 		array_size = lowmem_minfree_size;
 	for (i = 0; i < array_size; i++) {
-		if (other_free < lowmem_minfree[i] &&
-		    other_file < lowmem_minfree[i]) {
+		minfree = lowmem_minfree[i];
+		if (other_free < minfree && other_file < minfree) {
 			min_score_adj = lowmem_adj[i];
 			break;
 		}
@@ -153,10 +158,14 @@ static unsigned long lowmem_scan(struct shrinker *s, struct shrink_control *sc)
 		selected = p;
 		selected_tasksize = tasksize;
 		selected_oom_score_adj = oom_score_adj;
-		lowmem_print(2, "select %d (%s), adj %hd, size %d, to kill\n",
-			     p->pid, p->comm, oom_score_adj, tasksize);
+		lowmem_print(2, "select '%s' (%d), adj %hd, size %d, to kill\n",
+			     p->comm, p->pid, oom_score_adj, tasksize);
 	}
 	if (selected) {
+		long cache_size = other_file * (long)(PAGE_SIZE / 1024);
+		long cache_limit = minfree * (long)(PAGE_SIZE / 1024);
+		long free = other_free * (long)(PAGE_SIZE / 1024);
+
 		task_lock(selected);
 		send_sig(SIGKILL, selected, 0);
 		/*
@@ -167,9 +176,18 @@ static unsigned long lowmem_scan(struct shrinker *s, struct shrink_control *sc)
 		if (selected->mm)
 			mark_oom_victim(selected);
 		task_unlock(selected);
-		lowmem_print(1, "send sigkill to %d (%s), adj %hd, size %d\n",
-			     selected->pid, selected->comm,
-			     selected_oom_score_adj, selected_tasksize);
+		trace_lowmemory_kill(selected, cache_size, cache_limit, free);
+		lowmem_print(1, "Killing '%s' (%d) (tgid %d), adj %hd,\n" \
+			        "   to free %ldkB on behalf of '%s' (%d) because\n" \
+			        "   cache %ldkB is below limit %ldkB for oom_score_adj %hd\n" \
+			        "   Free memory is %ldkB above reserved\n",
+			     selected->comm, selected->pid, selected->tgid,
+			     selected_oom_score_adj,
+			     selected_tasksize * (long)(PAGE_SIZE / 1024),
+			     current->comm, current->pid,
+			     cache_size, cache_limit,
+			     min_score_adj,
+			     free);
 		lowmem_deathpending_timeout = jiffies + HZ;
 		rem += selected_tasksize;
 	}
@@ -193,13 +211,97 @@ static int __init lowmem_init(void)
 }
 device_initcall(lowmem_init);
 
+#ifdef CONFIG_ANDROID_LOW_MEMORY_KILLER_AUTODETECT_OOM_ADJ_VALUES
+static short lowmem_oom_adj_to_oom_score_adj(short oom_adj)
+{
+	if (oom_adj == OOM_ADJUST_MAX)
+		return OOM_SCORE_ADJ_MAX;
+	else
+		return (oom_adj * OOM_SCORE_ADJ_MAX) / -OOM_DISABLE;
+}
+
+static void lowmem_autodetect_oom_adj_values(void)
+{
+	int i;
+	short oom_adj;
+	short oom_score_adj;
+	int array_size = ARRAY_SIZE(lowmem_adj);
+
+	if (lowmem_adj_size < array_size)
+		array_size = lowmem_adj_size;
+
+	if (array_size <= 0)
+		return;
+
+	oom_adj = lowmem_adj[array_size - 1];
+	if (oom_adj > OOM_ADJUST_MAX)
+		return;
+
+	oom_score_adj = lowmem_oom_adj_to_oom_score_adj(oom_adj);
+	if (oom_score_adj <= OOM_ADJUST_MAX)
+		return;
+
+	lowmem_print(1, "lowmem_shrink: convert oom_adj to oom_score_adj:\n");
+	for (i = 0; i < array_size; i++) {
+		oom_adj = lowmem_adj[i];
+		oom_score_adj = lowmem_oom_adj_to_oom_score_adj(oom_adj);
+		lowmem_adj[i] = oom_score_adj;
+		lowmem_print(1, "oom_adj %d => oom_score_adj %d\n",
+			     oom_adj, oom_score_adj);
+	}
+}
+
+static int lowmem_adj_array_set(const char *val, const struct kernel_param *kp)
+{
+	int ret;
+
+	ret = param_array_ops.set(val, kp);
+
+	/* HACK: Autodetect oom_adj values in lowmem_adj array */
+	lowmem_autodetect_oom_adj_values();
+
+	return ret;
+}
+
+static int lowmem_adj_array_get(char *buffer, const struct kernel_param *kp)
+{
+	return param_array_ops.get(buffer, kp);
+}
+
+static void lowmem_adj_array_free(void *arg)
+{
+	param_array_ops.free(arg);
+}
+
+static struct kernel_param_ops lowmem_adj_array_ops = {
+	.set = lowmem_adj_array_set,
+	.get = lowmem_adj_array_get,
+	.free = lowmem_adj_array_free,
+};
+
+static const struct kparam_array __param_arr_adj = {
+	.max = ARRAY_SIZE(lowmem_adj),
+	.num = &lowmem_adj_size,
+	.ops = &param_ops_short,
+	.elemsize = sizeof(lowmem_adj[0]),
+	.elem = lowmem_adj,
+};
+#endif
+
 /*
  * not really modular, but the easiest way to keep compat with existing
  * bootargs behaviour is to continue using module_param here.
  */
 module_param_named(cost, lowmem_shrinker.seeks, int, S_IRUGO | S_IWUSR);
+#ifdef CONFIG_ANDROID_LOW_MEMORY_KILLER_AUTODETECT_OOM_ADJ_VALUES
+module_param_cb(adj, &lowmem_adj_array_ops,
+		.arr = &__param_arr_adj,
+		S_IRUGO | S_IWUSR);
+__MODULE_PARM_TYPE(adj, "array of short");
+#else
 module_param_array_named(adj, lowmem_adj, short, &lowmem_adj_size,
 			 S_IRUGO | S_IWUSR);
+#endif
 module_param_array_named(minfree, lowmem_minfree, uint, &lowmem_minfree_size,
 			 S_IRUGO | S_IWUSR);
 module_param_named(debug_level, lowmem_debug_level, uint, S_IRUGO | S_IWUSR);
diff --git a/drivers/staging/android/sync.c b/drivers/staging/android/sync.c
index f83e00c78051..da101a506cd2 100644
--- a/drivers/staging/android/sync.c
+++ b/drivers/staging/android/sync.c
@@ -465,6 +465,13 @@ static bool android_fence_enable_signaling(struct fence *fence)
 	return true;
 }
 
+static void android_fence_disable_signaling(struct fence *fence)
+{
+	struct sync_pt *pt = container_of(fence, struct sync_pt, base);
+
+	list_del_init(&pt->active_list);
+}
+
 static int android_fence_fill_driver_data(struct fence *fence,
 					  void *data, int size)
 {
@@ -508,6 +515,7 @@ static const struct fence_ops android_fence_ops = {
 	.get_driver_name = android_fence_get_driver_name,
 	.get_timeline_name = android_fence_get_timeline_name,
 	.enable_signaling = android_fence_enable_signaling,
+	.disable_signaling = android_fence_disable_signaling,
 	.signaled = android_fence_signaled,
 	.wait = fence_default_wait,
 	.release = android_fence_release,
@@ -519,12 +527,10 @@ static const struct fence_ops android_fence_ops = {
 static void sync_fence_free(struct kref *kref)
 {
 	struct sync_fence *fence = container_of(kref, struct sync_fence, kref);
-	int i, status = atomic_read(&fence->status);
+	int i;
 
 	for (i = 0; i < fence->num_fences; ++i) {
-		if (status)
-			fence_remove_callback(fence->cbs[i].sync_pt,
-					      &fence->cbs[i].cb);
+		fence_remove_callback(fence->cbs[i].sync_pt, &fence->cbs[i].cb);
 		fence_put(fence->cbs[i].sync_pt);
 	}
 
diff --git a/drivers/staging/android/trace/lowmemorykiller.h b/drivers/staging/android/trace/lowmemorykiller.h
new file mode 100644
index 000000000000..f43d3fae75ee
--- /dev/null
+++ b/drivers/staging/android/trace/lowmemorykiller.h
@@ -0,0 +1,41 @@
+#undef TRACE_SYSTEM
+#define TRACE_INCLUDE_PATH ../../drivers/staging/android/trace
+#define TRACE_SYSTEM lowmemorykiller
+
+#if !defined(_TRACE_LOWMEMORYKILLER_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_LOWMEMORYKILLER_H
+
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(lowmemory_kill,
+	TP_PROTO(struct task_struct *killed_task, long cache_size, \
+		 long cache_limit, long free),
+
+	TP_ARGS(killed_task, cache_size, cache_limit, free),
+
+	TP_STRUCT__entry(
+			__array(char, comm, TASK_COMM_LEN)
+			__field(pid_t, pid)
+			__field(long, pagecache_size)
+			__field(long, pagecache_limit)
+			__field(long, free)
+	),
+
+	TP_fast_assign(
+			memcpy(__entry->comm, killed_task->comm, TASK_COMM_LEN);
+			__entry->pid = killed_task->pid;
+			__entry->pagecache_size = cache_size;
+			__entry->pagecache_limit = cache_limit;
+			__entry->free = free;
+	),
+
+	TP_printk("%s (%d), page cache %ldkB (limit %ldkB), free %ldKb",
+		__entry->comm, __entry->pid, __entry->pagecache_size,
+		__entry->pagecache_limit, __entry->free)
+);
+
+
+#endif /* if !defined(_TRACE_LOWMEMORYKILLER_H) || defined(TRACE_HEADER_MULTI_READ) */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/drivers/staging/android/uapi/ashmem.h b/drivers/staging/android/uapi/ashmem.h
index ba4743c71d6b..13df42d200b7 100644
--- a/drivers/staging/android/uapi/ashmem.h
+++ b/drivers/staging/android/uapi/ashmem.h
@@ -13,6 +13,7 @@
 #define _UAPI_LINUX_ASHMEM_H
 
 #include <linux/ioctl.h>
+#include <linux/types.h>
 
 #define ASHMEM_NAME_LEN		256
 
diff --git a/drivers/staging/goldfish/Kconfig b/drivers/staging/goldfish/Kconfig
index 4e094602437c..c579141a7bed 100644
--- a/drivers/staging/goldfish/Kconfig
+++ b/drivers/staging/goldfish/Kconfig
@@ -4,6 +4,12 @@ config GOLDFISH_AUDIO
 	---help---
 	  Emulated audio channel for the Goldfish Android Virtual Device
 
+config GOLDFISH_SYNC
+    tristate "Goldfish AVD Sync Driver"
+    depends on GOLDFISH
+	---help---
+	  Emulated sync fences for the Goldfish Android Virtual Device
+
 config MTD_GOLDFISH_NAND
 	tristate "Goldfish NAND device"
 	depends on GOLDFISH
diff --git a/drivers/staging/goldfish/Makefile b/drivers/staging/goldfish/Makefile
index dec34ad58162..0cf525588210 100644
--- a/drivers/staging/goldfish/Makefile
+++ b/drivers/staging/goldfish/Makefile
@@ -4,3 +4,8 @@
 
 obj-$(CONFIG_GOLDFISH_AUDIO) += goldfish_audio.o
 obj-$(CONFIG_MTD_GOLDFISH_NAND)	+= goldfish_nand.o
+
+# and sync
+
+ccflags-y := -Idrivers/staging/android
+obj-$(CONFIG_GOLDFISH_SYNC) += goldfish_sync.o
diff --git a/drivers/staging/goldfish/goldfish_audio.c b/drivers/staging/goldfish/goldfish_audio.c
index b0927e49d0a8..63b79c09b41b 100644
--- a/drivers/staging/goldfish/goldfish_audio.c
+++ b/drivers/staging/goldfish/goldfish_audio.c
@@ -26,7 +26,9 @@
 #include <linux/sched.h>
 #include <linux/dma-mapping.h>
 #include <linux/uaccess.h>
+#include <linux/slab.h>
 #include <linux/goldfish.h>
+#include <linux/acpi.h>
 
 MODULE_AUTHOR("Google, Inc.");
 MODULE_DESCRIPTION("Android QEMU Audio Driver");
@@ -115,6 +117,7 @@ static ssize_t goldfish_audio_read(struct file *fp, char __user *buf,
 				   size_t count, loff_t *pos)
 {
 	struct goldfish_audio *data = fp->private_data;
+	unsigned long irq_flags;
 	int length;
 	int result = 0;
 
@@ -128,6 +131,10 @@ static ssize_t goldfish_audio_read(struct file *fp, char __user *buf,
 		wait_event_interruptible(data->wait, data->buffer_status &
 					 AUDIO_INT_READ_BUFFER_FULL);
 
+		spin_lock_irqsave(&data->lock, irq_flags);
+		data->buffer_status &= ~AUDIO_INT_READ_BUFFER_FULL;
+		spin_unlock_irqrestore(&data->lock, irq_flags);
+
 		length = AUDIO_READ(data, AUDIO_READ_BUFFER_AVAILABLE);
 
 		/* copy data to user space */
@@ -344,11 +351,25 @@ static int goldfish_audio_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id goldfish_audio_of_match[] = {
+	{ .compatible = "google,goldfish-audio", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, goldfish_audio_of_match);
+
+static const struct acpi_device_id goldfish_audio_acpi_match[] = {
+	{ "GFSH0005", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, goldfish_audio_acpi_match);
+
 static struct platform_driver goldfish_audio_driver = {
 	.probe		= goldfish_audio_probe,
 	.remove		= goldfish_audio_remove,
 	.driver = {
-		.name = "goldfish_audio"
+		.name = "goldfish_audio",
+		.of_match_table = goldfish_audio_of_match,
+		.acpi_match_table = ACPI_PTR(goldfish_audio_acpi_match),
 	}
 };
 
diff --git a/drivers/staging/goldfish/goldfish_sync.c b/drivers/staging/goldfish/goldfish_sync.c
new file mode 100644
index 000000000000..ba8def29901e
--- /dev/null
+++ b/drivers/staging/goldfish/goldfish_sync.c
@@ -0,0 +1,987 @@
+/*
+ * Copyright (C) 2016 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/fdtable.h>
+#include <linux/file.h>
+#include <linux/init.h>
+#include <linux/miscdevice.h>
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+
+#include <linux/interrupt.h>
+#include <linux/kref.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+
+#include <linux/io.h>
+#include <linux/mm.h>
+#include <linux/acpi.h>
+
+#include <linux/string.h>
+#include <linux/syscalls.h>
+
+#include "sw_sync.h"
+#include "sync.h"
+
+#define ERR(...) printk(KERN_ERR __VA_ARGS__);
+
+#define INFO(...) printk(KERN_INFO __VA_ARGS__);
+
+#define DPRINT(...) pr_debug(__VA_ARGS__);
+
+#define DTRACE() DPRINT("%s: enter", __func__)
+
+/* The Goldfish sync driver is designed to provide a interface
+ * between the underlying host's sync device and the kernel's
+ * sw_sync.
+ * The purpose of the device/driver is to enable lightweight
+ * creation and signaling of timelines and fences
+ * in order to synchronize the guest with host-side graphics events.
+ *
+ * Each time the interrupt trips, the driver
+ * may perform a sw_sync operation.
+ */
+
+/* The operations are: */
+
+/* Ready signal - used to mark when irq should lower */
+#define CMD_SYNC_READY            0
+
+/* Create a new timeline. writes timeline handle */
+#define CMD_CREATE_SYNC_TIMELINE  1
+
+/* Create a fence object. reads timeline handle and time argument.
+ * Writes fence fd to the SYNC_REG_HANDLE register. */
+#define CMD_CREATE_SYNC_FENCE     2
+
+/* Increments timeline. reads timeline handle and time argument */
+#define CMD_SYNC_TIMELINE_INC     3
+
+/* Destroys a timeline. reads timeline handle */
+#define CMD_DESTROY_SYNC_TIMELINE 4
+
+/* Starts a wait on the host with
+ * the given glsync object and sync thread handle. */
+#define CMD_TRIGGER_HOST_WAIT     5
+
+/* The register layout is: */
+
+#define SYNC_REG_BATCH_COMMAND                0x00 /* host->guest batch commands */
+#define SYNC_REG_BATCH_GUESTCOMMAND           0x04 /* guest->host batch commands */
+#define SYNC_REG_BATCH_COMMAND_ADDR           0x08 /* communicate physical address of host->guest batch commands */
+#define SYNC_REG_BATCH_COMMAND_ADDR_HIGH      0x0c /* 64-bit part */
+#define SYNC_REG_BATCH_GUESTCOMMAND_ADDR      0x10 /* communicate physical address of guest->host commands */
+#define SYNC_REG_BATCH_GUESTCOMMAND_ADDR_HIGH 0x14 /* 64-bit part */
+#define SYNC_REG_INIT                         0x18 /* signals that the device has been probed */
+
+/* There is an ioctl associated with goldfish sync driver.
+ * Make it conflict with ioctls that are not likely to be used
+ * in the emulator.
+ *
+ * '@'	00-0F	linux/radeonfb.h	conflict!
+ * '@'	00-0F	drivers/video/aty/aty128fb.c	conflict!
+ */
+#define GOLDFISH_SYNC_IOC_MAGIC	'@'
+
+#define GOLDFISH_SYNC_IOC_QUEUE_WORK	_IOWR(GOLDFISH_SYNC_IOC_MAGIC, 0, struct goldfish_sync_ioctl_info)
+
+/* The above definitions (command codes, register layout, ioctl definitions)
+ * need to be in sync with the following files:
+ *
+ * Host-side (emulator):
+ * external/qemu/android/emulation/goldfish_sync.h
+ * external/qemu-android/hw/misc/goldfish_sync.c
+ *
+ * Guest-side (system image):
+ * device/generic/goldfish-opengl/system/egl/goldfish_sync.h
+ * device/generic/goldfish/ueventd.ranchu.rc
+ * platform/build/target/board/generic/sepolicy/file_contexts
+ */
+struct goldfish_sync_hostcmd {
+	/* sorted for alignment */
+	uint64_t handle;
+	uint64_t hostcmd_handle;
+	uint32_t cmd;
+	uint32_t time_arg;
+};
+
+struct goldfish_sync_guestcmd {
+	uint64_t host_command; /* uint64_t for alignment */
+	uint64_t glsync_handle;
+	uint64_t thread_handle;
+	uint64_t guest_timeline_handle;
+};
+
+#define GOLDFISH_SYNC_MAX_CMDS 64
+
+struct goldfish_sync_state {
+	char __iomem *reg_base;
+	int irq;
+
+	/* Spinlock protects |to_do| / |to_do_end|. */
+	spinlock_t lock;
+	/* |mutex_lock| protects all concurrent access
+	 * to timelines for both kernel and user space. */
+	struct mutex mutex_lock;
+
+	/* Buffer holding commands issued from host. */
+	struct goldfish_sync_hostcmd to_do[GOLDFISH_SYNC_MAX_CMDS];
+	uint32_t to_do_end;
+
+	/* Addresses for the reading or writing
+	 * of individual commands. The host can directly write
+	 * to |batch_hostcmd| (and then this driver immediately
+	 * copies contents to |to_do|). This driver either replies
+	 * through |batch_hostcmd| or simply issues a
+	 * guest->host command through |batch_guestcmd|.
+	 */
+	struct goldfish_sync_hostcmd *batch_hostcmd;
+	struct goldfish_sync_guestcmd *batch_guestcmd;
+
+	/* Used to give this struct itself to a work queue
+	 * function for executing actual sync commands. */
+	struct work_struct work_item;
+};
+
+static struct goldfish_sync_state global_sync_state[1];
+
+struct goldfish_sync_timeline_obj {
+	struct sw_sync_timeline *sw_sync_tl;
+	uint32_t current_time;
+	/* We need to be careful about when we deallocate
+	 * this |goldfish_sync_timeline_obj| struct.
+	 * In order to ensure proper cleanup, we need to
+	 * consider the triggered host-side wait that may
+	 * still be in flight when the guest close()'s a
+	 * goldfish_sync device's sync context fd (and
+	 * destroys the |sw_sync_tl| field above).
+	 * The host-side wait may raise IRQ
+	 * and tell the kernel to increment the timeline _after_
+	 * the |sw_sync_tl| has already been set to null.
+	 *
+	 * From observations on OpenGL apps and CTS tests, this
+	 * happens at some very low probability upon context
+	 * destruction or process close, but it does happen
+	 * and it needs to be handled properly. Otherwise,
+	 * if we clean up the surrounding |goldfish_sync_timeline_obj|
+	 * too early, any |handle| field of any host->guest command
+	 * might not even point to a null |sw_sync_tl| field,
+	 * but to garbage memory or even a reclaimed |sw_sync_tl|.
+	 * If we do not count such "pending waits" and kfree the object
+	 * immediately upon |goldfish_sync_timeline_destroy|,
+	 * we might get mysterous RCU stalls after running a long
+	 * time because the garbage memory that is being read
+	 * happens to be interpretable as a |spinlock_t| struct
+	 * that is currently in the locked state.
+	 *
+	 * To track when to free the |goldfish_sync_timeline_obj|
+	 * itself, we maintain a kref.
+	 * The kref essentially counts the timeline itself plus
+	 * the number of waits in flight. kref_init/kref_put
+	 * are issued on
+	 * |goldfish_sync_timeline_create|/|goldfish_sync_timeline_destroy|
+	 * and kref_get/kref_put are issued on
+	 * |goldfish_sync_fence_create|/|goldfish_sync_timeline_inc|.
+	 *
+	 * The timeline is destroyed after reference count
+	 * reaches zero, which would happen after
+	 * |goldfish_sync_timeline_destroy| and all pending
+	 * |goldfish_sync_timeline_inc|'s are fulfilled.
+	 *
+	 * NOTE (1): We assume that |fence_create| and
+	 * |timeline_inc| calls are 1:1, otherwise the kref scheme
+	 * will not work. This is a valid assumption as long
+	 * as the host-side virtual device implementation
+	 * does not insert any timeline increments
+	 * that we did not trigger from here.
+	 *
+	 * NOTE (2): The use of kref by itself requires no locks,
+	 * but this does not mean everything works without locks.
+	 * Related timeline operations do require a lock of some sort,
+	 * or at least are not proven to work without it.
+	 * In particualr, we assume that all the operations
+	 * done on the |kref| field above are done in contexts where
+	 * |global_sync_state->mutex_lock| is held. Do not
+	 * remove that lock until everything is proven to work
+	 * without it!!! */
+	struct kref kref;
+};
+
+/* We will call |delete_timeline_obj| when the last reference count
+ * of the kref is decremented. This deletes the sw_sync
+ * timeline object along with the wrapper itself. */
+static void delete_timeline_obj(struct kref* kref) {
+	struct goldfish_sync_timeline_obj* obj =
+		container_of(kref, struct goldfish_sync_timeline_obj, kref);
+
+	sync_timeline_destroy(&obj->sw_sync_tl->obj);
+	obj->sw_sync_tl = NULL;
+	kfree(obj);
+}
+
+static uint64_t gensym_ctr;
+static void gensym(char *dst)
+{
+	sprintf(dst, "goldfish_sync:gensym:%llu", gensym_ctr);
+	gensym_ctr++;
+}
+
+/* |goldfish_sync_timeline_create| assumes that |global_sync_state->mutex_lock|
+ * is held. */
+static struct goldfish_sync_timeline_obj*
+goldfish_sync_timeline_create(void)
+{
+
+	char timeline_name[256];
+	struct sw_sync_timeline *res_sync_tl = NULL;
+	struct goldfish_sync_timeline_obj *res;
+
+	DTRACE();
+
+	gensym(timeline_name);
+
+	res_sync_tl = sw_sync_timeline_create(timeline_name);
+	if (!res_sync_tl) {
+		ERR("Failed to create sw_sync timeline.");
+		return NULL;
+	}
+
+	res = kzalloc(sizeof(struct goldfish_sync_timeline_obj), GFP_KERNEL);
+	res->sw_sync_tl = res_sync_tl;
+	res->current_time = 0;
+	kref_init(&res->kref);
+
+	DPRINT("new timeline_obj=0x%p", res);
+	return res;
+}
+
+/* |goldfish_sync_fence_create| assumes that |global_sync_state->mutex_lock|
+ * is held. */
+static int
+goldfish_sync_fence_create(struct goldfish_sync_timeline_obj *obj,
+							uint32_t val)
+{
+
+	int fd;
+	char fence_name[256];
+	struct sync_pt *syncpt = NULL;
+	struct sync_fence *sync_obj = NULL;
+	struct sw_sync_timeline *tl;
+
+	DTRACE();
+
+	if (!obj) return -1;
+
+	tl = obj->sw_sync_tl;
+
+	syncpt = sw_sync_pt_create(tl, val);
+	if (!syncpt) {
+		ERR("could not create sync point! "
+			"sync_timeline=0x%p val=%d",
+			   tl, val);
+		return -1;
+	}
+
+	fd = get_unused_fd_flags(O_CLOEXEC);
+	if (fd < 0) {
+		ERR("could not get unused fd for sync fence. "
+			"errno=%d", fd);
+		goto err_cleanup_pt;
+	}
+
+	gensym(fence_name);
+
+	sync_obj = sync_fence_create(fence_name, syncpt);
+	if (!sync_obj) {
+		ERR("could not create sync fence! "
+			"sync_timeline=0x%p val=%d sync_pt=0x%p",
+			   tl, val, syncpt);
+		goto err_cleanup_fd_pt;
+	}
+
+	DPRINT("installing sync fence into fd %d sync_obj=0x%p", fd, sync_obj);
+	sync_fence_install(sync_obj, fd);
+	kref_get(&obj->kref);
+
+	return fd;
+
+err_cleanup_fd_pt:
+	put_unused_fd(fd);
+err_cleanup_pt:
+	sync_pt_free(syncpt);
+	return -1;
+}
+
+/* |goldfish_sync_timeline_inc| assumes that |global_sync_state->mutex_lock|
+ * is held. */
+static void
+goldfish_sync_timeline_inc(struct goldfish_sync_timeline_obj *obj, uint32_t inc)
+{
+	DTRACE();
+	/* Just give up if someone else nuked the timeline.
+	 * Whoever it was won't care that it doesn't get signaled. */
+	if (!obj) return;
+
+	DPRINT("timeline_obj=0x%p", obj);
+	sw_sync_timeline_inc(obj->sw_sync_tl, inc);
+	DPRINT("incremented timeline. increment max_time");
+	obj->current_time += inc;
+
+	/* Here, we will end up deleting the timeline object if it
+	 * turns out that this call was a pending increment after
+	 * |goldfish_sync_timeline_destroy| was called. */
+	kref_put(&obj->kref, delete_timeline_obj);
+	DPRINT("done");
+}
+
+/* |goldfish_sync_timeline_destroy| assumes
+ * that |global_sync_state->mutex_lock| is held. */
+static void
+goldfish_sync_timeline_destroy(struct goldfish_sync_timeline_obj *obj)
+{
+	DTRACE();
+	/* See description of |goldfish_sync_timeline_obj| for why we
+	 * should not immediately destroy |obj| */
+	kref_put(&obj->kref, delete_timeline_obj);
+}
+
+static inline void
+goldfish_sync_cmd_queue(struct goldfish_sync_state *sync_state,
+						uint32_t cmd,
+						uint64_t handle,
+						uint32_t time_arg,
+						uint64_t hostcmd_handle)
+{
+	struct goldfish_sync_hostcmd *to_add;
+
+	DTRACE();
+
+	BUG_ON(sync_state->to_do_end == GOLDFISH_SYNC_MAX_CMDS);
+
+	to_add = &sync_state->to_do[sync_state->to_do_end];
+
+	to_add->cmd = cmd;
+	to_add->handle = handle;
+	to_add->time_arg = time_arg;
+	to_add->hostcmd_handle = hostcmd_handle;
+
+	sync_state->to_do_end += 1;
+}
+
+static inline void
+goldfish_sync_hostcmd_reply(struct goldfish_sync_state *sync_state,
+							uint32_t cmd,
+							uint64_t handle,
+							uint32_t time_arg,
+							uint64_t hostcmd_handle)
+{
+	unsigned long irq_flags;
+	struct goldfish_sync_hostcmd *batch_hostcmd =
+		sync_state->batch_hostcmd;
+
+	DTRACE();
+
+	spin_lock_irqsave(&sync_state->lock, irq_flags);
+
+	batch_hostcmd->cmd = cmd;
+	batch_hostcmd->handle = handle;
+	batch_hostcmd->time_arg = time_arg;
+	batch_hostcmd->hostcmd_handle = hostcmd_handle;
+	writel(0, sync_state->reg_base + SYNC_REG_BATCH_COMMAND);
+
+	spin_unlock_irqrestore(&sync_state->lock, irq_flags);
+}
+
+static inline void
+goldfish_sync_send_guestcmd(struct goldfish_sync_state *sync_state,
+							uint32_t cmd,
+							uint64_t glsync_handle,
+							uint64_t thread_handle,
+							uint64_t timeline_handle)
+{
+	unsigned long irq_flags;
+	struct goldfish_sync_guestcmd *batch_guestcmd =
+		sync_state->batch_guestcmd;
+
+	DTRACE();
+
+	spin_lock_irqsave(&sync_state->lock, irq_flags);
+
+	batch_guestcmd->host_command = (uint64_t)cmd;
+	batch_guestcmd->glsync_handle = (uint64_t)glsync_handle;
+	batch_guestcmd->thread_handle = (uint64_t)thread_handle;
+	batch_guestcmd->guest_timeline_handle = (uint64_t)timeline_handle;
+	writel(0, sync_state->reg_base + SYNC_REG_BATCH_GUESTCOMMAND);
+
+	spin_unlock_irqrestore(&sync_state->lock, irq_flags);
+}
+
+/* |goldfish_sync_interrupt| handles IRQ raises from the virtual device.
+ * In the context of OpenGL, this interrupt will fire whenever we need
+ * to signal a fence fd in the guest, with the command
+ * |CMD_SYNC_TIMELINE_INC|.
+ * However, because this function will be called in an interrupt context,
+ * it is necessary to do the actual work of signaling off of interrupt context.
+ * The shared work queue is used for this purpose. At the end when
+ * all pending commands are intercepted by the interrupt handler,
+ * we call |schedule_work|, which will later run the actual
+ * desired sync command in |goldfish_sync_work_item_fn|.
+ */
+static irqreturn_t goldfish_sync_interrupt(int irq, void *dev_id)
+{
+
+	struct goldfish_sync_state *sync_state = dev_id;
+
+	uint32_t nextcmd;
+	uint32_t command_r;
+	uint64_t handle_rw;
+	uint32_t time_r;
+	uint64_t hostcmd_handle_rw;
+
+	int count = 0;
+
+	DTRACE();
+
+	sync_state = dev_id;
+
+	spin_lock(&sync_state->lock);
+
+	for (;;) {
+
+		readl(sync_state->reg_base + SYNC_REG_BATCH_COMMAND);
+		nextcmd = sync_state->batch_hostcmd->cmd;
+
+		if (nextcmd == 0)
+			break;
+
+		command_r = nextcmd;
+		handle_rw = sync_state->batch_hostcmd->handle;
+		time_r = sync_state->batch_hostcmd->time_arg;
+		hostcmd_handle_rw = sync_state->batch_hostcmd->hostcmd_handle;
+
+		goldfish_sync_cmd_queue(
+				sync_state,
+				command_r,
+				handle_rw,
+				time_r,
+				hostcmd_handle_rw);
+
+		count++;
+	}
+
+	spin_unlock(&sync_state->lock);
+
+	schedule_work(&sync_state->work_item);
+
+	return (count == 0) ? IRQ_NONE : IRQ_HANDLED;
+}
+
+/* |goldfish_sync_work_item_fn| does the actual work of servicing
+ * host->guest sync commands. This function is triggered whenever
+ * the IRQ for the goldfish sync device is raised. Once it starts
+ * running, it grabs the contents of the buffer containing the
+ * commands it needs to execute (there may be multiple, because
+ * our IRQ is active high and not edge triggered), and then
+ * runs all of them one after the other.
+ */
+static void goldfish_sync_work_item_fn(struct work_struct *input)
+{
+
+	struct goldfish_sync_state *sync_state;
+	int sync_fence_fd;
+
+	struct goldfish_sync_timeline_obj *timeline;
+	uint64_t timeline_ptr;
+
+	uint64_t hostcmd_handle;
+
+	uint32_t cmd;
+	uint64_t handle;
+	uint32_t time_arg;
+
+	struct goldfish_sync_hostcmd *todo;
+	uint32_t todo_end;
+
+	unsigned long irq_flags;
+
+	struct goldfish_sync_hostcmd to_run[GOLDFISH_SYNC_MAX_CMDS];
+	uint32_t i = 0;
+
+	sync_state = container_of(input, struct goldfish_sync_state, work_item);
+
+	mutex_lock(&sync_state->mutex_lock);
+
+	spin_lock_irqsave(&sync_state->lock, irq_flags); {
+
+		todo_end = sync_state->to_do_end;
+
+		DPRINT("num sync todos: %u", sync_state->to_do_end);
+
+		for (i = 0; i < todo_end; i++)
+			to_run[i] = sync_state->to_do[i];
+
+		/* We expect that commands will come in at a slow enough rate
+		 * so that incoming items will not be more than
+		 * GOLDFISH_SYNC_MAX_CMDS.
+		 *
+		 * This is because the way the sync device is used,
+		 * it's only for managing buffer data transfers per frame,
+		 * with a sequential dependency between putting things in
+		 * to_do and taking them out. Once a set of commands is
+		 * queued up in to_do, the user of the device waits for
+		 * them to be processed before queuing additional commands,
+		 * which limits the rate at which commands come in
+		 * to the rate at which we take them out here.
+		 *
+		 * We also don't expect more than MAX_CMDS to be issued
+		 * at once; there is a correspondence between
+		 * which buffers need swapping to the (display / buffer queue)
+		 * to particular commands, and we don't expect there to be
+		 * enough display or buffer queues in operation at once
+		 * to overrun GOLDFISH_SYNC_MAX_CMDS.
+		 */
+		sync_state->to_do_end = 0;
+
+	} spin_unlock_irqrestore(&sync_state->lock, irq_flags);
+
+	for (i = 0; i < todo_end; i++) {
+		DPRINT("todo index: %u", i);
+
+		todo = &to_run[i];
+
+		cmd = todo->cmd;
+
+		handle = (uint64_t)todo->handle;
+		time_arg = todo->time_arg;
+		hostcmd_handle = (uint64_t)todo->hostcmd_handle;
+
+		DTRACE();
+
+		timeline = (struct goldfish_sync_timeline_obj *)(uintptr_t)handle;
+
+		switch (cmd) {
+		case CMD_SYNC_READY:
+			break;
+		case CMD_CREATE_SYNC_TIMELINE:
+			DPRINT("exec CMD_CREATE_SYNC_TIMELINE: "
+					"handle=0x%llx time_arg=%d",
+					handle, time_arg);
+			timeline = goldfish_sync_timeline_create();
+			timeline_ptr = (uintptr_t)timeline;
+			goldfish_sync_hostcmd_reply(sync_state, CMD_CREATE_SYNC_TIMELINE,
+										timeline_ptr,
+										0,
+										hostcmd_handle);
+			DPRINT("sync timeline created: %p", timeline);
+			break;
+		case CMD_CREATE_SYNC_FENCE:
+			DPRINT("exec CMD_CREATE_SYNC_FENCE: "
+					"handle=0x%llx time_arg=%d",
+					handle, time_arg);
+			sync_fence_fd = goldfish_sync_fence_create(timeline, time_arg);
+			goldfish_sync_hostcmd_reply(sync_state, CMD_CREATE_SYNC_FENCE,
+										sync_fence_fd,
+										0,
+										hostcmd_handle);
+			break;
+		case CMD_SYNC_TIMELINE_INC:
+			DPRINT("exec CMD_SYNC_TIMELINE_INC: "
+					"handle=0x%llx time_arg=%d",
+					handle, time_arg);
+			goldfish_sync_timeline_inc(timeline, time_arg);
+			break;
+		case CMD_DESTROY_SYNC_TIMELINE:
+			DPRINT("exec CMD_DESTROY_SYNC_TIMELINE: "
+					"handle=0x%llx time_arg=%d",
+					handle, time_arg);
+			goldfish_sync_timeline_destroy(timeline);
+			break;
+		}
+		DPRINT("Done executing sync command");
+	}
+	mutex_unlock(&sync_state->mutex_lock);
+}
+
+/* Guest-side interface: file operations */
+
+/* Goldfish sync context and ioctl info.
+ *
+ * When a sync context is created by open()-ing the goldfish sync device, we
+ * create a sync context (|goldfish_sync_context|).
+ *
+ * Currently, the only data required to track is the sync timeline itself
+ * along with the current time, which are all packed up in the
+ * |goldfish_sync_timeline_obj| field. We use a |goldfish_sync_context|
+ * as the filp->private_data.
+ *
+ * Next, when a sync context user requests that work be queued and a fence
+ * fd provided, we use the |goldfish_sync_ioctl_info| struct, which holds
+ * information about which host handles to touch for this particular
+ * queue-work operation. We need to know about the host-side sync thread
+ * and the particular host-side GLsync object. We also possibly write out
+ * a file descriptor.
+ */
+struct goldfish_sync_context {
+	struct goldfish_sync_timeline_obj *timeline;
+};
+
+struct goldfish_sync_ioctl_info {
+	uint64_t host_glsync_handle_in;
+	uint64_t host_syncthread_handle_in;
+	int fence_fd_out;
+};
+
+static int goldfish_sync_open(struct inode *inode, struct file *file)
+{
+
+	struct goldfish_sync_context *sync_context;
+
+	DTRACE();
+
+	mutex_lock(&global_sync_state->mutex_lock);
+
+	sync_context = kzalloc(sizeof(struct goldfish_sync_context), GFP_KERNEL);
+
+	if (sync_context == NULL) {
+		ERR("Creation of goldfish sync context failed!");
+		mutex_unlock(&global_sync_state->mutex_lock);
+		return -ENOMEM;
+	}
+
+	sync_context->timeline = NULL;
+
+	file->private_data = sync_context;
+
+	DPRINT("successfully create a sync context @0x%p", sync_context);
+
+	mutex_unlock(&global_sync_state->mutex_lock);
+
+	return 0;
+}
+
+static int goldfish_sync_release(struct inode *inode, struct file *file)
+{
+
+	struct goldfish_sync_context *sync_context;
+
+	DTRACE();
+
+	mutex_lock(&global_sync_state->mutex_lock);
+
+	sync_context = file->private_data;
+
+	if (sync_context->timeline)
+		goldfish_sync_timeline_destroy(sync_context->timeline);
+
+	sync_context->timeline = NULL;
+
+	kfree(sync_context);
+
+	mutex_unlock(&global_sync_state->mutex_lock);
+
+	return 0;
+}
+
+/* |goldfish_sync_ioctl| is the guest-facing interface of goldfish sync
+ * and is used in conjunction with eglCreateSyncKHR to queue up the
+ * actual work of waiting for the EGL sync command to complete,
+ * possibly returning a fence fd to the guest.
+ */
+static long goldfish_sync_ioctl(struct file *file,
+								unsigned int cmd,
+								unsigned long arg)
+{
+	struct goldfish_sync_context *sync_context_data;
+	struct goldfish_sync_timeline_obj *timeline;
+	int fd_out;
+	struct goldfish_sync_ioctl_info ioctl_data;
+
+	DTRACE();
+
+	sync_context_data = file->private_data;
+	fd_out = -1;
+
+	switch (cmd) {
+	case GOLDFISH_SYNC_IOC_QUEUE_WORK:
+
+		DPRINT("exec GOLDFISH_SYNC_IOC_QUEUE_WORK");
+
+		mutex_lock(&global_sync_state->mutex_lock);
+
+		if (copy_from_user(&ioctl_data,
+						(void __user *)arg,
+						sizeof(ioctl_data))) {
+			ERR("Failed to copy memory for ioctl_data from user.");
+			mutex_unlock(&global_sync_state->mutex_lock);
+			return -EFAULT;
+		}
+
+		if (ioctl_data.host_syncthread_handle_in == 0) {
+			DPRINT("Error: zero host syncthread handle!!!");
+			mutex_unlock(&global_sync_state->mutex_lock);
+			return -EFAULT;
+		}
+
+		if (!sync_context_data->timeline) {
+			DPRINT("no timeline yet, create one.");
+			sync_context_data->timeline = goldfish_sync_timeline_create();
+			DPRINT("timeline: 0x%p", &sync_context_data->timeline);
+		}
+
+		timeline = sync_context_data->timeline;
+		fd_out = goldfish_sync_fence_create(timeline,
+											timeline->current_time + 1);
+		DPRINT("Created fence with fd %d and current time %u (timeline: 0x%p)",
+			   fd_out,
+			   sync_context_data->timeline->current_time + 1,
+			   sync_context_data->timeline);
+
+		ioctl_data.fence_fd_out = fd_out;
+
+		if (copy_to_user((void __user *)arg,
+						&ioctl_data,
+						sizeof(ioctl_data))) {
+			DPRINT("Error, could not copy to user!!!");
+
+			sys_close(fd_out);
+			/* We won't be doing an increment, kref_put immediately. */
+			kref_put(&timeline->kref, delete_timeline_obj);
+			mutex_unlock(&global_sync_state->mutex_lock);
+			return -EFAULT;
+		}
+
+		/* We are now about to trigger a host-side wait;
+		 * accumulate on |pending_waits|. */
+		goldfish_sync_send_guestcmd(global_sync_state,
+				CMD_TRIGGER_HOST_WAIT,
+				ioctl_data.host_glsync_handle_in,
+				ioctl_data.host_syncthread_handle_in,
+				(uint64_t)(uintptr_t)(sync_context_data->timeline));
+
+		mutex_unlock(&global_sync_state->mutex_lock);
+		return 0;
+	default:
+		return -ENOTTY;
+	}
+}
+
+static const struct file_operations goldfish_sync_fops = {
+	.owner = THIS_MODULE,
+	.open = goldfish_sync_open,
+	.release = goldfish_sync_release,
+	.unlocked_ioctl = goldfish_sync_ioctl,
+	.compat_ioctl = goldfish_sync_ioctl,
+};
+
+static struct miscdevice goldfish_sync_device = {
+	.name = "goldfish_sync",
+	.fops = &goldfish_sync_fops,
+};
+
+
+static bool setup_verify_batch_cmd_addr(struct goldfish_sync_state *sync_state,
+										void *batch_addr,
+										uint32_t addr_offset,
+										uint32_t addr_offset_high)
+{
+	uint64_t batch_addr_phys;
+	uint32_t batch_addr_phys_test_lo;
+	uint32_t batch_addr_phys_test_hi;
+
+	if (!batch_addr) {
+		ERR("Could not use batch command address!");
+		return false;
+	}
+
+	batch_addr_phys = virt_to_phys(batch_addr);
+	writel((uint32_t)(batch_addr_phys),
+			sync_state->reg_base + addr_offset);
+	writel((uint32_t)(batch_addr_phys >> 32),
+			sync_state->reg_base + addr_offset_high);
+
+	batch_addr_phys_test_lo =
+		readl(sync_state->reg_base + addr_offset);
+	batch_addr_phys_test_hi =
+		readl(sync_state->reg_base + addr_offset_high);
+
+	if (virt_to_phys(batch_addr) !=
+			(((uint64_t)batch_addr_phys_test_hi << 32) |
+			 batch_addr_phys_test_lo)) {
+		ERR("Invalid batch command address!");
+		return false;
+	}
+
+	return true;
+}
+
+int goldfish_sync_probe(struct platform_device *pdev)
+{
+	struct resource *ioresource;
+	struct goldfish_sync_state *sync_state = global_sync_state;
+	int status;
+
+	DTRACE();
+
+	sync_state->to_do_end = 0;
+
+	spin_lock_init(&sync_state->lock);
+	mutex_init(&sync_state->mutex_lock);
+
+	platform_set_drvdata(pdev, sync_state);
+
+	ioresource = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (ioresource == NULL) {
+		ERR("platform_get_resource failed");
+		return -ENODEV;
+	}
+
+	sync_state->reg_base = devm_ioremap(&pdev->dev, ioresource->start, PAGE_SIZE);
+	if (sync_state->reg_base == NULL) {
+		ERR("Could not ioremap");
+		return -ENOMEM;
+	}
+
+	sync_state->irq = platform_get_irq(pdev, 0);
+	if (sync_state->irq < 0) {
+		ERR("Could not platform_get_irq");
+		return -ENODEV;
+	}
+
+	status = devm_request_irq(&pdev->dev,
+							sync_state->irq,
+							goldfish_sync_interrupt,
+							IRQF_SHARED,
+							pdev->name,
+							sync_state);
+	if (status) {
+		ERR("request_irq failed");
+		return -ENODEV;
+	}
+
+	INIT_WORK(&sync_state->work_item,
+			  goldfish_sync_work_item_fn);
+
+	misc_register(&goldfish_sync_device);
+
+	/* Obtain addresses for batch send/recv of commands. */
+	{
+		struct goldfish_sync_hostcmd *batch_addr_hostcmd;
+		struct goldfish_sync_guestcmd *batch_addr_guestcmd;
+
+		batch_addr_hostcmd = devm_kzalloc(&pdev->dev, sizeof(struct goldfish_sync_hostcmd),
+				GFP_KERNEL);
+		batch_addr_guestcmd = devm_kzalloc(&pdev->dev, sizeof(struct goldfish_sync_guestcmd),
+				GFP_KERNEL);
+
+		if (!setup_verify_batch_cmd_addr(sync_state,
+					batch_addr_hostcmd,
+					SYNC_REG_BATCH_COMMAND_ADDR,
+					SYNC_REG_BATCH_COMMAND_ADDR_HIGH)) {
+			ERR("goldfish_sync: Could not setup batch command address");
+			return -ENODEV;
+		}
+
+		if (!setup_verify_batch_cmd_addr(sync_state,
+					batch_addr_guestcmd,
+					SYNC_REG_BATCH_GUESTCOMMAND_ADDR,
+					SYNC_REG_BATCH_GUESTCOMMAND_ADDR_HIGH)) {
+			ERR("goldfish_sync: Could not setup batch guest command address");
+			return -ENODEV;
+		}
+
+		sync_state->batch_hostcmd = batch_addr_hostcmd;
+		sync_state->batch_guestcmd = batch_addr_guestcmd;
+	}
+
+	INFO("goldfish_sync: Initialized goldfish sync device");
+
+	writel(0, sync_state->reg_base + SYNC_REG_INIT);
+
+	return 0;
+}
+
+static int goldfish_sync_remove(struct platform_device *pdev)
+{
+	struct goldfish_sync_state *sync_state = global_sync_state;
+
+	DTRACE();
+
+	misc_deregister(&goldfish_sync_device);
+	memset(sync_state, 0, sizeof(struct goldfish_sync_state));
+	return 0;
+}
+
+static const struct of_device_id goldfish_sync_of_match[] = {
+	{ .compatible = "google,goldfish-sync", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, goldfish_sync_of_match);
+
+static const struct acpi_device_id goldfish_sync_acpi_match[] = {
+	{ "GFSH0006", 0 },
+	{ },
+};
+
+MODULE_DEVICE_TABLE(acpi, goldfish_sync_acpi_match);
+
+static struct platform_driver goldfish_sync = {
+	.probe = goldfish_sync_probe,
+	.remove = goldfish_sync_remove,
+	.driver = {
+		.name = "goldfish_sync",
+		.of_match_table = goldfish_sync_of_match,
+		.acpi_match_table = ACPI_PTR(goldfish_sync_acpi_match),
+	}
+};
+
+module_platform_driver(goldfish_sync);
+
+MODULE_AUTHOR("Google, Inc.");
+MODULE_DESCRIPTION("Android QEMU Sync Driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION("1.0");
+
+/* This function is only to run a basic test of sync framework.
+ * It creates a timeline and fence object whose signal point is at 1.
+ * The timeline is incremented, and we use the sync framework's
+ * sync_fence_wait on that fence object. If everything works out,
+ * we should not hang in the wait and return immediately.
+ * There is no way to explicitly run this test yet, but it
+ * can be used by inserting it at the end of goldfish_sync_probe.
+ */
+void test_kernel_sync(void)
+{
+	struct goldfish_sync_timeline_obj *test_timeline;
+	int test_fence_fd;
+
+	DTRACE();
+
+	DPRINT("test sw_sync");
+
+	test_timeline = goldfish_sync_timeline_create();
+	DPRINT("sw_sync_timeline_create -> 0x%p", test_timeline);
+
+	test_fence_fd = goldfish_sync_fence_create(test_timeline, 1);
+	DPRINT("sync_fence_create -> %d", test_fence_fd);
+
+	DPRINT("incrementing test timeline");
+	goldfish_sync_timeline_inc(test_timeline, 1);
+
+	DPRINT("test waiting (should NOT hang)");
+	sync_fence_wait(
+			sync_fence_fdget(test_fence_fd), -1);
+
+	DPRINT("test waiting (afterward)");
+}
diff --git a/drivers/tee/optee/rpc.c b/drivers/tee/optee/rpc.c
index 8814eca06021..8c1ab49c83c3 100644
--- a/drivers/tee/optee/rpc.c
+++ b/drivers/tee/optee/rpc.c
@@ -16,6 +16,7 @@
 
 #include <linux/delay.h>
 #include <linux/device.h>
+#include <linux/sched.h>
 #include <linux/slab.h>
 #include <linux/tee_drv.h>
 #include "optee_private.h"
diff --git a/drivers/tty/goldfish.c b/drivers/tty/goldfish.c
index 0f82c0b146f6..1e332855b933 100644
--- a/drivers/tty/goldfish.c
+++ b/drivers/tty/goldfish.c
@@ -68,8 +68,7 @@ static void goldfish_tty_do_write(int line, const char *buf, unsigned count)
 
 static irqreturn_t goldfish_tty_interrupt(int irq, void *dev_id)
 {
-	struct platform_device *pdev = dev_id;
-	struct goldfish_tty *qtty = &goldfish_ttys[pdev->id];
+	struct goldfish_tty *qtty = dev_id;
 	void __iomem *base = qtty->base;
 	unsigned long irq_flags;
 	unsigned char *buf;
@@ -233,6 +232,7 @@ static int goldfish_tty_probe(struct platform_device *pdev)
 	struct device *ttydev;
 	void __iomem *base;
 	u32 irq;
+	unsigned int line;
 
 	r = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	if (r == NULL)
@@ -248,10 +248,16 @@ static int goldfish_tty_probe(struct platform_device *pdev)
 
 	irq = r->start;
 
-	if (pdev->id >= goldfish_tty_line_count)
-		goto err_unmap;
-
 	mutex_lock(&goldfish_tty_lock);
+
+	if (pdev->id == PLATFORM_DEVID_NONE)
+		line = goldfish_tty_current_line_count;
+	else
+		line = pdev->id;
+
+	if (line >= goldfish_tty_line_count)
+		goto err_create_driver_failed;
+
 	if (goldfish_tty_current_line_count == 0) {
 		ret = goldfish_tty_create_driver();
 		if (ret)
@@ -259,7 +265,7 @@ static int goldfish_tty_probe(struct platform_device *pdev)
 	}
 	goldfish_tty_current_line_count++;
 
-	qtty = &goldfish_ttys[pdev->id];
+	qtty = &goldfish_ttys[line];
 	spin_lock_init(&qtty->lock);
 	tty_port_init(&qtty->port);
 	qtty->port.ops = &goldfish_port_ops;
@@ -269,13 +275,13 @@ static int goldfish_tty_probe(struct platform_device *pdev)
 	writel(GOLDFISH_TTY_CMD_INT_DISABLE, base + GOLDFISH_TTY_CMD);
 
 	ret = request_irq(irq, goldfish_tty_interrupt, IRQF_SHARED,
-						"goldfish_tty", pdev);
+						"goldfish_tty", qtty);
 	if (ret)
 		goto err_request_irq_failed;
 
 
 	ttydev = tty_port_register_device(&qtty->port, goldfish_tty_driver,
-							pdev->id, &pdev->dev);
+							line, &pdev->dev);
 	if (IS_ERR(ttydev)) {
 		ret = PTR_ERR(ttydev);
 		goto err_tty_register_device_failed;
@@ -286,8 +292,9 @@ static int goldfish_tty_probe(struct platform_device *pdev)
 	qtty->console.device = goldfish_tty_console_device;
 	qtty->console.setup = goldfish_tty_console_setup;
 	qtty->console.flags = CON_PRINTBUFFER;
-	qtty->console.index = pdev->id;
+	qtty->console.index = line;
 	register_console(&qtty->console);
+	platform_set_drvdata(pdev, qtty);
 
 	mutex_unlock(&goldfish_tty_lock);
 	return 0;
@@ -307,13 +314,12 @@ err_unmap:
 
 static int goldfish_tty_remove(struct platform_device *pdev)
 {
-	struct goldfish_tty *qtty;
+	struct goldfish_tty *qtty = platform_get_drvdata(pdev);
 
 	mutex_lock(&goldfish_tty_lock);
 
-	qtty = &goldfish_ttys[pdev->id];
 	unregister_console(&qtty->console);
-	tty_unregister_device(goldfish_tty_driver, pdev->id);
+	tty_unregister_device(goldfish_tty_driver, qtty->console.index);
 	iounmap(qtty->base);
 	qtty->base = NULL;
 	free_irq(qtty->irq, pdev);
@@ -324,11 +330,19 @@ static int goldfish_tty_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id goldfish_tty_of_match[] = {
+	{ .compatible = "google,goldfish-tty", },
+	{},
+};
+
+MODULE_DEVICE_TABLE(of, goldfish_tty_of_match);
+
 static struct platform_driver goldfish_tty_platform_driver = {
 	.probe = goldfish_tty_probe,
 	.remove = goldfish_tty_remove,
 	.driver = {
-		.name = "goldfish_tty"
+		.name = "goldfish_tty",
+		.of_match_table = goldfish_tty_of_match,
 	}
 };
 
diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c
index def5199ca004..62fe36858a75 100644
--- a/drivers/tty/serial/serial_core.c
+++ b/drivers/tty/serial/serial_core.c
@@ -95,6 +95,9 @@ static void __uart_start(struct tty_struct *tty)
 	struct uart_state *state = tty->driver_data;
 	struct uart_port *port = state->uart_port;
 
+	if (port->ops->wake_peer)
+		port->ops->wake_peer(port);
+
 	if (!uart_tx_stopped(port))
 		port->ops->start_tx(port);
 }
diff --git a/drivers/usb/dwc3/core.h b/drivers/usb/dwc3/core.h
index 68d11d7d4028..d5b604a2c325 100644
--- a/drivers/usb/dwc3/core.h
+++ b/drivers/usb/dwc3/core.h
@@ -37,6 +37,7 @@
 #define DWC3_MSG_MAX	500
 
 /* Global constants */
+#define DWC3_ZLP_BUF_SIZE	1024	/* size of a superspeed bulk */
 #define DWC3_EP0_BOUNCE_SIZE	512
 #define DWC3_ENDPOINTS_NUM	32
 #define DWC3_XHCI_RESOURCES_NUM	2
@@ -645,6 +646,7 @@ struct dwc3_scratchpad_array {
  * @ctrl_req: usb control request which is used for ep0
  * @ep0_trb: trb which is used for the ctrl_req
  * @ep0_bounce: bounce buffer for ep0
+ * @zlp_buf: used when request->zero is set
  * @setup_buf: used while precessing STD USB requests
  * @ctrl_req_addr: dma address of ctrl_req
  * @ep0_trb: dma address of ep0_trb
@@ -732,6 +734,7 @@ struct dwc3 {
 	struct usb_ctrlrequest	*ctrl_req;
 	struct dwc3_trb		*ep0_trb;
 	void			*ep0_bounce;
+	void			*zlp_buf;
 	void			*scratchbuf;
 	u8			*setup_buf;
 	dma_addr_t		ctrl_req_addr;
diff --git a/drivers/usb/dwc3/gadget.c b/drivers/usb/dwc3/gadget.c
index d3bd1afd6302..3368c3a88a57 100644
--- a/drivers/usb/dwc3/gadget.c
+++ b/drivers/usb/dwc3/gadget.c
@@ -1200,6 +1200,32 @@ out:
 	return ret;
 }
 
+static void __dwc3_gadget_ep_zlp_complete(struct usb_ep *ep,
+		struct usb_request *request)
+{
+	dwc3_gadget_ep_free_request(ep, request);
+}
+
+static int __dwc3_gadget_ep_queue_zlp(struct dwc3 *dwc, struct dwc3_ep *dep)
+{
+	struct dwc3_request		*req;
+	struct usb_request		*request;
+	struct usb_ep			*ep = &dep->endpoint;
+
+	dwc3_trace(trace_dwc3_gadget, "queueing ZLP\n");
+	request = dwc3_gadget_ep_alloc_request(ep, GFP_ATOMIC);
+	if (!request)
+		return -ENOMEM;
+
+	request->length = 0;
+	request->buf = dwc->zlp_buf;
+	request->complete = __dwc3_gadget_ep_zlp_complete;
+
+	req = to_dwc3_request(request);
+
+	return __dwc3_gadget_ep_queue(dep, req);
+}
+
 static int dwc3_gadget_ep_queue(struct usb_ep *ep, struct usb_request *request,
 	gfp_t gfp_flags)
 {
@@ -1227,6 +1253,16 @@ static int dwc3_gadget_ep_queue(struct usb_ep *ep, struct usb_request *request,
 
 	ret = __dwc3_gadget_ep_queue(dep, req);
 
+	/*
+	 * Okay, here's the thing, if gadget driver has requested for a ZLP by
+	 * setting request->zero, instead of doing magic, we will just queue an
+	 * extra usb_request ourselves so that it gets handled the same way as
+	 * any other request.
+	 */
+	if (ret == 0 && request->zero && request->length &&
+	    (request->length % ep->maxpacket == 0))
+		ret = __dwc3_gadget_ep_queue_zlp(dwc, dep);
+
 out:
 	spin_unlock_irqrestore(&dwc->lock, flags);
 
@@ -2798,6 +2834,12 @@ int dwc3_gadget_init(struct dwc3 *dwc)
 		goto err3;
 	}
 
+	dwc->zlp_buf = kzalloc(DWC3_ZLP_BUF_SIZE, GFP_KERNEL);
+	if (!dwc->zlp_buf) {
+		ret = -ENOMEM;
+		goto err4;
+	}
+
 	dwc->gadget.ops			= &dwc3_gadget_ops;
 	dwc->gadget.speed		= USB_SPEED_UNKNOWN;
 	dwc->gadget.sg_supported	= true;
@@ -2839,16 +2881,19 @@ int dwc3_gadget_init(struct dwc3 *dwc)
 
 	ret = dwc3_gadget_init_endpoints(dwc);
 	if (ret)
-		goto err4;
+		goto err5;
 
 	ret = usb_add_gadget_udc(dwc->dev, &dwc->gadget);
 	if (ret) {
 		dev_err(dwc->dev, "failed to register udc\n");
-		goto err4;
+		goto err5;
 	}
 
 	return 0;
 
+err5:
+	kfree(dwc->zlp_buf);
+
 err4:
 	dwc3_gadget_free_endpoints(dwc);
 	dma_free_coherent(dwc->dev, DWC3_EP0_BOUNCE_SIZE,
@@ -2881,6 +2926,7 @@ void dwc3_gadget_exit(struct dwc3 *dwc)
 			dwc->ep0_bounce, dwc->ep0_bounce_addr);
 
 	kfree(dwc->setup_buf);
+	kfree(dwc->zlp_buf);
 
 	dma_free_coherent(dwc->dev, sizeof(*dwc->ep0_trb) * 2,
 			dwc->ep0_trb, dwc->ep0_trb_addr);
diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig
index 33834aa09ed4..5cf6802b02ae 100644
--- a/drivers/usb/gadget/Kconfig
+++ b/drivers/usb/gadget/Kconfig
@@ -199,6 +199,18 @@ config USB_F_HID
 config USB_F_PRINTER
 	tristate
 
+config USB_F_MTP
+	tristate
+
+config USB_F_PTP
+        tristate
+
+config USB_F_AUDIO_SRC
+	tristate
+
+config USB_F_ACC
+	tristate
+
 choice
 	tristate "USB Gadget Drivers"
 	default USB_ETH
@@ -371,6 +383,44 @@ config USB_CONFIGFS_F_FS
 	  implemented in kernel space (for instance Ethernet, serial or
 	  mass storage) and other are implemented in user space.
 
+config USB_CONFIGFS_F_MTP
+	boolean "MTP gadget"
+	depends on USB_CONFIGFS
+	select USB_F_MTP
+	help
+	  USB gadget MTP support
+
+config USB_CONFIGFS_F_PTP
+	boolean "PTP gadget"
+	depends on USB_CONFIGFS && USB_CONFIGFS_F_MTP
+	select USB_F_PTP
+	help
+	  USB gadget PTP support
+
+config USB_CONFIGFS_F_ACC
+	boolean "Accessory gadget"
+	depends on USB_CONFIGFS
+	select USB_F_ACC
+	help
+	  USB gadget Accessory support
+
+config USB_CONFIGFS_F_AUDIO_SRC
+	boolean "Audio Source gadget"
+	depends on USB_CONFIGFS && USB_CONFIGFS_F_ACC
+	depends on SND
+	select SND_PCM
+	select USB_F_AUDIO_SRC
+	help
+	  USB gadget Audio Source support
+
+config USB_CONFIGFS_UEVENT
+	boolean "Uevent notification of Gadget state"
+	depends on USB_CONFIGFS
+	help
+	  Enable uevent notifications to userspace when the gadget
+	  state changes. The gadget can be in any of the following
+	  three states: "CONNECTED/DISCONNECTED/CONFIGURED"
+
 config USB_CONFIGFS_F_UAC1
 	bool "Audio Class 1.0"
 	depends on USB_CONFIGFS
diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c
index d186d0282a38..a1fccfeb75d3 100644
--- a/drivers/usb/gadget/composite.c
+++ b/drivers/usb/gadget/composite.c
@@ -1879,6 +1879,12 @@ void composite_disconnect(struct usb_gadget *gadget)
 	struct usb_composite_dev	*cdev = get_gadget_data(gadget);
 	unsigned long			flags;
 
+	if (cdev == NULL) {
+		WARN(1, "%s: Calling disconnect on a Gadget that is \
+			 not connected\n", __func__);
+		return;
+	}
+
 	/* REVISIT:  should we have config and device level
 	 * disconnect callbacks?
 	 */
diff --git a/drivers/usb/gadget/configfs.c b/drivers/usb/gadget/configfs.c
index 163d305e1200..a0bd3c542d06 100644
--- a/drivers/usb/gadget/configfs.c
+++ b/drivers/usb/gadget/configfs.c
@@ -9,6 +9,31 @@
 #include "u_f.h"
 #include "u_os_desc.h"
 
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+#include <linux/platform_device.h>
+#include <linux/kdev_t.h>
+#include <linux/usb/ch9.h>
+
+#ifdef CONFIG_USB_CONFIGFS_F_ACC
+extern int acc_ctrlrequest(struct usb_composite_dev *cdev,
+				const struct usb_ctrlrequest *ctrl);
+void acc_disconnect(void);
+#endif
+static struct class *android_class;
+static struct device *android_device;
+static int index;
+
+struct device *create_function_device(char *name)
+{
+	if (android_device && !IS_ERR(android_device))
+		return device_create(android_class, android_device,
+			MKDEV(0, index++), NULL, name);
+	else
+		return ERR_PTR(-EINVAL);
+}
+EXPORT_SYMBOL_GPL(create_function_device);
+#endif
+
 int check_user_usb_string(const char *name,
 		struct usb_gadget_strings *stringtab_dev)
 {
@@ -62,6 +87,12 @@ struct gadget_info {
 	bool use_os_desc;
 	char b_vendor_code;
 	char qw_sign[OS_STRING_QW_SIGN_LEN];
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+	bool connected;
+	bool sw_connected;
+	struct work_struct work;
+	struct device *dev;
+#endif
 };
 
 static inline struct gadget_info *to_gadget_info(struct config_item *item)
@@ -266,7 +297,7 @@ static ssize_t gadget_dev_desc_UDC_store(struct config_item *item,
 
 	mutex_lock(&gi->lock);
 
-	if (!strlen(name)) {
+	if (!strlen(name) || strcmp(name, "none") == 0) {
 		ret = unregister_gadget(gi);
 		if (ret)
 			goto err;
@@ -1373,6 +1404,60 @@ err_comp_cleanup:
 	return ret;
 }
 
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+static void android_work(struct work_struct *data)
+{
+	struct gadget_info *gi = container_of(data, struct gadget_info, work);
+	struct usb_composite_dev *cdev = &gi->cdev;
+	char *disconnected[2] = { "USB_STATE=DISCONNECTED", NULL };
+	char *connected[2]    = { "USB_STATE=CONNECTED", NULL };
+	char *configured[2]   = { "USB_STATE=CONFIGURED", NULL };
+	/* 0-connected 1-configured 2-disconnected*/
+	bool status[3] = { false, false, false };
+	unsigned long flags;
+	bool uevent_sent = false;
+
+	spin_lock_irqsave(&cdev->lock, flags);
+	if (cdev->config)
+		status[1] = true;
+
+	if (gi->connected != gi->sw_connected) {
+		if (gi->connected)
+			status[0] = true;
+		else
+			status[2] = true;
+		gi->sw_connected = gi->connected;
+	}
+	spin_unlock_irqrestore(&cdev->lock, flags);
+
+	if (status[0]) {
+		kobject_uevent_env(&android_device->kobj,
+					KOBJ_CHANGE, connected);
+		pr_info("%s: sent uevent %s\n", __func__, connected[0]);
+		uevent_sent = true;
+	}
+
+	if (status[1]) {
+		kobject_uevent_env(&android_device->kobj,
+					KOBJ_CHANGE, configured);
+		pr_info("%s: sent uevent %s\n", __func__, configured[0]);
+		uevent_sent = true;
+	}
+
+	if (status[2]) {
+		kobject_uevent_env(&android_device->kobj,
+					KOBJ_CHANGE, disconnected);
+		pr_info("%s: sent uevent %s\n", __func__, disconnected[0]);
+		uevent_sent = true;
+	}
+
+	if (!uevent_sent) {
+		pr_info("%s: did not send uevent (%d %d %p)\n", __func__,
+			gi->connected, gi->sw_connected, cdev->config);
+	}
+}
+#endif
+
 static void configfs_composite_unbind(struct usb_gadget *gadget)
 {
 	struct usb_composite_dev	*cdev;
@@ -1392,14 +1477,91 @@ static void configfs_composite_unbind(struct usb_gadget *gadget)
 	set_gadget_data(gadget, NULL);
 }
 
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+static int android_setup(struct usb_gadget *gadget,
+			const struct usb_ctrlrequest *c)
+{
+	struct usb_composite_dev *cdev = get_gadget_data(gadget);
+	unsigned long flags;
+	struct gadget_info *gi = container_of(cdev, struct gadget_info, cdev);
+	int value = -EOPNOTSUPP;
+	struct usb_function_instance *fi;
+
+	spin_lock_irqsave(&cdev->lock, flags);
+	if (!gi->connected) {
+		gi->connected = 1;
+		schedule_work(&gi->work);
+	}
+	spin_unlock_irqrestore(&cdev->lock, flags);
+	list_for_each_entry(fi, &gi->available_func, cfs_list) {
+		if (fi != NULL && fi->f != NULL && fi->f->setup != NULL) {
+			value = fi->f->setup(fi->f, c);
+			if (value >= 0)
+				break;
+		}
+	}
+
+#ifdef CONFIG_USB_CONFIGFS_F_ACC
+	if (value < 0)
+		value = acc_ctrlrequest(cdev, c);
+#endif
+
+	if (value < 0)
+		value = composite_setup(gadget, c);
+
+	spin_lock_irqsave(&cdev->lock, flags);
+	if (c->bRequest == USB_REQ_SET_CONFIGURATION &&
+						cdev->config) {
+		schedule_work(&gi->work);
+	}
+	spin_unlock_irqrestore(&cdev->lock, flags);
+
+	return value;
+}
+
+static void android_disconnect(struct usb_gadget *gadget)
+{
+	struct usb_composite_dev        *cdev = get_gadget_data(gadget);
+	struct gadget_info *gi = container_of(cdev, struct gadget_info, cdev);
+
+	/* FIXME: There's a race between usb_gadget_udc_stop() which is likely
+	 * to set the gadget driver to NULL in the udc driver and this drivers
+	 * gadget disconnect fn which likely checks for the gadget driver to
+	 * be a null ptr. It happens that unbind (doing set_gadget_data(NULL))
+	 * is called before the gadget driver is set to NULL and the udc driver
+	 * calls disconnect fn which results in cdev being a null ptr.
+	 */
+	if (cdev == NULL) {
+		WARN(1, "%s: gadget driver already disconnected\n", __func__);
+		return;
+	}
+
+	/* accessory HID support can be active while the
+		accessory function is not actually enabled,
+		so we need to inform it when we are disconnected.
+	*/
+
+#ifdef CONFIG_USB_CONFIGFS_F_ACC
+	acc_disconnect();
+#endif
+	gi->connected = 0;
+	schedule_work(&gi->work);
+	composite_disconnect(gadget);
+}
+#endif
+
 static const struct usb_gadget_driver configfs_driver_template = {
 	.bind           = configfs_composite_bind,
 	.unbind         = configfs_composite_unbind,
-
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+	.setup          = android_setup,
+	.reset          = android_disconnect,
+	.disconnect     = android_disconnect,
+#else
 	.setup          = composite_setup,
 	.reset          = composite_disconnect,
 	.disconnect     = composite_disconnect,
-
+#endif
 	.suspend	= composite_suspend,
 	.resume		= composite_resume,
 
@@ -1410,6 +1572,89 @@ static const struct usb_gadget_driver configfs_driver_template = {
 	},
 };
 
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+static ssize_t state_show(struct device *pdev, struct device_attribute *attr,
+			char *buf)
+{
+	struct gadget_info *dev = dev_get_drvdata(pdev);
+	struct usb_composite_dev *cdev;
+	char *state = "DISCONNECTED";
+	unsigned long flags;
+
+	if (!dev)
+		goto out;
+
+	cdev = &dev->cdev;
+
+	if (!cdev)
+		goto out;
+
+	spin_lock_irqsave(&cdev->lock, flags);
+	if (cdev->config)
+		state = "CONFIGURED";
+	else if (dev->connected)
+		state = "CONNECTED";
+	spin_unlock_irqrestore(&cdev->lock, flags);
+out:
+	return sprintf(buf, "%s\n", state);
+}
+
+static DEVICE_ATTR(state, S_IRUGO, state_show, NULL);
+
+static struct device_attribute *android_usb_attributes[] = {
+	&dev_attr_state,
+	NULL
+};
+
+static int android_device_create(struct gadget_info *gi)
+{
+	struct device_attribute **attrs;
+	struct device_attribute *attr;
+
+	INIT_WORK(&gi->work, android_work);
+	android_device = device_create(android_class, NULL,
+				MKDEV(0, 0), NULL, "android0");
+	if (IS_ERR(android_device))
+		return PTR_ERR(android_device);
+
+	dev_set_drvdata(android_device, gi);
+
+	attrs = android_usb_attributes;
+	while ((attr = *attrs++)) {
+		int err;
+
+		err = device_create_file(android_device, attr);
+		if (err) {
+			device_destroy(android_device->class,
+				       android_device->devt);
+			return err;
+		}
+	}
+
+	return 0;
+}
+
+static void android_device_destroy(void)
+{
+	struct device_attribute **attrs;
+	struct device_attribute *attr;
+
+	attrs = android_usb_attributes;
+	while ((attr = *attrs++))
+		device_remove_file(android_device, attr);
+	device_destroy(android_device->class, android_device->devt);
+}
+#else
+static inline int android_device_create(struct gadget_info *gi)
+{
+	return 0;
+}
+
+static inline void android_device_destroy(void)
+{
+}
+#endif
+
 static struct config_group *gadgets_make(
 		struct config_group *group,
 		const char *name)
@@ -1419,7 +1664,6 @@ static struct config_group *gadgets_make(
 	gi = kzalloc(sizeof(*gi), GFP_KERNEL);
 	if (!gi)
 		return ERR_PTR(-ENOMEM);
-
 	gi->group.default_groups = gi->default_groups;
 	gi->group.default_groups[0] = &gi->functions_group;
 	gi->group.default_groups[1] = &gi->configs_group;
@@ -1458,9 +1702,13 @@ static struct config_group *gadgets_make(
 	if (!gi->composite.gadget_driver.function)
 		goto err;
 
+	if (android_device_create(gi) < 0)
+		goto err;
+
 	config_group_init_type_name(&gi->group, name,
 				&gadget_root_type);
 	return &gi->group;
+
 err:
 	kfree(gi);
 	return ERR_PTR(-ENOMEM);
@@ -1469,6 +1717,7 @@ err:
 static void gadgets_drop(struct config_group *group, struct config_item *item)
 {
 	config_item_put(item);
+	android_device_destroy();
 }
 
 static struct configfs_group_operations gadgets_ops = {
@@ -1495,7 +1744,9 @@ void unregister_gadget_item(struct config_item *item)
 {
 	struct gadget_info *gi = to_gadget_info(item);
 
+	mutex_lock(&gi->lock);
 	unregister_gadget(gi);
+	mutex_unlock(&gi->lock);
 }
 EXPORT_SYMBOL_GPL(unregister_gadget_item);
 
@@ -1506,6 +1757,13 @@ static int __init gadget_cfs_init(void)
 	config_group_init(&gadget_subsys.su_group);
 
 	ret = configfs_register_subsystem(&gadget_subsys);
+
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+	android_class = class_create(THIS_MODULE, "android_usb");
+	if (IS_ERR(android_class))
+		return PTR_ERR(android_class);
+#endif
+
 	return ret;
 }
 module_init(gadget_cfs_init);
@@ -1513,5 +1771,10 @@ module_init(gadget_cfs_init);
 static void __exit gadget_cfs_exit(void)
 {
 	configfs_unregister_subsystem(&gadget_subsys);
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+	if (!IS_ERR(android_class))
+		class_destroy(android_class);
+#endif
+
 }
 module_exit(gadget_cfs_exit);
diff --git a/drivers/usb/gadget/function/Makefile b/drivers/usb/gadget/function/Makefile
index bd7def576955..1cd544beef63 100644
--- a/drivers/usb/gadget/function/Makefile
+++ b/drivers/usb/gadget/function/Makefile
@@ -44,3 +44,11 @@ usb_f_hid-y			:= f_hid.o
 obj-$(CONFIG_USB_F_HID)		+= usb_f_hid.o
 usb_f_printer-y			:= f_printer.o
 obj-$(CONFIG_USB_F_PRINTER)	+= usb_f_printer.o
+usb_f_mtp-y                     := f_mtp.o
+obj-$(CONFIG_USB_F_MTP)         += usb_f_mtp.o
+usb_f_ptp-y                     := f_ptp.o
+obj-$(CONFIG_USB_F_PTP)         += usb_f_ptp.o
+usb_f_audio_source-y            := f_audio_source.o
+obj-$(CONFIG_USB_F_AUDIO_SRC)   += usb_f_audio_source.o
+usb_f_accessory-y               := f_accessory.o
+obj-$(CONFIG_USB_F_ACC)         += usb_f_accessory.o
diff --git a/drivers/usb/gadget/function/f_accessory.c b/drivers/usb/gadget/function/f_accessory.c
new file mode 100644
index 000000000000..d31c0809046f
--- /dev/null
+++ b/drivers/usb/gadget/function/f_accessory.c
@@ -0,0 +1,1353 @@
+/*
+ * Gadget Function Driver for Android USB accessories
+ *
+ * Copyright (C) 2011 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+/* #define DEBUG */
+/* #define VERBOSE_DEBUG */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/poll.h>
+#include <linux/delay.h>
+#include <linux/wait.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+#include <linux/kthread.h>
+#include <linux/freezer.h>
+
+#include <linux/types.h>
+#include <linux/file.h>
+#include <linux/device.h>
+#include <linux/miscdevice.h>
+
+#include <linux/hid.h>
+#include <linux/hiddev.h>
+#include <linux/usb.h>
+#include <linux/usb/ch9.h>
+#include <linux/usb/f_accessory.h>
+
+#include <linux/configfs.h>
+#include <linux/usb/composite.h>
+
+#define MAX_INST_NAME_LEN        40
+#define BULK_BUFFER_SIZE    16384
+#define ACC_STRING_SIZE     256
+
+#define PROTOCOL_VERSION    2
+
+/* String IDs */
+#define INTERFACE_STRING_INDEX	0
+
+/* number of tx and rx requests to allocate */
+#define TX_REQ_MAX 4
+#define RX_REQ_MAX 2
+
+struct acc_hid_dev {
+	struct list_head	list;
+	struct hid_device *hid;
+	struct acc_dev *dev;
+	/* accessory defined ID */
+	int id;
+	/* HID report descriptor */
+	u8 *report_desc;
+	/* length of HID report descriptor */
+	int report_desc_len;
+	/* number of bytes of report_desc we have received so far */
+	int report_desc_offset;
+};
+
+struct acc_dev {
+	struct usb_function function;
+	struct usb_composite_dev *cdev;
+	spinlock_t lock;
+
+	struct usb_ep *ep_in;
+	struct usb_ep *ep_out;
+
+	/* online indicates state of function_set_alt & function_unbind
+	 * set to 1 when we connect
+	 */
+	int online:1;
+
+	/* disconnected indicates state of open & release
+	 * Set to 1 when we disconnect.
+	 * Not cleared until our file is closed.
+	 */
+	int disconnected:1;
+
+	/* strings sent by the host */
+	char manufacturer[ACC_STRING_SIZE];
+	char model[ACC_STRING_SIZE];
+	char description[ACC_STRING_SIZE];
+	char version[ACC_STRING_SIZE];
+	char uri[ACC_STRING_SIZE];
+	char serial[ACC_STRING_SIZE];
+
+	/* for acc_complete_set_string */
+	int string_index;
+
+	/* set to 1 if we have a pending start request */
+	int start_requested;
+
+	int audio_mode;
+
+	/* synchronize access to our device file */
+	atomic_t open_excl;
+
+	struct list_head tx_idle;
+
+	wait_queue_head_t read_wq;
+	wait_queue_head_t write_wq;
+	struct usb_request *rx_req[RX_REQ_MAX];
+	int rx_done;
+
+	/* delayed work for handling ACCESSORY_START */
+	struct delayed_work start_work;
+
+	/* worker for registering and unregistering hid devices */
+	struct work_struct hid_work;
+
+	/* list of active HID devices */
+	struct list_head	hid_list;
+
+	/* list of new HID devices to register */
+	struct list_head	new_hid_list;
+
+	/* list of dead HID devices to unregister */
+	struct list_head	dead_hid_list;
+};
+
+static struct usb_interface_descriptor acc_interface_desc = {
+	.bLength                = USB_DT_INTERFACE_SIZE,
+	.bDescriptorType        = USB_DT_INTERFACE,
+	.bInterfaceNumber       = 0,
+	.bNumEndpoints          = 2,
+	.bInterfaceClass        = USB_CLASS_VENDOR_SPEC,
+	.bInterfaceSubClass     = USB_SUBCLASS_VENDOR_SPEC,
+	.bInterfaceProtocol     = 0,
+};
+
+static struct usb_endpoint_descriptor acc_highspeed_in_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_IN,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+	.wMaxPacketSize         = __constant_cpu_to_le16(512),
+};
+
+static struct usb_endpoint_descriptor acc_highspeed_out_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_OUT,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+	.wMaxPacketSize         = __constant_cpu_to_le16(512),
+};
+
+static struct usb_endpoint_descriptor acc_fullspeed_in_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_IN,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+};
+
+static struct usb_endpoint_descriptor acc_fullspeed_out_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_OUT,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+};
+
+static struct usb_descriptor_header *fs_acc_descs[] = {
+	(struct usb_descriptor_header *) &acc_interface_desc,
+	(struct usb_descriptor_header *) &acc_fullspeed_in_desc,
+	(struct usb_descriptor_header *) &acc_fullspeed_out_desc,
+	NULL,
+};
+
+static struct usb_descriptor_header *hs_acc_descs[] = {
+	(struct usb_descriptor_header *) &acc_interface_desc,
+	(struct usb_descriptor_header *) &acc_highspeed_in_desc,
+	(struct usb_descriptor_header *) &acc_highspeed_out_desc,
+	NULL,
+};
+
+static struct usb_string acc_string_defs[] = {
+	[INTERFACE_STRING_INDEX].s	= "Android Accessory Interface",
+	{  },	/* end of list */
+};
+
+static struct usb_gadget_strings acc_string_table = {
+	.language		= 0x0409,	/* en-US */
+	.strings		= acc_string_defs,
+};
+
+static struct usb_gadget_strings *acc_strings[] = {
+	&acc_string_table,
+	NULL,
+};
+
+/* temporary variable used between acc_open() and acc_gadget_bind() */
+static struct acc_dev *_acc_dev;
+
+struct acc_instance {
+	struct usb_function_instance func_inst;
+	const char *name;
+};
+
+static inline struct acc_dev *func_to_dev(struct usb_function *f)
+{
+	return container_of(f, struct acc_dev, function);
+}
+
+static struct usb_request *acc_request_new(struct usb_ep *ep, int buffer_size)
+{
+	struct usb_request *req = usb_ep_alloc_request(ep, GFP_KERNEL);
+
+	if (!req)
+		return NULL;
+
+	/* now allocate buffers for the requests */
+	req->buf = kmalloc(buffer_size, GFP_KERNEL);
+	if (!req->buf) {
+		usb_ep_free_request(ep, req);
+		return NULL;
+	}
+
+	return req;
+}
+
+static void acc_request_free(struct usb_request *req, struct usb_ep *ep)
+{
+	if (req) {
+		kfree(req->buf);
+		usb_ep_free_request(ep, req);
+	}
+}
+
+/* add a request to the tail of a list */
+static void req_put(struct acc_dev *dev, struct list_head *head,
+		struct usb_request *req)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	list_add_tail(&req->list, head);
+	spin_unlock_irqrestore(&dev->lock, flags);
+}
+
+/* remove a request from the head of a list */
+static struct usb_request *req_get(struct acc_dev *dev, struct list_head *head)
+{
+	unsigned long flags;
+	struct usb_request *req;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	if (list_empty(head)) {
+		req = 0;
+	} else {
+		req = list_first_entry(head, struct usb_request, list);
+		list_del(&req->list);
+	}
+	spin_unlock_irqrestore(&dev->lock, flags);
+	return req;
+}
+
+static void acc_set_disconnected(struct acc_dev *dev)
+{
+	dev->disconnected = 1;
+}
+
+static void acc_complete_in(struct usb_ep *ep, struct usb_request *req)
+{
+	struct acc_dev *dev = _acc_dev;
+
+	if (req->status == -ESHUTDOWN) {
+		pr_debug("acc_complete_in set disconnected");
+		acc_set_disconnected(dev);
+	}
+
+	req_put(dev, &dev->tx_idle, req);
+
+	wake_up(&dev->write_wq);
+}
+
+static void acc_complete_out(struct usb_ep *ep, struct usb_request *req)
+{
+	struct acc_dev *dev = _acc_dev;
+
+	dev->rx_done = 1;
+	if (req->status == -ESHUTDOWN) {
+		pr_debug("acc_complete_out set disconnected");
+		acc_set_disconnected(dev);
+	}
+
+	wake_up(&dev->read_wq);
+}
+
+static void acc_complete_set_string(struct usb_ep *ep, struct usb_request *req)
+{
+	struct acc_dev	*dev = ep->driver_data;
+	char *string_dest = NULL;
+	int length = req->actual;
+
+	if (req->status != 0) {
+		pr_err("acc_complete_set_string, err %d\n", req->status);
+		return;
+	}
+
+	switch (dev->string_index) {
+	case ACCESSORY_STRING_MANUFACTURER:
+		string_dest = dev->manufacturer;
+		break;
+	case ACCESSORY_STRING_MODEL:
+		string_dest = dev->model;
+		break;
+	case ACCESSORY_STRING_DESCRIPTION:
+		string_dest = dev->description;
+		break;
+	case ACCESSORY_STRING_VERSION:
+		string_dest = dev->version;
+		break;
+	case ACCESSORY_STRING_URI:
+		string_dest = dev->uri;
+		break;
+	case ACCESSORY_STRING_SERIAL:
+		string_dest = dev->serial;
+		break;
+	}
+	if (string_dest) {
+		unsigned long flags;
+
+		if (length >= ACC_STRING_SIZE)
+			length = ACC_STRING_SIZE - 1;
+
+		spin_lock_irqsave(&dev->lock, flags);
+		memcpy(string_dest, req->buf, length);
+		/* ensure zero termination */
+		string_dest[length] = 0;
+		spin_unlock_irqrestore(&dev->lock, flags);
+	} else {
+		pr_err("unknown accessory string index %d\n",
+			dev->string_index);
+	}
+}
+
+static void acc_complete_set_hid_report_desc(struct usb_ep *ep,
+		struct usb_request *req)
+{
+	struct acc_hid_dev *hid = req->context;
+	struct acc_dev *dev = hid->dev;
+	int length = req->actual;
+
+	if (req->status != 0) {
+		pr_err("acc_complete_set_hid_report_desc, err %d\n",
+			req->status);
+		return;
+	}
+
+	memcpy(hid->report_desc + hid->report_desc_offset, req->buf, length);
+	hid->report_desc_offset += length;
+	if (hid->report_desc_offset == hid->report_desc_len) {
+		/* After we have received the entire report descriptor
+		 * we schedule work to initialize the HID device
+		 */
+		schedule_work(&dev->hid_work);
+	}
+}
+
+static void acc_complete_send_hid_event(struct usb_ep *ep,
+		struct usb_request *req)
+{
+	struct acc_hid_dev *hid = req->context;
+	int length = req->actual;
+
+	if (req->status != 0) {
+		pr_err("acc_complete_send_hid_event, err %d\n", req->status);
+		return;
+	}
+
+	hid_report_raw_event(hid->hid, HID_INPUT_REPORT, req->buf, length, 1);
+}
+
+static int acc_hid_parse(struct hid_device *hid)
+{
+	struct acc_hid_dev *hdev = hid->driver_data;
+
+	hid_parse_report(hid, hdev->report_desc, hdev->report_desc_len);
+	return 0;
+}
+
+static int acc_hid_start(struct hid_device *hid)
+{
+	return 0;
+}
+
+static void acc_hid_stop(struct hid_device *hid)
+{
+}
+
+static int acc_hid_open(struct hid_device *hid)
+{
+	return 0;
+}
+
+static void acc_hid_close(struct hid_device *hid)
+{
+}
+
+static int acc_hid_raw_request(struct hid_device *hid, unsigned char reportnum,
+	__u8 *buf, size_t len, unsigned char rtype, int reqtype)
+{
+	return 0;
+}
+
+static struct hid_ll_driver acc_hid_ll_driver = {
+	.parse = acc_hid_parse,
+	.start = acc_hid_start,
+	.stop = acc_hid_stop,
+	.open = acc_hid_open,
+	.close = acc_hid_close,
+	.raw_request = acc_hid_raw_request,
+};
+
+static struct acc_hid_dev *acc_hid_new(struct acc_dev *dev,
+		int id, int desc_len)
+{
+	struct acc_hid_dev *hdev;
+
+	hdev = kzalloc(sizeof(*hdev), GFP_ATOMIC);
+	if (!hdev)
+		return NULL;
+	hdev->report_desc = kzalloc(desc_len, GFP_ATOMIC);
+	if (!hdev->report_desc) {
+		kfree(hdev);
+		return NULL;
+	}
+	hdev->dev = dev;
+	hdev->id = id;
+	hdev->report_desc_len = desc_len;
+
+	return hdev;
+}
+
+static struct acc_hid_dev *acc_hid_get(struct list_head *list, int id)
+{
+	struct acc_hid_dev *hid;
+
+	list_for_each_entry(hid, list, list) {
+		if (hid->id == id)
+			return hid;
+	}
+	return NULL;
+}
+
+static int acc_register_hid(struct acc_dev *dev, int id, int desc_length)
+{
+	struct acc_hid_dev *hid;
+	unsigned long flags;
+
+	/* report descriptor length must be > 0 */
+	if (desc_length <= 0)
+		return -EINVAL;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	/* replace HID if one already exists with this ID */
+	hid = acc_hid_get(&dev->hid_list, id);
+	if (!hid)
+		hid = acc_hid_get(&dev->new_hid_list, id);
+	if (hid)
+		list_move(&hid->list, &dev->dead_hid_list);
+
+	hid = acc_hid_new(dev, id, desc_length);
+	if (!hid) {
+		spin_unlock_irqrestore(&dev->lock, flags);
+		return -ENOMEM;
+	}
+
+	list_add(&hid->list, &dev->new_hid_list);
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	/* schedule work to register the HID device */
+	schedule_work(&dev->hid_work);
+	return 0;
+}
+
+static int acc_unregister_hid(struct acc_dev *dev, int id)
+{
+	struct acc_hid_dev *hid;
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	hid = acc_hid_get(&dev->hid_list, id);
+	if (!hid)
+		hid = acc_hid_get(&dev->new_hid_list, id);
+	if (!hid) {
+		spin_unlock_irqrestore(&dev->lock, flags);
+		return -EINVAL;
+	}
+
+	list_move(&hid->list, &dev->dead_hid_list);
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	schedule_work(&dev->hid_work);
+	return 0;
+}
+
+static int create_bulk_endpoints(struct acc_dev *dev,
+				struct usb_endpoint_descriptor *in_desc,
+				struct usb_endpoint_descriptor *out_desc)
+{
+	struct usb_composite_dev *cdev = dev->cdev;
+	struct usb_request *req;
+	struct usb_ep *ep;
+	int i;
+
+	DBG(cdev, "create_bulk_endpoints dev: %p\n", dev);
+
+	ep = usb_ep_autoconfig(cdev->gadget, in_desc);
+	if (!ep) {
+		DBG(cdev, "usb_ep_autoconfig for ep_in failed\n");
+		return -ENODEV;
+	}
+	DBG(cdev, "usb_ep_autoconfig for ep_in got %s\n", ep->name);
+	ep->driver_data = dev;		/* claim the endpoint */
+	dev->ep_in = ep;
+
+	ep = usb_ep_autoconfig(cdev->gadget, out_desc);
+	if (!ep) {
+		DBG(cdev, "usb_ep_autoconfig for ep_out failed\n");
+		return -ENODEV;
+	}
+	DBG(cdev, "usb_ep_autoconfig for ep_out got %s\n", ep->name);
+	ep->driver_data = dev;		/* claim the endpoint */
+	dev->ep_out = ep;
+
+	/* now allocate requests for our endpoints */
+	for (i = 0; i < TX_REQ_MAX; i++) {
+		req = acc_request_new(dev->ep_in, BULK_BUFFER_SIZE);
+		if (!req)
+			goto fail;
+		req->complete = acc_complete_in;
+		req_put(dev, &dev->tx_idle, req);
+	}
+	for (i = 0; i < RX_REQ_MAX; i++) {
+		req = acc_request_new(dev->ep_out, BULK_BUFFER_SIZE);
+		if (!req)
+			goto fail;
+		req->complete = acc_complete_out;
+		dev->rx_req[i] = req;
+	}
+
+	return 0;
+
+fail:
+	pr_err("acc_bind() could not allocate requests\n");
+	while ((req = req_get(dev, &dev->tx_idle)))
+		acc_request_free(req, dev->ep_in);
+	for (i = 0; i < RX_REQ_MAX; i++)
+		acc_request_free(dev->rx_req[i], dev->ep_out);
+	return -1;
+}
+
+static ssize_t acc_read(struct file *fp, char __user *buf,
+	size_t count, loff_t *pos)
+{
+	struct acc_dev *dev = fp->private_data;
+	struct usb_request *req;
+	ssize_t r = count;
+	unsigned xfer;
+	int ret = 0;
+
+	pr_debug("acc_read(%zu)\n", count);
+
+	if (dev->disconnected) {
+		pr_debug("acc_read disconnected");
+		return -ENODEV;
+	}
+
+	if (count > BULK_BUFFER_SIZE)
+		count = BULK_BUFFER_SIZE;
+
+	/* we will block until we're online */
+	pr_debug("acc_read: waiting for online\n");
+	ret = wait_event_interruptible(dev->read_wq, dev->online);
+	if (ret < 0) {
+		r = ret;
+		goto done;
+	}
+
+	if (dev->rx_done) {
+		// last req cancelled. try to get it.
+		req = dev->rx_req[0];
+		goto copy_data;
+	}
+
+requeue_req:
+	/* queue a request */
+	req = dev->rx_req[0];
+	req->length = count;
+	dev->rx_done = 0;
+	ret = usb_ep_queue(dev->ep_out, req, GFP_KERNEL);
+	if (ret < 0) {
+		r = -EIO;
+		goto done;
+	} else {
+		pr_debug("rx %p queue\n", req);
+	}
+
+	/* wait for a request to complete */
+	ret = wait_event_interruptible(dev->read_wq, dev->rx_done);
+	if (ret < 0) {
+		r = ret;
+		ret = usb_ep_dequeue(dev->ep_out, req);
+		if (ret != 0) {
+			// cancel failed. There can be a data already received.
+			// it will be retrieved in the next read.
+			pr_debug("acc_read: cancelling failed %d", ret);
+		}
+		goto done;
+	}
+
+copy_data:
+	dev->rx_done = 0;
+	if (dev->online) {
+		/* If we got a 0-len packet, throw it back and try again. */
+		if (req->actual == 0)
+			goto requeue_req;
+
+		pr_debug("rx %p %u\n", req, req->actual);
+		xfer = (req->actual < count) ? req->actual : count;
+		r = xfer;
+		if (copy_to_user(buf, req->buf, xfer))
+			r = -EFAULT;
+	} else
+		r = -EIO;
+
+done:
+	pr_debug("acc_read returning %zd\n", r);
+	return r;
+}
+
+static ssize_t acc_write(struct file *fp, const char __user *buf,
+	size_t count, loff_t *pos)
+{
+	struct acc_dev *dev = fp->private_data;
+	struct usb_request *req = 0;
+	ssize_t r = count;
+	unsigned xfer;
+	int ret;
+
+	pr_debug("acc_write(%zu)\n", count);
+
+	if (!dev->online || dev->disconnected) {
+		pr_debug("acc_write disconnected or not online");
+		return -ENODEV;
+	}
+
+	while (count > 0) {
+		if (!dev->online) {
+			pr_debug("acc_write dev->error\n");
+			r = -EIO;
+			break;
+		}
+
+		/* get an idle tx request to use */
+		req = 0;
+		ret = wait_event_interruptible(dev->write_wq,
+			((req = req_get(dev, &dev->tx_idle)) || !dev->online));
+		if (!req) {
+			r = ret;
+			break;
+		}
+
+		if (count > BULK_BUFFER_SIZE) {
+			xfer = BULK_BUFFER_SIZE;
+			/* ZLP, They will be more TX requests so not yet. */
+			req->zero = 0;
+		} else {
+			xfer = count;
+			/*
+			 * If the data length is a multple of the
+			 * maxpacket size then send a zero length packet(ZLP).
+			 */
+			req->zero = ((xfer % dev->ep_in->maxpacket) == 0);
+		}
+		if (copy_from_user(req->buf, buf, xfer)) {
+			r = -EFAULT;
+			break;
+		}
+
+		req->length = xfer;
+		ret = usb_ep_queue(dev->ep_in, req, GFP_KERNEL);
+		if (ret < 0) {
+			pr_debug("acc_write: xfer error %d\n", ret);
+			r = -EIO;
+			break;
+		}
+
+		buf += xfer;
+		count -= xfer;
+
+		/* zero this so we don't try to free it on error exit */
+		req = 0;
+	}
+
+	if (req)
+		req_put(dev, &dev->tx_idle, req);
+
+	pr_debug("acc_write returning %zd\n", r);
+	return r;
+}
+
+static long acc_ioctl(struct file *fp, unsigned code, unsigned long value)
+{
+	struct acc_dev *dev = fp->private_data;
+	char *src = NULL;
+	int ret;
+
+	switch (code) {
+	case ACCESSORY_GET_STRING_MANUFACTURER:
+		src = dev->manufacturer;
+		break;
+	case ACCESSORY_GET_STRING_MODEL:
+		src = dev->model;
+		break;
+	case ACCESSORY_GET_STRING_DESCRIPTION:
+		src = dev->description;
+		break;
+	case ACCESSORY_GET_STRING_VERSION:
+		src = dev->version;
+		break;
+	case ACCESSORY_GET_STRING_URI:
+		src = dev->uri;
+		break;
+	case ACCESSORY_GET_STRING_SERIAL:
+		src = dev->serial;
+		break;
+	case ACCESSORY_IS_START_REQUESTED:
+		return dev->start_requested;
+	case ACCESSORY_GET_AUDIO_MODE:
+		return dev->audio_mode;
+	}
+	if (!src)
+		return -EINVAL;
+
+	ret = strlen(src) + 1;
+	if (copy_to_user((void __user *)value, src, ret))
+		ret = -EFAULT;
+	return ret;
+}
+
+static int acc_open(struct inode *ip, struct file *fp)
+{
+	printk(KERN_INFO "acc_open\n");
+	if (atomic_xchg(&_acc_dev->open_excl, 1))
+		return -EBUSY;
+
+	_acc_dev->disconnected = 0;
+	fp->private_data = _acc_dev;
+	return 0;
+}
+
+static int acc_release(struct inode *ip, struct file *fp)
+{
+	printk(KERN_INFO "acc_release\n");
+
+	WARN_ON(!atomic_xchg(&_acc_dev->open_excl, 0));
+	/* indicate that we are disconnected
+	 * still could be online so don't touch online flag
+	 */
+	_acc_dev->disconnected = 1;
+	return 0;
+}
+
+/* file operations for /dev/usb_accessory */
+static const struct file_operations acc_fops = {
+	.owner = THIS_MODULE,
+	.read = acc_read,
+	.write = acc_write,
+	.unlocked_ioctl = acc_ioctl,
+	.open = acc_open,
+	.release = acc_release,
+};
+
+static int acc_hid_probe(struct hid_device *hdev,
+		const struct hid_device_id *id)
+{
+	int ret;
+
+	ret = hid_parse(hdev);
+	if (ret)
+		return ret;
+	return hid_hw_start(hdev, HID_CONNECT_DEFAULT);
+}
+
+static struct miscdevice acc_device = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = "usb_accessory",
+	.fops = &acc_fops,
+};
+
+static const struct hid_device_id acc_hid_table[] = {
+	{ HID_USB_DEVICE(HID_ANY_ID, HID_ANY_ID) },
+	{ }
+};
+
+static struct hid_driver acc_hid_driver = {
+	.name = "USB accessory",
+	.id_table = acc_hid_table,
+	.probe = acc_hid_probe,
+};
+
+static void acc_complete_setup_noop(struct usb_ep *ep, struct usb_request *req)
+{
+	/*
+	 * Default no-op function when nothing needs to be done for the
+	 * setup request
+	 */
+}
+
+int acc_ctrlrequest(struct usb_composite_dev *cdev,
+				const struct usb_ctrlrequest *ctrl)
+{
+	struct acc_dev	*dev = _acc_dev;
+	int	value = -EOPNOTSUPP;
+	struct acc_hid_dev *hid;
+	int offset;
+	u8 b_requestType = ctrl->bRequestType;
+	u8 b_request = ctrl->bRequest;
+	u16	w_index = le16_to_cpu(ctrl->wIndex);
+	u16	w_value = le16_to_cpu(ctrl->wValue);
+	u16	w_length = le16_to_cpu(ctrl->wLength);
+	unsigned long flags;
+
+/*
+ *	printk(KERN_INFO "acc_ctrlrequest "
+ *			"%02x.%02x v%04x i%04x l%u\n",
+ *			b_requestType, b_request,
+ *			w_value, w_index, w_length);
+ */
+
+	if (b_requestType == (USB_DIR_OUT | USB_TYPE_VENDOR)) {
+		if (b_request == ACCESSORY_START) {
+			dev->start_requested = 1;
+			schedule_delayed_work(
+				&dev->start_work, msecs_to_jiffies(10));
+			value = 0;
+			cdev->req->complete = acc_complete_setup_noop;
+		} else if (b_request == ACCESSORY_SEND_STRING) {
+			dev->string_index = w_index;
+			cdev->gadget->ep0->driver_data = dev;
+			cdev->req->complete = acc_complete_set_string;
+			value = w_length;
+		} else if (b_request == ACCESSORY_SET_AUDIO_MODE &&
+				w_index == 0 && w_length == 0) {
+			dev->audio_mode = w_value;
+			cdev->req->complete = acc_complete_setup_noop;
+			value = 0;
+		} else if (b_request == ACCESSORY_REGISTER_HID) {
+			cdev->req->complete = acc_complete_setup_noop;
+			value = acc_register_hid(dev, w_value, w_index);
+		} else if (b_request == ACCESSORY_UNREGISTER_HID) {
+			cdev->req->complete = acc_complete_setup_noop;
+			value = acc_unregister_hid(dev, w_value);
+		} else if (b_request == ACCESSORY_SET_HID_REPORT_DESC) {
+			spin_lock_irqsave(&dev->lock, flags);
+			hid = acc_hid_get(&dev->new_hid_list, w_value);
+			spin_unlock_irqrestore(&dev->lock, flags);
+			if (!hid) {
+				value = -EINVAL;
+				goto err;
+			}
+			offset = w_index;
+			if (offset != hid->report_desc_offset
+				|| offset + w_length > hid->report_desc_len) {
+				value = -EINVAL;
+				goto err;
+			}
+			cdev->req->context = hid;
+			cdev->req->complete = acc_complete_set_hid_report_desc;
+			value = w_length;
+		} else if (b_request == ACCESSORY_SEND_HID_EVENT) {
+			spin_lock_irqsave(&dev->lock, flags);
+			hid = acc_hid_get(&dev->hid_list, w_value);
+			spin_unlock_irqrestore(&dev->lock, flags);
+			if (!hid) {
+				value = -EINVAL;
+				goto err;
+			}
+			cdev->req->context = hid;
+			cdev->req->complete = acc_complete_send_hid_event;
+			value = w_length;
+		}
+	} else if (b_requestType == (USB_DIR_IN | USB_TYPE_VENDOR)) {
+		if (b_request == ACCESSORY_GET_PROTOCOL) {
+			*((u16 *)cdev->req->buf) = PROTOCOL_VERSION;
+			value = sizeof(u16);
+			cdev->req->complete = acc_complete_setup_noop;
+			/* clear any string left over from a previous session */
+			memset(dev->manufacturer, 0, sizeof(dev->manufacturer));
+			memset(dev->model, 0, sizeof(dev->model));
+			memset(dev->description, 0, sizeof(dev->description));
+			memset(dev->version, 0, sizeof(dev->version));
+			memset(dev->uri, 0, sizeof(dev->uri));
+			memset(dev->serial, 0, sizeof(dev->serial));
+			dev->start_requested = 0;
+			dev->audio_mode = 0;
+		}
+	}
+
+	if (value >= 0) {
+		cdev->req->zero = 0;
+		cdev->req->length = value;
+		value = usb_ep_queue(cdev->gadget->ep0, cdev->req, GFP_ATOMIC);
+		if (value < 0)
+			ERROR(cdev, "%s setup response queue error\n",
+				__func__);
+	}
+
+err:
+	if (value == -EOPNOTSUPP)
+		VDBG(cdev,
+			"unknown class-specific control req "
+			"%02x.%02x v%04x i%04x l%u\n",
+			ctrl->bRequestType, ctrl->bRequest,
+			w_value, w_index, w_length);
+	return value;
+}
+EXPORT_SYMBOL_GPL(acc_ctrlrequest);
+
+static int
+__acc_function_bind(struct usb_configuration *c,
+			struct usb_function *f, bool configfs)
+{
+	struct usb_composite_dev *cdev = c->cdev;
+	struct acc_dev	*dev = func_to_dev(f);
+	int			id;
+	int			ret;
+
+	DBG(cdev, "acc_function_bind dev: %p\n", dev);
+
+	if (configfs) {
+		if (acc_string_defs[INTERFACE_STRING_INDEX].id == 0) {
+			ret = usb_string_id(c->cdev);
+			if (ret < 0)
+				return ret;
+			acc_string_defs[INTERFACE_STRING_INDEX].id = ret;
+			acc_interface_desc.iInterface = ret;
+		}
+		dev->cdev = c->cdev;
+	}
+	ret = hid_register_driver(&acc_hid_driver);
+	if (ret)
+		return ret;
+
+	dev->start_requested = 0;
+
+	/* allocate interface ID(s) */
+	id = usb_interface_id(c, f);
+	if (id < 0)
+		return id;
+	acc_interface_desc.bInterfaceNumber = id;
+
+	/* allocate endpoints */
+	ret = create_bulk_endpoints(dev, &acc_fullspeed_in_desc,
+			&acc_fullspeed_out_desc);
+	if (ret)
+		return ret;
+
+	/* support high speed hardware */
+	if (gadget_is_dualspeed(c->cdev->gadget)) {
+		acc_highspeed_in_desc.bEndpointAddress =
+			acc_fullspeed_in_desc.bEndpointAddress;
+		acc_highspeed_out_desc.bEndpointAddress =
+			acc_fullspeed_out_desc.bEndpointAddress;
+	}
+
+	DBG(cdev, "%s speed %s: IN/%s, OUT/%s\n",
+			gadget_is_dualspeed(c->cdev->gadget) ? "dual" : "full",
+			f->name, dev->ep_in->name, dev->ep_out->name);
+	return 0;
+}
+
+static int
+acc_function_bind_configfs(struct usb_configuration *c,
+			struct usb_function *f) {
+	return __acc_function_bind(c, f, true);
+}
+
+static void
+kill_all_hid_devices(struct acc_dev *dev)
+{
+	struct acc_hid_dev *hid;
+	struct list_head *entry, *temp;
+	unsigned long flags;
+
+	/* do nothing if usb accessory device doesn't exist */
+	if (!dev)
+		return;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	list_for_each_safe(entry, temp, &dev->hid_list) {
+		hid = list_entry(entry, struct acc_hid_dev, list);
+		list_del(&hid->list);
+		list_add(&hid->list, &dev->dead_hid_list);
+	}
+	list_for_each_safe(entry, temp, &dev->new_hid_list) {
+		hid = list_entry(entry, struct acc_hid_dev, list);
+		list_del(&hid->list);
+		list_add(&hid->list, &dev->dead_hid_list);
+	}
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	schedule_work(&dev->hid_work);
+}
+
+static void
+acc_hid_unbind(struct acc_dev *dev)
+{
+	hid_unregister_driver(&acc_hid_driver);
+	kill_all_hid_devices(dev);
+}
+
+static void
+acc_function_unbind(struct usb_configuration *c, struct usb_function *f)
+{
+	struct acc_dev	*dev = func_to_dev(f);
+	struct usb_request *req;
+	int i;
+
+	dev->online = 0;		/* clear online flag */
+	wake_up(&dev->read_wq);		/* unblock reads on closure */
+	wake_up(&dev->write_wq);	/* likewise for writes */
+
+	while ((req = req_get(dev, &dev->tx_idle)))
+		acc_request_free(req, dev->ep_in);
+	for (i = 0; i < RX_REQ_MAX; i++)
+		acc_request_free(dev->rx_req[i], dev->ep_out);
+
+	acc_hid_unbind(dev);
+}
+
+static void acc_start_work(struct work_struct *data)
+{
+	char *envp[2] = { "ACCESSORY=START", NULL };
+
+	kobject_uevent_env(&acc_device.this_device->kobj, KOBJ_CHANGE, envp);
+}
+
+static int acc_hid_init(struct acc_hid_dev *hdev)
+{
+	struct hid_device *hid;
+	int ret;
+
+	hid = hid_allocate_device();
+	if (IS_ERR(hid))
+		return PTR_ERR(hid);
+
+	hid->ll_driver = &acc_hid_ll_driver;
+	hid->dev.parent = acc_device.this_device;
+
+	hid->bus = BUS_USB;
+	hid->vendor = HID_ANY_ID;
+	hid->product = HID_ANY_ID;
+	hid->driver_data = hdev;
+	ret = hid_add_device(hid);
+	if (ret) {
+		pr_err("can't add hid device: %d\n", ret);
+		hid_destroy_device(hid);
+		return ret;
+	}
+
+	hdev->hid = hid;
+	return 0;
+}
+
+static void acc_hid_delete(struct acc_hid_dev *hid)
+{
+	kfree(hid->report_desc);
+	kfree(hid);
+}
+
+static void acc_hid_work(struct work_struct *data)
+{
+	struct acc_dev *dev = _acc_dev;
+	struct list_head	*entry, *temp;
+	struct acc_hid_dev *hid;
+	struct list_head	new_list, dead_list;
+	unsigned long flags;
+
+	INIT_LIST_HEAD(&new_list);
+
+	spin_lock_irqsave(&dev->lock, flags);
+
+	/* copy hids that are ready for initialization to new_list */
+	list_for_each_safe(entry, temp, &dev->new_hid_list) {
+		hid = list_entry(entry, struct acc_hid_dev, list);
+		if (hid->report_desc_offset == hid->report_desc_len)
+			list_move(&hid->list, &new_list);
+	}
+
+	if (list_empty(&dev->dead_hid_list)) {
+		INIT_LIST_HEAD(&dead_list);
+	} else {
+		/* move all of dev->dead_hid_list to dead_list */
+		dead_list.prev = dev->dead_hid_list.prev;
+		dead_list.next = dev->dead_hid_list.next;
+		dead_list.next->prev = &dead_list;
+		dead_list.prev->next = &dead_list;
+		INIT_LIST_HEAD(&dev->dead_hid_list);
+	}
+
+	spin_unlock_irqrestore(&dev->lock, flags);
+
+	/* register new HID devices */
+	list_for_each_safe(entry, temp, &new_list) {
+		hid = list_entry(entry, struct acc_hid_dev, list);
+		if (acc_hid_init(hid)) {
+			pr_err("can't add HID device %p\n", hid);
+			acc_hid_delete(hid);
+		} else {
+			spin_lock_irqsave(&dev->lock, flags);
+			list_move(&hid->list, &dev->hid_list);
+			spin_unlock_irqrestore(&dev->lock, flags);
+		}
+	}
+
+	/* remove dead HID devices */
+	list_for_each_safe(entry, temp, &dead_list) {
+		hid = list_entry(entry, struct acc_hid_dev, list);
+		list_del(&hid->list);
+		if (hid->hid)
+			hid_destroy_device(hid->hid);
+		acc_hid_delete(hid);
+	}
+}
+
+static int acc_function_set_alt(struct usb_function *f,
+		unsigned intf, unsigned alt)
+{
+	struct acc_dev	*dev = func_to_dev(f);
+	struct usb_composite_dev *cdev = f->config->cdev;
+	int ret;
+
+	DBG(cdev, "acc_function_set_alt intf: %d alt: %d\n", intf, alt);
+
+	ret = config_ep_by_speed(cdev->gadget, f, dev->ep_in);
+	if (ret)
+		return ret;
+
+	ret = usb_ep_enable(dev->ep_in);
+	if (ret)
+		return ret;
+
+	ret = config_ep_by_speed(cdev->gadget, f, dev->ep_out);
+	if (ret)
+		return ret;
+
+	ret = usb_ep_enable(dev->ep_out);
+	if (ret) {
+		usb_ep_disable(dev->ep_in);
+		return ret;
+	}
+
+	dev->online = 1;
+	dev->disconnected = 0; /* if online then not disconnected */
+
+	/* readers may be blocked waiting for us to go online */
+	wake_up(&dev->read_wq);
+	return 0;
+}
+
+static void acc_function_disable(struct usb_function *f)
+{
+	struct acc_dev	*dev = func_to_dev(f);
+	struct usb_composite_dev	*cdev = dev->cdev;
+
+	DBG(cdev, "acc_function_disable\n");
+	acc_set_disconnected(dev); /* this now only sets disconnected */
+	dev->online = 0; /* so now need to clear online flag here too */
+	usb_ep_disable(dev->ep_in);
+	usb_ep_disable(dev->ep_out);
+
+	/* readers may be blocked waiting for us to go online */
+	wake_up(&dev->read_wq);
+
+	VDBG(cdev, "%s disabled\n", dev->function.name);
+}
+
+static int acc_setup(void)
+{
+	struct acc_dev *dev;
+	int ret;
+
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+	if (!dev)
+		return -ENOMEM;
+
+	spin_lock_init(&dev->lock);
+	init_waitqueue_head(&dev->read_wq);
+	init_waitqueue_head(&dev->write_wq);
+	atomic_set(&dev->open_excl, 0);
+	INIT_LIST_HEAD(&dev->tx_idle);
+	INIT_LIST_HEAD(&dev->hid_list);
+	INIT_LIST_HEAD(&dev->new_hid_list);
+	INIT_LIST_HEAD(&dev->dead_hid_list);
+	INIT_DELAYED_WORK(&dev->start_work, acc_start_work);
+	INIT_WORK(&dev->hid_work, acc_hid_work);
+
+	/* _acc_dev must be set before calling usb_gadget_register_driver */
+	_acc_dev = dev;
+
+	ret = misc_register(&acc_device);
+	if (ret)
+		goto err;
+
+	return 0;
+
+err:
+	kfree(dev);
+	pr_err("USB accessory gadget driver failed to initialize\n");
+	return ret;
+}
+
+void acc_disconnect(void)
+{
+	/* unregister all HID devices if USB is disconnected */
+	kill_all_hid_devices(_acc_dev);
+}
+EXPORT_SYMBOL_GPL(acc_disconnect);
+
+static void acc_cleanup(void)
+{
+	misc_deregister(&acc_device);
+	kfree(_acc_dev);
+	_acc_dev = NULL;
+}
+static struct acc_instance *to_acc_instance(struct config_item *item)
+{
+	return container_of(to_config_group(item), struct acc_instance,
+		func_inst.group);
+}
+
+static void acc_attr_release(struct config_item *item)
+{
+	struct acc_instance *fi_acc = to_acc_instance(item);
+
+	usb_put_function_instance(&fi_acc->func_inst);
+}
+
+static struct configfs_item_operations acc_item_ops = {
+	.release        = acc_attr_release,
+};
+
+static struct config_item_type acc_func_type = {
+	.ct_item_ops    = &acc_item_ops,
+	.ct_owner       = THIS_MODULE,
+};
+
+static struct acc_instance *to_fi_acc(struct usb_function_instance *fi)
+{
+	return container_of(fi, struct acc_instance, func_inst);
+}
+
+static int acc_set_inst_name(struct usb_function_instance *fi, const char *name)
+{
+	struct acc_instance *fi_acc;
+	char *ptr;
+	int name_len;
+
+	name_len = strlen(name) + 1;
+	if (name_len > MAX_INST_NAME_LEN)
+		return -ENAMETOOLONG;
+
+	ptr = kstrndup(name, name_len, GFP_KERNEL);
+	if (!ptr)
+		return -ENOMEM;
+
+	fi_acc = to_fi_acc(fi);
+	fi_acc->name = ptr;
+	return 0;
+}
+
+static void acc_free_inst(struct usb_function_instance *fi)
+{
+	struct acc_instance *fi_acc;
+
+	fi_acc = to_fi_acc(fi);
+	kfree(fi_acc->name);
+	acc_cleanup();
+}
+
+static struct usb_function_instance *acc_alloc_inst(void)
+{
+	struct acc_instance *fi_acc;
+	struct acc_dev *dev;
+	int err;
+
+	fi_acc = kzalloc(sizeof(*fi_acc), GFP_KERNEL);
+	if (!fi_acc)
+		return ERR_PTR(-ENOMEM);
+	fi_acc->func_inst.set_inst_name = acc_set_inst_name;
+	fi_acc->func_inst.free_func_inst = acc_free_inst;
+
+	err = acc_setup();
+	if (err) {
+		kfree(fi_acc);
+		pr_err("Error setting ACCESSORY\n");
+		return ERR_PTR(err);
+	}
+
+	config_group_init_type_name(&fi_acc->func_inst.group,
+					"", &acc_func_type);
+	dev = _acc_dev;
+	return  &fi_acc->func_inst;
+}
+
+static void acc_free(struct usb_function *f)
+{
+/*NO-OP: no function specific resource allocation in mtp_alloc*/
+}
+
+int acc_ctrlrequest_configfs(struct usb_function *f,
+			const struct usb_ctrlrequest *ctrl) {
+	if (f->config != NULL && f->config->cdev != NULL)
+		return acc_ctrlrequest(f->config->cdev, ctrl);
+	else
+		return -1;
+}
+
+static struct usb_function *acc_alloc(struct usb_function_instance *fi)
+{
+	struct acc_dev *dev = _acc_dev;
+
+	pr_info("acc_alloc\n");
+
+	dev->function.name = "accessory";
+	dev->function.strings = acc_strings,
+	dev->function.fs_descriptors = fs_acc_descs;
+	dev->function.hs_descriptors = hs_acc_descs;
+	dev->function.bind = acc_function_bind_configfs;
+	dev->function.unbind = acc_function_unbind;
+	dev->function.set_alt = acc_function_set_alt;
+	dev->function.disable = acc_function_disable;
+	dev->function.free_func = acc_free;
+	dev->function.setup = acc_ctrlrequest_configfs;
+
+	return &dev->function;
+}
+DECLARE_USB_FUNCTION_INIT(accessory, acc_alloc_inst, acc_alloc);
+MODULE_LICENSE("GPL");
diff --git a/drivers/usb/gadget/function/f_audio_source.c b/drivers/usb/gadget/function/f_audio_source.c
new file mode 100644
index 000000000000..8124af33b738
--- /dev/null
+++ b/drivers/usb/gadget/function/f_audio_source.c
@@ -0,0 +1,1071 @@
+/*
+ * Gadget Function Driver for USB audio source device
+ *
+ * Copyright (C) 2012 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/device.h>
+#include <linux/usb/audio.h>
+#include <linux/wait.h>
+#include <linux/pm_qos.h>
+#include <sound/core.h>
+#include <sound/initval.h>
+#include <sound/pcm.h>
+
+#include <linux/usb.h>
+#include <linux/usb_usual.h>
+#include <linux/usb/ch9.h>
+#include <linux/configfs.h>
+#include <linux/usb/composite.h>
+#include <linux/module.h>
+#include <linux/moduleparam.h>
+#define SAMPLE_RATE 44100
+#define FRAMES_PER_MSEC (SAMPLE_RATE / 1000)
+
+#define IN_EP_MAX_PACKET_SIZE 256
+
+/* Number of requests to allocate */
+#define IN_EP_REQ_COUNT 4
+
+#define AUDIO_AC_INTERFACE	0
+#define AUDIO_AS_INTERFACE	1
+#define AUDIO_NUM_INTERFACES	2
+#define MAX_INST_NAME_LEN     40
+
+/* B.3.1  Standard AC Interface Descriptor */
+static struct usb_interface_descriptor ac_interface_desc = {
+	.bLength =		USB_DT_INTERFACE_SIZE,
+	.bDescriptorType =	USB_DT_INTERFACE,
+	.bNumEndpoints =	0,
+	.bInterfaceClass =	USB_CLASS_AUDIO,
+	.bInterfaceSubClass =	USB_SUBCLASS_AUDIOCONTROL,
+};
+
+DECLARE_UAC_AC_HEADER_DESCRIPTOR(2);
+
+#define UAC_DT_AC_HEADER_LENGTH	UAC_DT_AC_HEADER_SIZE(AUDIO_NUM_INTERFACES)
+/* 1 input terminal, 1 output terminal and 1 feature unit */
+#define UAC_DT_TOTAL_LENGTH (UAC_DT_AC_HEADER_LENGTH \
+	+ UAC_DT_INPUT_TERMINAL_SIZE + UAC_DT_OUTPUT_TERMINAL_SIZE \
+	+ UAC_DT_FEATURE_UNIT_SIZE(0))
+/* B.3.2  Class-Specific AC Interface Descriptor */
+static struct uac1_ac_header_descriptor_2 ac_header_desc = {
+	.bLength =		UAC_DT_AC_HEADER_LENGTH,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	UAC_HEADER,
+	.bcdADC =		__constant_cpu_to_le16(0x0100),
+	.wTotalLength =		__constant_cpu_to_le16(UAC_DT_TOTAL_LENGTH),
+	.bInCollection =	AUDIO_NUM_INTERFACES,
+	.baInterfaceNr = {
+		[0] =		AUDIO_AC_INTERFACE,
+		[1] =		AUDIO_AS_INTERFACE,
+	}
+};
+
+#define INPUT_TERMINAL_ID	1
+static struct uac_input_terminal_descriptor input_terminal_desc = {
+	.bLength =		UAC_DT_INPUT_TERMINAL_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	UAC_INPUT_TERMINAL,
+	.bTerminalID =		INPUT_TERMINAL_ID,
+	.wTerminalType =	UAC_INPUT_TERMINAL_MICROPHONE,
+	.bAssocTerminal =	0,
+	.wChannelConfig =	0x3,
+};
+
+DECLARE_UAC_FEATURE_UNIT_DESCRIPTOR(0);
+
+#define FEATURE_UNIT_ID		2
+static struct uac_feature_unit_descriptor_0 feature_unit_desc = {
+	.bLength		= UAC_DT_FEATURE_UNIT_SIZE(0),
+	.bDescriptorType	= USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype	= UAC_FEATURE_UNIT,
+	.bUnitID		= FEATURE_UNIT_ID,
+	.bSourceID		= INPUT_TERMINAL_ID,
+	.bControlSize		= 2,
+};
+
+#define OUTPUT_TERMINAL_ID	3
+static struct uac1_output_terminal_descriptor output_terminal_desc = {
+	.bLength		= UAC_DT_OUTPUT_TERMINAL_SIZE,
+	.bDescriptorType	= USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype	= UAC_OUTPUT_TERMINAL,
+	.bTerminalID		= OUTPUT_TERMINAL_ID,
+	.wTerminalType		= UAC_TERMINAL_STREAMING,
+	.bAssocTerminal		= FEATURE_UNIT_ID,
+	.bSourceID		= FEATURE_UNIT_ID,
+};
+
+/* B.4.1  Standard AS Interface Descriptor */
+static struct usb_interface_descriptor as_interface_alt_0_desc = {
+	.bLength =		USB_DT_INTERFACE_SIZE,
+	.bDescriptorType =	USB_DT_INTERFACE,
+	.bAlternateSetting =	0,
+	.bNumEndpoints =	0,
+	.bInterfaceClass =	USB_CLASS_AUDIO,
+	.bInterfaceSubClass =	USB_SUBCLASS_AUDIOSTREAMING,
+};
+
+static struct usb_interface_descriptor as_interface_alt_1_desc = {
+	.bLength =		USB_DT_INTERFACE_SIZE,
+	.bDescriptorType =	USB_DT_INTERFACE,
+	.bAlternateSetting =	1,
+	.bNumEndpoints =	1,
+	.bInterfaceClass =	USB_CLASS_AUDIO,
+	.bInterfaceSubClass =	USB_SUBCLASS_AUDIOSTREAMING,
+};
+
+/* B.4.2  Class-Specific AS Interface Descriptor */
+static struct uac1_as_header_descriptor as_header_desc = {
+	.bLength =		UAC_DT_AS_HEADER_SIZE,
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	UAC_AS_GENERAL,
+	.bTerminalLink =	INPUT_TERMINAL_ID,
+	.bDelay =		1,
+	.wFormatTag =		UAC_FORMAT_TYPE_I_PCM,
+};
+
+DECLARE_UAC_FORMAT_TYPE_I_DISCRETE_DESC(1);
+
+static struct uac_format_type_i_discrete_descriptor_1 as_type_i_desc = {
+	.bLength =		UAC_FORMAT_TYPE_I_DISCRETE_DESC_SIZE(1),
+	.bDescriptorType =	USB_DT_CS_INTERFACE,
+	.bDescriptorSubtype =	UAC_FORMAT_TYPE,
+	.bFormatType =		UAC_FORMAT_TYPE_I,
+	.bSubframeSize =	2,
+	.bBitResolution =	16,
+	.bSamFreqType =		1,
+};
+
+/* Standard ISO IN Endpoint Descriptor for highspeed */
+static struct usb_endpoint_descriptor hs_as_in_ep_desc  = {
+	.bLength =		USB_DT_ENDPOINT_AUDIO_SIZE,
+	.bDescriptorType =	USB_DT_ENDPOINT,
+	.bEndpointAddress =	USB_DIR_IN,
+	.bmAttributes =		USB_ENDPOINT_SYNC_SYNC
+				| USB_ENDPOINT_XFER_ISOC,
+	.wMaxPacketSize =	__constant_cpu_to_le16(IN_EP_MAX_PACKET_SIZE),
+	.bInterval =		4, /* poll 1 per millisecond */
+};
+
+/* Standard ISO IN Endpoint Descriptor for highspeed */
+static struct usb_endpoint_descriptor fs_as_in_ep_desc  = {
+	.bLength =		USB_DT_ENDPOINT_AUDIO_SIZE,
+	.bDescriptorType =	USB_DT_ENDPOINT,
+	.bEndpointAddress =	USB_DIR_IN,
+	.bmAttributes =		USB_ENDPOINT_SYNC_SYNC
+				| USB_ENDPOINT_XFER_ISOC,
+	.wMaxPacketSize =	__constant_cpu_to_le16(IN_EP_MAX_PACKET_SIZE),
+	.bInterval =		1, /* poll 1 per millisecond */
+};
+
+/* Class-specific AS ISO OUT Endpoint Descriptor */
+static struct uac_iso_endpoint_descriptor as_iso_in_desc = {
+	.bLength =		UAC_ISO_ENDPOINT_DESC_SIZE,
+	.bDescriptorType =	USB_DT_CS_ENDPOINT,
+	.bDescriptorSubtype =	UAC_EP_GENERAL,
+	.bmAttributes =		1,
+	.bLockDelayUnits =	1,
+	.wLockDelay =		__constant_cpu_to_le16(1),
+};
+
+static struct usb_descriptor_header *hs_audio_desc[] = {
+	(struct usb_descriptor_header *)&ac_interface_desc,
+	(struct usb_descriptor_header *)&ac_header_desc,
+
+	(struct usb_descriptor_header *)&input_terminal_desc,
+	(struct usb_descriptor_header *)&output_terminal_desc,
+	(struct usb_descriptor_header *)&feature_unit_desc,
+
+	(struct usb_descriptor_header *)&as_interface_alt_0_desc,
+	(struct usb_descriptor_header *)&as_interface_alt_1_desc,
+	(struct usb_descriptor_header *)&as_header_desc,
+
+	(struct usb_descriptor_header *)&as_type_i_desc,
+
+	(struct usb_descriptor_header *)&hs_as_in_ep_desc,
+	(struct usb_descriptor_header *)&as_iso_in_desc,
+	NULL,
+};
+
+static struct usb_descriptor_header *fs_audio_desc[] = {
+	(struct usb_descriptor_header *)&ac_interface_desc,
+	(struct usb_descriptor_header *)&ac_header_desc,
+
+	(struct usb_descriptor_header *)&input_terminal_desc,
+	(struct usb_descriptor_header *)&output_terminal_desc,
+	(struct usb_descriptor_header *)&feature_unit_desc,
+
+	(struct usb_descriptor_header *)&as_interface_alt_0_desc,
+	(struct usb_descriptor_header *)&as_interface_alt_1_desc,
+	(struct usb_descriptor_header *)&as_header_desc,
+
+	(struct usb_descriptor_header *)&as_type_i_desc,
+
+	(struct usb_descriptor_header *)&fs_as_in_ep_desc,
+	(struct usb_descriptor_header *)&as_iso_in_desc,
+	NULL,
+};
+
+static struct snd_pcm_hardware audio_hw_info = {
+	.info =			SNDRV_PCM_INFO_MMAP |
+				SNDRV_PCM_INFO_MMAP_VALID |
+				SNDRV_PCM_INFO_BATCH |
+				SNDRV_PCM_INFO_INTERLEAVED |
+				SNDRV_PCM_INFO_BLOCK_TRANSFER,
+
+	.formats		= SNDRV_PCM_FMTBIT_S16_LE,
+	.channels_min		= 2,
+	.channels_max		= 2,
+	.rate_min		= SAMPLE_RATE,
+	.rate_max		= SAMPLE_RATE,
+
+	.buffer_bytes_max =	1024 * 1024,
+	.period_bytes_min =	64,
+	.period_bytes_max =	512 * 1024,
+	.periods_min =		2,
+	.periods_max =		1024,
+};
+
+/*-------------------------------------------------------------------------*/
+
+struct audio_source_config {
+	int	card;
+	int	device;
+};
+
+struct audio_dev {
+	struct usb_function		func;
+	struct snd_card			*card;
+	struct snd_pcm			*pcm;
+	struct snd_pcm_substream *substream;
+
+	struct list_head		idle_reqs;
+	struct usb_ep			*in_ep;
+
+	spinlock_t			lock;
+
+	/* beginning, end and current position in our buffer */
+	void				*buffer_start;
+	void				*buffer_end;
+	void				*buffer_pos;
+
+	/* byte size of a "period" */
+	unsigned int			period;
+	/* bytes sent since last call to snd_pcm_period_elapsed */
+	unsigned int			period_offset;
+	/* time we started playing */
+	ktime_t				start_time;
+	/* number of frames sent since start_time */
+	s64				frames_sent;
+	struct audio_source_config	*config;
+	/* for creating and issuing QoS requests */
+	struct pm_qos_request pm_qos;
+};
+
+static inline struct audio_dev *func_to_audio(struct usb_function *f)
+{
+	return container_of(f, struct audio_dev, func);
+}
+
+/*-------------------------------------------------------------------------*/
+
+struct audio_source_instance {
+	struct usb_function_instance func_inst;
+	const char *name;
+	struct audio_source_config *config;
+	struct device *audio_device;
+};
+
+static void audio_source_attr_release(struct config_item *item);
+
+static struct configfs_item_operations audio_source_item_ops = {
+	.release        = audio_source_attr_release,
+};
+
+static struct config_item_type audio_source_func_type = {
+	.ct_item_ops    = &audio_source_item_ops,
+	.ct_owner       = THIS_MODULE,
+};
+
+static ssize_t audio_source_pcm_show(struct device *dev,
+		struct device_attribute *attr, char *buf);
+
+static DEVICE_ATTR(pcm, S_IRUGO, audio_source_pcm_show, NULL);
+
+static struct device_attribute *audio_source_function_attributes[] = {
+	&dev_attr_pcm,
+	NULL
+};
+
+/*--------------------------------------------------------------------------*/
+
+static struct usb_request *audio_request_new(struct usb_ep *ep, int buffer_size)
+{
+	struct usb_request *req = usb_ep_alloc_request(ep, GFP_KERNEL);
+
+	if (!req)
+		return NULL;
+
+	req->buf = kmalloc(buffer_size, GFP_KERNEL);
+	if (!req->buf) {
+		usb_ep_free_request(ep, req);
+		return NULL;
+	}
+	req->length = buffer_size;
+	return req;
+}
+
+static void audio_request_free(struct usb_request *req, struct usb_ep *ep)
+{
+	if (req) {
+		kfree(req->buf);
+		usb_ep_free_request(ep, req);
+	}
+}
+
+static void audio_req_put(struct audio_dev *audio, struct usb_request *req)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&audio->lock, flags);
+	list_add_tail(&req->list, &audio->idle_reqs);
+	spin_unlock_irqrestore(&audio->lock, flags);
+}
+
+static struct usb_request *audio_req_get(struct audio_dev *audio)
+{
+	unsigned long flags;
+	struct usb_request *req;
+
+	spin_lock_irqsave(&audio->lock, flags);
+	if (list_empty(&audio->idle_reqs)) {
+		req = 0;
+	} else {
+		req = list_first_entry(&audio->idle_reqs, struct usb_request,
+				list);
+		list_del(&req->list);
+	}
+	spin_unlock_irqrestore(&audio->lock, flags);
+	return req;
+}
+
+/* send the appropriate number of packets to match our bitrate */
+static void audio_send(struct audio_dev *audio)
+{
+	struct snd_pcm_runtime *runtime;
+	struct usb_request *req;
+	int length, length1, length2, ret;
+	s64 msecs;
+	s64 frames;
+	ktime_t now;
+
+	/* audio->substream will be null if we have been closed */
+	if (!audio->substream)
+		return;
+	/* audio->buffer_pos will be null if we have been stopped */
+	if (!audio->buffer_pos)
+		return;
+
+	runtime = audio->substream->runtime;
+
+	/* compute number of frames to send */
+	now = ktime_get();
+	msecs = div_s64((ktime_to_ns(now) - ktime_to_ns(audio->start_time)),
+			1000000);
+	frames = div_s64((msecs * SAMPLE_RATE), 1000);
+
+	/* Readjust our frames_sent if we fall too far behind.
+	 * If we get too far behind it is better to drop some frames than
+	 * to keep sending data too fast in an attempt to catch up.
+	 */
+	if (frames - audio->frames_sent > 10 * FRAMES_PER_MSEC)
+		audio->frames_sent = frames - FRAMES_PER_MSEC;
+
+	frames -= audio->frames_sent;
+
+	/* We need to send something to keep the pipeline going */
+	if (frames <= 0)
+		frames = FRAMES_PER_MSEC;
+
+	while (frames > 0) {
+		req = audio_req_get(audio);
+		if (!req)
+			break;
+
+		length = frames_to_bytes(runtime, frames);
+		if (length > IN_EP_MAX_PACKET_SIZE)
+			length = IN_EP_MAX_PACKET_SIZE;
+
+		if (audio->buffer_pos + length > audio->buffer_end)
+			length1 = audio->buffer_end - audio->buffer_pos;
+		else
+			length1 = length;
+		memcpy(req->buf, audio->buffer_pos, length1);
+		if (length1 < length) {
+			/* Wrap around and copy remaining length
+			 * at beginning of buffer.
+			 */
+			length2 = length - length1;
+			memcpy(req->buf + length1, audio->buffer_start,
+					length2);
+			audio->buffer_pos = audio->buffer_start + length2;
+		} else {
+			audio->buffer_pos += length1;
+			if (audio->buffer_pos >= audio->buffer_end)
+				audio->buffer_pos = audio->buffer_start;
+		}
+
+		req->length = length;
+		ret = usb_ep_queue(audio->in_ep, req, GFP_ATOMIC);
+		if (ret < 0) {
+			pr_err("usb_ep_queue failed ret: %d\n", ret);
+			audio_req_put(audio, req);
+			break;
+		}
+
+		frames -= bytes_to_frames(runtime, length);
+		audio->frames_sent += bytes_to_frames(runtime, length);
+	}
+}
+
+static void audio_control_complete(struct usb_ep *ep, struct usb_request *req)
+{
+	/* nothing to do here */
+}
+
+static void audio_data_complete(struct usb_ep *ep, struct usb_request *req)
+{
+	struct audio_dev *audio = req->context;
+
+	pr_debug("audio_data_complete req->status %d req->actual %d\n",
+		req->status, req->actual);
+
+	audio_req_put(audio, req);
+
+	if (!audio->buffer_start || req->status)
+		return;
+
+	audio->period_offset += req->actual;
+	if (audio->period_offset >= audio->period) {
+		snd_pcm_period_elapsed(audio->substream);
+		audio->period_offset = 0;
+	}
+	audio_send(audio);
+}
+
+static int audio_set_endpoint_req(struct usb_function *f,
+		const struct usb_ctrlrequest *ctrl)
+{
+	int value = -EOPNOTSUPP;
+	u16 ep = le16_to_cpu(ctrl->wIndex);
+	u16 len = le16_to_cpu(ctrl->wLength);
+	u16 w_value = le16_to_cpu(ctrl->wValue);
+
+	pr_debug("bRequest 0x%x, w_value 0x%04x, len %d, endpoint %d\n",
+			ctrl->bRequest, w_value, len, ep);
+
+	switch (ctrl->bRequest) {
+	case UAC_SET_CUR:
+	case UAC_SET_MIN:
+	case UAC_SET_MAX:
+	case UAC_SET_RES:
+		value = len;
+		break;
+	default:
+		break;
+	}
+
+	return value;
+}
+
+static int audio_get_endpoint_req(struct usb_function *f,
+		const struct usb_ctrlrequest *ctrl)
+{
+	struct usb_composite_dev *cdev = f->config->cdev;
+	int value = -EOPNOTSUPP;
+	u8 ep = ((le16_to_cpu(ctrl->wIndex) >> 8) & 0xFF);
+	u16 len = le16_to_cpu(ctrl->wLength);
+	u16 w_value = le16_to_cpu(ctrl->wValue);
+	u8 *buf = cdev->req->buf;
+
+	pr_debug("bRequest 0x%x, w_value 0x%04x, len %d, endpoint %d\n",
+			ctrl->bRequest, w_value, len, ep);
+
+	if (w_value == UAC_EP_CS_ATTR_SAMPLE_RATE << 8) {
+		switch (ctrl->bRequest) {
+		case UAC_GET_CUR:
+		case UAC_GET_MIN:
+		case UAC_GET_MAX:
+		case UAC_GET_RES:
+			/* return our sample rate */
+			buf[0] = (u8)SAMPLE_RATE;
+			buf[1] = (u8)(SAMPLE_RATE >> 8);
+			buf[2] = (u8)(SAMPLE_RATE >> 16);
+			value = 3;
+			break;
+		default:
+			break;
+		}
+	}
+
+	return value;
+}
+
+static int
+audio_setup(struct usb_function *f, const struct usb_ctrlrequest *ctrl)
+{
+	struct usb_composite_dev *cdev = f->config->cdev;
+	struct usb_request *req = cdev->req;
+	int value = -EOPNOTSUPP;
+	u16 w_index = le16_to_cpu(ctrl->wIndex);
+	u16 w_value = le16_to_cpu(ctrl->wValue);
+	u16 w_length = le16_to_cpu(ctrl->wLength);
+
+	/* composite driver infrastructure handles everything; interface
+	 * activation uses set_alt().
+	 */
+	switch (ctrl->bRequestType) {
+	case USB_DIR_OUT | USB_TYPE_CLASS | USB_RECIP_ENDPOINT:
+		value = audio_set_endpoint_req(f, ctrl);
+		break;
+
+	case USB_DIR_IN | USB_TYPE_CLASS | USB_RECIP_ENDPOINT:
+		value = audio_get_endpoint_req(f, ctrl);
+		break;
+	}
+
+	/* respond with data transfer or status phase? */
+	if (value >= 0) {
+		pr_debug("audio req%02x.%02x v%04x i%04x l%d\n",
+			ctrl->bRequestType, ctrl->bRequest,
+			w_value, w_index, w_length);
+		req->zero = 0;
+		req->length = value;
+		req->complete = audio_control_complete;
+		value = usb_ep_queue(cdev->gadget->ep0, req, GFP_ATOMIC);
+		if (value < 0)
+			pr_err("audio response on err %d\n", value);
+	}
+
+	/* device either stalls (value < 0) or reports success */
+	return value;
+}
+
+static int audio_set_alt(struct usb_function *f, unsigned intf, unsigned alt)
+{
+	struct audio_dev *audio = func_to_audio(f);
+	struct usb_composite_dev *cdev = f->config->cdev;
+	int ret;
+
+	pr_debug("audio_set_alt intf %d, alt %d\n", intf, alt);
+
+	ret = config_ep_by_speed(cdev->gadget, f, audio->in_ep);
+	if (ret)
+		return ret;
+
+	usb_ep_enable(audio->in_ep);
+	return 0;
+}
+
+static void audio_disable(struct usb_function *f)
+{
+	struct audio_dev	*audio = func_to_audio(f);
+
+	pr_debug("audio_disable\n");
+	usb_ep_disable(audio->in_ep);
+}
+
+static void audio_free_func(struct usb_function *f)
+{
+	/* no-op */
+}
+
+/*-------------------------------------------------------------------------*/
+
+static void audio_build_desc(struct audio_dev *audio)
+{
+	u8 *sam_freq;
+	int rate;
+
+	/* Set channel numbers */
+	input_terminal_desc.bNrChannels = 2;
+	as_type_i_desc.bNrChannels = 2;
+
+	/* Set sample rates */
+	rate = SAMPLE_RATE;
+	sam_freq = as_type_i_desc.tSamFreq[0];
+	memcpy(sam_freq, &rate, 3);
+}
+
+
+static int snd_card_setup(struct usb_configuration *c,
+	struct audio_source_config *config);
+static struct audio_source_instance *to_fi_audio_source(
+	const struct usb_function_instance *fi);
+
+
+/* audio function driver setup/binding */
+static int
+audio_bind(struct usb_configuration *c, struct usb_function *f)
+{
+	struct usb_composite_dev *cdev = c->cdev;
+	struct audio_dev *audio = func_to_audio(f);
+	int status;
+	struct usb_ep *ep;
+	struct usb_request *req;
+	int i;
+	int err;
+
+	if (IS_ENABLED(CONFIG_USB_CONFIGFS)) {
+		struct audio_source_instance *fi_audio =
+				to_fi_audio_source(f->fi);
+		struct audio_source_config *config =
+				fi_audio->config;
+
+		err = snd_card_setup(c, config);
+		if (err)
+			return err;
+	}
+
+	audio_build_desc(audio);
+
+	/* allocate instance-specific interface IDs, and patch descriptors */
+	status = usb_interface_id(c, f);
+	if (status < 0)
+		goto fail;
+	ac_interface_desc.bInterfaceNumber = status;
+
+	/* AUDIO_AC_INTERFACE */
+	ac_header_desc.baInterfaceNr[0] = status;
+
+	status = usb_interface_id(c, f);
+	if (status < 0)
+		goto fail;
+	as_interface_alt_0_desc.bInterfaceNumber = status;
+	as_interface_alt_1_desc.bInterfaceNumber = status;
+
+	/* AUDIO_AS_INTERFACE */
+	ac_header_desc.baInterfaceNr[1] = status;
+
+	status = -ENODEV;
+
+	/* allocate our endpoint */
+	ep = usb_ep_autoconfig(cdev->gadget, &fs_as_in_ep_desc);
+	if (!ep)
+		goto fail;
+	audio->in_ep = ep;
+	ep->driver_data = audio; /* claim */
+
+	if (gadget_is_dualspeed(c->cdev->gadget))
+		hs_as_in_ep_desc.bEndpointAddress =
+			fs_as_in_ep_desc.bEndpointAddress;
+
+	f->fs_descriptors = fs_audio_desc;
+	f->hs_descriptors = hs_audio_desc;
+
+	for (i = 0, status = 0; i < IN_EP_REQ_COUNT && status == 0; i++) {
+		req = audio_request_new(ep, IN_EP_MAX_PACKET_SIZE);
+		if (req) {
+			req->context = audio;
+			req->complete = audio_data_complete;
+			audio_req_put(audio, req);
+		} else
+			status = -ENOMEM;
+	}
+
+fail:
+	return status;
+}
+
+static void
+audio_unbind(struct usb_configuration *c, struct usb_function *f)
+{
+	struct audio_dev *audio = func_to_audio(f);
+	struct usb_request *req;
+
+	while ((req = audio_req_get(audio)))
+		audio_request_free(req, audio->in_ep);
+
+	snd_card_free_when_closed(audio->card);
+	audio->card = NULL;
+	audio->pcm = NULL;
+	audio->substream = NULL;
+	audio->in_ep = NULL;
+
+	if (IS_ENABLED(CONFIG_USB_CONFIGFS)) {
+		struct audio_source_instance *fi_audio =
+				to_fi_audio_source(f->fi);
+		struct audio_source_config *config =
+				fi_audio->config;
+
+		config->card = -1;
+		config->device = -1;
+	}
+}
+
+static void audio_pcm_playback_start(struct audio_dev *audio)
+{
+	audio->start_time = ktime_get();
+	audio->frames_sent = 0;
+	audio_send(audio);
+}
+
+static void audio_pcm_playback_stop(struct audio_dev *audio)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&audio->lock, flags);
+	audio->buffer_start = 0;
+	audio->buffer_end = 0;
+	audio->buffer_pos = 0;
+	spin_unlock_irqrestore(&audio->lock, flags);
+}
+
+static int audio_pcm_open(struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct audio_dev *audio = substream->private_data;
+
+	runtime->private_data = audio;
+	runtime->hw = audio_hw_info;
+	snd_pcm_limit_hw_rates(runtime);
+	runtime->hw.channels_max = 2;
+
+	audio->substream = substream;
+
+	/* Add the QoS request and set the latency to 0 */
+	pm_qos_add_request(&audio->pm_qos, PM_QOS_CPU_DMA_LATENCY, 0);
+
+	return 0;
+}
+
+static int audio_pcm_close(struct snd_pcm_substream *substream)
+{
+	struct audio_dev *audio = substream->private_data;
+	unsigned long flags;
+
+	spin_lock_irqsave(&audio->lock, flags);
+
+	/* Remove the QoS request */
+	pm_qos_remove_request(&audio->pm_qos);
+
+	audio->substream = NULL;
+	spin_unlock_irqrestore(&audio->lock, flags);
+
+	return 0;
+}
+
+static int audio_pcm_hw_params(struct snd_pcm_substream *substream,
+				struct snd_pcm_hw_params *params)
+{
+	unsigned int channels = params_channels(params);
+	unsigned int rate = params_rate(params);
+
+	if (rate != SAMPLE_RATE)
+		return -EINVAL;
+	if (channels != 2)
+		return -EINVAL;
+
+	return snd_pcm_lib_alloc_vmalloc_buffer(substream,
+		params_buffer_bytes(params));
+}
+
+static int audio_pcm_hw_free(struct snd_pcm_substream *substream)
+{
+	return snd_pcm_lib_free_vmalloc_buffer(substream);
+}
+
+static int audio_pcm_prepare(struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct audio_dev *audio = runtime->private_data;
+
+	audio->period = snd_pcm_lib_period_bytes(substream);
+	audio->period_offset = 0;
+	audio->buffer_start = runtime->dma_area;
+	audio->buffer_end = audio->buffer_start
+		+ snd_pcm_lib_buffer_bytes(substream);
+	audio->buffer_pos = audio->buffer_start;
+
+	return 0;
+}
+
+static snd_pcm_uframes_t audio_pcm_pointer(struct snd_pcm_substream *substream)
+{
+	struct snd_pcm_runtime *runtime = substream->runtime;
+	struct audio_dev *audio = runtime->private_data;
+	ssize_t bytes = audio->buffer_pos - audio->buffer_start;
+
+	/* return offset of next frame to fill in our buffer */
+	return bytes_to_frames(runtime, bytes);
+}
+
+static int audio_pcm_playback_trigger(struct snd_pcm_substream *substream,
+					int cmd)
+{
+	struct audio_dev *audio = substream->runtime->private_data;
+	int ret = 0;
+
+	switch (cmd) {
+	case SNDRV_PCM_TRIGGER_START:
+	case SNDRV_PCM_TRIGGER_RESUME:
+		audio_pcm_playback_start(audio);
+		break;
+
+	case SNDRV_PCM_TRIGGER_STOP:
+	case SNDRV_PCM_TRIGGER_SUSPEND:
+		audio_pcm_playback_stop(audio);
+		break;
+
+	default:
+		ret = -EINVAL;
+	}
+
+	return ret;
+}
+
+static struct audio_dev _audio_dev = {
+	.func = {
+		.name = "audio_source",
+		.bind = audio_bind,
+		.unbind = audio_unbind,
+		.set_alt = audio_set_alt,
+		.setup = audio_setup,
+		.disable = audio_disable,
+		.free_func = audio_free_func,
+	},
+	.lock = __SPIN_LOCK_UNLOCKED(_audio_dev.lock),
+	.idle_reqs = LIST_HEAD_INIT(_audio_dev.idle_reqs),
+};
+
+static struct snd_pcm_ops audio_playback_ops = {
+	.open		= audio_pcm_open,
+	.close		= audio_pcm_close,
+	.ioctl		= snd_pcm_lib_ioctl,
+	.hw_params	= audio_pcm_hw_params,
+	.hw_free	= audio_pcm_hw_free,
+	.prepare	= audio_pcm_prepare,
+	.trigger	= audio_pcm_playback_trigger,
+	.pointer	= audio_pcm_pointer,
+};
+
+int audio_source_bind_config(struct usb_configuration *c,
+		struct audio_source_config *config)
+{
+	struct audio_dev *audio;
+	int err;
+
+	config->card = -1;
+	config->device = -1;
+
+	audio = &_audio_dev;
+
+	err = snd_card_setup(c, config);
+	if (err)
+		return err;
+
+	err = usb_add_function(c, &audio->func);
+	if (err)
+		goto add_fail;
+
+	return 0;
+
+add_fail:
+	snd_card_free(audio->card);
+	return err;
+}
+
+static int snd_card_setup(struct usb_configuration *c,
+		struct audio_source_config *config)
+{
+	struct audio_dev *audio;
+	struct snd_card *card;
+	struct snd_pcm *pcm;
+	int err;
+
+	audio = &_audio_dev;
+
+	err = snd_card_new(&c->cdev->gadget->dev,
+			SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1,
+			THIS_MODULE, 0, &card);
+	if (err)
+		return err;
+
+	err = snd_pcm_new(card, "USB audio source", 0, 1, 0, &pcm);
+	if (err)
+		goto pcm_fail;
+
+	pcm->private_data = audio;
+	pcm->info_flags = 0;
+	audio->pcm = pcm;
+
+	strlcpy(pcm->name, "USB gadget audio", sizeof(pcm->name));
+
+	snd_pcm_set_ops(pcm, SNDRV_PCM_STREAM_PLAYBACK, &audio_playback_ops);
+	snd_pcm_lib_preallocate_pages_for_all(pcm, SNDRV_DMA_TYPE_DEV,
+				NULL, 0, 64 * 1024);
+
+	strlcpy(card->driver, "audio_source", sizeof(card->driver));
+	strlcpy(card->shortname, card->driver, sizeof(card->shortname));
+	strlcpy(card->longname, "USB accessory audio source",
+		sizeof(card->longname));
+
+	err = snd_card_register(card);
+	if (err)
+		goto register_fail;
+
+	config->card = pcm->card->number;
+	config->device = pcm->device;
+	audio->card = card;
+	return 0;
+
+register_fail:
+pcm_fail:
+	snd_card_free(audio->card);
+	return err;
+}
+
+static struct audio_source_instance *to_audio_source_instance(
+					struct config_item *item)
+{
+	return container_of(to_config_group(item), struct audio_source_instance,
+		func_inst.group);
+}
+
+static struct audio_source_instance *to_fi_audio_source(
+					const struct usb_function_instance *fi)
+{
+	return container_of(fi, struct audio_source_instance, func_inst);
+}
+
+static void audio_source_attr_release(struct config_item *item)
+{
+	struct audio_source_instance *fi_audio = to_audio_source_instance(item);
+
+	usb_put_function_instance(&fi_audio->func_inst);
+}
+
+static int audio_source_set_inst_name(struct usb_function_instance *fi,
+					const char *name)
+{
+	struct audio_source_instance *fi_audio;
+	char *ptr;
+	int name_len;
+
+	name_len = strlen(name) + 1;
+	if (name_len > MAX_INST_NAME_LEN)
+		return -ENAMETOOLONG;
+
+	ptr = kstrndup(name, name_len, GFP_KERNEL);
+	if (!ptr)
+		return -ENOMEM;
+
+	fi_audio = to_fi_audio_source(fi);
+	fi_audio->name = ptr;
+
+	return 0;
+}
+
+static void audio_source_free_inst(struct usb_function_instance *fi)
+{
+	struct audio_source_instance *fi_audio;
+
+	fi_audio = to_fi_audio_source(fi);
+	device_destroy(fi_audio->audio_device->class,
+			fi_audio->audio_device->devt);
+	kfree(fi_audio->name);
+	kfree(fi_audio->config);
+}
+
+static ssize_t audio_source_pcm_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct audio_source_instance *fi_audio = dev_get_drvdata(dev);
+	struct audio_source_config *config = fi_audio->config;
+
+	/* print PCM card and device numbers */
+	return sprintf(buf, "%d %d\n", config->card, config->device);
+}
+
+struct device *create_function_device(char *name);
+
+static struct usb_function_instance *audio_source_alloc_inst(void)
+{
+	struct audio_source_instance *fi_audio;
+	struct device_attribute **attrs;
+	struct device_attribute *attr;
+	struct device *dev;
+	void *err_ptr;
+	int err = 0;
+
+	fi_audio = kzalloc(sizeof(*fi_audio), GFP_KERNEL);
+	if (!fi_audio)
+		return ERR_PTR(-ENOMEM);
+
+	fi_audio->func_inst.set_inst_name = audio_source_set_inst_name;
+	fi_audio->func_inst.free_func_inst = audio_source_free_inst;
+
+	fi_audio->config = kzalloc(sizeof(struct audio_source_config),
+							GFP_KERNEL);
+	if (!fi_audio->config) {
+		err_ptr = ERR_PTR(-ENOMEM);
+		goto fail_audio;
+	}
+
+	config_group_init_type_name(&fi_audio->func_inst.group, "",
+						&audio_source_func_type);
+	dev = create_function_device("f_audio_source");
+
+	if (IS_ERR(dev)) {
+		err_ptr = dev;
+		goto fail_audio_config;
+	}
+
+	fi_audio->config->card = -1;
+	fi_audio->config->device = -1;
+	fi_audio->audio_device = dev;
+
+	attrs = audio_source_function_attributes;
+	if (attrs) {
+		while ((attr = *attrs++) && !err)
+			err = device_create_file(dev, attr);
+		if (err) {
+			err_ptr = ERR_PTR(-EINVAL);
+			goto fail_device;
+		}
+	}
+
+	dev_set_drvdata(dev, fi_audio);
+	_audio_dev.config = fi_audio->config;
+
+	return  &fi_audio->func_inst;
+
+fail_device:
+	device_destroy(dev->class, dev->devt);
+fail_audio_config:
+	kfree(fi_audio->config);
+fail_audio:
+	kfree(fi_audio);
+	return err_ptr;
+
+}
+
+static struct usb_function *audio_source_alloc(struct usb_function_instance *fi)
+{
+	return &_audio_dev.func;
+}
+
+DECLARE_USB_FUNCTION_INIT(audio_source, audio_source_alloc_inst,
+			audio_source_alloc);
+MODULE_LICENSE("GPL");
diff --git a/drivers/usb/gadget/function/f_midi.c b/drivers/usb/gadget/function/f_midi.c
index af60cc3714c1..315c2690afe1 100644
--- a/drivers/usb/gadget/function/f_midi.c
+++ b/drivers/usb/gadget/function/f_midi.c
@@ -1042,6 +1042,65 @@ static void f_midi_free_inst(struct usb_function_instance *f)
 	kfree(opts);
 }
 
+#ifdef CONFIG_USB_CONFIGFS_UEVENT
+extern struct device *create_function_device(char *name);
+static ssize_t alsa_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct usb_function_instance *fi_midi = dev_get_drvdata(dev);
+	struct f_midi *midi;
+
+	if (!fi_midi->f)
+		dev_warn(dev, "f_midi: function not set\n");
+
+	if (fi_midi && fi_midi->f) {
+		midi = func_to_midi(fi_midi->f);
+		if (midi->rmidi && midi->rmidi->card)
+			return sprintf(buf, "%d %d\n",
+			midi->rmidi->card->number, midi->rmidi->device);
+	}
+
+	/* print PCM card and device numbers */
+	return sprintf(buf, "%d %d\n", -1, -1);
+}
+
+static DEVICE_ATTR(alsa, S_IRUGO, alsa_show, NULL);
+
+static struct device_attribute *alsa_function_attributes[] = {
+	&dev_attr_alsa,
+	NULL
+};
+
+static int create_alsa_device(struct usb_function_instance *fi)
+{
+	struct device *dev;
+	struct device_attribute **attrs;
+	struct device_attribute *attr;
+	int err = 0;
+
+	dev = create_function_device("f_midi");
+	if (IS_ERR(dev))
+		return PTR_ERR(dev);
+
+	attrs = alsa_function_attributes;
+	if (attrs) {
+		while ((attr = *attrs++) && !err)
+			err = device_create_file(dev, attr);
+		if (err) {
+			device_destroy(dev->class, dev->devt);
+			return -EINVAL;
+		}
+	}
+	dev_set_drvdata(dev, fi);
+	return 0;
+}
+#else
+static int create_alsa_device(struct usb_function_instance *fi)
+{
+	return 0;
+}
+#endif
+
 static struct usb_function_instance *f_midi_alloc_inst(void)
 {
 	struct f_midi_opts *opts;
@@ -1059,6 +1118,11 @@ static struct usb_function_instance *f_midi_alloc_inst(void)
 	opts->in_ports = 1;
 	opts->out_ports = 1;
 
+	if (create_alsa_device(&opts->func_inst)) {
+		kfree(opts);
+		return ERR_PTR(-ENODEV);
+	}
+
 	config_group_init_type_name(&opts->func_inst.group, "",
 				    &midi_func_type);
 
@@ -1078,6 +1142,7 @@ static void f_midi_free(struct usb_function *f)
 	for (i = opts->in_ports - 1; i >= 0; --i)
 		kfree(midi->in_port[i]);
 	kfree(midi);
+	opts->func_inst.f = NULL;
 	--opts->refcnt;
 	mutex_unlock(&opts->lock);
 }
@@ -1160,6 +1225,7 @@ static struct usb_function *f_midi_alloc(struct usb_function_instance *fi)
 	midi->func.disable	= f_midi_disable;
 	midi->func.free_func	= f_midi_free;
 
+	fi->f = &midi->func;
 	return &midi->func;
 
 setup_fail:
diff --git a/drivers/usb/gadget/function/f_mtp.c b/drivers/usb/gadget/function/f_mtp.c
new file mode 100644
index 000000000000..b25cb3594d01
--- /dev/null
+++ b/drivers/usb/gadget/function/f_mtp.c
@@ -0,0 +1,1553 @@
+/*
+ * Gadget Function Driver for MTP
+ *
+ * Copyright (C) 2010 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+/* #define DEBUG */
+/* #define VERBOSE_DEBUG */
+
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/poll.h>
+#include <linux/delay.h>
+#include <linux/wait.h>
+#include <linux/err.h>
+#include <linux/interrupt.h>
+
+#include <linux/types.h>
+#include <linux/file.h>
+#include <linux/device.h>
+#include <linux/miscdevice.h>
+
+#include <linux/usb.h>
+#include <linux/usb_usual.h>
+#include <linux/usb/ch9.h>
+#include <linux/usb/f_mtp.h>
+#include <linux/configfs.h>
+#include <linux/usb/composite.h>
+
+#include "configfs.h"
+
+#define MTP_BULK_BUFFER_SIZE       16384
+#define INTR_BUFFER_SIZE           28
+#define MAX_INST_NAME_LEN          40
+#define MTP_MAX_FILE_SIZE          0xFFFFFFFFL
+
+/* String IDs */
+#define INTERFACE_STRING_INDEX	0
+
+/* values for mtp_dev.state */
+#define STATE_OFFLINE               0   /* initial state, disconnected */
+#define STATE_READY                 1   /* ready for userspace calls */
+#define STATE_BUSY                  2   /* processing userspace calls */
+#define STATE_CANCELED              3   /* transaction canceled by host */
+#define STATE_ERROR                 4   /* error from completion routine */
+
+/* number of tx and rx requests to allocate */
+#define TX_REQ_MAX 4
+#define RX_REQ_MAX 2
+#define INTR_REQ_MAX 5
+
+/* ID for Microsoft MTP OS String */
+#define MTP_OS_STRING_ID   0xEE
+
+/* MTP class reqeusts */
+#define MTP_REQ_CANCEL              0x64
+#define MTP_REQ_GET_EXT_EVENT_DATA  0x65
+#define MTP_REQ_RESET               0x66
+#define MTP_REQ_GET_DEVICE_STATUS   0x67
+
+/* constants for device status */
+#define MTP_RESPONSE_OK             0x2001
+#define MTP_RESPONSE_DEVICE_BUSY    0x2019
+#define DRIVER_NAME "mtp"
+
+static const char mtp_shortname[] = DRIVER_NAME "_usb";
+
+struct mtp_dev {
+	struct usb_function function;
+	struct usb_composite_dev *cdev;
+	spinlock_t lock;
+
+	struct usb_ep *ep_in;
+	struct usb_ep *ep_out;
+	struct usb_ep *ep_intr;
+
+	int state;
+
+	/* synchronize access to our device file */
+	atomic_t open_excl;
+	/* to enforce only one ioctl at a time */
+	atomic_t ioctl_excl;
+
+	struct list_head tx_idle;
+	struct list_head intr_idle;
+
+	wait_queue_head_t read_wq;
+	wait_queue_head_t write_wq;
+	wait_queue_head_t intr_wq;
+	struct usb_request *rx_req[RX_REQ_MAX];
+	int rx_done;
+
+	/* for processing MTP_SEND_FILE, MTP_RECEIVE_FILE and
+	 * MTP_SEND_FILE_WITH_HEADER ioctls on a work queue
+	 */
+	struct workqueue_struct *wq;
+	struct work_struct send_file_work;
+	struct work_struct receive_file_work;
+	struct file *xfer_file;
+	loff_t xfer_file_offset;
+	int64_t xfer_file_length;
+	unsigned xfer_send_header;
+	uint16_t xfer_command;
+	uint32_t xfer_transaction_id;
+	int xfer_result;
+};
+
+static struct usb_interface_descriptor mtp_interface_desc = {
+	.bLength                = USB_DT_INTERFACE_SIZE,
+	.bDescriptorType        = USB_DT_INTERFACE,
+	.bInterfaceNumber       = 0,
+	.bNumEndpoints          = 3,
+	.bInterfaceClass        = USB_CLASS_VENDOR_SPEC,
+	.bInterfaceSubClass     = USB_SUBCLASS_VENDOR_SPEC,
+	.bInterfaceProtocol     = 0,
+};
+
+static struct usb_interface_descriptor ptp_interface_desc = {
+	.bLength                = USB_DT_INTERFACE_SIZE,
+	.bDescriptorType        = USB_DT_INTERFACE,
+	.bInterfaceNumber       = 0,
+	.bNumEndpoints          = 3,
+	.bInterfaceClass        = USB_CLASS_STILL_IMAGE,
+	.bInterfaceSubClass     = 1,
+	.bInterfaceProtocol     = 1,
+};
+
+static struct usb_endpoint_descriptor mtp_ss_in_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_IN,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+	.wMaxPacketSize         = __constant_cpu_to_le16(1024),
+};
+
+static struct usb_ss_ep_comp_descriptor mtp_ss_in_comp_desc = {
+	.bLength                = sizeof(mtp_ss_in_comp_desc),
+	.bDescriptorType        = USB_DT_SS_ENDPOINT_COMP,
+	/* .bMaxBurst           = DYNAMIC, */
+};
+
+static struct usb_endpoint_descriptor mtp_ss_out_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_OUT,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+	.wMaxPacketSize         = __constant_cpu_to_le16(1024),
+};
+
+static struct usb_ss_ep_comp_descriptor mtp_ss_out_comp_desc = {
+	.bLength                = sizeof(mtp_ss_out_comp_desc),
+	.bDescriptorType        = USB_DT_SS_ENDPOINT_COMP,
+	/* .bMaxBurst           = DYNAMIC, */
+};
+
+static struct usb_endpoint_descriptor mtp_highspeed_in_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_IN,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+	.wMaxPacketSize         = __constant_cpu_to_le16(512),
+};
+
+static struct usb_endpoint_descriptor mtp_highspeed_out_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_OUT,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+	.wMaxPacketSize         = __constant_cpu_to_le16(512),
+};
+
+static struct usb_endpoint_descriptor mtp_fullspeed_in_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_IN,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+};
+
+static struct usb_endpoint_descriptor mtp_fullspeed_out_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_OUT,
+	.bmAttributes           = USB_ENDPOINT_XFER_BULK,
+};
+
+static struct usb_endpoint_descriptor mtp_intr_desc = {
+	.bLength                = USB_DT_ENDPOINT_SIZE,
+	.bDescriptorType        = USB_DT_ENDPOINT,
+	.bEndpointAddress       = USB_DIR_IN,
+	.bmAttributes           = USB_ENDPOINT_XFER_INT,
+	.wMaxPacketSize         = __constant_cpu_to_le16(INTR_BUFFER_SIZE),
+	.bInterval              = 6,
+};
+
+static struct usb_ss_ep_comp_descriptor mtp_intr_ss_comp_desc = {
+	.bLength                = sizeof(mtp_intr_ss_comp_desc),
+	.bDescriptorType        = USB_DT_SS_ENDPOINT_COMP,
+	.wBytesPerInterval      = cpu_to_le16(INTR_BUFFER_SIZE),
+};
+
+static struct usb_descriptor_header *fs_mtp_descs[] = {
+	(struct usb_descriptor_header *) &mtp_interface_desc,
+	(struct usb_descriptor_header *) &mtp_fullspeed_in_desc,
+	(struct usb_descriptor_header *) &mtp_fullspeed_out_desc,
+	(struct usb_descriptor_header *) &mtp_intr_desc,
+	NULL,
+};
+
+static struct usb_descriptor_header *hs_mtp_descs[] = {
+	(struct usb_descriptor_header *) &mtp_interface_desc,
+	(struct usb_descriptor_header *) &mtp_highspeed_in_desc,
+	(struct usb_descriptor_header *) &mtp_highspeed_out_desc,
+	(struct usb_descriptor_header *) &mtp_intr_desc,
+	NULL,
+};
+
+static struct usb_descriptor_header *ss_mtp_descs[] = {
+	(struct usb_descriptor_header *) &mtp_interface_desc,
+	(struct usb_descriptor_header *) &mtp_ss_in_desc,
+	(struct usb_descriptor_header *) &mtp_ss_in_comp_desc,
+	(struct usb_descriptor_header *) &mtp_ss_out_desc,
+	(struct usb_descriptor_header *) &mtp_ss_out_comp_desc,
+	(struct usb_descriptor_header *) &mtp_intr_desc,
+	(struct usb_descriptor_header *) &mtp_intr_ss_comp_desc,
+	NULL,
+};
+
+static struct usb_descriptor_header *fs_ptp_descs[] = {
+	(struct usb_descriptor_header *) &ptp_interface_desc,
+	(struct usb_descriptor_header *) &mtp_fullspeed_in_desc,
+	(struct usb_descriptor_header *) &mtp_fullspeed_out_desc,
+	(struct usb_descriptor_header *) &mtp_intr_desc,
+	NULL,
+};
+
+static struct usb_descriptor_header *hs_ptp_descs[] = {
+	(struct usb_descriptor_header *) &ptp_interface_desc,
+	(struct usb_descriptor_header *) &mtp_highspeed_in_desc,
+	(struct usb_descriptor_header *) &mtp_highspeed_out_desc,
+	(struct usb_descriptor_header *) &mtp_intr_desc,
+	NULL,
+};
+
+static struct usb_descriptor_header *ss_ptp_descs[] = {
+	(struct usb_descriptor_header *) &ptp_interface_desc,
+	(struct usb_descriptor_header *) &mtp_ss_in_desc,
+	(struct usb_descriptor_header *) &mtp_ss_in_comp_desc,
+	(struct usb_descriptor_header *) &mtp_ss_out_desc,
+	(struct usb_descriptor_header *) &mtp_ss_out_comp_desc,
+	(struct usb_descriptor_header *) &mtp_intr_desc,
+	(struct usb_descriptor_header *) &mtp_intr_ss_comp_desc,
+	NULL,
+};
+
+static struct usb_string mtp_string_defs[] = {
+	/* Naming interface "MTP" so libmtp will recognize us */
+	[INTERFACE_STRING_INDEX].s	= "MTP",
+	{  },	/* end of list */
+};
+
+static struct usb_gadget_strings mtp_string_table = {
+	.language		= 0x0409,	/* en-US */
+	.strings		= mtp_string_defs,
+};
+
+static struct usb_gadget_strings *mtp_strings[] = {
+	&mtp_string_table,
+	NULL,
+};
+
+/* Microsoft MTP OS String */
+static u8 mtp_os_string[] = {
+	18, /* sizeof(mtp_os_string) */
+	USB_DT_STRING,
+	/* Signature field: "MSFT100" */
+	'M', 0, 'S', 0, 'F', 0, 'T', 0, '1', 0, '0', 0, '0', 0,
+	/* vendor code */
+	1,
+	/* padding */
+	0
+};
+
+/* Microsoft Extended Configuration Descriptor Header Section */
+struct mtp_ext_config_desc_header {
+	__le32	dwLength;
+	__u16	bcdVersion;
+	__le16	wIndex;
+	__u8	bCount;
+	__u8	reserved[7];
+};
+
+/* Microsoft Extended Configuration Descriptor Function Section */
+struct mtp_ext_config_desc_function {
+	__u8	bFirstInterfaceNumber;
+	__u8	bInterfaceCount;
+	__u8	compatibleID[8];
+	__u8	subCompatibleID[8];
+	__u8	reserved[6];
+};
+
+/* MTP Extended Configuration Descriptor */
+struct {
+	struct mtp_ext_config_desc_header	header;
+	struct mtp_ext_config_desc_function    function;
+} mtp_ext_config_desc = {
+	.header = {
+		.dwLength = __constant_cpu_to_le32(sizeof(mtp_ext_config_desc)),
+		.bcdVersion = __constant_cpu_to_le16(0x0100),
+		.wIndex = __constant_cpu_to_le16(4),
+		.bCount = 1,
+	},
+	.function = {
+		.bFirstInterfaceNumber = 0,
+		.bInterfaceCount = 1,
+		.compatibleID = { 'M', 'T', 'P' },
+	},
+};
+
+struct mtp_device_status {
+	__le16	wLength;
+	__le16	wCode;
+};
+
+struct mtp_data_header {
+	/* length of packet, including this header */
+	__le32	length;
+	/* container type (2 for data packet) */
+	__le16	type;
+	/* MTP command code */
+	__le16	command;
+	/* MTP transaction ID */
+	__le32	transaction_id;
+};
+
+struct mtp_instance {
+	struct usb_function_instance func_inst;
+	const char *name;
+	struct mtp_dev *dev;
+	char mtp_ext_compat_id[16];
+	struct usb_os_desc mtp_os_desc;
+};
+
+/* temporary variable used between mtp_open() and mtp_gadget_bind() */
+static struct mtp_dev *_mtp_dev;
+
+static inline struct mtp_dev *func_to_mtp(struct usb_function *f)
+{
+	return container_of(f, struct mtp_dev, function);
+}
+
+static struct usb_request *mtp_request_new(struct usb_ep *ep, int buffer_size)
+{
+	struct usb_request *req = usb_ep_alloc_request(ep, GFP_KERNEL);
+
+	if (!req)
+		return NULL;
+
+	/* now allocate buffers for the requests */
+	req->buf = kmalloc(buffer_size, GFP_KERNEL);
+	if (!req->buf) {
+		usb_ep_free_request(ep, req);
+		return NULL;
+	}
+
+	return req;
+}
+
+static void mtp_request_free(struct usb_request *req, struct usb_ep *ep)
+{
+	if (req) {
+		kfree(req->buf);
+		usb_ep_free_request(ep, req);
+	}
+}
+
+static inline int mtp_lock(atomic_t *excl)
+{
+	if (atomic_inc_return(excl) == 1) {
+		return 0;
+	} else {
+		atomic_dec(excl);
+		return -1;
+	}
+}
+
+static inline void mtp_unlock(atomic_t *excl)
+{
+	atomic_dec(excl);
+}
+
+/* add a request to the tail of a list */
+static void mtp_req_put(struct mtp_dev *dev, struct list_head *head,
+		struct usb_request *req)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	list_add_tail(&req->list, head);
+	spin_unlock_irqrestore(&dev->lock, flags);
+}
+
+/* remove a request from the head of a list */
+static struct usb_request
+*mtp_req_get(struct mtp_dev *dev, struct list_head *head)
+{
+	unsigned long flags;
+	struct usb_request *req;
+
+	spin_lock_irqsave(&dev->lock, flags);
+	if (list_empty(head)) {
+		req = 0;
+	} else {
+		req = list_first_entry(head, struct usb_request, list);
+		list_del(&req->list);
+	}
+	spin_unlock_irqrestore(&dev->lock, flags);
+	return req;
+}
+
+static void mtp_complete_in(struct usb_ep *ep, struct usb_request *req)
+{
+	struct mtp_dev *dev = _mtp_dev;
+
+	if (req->status != 0)
+		dev->state = STATE_ERROR;
+
+	mtp_req_put(dev, &dev->tx_idle, req);
+
+	wake_up(&dev->write_wq);
+}
+
+static void mtp_complete_out(struct usb_ep *ep, struct usb_request *req)
+{
+	struct mtp_dev *dev = _mtp_dev;
+
+	dev->rx_done = 1;
+	if (req->status != 0)
+		dev->state = STATE_ERROR;
+
+	wake_up(&dev->read_wq);
+}
+
+static void mtp_complete_intr(struct usb_ep *ep, struct usb_request *req)
+{
+	struct mtp_dev *dev = _mtp_dev;
+
+	if (req->status != 0)
+		dev->state = STATE_ERROR;
+
+	mtp_req_put(dev, &dev->intr_idle, req);
+
+	wake_up(&dev->intr_wq);
+}
+
+static int mtp_create_bulk_endpoints(struct mtp_dev *dev,
+				struct usb_endpoint_descriptor *in_desc,
+				struct usb_endpoint_descriptor *out_desc,
+				struct usb_endpoint_descriptor *intr_desc)
+{
+	struct usb_composite_dev *cdev = dev->cdev;
+	struct usb_request *req;
+	struct usb_ep *ep;
+	int i;
+
+	DBG(cdev, "create_bulk_endpoints dev: %p\n", dev);
+
+	ep = usb_ep_autoconfig(cdev->gadget, in_desc);
+	if (!ep) {
+		DBG(cdev, "usb_ep_autoconfig for ep_in failed\n");
+		return -ENODEV;
+	}
+	DBG(cdev, "usb_ep_autoconfig for ep_in got %s\n", ep->name);
+	ep->driver_data = dev;		/* claim the endpoint */
+	dev->ep_in = ep;
+
+	ep = usb_ep_autoconfig(cdev->gadget, out_desc);
+	if (!ep) {
+		DBG(cdev, "usb_ep_autoconfig for ep_out failed\n");
+		return -ENODEV;
+	}
+	DBG(cdev, "usb_ep_autoconfig for mtp ep_out got %s\n", ep->name);
+	ep->driver_data = dev;		/* claim the endpoint */
+	dev->ep_out = ep;
+
+	ep = usb_ep_autoconfig(cdev->gadget, intr_desc);
+	if (!ep) {
+		DBG(cdev, "usb_ep_autoconfig for ep_intr failed\n");
+		return -ENODEV;
+	}
+	DBG(cdev, "usb_ep_autoconfig for mtp ep_intr got %s\n", ep->name);
+	ep->driver_data = dev;		/* claim the endpoint */
+	dev->ep_intr = ep;
+
+	/* now allocate requests for our endpoints */
+	for (i = 0; i < TX_REQ_MAX; i++) {
+		req = mtp_request_new(dev->ep_in, MTP_BULK_BUFFER_SIZE);
+		if (!req)
+			goto fail;
+		req->complete = mtp_complete_in;
+		mtp_req_put(dev, &dev->tx_idle, req);
+	}
+	for (i = 0; i < RX_REQ_MAX; i++) {
+		req = mtp_request_new(dev->ep_out, MTP_BULK_BUFFER_SIZE);
+		if (!req)
+			goto fail;
+		req->complete = mtp_complete_out;
+		dev->rx_req[i] = req;
+	}
+	for (i = 0; i < INTR_REQ_MAX; i++) {
+		req = mtp_request_new(dev->ep_intr, INTR_BUFFER_SIZE);
+		if (!req)
+			goto fail;
+		req->complete = mtp_complete_intr;
+		mtp_req_put(dev, &dev->intr_idle, req);
+	}
+
+	return 0;
+
+fail:
+	pr_err("mtp_bind() could not allocate requests\n");
+	return -1;
+}
+
+static ssize_t mtp_read(struct file *fp, char __user *buf,
+	size_t count, loff_t *pos)
+{
+	struct mtp_dev *dev = fp->private_data;
+	struct usb_composite_dev *cdev = dev->cdev;
+	struct usb_request *req;
+	ssize_t r = count;
+	unsigned xfer;
+	int ret = 0;
+	size_t len = 0;
+
+	DBG(cdev, "mtp_read(%zu)\n", count);
+
+	/* we will block until we're online */
+	DBG(cdev, "mtp_read: waiting for online state\n");
+	ret = wait_event_interruptible(dev->read_wq,
+		dev->state != STATE_OFFLINE);
+	if (ret < 0) {
+		r = ret;
+		goto done;
+	}
+	spin_lock_irq(&dev->lock);
+	if (dev->ep_out->desc) {
+		len = usb_ep_align_maybe(cdev->gadget, dev->ep_out, count);
+		if (len > MTP_BULK_BUFFER_SIZE) {
+			spin_unlock_irq(&dev->lock);
+			return -EINVAL;
+		}
+	}
+
+	if (dev->state == STATE_CANCELED) {
+		/* report cancelation to userspace */
+		dev->state = STATE_READY;
+		spin_unlock_irq(&dev->lock);
+		return -ECANCELED;
+	}
+	dev->state = STATE_BUSY;
+	spin_unlock_irq(&dev->lock);
+
+requeue_req:
+	/* queue a request */
+	req = dev->rx_req[0];
+	req->length = len;
+	dev->rx_done = 0;
+	ret = usb_ep_queue(dev->ep_out, req, GFP_KERNEL);
+	if (ret < 0) {
+		r = -EIO;
+		goto done;
+	} else {
+		DBG(cdev, "rx %p queue\n", req);
+	}
+
+	/* wait for a request to complete */
+	ret = wait_event_interruptible(dev->read_wq, dev->rx_done);
+	if (ret < 0) {
+		r = ret;
+		usb_ep_dequeue(dev->ep_out, req);
+		goto done;
+	}
+	if (dev->state == STATE_BUSY) {
+		/* If we got a 0-len packet, throw it back and try again. */
+		if (req->actual == 0)
+			goto requeue_req;
+
+		DBG(cdev, "rx %p %d\n", req, req->actual);
+		xfer = (req->actual < count) ? req->actual : count;
+		r = xfer;
+		if (copy_to_user(buf, req->buf, xfer))
+			r = -EFAULT;
+	} else
+		r = -EIO;
+
+done:
+	spin_lock_irq(&dev->lock);
+	if (dev->state == STATE_CANCELED)
+		r = -ECANCELED;
+	else if (dev->state != STATE_OFFLINE)
+		dev->state = STATE_READY;
+	spin_unlock_irq(&dev->lock);
+
+	DBG(cdev, "mtp_read returning %zd\n", r);
+	return r;
+}
+
+static ssize_t mtp_write(struct file *fp, const char __user *buf,
+	size_t count, loff_t *pos)
+{
+	struct mtp_dev *dev = fp->private_data;
+	struct usb_composite_dev *cdev = dev->cdev;
+	struct usb_request *req = 0;
+	ssize_t r = count;
+	unsigned xfer;
+	int sendZLP = 0;
+	int ret;
+
+	DBG(cdev, "mtp_write(%zu)\n", count);
+
+	spin_lock_irq(&dev->lock);
+	if (dev->state == STATE_CANCELED) {
+		/* report cancelation to userspace */
+		dev->state = STATE_READY;
+		spin_unlock_irq(&dev->lock);
+		return -ECANCELED;
+	}
+	if (dev->state == STATE_OFFLINE) {
+		spin_unlock_irq(&dev->lock);
+		return -ENODEV;
+	}
+	dev->state = STATE_BUSY;
+	spin_unlock_irq(&dev->lock);
+
+	/* we need to send a zero length packet to signal the end of transfer
+	 * if the transfer size is aligned to a packet boundary.
+	 */
+	if ((count & (dev->ep_in->maxpacket - 1)) == 0)
+		sendZLP = 1;
+
+	while (count > 0 || sendZLP) {
+		/* so we exit after sending ZLP */
+		if (count == 0)
+			sendZLP = 0;
+
+		if (dev->state != STATE_BUSY) {
+			DBG(cdev, "mtp_write dev->error\n");
+			r = -EIO;
+			break;
+		}
+
+		/* get an idle tx request to use */
+		req = 0;
+		ret = wait_event_interruptible(dev->write_wq,
+			((req = mtp_req_get(dev, &dev->tx_idle))
+				|| dev->state != STATE_BUSY));
+		if (!req) {
+			r = ret;
+			break;
+		}
+
+		if (count > MTP_BULK_BUFFER_SIZE)
+			xfer = MTP_BULK_BUFFER_SIZE;
+		else
+			xfer = count;
+		if (xfer && copy_from_user(req->buf, buf, xfer)) {
+			r = -EFAULT;
+			break;
+		}
+
+		req->length = xfer;
+		ret = usb_ep_queue(dev->ep_in, req, GFP_KERNEL);
+		if (ret < 0) {
+			DBG(cdev, "mtp_write: xfer error %d\n", ret);
+			r = -EIO;
+			break;
+		}
+
+		buf += xfer;
+		count -= xfer;
+
+		/* zero this so we don't try to free it on error exit */
+		req = 0;
+	}
+
+	if (req)
+		mtp_req_put(dev, &dev->tx_idle, req);
+
+	spin_lock_irq(&dev->lock);
+	if (dev->state == STATE_CANCELED)
+		r = -ECANCELED;
+	else if (dev->state != STATE_OFFLINE)
+		dev->state = STATE_READY;
+	spin_unlock_irq(&dev->lock);
+
+	DBG(cdev, "mtp_write returning %zd\n", r);
+	return r;
+}
+
+/* read from a local file and write to USB */
+static void send_file_work(struct work_struct *data)
+{
+	struct mtp_dev *dev = container_of(data, struct mtp_dev,
+						send_file_work);
+	struct usb_composite_dev *cdev = dev->cdev;
+	struct usb_request *req = 0;
+	struct mtp_data_header *header;
+	struct file *filp;
+	loff_t offset;
+	int64_t count;
+	int xfer, ret, hdr_size;
+	int r = 0;
+	int sendZLP = 0;
+
+	/* read our parameters */
+	smp_rmb();
+	filp = dev->xfer_file;
+	offset = dev->xfer_file_offset;
+	count = dev->xfer_file_length;
+
+	DBG(cdev, "send_file_work(%lld %lld)\n", offset, count);
+
+	if (dev->xfer_send_header) {
+		hdr_size = sizeof(struct mtp_data_header);
+		count += hdr_size;
+	} else {
+		hdr_size = 0;
+	}
+
+	/* we need to send a zero length packet to signal the end of transfer
+	 * if the transfer size is aligned to a packet boundary.
+	 */
+	if ((count & (dev->ep_in->maxpacket - 1)) == 0)
+		sendZLP = 1;
+
+	while (count > 0 || sendZLP) {
+		/* so we exit after sending ZLP */
+		if (count == 0)
+			sendZLP = 0;
+
+		/* get an idle tx request to use */
+		req = 0;
+		ret = wait_event_interruptible(dev->write_wq,
+			(req = mtp_req_get(dev, &dev->tx_idle))
+			|| dev->state != STATE_BUSY);
+		if (dev->state == STATE_CANCELED) {
+			r = -ECANCELED;
+			break;
+		}
+		if (!req) {
+			r = ret;
+			break;
+		}
+
+		if (count > MTP_BULK_BUFFER_SIZE)
+			xfer = MTP_BULK_BUFFER_SIZE;
+		else
+			xfer = count;
+
+		if (hdr_size) {
+			/* prepend MTP data header */
+			header = (struct mtp_data_header *)req->buf;
+			/*
+                         * set file size with header according to
+                         * MTP Specification v1.0
+                         */
+			header->length = (count > MTP_MAX_FILE_SIZE) ?
+				MTP_MAX_FILE_SIZE : __cpu_to_le32(count);
+			header->type = __cpu_to_le16(2); /* data packet */
+			header->command = __cpu_to_le16(dev->xfer_command);
+			header->transaction_id =
+					__cpu_to_le32(dev->xfer_transaction_id);
+		}
+
+		ret = vfs_read(filp, req->buf + hdr_size, xfer - hdr_size,
+								&offset);
+		if (ret < 0) {
+			r = ret;
+			break;
+		}
+		xfer = ret + hdr_size;
+		hdr_size = 0;
+
+		req->length = xfer;
+		ret = usb_ep_queue(dev->ep_in, req, GFP_KERNEL);
+		if (ret < 0) {
+			DBG(cdev, "send_file_work: xfer error %d\n", ret);
+			dev->state = STATE_ERROR;
+			r = -EIO;
+			break;
+		}
+
+		count -= xfer;
+
+		/* zero this so we don't try to free it on error exit */
+		req = 0;
+	}
+
+	if (req)
+		mtp_req_put(dev, &dev->tx_idle, req);
+
+	DBG(cdev, "send_file_work returning %d\n", r);
+	/* write the result */
+	dev->xfer_result = r;
+	smp_wmb();
+}
+
+/* read from USB and write to a local file */
+static void receive_file_work(struct work_struct *data)
+{
+	struct mtp_dev *dev = container_of(data, struct mtp_dev,
+						receive_file_work);
+	struct usb_composite_dev *cdev = dev->cdev;
+	struct usb_request *read_req = NULL, *write_req = NULL;
+	struct file *filp;
+	loff_t offset;
+	int64_t count;
+	int ret, cur_buf = 0;
+	int r = 0;
+
+	/* read our parameters */
+	smp_rmb();
+	filp = dev->xfer_file;
+	offset = dev->xfer_file_offset;
+	count = dev->xfer_file_length;
+
+	DBG(cdev, "receive_file_work(%lld)\n", count);
+
+	while (count > 0 || write_req) {
+		if (count > 0) {
+			/* queue a request */
+			read_req = dev->rx_req[cur_buf];
+			cur_buf = (cur_buf + 1) % RX_REQ_MAX;
+
+			read_req->length = (count > MTP_BULK_BUFFER_SIZE
+					? MTP_BULK_BUFFER_SIZE : count);
+			dev->rx_done = 0;
+			ret = usb_ep_queue(dev->ep_out, read_req, GFP_KERNEL);
+			if (ret < 0) {
+				r = -EIO;
+				dev->state = STATE_ERROR;
+				break;
+			}
+		}
+
+		if (write_req) {
+			DBG(cdev, "rx %p %d\n", write_req, write_req->actual);
+			ret = vfs_write(filp, write_req->buf, write_req->actual,
+				&offset);
+			DBG(cdev, "vfs_write %d\n", ret);
+			if (ret != write_req->actual) {
+				r = -EIO;
+				dev->state = STATE_ERROR;
+				break;
+			}
+			write_req = NULL;
+		}
+
+		if (read_req) {
+			/* wait for our last read to complete */
+			ret = wait_event_interruptible(dev->read_wq,
+				dev->rx_done || dev->state != STATE_BUSY);
+			if (dev->state == STATE_CANCELED) {
+				r = -ECANCELED;
+				if (!dev->rx_done)
+					usb_ep_dequeue(dev->ep_out, read_req);
+				break;
+			}
+			if (read_req->status) {
+				r = read_req->status;
+				break;
+			}
+			/* if xfer_file_length is 0xFFFFFFFF, then we read until
+			 * we get a zero length packet
+			 */
+			if (count != 0xFFFFFFFF)
+				count -= read_req->actual;
+			if (read_req->actual < read_req->length) {
+				/*
+				 * short packet is used to signal EOF for
+				 * sizes > 4 gig
+				 */
+				DBG(cdev, "got short packet\n");
+				count = 0;
+			}
+
+			write_req = read_req;
+			read_req = NULL;
+		}
+	}
+
+	DBG(cdev, "receive_file_work returning %d\n", r);
+	/* write the result */
+	dev->xfer_result = r;
+	smp_wmb();
+}
+
+static int mtp_send_event(struct mtp_dev *dev, struct mtp_event *event)
+{
+	struct usb_request *req = NULL;
+	int ret;
+	int length = event->length;
+
+	DBG(dev->cdev, "mtp_send_event(%zu)\n", event->length);
+
+	if (length < 0 || length > INTR_BUFFER_SIZE)
+		return -EINVAL;
+	if (dev->state == STATE_OFFLINE)
+		return -ENODEV;
+
+	ret = wait_event_interruptible_timeout(dev->intr_wq,
+			(req = mtp_req_get(dev, &dev->intr_idle)),
+			msecs_to_jiffies(1000));
+	if (!req)
+		return -ETIME;
+
+	if (copy_from_user(req->buf, (void __user *)event->data, length)) {
+		mtp_req_put(dev, &dev->intr_idle, req);
+		return -EFAULT;
+	}
+	req->length = length;
+	ret = usb_ep_queue(dev->ep_intr, req, GFP_KERNEL);
+	if (ret)
+		mtp_req_put(dev, &dev->intr_idle, req);
+
+	return ret;
+}
+
+static long mtp_ioctl(struct file *fp, unsigned code, unsigned long value)
+{
+	struct mtp_dev *dev = fp->private_data;
+	struct file *filp = NULL;
+	int ret = -EINVAL;
+
+	if (mtp_lock(&dev->ioctl_excl))
+		return -EBUSY;
+
+	switch (code) {
+	case MTP_SEND_FILE:
+	case MTP_RECEIVE_FILE:
+	case MTP_SEND_FILE_WITH_HEADER:
+	{
+		struct mtp_file_range	mfr;
+		struct work_struct *work;
+
+		spin_lock_irq(&dev->lock);
+		if (dev->state == STATE_CANCELED) {
+			/* report cancelation to userspace */
+			dev->state = STATE_READY;
+			spin_unlock_irq(&dev->lock);
+			ret = -ECANCELED;
+			goto out;
+		}
+		if (dev->state == STATE_OFFLINE) {
+			spin_unlock_irq(&dev->lock);
+			ret = -ENODEV;
+			goto out;
+		}
+		dev->state = STATE_BUSY;
+		spin_unlock_irq(&dev->lock);
+
+		if (copy_from_user(&mfr, (void __user *)value, sizeof(mfr))) {
+			ret = -EFAULT;
+			goto fail;
+		}
+		/* hold a reference to the file while we are working with it */
+		filp = fget(mfr.fd);
+		if (!filp) {
+			ret = -EBADF;
+			goto fail;
+		}
+
+		/* write the parameters */
+		dev->xfer_file = filp;
+		dev->xfer_file_offset = mfr.offset;
+		dev->xfer_file_length = mfr.length;
+		smp_wmb();
+
+		if (code == MTP_SEND_FILE_WITH_HEADER) {
+			work = &dev->send_file_work;
+			dev->xfer_send_header = 1;
+			dev->xfer_command = mfr.command;
+			dev->xfer_transaction_id = mfr.transaction_id;
+		} else if (code == MTP_SEND_FILE) {
+			work = &dev->send_file_work;
+			dev->xfer_send_header = 0;
+		} else {
+			work = &dev->receive_file_work;
+		}
+
+		/* We do the file transfer on a work queue so it will run
+		 * in kernel context, which is necessary for vfs_read and
+		 * vfs_write to use our buffers in the kernel address space.
+		 */
+		queue_work(dev->wq, work);
+		/* wait for operation to complete */
+		flush_workqueue(dev->wq);
+		fput(filp);
+
+		/* read the result */
+		smp_rmb();
+		ret = dev->xfer_result;
+		break;
+	}
+	case MTP_SEND_EVENT:
+	{
+		struct mtp_event	event;
+		/* return here so we don't change dev->state below,
+		 * which would interfere with bulk transfer state.
+		 */
+		if (copy_from_user(&event, (void __user *)value, sizeof(event)))
+			ret = -EFAULT;
+		else
+			ret = mtp_send_event(dev, &event);
+		goto out;
+	}
+	}
+
+fail:
+	spin_lock_irq(&dev->lock);
+	if (dev->state == STATE_CANCELED)
+		ret = -ECANCELED;
+	else if (dev->state != STATE_OFFLINE)
+		dev->state = STATE_READY;
+	spin_unlock_irq(&dev->lock);
+out:
+	mtp_unlock(&dev->ioctl_excl);
+	DBG(dev->cdev, "ioctl returning %d\n", ret);
+	return ret;
+}
+
+static int mtp_open(struct inode *ip, struct file *fp)
+{
+	printk(KERN_INFO "mtp_open\n");
+	if (mtp_lock(&_mtp_dev->open_excl))
+		return -EBUSY;
+
+	/* clear any error condition */
+	if (_mtp_dev->state != STATE_OFFLINE)
+		_mtp_dev->state = STATE_READY;
+
+	fp->private_data = _mtp_dev;
+	return 0;
+}
+
+static int mtp_release(struct inode *ip, struct file *fp)
+{
+	printk(KERN_INFO "mtp_release\n");
+
+	mtp_unlock(&_mtp_dev->open_excl);
+	return 0;
+}
+
+/* file operations for /dev/mtp_usb */
+static const struct file_operations mtp_fops = {
+	.owner = THIS_MODULE,
+	.read = mtp_read,
+	.write = mtp_write,
+	.unlocked_ioctl = mtp_ioctl,
+	.open = mtp_open,
+	.release = mtp_release,
+};
+
+static struct miscdevice mtp_device = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = mtp_shortname,
+	.fops = &mtp_fops,
+};
+
+static int mtp_ctrlrequest(struct usb_composite_dev *cdev,
+				const struct usb_ctrlrequest *ctrl)
+{
+	struct mtp_dev *dev = _mtp_dev;
+	int	value = -EOPNOTSUPP;
+	u16	w_index = le16_to_cpu(ctrl->wIndex);
+	u16	w_value = le16_to_cpu(ctrl->wValue);
+	u16	w_length = le16_to_cpu(ctrl->wLength);
+	unsigned long	flags;
+
+	VDBG(cdev, "mtp_ctrlrequest "
+			"%02x.%02x v%04x i%04x l%u\n",
+			ctrl->bRequestType, ctrl->bRequest,
+			w_value, w_index, w_length);
+
+	/* Handle MTP OS string */
+	if (ctrl->bRequestType ==
+			(USB_DIR_IN | USB_TYPE_STANDARD | USB_RECIP_DEVICE)
+			&& ctrl->bRequest == USB_REQ_GET_DESCRIPTOR
+			&& (w_value >> 8) == USB_DT_STRING
+			&& (w_value & 0xFF) == MTP_OS_STRING_ID) {
+		value = (w_length < sizeof(mtp_os_string)
+				? w_length : sizeof(mtp_os_string));
+		memcpy(cdev->req->buf, mtp_os_string, value);
+	} else if ((ctrl->bRequestType & USB_TYPE_MASK) == USB_TYPE_VENDOR) {
+		/* Handle MTP OS descriptor */
+		DBG(cdev, "vendor request: %d index: %d value: %d length: %d\n",
+			ctrl->bRequest, w_index, w_value, w_length);
+
+		if (ctrl->bRequest == 1
+				&& (ctrl->bRequestType & USB_DIR_IN)
+				&& (w_index == 4 || w_index == 5)) {
+			value = (w_length < sizeof(mtp_ext_config_desc) ?
+					w_length : sizeof(mtp_ext_config_desc));
+			memcpy(cdev->req->buf, &mtp_ext_config_desc, value);
+		}
+	} else if ((ctrl->bRequestType & USB_TYPE_MASK) == USB_TYPE_CLASS) {
+		DBG(cdev, "class request: %d index: %d value: %d length: %d\n",
+			ctrl->bRequest, w_index, w_value, w_length);
+
+		if (ctrl->bRequest == MTP_REQ_CANCEL && w_index == 0
+				&& w_value == 0) {
+			DBG(cdev, "MTP_REQ_CANCEL\n");
+
+			spin_lock_irqsave(&dev->lock, flags);
+			if (dev->state == STATE_BUSY) {
+				dev->state = STATE_CANCELED;
+				wake_up(&dev->read_wq);
+				wake_up(&dev->write_wq);
+			}
+			spin_unlock_irqrestore(&dev->lock, flags);
+
+			/* We need to queue a request to read the remaining
+			 *  bytes, but we don't actually need to look at
+			 * the contents.
+			 */
+			value = w_length;
+		} else if (ctrl->bRequest == MTP_REQ_GET_DEVICE_STATUS
+				&& w_index == 0 && w_value == 0) {
+			struct mtp_device_status *status = cdev->req->buf;
+
+			status->wLength =
+				__constant_cpu_to_le16(sizeof(*status));
+
+			DBG(cdev, "MTP_REQ_GET_DEVICE_STATUS\n");
+			spin_lock_irqsave(&dev->lock, flags);
+			/* device status is "busy" until we report
+			 * the cancelation to userspace
+			 */
+			if (dev->state == STATE_CANCELED)
+				status->wCode =
+					__cpu_to_le16(MTP_RESPONSE_DEVICE_BUSY);
+			else
+				status->wCode =
+					__cpu_to_le16(MTP_RESPONSE_OK);
+			spin_unlock_irqrestore(&dev->lock, flags);
+			value = sizeof(*status);
+		}
+	}
+
+	/* respond with data transfer or status phase? */
+	if (value >= 0) {
+		int rc;
+
+		cdev->req->zero = value < w_length;
+		cdev->req->length = value;
+		rc = usb_ep_queue(cdev->gadget->ep0, cdev->req, GFP_ATOMIC);
+		if (rc < 0)
+			ERROR(cdev, "%s: response queue error\n", __func__);
+	}
+	return value;
+}
+
+static int
+mtp_function_bind(struct usb_configuration *c, struct usb_function *f)
+{
+	struct usb_composite_dev *cdev = c->cdev;
+	struct mtp_dev	*dev = func_to_mtp(f);
+	int			id;
+	int			ret;
+	struct mtp_instance *fi_mtp;
+
+	dev->cdev = cdev;
+	DBG(cdev, "mtp_function_bind dev: %p\n", dev);
+
+	/* allocate interface ID(s) */
+	id = usb_interface_id(c, f);
+	if (id < 0)
+		return id;
+	mtp_interface_desc.bInterfaceNumber = id;
+
+	if (mtp_string_defs[INTERFACE_STRING_INDEX].id == 0) {
+		ret = usb_string_id(c->cdev);
+		if (ret < 0)
+			return ret;
+		mtp_string_defs[INTERFACE_STRING_INDEX].id = ret;
+		mtp_interface_desc.iInterface = ret;
+	}
+
+	fi_mtp = container_of(f->fi, struct mtp_instance, func_inst);
+
+	if (cdev->use_os_string) {
+		f->os_desc_table = kzalloc(sizeof(*f->os_desc_table),
+					GFP_KERNEL);
+		if (!f->os_desc_table)
+			return -ENOMEM;
+		f->os_desc_n = 1;
+		f->os_desc_table[0].os_desc = &fi_mtp->mtp_os_desc;
+	}
+
+	/* allocate endpoints */
+	ret = mtp_create_bulk_endpoints(dev, &mtp_fullspeed_in_desc,
+			&mtp_fullspeed_out_desc, &mtp_intr_desc);
+	if (ret)
+		return ret;
+
+	/* support high speed hardware */
+	if (gadget_is_dualspeed(c->cdev->gadget)) {
+		mtp_highspeed_in_desc.bEndpointAddress =
+			mtp_fullspeed_in_desc.bEndpointAddress;
+		mtp_highspeed_out_desc.bEndpointAddress =
+			mtp_fullspeed_out_desc.bEndpointAddress;
+	}
+	/* support super speed hardware */
+	if (gadget_is_superspeed(c->cdev->gadget)) {
+		unsigned max_burst;
+
+		/* Calculate bMaxBurst, we know packet size is 1024 */
+		max_burst = min_t(unsigned, MTP_BULK_BUFFER_SIZE / 1024, 15);
+		mtp_ss_in_desc.bEndpointAddress =
+			mtp_fullspeed_in_desc.bEndpointAddress;
+		mtp_ss_in_comp_desc.bMaxBurst = max_burst;
+		mtp_ss_out_desc.bEndpointAddress =
+			mtp_fullspeed_out_desc.bEndpointAddress;
+		mtp_ss_out_comp_desc.bMaxBurst = max_burst;
+	}
+
+	DBG(cdev, "%s speed %s: IN/%s, OUT/%s\n",
+		gadget_is_superspeed(c->cdev->gadget) ? "super" :
+		(gadget_is_dualspeed(c->cdev->gadget) ? "dual" : "full"),
+		f->name, dev->ep_in->name, dev->ep_out->name);
+	return 0;
+}
+
+static void
+mtp_function_unbind(struct usb_configuration *c, struct usb_function *f)
+{
+	struct mtp_dev	*dev = func_to_mtp(f);
+	struct usb_request *req;
+	int i;
+
+	mtp_string_defs[INTERFACE_STRING_INDEX].id = 0;
+	while ((req = mtp_req_get(dev, &dev->tx_idle)))
+		mtp_request_free(req, dev->ep_in);
+	for (i = 0; i < RX_REQ_MAX; i++)
+		mtp_request_free(dev->rx_req[i], dev->ep_out);
+	while ((req = mtp_req_get(dev, &dev->intr_idle)))
+		mtp_request_free(req, dev->ep_intr);
+	dev->state = STATE_OFFLINE;
+	kfree(f->os_desc_table);
+	f->os_desc_n = 0;
+}
+
+static int mtp_function_set_alt(struct usb_function *f,
+		unsigned intf, unsigned alt)
+{
+	struct mtp_dev	*dev = func_to_mtp(f);
+	struct usb_composite_dev *cdev = f->config->cdev;
+	int ret;
+
+	DBG(cdev, "mtp_function_set_alt intf: %d alt: %d\n", intf, alt);
+
+	ret = config_ep_by_speed(cdev->gadget, f, dev->ep_in);
+	if (ret)
+		return ret;
+
+	ret = usb_ep_enable(dev->ep_in);
+	if (ret)
+		return ret;
+
+	ret = config_ep_by_speed(cdev->gadget, f, dev->ep_out);
+	if (ret)
+		return ret;
+
+	ret = usb_ep_enable(dev->ep_out);
+	if (ret) {
+		usb_ep_disable(dev->ep_in);
+		return ret;
+	}
+
+	ret = config_ep_by_speed(cdev->gadget, f, dev->ep_intr);
+	if (ret)
+		return ret;
+
+	ret = usb_ep_enable(dev->ep_intr);
+	if (ret) {
+		usb_ep_disable(dev->ep_out);
+		usb_ep_disable(dev->ep_in);
+		return ret;
+	}
+	dev->state = STATE_READY;
+
+	/* readers may be blocked waiting for us to go online */
+	wake_up(&dev->read_wq);
+	return 0;
+}
+
+static void mtp_function_disable(struct usb_function *f)
+{
+	struct mtp_dev	*dev = func_to_mtp(f);
+	struct usb_composite_dev	*cdev = dev->cdev;
+
+	DBG(cdev, "mtp_function_disable\n");
+	dev->state = STATE_OFFLINE;
+	usb_ep_disable(dev->ep_in);
+	usb_ep_disable(dev->ep_out);
+	usb_ep_disable(dev->ep_intr);
+
+	/* readers may be blocked waiting for us to go online */
+	wake_up(&dev->read_wq);
+
+	VDBG(cdev, "%s disabled\n", dev->function.name);
+}
+
+static int __mtp_setup(struct mtp_instance *fi_mtp)
+{
+	struct mtp_dev *dev;
+	int ret;
+
+	dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+
+	if (fi_mtp != NULL)
+		fi_mtp->dev = dev;
+
+	if (!dev)
+		return -ENOMEM;
+
+	spin_lock_init(&dev->lock);
+	init_waitqueue_head(&dev->read_wq);
+	init_waitqueue_head(&dev->write_wq);
+	init_waitqueue_head(&dev->intr_wq);
+	atomic_set(&dev->open_excl, 0);
+	atomic_set(&dev->ioctl_excl, 0);
+	INIT_LIST_HEAD(&dev->tx_idle);
+	INIT_LIST_HEAD(&dev->intr_idle);
+
+	dev->wq = create_singlethread_workqueue("f_mtp");
+	if (!dev->wq) {
+		ret = -ENOMEM;
+		goto err1;
+	}
+	INIT_WORK(&dev->send_file_work, send_file_work);
+	INIT_WORK(&dev->receive_file_work, receive_file_work);
+
+	_mtp_dev = dev;
+
+	ret = misc_register(&mtp_device);
+	if (ret)
+		goto err2;
+
+	return 0;
+
+err2:
+	destroy_workqueue(dev->wq);
+err1:
+	_mtp_dev = NULL;
+	kfree(dev);
+	printk(KERN_ERR "mtp gadget driver failed to initialize\n");
+	return ret;
+}
+
+static int mtp_setup_configfs(struct mtp_instance *fi_mtp)
+{
+	return __mtp_setup(fi_mtp);
+}
+
+
+static void mtp_cleanup(void)
+{
+	struct mtp_dev *dev = _mtp_dev;
+
+	if (!dev)
+		return;
+
+	misc_deregister(&mtp_device);
+	destroy_workqueue(dev->wq);
+	_mtp_dev = NULL;
+	kfree(dev);
+}
+
+static struct mtp_instance *to_mtp_instance(struct config_item *item)
+{
+	return container_of(to_config_group(item), struct mtp_instance,
+		func_inst.group);
+}
+
+static void mtp_attr_release(struct config_item *item)
+{
+	struct mtp_instance *fi_mtp = to_mtp_instance(item);
+
+	usb_put_function_instance(&fi_mtp->func_inst);
+}
+
+static struct configfs_item_operations mtp_item_ops = {
+	.release        = mtp_attr_release,
+};
+
+static struct config_item_type mtp_func_type = {
+	.ct_item_ops    = &mtp_item_ops,
+	.ct_owner       = THIS_MODULE,
+};
+
+
+static struct mtp_instance *to_fi_mtp(struct usb_function_instance *fi)
+{
+	return container_of(fi, struct mtp_instance, func_inst);
+}
+
+static int mtp_set_inst_name(struct usb_function_instance *fi, const char *name)
+{
+	struct mtp_instance *fi_mtp;
+	char *ptr;
+	int name_len;
+
+	name_len = strlen(name) + 1;
+	if (name_len > MAX_INST_NAME_LEN)
+		return -ENAMETOOLONG;
+
+	ptr = kstrndup(name, name_len, GFP_KERNEL);
+	if (!ptr)
+		return -ENOMEM;
+
+	fi_mtp = to_fi_mtp(fi);
+	fi_mtp->name = ptr;
+
+	return 0;
+}
+
+static void mtp_free_inst(struct usb_function_instance *fi)
+{
+	struct mtp_instance *fi_mtp;
+
+	fi_mtp = to_fi_mtp(fi);
+	kfree(fi_mtp->name);
+	mtp_cleanup();
+	kfree(fi_mtp->mtp_os_desc.group.default_groups);
+	kfree(fi_mtp);
+}
+
+struct usb_function_instance *alloc_inst_mtp_ptp(bool mtp_config)
+{
+	struct mtp_instance *fi_mtp;
+	int ret = 0;
+	struct usb_os_desc *descs[1];
+	char *names[1];
+
+	fi_mtp = kzalloc(sizeof(*fi_mtp), GFP_KERNEL);
+	if (!fi_mtp)
+		return ERR_PTR(-ENOMEM);
+	fi_mtp->func_inst.set_inst_name = mtp_set_inst_name;
+	fi_mtp->func_inst.free_func_inst = mtp_free_inst;
+
+	fi_mtp->mtp_os_desc.ext_compat_id = fi_mtp->mtp_ext_compat_id;
+	INIT_LIST_HEAD(&fi_mtp->mtp_os_desc.ext_prop);
+	descs[0] = &fi_mtp->mtp_os_desc;
+	names[0] = "MTP";
+	usb_os_desc_prepare_interf_dir(&fi_mtp->func_inst.group, 1,
+					descs, names, THIS_MODULE);
+
+	if (mtp_config) {
+		ret = mtp_setup_configfs(fi_mtp);
+		if (ret) {
+			kfree(fi_mtp);
+			pr_err("Error setting MTP\n");
+			return ERR_PTR(ret);
+		}
+	} else
+		fi_mtp->dev = _mtp_dev;
+
+	config_group_init_type_name(&fi_mtp->func_inst.group,
+					"", &mtp_func_type);
+
+	return  &fi_mtp->func_inst;
+}
+EXPORT_SYMBOL_GPL(alloc_inst_mtp_ptp);
+
+static struct usb_function_instance *mtp_alloc_inst(void)
+{
+		return alloc_inst_mtp_ptp(true);
+}
+
+static int mtp_ctrlreq_configfs(struct usb_function *f,
+				const struct usb_ctrlrequest *ctrl)
+{
+	return mtp_ctrlrequest(f->config->cdev, ctrl);
+}
+
+static void mtp_free(struct usb_function *f)
+{
+	/*NO-OP: no function specific resource allocation in mtp_alloc*/
+}
+
+struct usb_function *function_alloc_mtp_ptp(struct usb_function_instance *fi,
+					bool mtp_config)
+{
+	struct mtp_instance *fi_mtp = to_fi_mtp(fi);
+	struct mtp_dev *dev;
+
+	/*
+	 * PTP piggybacks on MTP function so make sure we have
+	 * created MTP function before we associate this PTP
+	 * function with a gadget configuration.
+	 */
+	if (fi_mtp->dev == NULL) {
+		pr_err("Error: Create MTP function before linking"
+				" PTP function with a gadget configuration\n");
+		pr_err("\t1: Delete existing PTP function if any\n");
+		pr_err("\t2: Create MTP function\n");
+		pr_err("\t3: Create and symlink PTP function"
+				" with a gadget configuration\n");
+		return ERR_PTR(-EINVAL); /* Invalid Configuration */
+	}
+
+	dev = fi_mtp->dev;
+	dev->function.name = DRIVER_NAME;
+	dev->function.strings = mtp_strings;
+	if (mtp_config) {
+		dev->function.fs_descriptors = fs_mtp_descs;
+		dev->function.hs_descriptors = hs_mtp_descs;
+		dev->function.ss_descriptors = ss_mtp_descs;
+	} else {
+		dev->function.fs_descriptors = fs_ptp_descs;
+		dev->function.hs_descriptors = hs_ptp_descs;
+		dev->function.ss_descriptors = ss_ptp_descs;
+	}
+	dev->function.bind = mtp_function_bind;
+	dev->function.unbind = mtp_function_unbind;
+	dev->function.set_alt = mtp_function_set_alt;
+	dev->function.disable = mtp_function_disable;
+	dev->function.setup = mtp_ctrlreq_configfs;
+	dev->function.free_func = mtp_free;
+
+	return &dev->function;
+}
+EXPORT_SYMBOL_GPL(function_alloc_mtp_ptp);
+
+static struct usb_function *mtp_alloc(struct usb_function_instance *fi)
+{
+	return function_alloc_mtp_ptp(fi, true);
+}
+
+DECLARE_USB_FUNCTION_INIT(mtp, mtp_alloc_inst, mtp_alloc);
+MODULE_LICENSE("GPL");
diff --git a/drivers/usb/gadget/function/f_mtp.h b/drivers/usb/gadget/function/f_mtp.h
new file mode 100644
index 000000000000..7adb1ff08eff
--- /dev/null
+++ b/drivers/usb/gadget/function/f_mtp.h
@@ -0,0 +1,18 @@
+/*
+ * Copyright (C) 2014 Google, Inc.
+ * Author: Badhri Jagan Sridharan <badhri@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+extern struct usb_function_instance *alloc_inst_mtp_ptp(bool mtp_config);
+extern struct usb_function *function_alloc_mtp_ptp(
+			struct usb_function_instance *fi, bool mtp_config);
diff --git a/drivers/usb/gadget/function/f_ptp.c b/drivers/usb/gadget/function/f_ptp.c
new file mode 100644
index 000000000000..da3e4d53e085
--- /dev/null
+++ b/drivers/usb/gadget/function/f_ptp.c
@@ -0,0 +1,38 @@
+/*
+ * Gadget Function Driver for PTP
+ *
+ * Copyright (C) 2014 Google, Inc.
+ * Author: Badhri Jagan Sridharan <badhri@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/types.h>
+
+#include <linux/configfs.h>
+#include <linux/usb/composite.h>
+
+#include "f_mtp.h"
+
+static struct usb_function_instance *ptp_alloc_inst(void)
+{
+	return alloc_inst_mtp_ptp(false);
+}
+
+static struct usb_function *ptp_alloc(struct usb_function_instance *fi)
+{
+	return function_alloc_mtp_ptp(fi, false);
+}
+
+DECLARE_USB_FUNCTION_INIT(ptp, ptp_alloc_inst, ptp_alloc);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Badhri Jagan Sridharan");
diff --git a/drivers/usb/gadget/function/f_rndis.c b/drivers/usb/gadget/function/f_rndis.c
index e587767e374c..3a28d8ac3b3d 100644
--- a/drivers/usb/gadget/function/f_rndis.c
+++ b/drivers/usb/gadget/function/f_rndis.c
@@ -70,6 +70,16 @@
  *   - MS-Windows drivers sometimes emit undocumented requests.
  */
 
+static unsigned int rndis_dl_max_pkt_per_xfer = 3;
+module_param(rndis_dl_max_pkt_per_xfer, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(rndis_dl_max_pkt_per_xfer,
+	"Maximum packets per transfer for DL aggregation");
+
+static unsigned int rndis_ul_max_pkt_per_xfer = 3;
+module_param(rndis_ul_max_pkt_per_xfer, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(rndis_ul_max_pkt_per_xfer,
+       "Maximum packets per transfer for UL aggregation");
+
 struct f_rndis {
 	struct gether			port;
 	u8				ctrl_id, data_id;
@@ -449,7 +459,9 @@ static void rndis_response_complete(struct usb_ep *ep, struct usb_request *req)
 static void rndis_command_complete(struct usb_ep *ep, struct usb_request *req)
 {
 	struct f_rndis			*rndis = req->context;
+	struct usb_composite_dev	*cdev = rndis->port.func.config->cdev;
 	int				status;
+	rndis_init_msg_type		*buf;
 
 	/* received RNDIS command from USB_CDC_SEND_ENCAPSULATED_COMMAND */
 //	spin_lock(&dev->lock);
@@ -457,6 +469,21 @@ static void rndis_command_complete(struct usb_ep *ep, struct usb_request *req)
 	if (status < 0)
 		pr_err("RNDIS command error %d, %d/%d\n",
 			status, req->actual, req->length);
+
+	buf = (rndis_init_msg_type *)req->buf;
+
+	if (buf->MessageType == RNDIS_MSG_INIT) {
+		if (buf->MaxTransferSize > 2048)
+			rndis->port.multi_pkt_xfer = 1;
+		else
+			rndis->port.multi_pkt_xfer = 0;
+		DBG(cdev, "%s: MaxTransferSize: %d : Multi_pkt_txr: %s\n",
+				__func__, buf->MaxTransferSize,
+				rndis->port.multi_pkt_xfer ? "enabled" :
+							    "disabled");
+		if (rndis_dl_max_pkt_per_xfer <= 1)
+			rndis->port.multi_pkt_xfer = 0;
+	}
 //	spin_unlock(&dev->lock);
 }
 
@@ -792,6 +819,7 @@ rndis_bind(struct usb_configuration *c, struct usb_function *f)
 
 	rndis_set_param_medium(rndis->params, RNDIS_MEDIUM_802_3, 0);
 	rndis_set_host_mac(rndis->params, rndis->ethaddr);
+	rndis_set_max_pkt_xfer(rndis->params, rndis_ul_max_pkt_per_xfer);
 
 	if (rndis->manufacturer && rndis->vendorID &&
 			rndis_set_param_vendor(rndis->params, rndis->vendorID,
@@ -978,6 +1006,8 @@ static struct usb_function *rndis_alloc(struct usb_function_instance *fi)
 	rndis->port.header_len = sizeof(struct rndis_packet_msg_type);
 	rndis->port.wrap = rndis_add_header;
 	rndis->port.unwrap = rndis_rm_hdr;
+	rndis->port.ul_max_pkts_per_xfer = rndis_ul_max_pkt_per_xfer;
+	rndis->port.dl_max_pkts_per_xfer = rndis_dl_max_pkt_per_xfer;
 
 	rndis->port.func.name = "rndis";
 	/* descriptors are per-instance copies */
diff --git a/drivers/usb/gadget/function/rndis.c b/drivers/usb/gadget/function/rndis.c
index 70d3917cc003..2ec7171b3f04 100644
--- a/drivers/usb/gadget/function/rndis.c
+++ b/drivers/usb/gadget/function/rndis.c
@@ -42,6 +42,16 @@
 
 #include "rndis.h"
 
+int rndis_ul_max_pkt_per_xfer_rcvd;
+module_param(rndis_ul_max_pkt_per_xfer_rcvd, int, S_IRUGO);
+MODULE_PARM_DESC(rndis_ul_max_pkt_per_xfer_rcvd,
+		"Max num of REMOTE_NDIS_PACKET_MSGs received in a single transfer");
+
+int rndis_ul_max_xfer_size_rcvd;
+module_param(rndis_ul_max_xfer_size_rcvd, int, S_IRUGO);
+MODULE_PARM_DESC(rndis_ul_max_xfer_size_rcvd,
+		"Max size of bus transfer received");
+
 
 /* The driver for your USB chip needs to support ep0 OUT to work with
  * RNDIS, plus all three CDC Ethernet endpoints (interrupt not optional).
@@ -579,12 +589,12 @@ static int rndis_init_response(struct rndis_params *params,
 	resp->MinorVersion = cpu_to_le32(RNDIS_MINOR_VERSION);
 	resp->DeviceFlags = cpu_to_le32(RNDIS_DF_CONNECTIONLESS);
 	resp->Medium = cpu_to_le32(RNDIS_MEDIUM_802_3);
-	resp->MaxPacketsPerTransfer = cpu_to_le32(1);
-	resp->MaxTransferSize = cpu_to_le32(
-		  params->dev->mtu
+	resp->MaxPacketsPerTransfer = cpu_to_le32(params->max_pkt_per_xfer);
+	resp->MaxTransferSize = cpu_to_le32(params->max_pkt_per_xfer *
+		(params->dev->mtu
 		+ sizeof(struct ethhdr)
 		+ sizeof(struct rndis_packet_msg_type)
-		+ 22);
+		+ 22));
 	resp->PacketAlignmentFactor = cpu_to_le32(0);
 	resp->AFListOffset = cpu_to_le32(0);
 	resp->AFListSize = cpu_to_le32(0);
@@ -681,6 +691,13 @@ static int rndis_reset_response(struct rndis_params *params,
 	rndis_reset_cmplt_type *resp;
 	rndis_resp_t *r;
 
+	u32 length;
+	u8 *xbuf;
+
+	/* drain the response queue */
+	while ((xbuf = rndis_get_next_response(params, &length)))
+		rndis_free_response(params, xbuf);
+
 	r = rndis_add_response(params, sizeof(rndis_reset_cmplt_type));
 	if (!r)
 		return -ENOMEM;
@@ -957,6 +974,8 @@ int rndis_set_param_dev(struct rndis_params *params, struct net_device *dev,
 	params->dev = dev;
 	params->filter = cdc_filter;
 
+	rndis_ul_max_xfer_size_rcvd = 0;
+	rndis_ul_max_pkt_per_xfer_rcvd = 0;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(rndis_set_param_dev);
@@ -989,6 +1008,13 @@ int rndis_set_param_medium(struct rndis_params *params, u32 medium, u32 speed)
 }
 EXPORT_SYMBOL_GPL(rndis_set_param_medium);
 
+void rndis_set_max_pkt_xfer(struct rndis_params *params, u8 max_pkt_per_xfer)
+{
+	pr_debug("%s:\n", __func__);
+
+	params->max_pkt_per_xfer = max_pkt_per_xfer;
+}
+
 void rndis_add_hdr(struct sk_buff *skb)
 {
 	struct rndis_packet_msg_type *header;
@@ -1061,23 +1087,73 @@ int rndis_rm_hdr(struct gether *port,
 			struct sk_buff *skb,
 			struct sk_buff_head *list)
 {
-	/* tmp points to a struct rndis_packet_msg_type */
-	__le32 *tmp = (void *)skb->data;
+	int num_pkts = 1;
 
-	/* MessageType, MessageLength */
-	if (cpu_to_le32(RNDIS_MSG_PACKET)
-			!= get_unaligned(tmp++)) {
-		dev_kfree_skb_any(skb);
-		return -EINVAL;
-	}
-	tmp++;
+	if (skb->len > rndis_ul_max_xfer_size_rcvd)
+		rndis_ul_max_xfer_size_rcvd = skb->len;
+
+	while (skb->len) {
+		struct rndis_packet_msg_type *hdr;
+		struct sk_buff          *skb2;
+		u32             msg_len, data_offset, data_len;
 
-	/* DataOffset, DataLength */
-	if (!skb_pull(skb, get_unaligned_le32(tmp++) + 8)) {
-		dev_kfree_skb_any(skb);
-		return -EOVERFLOW;
+		/* some rndis hosts send extra byte to avoid zlp, ignore it */
+		if (skb->len == 1) {
+			dev_kfree_skb_any(skb);
+			return 0;
+		}
+
+		if (skb->len < sizeof *hdr) {
+			pr_err("invalid rndis pkt: skblen:%u hdr_len:%zu",
+					skb->len, sizeof *hdr);
+			dev_kfree_skb_any(skb);
+			return -EINVAL;
+		}
+
+		hdr = (void *)skb->data;
+		msg_len = le32_to_cpu(hdr->MessageLength);
+		data_offset = le32_to_cpu(hdr->DataOffset);
+		data_len = le32_to_cpu(hdr->DataLength);
+
+		if (skb->len < msg_len ||
+				((data_offset + data_len + 8) > msg_len)) {
+			pr_err("invalid rndis message: %d/%d/%d/%d, len:%d\n",
+					le32_to_cpu(hdr->MessageType),
+					msg_len, data_offset, data_len, skb->len);
+			dev_kfree_skb_any(skb);
+			return -EOVERFLOW;
+		}
+		if (le32_to_cpu(hdr->MessageType) != RNDIS_MSG_PACKET) {
+			pr_err("invalid rndis message: %d/%d/%d/%d, len:%d\n",
+					le32_to_cpu(hdr->MessageType),
+					msg_len, data_offset, data_len, skb->len);
+			dev_kfree_skb_any(skb);
+			return -EINVAL;
+		}
+
+		skb_pull(skb, data_offset + 8);
+
+		if (msg_len == skb->len) {
+			skb_trim(skb, data_len);
+			break;
+		}
+
+		skb2 = skb_clone(skb, GFP_ATOMIC);
+		if (!skb2) {
+			pr_err("%s:skb clone failed\n", __func__);
+			dev_kfree_skb_any(skb);
+			return -ENOMEM;
+		}
+
+		skb_pull(skb, msg_len - sizeof *hdr);
+		skb_trim(skb2, data_len);
+		skb_queue_tail(list, skb2);
+
+		num_pkts++;
 	}
-	skb_trim(skb, get_unaligned_le32(tmp++));
+
+	if (num_pkts > rndis_ul_max_pkt_per_xfer_rcvd)
+		rndis_ul_max_pkt_per_xfer_rcvd = num_pkts;
 
 	skb_queue_tail(list, skb);
 	return 0;
diff --git a/drivers/usb/gadget/function/rndis.h b/drivers/usb/gadget/function/rndis.h
index ef92eb66d8ad..310cac3f088e 100644
--- a/drivers/usb/gadget/function/rndis.h
+++ b/drivers/usb/gadget/function/rndis.h
@@ -190,6 +190,7 @@ typedef struct rndis_params
 	struct net_device	*dev;
 
 	u32			vendorID;
+	u8			max_pkt_per_xfer;
 	const char		*vendorDescr;
 	void			(*resp_avail)(void *v);
 	void			*v;
@@ -206,6 +207,7 @@ int  rndis_set_param_vendor(struct rndis_params *params, u32 vendorID,
 			    const char *vendorDescr);
 int  rndis_set_param_medium(struct rndis_params *params, u32 medium,
 			     u32 speed);
+void rndis_set_max_pkt_xfer(struct rndis_params *params, u8 max_pkt_per_xfer);
 void rndis_add_hdr(struct sk_buff *skb);
 int rndis_rm_hdr(struct gether *port, struct sk_buff *skb,
 			struct sk_buff_head *list);
diff --git a/drivers/usb/gadget/function/u_ether.c b/drivers/usb/gadget/function/u_ether.c
index 7413f89660f7..21bf0a8423d5 100644
--- a/drivers/usb/gadget/function/u_ether.c
+++ b/drivers/usb/gadget/function/u_ether.c
@@ -53,6 +53,8 @@
  * blocks and still have efficient handling. */
 #define GETHER_MAX_ETH_FRAME_LEN 15412
 
+static struct workqueue_struct	*uether_wq;
+
 struct eth_dev {
 	/* lock is held while accessing port_usb
 	 */
@@ -65,18 +67,26 @@ struct eth_dev {
 	spinlock_t		req_lock;	/* guard {rx,tx}_reqs */
 	struct list_head	tx_reqs, rx_reqs;
 	atomic_t		tx_qlen;
+/* Minimum number of TX USB request queued to UDC */
+#define TX_REQ_THRESHOLD	5
+	int			no_tx_req_used;
+	int			tx_skb_hold_count;
+	u32			tx_req_bufsize;
 
 	struct sk_buff_head	rx_frames;
 
 	unsigned		qmult;
 
 	unsigned		header_len;
+	unsigned		ul_max_pkts_per_xfer;
+	unsigned		dl_max_pkts_per_xfer;
 	struct sk_buff		*(*wrap)(struct gether *, struct sk_buff *skb);
 	int			(*unwrap)(struct gether *,
 						struct sk_buff *skb,
 						struct sk_buff_head *list);
 
 	struct work_struct	work;
+	struct work_struct	rx_work;
 
 	unsigned long		todo;
 #define	WORK_RX_MEMORY		0
@@ -230,9 +240,13 @@ rx_submit(struct eth_dev *dev, struct usb_request *req, gfp_t gfp_flags)
 	size += out->maxpacket - 1;
 	size -= size % out->maxpacket;
 
+	if (dev->ul_max_pkts_per_xfer)
+		size *= dev->ul_max_pkts_per_xfer;
+
 	if (dev->port_usb->is_fixed)
 		size = max_t(size_t, size, dev->port_usb->fixed_out_len);
 
+	DBG(dev, "%s: size: %zd\n", __func__, size);
 	skb = alloc_skb(size + NET_IP_ALIGN, gfp_flags);
 	if (skb == NULL) {
 		DBG(dev, "no rx skb\n");
@@ -258,18 +272,16 @@ enomem:
 		DBG(dev, "rx submit --> %d\n", retval);
 		if (skb)
 			dev_kfree_skb_any(skb);
-		spin_lock_irqsave(&dev->req_lock, flags);
-		list_add(&req->list, &dev->rx_reqs);
-		spin_unlock_irqrestore(&dev->req_lock, flags);
 	}
 	return retval;
 }
 
 static void rx_complete(struct usb_ep *ep, struct usb_request *req)
 {
-	struct sk_buff	*skb = req->context, *skb2;
+	struct sk_buff	*skb = req->context;
 	struct eth_dev	*dev = ep->driver_data;
 	int		status = req->status;
+	bool		queue = 0;
 
 	switch (status) {
 
@@ -285,6 +297,10 @@ static void rx_complete(struct usb_ep *ep, struct usb_request *req)
 				status = dev->unwrap(dev->port_usb,
 							skb,
 							&dev->rx_frames);
+				if (status == -EINVAL)
+					dev->net->stats.rx_errors++;
+				else if (status == -EOVERFLOW)
+					dev->net->stats.rx_over_errors++;
 			} else {
 				dev_kfree_skb_any(skb);
 				status = -ENOTCONN;
@@ -293,30 +309,8 @@ static void rx_complete(struct usb_ep *ep, struct usb_request *req)
 		} else {
 			skb_queue_tail(&dev->rx_frames, skb);
 		}
-		skb = NULL;
-
-		skb2 = skb_dequeue(&dev->rx_frames);
-		while (skb2) {
-			if (status < 0
-					|| ETH_HLEN > skb2->len
-					|| skb2->len > GETHER_MAX_ETH_FRAME_LEN) {
-				dev->net->stats.rx_errors++;
-				dev->net->stats.rx_length_errors++;
-				DBG(dev, "rx length %d\n", skb2->len);
-				dev_kfree_skb_any(skb2);
-				goto next_frame;
-			}
-			skb2->protocol = eth_type_trans(skb2, dev->net);
-			dev->net->stats.rx_packets++;
-			dev->net->stats.rx_bytes += skb2->len;
-
-			/* no buffer copies needed, unless hardware can't
-			 * use skb buffers.
-			 */
-			status = netif_rx(skb2);
-next_frame:
-			skb2 = skb_dequeue(&dev->rx_frames);
-		}
+		if (!status)
+			queue = 1;
 		break;
 
 	/* software-driven interface shutdown */
@@ -339,22 +333,20 @@ quiesce:
 		/* FALLTHROUGH */
 
 	default:
+		queue = 1;
+		dev_kfree_skb_any(skb);
 		dev->net->stats.rx_errors++;
 		DBG(dev, "rx status %d\n", status);
 		break;
 	}
 
-	if (skb)
-		dev_kfree_skb_any(skb);
-	if (!netif_running(dev->net)) {
 clean:
-		spin_lock(&dev->req_lock);
-		list_add(&req->list, &dev->rx_reqs);
-		spin_unlock(&dev->req_lock);
-		req = NULL;
-	}
-	if (req)
-		rx_submit(dev, req, GFP_ATOMIC);
+	spin_lock(&dev->req_lock);
+	list_add(&req->list, &dev->rx_reqs);
+	spin_unlock(&dev->req_lock);
+
+	if (queue)
+		queue_work(uether_wq, &dev->rx_work);
 }
 
 static int prealloc(struct list_head *list, struct usb_ep *ep, unsigned n)
@@ -419,16 +411,24 @@ static void rx_fill(struct eth_dev *dev, gfp_t gfp_flags)
 {
 	struct usb_request	*req;
 	unsigned long		flags;
+	int			req_cnt = 0;
 
 	/* fill unused rxq slots with some skb */
 	spin_lock_irqsave(&dev->req_lock, flags);
 	while (!list_empty(&dev->rx_reqs)) {
+		/* break the nexus of continuous completion and re-submission*/
+		if (++req_cnt > qlen(dev->gadget, dev->qmult))
+			break;
+
 		req = container_of(dev->rx_reqs.next,
 				struct usb_request, list);
 		list_del_init(&req->list);
 		spin_unlock_irqrestore(&dev->req_lock, flags);
 
 		if (rx_submit(dev, req, gfp_flags) < 0) {
+			spin_lock_irqsave(&dev->req_lock, flags);
+			list_add(&req->list, &dev->rx_reqs);
+			spin_unlock_irqrestore(&dev->req_lock, flags);
 			defer_kevent(dev, WORK_RX_MEMORY);
 			return;
 		}
@@ -438,6 +438,36 @@ static void rx_fill(struct eth_dev *dev, gfp_t gfp_flags)
 	spin_unlock_irqrestore(&dev->req_lock, flags);
 }
 
+static void process_rx_w(struct work_struct *work)
+{
+	struct eth_dev	*dev = container_of(work, struct eth_dev, rx_work);
+	struct sk_buff	*skb;
+	int		status = 0;
+
+	if (!dev->port_usb)
+		return;
+
+	while ((skb = skb_dequeue(&dev->rx_frames))) {
+		if (status < 0
+				|| ETH_HLEN > skb->len
+				|| skb->len > ETH_FRAME_LEN) {
+			dev->net->stats.rx_errors++;
+			dev->net->stats.rx_length_errors++;
+			DBG(dev, "rx length %d\n", skb->len);
+			dev_kfree_skb_any(skb);
+			continue;
+		}
+		skb->protocol = eth_type_trans(skb, dev->net);
+		dev->net->stats.rx_packets++;
+		dev->net->stats.rx_bytes += skb->len;
+
+		status = netif_rx_ni(skb);
+	}
+
+	if (netif_running(dev->net))
+		rx_fill(dev, GFP_KERNEL);
+}
+
 static void eth_work(struct work_struct *work)
 {
 	struct eth_dev	*dev = container_of(work, struct eth_dev, work);
@@ -455,6 +485,11 @@ static void tx_complete(struct usb_ep *ep, struct usb_request *req)
 {
 	struct sk_buff	*skb = req->context;
 	struct eth_dev	*dev = ep->driver_data;
+	struct net_device *net = dev->net;
+	struct usb_request *new_req;
+	struct usb_ep *in;
+	int length;
+	int retval;
 
 	switch (req->status) {
 	default:
@@ -465,14 +500,73 @@ static void tx_complete(struct usb_ep *ep, struct usb_request *req)
 	case -ESHUTDOWN:		/* disconnect etc */
 		break;
 	case 0:
-		dev->net->stats.tx_bytes += skb->len;
+		if (!req->zero)
+			dev->net->stats.tx_bytes += req->length-1;
+		else
+			dev->net->stats.tx_bytes += req->length;
 	}
 	dev->net->stats.tx_packets++;
 
 	spin_lock(&dev->req_lock);
-	list_add(&req->list, &dev->tx_reqs);
-	spin_unlock(&dev->req_lock);
-	dev_kfree_skb_any(skb);
+	list_add_tail(&req->list, &dev->tx_reqs);
+
+	if (dev->port_usb->multi_pkt_xfer) {
+		dev->no_tx_req_used--;
+		req->length = 0;
+		in = dev->port_usb->in_ep;
+
+		if (!list_empty(&dev->tx_reqs)) {
+			new_req = container_of(dev->tx_reqs.next,
+					struct usb_request, list);
+			list_del(&new_req->list);
+			spin_unlock(&dev->req_lock);
+			if (new_req->length > 0) {
+				length = new_req->length;
+
+				/* NCM requires no zlp if transfer is
+				 * dwNtbInMaxSize */
+				if (dev->port_usb->is_fixed &&
+					length == dev->port_usb->fixed_in_len &&
+					(length % in->maxpacket) == 0)
+					new_req->zero = 0;
+				else
+					new_req->zero = 1;
+
+				/* use zlp framing on tx for strict CDC-Ether
+				 * conformance, though any robust network rx
+				 * path ignores extra padding. and some hardware
+				 * doesn't like to write zlps.
+				 */
+				if (new_req->zero && !dev->zlp &&
+						(length % in->maxpacket) == 0) {
+					new_req->zero = 0;
+					length++;
+				}
+
+				new_req->length = length;
+				retval = usb_ep_queue(in, new_req, GFP_ATOMIC);
+				switch (retval) {
+				default:
+					DBG(dev, "tx queue err %d\n", retval);
+					break;
+				case 0:
+					spin_lock(&dev->req_lock);
+					dev->no_tx_req_used++;
+					spin_unlock(&dev->req_lock);
+					net->trans_start = jiffies;
+				}
+			} else {
+				spin_lock(&dev->req_lock);
+				list_add(&new_req->list, &dev->tx_reqs);
+				spin_unlock(&dev->req_lock);
+			}
+		} else {
+			spin_unlock(&dev->req_lock);
+		}
+	} else {
+		spin_unlock(&dev->req_lock);
+		dev_kfree_skb_any(skb);
+	}
 
 	atomic_dec(&dev->tx_qlen);
 	if (netif_carrier_ok(dev->net))
@@ -484,6 +578,26 @@ static inline int is_promisc(u16 cdc_filter)
 	return cdc_filter & USB_CDC_PACKET_TYPE_PROMISCUOUS;
 }
 
+static void alloc_tx_buffer(struct eth_dev *dev)
+{
+	struct list_head	*act;
+	struct usb_request	*req;
+
+	dev->tx_req_bufsize = (dev->dl_max_pkts_per_xfer *
+				(dev->net->mtu
+				+ sizeof(struct ethhdr)
+				/* size of rndis_packet_msg_type */
+				+ 44
+				+ 22));
+
+	list_for_each(act, &dev->tx_reqs) {
+		req = container_of(act, struct usb_request, list);
+		if (!req->buf)
+			req->buf = kmalloc(dev->tx_req_bufsize,
+						GFP_ATOMIC);
+	}
+}
+
 static netdev_tx_t eth_start_xmit(struct sk_buff *skb,
 					struct net_device *net)
 {
@@ -510,6 +624,10 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb,
 		return NETDEV_TX_OK;
 	}
 
+	/* Allocate memory for tx_reqs to support multi packet transfer */
+	if (dev->port_usb->multi_pkt_xfer && !dev->tx_req_bufsize)
+		alloc_tx_buffer(dev);
+
 	/* apply outgoing CDC or RNDIS filters */
 	if (skb && !is_promisc(cdc_filter)) {
 		u8		*dest = skb->data;
@@ -572,9 +690,37 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb,
 		}
 	}
 
-	length = skb->len;
-	req->buf = skb->data;
-	req->context = skb;
+	spin_lock_irqsave(&dev->req_lock, flags);
+	dev->tx_skb_hold_count++;
+	spin_unlock_irqrestore(&dev->req_lock, flags);
+
+	if (dev->port_usb->multi_pkt_xfer) {
+		memcpy(req->buf + req->length, skb->data, skb->len);
+		req->length = req->length + skb->len;
+		length = req->length;
+		dev_kfree_skb_any(skb);
+
+		spin_lock_irqsave(&dev->req_lock, flags);
+		if (dev->tx_skb_hold_count < dev->dl_max_pkts_per_xfer) {
+			if (dev->no_tx_req_used > TX_REQ_THRESHOLD) {
+				list_add(&req->list, &dev->tx_reqs);
+				spin_unlock_irqrestore(&dev->req_lock, flags);
+				goto success;
+			}
+		}
+
+		dev->no_tx_req_used++;
+		spin_unlock_irqrestore(&dev->req_lock, flags);
+
+		spin_lock_irqsave(&dev->lock, flags);
+		dev->tx_skb_hold_count = 0;
+		spin_unlock_irqrestore(&dev->lock, flags);
+	} else {
+		length = skb->len;
+		req->buf = skb->data;
+		req->context = skb;
+	}
+
 	req->complete = tx_complete;
 
 	/* NCM requires no zlp if transfer is dwNtbInMaxSize */
@@ -589,11 +735,21 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb,
 	 * though any robust network rx path ignores extra padding.
 	 * and some hardware doesn't like to write zlps.
 	 */
-	if (req->zero && !dev->zlp && (length % in->maxpacket) == 0)
+	if (req->zero && !dev->zlp && (length % in->maxpacket) == 0) {
+		req->zero = 0;
 		length++;
+	}
 
 	req->length = length;
 
+	/* throttle high/super speed IRQ rate back slightly */
+	if (gadget_is_dualspeed(dev->gadget))
+		req->no_interrupt = (((dev->gadget->speed == USB_SPEED_HIGH ||
+				       dev->gadget->speed == USB_SPEED_SUPER)) &&
+					!list_empty(&dev->tx_reqs))
+			? ((atomic_read(&dev->tx_qlen) % dev->qmult) != 0)
+			: 0;
+
 	retval = usb_ep_queue(in, req, GFP_ATOMIC);
 	switch (retval) {
 	default:
@@ -605,7 +761,8 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb,
 	}
 
 	if (retval) {
-		dev_kfree_skb_any(skb);
+		if (!dev->port_usb->multi_pkt_xfer)
+			dev_kfree_skb_any(skb);
 drop:
 		dev->net->stats.tx_dropped++;
 multiframe:
@@ -615,6 +772,7 @@ multiframe:
 		list_add(&req->list, &dev->tx_reqs);
 		spin_unlock_irqrestore(&dev->req_lock, flags);
 	}
+success:
 	return NETDEV_TX_OK;
 }
 
@@ -774,6 +932,7 @@ struct eth_dev *gether_setup_name(struct usb_gadget *g,
 	spin_lock_init(&dev->lock);
 	spin_lock_init(&dev->req_lock);
 	INIT_WORK(&dev->work, eth_work);
+	INIT_WORK(&dev->rx_work, process_rx_w);
 	INIT_LIST_HEAD(&dev->tx_reqs);
 	INIT_LIST_HEAD(&dev->rx_reqs);
 
@@ -836,6 +995,7 @@ struct net_device *gether_setup_name_default(const char *netname)
 	spin_lock_init(&dev->lock);
 	spin_lock_init(&dev->req_lock);
 	INIT_WORK(&dev->work, eth_work);
+	INIT_WORK(&dev->rx_work, process_rx_w);
 	INIT_LIST_HEAD(&dev->tx_reqs);
 	INIT_LIST_HEAD(&dev->rx_reqs);
 
@@ -1070,8 +1230,13 @@ struct net_device *gether_connect(struct gether *link)
 		dev->header_len = link->header_len;
 		dev->unwrap = link->unwrap;
 		dev->wrap = link->wrap;
+		dev->ul_max_pkts_per_xfer = link->ul_max_pkts_per_xfer;
+		dev->dl_max_pkts_per_xfer = link->dl_max_pkts_per_xfer;
 
 		spin_lock(&dev->lock);
+		dev->tx_skb_hold_count = 0;
+		dev->no_tx_req_used = 0;
+		dev->tx_req_bufsize = 0;
 		dev->port_usb = link;
 		if (netif_running(dev->net)) {
 			if (link->open)
@@ -1116,6 +1281,7 @@ void gether_disconnect(struct gether *link)
 {
 	struct eth_dev		*dev = link->ioport;
 	struct usb_request	*req;
+	struct sk_buff		*skb;
 
 	WARN_ON(!dev);
 	if (!dev)
@@ -1138,6 +1304,8 @@ void gether_disconnect(struct gether *link)
 		list_del(&req->list);
 
 		spin_unlock(&dev->req_lock);
+		if (link->multi_pkt_xfer)
+			kfree(req->buf);
 		usb_ep_free_request(link->in_ep, req);
 		spin_lock(&dev->req_lock);
 	}
@@ -1156,6 +1324,12 @@ void gether_disconnect(struct gether *link)
 		spin_lock(&dev->req_lock);
 	}
 	spin_unlock(&dev->req_lock);
+
+	spin_lock(&dev->rx_frames.lock);
+	while ((skb = __skb_dequeue(&dev->rx_frames)))
+		dev_kfree_skb_any(skb);
+	spin_unlock(&dev->rx_frames.lock);
+
 	link->out_ep->desc = NULL;
 
 	/* finish forgetting about this USB link episode */
@@ -1169,5 +1343,23 @@ void gether_disconnect(struct gether *link)
 }
 EXPORT_SYMBOL_GPL(gether_disconnect);
 
-MODULE_LICENSE("GPL");
+static int __init gether_init(void)
+{
+	uether_wq  = create_singlethread_workqueue("uether");
+	if (!uether_wq) {
+		pr_err("%s: Unable to create workqueue: uether\n", __func__);
+		return -ENOMEM;
+	}
+	return 0;
+}
+module_init(gether_init);
+
+static void __exit gether_exit(void)
+{
+	destroy_workqueue(uether_wq);
+
+}
+module_exit(gether_exit);
 MODULE_AUTHOR("David Brownell");
+MODULE_DESCRIPTION("ethernet over USB driver");
+MODULE_LICENSE("GPL v2");
diff --git a/drivers/usb/gadget/function/u_ether.h b/drivers/usb/gadget/function/u_ether.h
index c77145bd6b5b..e9fd14b46102 100644
--- a/drivers/usb/gadget/function/u_ether.h
+++ b/drivers/usb/gadget/function/u_ether.h
@@ -73,6 +73,9 @@ struct gether {
 	bool				is_fixed;
 	u32				fixed_out_len;
 	u32				fixed_in_len;
+	unsigned		ul_max_pkts_per_xfer;
+	unsigned		dl_max_pkts_per_xfer;
+	bool				multi_pkt_xfer;
 	bool				supports_multi_frame;
 	struct sk_buff			*(*wrap)(struct gether *port,
 						struct sk_buff *skb);
diff --git a/drivers/usb/phy/Kconfig b/drivers/usb/phy/Kconfig
index 22e8ecb6bfbd..795485eac7b0 100644
--- a/drivers/usb/phy/Kconfig
+++ b/drivers/usb/phy/Kconfig
@@ -6,6 +6,14 @@ menu "USB Physical Layer drivers"
 config USB_PHY
 	def_bool n
 
+config USB_OTG_WAKELOCK
+	bool "Hold a wakelock when USB connected"
+	depends on WAKELOCK
+	select USB_OTG_UTILS
+	help
+	  Select this to automatically hold a wakelock when USB is
+	  connected, preventing suspend.
+
 #
 # USB Transceiver Drivers
 #
@@ -213,4 +221,13 @@ config USB_ULPI_VIEWPORT
 	  Provides read/write operations to the ULPI phy register set for
 	  controllers with a viewport register (e.g. Chipidea/ARC controllers).
 
+config DUAL_ROLE_USB_INTF
+	bool "Generic DUAL ROLE sysfs interface"
+	depends on SYSFS && USB_PHY
+	help
+	  A generic sysfs interface to track and change the state of
+	  dual role usb phys. The usb phy drivers can register to
+	  this interface to expose it capabilities to the userspace
+	  and thereby allowing userspace to change the port mode.
+
 endmenu
diff --git a/drivers/usb/phy/Makefile b/drivers/usb/phy/Makefile
index 19c0dccbb116..f7543f3b9943 100644
--- a/drivers/usb/phy/Makefile
+++ b/drivers/usb/phy/Makefile
@@ -3,6 +3,8 @@
 #
 obj-$(CONFIG_USB_PHY)			+= phy.o
 obj-$(CONFIG_OF)			+= of.o
+obj-$(CONFIG_USB_OTG_WAKELOCK)		+= otg-wakelock.o
+obj-$(CONFIG_DUAL_ROLE_USB_INTF)	+= class-dual-role.o
 
 # transceiver drivers, keep the list sorted
 
diff --git a/drivers/usb/phy/class-dual-role.c b/drivers/usb/phy/class-dual-role.c
new file mode 100644
index 000000000000..51fcb545a9d5
--- /dev/null
+++ b/drivers/usb/phy/class-dual-role.c
@@ -0,0 +1,529 @@
+/*
+ * class-dual-role.c
+ *
+ * Copyright (C) 2015 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/ctype.h>
+#include <linux/device.h>
+#include <linux/usb/class-dual-role.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+#include <linux/stat.h>
+#include <linux/types.h>
+
+#define DUAL_ROLE_NOTIFICATION_TIMEOUT 2000
+
+static ssize_t dual_role_store_property(struct device *dev,
+					struct device_attribute *attr,
+					const char *buf, size_t count);
+static ssize_t dual_role_show_property(struct device *dev,
+				       struct device_attribute *attr,
+				       char *buf);
+
+#define DUAL_ROLE_ATTR(_name)				\
+{							\
+	.attr = { .name = #_name },			\
+	.show = dual_role_show_property,		\
+	.store = dual_role_store_property,		\
+}
+
+static struct device_attribute dual_role_attrs[] = {
+	DUAL_ROLE_ATTR(supported_modes),
+	DUAL_ROLE_ATTR(mode),
+	DUAL_ROLE_ATTR(power_role),
+	DUAL_ROLE_ATTR(data_role),
+	DUAL_ROLE_ATTR(powers_vconn),
+};
+
+struct class *dual_role_class;
+EXPORT_SYMBOL_GPL(dual_role_class);
+
+static struct device_type dual_role_dev_type;
+
+static char *kstrdupcase(const char *str, gfp_t gfp, bool to_upper)
+{
+	char *ret, *ustr;
+
+	ustr = ret = kmalloc(strlen(str) + 1, gfp);
+
+	if (!ret)
+		return NULL;
+
+	while (*str)
+		*ustr++ = to_upper ? toupper(*str++) : tolower(*str++);
+
+	*ustr = 0;
+
+	return ret;
+}
+
+static void dual_role_changed_work(struct work_struct *work)
+{
+	struct dual_role_phy_instance *dual_role =
+	    container_of(work, struct dual_role_phy_instance,
+			 changed_work);
+
+	dev_dbg(&dual_role->dev, "%s\n", __func__);
+	kobject_uevent(&dual_role->dev.kobj, KOBJ_CHANGE);
+}
+
+void dual_role_instance_changed(struct dual_role_phy_instance *dual_role)
+{
+	dev_dbg(&dual_role->dev, "%s\n", __func__);
+	pm_wakeup_event(&dual_role->dev, DUAL_ROLE_NOTIFICATION_TIMEOUT);
+	schedule_work(&dual_role->changed_work);
+}
+EXPORT_SYMBOL_GPL(dual_role_instance_changed);
+
+int dual_role_get_property(struct dual_role_phy_instance *dual_role,
+			   enum dual_role_property prop,
+			   unsigned int *val)
+{
+	return dual_role->desc->get_property(dual_role, prop, val);
+}
+EXPORT_SYMBOL_GPL(dual_role_get_property);
+
+int dual_role_set_property(struct dual_role_phy_instance *dual_role,
+			   enum dual_role_property prop,
+			   const unsigned int *val)
+{
+	if (!dual_role->desc->set_property)
+		return -ENODEV;
+
+	return dual_role->desc->set_property(dual_role, prop, val);
+}
+EXPORT_SYMBOL_GPL(dual_role_set_property);
+
+int dual_role_property_is_writeable(struct dual_role_phy_instance *dual_role,
+				    enum dual_role_property prop)
+{
+	if (!dual_role->desc->property_is_writeable)
+		return -ENODEV;
+
+	return dual_role->desc->property_is_writeable(dual_role, prop);
+}
+EXPORT_SYMBOL_GPL(dual_role_property_is_writeable);
+
+static void dual_role_dev_release(struct device *dev)
+{
+	struct dual_role_phy_instance *dual_role =
+	    container_of(dev, struct dual_role_phy_instance, dev);
+	pr_debug("device: '%s': %s\n", dev_name(dev), __func__);
+	kfree(dual_role);
+}
+
+static struct dual_role_phy_instance *__must_check
+__dual_role_register(struct device *parent,
+		     const struct dual_role_phy_desc *desc)
+{
+	struct device *dev;
+	struct dual_role_phy_instance *dual_role;
+	int rc;
+
+	dual_role = kzalloc(sizeof(*dual_role), GFP_KERNEL);
+	if (!dual_role)
+		return ERR_PTR(-ENOMEM);
+
+	dev = &dual_role->dev;
+
+	device_initialize(dev);
+
+	dev->class = dual_role_class;
+	dev->type = &dual_role_dev_type;
+	dev->parent = parent;
+	dev->release = dual_role_dev_release;
+	dev_set_drvdata(dev, dual_role);
+	dual_role->desc = desc;
+
+	rc = dev_set_name(dev, "%s", desc->name);
+	if (rc)
+		goto dev_set_name_failed;
+
+	INIT_WORK(&dual_role->changed_work, dual_role_changed_work);
+
+	rc = device_init_wakeup(dev, true);
+	if (rc)
+		goto wakeup_init_failed;
+
+	rc = device_add(dev);
+	if (rc)
+		goto device_add_failed;
+
+	dual_role_instance_changed(dual_role);
+
+	return dual_role;
+
+device_add_failed:
+	device_init_wakeup(dev, false);
+wakeup_init_failed:
+dev_set_name_failed:
+	put_device(dev);
+	kfree(dual_role);
+
+	return ERR_PTR(rc);
+}
+
+static void dual_role_instance_unregister(struct dual_role_phy_instance
+					  *dual_role)
+{
+	cancel_work_sync(&dual_role->changed_work);
+	device_init_wakeup(&dual_role->dev, false);
+	device_unregister(&dual_role->dev);
+}
+
+static void devm_dual_role_release(struct device *dev, void *res)
+{
+	struct dual_role_phy_instance **dual_role = res;
+
+	dual_role_instance_unregister(*dual_role);
+}
+
+struct dual_role_phy_instance *__must_check
+devm_dual_role_instance_register(struct device *parent,
+				 const struct dual_role_phy_desc *desc)
+{
+	struct dual_role_phy_instance **ptr, *dual_role;
+
+	ptr = devres_alloc(devm_dual_role_release, sizeof(*ptr), GFP_KERNEL);
+
+	if (!ptr)
+		return ERR_PTR(-ENOMEM);
+	dual_role = __dual_role_register(parent, desc);
+	if (IS_ERR(dual_role)) {
+		devres_free(ptr);
+	} else {
+		*ptr = dual_role;
+		devres_add(parent, ptr);
+	}
+	return dual_role;
+}
+EXPORT_SYMBOL_GPL(devm_dual_role_instance_register);
+
+static int devm_dual_role_match(struct device *dev, void *res, void *data)
+{
+	struct dual_role_phy_instance **r = res;
+
+	if (WARN_ON(!r || !*r))
+		return 0;
+
+	return *r == data;
+}
+
+void devm_dual_role_instance_unregister(struct device *dev,
+					struct dual_role_phy_instance
+					*dual_role)
+{
+	int rc;
+
+	rc = devres_release(dev, devm_dual_role_release,
+			    devm_dual_role_match, dual_role);
+	WARN_ON(rc);
+}
+EXPORT_SYMBOL_GPL(devm_dual_role_instance_unregister);
+
+void *dual_role_get_drvdata(struct dual_role_phy_instance *dual_role)
+{
+	return dual_role->drv_data;
+}
+EXPORT_SYMBOL_GPL(dual_role_get_drvdata);
+
+/***************** Device attribute functions **************************/
+
+/* port type */
+static char *supported_modes_text[] = {
+	"ufp dfp", "dfp", "ufp"
+};
+
+/* current mode */
+static char *mode_text[] = {
+	"ufp", "dfp", "none"
+};
+
+/* Power role */
+static char *pr_text[] = {
+	"source", "sink", "none"
+};
+
+/* Data role */
+static char *dr_text[] = {
+	"host", "device", "none"
+};
+
+/* Vconn supply */
+static char *vconn_supply_text[] = {
+	"n", "y"
+};
+
+static ssize_t dual_role_show_property(struct device *dev,
+				       struct device_attribute *attr, char *buf)
+{
+	ssize_t ret = 0;
+	struct dual_role_phy_instance *dual_role = dev_get_drvdata(dev);
+	const ptrdiff_t off = attr - dual_role_attrs;
+	unsigned int value;
+
+	if (off == DUAL_ROLE_PROP_SUPPORTED_MODES) {
+		value = dual_role->desc->supported_modes;
+	} else {
+		ret = dual_role_get_property(dual_role, off, &value);
+
+		if (ret < 0) {
+			if (ret == -ENODATA)
+				dev_dbg(dev,
+					"driver has no data for `%s' property\n",
+					attr->attr.name);
+			else if (ret != -ENODEV)
+				dev_err(dev,
+					"driver failed to report `%s' property: %zd\n",
+					attr->attr.name, ret);
+			return ret;
+		}
+	}
+
+	if (off == DUAL_ROLE_PROP_SUPPORTED_MODES) {
+		BUILD_BUG_ON(DUAL_ROLE_PROP_SUPPORTED_MODES_TOTAL !=
+			ARRAY_SIZE(supported_modes_text));
+		if (value < DUAL_ROLE_PROP_SUPPORTED_MODES_TOTAL)
+			return snprintf(buf, PAGE_SIZE, "%s\n",
+					supported_modes_text[value]);
+		else
+			return -EIO;
+	} else if (off == DUAL_ROLE_PROP_MODE) {
+		BUILD_BUG_ON(DUAL_ROLE_PROP_MODE_TOTAL !=
+			ARRAY_SIZE(mode_text));
+		if (value < DUAL_ROLE_PROP_MODE_TOTAL)
+			return snprintf(buf, PAGE_SIZE, "%s\n",
+					mode_text[value]);
+		else
+			return -EIO;
+	} else if (off == DUAL_ROLE_PROP_PR) {
+		BUILD_BUG_ON(DUAL_ROLE_PROP_PR_TOTAL != ARRAY_SIZE(pr_text));
+		if (value < DUAL_ROLE_PROP_PR_TOTAL)
+			return snprintf(buf, PAGE_SIZE, "%s\n",
+					pr_text[value]);
+		else
+			return -EIO;
+	} else if (off == DUAL_ROLE_PROP_DR) {
+		BUILD_BUG_ON(DUAL_ROLE_PROP_DR_TOTAL != ARRAY_SIZE(dr_text));
+		if (value < DUAL_ROLE_PROP_DR_TOTAL)
+			return snprintf(buf, PAGE_SIZE, "%s\n",
+					dr_text[value]);
+		else
+			return -EIO;
+	} else if (off == DUAL_ROLE_PROP_VCONN_SUPPLY) {
+		BUILD_BUG_ON(DUAL_ROLE_PROP_VCONN_SUPPLY_TOTAL !=
+				ARRAY_SIZE(vconn_supply_text));
+		if (value < DUAL_ROLE_PROP_VCONN_SUPPLY_TOTAL)
+			return snprintf(buf, PAGE_SIZE, "%s\n",
+					vconn_supply_text[value]);
+		else
+			return -EIO;
+	} else
+		return -EIO;
+}
+
+static ssize_t dual_role_store_property(struct device *dev,
+					struct device_attribute *attr,
+					const char *buf, size_t count)
+{
+	ssize_t ret;
+	struct dual_role_phy_instance *dual_role = dev_get_drvdata(dev);
+	const ptrdiff_t off = attr - dual_role_attrs;
+	unsigned int value;
+	int total, i;
+	char *dup_buf, **text_array;
+	bool result = false;
+
+	dup_buf = kstrdupcase(buf, GFP_KERNEL, false);
+	switch (off) {
+	case DUAL_ROLE_PROP_MODE:
+		total = DUAL_ROLE_PROP_MODE_TOTAL;
+		text_array = mode_text;
+		break;
+	case DUAL_ROLE_PROP_PR:
+		total = DUAL_ROLE_PROP_PR_TOTAL;
+		text_array = pr_text;
+		break;
+	case DUAL_ROLE_PROP_DR:
+		total = DUAL_ROLE_PROP_DR_TOTAL;
+		text_array = dr_text;
+		break;
+	case DUAL_ROLE_PROP_VCONN_SUPPLY:
+		ret = strtobool(dup_buf, &result);
+		value = result;
+		if (!ret)
+			goto setprop;
+	default:
+		ret = -EINVAL;
+		goto error;
+	}
+
+	for (i = 0; i <= total; i++) {
+		if (i == total) {
+			ret = -ENOTSUPP;
+			goto error;
+		}
+		if (!strncmp(*(text_array + i), dup_buf,
+			     strlen(*(text_array + i)))) {
+			value = i;
+			break;
+		}
+	}
+
+setprop:
+	ret = dual_role->desc->set_property(dual_role, off, &value);
+
+error:
+	kfree(dup_buf);
+
+	if (ret < 0)
+		return ret;
+
+	return count;
+}
+
+static umode_t dual_role_attr_is_visible(struct kobject *kobj,
+					 struct attribute *attr, int attrno)
+{
+	struct device *dev = container_of(kobj, struct device, kobj);
+	struct dual_role_phy_instance *dual_role = dev_get_drvdata(dev);
+	umode_t mode = S_IRUSR | S_IRGRP | S_IROTH;
+	int i;
+
+	if (attrno == DUAL_ROLE_PROP_SUPPORTED_MODES)
+		return mode;
+
+	for (i = 0; i < dual_role->desc->num_properties; i++) {
+		int property = dual_role->desc->properties[i];
+
+		if (property == attrno) {
+			if (dual_role->desc->property_is_writeable &&
+			    dual_role_property_is_writeable(dual_role, property)
+			    > 0)
+				mode |= S_IWUSR;
+
+			return mode;
+		}
+	}
+
+	return 0;
+}
+
+static struct attribute *__dual_role_attrs[ARRAY_SIZE(dual_role_attrs) + 1];
+
+static struct attribute_group dual_role_attr_group = {
+	.attrs = __dual_role_attrs,
+	.is_visible = dual_role_attr_is_visible,
+};
+
+static const struct attribute_group *dual_role_attr_groups[] = {
+	&dual_role_attr_group,
+	NULL,
+};
+
+void dual_role_init_attrs(struct device_type *dev_type)
+{
+	int i;
+
+	dev_type->groups = dual_role_attr_groups;
+
+	for (i = 0; i < ARRAY_SIZE(dual_role_attrs); i++)
+		__dual_role_attrs[i] = &dual_role_attrs[i].attr;
+}
+
+int dual_role_uevent(struct device *dev, struct kobj_uevent_env *env)
+{
+	struct dual_role_phy_instance *dual_role = dev_get_drvdata(dev);
+	int ret = 0, j;
+	char *prop_buf;
+	char *attrname;
+
+	dev_dbg(dev, "uevent\n");
+
+	if (!dual_role || !dual_role->desc) {
+		dev_dbg(dev, "No dual_role phy yet\n");
+		return ret;
+	}
+
+	dev_dbg(dev, "DUAL_ROLE_NAME=%s\n", dual_role->desc->name);
+
+	ret = add_uevent_var(env, "DUAL_ROLE_NAME=%s", dual_role->desc->name);
+	if (ret)
+		return ret;
+
+	prop_buf = (char *)get_zeroed_page(GFP_KERNEL);
+	if (!prop_buf)
+		return -ENOMEM;
+
+	for (j = 0; j < dual_role->desc->num_properties; j++) {
+		struct device_attribute *attr;
+		char *line;
+
+		attr = &dual_role_attrs[dual_role->desc->properties[j]];
+
+		ret = dual_role_show_property(dev, attr, prop_buf);
+		if (ret == -ENODEV || ret == -ENODATA) {
+			ret = 0;
+			continue;
+		}
+
+		if (ret < 0)
+			goto out;
+		line = strnchr(prop_buf, PAGE_SIZE, '\n');
+		if (line)
+			*line = 0;
+
+		attrname = kstrdupcase(attr->attr.name, GFP_KERNEL, true);
+		if (!attrname)
+			ret = -ENOMEM;
+
+		dev_dbg(dev, "prop %s=%s\n", attrname, prop_buf);
+
+		ret = add_uevent_var(env, "DUAL_ROLE_%s=%s", attrname,
+				     prop_buf);
+		kfree(attrname);
+		if (ret)
+			goto out;
+	}
+
+out:
+	free_page((unsigned long)prop_buf);
+
+	return ret;
+}
+
+/******************* Module Init ***********************************/
+
+static int __init dual_role_class_init(void)
+{
+	dual_role_class = class_create(THIS_MODULE, "dual_role_usb");
+
+	if (IS_ERR(dual_role_class))
+		return PTR_ERR(dual_role_class);
+
+	dual_role_class->dev_uevent = dual_role_uevent;
+	dual_role_init_attrs(&dual_role_dev_type);
+
+	return 0;
+}
+
+static void __exit dual_role_class_exit(void)
+{
+	class_destroy(dual_role_class);
+}
+
+subsys_initcall(dual_role_class_init);
+module_exit(dual_role_class_exit);
diff --git a/drivers/usb/phy/otg-wakelock.c b/drivers/usb/phy/otg-wakelock.c
new file mode 100644
index 000000000000..479376bfa484
--- /dev/null
+++ b/drivers/usb/phy/otg-wakelock.c
@@ -0,0 +1,173 @@
+/*
+ * otg-wakelock.c
+ *
+ * Copyright (C) 2011 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/module.h>
+#include <linux/notifier.h>
+#include <linux/wakelock.h>
+#include <linux/spinlock.h>
+#include <linux/usb/otg.h>
+
+#define TEMPORARY_HOLD_TIME	2000
+
+static bool enabled = true;
+static struct usb_phy *otgwl_xceiv;
+static struct notifier_block otgwl_nb;
+
+/*
+ * otgwl_spinlock is held while the VBUS lock is grabbed or dropped and the
+ * held field is updated to match.
+ */
+
+static DEFINE_SPINLOCK(otgwl_spinlock);
+
+/*
+ * Only one lock, but since these 3 fields are associated with each other...
+ */
+
+struct otgwl_lock {
+	char name[40];
+	struct wake_lock wakelock;
+	bool held;
+};
+
+/*
+ * VBUS present lock.  Also used as a timed lock on charger
+ * connect/disconnect and USB host disconnect, to allow the system
+ * to react to the change in power.
+ */
+
+static struct otgwl_lock vbus_lock;
+
+static void otgwl_hold(struct otgwl_lock *lock)
+{
+	if (!lock->held) {
+		wake_lock(&lock->wakelock);
+		lock->held = true;
+	}
+}
+
+static void otgwl_temporary_hold(struct otgwl_lock *lock)
+{
+	wake_lock_timeout(&lock->wakelock,
+			  msecs_to_jiffies(TEMPORARY_HOLD_TIME));
+	lock->held = false;
+}
+
+static void otgwl_drop(struct otgwl_lock *lock)
+{
+	if (lock->held) {
+		wake_unlock(&lock->wakelock);
+		lock->held = false;
+	}
+}
+
+static void otgwl_handle_event(unsigned long event)
+{
+	unsigned long irqflags;
+
+	spin_lock_irqsave(&otgwl_spinlock, irqflags);
+
+	if (!enabled) {
+		otgwl_drop(&vbus_lock);
+		spin_unlock_irqrestore(&otgwl_spinlock, irqflags);
+		return;
+	}
+
+	switch (event) {
+	case USB_EVENT_VBUS:
+	case USB_EVENT_ENUMERATED:
+		otgwl_hold(&vbus_lock);
+		break;
+
+	case USB_EVENT_NONE:
+	case USB_EVENT_ID:
+	case USB_EVENT_CHARGER:
+		otgwl_temporary_hold(&vbus_lock);
+		break;
+
+	default:
+		break;
+	}
+
+	spin_unlock_irqrestore(&otgwl_spinlock, irqflags);
+}
+
+static int otgwl_otg_notifications(struct notifier_block *nb,
+				   unsigned long event, void *unused)
+{
+	otgwl_handle_event(event);
+	return NOTIFY_OK;
+}
+
+static int set_enabled(const char *val, const struct kernel_param *kp)
+{
+	int rv = param_set_bool(val, kp);
+
+	if (rv)
+		return rv;
+
+	if (otgwl_xceiv)
+		otgwl_handle_event(otgwl_xceiv->last_event);
+
+	return 0;
+}
+
+static struct kernel_param_ops enabled_param_ops = {
+	.set = set_enabled,
+	.get = param_get_bool,
+};
+
+module_param_cb(enabled, &enabled_param_ops, &enabled, 0644);
+MODULE_PARM_DESC(enabled, "enable wakelock when VBUS present");
+
+static int __init otg_wakelock_init(void)
+{
+	int ret;
+	struct usb_phy *phy;
+
+	phy = usb_get_phy(USB_PHY_TYPE_USB2);
+
+	if (IS_ERR(phy)) {
+		pr_err("%s: No USB transceiver found\n", __func__);
+		return PTR_ERR(phy);
+	}
+	otgwl_xceiv = phy;
+
+	snprintf(vbus_lock.name, sizeof(vbus_lock.name), "vbus-%s",
+		 dev_name(otgwl_xceiv->dev));
+	wake_lock_init(&vbus_lock.wakelock, WAKE_LOCK_SUSPEND,
+		       vbus_lock.name);
+
+	otgwl_nb.notifier_call = otgwl_otg_notifications;
+	ret = usb_register_notifier(otgwl_xceiv, &otgwl_nb);
+
+	if (ret) {
+		pr_err("%s: usb_register_notifier on transceiver %s"
+		       " failed\n", __func__,
+		       dev_name(otgwl_xceiv->dev));
+		otgwl_xceiv = NULL;
+		wake_lock_destroy(&vbus_lock.wakelock);
+		return ret;
+	}
+
+	otgwl_handle_event(otgwl_xceiv->last_event);
+	return ret;
+}
+
+late_initcall(otg_wakelock_init);
diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
index e0606c01e8ac..35d239ebf912 100644
--- a/drivers/video/Kconfig
+++ b/drivers/video/Kconfig
@@ -29,6 +29,7 @@ source "drivers/video/fbdev/Kconfig"
 endmenu
 
 source "drivers/video/backlight/Kconfig"
+source "drivers/video/adf/Kconfig"
 
 config VGASTATE
        tristate
diff --git a/drivers/video/Makefile b/drivers/video/Makefile
index 9ad3c17d6456..1a8c4ced39b2 100644
--- a/drivers/video/Makefile
+++ b/drivers/video/Makefile
@@ -1,6 +1,7 @@
 obj-$(CONFIG_VGASTATE)            += vgastate.o
 obj-$(CONFIG_HDMI)                += hdmi.o
 
+obj-$(CONFIG_ADF)		  += adf/
 obj-$(CONFIG_VT)		  += console/
 obj-$(CONFIG_LOGO)		  += logo/
 obj-y				  += backlight/
diff --git a/drivers/video/adf/Kconfig b/drivers/video/adf/Kconfig
new file mode 100644
index 000000000000..2777db48fae0
--- /dev/null
+++ b/drivers/video/adf/Kconfig
@@ -0,0 +1,14 @@
+menuconfig ADF
+	depends on SYNC
+	depends on DMA_SHARED_BUFFER
+	tristate "Atomic Display Framework"
+
+menuconfig ADF_FBDEV
+	depends on ADF
+	depends on FB
+	tristate "Helper for implementing the fbdev API in ADF drivers"
+
+menuconfig ADF_MEMBLOCK
+	depends on ADF
+	depends on HAVE_MEMBLOCK
+	bool "Helper for using memblocks as buffers in ADF drivers"
diff --git a/drivers/video/adf/Makefile b/drivers/video/adf/Makefile
new file mode 100644
index 000000000000..cdf34a666dc7
--- /dev/null
+++ b/drivers/video/adf/Makefile
@@ -0,0 +1,17 @@
+ccflags-y := -Idrivers/staging/android
+
+CFLAGS_adf.o := -I$(src)
+
+obj-$(CONFIG_ADF) += adf_core.o
+
+adf_core-y := adf.o \
+	adf_client.o \
+	adf_fops.o \
+	adf_format.o \
+	adf_sysfs.o
+
+adf_core-$(CONFIG_COMPAT) += adf_fops32.o
+
+obj-$(CONFIG_ADF_FBDEV) += adf_fbdev.o
+
+obj-$(CONFIG_ADF_MEMBLOCK) += adf_memblock.o
diff --git a/drivers/video/adf/adf.c b/drivers/video/adf/adf.c
new file mode 100644
index 000000000000..42c30c05826a
--- /dev/null
+++ b/drivers/video/adf/adf.c
@@ -0,0 +1,1188 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ * adf_modeinfo_{set_name,set_vrefresh} modified from
+ * drivers/gpu/drm/drm_modes.c
+ * adf_format_validate_yuv modified from framebuffer_check in
+ * drivers/gpu/drm/drm_crtc.c
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/device.h>
+#include <linux/idr.h>
+#include <linux/highmem.h>
+#include <linux/memblock.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+#include <video/adf_format.h>
+
+#include "sw_sync.h"
+#include "sync.h"
+
+#include "adf.h"
+#include "adf_fops.h"
+#include "adf_sysfs.h"
+
+#define CREATE_TRACE_POINTS
+#include "adf_trace.h"
+
+#define ADF_SHORT_FENCE_TIMEOUT (1 * MSEC_PER_SEC)
+#define ADF_LONG_FENCE_TIMEOUT (10 * MSEC_PER_SEC)
+
+static DEFINE_IDR(adf_devices);
+
+static void adf_fence_wait(struct adf_device *dev, struct sync_fence *fence)
+{
+	/* sync_fence_wait() dumps debug information on timeout.  Experience
+	   has shown that if the pipeline gets stuck, a short timeout followed
+	   by a longer one provides useful information for debugging. */
+	int err = sync_fence_wait(fence, ADF_SHORT_FENCE_TIMEOUT);
+	if (err >= 0)
+		return;
+
+	if (err == -ETIME)
+		err = sync_fence_wait(fence, ADF_LONG_FENCE_TIMEOUT);
+
+	if (err < 0)
+		dev_warn(&dev->base.dev, "error waiting on fence: %d\n", err);
+}
+
+void adf_buffer_cleanup(struct adf_buffer *buf)
+{
+	size_t i;
+	for (i = 0; i < ARRAY_SIZE(buf->dma_bufs); i++)
+		if (buf->dma_bufs[i])
+			dma_buf_put(buf->dma_bufs[i]);
+
+	if (buf->acquire_fence)
+		sync_fence_put(buf->acquire_fence);
+}
+
+void adf_buffer_mapping_cleanup(struct adf_buffer_mapping *mapping,
+		struct adf_buffer *buf)
+{
+	/* calling adf_buffer_mapping_cleanup() is safe even if mapping is
+	   uninitialized or partially-initialized, as long as it was
+	   zeroed on allocation */
+	size_t i;
+	for (i = 0; i < ARRAY_SIZE(mapping->sg_tables); i++) {
+		if (mapping->sg_tables[i])
+			dma_buf_unmap_attachment(mapping->attachments[i],
+					mapping->sg_tables[i], DMA_TO_DEVICE);
+		if (mapping->attachments[i])
+			dma_buf_detach(buf->dma_bufs[i],
+					mapping->attachments[i]);
+	}
+}
+
+void adf_post_cleanup(struct adf_device *dev, struct adf_pending_post *post)
+{
+	size_t i;
+
+	if (post->state)
+		dev->ops->state_free(dev, post->state);
+
+	for (i = 0; i < post->config.n_bufs; i++) {
+		adf_buffer_mapping_cleanup(&post->config.mappings[i],
+				&post->config.bufs[i]);
+		adf_buffer_cleanup(&post->config.bufs[i]);
+	}
+
+	kfree(post->config.custom_data);
+	kfree(post->config.mappings);
+	kfree(post->config.bufs);
+	kfree(post);
+}
+
+static void adf_sw_advance_timeline(struct adf_device *dev)
+{
+#ifdef CONFIG_SW_SYNC
+	sw_sync_timeline_inc(dev->timeline, 1);
+#else
+	BUG();
+#endif
+}
+
+static void adf_post_work_func(struct kthread_work *work)
+{
+	struct adf_device *dev =
+			container_of(work, struct adf_device, post_work);
+	struct adf_pending_post *post, *next;
+	struct list_head saved_list;
+
+	mutex_lock(&dev->post_lock);
+	memcpy(&saved_list, &dev->post_list, sizeof(saved_list));
+	list_replace_init(&dev->post_list, &saved_list);
+	mutex_unlock(&dev->post_lock);
+
+	list_for_each_entry_safe(post, next, &saved_list, head) {
+		int i;
+
+		for (i = 0; i < post->config.n_bufs; i++) {
+			struct sync_fence *fence =
+					post->config.bufs[i].acquire_fence;
+			if (fence)
+				adf_fence_wait(dev, fence);
+		}
+
+		dev->ops->post(dev, &post->config, post->state);
+
+		if (dev->ops->advance_timeline)
+			dev->ops->advance_timeline(dev, &post->config,
+					post->state);
+		else
+			adf_sw_advance_timeline(dev);
+
+		list_del(&post->head);
+		if (dev->onscreen)
+			adf_post_cleanup(dev, dev->onscreen);
+		dev->onscreen = post;
+	}
+}
+
+void adf_attachment_free(struct adf_attachment_list *attachment)
+{
+	list_del(&attachment->head);
+	kfree(attachment);
+}
+
+struct adf_event_refcount *adf_obj_find_event_refcount(struct adf_obj *obj,
+		enum adf_event_type type)
+{
+	struct rb_root *root = &obj->event_refcount;
+	struct rb_node **new = &(root->rb_node);
+	struct rb_node *parent = NULL;
+	struct adf_event_refcount *refcount;
+
+	while (*new) {
+		refcount = container_of(*new, struct adf_event_refcount, node);
+		parent = *new;
+
+		if (refcount->type > type)
+			new = &(*new)->rb_left;
+		else if (refcount->type < type)
+			new = &(*new)->rb_right;
+		else
+			return refcount;
+	}
+
+	refcount = kzalloc(sizeof(*refcount), GFP_KERNEL);
+	if (!refcount)
+		return NULL;
+	refcount->type = type;
+
+	rb_link_node(&refcount->node, parent, new);
+	rb_insert_color(&refcount->node, root);
+	return refcount;
+}
+
+/**
+ * adf_event_get - increase the refcount for an event
+ *
+ * @obj: the object that produces the event
+ * @type: the event type
+ *
+ * ADF will call the object's set_event() op if needed.  ops are allowed
+ * to sleep, so adf_event_get() must NOT be called from an atomic context.
+ *
+ * Returns 0 if successful, or -%EINVAL if the object does not support the
+ * requested event type.
+ */
+int adf_event_get(struct adf_obj *obj, enum adf_event_type type)
+{
+	struct adf_event_refcount *refcount;
+	int old_refcount;
+	int ret;
+
+	ret = adf_obj_check_supports_event(obj, type);
+	if (ret < 0)
+		return ret;
+
+	mutex_lock(&obj->event_lock);
+
+	refcount = adf_obj_find_event_refcount(obj, type);
+	if (!refcount) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	old_refcount = refcount->refcount++;
+
+	if (old_refcount == 0) {
+		obj->ops->set_event(obj, type, true);
+		trace_adf_event_enable(obj, type);
+	}
+
+done:
+	mutex_unlock(&obj->event_lock);
+	return ret;
+}
+EXPORT_SYMBOL(adf_event_get);
+
+/**
+ * adf_event_put - decrease the refcount for an event
+ *
+ * @obj: the object that produces the event
+ * @type: the event type
+ *
+ * ADF will call the object's set_event() op if needed.  ops are allowed
+ * to sleep, so adf_event_put() must NOT be called from an atomic context.
+ *
+ * Returns 0 if successful, -%EINVAL if the object does not support the
+ * requested event type, or -%EALREADY if the refcount is already 0.
+ */
+int adf_event_put(struct adf_obj *obj, enum adf_event_type type)
+{
+	struct adf_event_refcount *refcount;
+	int old_refcount;
+	int ret;
+
+	ret = adf_obj_check_supports_event(obj, type);
+	if (ret < 0)
+		return ret;
+
+
+	mutex_lock(&obj->event_lock);
+
+	refcount = adf_obj_find_event_refcount(obj, type);
+	if (!refcount) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	old_refcount = refcount->refcount--;
+
+	if (WARN_ON(old_refcount == 0)) {
+		refcount->refcount++;
+		ret = -EALREADY;
+	} else if (old_refcount == 1) {
+		obj->ops->set_event(obj, type, false);
+		trace_adf_event_disable(obj, type);
+	}
+
+done:
+	mutex_unlock(&obj->event_lock);
+	return ret;
+}
+EXPORT_SYMBOL(adf_event_put);
+
+/**
+ * adf_vsync_wait - wait for a vsync event on a display interface
+ *
+ * @intf: the display interface
+ * @timeout: timeout in jiffies (0 = wait indefinitely)
+ *
+ * adf_vsync_wait() may sleep, so it must NOT be called from an atomic context.
+ *
+ * This function returns -%ERESTARTSYS if it is interrupted by a signal.
+ * If @timeout == 0 then this function returns 0 on vsync. If @timeout > 0 then
+ * this function returns the number of remaining jiffies or -%ETIMEDOUT on
+ * timeout.
+ */
+int adf_vsync_wait(struct adf_interface *intf, long timeout)
+{
+	ktime_t timestamp;
+	int ret;
+	unsigned long flags;
+
+	read_lock_irqsave(&intf->vsync_lock, flags);
+	timestamp = intf->vsync_timestamp;
+	read_unlock_irqrestore(&intf->vsync_lock, flags);
+
+	adf_vsync_get(intf);
+	if (timeout) {
+		ret = wait_event_interruptible_timeout(intf->vsync_wait,
+				!ktime_equal(timestamp,
+						intf->vsync_timestamp),
+				msecs_to_jiffies(timeout));
+		if (ret == 0 && ktime_equal(timestamp, intf->vsync_timestamp))
+			ret = -ETIMEDOUT;
+	} else {
+		ret = wait_event_interruptible(intf->vsync_wait,
+				!ktime_equal(timestamp,
+						intf->vsync_timestamp));
+	}
+	adf_vsync_put(intf);
+
+	return ret;
+}
+EXPORT_SYMBOL(adf_vsync_wait);
+
+static void adf_event_queue(struct adf_obj *obj, struct adf_event *event)
+{
+	struct adf_file *file;
+	unsigned long flags;
+
+	trace_adf_event(obj, event->type);
+
+	spin_lock_irqsave(&obj->file_lock, flags);
+
+	list_for_each_entry(file, &obj->file_list, head)
+		if (test_bit(event->type, file->event_subscriptions))
+			adf_file_queue_event(file, event);
+
+	spin_unlock_irqrestore(&obj->file_lock, flags);
+}
+
+/**
+ * adf_event_notify - notify userspace of a driver-private event
+ *
+ * @obj: the ADF object that produced the event
+ * @event: the event
+ *
+ * adf_event_notify() may be called safely from an atomic context.  It will
+ * copy @event if needed, so @event may point to a variable on the stack.
+ *
+ * Drivers must NOT call adf_event_notify() for vsync and hotplug events.
+ * ADF provides adf_vsync_notify() and
+ * adf_hotplug_notify_{connected,disconnected}() for these events.
+ */
+int adf_event_notify(struct adf_obj *obj, struct adf_event *event)
+{
+	if (WARN_ON(event->type == ADF_EVENT_VSYNC ||
+			event->type == ADF_EVENT_HOTPLUG))
+		return -EINVAL;
+
+	adf_event_queue(obj, event);
+	return 0;
+}
+EXPORT_SYMBOL(adf_event_notify);
+
+/**
+ * adf_vsync_notify - notify ADF of a display interface's vsync event
+ *
+ * @intf: the display interface
+ * @timestamp: the time the vsync occurred
+ *
+ * adf_vsync_notify() may be called safely from an atomic context.
+ */
+void adf_vsync_notify(struct adf_interface *intf, ktime_t timestamp)
+{
+	unsigned long flags;
+	struct adf_vsync_event event;
+
+	write_lock_irqsave(&intf->vsync_lock, flags);
+	intf->vsync_timestamp = timestamp;
+	write_unlock_irqrestore(&intf->vsync_lock, flags);
+
+	wake_up_interruptible_all(&intf->vsync_wait);
+
+	event.base.type = ADF_EVENT_VSYNC;
+	event.base.length = sizeof(event);
+	event.timestamp = ktime_to_ns(timestamp);
+	adf_event_queue(&intf->base, &event.base);
+}
+EXPORT_SYMBOL(adf_vsync_notify);
+
+void adf_hotplug_notify(struct adf_interface *intf, bool connected,
+		struct drm_mode_modeinfo *modelist, size_t n_modes)
+{
+	unsigned long flags;
+	struct adf_hotplug_event event;
+	struct drm_mode_modeinfo *old_modelist;
+
+	write_lock_irqsave(&intf->hotplug_modelist_lock, flags);
+	old_modelist = intf->modelist;
+	intf->hotplug_detect = connected;
+	intf->modelist = modelist;
+	intf->n_modes = n_modes;
+	write_unlock_irqrestore(&intf->hotplug_modelist_lock, flags);
+
+	kfree(old_modelist);
+
+	event.base.length = sizeof(event);
+	event.base.type = ADF_EVENT_HOTPLUG;
+	event.connected = connected;
+	adf_event_queue(&intf->base, &event.base);
+}
+
+/**
+ * adf_hotplug_notify_connected - notify ADF of a display interface being
+ * connected to a display
+ *
+ * @intf: the display interface
+ * @modelist: hardware modes supported by display
+ * @n_modes: length of modelist
+ *
+ * @modelist is copied as needed, so it may point to a variable on the stack.
+ *
+ * adf_hotplug_notify_connected() may NOT be called safely from an atomic
+ * context.
+ *
+ * Returns 0 on success or error code (<0) on error.
+ */
+int adf_hotplug_notify_connected(struct adf_interface *intf,
+		struct drm_mode_modeinfo *modelist, size_t n_modes)
+{
+	struct drm_mode_modeinfo *modelist_copy;
+
+	if (n_modes > ADF_MAX_MODES)
+		return -ENOMEM;
+
+	modelist_copy = kzalloc(sizeof(modelist_copy[0]) * n_modes,
+			GFP_KERNEL);
+	if (!modelist_copy)
+		return -ENOMEM;
+	memcpy(modelist_copy, modelist, sizeof(modelist_copy[0]) * n_modes);
+
+	adf_hotplug_notify(intf, true, modelist_copy, n_modes);
+	return 0;
+}
+EXPORT_SYMBOL(adf_hotplug_notify_connected);
+
+/**
+ * adf_hotplug_notify_disconnected - notify ADF of a display interface being
+ * disconnected from a display
+ *
+ * @intf: the display interface
+ *
+ * adf_hotplug_notify_disconnected() may be called safely from an atomic
+ * context.
+ */
+void adf_hotplug_notify_disconnected(struct adf_interface *intf)
+{
+	adf_hotplug_notify(intf, false, NULL, 0);
+}
+EXPORT_SYMBOL(adf_hotplug_notify_disconnected);
+
+static int adf_obj_init(struct adf_obj *obj, enum adf_obj_type type,
+		struct idr *idr, struct adf_device *parent,
+		const struct adf_obj_ops *ops, const char *fmt, va_list args)
+{
+	int ret;
+
+	if (ops && ops->supports_event && !ops->set_event) {
+		pr_err("%s: %s implements supports_event but not set_event\n",
+				__func__, adf_obj_type_str(type));
+		return -EINVAL;
+	}
+
+	ret = idr_alloc(idr, obj, 0, 0, GFP_KERNEL);
+	if (ret < 0) {
+		pr_err("%s: allocating object id failed: %d\n", __func__, ret);
+		return ret;
+	}
+	obj->id = ret;
+
+	vscnprintf(obj->name, sizeof(obj->name), fmt, args);
+
+	obj->type = type;
+	obj->ops = ops;
+	obj->parent = parent;
+	mutex_init(&obj->event_lock);
+	obj->event_refcount = RB_ROOT;
+	spin_lock_init(&obj->file_lock);
+	INIT_LIST_HEAD(&obj->file_list);
+	return 0;
+}
+
+static void adf_obj_destroy(struct adf_obj *obj, struct idr *idr)
+{
+	struct rb_node *node = rb_first(&obj->event_refcount);
+
+	while (node) {
+		struct adf_event_refcount *refcount =
+				container_of(node, struct adf_event_refcount,
+						node);
+		rb_erase(&refcount->node, &obj->event_refcount);
+		kfree(refcount);
+		node = rb_first(&obj->event_refcount);
+	}
+
+	mutex_destroy(&obj->event_lock);
+	idr_remove(idr, obj->id);
+}
+
+/**
+ * adf_device_init - initialize ADF-internal data for a display device
+ * and create sysfs entries
+ *
+ * @dev: the display device
+ * @parent: the device's parent device
+ * @ops: the device's associated ops
+ * @fmt: formatting string for the display device's name
+ *
+ * @fmt specifies the device's sysfs filename and the name returned to
+ * userspace through the %ADF_GET_DEVICE_DATA ioctl.
+ *
+ * Returns 0 on success or error code (<0) on failure.
+ */
+int adf_device_init(struct adf_device *dev, struct device *parent,
+		const struct adf_device_ops *ops, const char *fmt, ...)
+{
+	int ret;
+	va_list args;
+
+	if (!ops->validate || !ops->post) {
+		pr_err("%s: device must implement validate and post\n",
+				__func__);
+		return -EINVAL;
+	}
+
+	if (!ops->complete_fence && !ops->advance_timeline) {
+		if (!IS_ENABLED(CONFIG_SW_SYNC)) {
+			pr_err("%s: device requires sw_sync but it is not enabled in the kernel\n",
+					__func__);
+			return -EINVAL;
+		}
+	} else if (!(ops->complete_fence && ops->advance_timeline)) {
+		pr_err("%s: device must implement both complete_fence and advance_timeline, or implement neither\n",
+				__func__);
+		return -EINVAL;
+	}
+
+	memset(dev, 0, sizeof(*dev));
+
+	va_start(args, fmt);
+	ret = adf_obj_init(&dev->base, ADF_OBJ_DEVICE, &adf_devices, dev,
+			&ops->base, fmt, args);
+	va_end(args);
+	if (ret < 0)
+		return ret;
+
+	dev->dev = parent;
+	dev->ops = ops;
+	idr_init(&dev->overlay_engines);
+	idr_init(&dev->interfaces);
+	mutex_init(&dev->client_lock);
+	INIT_LIST_HEAD(&dev->post_list);
+	mutex_init(&dev->post_lock);
+	init_kthread_worker(&dev->post_worker);
+	INIT_LIST_HEAD(&dev->attached);
+	INIT_LIST_HEAD(&dev->attach_allowed);
+
+	dev->post_thread = kthread_run(kthread_worker_fn,
+			&dev->post_worker, dev->base.name);
+	if (IS_ERR(dev->post_thread)) {
+		ret = PTR_ERR(dev->post_thread);
+		dev->post_thread = NULL;
+
+		pr_err("%s: failed to run config posting thread: %d\n",
+				__func__, ret);
+		goto err;
+	}
+	init_kthread_work(&dev->post_work, adf_post_work_func);
+
+	ret = adf_device_sysfs_init(dev);
+	if (ret < 0)
+		goto err;
+
+	return 0;
+
+err:
+	adf_device_destroy(dev);
+	return ret;
+}
+EXPORT_SYMBOL(adf_device_init);
+
+/**
+ * adf_device_destroy - clean up ADF-internal data for a display device
+ *
+ * @dev: the display device
+ */
+void adf_device_destroy(struct adf_device *dev)
+{
+	struct adf_attachment_list *entry, *next;
+
+	idr_destroy(&dev->interfaces);
+	idr_destroy(&dev->overlay_engines);
+
+	if (dev->post_thread) {
+		flush_kthread_worker(&dev->post_worker);
+		kthread_stop(dev->post_thread);
+	}
+
+	if (dev->onscreen)
+		adf_post_cleanup(dev, dev->onscreen);
+	adf_device_sysfs_destroy(dev);
+	list_for_each_entry_safe(entry, next, &dev->attach_allowed, head) {
+		adf_attachment_free(entry);
+	}
+	list_for_each_entry_safe(entry, next, &dev->attached, head) {
+		adf_attachment_free(entry);
+	}
+	mutex_destroy(&dev->post_lock);
+	mutex_destroy(&dev->client_lock);
+
+	if (dev->timeline)
+		sync_timeline_destroy(&dev->timeline->obj);
+
+	adf_obj_destroy(&dev->base, &adf_devices);
+}
+EXPORT_SYMBOL(adf_device_destroy);
+
+/**
+ * adf_interface_init - initialize ADF-internal data for a display interface
+ * and create sysfs entries
+ *
+ * @intf: the display interface
+ * @dev: the interface's "parent" display device
+ * @type: interface type (see enum @adf_interface_type)
+ * @idx: which interface of type @type;
+ *	e.g. interface DSI.1 -> @type=%ADF_INTF_TYPE_DSI, @idx=1
+ * @flags: informational flags (bitmask of %ADF_INTF_FLAG_* values)
+ * @ops: the interface's associated ops
+ * @fmt: formatting string for the display interface's name
+ *
+ * @dev must have previously been initialized with adf_device_init().
+ *
+ * @fmt affects the name returned to userspace through the
+ * %ADF_GET_INTERFACE_DATA ioctl.  It does not affect the sysfs filename,
+ * which is derived from @dev's name.
+ *
+ * Returns 0 on success or error code (<0) on failure.
+ */
+int adf_interface_init(struct adf_interface *intf, struct adf_device *dev,
+		enum adf_interface_type type, u32 idx, u32 flags,
+		const struct adf_interface_ops *ops, const char *fmt, ...)
+{
+	int ret;
+	va_list args;
+	const u32 allowed_flags = ADF_INTF_FLAG_PRIMARY |
+			ADF_INTF_FLAG_EXTERNAL;
+
+	if (dev->n_interfaces == ADF_MAX_INTERFACES) {
+		pr_err("%s: parent device %s has too many interfaces\n",
+				__func__, dev->base.name);
+		return -ENOMEM;
+	}
+
+	if (type >= ADF_INTF_MEMORY && type <= ADF_INTF_TYPE_DEVICE_CUSTOM) {
+		pr_err("%s: invalid interface type %u\n", __func__, type);
+		return -EINVAL;
+	}
+
+	if (flags & ~allowed_flags) {
+		pr_err("%s: invalid interface flags 0x%X\n", __func__,
+				flags & ~allowed_flags);
+		return -EINVAL;
+	}
+
+	memset(intf, 0, sizeof(*intf));
+
+	va_start(args, fmt);
+	ret = adf_obj_init(&intf->base, ADF_OBJ_INTERFACE, &dev->interfaces,
+			dev, ops ? &ops->base : NULL, fmt, args);
+	va_end(args);
+	if (ret < 0)
+		return ret;
+
+	intf->type = type;
+	intf->idx = idx;
+	intf->flags = flags;
+	intf->ops = ops;
+	intf->dpms_state = DRM_MODE_DPMS_OFF;
+	init_waitqueue_head(&intf->vsync_wait);
+	rwlock_init(&intf->vsync_lock);
+	rwlock_init(&intf->hotplug_modelist_lock);
+
+	ret = adf_interface_sysfs_init(intf);
+	if (ret < 0)
+		goto err;
+	dev->n_interfaces++;
+
+	return 0;
+
+err:
+	adf_obj_destroy(&intf->base, &dev->interfaces);
+	return ret;
+}
+EXPORT_SYMBOL(adf_interface_init);
+
+/**
+ * adf_interface_destroy - clean up ADF-internal data for a display interface
+ *
+ * @intf: the display interface
+ */
+void adf_interface_destroy(struct adf_interface *intf)
+{
+	struct adf_device *dev = adf_interface_parent(intf);
+	struct adf_attachment_list *entry, *next;
+
+	mutex_lock(&dev->client_lock);
+	list_for_each_entry_safe(entry, next, &dev->attach_allowed, head) {
+		if (entry->attachment.interface == intf) {
+			adf_attachment_free(entry);
+			dev->n_attach_allowed--;
+		}
+	}
+	list_for_each_entry_safe(entry, next, &dev->attached, head) {
+		if (entry->attachment.interface == intf) {
+			adf_device_detach_op(dev,
+					entry->attachment.overlay_engine, intf);
+			adf_attachment_free(entry);
+			dev->n_attached--;
+		}
+	}
+	kfree(intf->modelist);
+	adf_interface_sysfs_destroy(intf);
+	adf_obj_destroy(&intf->base, &dev->interfaces);
+	dev->n_interfaces--;
+	mutex_unlock(&dev->client_lock);
+}
+EXPORT_SYMBOL(adf_interface_destroy);
+
+static bool adf_overlay_engine_has_custom_formats(
+		const struct adf_overlay_engine_ops *ops)
+{
+	size_t i;
+	for (i = 0; i < ops->n_supported_formats; i++)
+		if (!adf_format_is_standard(ops->supported_formats[i]))
+			return true;
+	return false;
+}
+
+/**
+ * adf_overlay_engine_init - initialize ADF-internal data for an
+ * overlay engine and create sysfs entries
+ *
+ * @eng: the overlay engine
+ * @dev: the overlay engine's "parent" display device
+ * @ops: the overlay engine's associated ops
+ * @fmt: formatting string for the overlay engine's name
+ *
+ * @dev must have previously been initialized with adf_device_init().
+ *
+ * @fmt affects the name returned to userspace through the
+ * %ADF_GET_OVERLAY_ENGINE_DATA ioctl.  It does not affect the sysfs filename,
+ * which is derived from @dev's name.
+ *
+ * Returns 0 on success or error code (<0) on failure.
+ */
+int adf_overlay_engine_init(struct adf_overlay_engine *eng,
+		struct adf_device *dev,
+		const struct adf_overlay_engine_ops *ops, const char *fmt, ...)
+{
+	int ret;
+	va_list args;
+
+	if (!ops->supported_formats) {
+		pr_err("%s: overlay engine must support at least one format\n",
+				__func__);
+		return -EINVAL;
+	}
+
+	if (ops->n_supported_formats > ADF_MAX_SUPPORTED_FORMATS) {
+		pr_err("%s: overlay engine supports too many formats\n",
+				__func__);
+		return -EINVAL;
+	}
+
+	if (adf_overlay_engine_has_custom_formats(ops) &&
+			!dev->ops->validate_custom_format) {
+		pr_err("%s: overlay engine has custom formats but parent device %s does not implement validate_custom_format\n",
+				__func__, dev->base.name);
+		return -EINVAL;
+	}
+
+	memset(eng, 0, sizeof(*eng));
+
+	va_start(args, fmt);
+	ret = adf_obj_init(&eng->base, ADF_OBJ_OVERLAY_ENGINE,
+			&dev->overlay_engines, dev, &ops->base, fmt, args);
+	va_end(args);
+	if (ret < 0)
+		return ret;
+
+	eng->ops = ops;
+
+	ret = adf_overlay_engine_sysfs_init(eng);
+	if (ret < 0)
+		goto err;
+
+	return 0;
+
+err:
+	adf_obj_destroy(&eng->base, &dev->overlay_engines);
+	return ret;
+}
+EXPORT_SYMBOL(adf_overlay_engine_init);
+
+/**
+ * adf_interface_destroy - clean up ADF-internal data for an overlay engine
+ *
+ * @eng: the overlay engine
+ */
+void adf_overlay_engine_destroy(struct adf_overlay_engine *eng)
+{
+	struct adf_device *dev = adf_overlay_engine_parent(eng);
+	struct adf_attachment_list *entry, *next;
+
+	mutex_lock(&dev->client_lock);
+	list_for_each_entry_safe(entry, next, &dev->attach_allowed, head) {
+		if (entry->attachment.overlay_engine == eng) {
+			adf_attachment_free(entry);
+			dev->n_attach_allowed--;
+		}
+	}
+	list_for_each_entry_safe(entry, next, &dev->attached, head) {
+		if (entry->attachment.overlay_engine == eng) {
+			adf_device_detach_op(dev, eng,
+					entry->attachment.interface);
+			adf_attachment_free(entry);
+			dev->n_attached--;
+		}
+	}
+	adf_overlay_engine_sysfs_destroy(eng);
+	adf_obj_destroy(&eng->base, &dev->overlay_engines);
+	mutex_unlock(&dev->client_lock);
+}
+EXPORT_SYMBOL(adf_overlay_engine_destroy);
+
+struct adf_attachment_list *adf_attachment_find(struct list_head *list,
+		struct adf_overlay_engine *eng, struct adf_interface *intf)
+{
+	struct adf_attachment_list *entry;
+	list_for_each_entry(entry, list, head) {
+		if (entry->attachment.interface == intf &&
+				entry->attachment.overlay_engine == eng)
+			return entry;
+	}
+	return NULL;
+}
+
+int adf_attachment_validate(struct adf_device *dev,
+		struct adf_overlay_engine *eng, struct adf_interface *intf)
+{
+	struct adf_device *intf_dev = adf_interface_parent(intf);
+	struct adf_device *eng_dev = adf_overlay_engine_parent(eng);
+
+	if (intf_dev != dev) {
+		dev_err(&dev->base.dev, "can't attach interface %s belonging to device %s\n",
+				intf->base.name, intf_dev->base.name);
+		return -EINVAL;
+	}
+
+	if (eng_dev != dev) {
+		dev_err(&dev->base.dev, "can't attach overlay engine %s belonging to device %s\n",
+				eng->base.name, eng_dev->base.name);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+/**
+ * adf_attachment_allow - add a new entry to the list of allowed
+ * attachments
+ *
+ * @dev: the parent device
+ * @eng: the overlay engine
+ * @intf: the interface
+ *
+ * adf_attachment_allow() indicates that the underlying display hardware allows
+ * @intf to scan out @eng's output.  It is intended to be called at
+ * driver initialization for each supported overlay engine + interface pair.
+ *
+ * Returns 0 on success, -%EALREADY if the entry already exists, or -errno on
+ * any other failure.
+ */
+int adf_attachment_allow(struct adf_device *dev,
+		struct adf_overlay_engine *eng, struct adf_interface *intf)
+{
+	int ret;
+	struct adf_attachment_list *entry = NULL;
+
+	ret = adf_attachment_validate(dev, eng, intf);
+	if (ret < 0)
+		return ret;
+
+	mutex_lock(&dev->client_lock);
+
+	if (dev->n_attach_allowed == ADF_MAX_ATTACHMENTS) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	if (adf_attachment_find(&dev->attach_allowed, eng, intf)) {
+		ret = -EALREADY;
+		goto done;
+	}
+
+	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
+	if (!entry) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	entry->attachment.interface = intf;
+	entry->attachment.overlay_engine = eng;
+	list_add_tail(&entry->head, &dev->attach_allowed);
+	dev->n_attach_allowed++;
+
+done:
+	mutex_unlock(&dev->client_lock);
+	if (ret < 0)
+		kfree(entry);
+
+	return ret;
+}
+EXPORT_SYMBOL(adf_attachment_allow);
+
+/**
+ * adf_obj_type_str - string representation of an adf_obj_type
+ *
+ * @type: the object type
+ */
+const char *adf_obj_type_str(enum adf_obj_type type)
+{
+	switch (type) {
+	case ADF_OBJ_OVERLAY_ENGINE:
+		return "overlay engine";
+
+	case ADF_OBJ_INTERFACE:
+		return "interface";
+
+	case ADF_OBJ_DEVICE:
+		return "device";
+
+	default:
+		return "unknown";
+	}
+}
+EXPORT_SYMBOL(adf_obj_type_str);
+
+/**
+ * adf_interface_type_str - string representation of an adf_interface's type
+ *
+ * @intf: the interface
+ */
+const char *adf_interface_type_str(struct adf_interface *intf)
+{
+	switch (intf->type) {
+	case ADF_INTF_DSI:
+		return "DSI";
+
+	case ADF_INTF_eDP:
+		return "eDP";
+
+	case ADF_INTF_DPI:
+		return "DPI";
+
+	case ADF_INTF_VGA:
+		return "VGA";
+
+	case ADF_INTF_DVI:
+		return "DVI";
+
+	case ADF_INTF_HDMI:
+		return "HDMI";
+
+	case ADF_INTF_MEMORY:
+		return "memory";
+
+	default:
+		if (intf->type >= ADF_INTF_TYPE_DEVICE_CUSTOM) {
+			if (intf->ops && intf->ops->type_str)
+				return intf->ops->type_str(intf);
+			return "custom";
+		}
+		return "unknown";
+	}
+}
+EXPORT_SYMBOL(adf_interface_type_str);
+
+/**
+ * adf_event_type_str - string representation of an adf_event_type
+ *
+ * @obj: ADF object that produced the event
+ * @type: event type
+ */
+const char *adf_event_type_str(struct adf_obj *obj, enum adf_event_type type)
+{
+	switch (type) {
+	case ADF_EVENT_VSYNC:
+		return "vsync";
+
+	case ADF_EVENT_HOTPLUG:
+		return "hotplug";
+
+	default:
+		if (type >= ADF_EVENT_DEVICE_CUSTOM) {
+			if (obj->ops && obj->ops->event_type_str)
+				return obj->ops->event_type_str(obj, type);
+			return "custom";
+		}
+		return "unknown";
+	}
+}
+EXPORT_SYMBOL(adf_event_type_str);
+
+/**
+ * adf_format_str - string representation of an ADF/DRM fourcc format
+ *
+ * @format: format fourcc
+ * @buf: target buffer for the format's string representation
+ */
+void adf_format_str(u32 format, char buf[ADF_FORMAT_STR_SIZE])
+{
+	buf[0] = format & 0xFF;
+	buf[1] = (format >> 8) & 0xFF;
+	buf[2] = (format >> 16) & 0xFF;
+	buf[3] = (format >> 24) & 0xFF;
+	buf[4] = '\0';
+}
+EXPORT_SYMBOL(adf_format_str);
+
+/**
+ * adf_format_validate_yuv - validate the number and size of planes in buffers
+ * with a custom YUV format.
+ *
+ * @dev: ADF device performing the validation
+ * @buf: buffer to validate
+ * @num_planes: expected number of planes
+ * @hsub: expected horizontal chroma subsampling factor, in pixels
+ * @vsub: expected vertical chroma subsampling factor, in pixels
+ * @cpp: expected bytes per pixel for each plane (length @num_planes)
+ *
+ * adf_format_validate_yuv() is intended to be called as a helper from @dev's
+ * validate_custom_format() op.
+ *
+ * Returns 0 if @buf has the expected number of planes and each plane
+ * has sufficient size, or -EINVAL otherwise.
+ */
+int adf_format_validate_yuv(struct adf_device *dev, struct adf_buffer *buf,
+		u8 num_planes, u8 hsub, u8 vsub, u8 cpp[])
+{
+	u8 i;
+
+	if (num_planes != buf->n_planes) {
+		char format_str[ADF_FORMAT_STR_SIZE];
+		adf_format_str(buf->format, format_str);
+		dev_err(&dev->base.dev, "%u planes expected for format %s but %u planes provided\n",
+				num_planes, format_str, buf->n_planes);
+		return -EINVAL;
+	}
+
+	if (buf->w == 0 || buf->w % hsub) {
+		dev_err(&dev->base.dev, "bad buffer width %u\n", buf->w);
+		return -EINVAL;
+	}
+
+	if (buf->h == 0 || buf->h % vsub) {
+		dev_err(&dev->base.dev, "bad buffer height %u\n", buf->h);
+		return -EINVAL;
+	}
+
+	for (i = 0; i < num_planes; i++) {
+		u32 width = buf->w / (i != 0 ? hsub : 1);
+		u32 height = buf->h / (i != 0 ? vsub : 1);
+		u8 cpp = adf_format_plane_cpp(buf->format, i);
+		u32 last_line_size;
+
+		if (buf->pitch[i] < (u64) width * cpp) {
+			dev_err(&dev->base.dev, "plane %u pitch is shorter than buffer width (pitch = %u, width = %u, bpp = %u)\n",
+					i, buf->pitch[i], width, cpp * 8);
+			return -EINVAL;
+		}
+
+		switch (dev->ops->quirks.buffer_padding) {
+		case ADF_BUFFER_PADDED_TO_PITCH:
+			last_line_size = buf->pitch[i];
+			break;
+
+		case ADF_BUFFER_UNPADDED:
+			last_line_size = width * cpp;
+			break;
+
+		default:
+			BUG();
+		}
+
+		if ((u64) (height - 1) * buf->pitch[i] + last_line_size +
+				buf->offset[i] > buf->dma_bufs[i]->size) {
+			dev_err(&dev->base.dev, "plane %u buffer too small (height = %u, pitch = %u, offset = %u, size = %zu)\n",
+					i, height, buf->pitch[i],
+					buf->offset[i], buf->dma_bufs[i]->size);
+			return -EINVAL;
+		}
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(adf_format_validate_yuv);
+
+/**
+ * adf_modeinfo_set_name - sets the name of a mode from its display resolution
+ *
+ * @mode: mode
+ *
+ * adf_modeinfo_set_name() fills in @mode->name in the format
+ * "[hdisplay]x[vdisplay](i)".  It is intended to help drivers create
+ * ADF/DRM-style modelists from other mode formats.
+ */
+void adf_modeinfo_set_name(struct drm_mode_modeinfo *mode)
+{
+	bool interlaced = mode->flags & DRM_MODE_FLAG_INTERLACE;
+
+	snprintf(mode->name, DRM_DISPLAY_MODE_LEN, "%dx%d%s",
+		 mode->hdisplay, mode->vdisplay,
+		 interlaced ? "i" : "");
+}
+EXPORT_SYMBOL(adf_modeinfo_set_name);
+
+/**
+ * adf_modeinfo_set_vrefresh - sets the vrefresh of a mode from its other
+ * timing data
+ *
+ * @mode: mode
+ *
+ * adf_modeinfo_set_vrefresh() calculates @mode->vrefresh from
+ * @mode->{h,v}display and @mode->flags.  It is intended to help drivers
+ * create ADF/DRM-style modelists from other mode formats.
+ */
+void adf_modeinfo_set_vrefresh(struct drm_mode_modeinfo *mode)
+{
+	int refresh = 0;
+	unsigned int calc_val;
+
+	if (mode->vrefresh > 0)
+		return;
+
+	if (mode->htotal <= 0 || mode->vtotal <= 0)
+		return;
+
+	/* work out vrefresh the value will be x1000 */
+	calc_val = (mode->clock * 1000);
+	calc_val /= mode->htotal;
+	refresh = (calc_val + mode->vtotal / 2) / mode->vtotal;
+
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		refresh *= 2;
+	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
+		refresh /= 2;
+	if (mode->vscan > 1)
+		refresh /= mode->vscan;
+
+	mode->vrefresh = refresh;
+}
+EXPORT_SYMBOL(adf_modeinfo_set_vrefresh);
+
+static int __init adf_init(void)
+{
+	int err;
+
+	err = adf_sysfs_init();
+	if (err < 0)
+		return err;
+
+	return 0;
+}
+
+static void __exit adf_exit(void)
+{
+	adf_sysfs_destroy();
+}
+
+module_init(adf_init);
+module_exit(adf_exit);
diff --git a/drivers/video/adf/adf.h b/drivers/video/adf/adf.h
new file mode 100644
index 000000000000..3bcf1fabc23c
--- /dev/null
+++ b/drivers/video/adf/adf.h
@@ -0,0 +1,71 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __VIDEO_ADF_ADF_H
+#define __VIDEO_ADF_ADF_H
+
+#include <linux/idr.h>
+#include <linux/list.h>
+#include <video/adf.h>
+#include "sync.h"
+
+struct adf_event_refcount {
+	struct rb_node node;
+	enum adf_event_type type;
+	int refcount;
+};
+
+void adf_buffer_cleanup(struct adf_buffer *buf);
+void adf_buffer_mapping_cleanup(struct adf_buffer_mapping *mapping,
+		struct adf_buffer *buf);
+void adf_post_cleanup(struct adf_device *dev, struct adf_pending_post *post);
+
+struct adf_attachment_list *adf_attachment_find(struct list_head *list,
+		struct adf_overlay_engine *eng, struct adf_interface *intf);
+int adf_attachment_validate(struct adf_device *dev,
+		struct adf_overlay_engine *eng, struct adf_interface *intf);
+void adf_attachment_free(struct adf_attachment_list *attachment);
+
+struct adf_event_refcount *adf_obj_find_event_refcount(struct adf_obj *obj,
+		enum adf_event_type type);
+
+static inline int adf_obj_check_supports_event(struct adf_obj *obj,
+		enum adf_event_type type)
+{
+	if (!obj->ops || !obj->ops->supports_event)
+		return -EOPNOTSUPP;
+	if (!obj->ops->supports_event(obj, type))
+		return -EINVAL;
+	return 0;
+}
+
+static inline int adf_device_attach_op(struct adf_device *dev,
+		struct adf_overlay_engine *eng, struct adf_interface *intf)
+{
+	if (!dev->ops->attach)
+		return 0;
+
+	return dev->ops->attach(dev, eng, intf);
+}
+
+static inline int adf_device_detach_op(struct adf_device *dev,
+		struct adf_overlay_engine *eng, struct adf_interface *intf)
+{
+	if (!dev->ops->detach)
+		return 0;
+
+	return dev->ops->detach(dev, eng, intf);
+}
+
+#endif /* __VIDEO_ADF_ADF_H */
diff --git a/drivers/video/adf/adf_client.c b/drivers/video/adf/adf_client.c
new file mode 100644
index 000000000000..8061d8e6b9fb
--- /dev/null
+++ b/drivers/video/adf/adf_client.c
@@ -0,0 +1,811 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/kthread.h>
+#include <linux/mutex.h>
+#include <linux/slab.h>
+
+#include "sw_sync.h"
+
+#include <video/adf.h>
+#include <video/adf_client.h>
+#include <video/adf_format.h>
+
+#include "adf.h"
+
+static inline bool vsync_active(u8 state)
+{
+	return state == DRM_MODE_DPMS_ON || state == DRM_MODE_DPMS_STANDBY;
+}
+
+/**
+ * adf_interface_blank - set interface's DPMS state
+ *
+ * @intf: the interface
+ * @state: one of %DRM_MODE_DPMS_*
+ *
+ * Returns 0 on success or -errno on failure.
+ */
+int adf_interface_blank(struct adf_interface *intf, u8 state)
+{
+	struct adf_device *dev = adf_interface_parent(intf);
+	u8 prev_state;
+	bool disable_vsync;
+	bool enable_vsync;
+	int ret = 0;
+	struct adf_event_refcount *vsync_refcount;
+
+	if (!intf->ops || !intf->ops->blank)
+		return -EOPNOTSUPP;
+
+	if (state > DRM_MODE_DPMS_OFF)
+		return -EINVAL;
+
+	mutex_lock(&dev->client_lock);
+	if (state != DRM_MODE_DPMS_ON)
+		flush_kthread_worker(&dev->post_worker);
+	mutex_lock(&intf->base.event_lock);
+
+	vsync_refcount = adf_obj_find_event_refcount(&intf->base,
+			ADF_EVENT_VSYNC);
+	if (!vsync_refcount) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	prev_state = intf->dpms_state;
+	if (prev_state == state) {
+		ret = -EBUSY;
+		goto done;
+	}
+
+	disable_vsync = vsync_active(prev_state) &&
+			!vsync_active(state) &&
+			vsync_refcount->refcount;
+	enable_vsync = !vsync_active(prev_state) &&
+			vsync_active(state) &&
+			vsync_refcount->refcount;
+
+	if (disable_vsync)
+		intf->base.ops->set_event(&intf->base, ADF_EVENT_VSYNC,
+				false);
+
+	ret = intf->ops->blank(intf, state);
+	if (ret < 0) {
+		if (disable_vsync)
+			intf->base.ops->set_event(&intf->base, ADF_EVENT_VSYNC,
+					true);
+		goto done;
+	}
+
+	if (enable_vsync)
+		intf->base.ops->set_event(&intf->base, ADF_EVENT_VSYNC,
+				true);
+
+	intf->dpms_state = state;
+done:
+	mutex_unlock(&intf->base.event_lock);
+	mutex_unlock(&dev->client_lock);
+	return ret;
+}
+EXPORT_SYMBOL(adf_interface_blank);
+
+/**
+ * adf_interface_blank - get interface's current DPMS state
+ *
+ * @intf: the interface
+ *
+ * Returns one of %DRM_MODE_DPMS_*.
+ */
+u8 adf_interface_dpms_state(struct adf_interface *intf)
+{
+	struct adf_device *dev = adf_interface_parent(intf);
+	u8 dpms_state;
+
+	mutex_lock(&dev->client_lock);
+	dpms_state = intf->dpms_state;
+	mutex_unlock(&dev->client_lock);
+
+	return dpms_state;
+}
+EXPORT_SYMBOL(adf_interface_dpms_state);
+
+/**
+ * adf_interface_current_mode - get interface's current display mode
+ *
+ * @intf: the interface
+ * @mode: returns the current mode
+ */
+void adf_interface_current_mode(struct adf_interface *intf,
+		struct drm_mode_modeinfo *mode)
+{
+	struct adf_device *dev = adf_interface_parent(intf);
+
+	mutex_lock(&dev->client_lock);
+	memcpy(mode, &intf->current_mode, sizeof(*mode));
+	mutex_unlock(&dev->client_lock);
+}
+EXPORT_SYMBOL(adf_interface_current_mode);
+
+/**
+ * adf_interface_modelist - get interface's modelist
+ *
+ * @intf: the interface
+ * @modelist: storage for the modelist (optional)
+ * @n_modes: length of @modelist
+ *
+ * If @modelist is not NULL, adf_interface_modelist() will copy up to @n_modes
+ * modelist entries into @modelist.
+ *
+ * Returns the length of the modelist.
+ */
+size_t adf_interface_modelist(struct adf_interface *intf,
+		struct drm_mode_modeinfo *modelist, size_t n_modes)
+{
+	unsigned long flags;
+	size_t retval;
+
+	read_lock_irqsave(&intf->hotplug_modelist_lock, flags);
+	if (modelist)
+		memcpy(modelist, intf->modelist, sizeof(modelist[0]) *
+				min(n_modes, intf->n_modes));
+	retval = intf->n_modes;
+	read_unlock_irqrestore(&intf->hotplug_modelist_lock, flags);
+
+	return retval;
+}
+EXPORT_SYMBOL(adf_interface_modelist);
+
+/**
+ * adf_interface_set_mode - set interface's display mode
+ *
+ * @intf: the interface
+ * @mode: the new mode
+ *
+ * Returns 0 on success or -errno on failure.
+ */
+int adf_interface_set_mode(struct adf_interface *intf,
+		struct drm_mode_modeinfo *mode)
+{
+	struct adf_device *dev = adf_interface_parent(intf);
+	int ret = 0;
+
+	if (!intf->ops || !intf->ops->modeset)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&dev->client_lock);
+	flush_kthread_worker(&dev->post_worker);
+
+	ret = intf->ops->modeset(intf, mode);
+	if (ret < 0)
+		goto done;
+
+	memcpy(&intf->current_mode, mode, sizeof(*mode));
+done:
+	mutex_unlock(&dev->client_lock);
+	return ret;
+}
+EXPORT_SYMBOL(adf_interface_set_mode);
+
+/**
+ * adf_interface_screen_size - get size of screen connected to interface
+ *
+ * @intf: the interface
+ * @width_mm: returns the screen width in mm
+ * @height_mm: returns the screen width in mm
+ *
+ * Returns 0 on success or -errno on failure.
+ */
+int adf_interface_get_screen_size(struct adf_interface *intf, u16 *width_mm,
+		u16 *height_mm)
+{
+	struct adf_device *dev = adf_interface_parent(intf);
+	int ret;
+
+	if (!intf->ops || !intf->ops->screen_size)
+		return -EOPNOTSUPP;
+
+	mutex_lock(&dev->client_lock);
+	ret = intf->ops->screen_size(intf, width_mm, height_mm);
+	mutex_unlock(&dev->client_lock);
+
+	return ret;
+}
+EXPORT_SYMBOL(adf_interface_get_screen_size);
+
+/**
+ * adf_overlay_engine_supports_format - returns whether a format is in an
+ * overlay engine's supported list
+ *
+ * @eng: the overlay engine
+ * @format: format fourcc
+ */
+bool adf_overlay_engine_supports_format(struct adf_overlay_engine *eng,
+		u32 format)
+{
+	size_t i;
+	for (i = 0; i < eng->ops->n_supported_formats; i++)
+		if (format == eng->ops->supported_formats[i])
+			return true;
+
+	return false;
+}
+EXPORT_SYMBOL(adf_overlay_engine_supports_format);
+
+static int adf_buffer_validate(struct adf_buffer *buf)
+{
+	struct adf_overlay_engine *eng = buf->overlay_engine;
+	struct device *dev = &eng->base.dev;
+	struct adf_device *parent = adf_overlay_engine_parent(eng);
+	u8 hsub, vsub, num_planes, cpp[ADF_MAX_PLANES], i;
+
+	if (!adf_overlay_engine_supports_format(eng, buf->format)) {
+		char format_str[ADF_FORMAT_STR_SIZE];
+		adf_format_str(buf->format, format_str);
+		dev_err(dev, "unsupported format %s\n", format_str);
+		return -EINVAL;
+	}
+
+	if (!adf_format_is_standard(buf->format))
+		return parent->ops->validate_custom_format(parent, buf);
+
+	hsub = adf_format_horz_chroma_subsampling(buf->format);
+	vsub = adf_format_vert_chroma_subsampling(buf->format);
+	num_planes = adf_format_num_planes(buf->format);
+	for (i = 0; i < num_planes; i++)
+		cpp[i] = adf_format_plane_cpp(buf->format, i);
+
+	return adf_format_validate_yuv(parent, buf, num_planes, hsub, vsub,
+			cpp);
+}
+
+static int adf_buffer_map(struct adf_device *dev, struct adf_buffer *buf,
+		struct adf_buffer_mapping *mapping)
+{
+	int ret = 0;
+	size_t i;
+
+	for (i = 0; i < buf->n_planes; i++) {
+		struct dma_buf_attachment *attachment;
+		struct sg_table *sg_table;
+
+		attachment = dma_buf_attach(buf->dma_bufs[i], dev->dev);
+		if (IS_ERR(attachment)) {
+			ret = PTR_ERR(attachment);
+			dev_err(&dev->base.dev, "attaching plane %zu failed: %d\n",
+					i, ret);
+			goto done;
+		}
+		mapping->attachments[i] = attachment;
+
+		sg_table = dma_buf_map_attachment(attachment, DMA_TO_DEVICE);
+		if (IS_ERR(sg_table)) {
+			ret = PTR_ERR(sg_table);
+			dev_err(&dev->base.dev, "mapping plane %zu failed: %d",
+					i, ret);
+			goto done;
+		} else if (!sg_table) {
+			ret = -ENOMEM;
+			dev_err(&dev->base.dev, "mapping plane %zu failed\n",
+					i);
+			goto done;
+		}
+		mapping->sg_tables[i] = sg_table;
+	}
+
+done:
+	if (ret < 0)
+		adf_buffer_mapping_cleanup(mapping, buf);
+
+	return ret;
+}
+
+static struct sync_fence *adf_sw_complete_fence(struct adf_device *dev)
+{
+	struct sync_pt *pt;
+	struct sync_fence *complete_fence;
+
+	if (!dev->timeline) {
+		dev->timeline = sw_sync_timeline_create(dev->base.name);
+		if (!dev->timeline)
+			return ERR_PTR(-ENOMEM);
+		dev->timeline_max = 1;
+	}
+
+	dev->timeline_max++;
+	pt = sw_sync_pt_create(dev->timeline, dev->timeline_max);
+	if (!pt)
+		goto err_pt_create;
+	complete_fence = sync_fence_create(dev->base.name, pt);
+	if (!complete_fence)
+		goto err_fence_create;
+
+	return complete_fence;
+
+err_fence_create:
+	sync_pt_free(pt);
+err_pt_create:
+	dev->timeline_max--;
+	return ERR_PTR(-ENOSYS);
+}
+
+/**
+ * adf_device_post - flip to a new set of buffers
+ *
+ * @dev: device targeted by the flip
+ * @intfs: interfaces targeted by the flip
+ * @n_intfs: number of targeted interfaces
+ * @bufs: description of buffers displayed
+ * @n_bufs: number of buffers displayed
+ * @custom_data: driver-private data
+ * @custom_data_size: size of driver-private data
+ *
+ * adf_device_post() will copy @intfs, @bufs, and @custom_data, so they may
+ * point to variables on the stack.  adf_device_post() also takes its own
+ * reference on each of the dma-bufs in @bufs.  The adf_device_post_nocopy()
+ * variant transfers ownership of these resources to ADF instead.
+ *
+ * On success, returns a sync fence which signals when the buffers are removed
+ * from the screen.  On failure, returns ERR_PTR(-errno).
+ */
+struct sync_fence *adf_device_post(struct adf_device *dev,
+		struct adf_interface **intfs, size_t n_intfs,
+		struct adf_buffer *bufs, size_t n_bufs, void *custom_data,
+		size_t custom_data_size)
+{
+	struct adf_interface **intfs_copy = NULL;
+	struct adf_buffer *bufs_copy = NULL;
+	void *custom_data_copy = NULL;
+	struct sync_fence *ret;
+	size_t i;
+
+	intfs_copy = kzalloc(sizeof(intfs_copy[0]) * n_intfs, GFP_KERNEL);
+	if (!intfs_copy)
+		return ERR_PTR(-ENOMEM);
+
+	bufs_copy = kzalloc(sizeof(bufs_copy[0]) * n_bufs, GFP_KERNEL);
+	if (!bufs_copy) {
+		ret = ERR_PTR(-ENOMEM);
+		goto err_alloc;
+	}
+
+	custom_data_copy = kzalloc(custom_data_size, GFP_KERNEL);
+	if (!custom_data_copy) {
+		ret = ERR_PTR(-ENOMEM);
+		goto err_alloc;
+	}
+
+	for (i = 0; i < n_bufs; i++) {
+		size_t j;
+		for (j = 0; j < bufs[i].n_planes; j++)
+			get_dma_buf(bufs[i].dma_bufs[j]);
+	}
+
+	memcpy(intfs_copy, intfs, sizeof(intfs_copy[0]) * n_intfs);
+	memcpy(bufs_copy, bufs, sizeof(bufs_copy[0]) * n_bufs);
+	memcpy(custom_data_copy, custom_data, custom_data_size);
+
+	ret = adf_device_post_nocopy(dev, intfs_copy, n_intfs, bufs_copy,
+			n_bufs, custom_data_copy, custom_data_size);
+	if (IS_ERR(ret))
+		goto err_post;
+
+	return ret;
+
+err_post:
+	for (i = 0; i < n_bufs; i++) {
+		size_t j;
+		for (j = 0; j < bufs[i].n_planes; j++)
+			dma_buf_put(bufs[i].dma_bufs[j]);
+	}
+err_alloc:
+	kfree(custom_data_copy);
+	kfree(bufs_copy);
+	kfree(intfs_copy);
+	return ret;
+}
+EXPORT_SYMBOL(adf_device_post);
+
+/**
+ * adf_device_post_nocopy - flip to a new set of buffers
+ *
+ * adf_device_post_nocopy() has the same behavior as adf_device_post(),
+ * except ADF does not copy @intfs, @bufs, or @custom_data, and it does
+ * not take an extra reference on the dma-bufs in @bufs.
+ *
+ * @intfs, @bufs, and @custom_data must point to buffers allocated by
+ * kmalloc().  On success, ADF takes ownership of these buffers and the dma-bufs
+ * in @bufs, and will kfree()/dma_buf_put() them when they are no longer needed.
+ * On failure, adf_device_post_nocopy() does NOT take ownership of these
+ * buffers or the dma-bufs, and the caller must clean them up.
+ *
+ * adf_device_post_nocopy() is mainly intended for implementing ADF's ioctls.
+ * Clients may find the nocopy variant useful in limited cases, but most should
+ * call adf_device_post() instead.
+ */
+struct sync_fence *adf_device_post_nocopy(struct adf_device *dev,
+		struct adf_interface **intfs, size_t n_intfs,
+		struct adf_buffer *bufs, size_t n_bufs,
+		void *custom_data, size_t custom_data_size)
+{
+	struct adf_pending_post *cfg;
+	struct adf_buffer_mapping *mappings;
+	struct sync_fence *ret;
+	size_t i;
+	int err;
+
+	cfg = kzalloc(sizeof(*cfg), GFP_KERNEL);
+	if (!cfg)
+		return ERR_PTR(-ENOMEM);
+
+	mappings = kzalloc(sizeof(mappings[0]) * n_bufs, GFP_KERNEL);
+	if (!mappings) {
+		ret = ERR_PTR(-ENOMEM);
+		goto err_alloc;
+	}
+
+	mutex_lock(&dev->client_lock);
+
+	for (i = 0; i < n_bufs; i++) {
+		err = adf_buffer_validate(&bufs[i]);
+		if (err < 0) {
+			ret = ERR_PTR(err);
+			goto err_buf;
+		}
+
+		err = adf_buffer_map(dev, &bufs[i], &mappings[i]);
+		if (err < 0) {
+			ret = ERR_PTR(err);
+			goto err_buf;
+		}
+	}
+
+	INIT_LIST_HEAD(&cfg->head);
+	cfg->config.n_bufs = n_bufs;
+	cfg->config.bufs = bufs;
+	cfg->config.mappings = mappings;
+	cfg->config.custom_data = custom_data;
+	cfg->config.custom_data_size = custom_data_size;
+
+	err = dev->ops->validate(dev, &cfg->config, &cfg->state);
+	if (err < 0) {
+		ret = ERR_PTR(err);
+		goto err_buf;
+	}
+
+	mutex_lock(&dev->post_lock);
+
+	if (dev->ops->complete_fence)
+		ret = dev->ops->complete_fence(dev, &cfg->config,
+				cfg->state);
+	else
+		ret = adf_sw_complete_fence(dev);
+
+	if (IS_ERR(ret))
+		goto err_fence;
+
+	list_add_tail(&cfg->head, &dev->post_list);
+	queue_kthread_work(&dev->post_worker, &dev->post_work);
+	mutex_unlock(&dev->post_lock);
+	mutex_unlock(&dev->client_lock);
+	kfree(intfs);
+	return ret;
+
+err_fence:
+	mutex_unlock(&dev->post_lock);
+
+err_buf:
+	for (i = 0; i < n_bufs; i++)
+		adf_buffer_mapping_cleanup(&mappings[i], &bufs[i]);
+
+	mutex_unlock(&dev->client_lock);
+	kfree(mappings);
+
+err_alloc:
+	kfree(cfg);
+	return ret;
+}
+EXPORT_SYMBOL(adf_device_post_nocopy);
+
+static void adf_attachment_list_to_array(struct adf_device *dev,
+		struct list_head *src, struct adf_attachment *dst, size_t size)
+{
+	struct adf_attachment_list *entry;
+	size_t i = 0;
+
+	if (!dst)
+		return;
+
+	list_for_each_entry(entry, src, head) {
+		if (i == size)
+			return;
+		dst[i] = entry->attachment;
+		i++;
+	}
+}
+
+/**
+ * adf_device_attachments - get device's list of active attachments
+ *
+ * @dev: the device
+ * @attachments: storage for the attachment list (optional)
+ * @n_attachments: length of @attachments
+ *
+ * If @attachments is not NULL, adf_device_attachments() will copy up to
+ * @n_attachments entries into @attachments.
+ *
+ * Returns the length of the active attachment list.
+ */
+size_t adf_device_attachments(struct adf_device *dev,
+		struct adf_attachment *attachments, size_t n_attachments)
+{
+	size_t retval;
+
+	mutex_lock(&dev->client_lock);
+	adf_attachment_list_to_array(dev, &dev->attached, attachments,
+			n_attachments);
+	retval = dev->n_attached;
+	mutex_unlock(&dev->client_lock);
+
+	return retval;
+}
+EXPORT_SYMBOL(adf_device_attachments);
+
+/**
+ * adf_device_attachments_allowed - get device's list of allowed attachments
+ *
+ * @dev: the device
+ * @attachments: storage for the attachment list (optional)
+ * @n_attachments: length of @attachments
+ *
+ * If @attachments is not NULL, adf_device_attachments_allowed() will copy up to
+ * @n_attachments entries into @attachments.
+ *
+ * Returns the length of the allowed attachment list.
+ */
+size_t adf_device_attachments_allowed(struct adf_device *dev,
+		struct adf_attachment *attachments, size_t n_attachments)
+{
+	size_t retval;
+
+	mutex_lock(&dev->client_lock);
+	adf_attachment_list_to_array(dev, &dev->attach_allowed, attachments,
+			n_attachments);
+	retval = dev->n_attach_allowed;
+	mutex_unlock(&dev->client_lock);
+
+	return retval;
+}
+EXPORT_SYMBOL(adf_device_attachments_allowed);
+
+/**
+ * adf_device_attached - return whether an overlay engine and interface are
+ * attached
+ *
+ * @dev: the parent device
+ * @eng: the overlay engine
+ * @intf: the interface
+ */
+bool adf_device_attached(struct adf_device *dev, struct adf_overlay_engine *eng,
+		struct adf_interface *intf)
+{
+	struct adf_attachment_list *attachment;
+
+	mutex_lock(&dev->client_lock);
+	attachment = adf_attachment_find(&dev->attached, eng, intf);
+	mutex_unlock(&dev->client_lock);
+
+	return attachment != NULL;
+}
+EXPORT_SYMBOL(adf_device_attached);
+
+/**
+ * adf_device_attach_allowed - return whether the ADF device supports attaching
+ * an overlay engine and interface
+ *
+ * @dev: the parent device
+ * @eng: the overlay engine
+ * @intf: the interface
+ */
+bool adf_device_attach_allowed(struct adf_device *dev,
+		struct adf_overlay_engine *eng, struct adf_interface *intf)
+{
+	struct adf_attachment_list *attachment;
+
+	mutex_lock(&dev->client_lock);
+	attachment = adf_attachment_find(&dev->attach_allowed, eng, intf);
+	mutex_unlock(&dev->client_lock);
+
+	return attachment != NULL;
+}
+EXPORT_SYMBOL(adf_device_attach_allowed);
+/**
+ * adf_device_attach - attach an overlay engine to an interface
+ *
+ * @dev: the parent device
+ * @eng: the overlay engine
+ * @intf: the interface
+ *
+ * Returns 0 on success, -%EINVAL if attaching @intf and @eng is not allowed,
+ * -%EALREADY if @intf and @eng are already attached, or -errno on any other
+ * failure.
+ */
+int adf_device_attach(struct adf_device *dev, struct adf_overlay_engine *eng,
+		struct adf_interface *intf)
+{
+	int ret;
+	struct adf_attachment_list *attachment = NULL;
+
+	ret = adf_attachment_validate(dev, eng, intf);
+	if (ret < 0)
+		return ret;
+
+	mutex_lock(&dev->client_lock);
+
+	if (dev->n_attached == ADF_MAX_ATTACHMENTS) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	if (!adf_attachment_find(&dev->attach_allowed, eng, intf)) {
+		ret = -EINVAL;
+		goto done;
+	}
+
+	if (adf_attachment_find(&dev->attached, eng, intf)) {
+		ret = -EALREADY;
+		goto done;
+	}
+
+	ret = adf_device_attach_op(dev, eng, intf);
+	if (ret < 0)
+		goto done;
+
+	attachment = kzalloc(sizeof(*attachment), GFP_KERNEL);
+	if (!attachment) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	attachment->attachment.interface = intf;
+	attachment->attachment.overlay_engine = eng;
+	list_add_tail(&attachment->head, &dev->attached);
+	dev->n_attached++;
+
+done:
+	mutex_unlock(&dev->client_lock);
+	if (ret < 0)
+		kfree(attachment);
+
+	return ret;
+}
+EXPORT_SYMBOL(adf_device_attach);
+
+/**
+ * adf_device_detach - detach an overlay engine from an interface
+ *
+ * @dev: the parent device
+ * @eng: the overlay engine
+ * @intf: the interface
+ *
+ * Returns 0 on success, -%EINVAL if @intf and @eng are not attached,
+ * or -errno on any other failure.
+ */
+int adf_device_detach(struct adf_device *dev, struct adf_overlay_engine *eng,
+		struct adf_interface *intf)
+{
+	int ret;
+	struct adf_attachment_list *attachment;
+
+	ret = adf_attachment_validate(dev, eng, intf);
+	if (ret < 0)
+		return ret;
+
+	mutex_lock(&dev->client_lock);
+
+	attachment = adf_attachment_find(&dev->attached, eng, intf);
+	if (!attachment) {
+		ret = -EINVAL;
+		goto done;
+	}
+
+	ret = adf_device_detach_op(dev, eng, intf);
+	if (ret < 0)
+		goto done;
+
+	adf_attachment_free(attachment);
+	dev->n_attached--;
+done:
+	mutex_unlock(&dev->client_lock);
+	return ret;
+}
+EXPORT_SYMBOL(adf_device_detach);
+
+/**
+ * adf_interface_simple_buffer_alloc - allocate a simple buffer
+ *
+ * @intf: target interface
+ * @w: width in pixels
+ * @h: height in pixels
+ * @format: format fourcc
+ * @dma_buf: returns the allocated buffer
+ * @offset: returns the byte offset of the allocated buffer's first pixel
+ * @pitch: returns the allocated buffer's pitch
+ *
+ * See &struct adf_simple_buffer_alloc for a description of simple buffers and
+ * their limitations.
+ *
+ * Returns 0 on success or -errno on failure.
+ */
+int adf_interface_simple_buffer_alloc(struct adf_interface *intf, u16 w, u16 h,
+		u32 format, struct dma_buf **dma_buf, u32 *offset, u32 *pitch)
+{
+	if (!intf->ops || !intf->ops->alloc_simple_buffer)
+		return -EOPNOTSUPP;
+
+	if (!adf_format_is_rgb(format))
+		return -EINVAL;
+
+	return intf->ops->alloc_simple_buffer(intf, w, h, format, dma_buf,
+			offset, pitch);
+}
+EXPORT_SYMBOL(adf_interface_simple_buffer_alloc);
+
+/**
+ * adf_interface_simple_post - flip to a single buffer
+ *
+ * @intf: interface targeted by the flip
+ * @buf: buffer to display
+ *
+ * adf_interface_simple_post() can be used generically for simple display
+ * configurations, since the client does not need to provide any driver-private
+ * configuration data.
+ *
+ * adf_interface_simple_post() has the same copying semantics as
+ * adf_device_post().
+ *
+ * On success, returns a sync fence which signals when the buffer is removed
+ * from the screen.  On failure, returns ERR_PTR(-errno).
+ */
+struct sync_fence *adf_interface_simple_post(struct adf_interface *intf,
+		struct adf_buffer *buf)
+{
+	size_t custom_data_size = 0;
+	void *custom_data = NULL;
+	struct sync_fence *ret;
+
+	if (intf->ops && intf->ops->describe_simple_post) {
+		int err;
+
+		custom_data = kzalloc(ADF_MAX_CUSTOM_DATA_SIZE, GFP_KERNEL);
+		if (!custom_data) {
+			ret = ERR_PTR(-ENOMEM);
+			goto done;
+		}
+
+		err = intf->ops->describe_simple_post(intf, buf, custom_data,
+				&custom_data_size);
+		if (err < 0) {
+			ret = ERR_PTR(err);
+			goto done;
+		}
+	}
+
+	ret = adf_device_post(adf_interface_parent(intf), &intf, 1, buf, 1,
+			custom_data, custom_data_size);
+done:
+	kfree(custom_data);
+	return ret;
+}
+EXPORT_SYMBOL(adf_interface_simple_post);
diff --git a/drivers/video/adf/adf_fbdev.c b/drivers/video/adf/adf_fbdev.c
new file mode 100644
index 000000000000..a5b53bc08c3f
--- /dev/null
+++ b/drivers/video/adf/adf_fbdev.c
@@ -0,0 +1,665 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/vmalloc.h>
+
+#include <video/adf.h>
+#include <video/adf_client.h>
+#include <video/adf_fbdev.h>
+#include <video/adf_format.h>
+
+#include "adf.h"
+
+struct adf_fbdev_format {
+	u32 fourcc;
+	u32 bpp;
+	u32 r_length;
+	u32 g_length;
+	u32 b_length;
+	u32 a_length;
+	u32 r_offset;
+	u32 g_offset;
+	u32 b_offset;
+	u32 a_offset;
+};
+
+static const struct adf_fbdev_format format_table[] = {
+	{DRM_FORMAT_RGB332, 8, 3, 3, 2, 0, 5, 2, 0, 0},
+	{DRM_FORMAT_BGR233, 8, 3, 3, 2, 0, 0, 3, 5, 0},
+
+	{DRM_FORMAT_XRGB4444, 16, 4, 4, 4, 0, 8, 4, 0, 0},
+	{DRM_FORMAT_XBGR4444, 16, 4, 4, 4, 0, 0, 4, 8, 0},
+	{DRM_FORMAT_RGBX4444, 16, 4, 4, 4, 0, 12, 8, 4, 0},
+	{DRM_FORMAT_BGRX4444, 16, 4, 4, 4, 0, 0, 4, 8, 0},
+
+	{DRM_FORMAT_ARGB4444, 16, 4, 4, 4, 4, 8, 4, 0, 12},
+	{DRM_FORMAT_ABGR4444, 16, 4, 4, 4, 4, 0, 4, 8, 12},
+	{DRM_FORMAT_RGBA4444, 16, 4, 4, 4, 4, 12, 8, 4, 0},
+	{DRM_FORMAT_BGRA4444, 16, 4, 4, 4, 4, 0, 4, 8, 0},
+
+	{DRM_FORMAT_XRGB1555, 16, 5, 5, 5, 0, 10, 5, 0, 0},
+	{DRM_FORMAT_XBGR1555, 16, 5, 5, 5, 0, 0, 5, 10, 0},
+	{DRM_FORMAT_RGBX5551, 16, 5, 5, 5, 0, 11, 6, 1, 0},
+	{DRM_FORMAT_BGRX5551, 16, 5, 5, 5, 0, 1, 6, 11, 0},
+
+	{DRM_FORMAT_ARGB1555, 16, 5, 5, 5, 1, 10, 5, 0, 15},
+	{DRM_FORMAT_ABGR1555, 16, 5, 5, 5, 1, 0, 5, 10, 15},
+	{DRM_FORMAT_RGBA5551, 16, 5, 5, 5, 1, 11, 6, 1, 0},
+	{DRM_FORMAT_BGRA5551, 16, 5, 5, 5, 1, 1, 6, 11, 0},
+
+	{DRM_FORMAT_RGB565, 16, 5, 6, 5, 0, 11, 5, 0, 0},
+	{DRM_FORMAT_BGR565, 16, 5, 6, 5, 0, 0, 5, 11, 0},
+
+	{DRM_FORMAT_RGB888, 24, 8, 8, 8, 0, 16, 8, 0, 0},
+	{DRM_FORMAT_BGR888, 24, 8, 8, 8, 0, 0, 8, 16, 0},
+
+	{DRM_FORMAT_XRGB8888, 32, 8, 8, 8, 0, 16, 8, 0, 0},
+	{DRM_FORMAT_XBGR8888, 32, 8, 8, 8, 0, 0, 8, 16, 0},
+	{DRM_FORMAT_RGBX8888, 32, 8, 8, 8, 0, 24, 16, 8, 0},
+	{DRM_FORMAT_BGRX8888, 32, 8, 8, 8, 0, 8, 16, 24, 0},
+
+	{DRM_FORMAT_ARGB8888, 32, 8, 8, 8, 8, 16, 8, 0, 24},
+	{DRM_FORMAT_ABGR8888, 32, 8, 8, 8, 8, 0, 8, 16, 24},
+	{DRM_FORMAT_RGBA8888, 32, 8, 8, 8, 8, 24, 16, 8, 0},
+	{DRM_FORMAT_BGRA8888, 32, 8, 8, 8, 8, 8, 16, 24, 0},
+
+	{DRM_FORMAT_XRGB2101010, 32, 10, 10, 10, 0, 20, 10, 0, 0},
+	{DRM_FORMAT_XBGR2101010, 32, 10, 10, 10, 0, 0, 10, 20, 0},
+	{DRM_FORMAT_RGBX1010102, 32, 10, 10, 10, 0, 22, 12, 2, 0},
+	{DRM_FORMAT_BGRX1010102, 32, 10, 10, 10, 0, 2, 12, 22, 0},
+
+	{DRM_FORMAT_ARGB2101010, 32, 10, 10, 10, 2, 20, 10, 0, 30},
+	{DRM_FORMAT_ABGR2101010, 32, 10, 10, 10, 2, 0, 10, 20, 30},
+	{DRM_FORMAT_RGBA1010102, 32, 10, 10, 10, 2, 22, 12, 2, 0},
+	{DRM_FORMAT_BGRA1010102, 32, 10, 10, 10, 2, 2, 12, 22, 0},
+};
+
+static u32 drm_fourcc_from_fb_var(struct fb_var_screeninfo *var)
+{
+	size_t i;
+	for (i = 0; i < ARRAY_SIZE(format_table); i++) {
+		const struct adf_fbdev_format *f = &format_table[i];
+		if (var->red.length == f->r_length &&
+			var->red.offset == f->r_offset &&
+			var->green.length == f->g_length &&
+			var->green.offset == f->g_offset &&
+			var->blue.length == f->b_length &&
+			var->blue.offset == f->b_offset &&
+			var->transp.length == f->a_length &&
+			(var->transp.length == 0 ||
+					var->transp.offset == f->a_offset))
+			return f->fourcc;
+	}
+
+	return 0;
+}
+
+static const struct adf_fbdev_format *fbdev_format_info(u32 format)
+{
+	size_t i;
+	for (i = 0; i < ARRAY_SIZE(format_table); i++) {
+		const struct adf_fbdev_format *f = &format_table[i];
+		if (f->fourcc == format)
+			return f;
+	}
+
+	BUG();
+}
+
+void adf_modeinfo_to_fb_videomode(const struct drm_mode_modeinfo *mode,
+		struct fb_videomode *vmode)
+{
+	memset(vmode, 0, sizeof(*vmode));
+
+	vmode->refresh = mode->vrefresh;
+
+	vmode->xres = mode->hdisplay;
+	vmode->yres = mode->vdisplay;
+
+	vmode->pixclock = mode->clock ? KHZ2PICOS(mode->clock) : 0;
+	vmode->left_margin = mode->htotal - mode->hsync_end;
+	vmode->right_margin = mode->hsync_start - mode->hdisplay;
+	vmode->upper_margin = mode->vtotal - mode->vsync_end;
+	vmode->lower_margin = mode->vsync_start - mode->vdisplay;
+	vmode->hsync_len = mode->hsync_end - mode->hsync_start;
+	vmode->vsync_len = mode->vsync_end - mode->vsync_start;
+
+	vmode->sync = 0;
+	if (mode->flags & DRM_MODE_FLAG_PHSYNC)
+		vmode->sync |= FB_SYNC_HOR_HIGH_ACT;
+	if (mode->flags & DRM_MODE_FLAG_PVSYNC)
+		vmode->sync |= FB_SYNC_VERT_HIGH_ACT;
+	if (mode->flags & DRM_MODE_FLAG_PCSYNC)
+		vmode->sync |= FB_SYNC_COMP_HIGH_ACT;
+	if (mode->flags & DRM_MODE_FLAG_BCAST)
+		vmode->sync |= FB_SYNC_BROADCAST;
+
+	vmode->vmode = 0;
+	if (mode->flags & DRM_MODE_FLAG_INTERLACE)
+		vmode->vmode |= FB_VMODE_INTERLACED;
+	if (mode->flags & DRM_MODE_FLAG_DBLSCAN)
+		vmode->vmode |= FB_VMODE_DOUBLE;
+}
+EXPORT_SYMBOL(adf_modeinfo_to_fb_videomode);
+
+void adf_modeinfo_from_fb_videomode(const struct fb_videomode *vmode,
+		struct drm_mode_modeinfo *mode)
+{
+	memset(mode, 0, sizeof(*mode));
+
+	mode->hdisplay = vmode->xres;
+	mode->hsync_start = mode->hdisplay + vmode->right_margin;
+	mode->hsync_end = mode->hsync_start + vmode->hsync_len;
+	mode->htotal = mode->hsync_end + vmode->left_margin;
+
+	mode->vdisplay = vmode->yres;
+	mode->vsync_start = mode->vdisplay + vmode->lower_margin;
+	mode->vsync_end = mode->vsync_start + vmode->vsync_len;
+	mode->vtotal = mode->vsync_end + vmode->upper_margin;
+
+	mode->clock = vmode->pixclock ? PICOS2KHZ(vmode->pixclock) : 0;
+
+	mode->flags = 0;
+	if (vmode->sync & FB_SYNC_HOR_HIGH_ACT)
+		mode->flags |= DRM_MODE_FLAG_PHSYNC;
+	if (vmode->sync & FB_SYNC_VERT_HIGH_ACT)
+		mode->flags |= DRM_MODE_FLAG_PVSYNC;
+	if (vmode->sync & FB_SYNC_COMP_HIGH_ACT)
+		mode->flags |= DRM_MODE_FLAG_PCSYNC;
+	if (vmode->sync & FB_SYNC_BROADCAST)
+		mode->flags |= DRM_MODE_FLAG_BCAST;
+	if (vmode->vmode & FB_VMODE_INTERLACED)
+		mode->flags |= DRM_MODE_FLAG_INTERLACE;
+	if (vmode->vmode & FB_VMODE_DOUBLE)
+		mode->flags |= DRM_MODE_FLAG_DBLSCAN;
+
+	if (vmode->refresh)
+		mode->vrefresh = vmode->refresh;
+	else
+		adf_modeinfo_set_vrefresh(mode);
+
+	if (vmode->name)
+		strlcpy(mode->name, vmode->name, sizeof(mode->name));
+	else
+		adf_modeinfo_set_name(mode);
+}
+EXPORT_SYMBOL(adf_modeinfo_from_fb_videomode);
+
+static int adf_fbdev_post(struct adf_fbdev *fbdev)
+{
+	struct adf_buffer buf;
+	struct sync_fence *complete_fence;
+	int ret = 0;
+
+	memset(&buf, 0, sizeof(buf));
+	buf.overlay_engine = fbdev->eng;
+	buf.w = fbdev->info->var.xres;
+	buf.h = fbdev->info->var.yres;
+	buf.format = fbdev->format;
+	buf.dma_bufs[0] = fbdev->dma_buf;
+	buf.offset[0] = fbdev->offset +
+			fbdev->info->var.yoffset * fbdev->pitch +
+			fbdev->info->var.xoffset *
+			(fbdev->info->var.bits_per_pixel / 8);
+	buf.pitch[0] = fbdev->pitch;
+	buf.n_planes = 1;
+
+	complete_fence = adf_interface_simple_post(fbdev->intf, &buf);
+	if (IS_ERR(complete_fence)) {
+		ret = PTR_ERR(complete_fence);
+		goto done;
+	}
+
+	sync_fence_put(complete_fence);
+done:
+	return ret;
+}
+
+static const u16 vga_palette[][3] = {
+	{0x0000, 0x0000, 0x0000},
+	{0x0000, 0x0000, 0xAAAA},
+	{0x0000, 0xAAAA, 0x0000},
+	{0x0000, 0xAAAA, 0xAAAA},
+	{0xAAAA, 0x0000, 0x0000},
+	{0xAAAA, 0x0000, 0xAAAA},
+	{0xAAAA, 0x5555, 0x0000},
+	{0xAAAA, 0xAAAA, 0xAAAA},
+	{0x5555, 0x5555, 0x5555},
+	{0x5555, 0x5555, 0xFFFF},
+	{0x5555, 0xFFFF, 0x5555},
+	{0x5555, 0xFFFF, 0xFFFF},
+	{0xFFFF, 0x5555, 0x5555},
+	{0xFFFF, 0x5555, 0xFFFF},
+	{0xFFFF, 0xFFFF, 0x5555},
+	{0xFFFF, 0xFFFF, 0xFFFF},
+};
+
+static int adf_fb_alloc(struct adf_fbdev *fbdev)
+{
+	int ret;
+
+	ret = adf_interface_simple_buffer_alloc(fbdev->intf,
+			fbdev->default_xres_virtual,
+			fbdev->default_yres_virtual,
+			fbdev->default_format,
+			&fbdev->dma_buf, &fbdev->offset, &fbdev->pitch);
+	if (ret < 0) {
+		dev_err(fbdev->info->dev, "allocating fb failed: %d\n", ret);
+		return ret;
+	}
+
+	fbdev->vaddr = dma_buf_vmap(fbdev->dma_buf);
+	if (!fbdev->vaddr) {
+		ret = -ENOMEM;
+		dev_err(fbdev->info->dev, "vmapping fb failed\n");
+		goto err_vmap;
+	}
+	fbdev->info->fix.line_length = fbdev->pitch;
+	fbdev->info->var.xres_virtual = fbdev->default_xres_virtual;
+	fbdev->info->var.yres_virtual = fbdev->default_yres_virtual;
+	fbdev->info->fix.smem_len = fbdev->dma_buf->size;
+	fbdev->info->screen_base = fbdev->vaddr;
+
+	return 0;
+
+err_vmap:
+	dma_buf_put(fbdev->dma_buf);
+	return ret;
+}
+
+static void adf_fb_destroy(struct adf_fbdev *fbdev)
+{
+	dma_buf_vunmap(fbdev->dma_buf, fbdev->vaddr);
+	dma_buf_put(fbdev->dma_buf);
+}
+
+static void adf_fbdev_set_format(struct adf_fbdev *fbdev, u32 format)
+{
+	size_t i;
+	const struct adf_fbdev_format *info = fbdev_format_info(format);
+	for (i = 0; i < ARRAY_SIZE(vga_palette); i++) {
+		u16 r = vga_palette[i][0];
+		u16 g = vga_palette[i][1];
+		u16 b = vga_palette[i][2];
+
+		r >>= (16 - info->r_length);
+		g >>= (16 - info->g_length);
+		b >>= (16 - info->b_length);
+
+		fbdev->pseudo_palette[i] =
+			(r << info->r_offset) |
+			(g << info->g_offset) |
+			(b << info->b_offset);
+
+		if (info->a_length) {
+			u16 a = BIT(info->a_length) - 1;
+			fbdev->pseudo_palette[i] |= (a << info->a_offset);
+		}
+	}
+
+	fbdev->info->var.bits_per_pixel = adf_format_bpp(format);
+	fbdev->info->var.red.length = info->r_length;
+	fbdev->info->var.red.offset = info->r_offset;
+	fbdev->info->var.green.length = info->g_length;
+	fbdev->info->var.green.offset = info->g_offset;
+	fbdev->info->var.blue.length = info->b_length;
+	fbdev->info->var.blue.offset = info->b_offset;
+	fbdev->info->var.transp.length = info->a_length;
+	fbdev->info->var.transp.offset = info->a_offset;
+	fbdev->format = format;
+}
+
+static void adf_fbdev_fill_modelist(struct adf_fbdev *fbdev)
+{
+	struct drm_mode_modeinfo *modelist;
+	struct fb_videomode fbmode;
+	size_t n_modes, i;
+	int ret = 0;
+
+	n_modes = adf_interface_modelist(fbdev->intf, NULL, 0);
+	modelist = kzalloc(sizeof(modelist[0]) * n_modes, GFP_KERNEL);
+	if (!modelist) {
+		dev_warn(fbdev->info->dev, "allocating new modelist failed; keeping old modelist\n");
+		return;
+	}
+	adf_interface_modelist(fbdev->intf, modelist, n_modes);
+
+	fb_destroy_modelist(&fbdev->info->modelist);
+
+	for (i = 0; i < n_modes; i++) {
+		adf_modeinfo_to_fb_videomode(&modelist[i], &fbmode);
+		ret = fb_add_videomode(&fbmode, &fbdev->info->modelist);
+		if (ret < 0)
+			dev_warn(fbdev->info->dev, "adding mode %s to modelist failed: %d\n",
+					modelist[i].name, ret);
+	}
+
+	kfree(modelist);
+}
+
+/**
+ * adf_fbdev_open - default implementation of fbdev open op
+ */
+int adf_fbdev_open(struct fb_info *info, int user)
+{
+	struct adf_fbdev *fbdev = info->par;
+	int ret;
+
+	mutex_lock(&fbdev->refcount_lock);
+
+	if (unlikely(fbdev->refcount == UINT_MAX)) {
+		ret = -EMFILE;
+		goto done;
+	}
+
+	if (!fbdev->refcount) {
+		struct drm_mode_modeinfo mode;
+		struct fb_videomode fbmode;
+		struct adf_device *dev = adf_interface_parent(fbdev->intf);
+
+		ret = adf_device_attach(dev, fbdev->eng, fbdev->intf);
+		if (ret < 0 && ret != -EALREADY)
+			goto done;
+
+		ret = adf_fb_alloc(fbdev);
+		if (ret < 0)
+			goto done;
+
+		adf_interface_current_mode(fbdev->intf, &mode);
+		adf_modeinfo_to_fb_videomode(&mode, &fbmode);
+		fb_videomode_to_var(&fbdev->info->var, &fbmode);
+
+		adf_fbdev_set_format(fbdev, fbdev->default_format);
+		adf_fbdev_fill_modelist(fbdev);
+	}
+
+	ret = adf_fbdev_post(fbdev);
+	if (ret < 0) {
+		if (!fbdev->refcount)
+			adf_fb_destroy(fbdev);
+		goto done;
+	}
+
+	fbdev->refcount++;
+done:
+	mutex_unlock(&fbdev->refcount_lock);
+	return ret;
+}
+EXPORT_SYMBOL(adf_fbdev_open);
+
+/**
+ * adf_fbdev_release - default implementation of fbdev release op
+ */
+int adf_fbdev_release(struct fb_info *info, int user)
+{
+	struct adf_fbdev *fbdev = info->par;
+	mutex_lock(&fbdev->refcount_lock);
+	BUG_ON(!fbdev->refcount);
+	fbdev->refcount--;
+	if (!fbdev->refcount)
+		adf_fb_destroy(fbdev);
+	mutex_unlock(&fbdev->refcount_lock);
+	return 0;
+}
+EXPORT_SYMBOL(adf_fbdev_release);
+
+/**
+ * adf_fbdev_check_var - default implementation of fbdev check_var op
+ */
+int adf_fbdev_check_var(struct fb_var_screeninfo *var, struct fb_info *info)
+{
+	struct adf_fbdev *fbdev = info->par;
+	bool valid_format = true;
+	u32 format = drm_fourcc_from_fb_var(var);
+	u32 pitch = var->xres_virtual * var->bits_per_pixel / 8;
+
+	if (!format) {
+		dev_dbg(info->dev, "%s: unrecognized format\n", __func__);
+		valid_format = false;
+	}
+
+	if (valid_format && var->grayscale) {
+		dev_dbg(info->dev, "%s: grayscale modes not supported\n",
+				__func__);
+		valid_format = false;
+	}
+
+	if (valid_format && var->nonstd) {
+		dev_dbg(info->dev, "%s: nonstandard formats not supported\n",
+				__func__);
+		valid_format = false;
+	}
+
+	if (valid_format && !adf_overlay_engine_supports_format(fbdev->eng,
+			format)) {
+		char format_str[ADF_FORMAT_STR_SIZE];
+		adf_format_str(format, format_str);
+		dev_dbg(info->dev, "%s: format %s not supported by overlay engine %s\n",
+				__func__, format_str, fbdev->eng->base.name);
+		valid_format = false;
+	}
+
+	if (valid_format && pitch > fbdev->pitch) {
+		dev_dbg(info->dev, "%s: fb pitch too small for var (pitch = %u, xres_virtual = %u, bits_per_pixel = %u)\n",
+				__func__, fbdev->pitch, var->xres_virtual,
+				var->bits_per_pixel);
+		valid_format = false;
+	}
+
+	if (valid_format && var->yres_virtual > fbdev->default_yres_virtual) {
+		dev_dbg(info->dev, "%s: fb height too small for var (h = %u, yres_virtual = %u)\n",
+				__func__, fbdev->default_yres_virtual,
+				var->yres_virtual);
+		valid_format = false;
+	}
+
+	if (valid_format) {
+		var->activate = info->var.activate;
+		var->height = info->var.height;
+		var->width = info->var.width;
+		var->accel_flags = info->var.accel_flags;
+		var->rotate = info->var.rotate;
+		var->colorspace = info->var.colorspace;
+		/* userspace can't change these */
+	} else {
+		/* if any part of the format is invalid then fixing it up is
+		   impractical, so save just the modesetting bits and
+		   overwrite everything else */
+		struct fb_videomode mode;
+		fb_var_to_videomode(&mode, var);
+		memcpy(var, &info->var, sizeof(*var));
+		fb_videomode_to_var(var, &mode);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(adf_fbdev_check_var);
+
+/**
+ * adf_fbdev_set_par - default implementation of fbdev set_par op
+ */
+int adf_fbdev_set_par(struct fb_info *info)
+{
+	struct adf_fbdev *fbdev = info->par;
+	struct adf_interface *intf = fbdev->intf;
+	struct fb_videomode vmode;
+	struct drm_mode_modeinfo mode;
+	int ret;
+	u32 format = drm_fourcc_from_fb_var(&info->var);
+
+	fb_var_to_videomode(&vmode, &info->var);
+	adf_modeinfo_from_fb_videomode(&vmode, &mode);
+	ret = adf_interface_set_mode(intf, &mode);
+	if (ret < 0)
+		return ret;
+
+	ret = adf_fbdev_post(fbdev);
+	if (ret < 0)
+		return ret;
+
+	if (format != fbdev->format)
+		adf_fbdev_set_format(fbdev, format);
+
+	return 0;
+}
+EXPORT_SYMBOL(adf_fbdev_set_par);
+
+/**
+ * adf_fbdev_blank - default implementation of fbdev blank op
+ */
+int adf_fbdev_blank(int blank, struct fb_info *info)
+{
+	struct adf_fbdev *fbdev = info->par;
+	struct adf_interface *intf = fbdev->intf;
+	u8 dpms_state;
+
+	switch (blank) {
+	case FB_BLANK_UNBLANK:
+		dpms_state = DRM_MODE_DPMS_ON;
+		break;
+	case FB_BLANK_NORMAL:
+		dpms_state = DRM_MODE_DPMS_STANDBY;
+		break;
+	case FB_BLANK_VSYNC_SUSPEND:
+		dpms_state = DRM_MODE_DPMS_SUSPEND;
+		break;
+	case FB_BLANK_HSYNC_SUSPEND:
+		dpms_state = DRM_MODE_DPMS_STANDBY;
+		break;
+	case FB_BLANK_POWERDOWN:
+		dpms_state = DRM_MODE_DPMS_OFF;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	return adf_interface_blank(intf, dpms_state);
+}
+EXPORT_SYMBOL(adf_fbdev_blank);
+
+/**
+ * adf_fbdev_pan_display - default implementation of fbdev pan_display op
+ */
+int adf_fbdev_pan_display(struct fb_var_screeninfo *var, struct fb_info *info)
+{
+	struct adf_fbdev *fbdev = info->par;
+	return adf_fbdev_post(fbdev);
+}
+EXPORT_SYMBOL(adf_fbdev_pan_display);
+
+/**
+ * adf_fbdev_mmap - default implementation of fbdev mmap op
+ */
+int adf_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma)
+{
+	struct adf_fbdev *fbdev = info->par;
+
+	vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
+	return dma_buf_mmap(fbdev->dma_buf, vma, 0);
+}
+EXPORT_SYMBOL(adf_fbdev_mmap);
+
+/**
+ * adf_fbdev_init - initialize helper to wrap ADF device in fbdev API
+ *
+ * @fbdev: the fbdev helper
+ * @interface: the ADF interface that will display the framebuffer
+ * @eng: the ADF overlay engine that will scan out the framebuffer
+ * @xres_virtual: the virtual width of the framebuffer
+ * @yres_virtual: the virtual height of the framebuffer
+ * @format: the format of the framebuffer
+ * @fbops: the device's fbdev ops
+ * @fmt: formatting for the framebuffer identification string
+ * @...: variable arguments
+ *
+ * @format must be a standard, non-indexed RGB format, i.e.,
+ * adf_format_is_rgb(@format) && @format != @DRM_FORMAT_C8.
+ *
+ * Returns 0 on success or -errno on failure.
+ */
+int adf_fbdev_init(struct adf_fbdev *fbdev, struct adf_interface *interface,
+		struct adf_overlay_engine *eng,
+		u16 xres_virtual, u16 yres_virtual, u32 format,
+		struct fb_ops *fbops, const char *fmt, ...)
+{
+	struct adf_device *parent = adf_interface_parent(interface);
+	struct device *dev = &parent->base.dev;
+	u16 width_mm, height_mm;
+	va_list args;
+	int ret;
+
+	if (!adf_format_is_rgb(format) ||
+			format == DRM_FORMAT_C8) {
+		dev_err(dev, "fbdev helper does not support format %u\n",
+				format);
+		return -EINVAL;
+	}
+
+	memset(fbdev, 0, sizeof(*fbdev));
+	fbdev->intf = interface;
+	fbdev->eng = eng;
+	fbdev->info = framebuffer_alloc(0, dev);
+	if (!fbdev->info) {
+		dev_err(dev, "allocating framebuffer device failed\n");
+		return -ENOMEM;
+	}
+	mutex_init(&fbdev->refcount_lock);
+	fbdev->default_xres_virtual = xres_virtual;
+	fbdev->default_yres_virtual = yres_virtual;
+	fbdev->default_format = format;
+
+	fbdev->info->flags = FBINFO_FLAG_DEFAULT;
+	ret = adf_interface_get_screen_size(interface, &width_mm, &height_mm);
+	if (ret < 0) {
+		width_mm = 0;
+		height_mm = 0;
+	}
+	fbdev->info->var.width = width_mm;
+	fbdev->info->var.height = height_mm;
+	fbdev->info->var.activate = FB_ACTIVATE_VBL;
+	va_start(args, fmt);
+	vsnprintf(fbdev->info->fix.id, sizeof(fbdev->info->fix.id), fmt, args);
+	va_end(args);
+	fbdev->info->fix.type = FB_TYPE_PACKED_PIXELS;
+	fbdev->info->fix.visual = FB_VISUAL_TRUECOLOR;
+	fbdev->info->fix.xpanstep = 1;
+	fbdev->info->fix.ypanstep = 1;
+	INIT_LIST_HEAD(&fbdev->info->modelist);
+	fbdev->info->fbops = fbops;
+	fbdev->info->pseudo_palette = fbdev->pseudo_palette;
+	fbdev->info->par = fbdev;
+
+	ret = register_framebuffer(fbdev->info);
+	if (ret < 0) {
+		dev_err(dev, "registering framebuffer failed: %d\n", ret);
+		return ret;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(adf_fbdev_init);
+
+/**
+ * adf_fbdev_destroy - destroy helper to wrap ADF device in fbdev API
+ *
+ * @fbdev: the fbdev helper
+ */
+void adf_fbdev_destroy(struct adf_fbdev *fbdev)
+{
+	unregister_framebuffer(fbdev->info);
+	BUG_ON(fbdev->refcount);
+	mutex_destroy(&fbdev->refcount_lock);
+	framebuffer_release(fbdev->info);
+}
+EXPORT_SYMBOL(adf_fbdev_destroy);
diff --git a/drivers/video/adf/adf_fops.c b/drivers/video/adf/adf_fops.c
new file mode 100644
index 000000000000..705411bfaebb
--- /dev/null
+++ b/drivers/video/adf/adf_fops.c
@@ -0,0 +1,946 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/bitops.h>
+#include <linux/circ_buf.h>
+#include <linux/fs.h>
+#include <linux/module.h>
+#include <linux/poll.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+
+#include <video/adf_client.h>
+#include <video/adf_format.h>
+
+#include "sw_sync.h"
+#include "sync.h"
+
+#include "adf.h"
+#include "adf_fops.h"
+#include "adf_sysfs.h"
+
+#ifdef CONFIG_COMPAT
+#include "adf_fops32.h"
+#endif
+
+static int adf_obj_set_event(struct adf_obj *obj, struct adf_file *file,
+		struct adf_set_event __user *arg)
+{
+	struct adf_set_event data;
+	bool enabled;
+	unsigned long flags;
+	int err;
+
+	if (copy_from_user(&data, arg, sizeof(data)))
+		return -EFAULT;
+
+	err = adf_obj_check_supports_event(obj, data.type);
+	if (err < 0)
+		return err;
+
+	spin_lock_irqsave(&obj->file_lock, flags);
+	if (data.enabled)
+		enabled = test_and_set_bit(data.type,
+				file->event_subscriptions);
+	else
+		enabled = test_and_clear_bit(data.type,
+				file->event_subscriptions);
+	spin_unlock_irqrestore(&obj->file_lock, flags);
+
+	if (data.enabled == enabled)
+		return -EALREADY;
+
+	if (data.enabled)
+		adf_event_get(obj, data.type);
+	else
+		adf_event_put(obj, data.type);
+
+	return 0;
+}
+
+static int adf_obj_copy_custom_data_to_user(struct adf_obj *obj,
+		void __user *dst, size_t *dst_size)
+{
+	void *custom_data;
+	size_t custom_data_size;
+	int ret;
+
+	if (!obj->ops || !obj->ops->custom_data) {
+		dev_dbg(&obj->dev, "%s: no custom_data op\n", __func__);
+		return 0;
+	}
+
+	custom_data = kzalloc(ADF_MAX_CUSTOM_DATA_SIZE, GFP_KERNEL);
+	if (!custom_data)
+		return -ENOMEM;
+
+	ret = obj->ops->custom_data(obj, custom_data, &custom_data_size);
+	if (ret < 0)
+		goto done;
+
+	if (copy_to_user(dst, custom_data, min(*dst_size, custom_data_size))) {
+		ret = -EFAULT;
+		goto done;
+	}
+	*dst_size = custom_data_size;
+
+done:
+	kfree(custom_data);
+	return ret;
+}
+
+static int adf_eng_get_data(struct adf_overlay_engine *eng,
+		struct adf_overlay_engine_data __user *arg)
+{
+	struct adf_device *dev = adf_overlay_engine_parent(eng);
+	struct adf_overlay_engine_data data;
+	size_t n_supported_formats;
+	u32 *supported_formats = NULL;
+	int ret = 0;
+
+	if (copy_from_user(&data, arg, sizeof(data)))
+		return -EFAULT;
+
+	strlcpy(data.name, eng->base.name, sizeof(data.name));
+
+	if (data.n_supported_formats > ADF_MAX_SUPPORTED_FORMATS)
+		return -EINVAL;
+
+	n_supported_formats = data.n_supported_formats;
+	data.n_supported_formats = eng->ops->n_supported_formats;
+
+	if (n_supported_formats) {
+		supported_formats = kzalloc(n_supported_formats *
+				sizeof(supported_formats[0]), GFP_KERNEL);
+		if (!supported_formats)
+			return -ENOMEM;
+	}
+
+	memcpy(supported_formats, eng->ops->supported_formats,
+			sizeof(u32) * min(n_supported_formats,
+					eng->ops->n_supported_formats));
+
+	mutex_lock(&dev->client_lock);
+	ret = adf_obj_copy_custom_data_to_user(&eng->base, data.custom_data,
+			&data.custom_data_size);
+	mutex_unlock(&dev->client_lock);
+
+	if (ret < 0)
+		goto done;
+
+	if (copy_to_user(arg, &data, sizeof(data))) {
+		ret = -EFAULT;
+		goto done;
+	}
+
+	if (supported_formats && copy_to_user(data.supported_formats,
+			supported_formats,
+			n_supported_formats * sizeof(supported_formats[0])))
+		ret = -EFAULT;
+
+done:
+	kfree(supported_formats);
+	return ret;
+}
+
+static int adf_buffer_import(struct adf_device *dev,
+		struct adf_buffer_config __user *cfg, struct adf_buffer *buf)
+{
+	struct adf_buffer_config user_buf;
+	size_t i;
+	int ret = 0;
+
+	if (copy_from_user(&user_buf, cfg, sizeof(user_buf)))
+		return -EFAULT;
+
+	memset(buf, 0, sizeof(*buf));
+
+	if (user_buf.n_planes > ADF_MAX_PLANES) {
+		dev_err(&dev->base.dev, "invalid plane count %u\n",
+				user_buf.n_planes);
+		return -EINVAL;
+	}
+
+	buf->overlay_engine = idr_find(&dev->overlay_engines,
+			user_buf.overlay_engine);
+	if (!buf->overlay_engine) {
+		dev_err(&dev->base.dev, "invalid overlay engine id %u\n",
+				user_buf.overlay_engine);
+		return -ENOENT;
+	}
+
+	buf->w = user_buf.w;
+	buf->h = user_buf.h;
+	buf->format = user_buf.format;
+	for (i = 0; i < user_buf.n_planes; i++) {
+		buf->dma_bufs[i] = dma_buf_get(user_buf.fd[i]);
+		if (IS_ERR(buf->dma_bufs[i])) {
+			ret = PTR_ERR(buf->dma_bufs[i]);
+			dev_err(&dev->base.dev, "importing dma_buf fd %d failed: %d\n",
+					user_buf.fd[i], ret);
+			buf->dma_bufs[i] = NULL;
+			goto done;
+		}
+		buf->offset[i] = user_buf.offset[i];
+		buf->pitch[i] = user_buf.pitch[i];
+	}
+	buf->n_planes = user_buf.n_planes;
+
+	if (user_buf.acquire_fence >= 0) {
+		buf->acquire_fence = sync_fence_fdget(user_buf.acquire_fence);
+		if (!buf->acquire_fence) {
+			dev_err(&dev->base.dev, "getting fence fd %d failed\n",
+					user_buf.acquire_fence);
+			ret = -EINVAL;
+			goto done;
+		}
+	}
+
+done:
+	if (ret < 0)
+		adf_buffer_cleanup(buf);
+	return ret;
+}
+
+static int adf_device_post_config(struct adf_device *dev,
+		struct adf_post_config __user *arg)
+{
+	struct sync_fence *complete_fence;
+	int complete_fence_fd;
+	struct adf_buffer *bufs = NULL;
+	struct adf_interface **intfs = NULL;
+	struct adf_post_config data;
+	size_t i;
+	void *custom_data = NULL;
+	int ret = 0;
+
+	if (copy_from_user(&data, arg, sizeof(data)))
+		return -EFAULT;
+
+	complete_fence_fd = get_unused_fd_flags(O_CLOEXEC);
+	if (complete_fence_fd < 0)
+		return complete_fence_fd;
+
+	if (data.n_interfaces > ADF_MAX_INTERFACES) {
+		ret = -EINVAL;
+		goto err_get_user;
+	}
+
+	if (data.n_bufs > ADF_MAX_BUFFERS) {
+		ret = -EINVAL;
+		goto err_get_user;
+	}
+
+	if (data.custom_data_size > ADF_MAX_CUSTOM_DATA_SIZE) {
+		ret = -EINVAL;
+		goto err_get_user;
+	}
+
+	if (data.n_interfaces) {
+		intfs = kmalloc(sizeof(intfs[0]) * data.n_interfaces,
+			GFP_KERNEL);
+		if (!intfs) {
+			ret = -ENOMEM;
+			goto err_get_user;
+		}
+	}
+
+	for (i = 0; i < data.n_interfaces; i++) {
+		u32 intf_id;
+		if (get_user(intf_id, &data.interfaces[i])) {
+			ret = -EFAULT;
+			goto err_get_user;
+		}
+
+		intfs[i] = idr_find(&dev->interfaces, intf_id);
+		if (!intfs[i]) {
+			ret = -EINVAL;
+			goto err_get_user;
+		}
+	}
+
+	if (data.n_bufs) {
+		bufs = kzalloc(sizeof(bufs[0]) * data.n_bufs, GFP_KERNEL);
+		if (!bufs) {
+			ret = -ENOMEM;
+			goto err_get_user;
+		}
+	}
+
+	for (i = 0; i < data.n_bufs; i++) {
+		ret = adf_buffer_import(dev, &data.bufs[i], &bufs[i]);
+		if (ret < 0) {
+			memset(&bufs[i], 0, sizeof(bufs[i]));
+			goto err_import;
+		}
+	}
+
+	if (data.custom_data_size) {
+		custom_data = kzalloc(data.custom_data_size, GFP_KERNEL);
+		if (!custom_data) {
+			ret = -ENOMEM;
+			goto err_import;
+		}
+
+		if (copy_from_user(custom_data, data.custom_data,
+				data.custom_data_size)) {
+			ret = -EFAULT;
+			goto err_import;
+		}
+	}
+
+	if (put_user(complete_fence_fd, &arg->complete_fence)) {
+		ret = -EFAULT;
+		goto err_import;
+	}
+
+	complete_fence = adf_device_post_nocopy(dev, intfs, data.n_interfaces,
+			bufs, data.n_bufs, custom_data, data.custom_data_size);
+	if (IS_ERR(complete_fence)) {
+		ret = PTR_ERR(complete_fence);
+		goto err_import;
+	}
+
+	sync_fence_install(complete_fence, complete_fence_fd);
+	return 0;
+
+err_import:
+	for (i = 0; i < data.n_bufs; i++)
+		adf_buffer_cleanup(&bufs[i]);
+
+err_get_user:
+	kfree(custom_data);
+	kfree(bufs);
+	kfree(intfs);
+	put_unused_fd(complete_fence_fd);
+	return ret;
+}
+
+static int adf_intf_simple_post_config(struct adf_interface *intf,
+		struct adf_simple_post_config __user *arg)
+{
+	struct adf_device *dev = intf->base.parent;
+	struct sync_fence *complete_fence;
+	int complete_fence_fd;
+	struct adf_buffer buf;
+	int ret = 0;
+
+	complete_fence_fd = get_unused_fd_flags(O_CLOEXEC);
+	if (complete_fence_fd < 0)
+		return complete_fence_fd;
+
+	ret = adf_buffer_import(dev, &arg->buf, &buf);
+	if (ret < 0)
+		goto err_import;
+
+	if (put_user(complete_fence_fd, &arg->complete_fence)) {
+		ret = -EFAULT;
+		goto err_put_user;
+	}
+
+	complete_fence = adf_interface_simple_post(intf, &buf);
+	if (IS_ERR(complete_fence)) {
+		ret = PTR_ERR(complete_fence);
+		goto err_put_user;
+	}
+
+	sync_fence_install(complete_fence, complete_fence_fd);
+	return 0;
+
+err_put_user:
+	adf_buffer_cleanup(&buf);
+err_import:
+	put_unused_fd(complete_fence_fd);
+	return ret;
+}
+
+static int adf_intf_simple_buffer_alloc(struct adf_interface *intf,
+		struct adf_simple_buffer_alloc __user *arg)
+{
+	struct adf_simple_buffer_alloc data;
+	struct dma_buf *dma_buf;
+	int ret = 0;
+
+	if (copy_from_user(&data, arg, sizeof(data)))
+		return -EFAULT;
+
+	data.fd = get_unused_fd_flags(O_CLOEXEC);
+	if (data.fd < 0)
+		return data.fd;
+
+	ret = adf_interface_simple_buffer_alloc(intf, data.w, data.h,
+			data.format, &dma_buf, &data.offset, &data.pitch);
+	if (ret < 0)
+		goto err_alloc;
+
+	if (copy_to_user(arg, &data, sizeof(*arg))) {
+		ret = -EFAULT;
+		goto err_copy;
+	}
+
+	fd_install(data.fd, dma_buf->file);
+	return 0;
+
+err_copy:
+	dma_buf_put(dma_buf);
+
+err_alloc:
+	put_unused_fd(data.fd);
+	return ret;
+}
+
+static int adf_copy_attachment_list_to_user(
+		struct adf_attachment_config __user *to, size_t n_to,
+		struct adf_attachment *from, size_t n_from)
+{
+	struct adf_attachment_config *temp;
+	size_t n = min(n_to, n_from);
+	size_t i;
+	int ret = 0;
+
+	if (!n)
+		return 0;
+
+	temp = kzalloc(n * sizeof(temp[0]), GFP_KERNEL);
+	if (!temp)
+		return -ENOMEM;
+
+	for (i = 0; i < n; i++) {
+		temp[i].interface = from[i].interface->base.id;
+		temp[i].overlay_engine = from[i].overlay_engine->base.id;
+	}
+
+	if (copy_to_user(to, temp, n * sizeof(to[0]))) {
+		ret = -EFAULT;
+		goto done;
+	}
+
+done:
+	kfree(temp);
+	return ret;
+}
+
+static int adf_device_get_data(struct adf_device *dev,
+		struct adf_device_data __user *arg)
+{
+	struct adf_device_data data;
+	size_t n_attach;
+	struct adf_attachment *attach = NULL;
+	size_t n_allowed_attach;
+	struct adf_attachment *allowed_attach = NULL;
+	int ret = 0;
+
+	if (copy_from_user(&data, arg, sizeof(data)))
+		return -EFAULT;
+
+	if (data.n_attachments > ADF_MAX_ATTACHMENTS ||
+			data.n_allowed_attachments > ADF_MAX_ATTACHMENTS)
+		return -EINVAL;
+
+	strlcpy(data.name, dev->base.name, sizeof(data.name));
+
+	if (data.n_attachments) {
+		attach = kzalloc(data.n_attachments * sizeof(attach[0]),
+				GFP_KERNEL);
+		if (!attach)
+			return -ENOMEM;
+	}
+	n_attach = adf_device_attachments(dev, attach, data.n_attachments);
+
+	if (data.n_allowed_attachments) {
+		allowed_attach = kzalloc(data.n_allowed_attachments *
+				sizeof(allowed_attach[0]), GFP_KERNEL);
+		if (!allowed_attach) {
+			ret = -ENOMEM;
+			goto done;
+		}
+	}
+	n_allowed_attach = adf_device_attachments_allowed(dev, allowed_attach,
+			data.n_allowed_attachments);
+
+	mutex_lock(&dev->client_lock);
+	ret = adf_obj_copy_custom_data_to_user(&dev->base, data.custom_data,
+			&data.custom_data_size);
+	mutex_unlock(&dev->client_lock);
+
+	if (ret < 0)
+		goto done;
+
+	ret = adf_copy_attachment_list_to_user(data.attachments,
+			data.n_attachments, attach, n_attach);
+	if (ret < 0)
+		goto done;
+
+	ret = adf_copy_attachment_list_to_user(data.allowed_attachments,
+			data.n_allowed_attachments, allowed_attach,
+			n_allowed_attach);
+	if (ret < 0)
+		goto done;
+
+	data.n_attachments = n_attach;
+	data.n_allowed_attachments = n_allowed_attach;
+
+	if (copy_to_user(arg, &data, sizeof(data)))
+		ret = -EFAULT;
+
+done:
+	kfree(allowed_attach);
+	kfree(attach);
+	return ret;
+}
+
+static int adf_device_handle_attachment(struct adf_device *dev,
+		struct adf_attachment_config __user *arg, bool attach)
+{
+	struct adf_attachment_config data;
+	struct adf_overlay_engine *eng;
+	struct adf_interface *intf;
+
+	if (copy_from_user(&data, arg, sizeof(data)))
+		return -EFAULT;
+
+	eng = idr_find(&dev->overlay_engines, data.overlay_engine);
+	if (!eng) {
+		dev_err(&dev->base.dev, "invalid overlay engine id %u\n",
+				data.overlay_engine);
+		return -EINVAL;
+	}
+
+	intf = idr_find(&dev->interfaces, data.interface);
+	if (!intf) {
+		dev_err(&dev->base.dev, "invalid interface id %u\n",
+				data.interface);
+		return -EINVAL;
+	}
+
+	if (attach)
+		return adf_device_attach(dev, eng, intf);
+	else
+		return adf_device_detach(dev, eng, intf);
+}
+
+static int adf_intf_set_mode(struct adf_interface *intf,
+		struct drm_mode_modeinfo __user *arg)
+{
+	struct drm_mode_modeinfo mode;
+
+	if (copy_from_user(&mode, arg, sizeof(mode)))
+		return -EFAULT;
+
+	return adf_interface_set_mode(intf, &mode);
+}
+
+static int adf_intf_get_data(struct adf_interface *intf,
+		struct adf_interface_data __user *arg)
+{
+	struct adf_device *dev = adf_interface_parent(intf);
+	struct adf_interface_data data;
+	struct drm_mode_modeinfo *modelist;
+	size_t modelist_size;
+	int err;
+	int ret = 0;
+	unsigned long flags;
+
+	if (copy_from_user(&data, arg, sizeof(data)))
+		return -EFAULT;
+
+	strlcpy(data.name, intf->base.name, sizeof(data.name));
+
+	data.type = intf->type;
+	data.id = intf->idx;
+	data.flags = intf->flags;
+
+	err = adf_interface_get_screen_size(intf, &data.width_mm,
+			&data.height_mm);
+	if (err < 0) {
+		data.width_mm = 0;
+		data.height_mm = 0;
+	}
+
+	modelist = kmalloc(sizeof(modelist[0]) * ADF_MAX_MODES, GFP_KERNEL);
+	if (!modelist)
+		return -ENOMEM;
+
+	mutex_lock(&dev->client_lock);
+	read_lock_irqsave(&intf->hotplug_modelist_lock, flags);
+	data.hotplug_detect = intf->hotplug_detect;
+	modelist_size = min(data.n_available_modes, intf->n_modes) *
+			sizeof(intf->modelist[0]);
+	memcpy(modelist, intf->modelist, modelist_size);
+	data.n_available_modes = intf->n_modes;
+	read_unlock_irqrestore(&intf->hotplug_modelist_lock, flags);
+
+	if (copy_to_user(data.available_modes, modelist, modelist_size)) {
+		ret = -EFAULT;
+		goto done;
+	}
+
+	data.dpms_state = intf->dpms_state;
+	memcpy(&data.current_mode, &intf->current_mode,
+			sizeof(intf->current_mode));
+
+	ret = adf_obj_copy_custom_data_to_user(&intf->base, data.custom_data,
+			&data.custom_data_size);
+done:
+	mutex_unlock(&dev->client_lock);
+	kfree(modelist);
+
+	if (ret < 0)
+		return ret;
+
+	if (copy_to_user(arg, &data, sizeof(data)))
+		ret = -EFAULT;
+
+	return ret;
+}
+
+static inline long adf_obj_custom_ioctl(struct adf_obj *obj, unsigned int cmd,
+		unsigned long arg)
+{
+	if (obj->ops && obj->ops->ioctl)
+		return obj->ops->ioctl(obj, cmd, arg);
+	return -ENOTTY;
+}
+
+static long adf_overlay_engine_ioctl(struct adf_overlay_engine *eng,
+		struct adf_file *file, unsigned int cmd, unsigned long arg)
+{
+	switch (cmd) {
+	case ADF_SET_EVENT:
+		return adf_obj_set_event(&eng->base, file,
+				(struct adf_set_event __user *)arg);
+
+	case ADF_GET_OVERLAY_ENGINE_DATA:
+		return adf_eng_get_data(eng,
+			(struct adf_overlay_engine_data __user *)arg);
+
+	case ADF_BLANK:
+	case ADF_POST_CONFIG:
+	case ADF_SET_MODE:
+	case ADF_GET_DEVICE_DATA:
+	case ADF_GET_INTERFACE_DATA:
+	case ADF_SIMPLE_POST_CONFIG:
+	case ADF_SIMPLE_BUFFER_ALLOC:
+	case ADF_ATTACH:
+	case ADF_DETACH:
+		return -EINVAL;
+
+	default:
+		return adf_obj_custom_ioctl(&eng->base, cmd, arg);
+	}
+}
+
+static long adf_interface_ioctl(struct adf_interface *intf,
+		struct adf_file *file, unsigned int cmd, unsigned long arg)
+{
+	switch (cmd) {
+	case ADF_SET_EVENT:
+		return adf_obj_set_event(&intf->base, file,
+				(struct adf_set_event __user *)arg);
+
+	case ADF_BLANK:
+		return adf_interface_blank(intf, arg);
+
+	case ADF_SET_MODE:
+		return adf_intf_set_mode(intf,
+				(struct drm_mode_modeinfo __user *)arg);
+
+	case ADF_GET_INTERFACE_DATA:
+		return adf_intf_get_data(intf,
+				(struct adf_interface_data __user *)arg);
+
+	case ADF_SIMPLE_POST_CONFIG:
+		return adf_intf_simple_post_config(intf,
+				(struct adf_simple_post_config __user *)arg);
+
+	case ADF_SIMPLE_BUFFER_ALLOC:
+		return adf_intf_simple_buffer_alloc(intf,
+				(struct adf_simple_buffer_alloc __user *)arg);
+
+	case ADF_POST_CONFIG:
+	case ADF_GET_DEVICE_DATA:
+	case ADF_GET_OVERLAY_ENGINE_DATA:
+	case ADF_ATTACH:
+	case ADF_DETACH:
+		return -EINVAL;
+
+	default:
+		return adf_obj_custom_ioctl(&intf->base, cmd, arg);
+	}
+}
+
+static long adf_device_ioctl(struct adf_device *dev, struct adf_file *file,
+		unsigned int cmd, unsigned long arg)
+{
+	switch (cmd) {
+	case ADF_SET_EVENT:
+		return adf_obj_set_event(&dev->base, file,
+				(struct adf_set_event __user *)arg);
+
+	case ADF_POST_CONFIG:
+		return adf_device_post_config(dev,
+				(struct adf_post_config __user *)arg);
+
+	case ADF_GET_DEVICE_DATA:
+		return adf_device_get_data(dev,
+				(struct adf_device_data __user *)arg);
+
+	case ADF_ATTACH:
+		return adf_device_handle_attachment(dev,
+				(struct adf_attachment_config __user *)arg,
+				true);
+
+	case ADF_DETACH:
+		return adf_device_handle_attachment(dev,
+				(struct adf_attachment_config __user *)arg,
+				false);
+
+	case ADF_BLANK:
+	case ADF_SET_MODE:
+	case ADF_GET_INTERFACE_DATA:
+	case ADF_GET_OVERLAY_ENGINE_DATA:
+	case ADF_SIMPLE_POST_CONFIG:
+	case ADF_SIMPLE_BUFFER_ALLOC:
+		return -EINVAL;
+
+	default:
+		return adf_obj_custom_ioctl(&dev->base, cmd, arg);
+	}
+}
+
+static int adf_file_open(struct inode *inode, struct file *file)
+{
+	struct adf_obj *obj;
+	struct adf_file *fpriv = NULL;
+	unsigned long flags;
+	int ret = 0;
+
+	obj = adf_obj_sysfs_find(iminor(inode));
+	if (!obj)
+		return -ENODEV;
+
+	dev_dbg(&obj->dev, "opening %s\n", dev_name(&obj->dev));
+
+	if (!try_module_get(obj->parent->ops->owner)) {
+		dev_err(&obj->dev, "getting owner module failed\n");
+		return -ENODEV;
+	}
+
+	fpriv = kzalloc(sizeof(*fpriv), GFP_KERNEL);
+	if (!fpriv) {
+		ret = -ENOMEM;
+		goto done;
+	}
+
+	INIT_LIST_HEAD(&fpriv->head);
+	fpriv->obj = obj;
+	init_waitqueue_head(&fpriv->event_wait);
+
+	file->private_data = fpriv;
+
+	if (obj->ops && obj->ops->open) {
+		ret = obj->ops->open(obj, inode, file);
+		if (ret < 0)
+			goto done;
+	}
+
+	spin_lock_irqsave(&obj->file_lock, flags);
+	list_add_tail(&fpriv->head, &obj->file_list);
+	spin_unlock_irqrestore(&obj->file_lock, flags);
+
+done:
+	if (ret < 0) {
+		kfree(fpriv);
+		module_put(obj->parent->ops->owner);
+	}
+	return ret;
+}
+
+static int adf_file_release(struct inode *inode, struct file *file)
+{
+	struct adf_file *fpriv = file->private_data;
+	struct adf_obj *obj = fpriv->obj;
+	enum adf_event_type event_type;
+	unsigned long flags;
+
+	if (obj->ops && obj->ops->release)
+		obj->ops->release(obj, inode, file);
+
+	spin_lock_irqsave(&obj->file_lock, flags);
+	list_del(&fpriv->head);
+	spin_unlock_irqrestore(&obj->file_lock, flags);
+
+	for_each_set_bit(event_type, fpriv->event_subscriptions,
+			ADF_EVENT_TYPE_MAX) {
+		adf_event_put(obj, event_type);
+	}
+
+	kfree(fpriv);
+	module_put(obj->parent->ops->owner);
+
+	dev_dbg(&obj->dev, "released %s\n", dev_name(&obj->dev));
+	return 0;
+}
+
+long adf_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
+{
+	struct adf_file *fpriv = file->private_data;
+	struct adf_obj *obj = fpriv->obj;
+	long ret = -EINVAL;
+
+	dev_dbg(&obj->dev, "%s ioctl %u\n", dev_name(&obj->dev), _IOC_NR(cmd));
+
+	switch (obj->type) {
+	case ADF_OBJ_OVERLAY_ENGINE:
+		ret = adf_overlay_engine_ioctl(adf_obj_to_overlay_engine(obj),
+				fpriv, cmd, arg);
+		break;
+
+	case ADF_OBJ_INTERFACE:
+		ret = adf_interface_ioctl(adf_obj_to_interface(obj), fpriv, cmd,
+				arg);
+		break;
+
+	case ADF_OBJ_DEVICE:
+		ret = adf_device_ioctl(adf_obj_to_device(obj), fpriv, cmd, arg);
+		break;
+	}
+
+	return ret;
+}
+
+static inline bool adf_file_event_available(struct adf_file *fpriv)
+{
+	int head = fpriv->event_head;
+	int tail = fpriv->event_tail;
+	return CIRC_CNT(head, tail, sizeof(fpriv->event_buf)) != 0;
+}
+
+void adf_file_queue_event(struct adf_file *fpriv, struct adf_event *event)
+{
+	int head = fpriv->event_head;
+	int tail = fpriv->event_tail;
+	size_t space = CIRC_SPACE(head, tail, sizeof(fpriv->event_buf));
+	size_t space_to_end =
+			CIRC_SPACE_TO_END(head, tail, sizeof(fpriv->event_buf));
+
+	if (space < event->length) {
+		dev_dbg(&fpriv->obj->dev,
+				"insufficient buffer space for event %u\n",
+				event->type);
+		return;
+	}
+
+	if (space_to_end >= event->length) {
+		memcpy(fpriv->event_buf + head, event, event->length);
+	} else {
+		memcpy(fpriv->event_buf + head, event, space_to_end);
+		memcpy(fpriv->event_buf, (u8 *)event + space_to_end,
+				event->length - space_to_end);
+	}
+
+	smp_wmb();
+	fpriv->event_head = (fpriv->event_head + event->length) &
+			(sizeof(fpriv->event_buf) - 1);
+	wake_up_interruptible_all(&fpriv->event_wait);
+}
+
+static ssize_t adf_file_copy_to_user(struct adf_file *fpriv,
+		char __user *buffer, size_t buffer_size)
+{
+	int head, tail;
+	u8 *event_buf;
+	size_t cnt, cnt_to_end, copy_size = 0;
+	ssize_t ret = 0;
+	unsigned long flags;
+
+	event_buf = kmalloc(min(buffer_size, sizeof(fpriv->event_buf)),
+			GFP_KERNEL);
+	if (!event_buf)
+		return -ENOMEM;
+
+	spin_lock_irqsave(&fpriv->obj->file_lock, flags);
+
+	if (!adf_file_event_available(fpriv))
+		goto out;
+
+	head = fpriv->event_head;
+	tail = fpriv->event_tail;
+
+	cnt = CIRC_CNT(head, tail, sizeof(fpriv->event_buf));
+	cnt_to_end = CIRC_CNT_TO_END(head, tail, sizeof(fpriv->event_buf));
+	copy_size = min(buffer_size, cnt);
+
+	if (cnt_to_end >= copy_size) {
+		memcpy(event_buf, fpriv->event_buf + tail, copy_size);
+	} else {
+		memcpy(event_buf, fpriv->event_buf + tail, cnt_to_end);
+		memcpy(event_buf + cnt_to_end, fpriv->event_buf,
+				copy_size - cnt_to_end);
+	}
+
+	fpriv->event_tail = (fpriv->event_tail + copy_size) &
+			(sizeof(fpriv->event_buf) - 1);
+
+out:
+	spin_unlock_irqrestore(&fpriv->obj->file_lock, flags);
+	if (copy_size) {
+		if (copy_to_user(buffer, event_buf, copy_size))
+			ret = -EFAULT;
+		else
+			ret = copy_size;
+	}
+	kfree(event_buf);
+	return ret;
+}
+
+ssize_t adf_file_read(struct file *filp, char __user *buffer,
+		 size_t count, loff_t *offset)
+{
+	struct adf_file *fpriv = filp->private_data;
+	int err;
+
+	err = wait_event_interruptible(fpriv->event_wait,
+			adf_file_event_available(fpriv));
+	if (err < 0)
+		return err;
+
+	return adf_file_copy_to_user(fpriv, buffer, count);
+}
+
+unsigned int adf_file_poll(struct file *filp, struct poll_table_struct *wait)
+{
+	struct adf_file *fpriv = filp->private_data;
+	unsigned int mask = 0;
+
+	poll_wait(filp, &fpriv->event_wait, wait);
+
+	if (adf_file_event_available(fpriv))
+		mask |= POLLIN | POLLRDNORM;
+
+	return mask;
+}
+
+const struct file_operations adf_fops = {
+	.owner = THIS_MODULE,
+	.unlocked_ioctl = adf_file_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl = adf_file_compat_ioctl,
+#endif
+	.open = adf_file_open,
+	.release = adf_file_release,
+	.llseek = default_llseek,
+	.read = adf_file_read,
+	.poll = adf_file_poll,
+};
diff --git a/drivers/video/adf/adf_fops.h b/drivers/video/adf/adf_fops.h
new file mode 100644
index 000000000000..90a3a74796d6
--- /dev/null
+++ b/drivers/video/adf/adf_fops.h
@@ -0,0 +1,37 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __VIDEO_ADF_ADF_FOPS_H
+#define __VIDEO_ADF_ADF_FOPS_H
+
+#include <linux/bitmap.h>
+#include <linux/fs.h>
+
+extern const struct file_operations adf_fops;
+
+struct adf_file {
+	struct list_head head;
+	struct adf_obj *obj;
+
+	DECLARE_BITMAP(event_subscriptions, ADF_EVENT_TYPE_MAX);
+	u8 event_buf[4096];
+	int event_head;
+	int event_tail;
+	wait_queue_head_t event_wait;
+};
+
+void adf_file_queue_event(struct adf_file *file, struct adf_event *event);
+long adf_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
+
+#endif /* __VIDEO_ADF_ADF_FOPS_H */
diff --git a/drivers/video/adf/adf_fops32.c b/drivers/video/adf/adf_fops32.c
new file mode 100644
index 000000000000..d299a8161491
--- /dev/null
+++ b/drivers/video/adf/adf_fops32.c
@@ -0,0 +1,217 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/uaccess.h>
+#include <video/adf.h>
+
+#include "adf_fops.h"
+#include "adf_fops32.h"
+
+long adf_compat_post_config(struct file *file,
+		struct adf_post_config32 __user *arg)
+{
+	struct adf_post_config32 cfg32;
+	struct adf_post_config __user *cfg;
+	int ret;
+
+	if (copy_from_user(&cfg32, arg, sizeof(cfg32)))
+		return -EFAULT;
+
+	cfg = compat_alloc_user_space(sizeof(*cfg));
+	if (!access_ok(VERIFY_WRITE, cfg, sizeof(*cfg)))
+		return -EFAULT;
+
+	if (put_user(cfg32.n_interfaces, &cfg->n_interfaces) ||
+			put_user(compat_ptr(cfg32.interfaces),
+					&cfg->interfaces) ||
+			put_user(cfg32.n_bufs, &cfg->n_bufs) ||
+			put_user(compat_ptr(cfg32.bufs), &cfg->bufs) ||
+			put_user(cfg32.custom_data_size,
+					&cfg->custom_data_size) ||
+			put_user(compat_ptr(cfg32.custom_data),
+					&cfg->custom_data))
+		return -EFAULT;
+
+	ret = adf_file_ioctl(file, ADF_POST_CONFIG, (unsigned long)cfg);
+	if (ret < 0)
+		return ret;
+
+	if (copy_in_user(&arg->complete_fence, &cfg->complete_fence,
+			sizeof(cfg->complete_fence)))
+		return -EFAULT;
+
+	return 0;
+}
+
+long adf_compat_get_device_data(struct file *file,
+		struct adf_device_data32 __user *arg)
+{
+	struct adf_device_data32 data32;
+	struct adf_device_data __user *data;
+	int ret;
+
+	if (copy_from_user(&data32, arg, sizeof(data32)))
+		return -EFAULT;
+
+	data = compat_alloc_user_space(sizeof(*data));
+	if (!access_ok(VERIFY_WRITE, data, sizeof(*data)))
+		return -EFAULT;
+
+	if (put_user(data32.n_attachments, &data->n_attachments) ||
+			put_user(compat_ptr(data32.attachments),
+					&data->attachments) ||
+			put_user(data32.n_allowed_attachments,
+					&data->n_allowed_attachments) ||
+			put_user(compat_ptr(data32.allowed_attachments),
+					&data->allowed_attachments) ||
+			put_user(data32.custom_data_size,
+					&data->custom_data_size) ||
+			put_user(compat_ptr(data32.custom_data),
+					&data->custom_data))
+		return -EFAULT;
+
+	ret = adf_file_ioctl(file, ADF_GET_DEVICE_DATA, (unsigned long)data);
+	if (ret < 0)
+		return ret;
+
+	if (copy_in_user(arg->name, data->name, sizeof(arg->name)) ||
+			copy_in_user(&arg->n_attachments, &data->n_attachments,
+					sizeof(arg->n_attachments)) ||
+			copy_in_user(&arg->n_allowed_attachments,
+					&data->n_allowed_attachments,
+					sizeof(arg->n_allowed_attachments)) ||
+			copy_in_user(&arg->custom_data_size,
+					&data->custom_data_size,
+					sizeof(arg->custom_data_size)))
+		return -EFAULT;
+
+	return 0;
+}
+
+long adf_compat_get_interface_data(struct file *file,
+		struct adf_interface_data32 __user *arg)
+{
+	struct adf_interface_data32 data32;
+	struct adf_interface_data __user *data;
+	int ret;
+
+	if (copy_from_user(&data32, arg, sizeof(data32)))
+		return -EFAULT;
+
+	data = compat_alloc_user_space(sizeof(*data));
+	if (!access_ok(VERIFY_WRITE, data, sizeof(*data)))
+		return -EFAULT;
+
+	if (put_user(data32.n_available_modes, &data->n_available_modes) ||
+			put_user(compat_ptr(data32.available_modes),
+					&data->available_modes) ||
+			put_user(data32.custom_data_size,
+					&data->custom_data_size) ||
+			put_user(compat_ptr(data32.custom_data),
+					&data->custom_data))
+		return -EFAULT;
+
+	ret = adf_file_ioctl(file, ADF_GET_INTERFACE_DATA, (unsigned long)data);
+	if (ret < 0)
+		return ret;
+
+	if (copy_in_user(arg->name, data->name, sizeof(arg->name)) ||
+			copy_in_user(&arg->type, &data->type,
+					sizeof(arg->type)) ||
+			copy_in_user(&arg->id, &data->id, sizeof(arg->id)) ||
+			copy_in_user(&arg->flags, &data->flags,
+					sizeof(arg->flags)) ||
+			copy_in_user(&arg->dpms_state, &data->dpms_state,
+					sizeof(arg->dpms_state)) ||
+			copy_in_user(&arg->hotplug_detect,
+					&data->hotplug_detect,
+					sizeof(arg->hotplug_detect)) ||
+			copy_in_user(&arg->width_mm, &data->width_mm,
+					sizeof(arg->width_mm)) ||
+			copy_in_user(&arg->height_mm, &data->height_mm,
+					sizeof(arg->height_mm)) ||
+			copy_in_user(&arg->current_mode, &data->current_mode,
+					sizeof(arg->current_mode)) ||
+			copy_in_user(&arg->n_available_modes,
+					&data->n_available_modes,
+					sizeof(arg->n_available_modes)) ||
+			copy_in_user(&arg->custom_data_size,
+					&data->custom_data_size,
+					sizeof(arg->custom_data_size)))
+		return -EFAULT;
+
+	return 0;
+}
+
+long adf_compat_get_overlay_engine_data(struct file *file,
+		struct adf_overlay_engine_data32 __user *arg)
+{
+	struct adf_overlay_engine_data32 data32;
+	struct adf_overlay_engine_data __user *data;
+	int ret;
+
+	if (copy_from_user(&data32, arg, sizeof(data32)))
+		return -EFAULT;
+
+	data = compat_alloc_user_space(sizeof(*data));
+	if (!access_ok(VERIFY_WRITE, data, sizeof(*data)))
+		return -EFAULT;
+
+	if (put_user(data32.n_supported_formats, &data->n_supported_formats) ||
+			put_user(compat_ptr(data32.supported_formats),
+					&data->supported_formats) ||
+			put_user(data32.custom_data_size,
+					&data->custom_data_size) ||
+			put_user(compat_ptr(data32.custom_data),
+					&data->custom_data))
+		return -EFAULT;
+
+	ret = adf_file_ioctl(file, ADF_GET_OVERLAY_ENGINE_DATA,
+			(unsigned long)data);
+	if (ret < 0)
+		return ret;
+
+	if (copy_in_user(arg->name, data->name, sizeof(arg->name)) ||
+			copy_in_user(&arg->n_supported_formats,
+					&data->n_supported_formats,
+					sizeof(arg->n_supported_formats)) ||
+			copy_in_user(&arg->custom_data_size,
+					&data->custom_data_size,
+					sizeof(arg->custom_data_size)))
+		return -EFAULT;
+
+	return 0;
+}
+
+long adf_file_compat_ioctl(struct file *file, unsigned int cmd,
+		unsigned long arg)
+{
+	switch (cmd) {
+	case ADF_POST_CONFIG32:
+		return adf_compat_post_config(file, compat_ptr(arg));
+
+	case ADF_GET_DEVICE_DATA32:
+		return adf_compat_get_device_data(file, compat_ptr(arg));
+
+	case ADF_GET_INTERFACE_DATA32:
+		return adf_compat_get_interface_data(file, compat_ptr(arg));
+
+	case ADF_GET_OVERLAY_ENGINE_DATA32:
+		return adf_compat_get_overlay_engine_data(file,
+				compat_ptr(arg));
+
+	default:
+		return adf_file_ioctl(file, cmd, arg);
+	}
+}
diff --git a/drivers/video/adf/adf_fops32.h b/drivers/video/adf/adf_fops32.h
new file mode 100644
index 000000000000..64034ce33a6b
--- /dev/null
+++ b/drivers/video/adf/adf_fops32.h
@@ -0,0 +1,78 @@
+#ifndef __VIDEO_ADF_ADF_FOPS32_H
+#define __VIDEO_ADF_ADF_FOPS32_H
+
+#include <linux/compat.h>
+#include <linux/ioctl.h>
+
+#include <video/adf.h>
+
+#define ADF_POST_CONFIG32 \
+		_IOW(ADF_IOCTL_TYPE, 2, struct adf_post_config32)
+#define ADF_GET_DEVICE_DATA32 \
+		_IOR(ADF_IOCTL_TYPE, 4, struct adf_device_data32)
+#define ADF_GET_INTERFACE_DATA32 \
+		_IOR(ADF_IOCTL_TYPE, 5, struct adf_interface_data32)
+#define ADF_GET_OVERLAY_ENGINE_DATA32 \
+		_IOR(ADF_IOCTL_TYPE, 6, struct adf_overlay_engine_data32)
+
+struct adf_post_config32 {
+	compat_size_t n_interfaces;
+	compat_uptr_t interfaces;
+
+	compat_size_t n_bufs;
+	compat_uptr_t bufs;
+
+	compat_size_t custom_data_size;
+	compat_uptr_t custom_data;
+
+	__s32 complete_fence;
+};
+
+struct adf_device_data32 {
+	char name[ADF_NAME_LEN];
+
+	compat_size_t n_attachments;
+	compat_uptr_t attachments;
+
+	compat_size_t n_allowed_attachments;
+	compat_uptr_t allowed_attachments;
+
+	compat_size_t custom_data_size;
+	compat_uptr_t custom_data;
+};
+
+struct adf_interface_data32 {
+	char name[ADF_NAME_LEN];
+
+	__u8 type;
+	__u32 id;
+	/* e.g. type=ADF_INTF_TYPE_DSI, id=1 => DSI.1 */
+	__u32 flags;
+
+	__u8 dpms_state;
+	__u8 hotplug_detect;
+	__u16 width_mm;
+	__u16 height_mm;
+
+	struct drm_mode_modeinfo current_mode;
+	compat_size_t n_available_modes;
+	compat_uptr_t available_modes;
+
+	compat_size_t custom_data_size;
+	compat_uptr_t custom_data;
+};
+
+struct adf_overlay_engine_data32 {
+	char name[ADF_NAME_LEN];
+
+	compat_size_t n_supported_formats;
+	compat_uptr_t supported_formats;
+
+	compat_size_t custom_data_size;
+	compat_uptr_t custom_data;
+};
+
+long adf_file_compat_ioctl(struct file *file, unsigned int cmd,
+		unsigned long arg);
+
+#endif /* __VIDEO_ADF_ADF_FOPS32_H */
diff --git a/drivers/video/adf/adf_format.c b/drivers/video/adf/adf_format.c
new file mode 100644
index 000000000000..e3f22c7c85d9
--- /dev/null
+++ b/drivers/video/adf/adf_format.c
@@ -0,0 +1,280 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ * modified from drivers/gpu/drm/drm_crtc.c
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/export.h>
+#include <linux/kernel.h>
+#include <drm/drm_fourcc.h>
+#include <video/adf_format.h>
+
+bool adf_format_is_standard(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_C8:
+	case DRM_FORMAT_RGB332:
+	case DRM_FORMAT_BGR233:
+	case DRM_FORMAT_XRGB4444:
+	case DRM_FORMAT_XBGR4444:
+	case DRM_FORMAT_RGBX4444:
+	case DRM_FORMAT_BGRX4444:
+	case DRM_FORMAT_ARGB4444:
+	case DRM_FORMAT_ABGR4444:
+	case DRM_FORMAT_RGBA4444:
+	case DRM_FORMAT_BGRA4444:
+	case DRM_FORMAT_XRGB1555:
+	case DRM_FORMAT_XBGR1555:
+	case DRM_FORMAT_RGBX5551:
+	case DRM_FORMAT_BGRX5551:
+	case DRM_FORMAT_ARGB1555:
+	case DRM_FORMAT_ABGR1555:
+	case DRM_FORMAT_RGBA5551:
+	case DRM_FORMAT_BGRA5551:
+	case DRM_FORMAT_RGB565:
+	case DRM_FORMAT_BGR565:
+	case DRM_FORMAT_RGB888:
+	case DRM_FORMAT_BGR888:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_XBGR8888:
+	case DRM_FORMAT_RGBX8888:
+	case DRM_FORMAT_BGRX8888:
+	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_ABGR8888:
+	case DRM_FORMAT_RGBA8888:
+	case DRM_FORMAT_BGRA8888:
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_XBGR2101010:
+	case DRM_FORMAT_RGBX1010102:
+	case DRM_FORMAT_BGRX1010102:
+	case DRM_FORMAT_ARGB2101010:
+	case DRM_FORMAT_ABGR2101010:
+	case DRM_FORMAT_RGBA1010102:
+	case DRM_FORMAT_BGRA1010102:
+	case DRM_FORMAT_YUYV:
+	case DRM_FORMAT_YVYU:
+	case DRM_FORMAT_UYVY:
+	case DRM_FORMAT_VYUY:
+	case DRM_FORMAT_AYUV:
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+	case DRM_FORMAT_NV16:
+	case DRM_FORMAT_NV61:
+	case DRM_FORMAT_YUV410:
+	case DRM_FORMAT_YVU410:
+	case DRM_FORMAT_YUV411:
+	case DRM_FORMAT_YVU411:
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+	case DRM_FORMAT_YUV422:
+	case DRM_FORMAT_YVU422:
+	case DRM_FORMAT_YUV444:
+	case DRM_FORMAT_YVU444:
+		return true;
+	default:
+		return false;
+	}
+}
+EXPORT_SYMBOL(adf_format_is_standard);
+
+bool adf_format_is_rgb(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_C8:
+	case DRM_FORMAT_RGB332:
+	case DRM_FORMAT_BGR233:
+	case DRM_FORMAT_XRGB1555:
+	case DRM_FORMAT_XBGR1555:
+	case DRM_FORMAT_RGBX5551:
+	case DRM_FORMAT_BGRX5551:
+	case DRM_FORMAT_ARGB1555:
+	case DRM_FORMAT_ABGR1555:
+	case DRM_FORMAT_RGBA5551:
+	case DRM_FORMAT_BGRA5551:
+	case DRM_FORMAT_RGB565:
+	case DRM_FORMAT_BGR565:
+	case DRM_FORMAT_RGB888:
+	case DRM_FORMAT_BGR888:
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_XBGR8888:
+	case DRM_FORMAT_RGBX8888:
+	case DRM_FORMAT_BGRX8888:
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_XBGR2101010:
+	case DRM_FORMAT_RGBX1010102:
+	case DRM_FORMAT_BGRX1010102:
+	case DRM_FORMAT_ARGB2101010:
+	case DRM_FORMAT_ABGR2101010:
+	case DRM_FORMAT_RGBA1010102:
+	case DRM_FORMAT_BGRA1010102:
+	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_ABGR8888:
+	case DRM_FORMAT_RGBA8888:
+	case DRM_FORMAT_BGRA8888:
+		return true;
+
+	default:
+		return false;
+	}
+}
+EXPORT_SYMBOL(adf_format_is_rgb);
+
+u8 adf_format_num_planes(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_YUV410:
+	case DRM_FORMAT_YVU410:
+	case DRM_FORMAT_YUV411:
+	case DRM_FORMAT_YVU411:
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+	case DRM_FORMAT_YUV422:
+	case DRM_FORMAT_YVU422:
+	case DRM_FORMAT_YUV444:
+	case DRM_FORMAT_YVU444:
+		return 3;
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+	case DRM_FORMAT_NV16:
+	case DRM_FORMAT_NV61:
+		return 2;
+	default:
+		return 1;
+	}
+}
+EXPORT_SYMBOL(adf_format_num_planes);
+
+u8 adf_format_bpp(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_C8:
+	case DRM_FORMAT_RGB332:
+	case DRM_FORMAT_BGR233:
+		return 8;
+
+	case DRM_FORMAT_XRGB1555:
+	case DRM_FORMAT_XBGR1555:
+	case DRM_FORMAT_RGBX5551:
+	case DRM_FORMAT_BGRX5551:
+	case DRM_FORMAT_ARGB1555:
+	case DRM_FORMAT_ABGR1555:
+	case DRM_FORMAT_RGBA5551:
+	case DRM_FORMAT_BGRA5551:
+	case DRM_FORMAT_RGB565:
+	case DRM_FORMAT_BGR565:
+		return 16;
+
+	case DRM_FORMAT_RGB888:
+	case DRM_FORMAT_BGR888:
+		return 24;
+
+	case DRM_FORMAT_XRGB8888:
+	case DRM_FORMAT_XBGR8888:
+	case DRM_FORMAT_RGBX8888:
+	case DRM_FORMAT_BGRX8888:
+	case DRM_FORMAT_XRGB2101010:
+	case DRM_FORMAT_XBGR2101010:
+	case DRM_FORMAT_RGBX1010102:
+	case DRM_FORMAT_BGRX1010102:
+	case DRM_FORMAT_ARGB2101010:
+	case DRM_FORMAT_ABGR2101010:
+	case DRM_FORMAT_RGBA1010102:
+	case DRM_FORMAT_BGRA1010102:
+	case DRM_FORMAT_ARGB8888:
+	case DRM_FORMAT_ABGR8888:
+	case DRM_FORMAT_RGBA8888:
+	case DRM_FORMAT_BGRA8888:
+		return 32;
+
+	default:
+		pr_debug("%s: unsupported pixel format %u\n", __func__, format);
+		return 0;
+	}
+}
+EXPORT_SYMBOL(adf_format_bpp);
+
+u8 adf_format_plane_cpp(u32 format, int plane)
+{
+	if (plane >= adf_format_num_planes(format))
+		return 0;
+
+	switch (format) {
+	case DRM_FORMAT_YUYV:
+	case DRM_FORMAT_YVYU:
+	case DRM_FORMAT_UYVY:
+	case DRM_FORMAT_VYUY:
+		return 2;
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+	case DRM_FORMAT_NV16:
+	case DRM_FORMAT_NV61:
+		return plane ? 2 : 1;
+	case DRM_FORMAT_YUV410:
+	case DRM_FORMAT_YVU410:
+	case DRM_FORMAT_YUV411:
+	case DRM_FORMAT_YVU411:
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+	case DRM_FORMAT_YUV422:
+	case DRM_FORMAT_YVU422:
+	case DRM_FORMAT_YUV444:
+	case DRM_FORMAT_YVU444:
+		return 1;
+	default:
+		return adf_format_bpp(format) / 8;
+	}
+}
+EXPORT_SYMBOL(adf_format_plane_cpp);
+
+u8 adf_format_horz_chroma_subsampling(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_YUV411:
+	case DRM_FORMAT_YVU411:
+	case DRM_FORMAT_YUV410:
+	case DRM_FORMAT_YVU410:
+		return 4;
+	case DRM_FORMAT_YUYV:
+	case DRM_FORMAT_YVYU:
+	case DRM_FORMAT_UYVY:
+	case DRM_FORMAT_VYUY:
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+	case DRM_FORMAT_NV16:
+	case DRM_FORMAT_NV61:
+	case DRM_FORMAT_YUV422:
+	case DRM_FORMAT_YVU422:
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+		return 2;
+	default:
+		return 1;
+	}
+}
+EXPORT_SYMBOL(adf_format_horz_chroma_subsampling);
+
+u8 adf_format_vert_chroma_subsampling(u32 format)
+{
+	switch (format) {
+	case DRM_FORMAT_YUV410:
+	case DRM_FORMAT_YVU410:
+		return 4;
+	case DRM_FORMAT_YUV420:
+	case DRM_FORMAT_YVU420:
+	case DRM_FORMAT_NV12:
+	case DRM_FORMAT_NV21:
+		return 2;
+	default:
+		return 1;
+	}
+}
+EXPORT_SYMBOL(adf_format_vert_chroma_subsampling);
diff --git a/drivers/video/adf/adf_memblock.c b/drivers/video/adf/adf_memblock.c
new file mode 100644
index 000000000000..285218a08e8f
--- /dev/null
+++ b/drivers/video/adf/adf_memblock.c
@@ -0,0 +1,166 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/dma-buf.h>
+#include <linux/highmem.h>
+#include <linux/memblock.h>
+#include <linux/slab.h>
+
+struct adf_memblock_pdata {
+	phys_addr_t base;
+};
+
+static struct sg_table *adf_memblock_map(struct dma_buf_attachment *attach,
+		enum dma_data_direction direction)
+{
+	struct adf_memblock_pdata *pdata = attach->dmabuf->priv;
+	unsigned long pfn = PFN_DOWN(pdata->base);
+	struct page *page = pfn_to_page(pfn);
+	struct sg_table *table;
+	int nents, ret;
+
+	table = kzalloc(sizeof(*table), GFP_KERNEL);
+	if (!table)
+		return ERR_PTR(-ENOMEM);
+
+	ret = sg_alloc_table(table, 1, GFP_KERNEL);
+	if (ret < 0)
+		goto err_alloc;
+
+	sg_set_page(table->sgl, page, attach->dmabuf->size, 0);
+
+	nents = dma_map_sg(attach->dev, table->sgl, 1, direction);
+	if (!nents) {
+		ret = -EINVAL;
+		goto err_map;
+	}
+
+	return table;
+
+err_map:
+	sg_free_table(table);
+err_alloc:
+	kfree(table);
+	return ERR_PTR(ret);
+}
+
+static void adf_memblock_unmap(struct dma_buf_attachment *attach,
+		struct sg_table *table, enum dma_data_direction direction)
+{
+	dma_unmap_sg(attach->dev, table->sgl, 1, direction);
+	sg_free_table(table);
+}
+
+static void __init_memblock adf_memblock_release(struct dma_buf *buf)
+{
+	struct adf_memblock_pdata *pdata = buf->priv;
+	int err = memblock_free(pdata->base, buf->size);
+
+	if (err < 0)
+		pr_warn("%s: freeing memblock failed: %d\n", __func__, err);
+	kfree(pdata);
+}
+
+static void *adf_memblock_do_kmap(struct dma_buf *buf, unsigned long pgoffset,
+		bool atomic)
+{
+	struct adf_memblock_pdata *pdata = buf->priv;
+	unsigned long pfn = PFN_DOWN(pdata->base) + pgoffset;
+	struct page *page = pfn_to_page(pfn);
+
+	if (atomic)
+		return kmap_atomic(page);
+	else
+		return kmap(page);
+}
+
+static void *adf_memblock_kmap_atomic(struct dma_buf *buf,
+		unsigned long pgoffset)
+{
+	return adf_memblock_do_kmap(buf, pgoffset, true);
+}
+
+static void adf_memblock_kunmap_atomic(struct dma_buf *buf,
+		unsigned long pgoffset, void *vaddr)
+{
+	kunmap_atomic(vaddr);
+}
+
+static void *adf_memblock_kmap(struct dma_buf *buf, unsigned long pgoffset)
+{
+	return adf_memblock_do_kmap(buf, pgoffset, false);
+}
+
+static void adf_memblock_kunmap(struct dma_buf *buf, unsigned long pgoffset,
+		void *vaddr)
+{
+	kunmap(vaddr);
+}
+
+static int adf_memblock_mmap(struct dma_buf *buf, struct vm_area_struct *vma)
+{
+	struct adf_memblock_pdata *pdata = buf->priv;
+
+	return remap_pfn_range(vma, vma->vm_start, PFN_DOWN(pdata->base),
+			vma->vm_end - vma->vm_start, vma->vm_page_prot);
+}
+
+struct dma_buf_ops adf_memblock_ops = {
+	.map_dma_buf = adf_memblock_map,
+	.unmap_dma_buf = adf_memblock_unmap,
+	.release = adf_memblock_release,
+	.kmap_atomic = adf_memblock_kmap_atomic,
+	.kunmap_atomic = adf_memblock_kunmap_atomic,
+	.kmap = adf_memblock_kmap,
+	.kunmap = adf_memblock_kunmap,
+	.mmap = adf_memblock_mmap,
+};
+
+/**
+ * adf_memblock_export - export a memblock reserved area as a dma-buf
+ *
+ * @base: base physical address
+ * @size: memblock size
+ * @flags: mode flags for the dma-buf's file
+ *
+ * @base and @size must be page-aligned.
+ *
+ * Returns a dma-buf on success or ERR_PTR(-errno) on failure.
+ */
+struct dma_buf *adf_memblock_export(phys_addr_t base, size_t size, int flags)
+{
+	struct adf_memblock_pdata *pdata;
+	struct dma_buf *buf;
+	DEFINE_DMA_BUF_EXPORT_INFO(exp_info);
+
+	if (PAGE_ALIGN(base) != base || PAGE_ALIGN(size) != size)
+		return ERR_PTR(-EINVAL);
+
+	pdata = kzalloc(sizeof(*pdata), GFP_KERNEL);
+	if (!pdata)
+		return ERR_PTR(-ENOMEM);
+
+	pdata->base = base;
+	exp_info.ops = &adf_memblock_ops;
+	exp_info.size = size;
+	exp_info.flags = flags;
+	exp_info.priv = pdata;
+
+	buf = dma_buf_export(&exp_info);
+	if (IS_ERR(buf))
+		kfree(pdata);
+
+	return buf;
+}
+EXPORT_SYMBOL(adf_memblock_export);
diff --git a/drivers/video/adf/adf_sysfs.c b/drivers/video/adf/adf_sysfs.c
new file mode 100644
index 000000000000..8c659c71ffa8
--- /dev/null
+++ b/drivers/video/adf/adf_sysfs.c
@@ -0,0 +1,296 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <video/adf_client.h>
+
+#include "adf.h"
+#include "adf_fops.h"
+#include "adf_sysfs.h"
+
+static struct class *adf_class;
+static int adf_major;
+static DEFINE_IDR(adf_minors);
+
+#define dev_to_adf_interface(p) \
+	adf_obj_to_interface(container_of(p, struct adf_obj, dev))
+
+static ssize_t dpms_state_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct adf_interface *intf = dev_to_adf_interface(dev);
+	return scnprintf(buf, PAGE_SIZE, "%u\n",
+			adf_interface_dpms_state(intf));
+}
+
+static ssize_t dpms_state_store(struct device *dev,
+		struct device_attribute *attr, const char *buf, size_t count)
+{
+	struct adf_interface *intf = dev_to_adf_interface(dev);
+	u8 dpms_state;
+	int err;
+
+	err = kstrtou8(buf, 0, &dpms_state);
+	if (err < 0)
+		return err;
+
+	err = adf_interface_blank(intf, dpms_state);
+	if (err < 0)
+		return err;
+
+	return count;
+}
+
+static ssize_t current_mode_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct adf_interface *intf = dev_to_adf_interface(dev);
+	struct drm_mode_modeinfo mode;
+
+	adf_interface_current_mode(intf, &mode);
+
+	if (mode.name[0]) {
+		return scnprintf(buf, PAGE_SIZE, "%s\n", mode.name);
+	} else {
+		bool interlaced = !!(mode.flags & DRM_MODE_FLAG_INTERLACE);
+		return scnprintf(buf, PAGE_SIZE, "%ux%u%s\n", mode.hdisplay,
+				mode.vdisplay, interlaced ? "i" : "");
+	}
+}
+
+static ssize_t type_show(struct device *dev, struct device_attribute *attr,
+			char *buf)
+{
+	struct adf_interface *intf = dev_to_adf_interface(dev);
+	return scnprintf(buf, PAGE_SIZE, "%s\n",
+			adf_interface_type_str(intf));
+}
+
+static ssize_t vsync_timestamp_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct adf_interface *intf = dev_to_adf_interface(dev);
+	ktime_t timestamp;
+	unsigned long flags;
+
+	read_lock_irqsave(&intf->vsync_lock, flags);
+	memcpy(&timestamp, &intf->vsync_timestamp, sizeof(timestamp));
+	read_unlock_irqrestore(&intf->vsync_lock, flags);
+
+	return scnprintf(buf, PAGE_SIZE, "%llu\n", ktime_to_ns(timestamp));
+}
+
+static ssize_t hotplug_detect_show(struct device *dev,
+		struct device_attribute *attr, char *buf)
+{
+	struct adf_interface *intf = dev_to_adf_interface(dev);
+	return scnprintf(buf, PAGE_SIZE, "%u\n", intf->hotplug_detect);
+}
+
+static struct device_attribute adf_interface_attrs[] = {
+	__ATTR(dpms_state, S_IRUGO|S_IWUSR, dpms_state_show, dpms_state_store),
+	__ATTR_RO(current_mode),
+	__ATTR_RO(hotplug_detect),
+	__ATTR_RO(type),
+	__ATTR_RO(vsync_timestamp),
+};
+
+int adf_obj_sysfs_init(struct adf_obj *obj, struct device *parent)
+{
+	int ret = idr_alloc(&adf_minors, obj, 0, 0, GFP_KERNEL);
+	if (ret < 0) {
+		pr_err("%s: allocating adf minor failed: %d\n", __func__,
+				ret);
+		return ret;
+	}
+
+	obj->minor = ret;
+	obj->dev.parent = parent;
+	obj->dev.class = adf_class;
+	obj->dev.devt = MKDEV(adf_major, obj->minor);
+
+	ret = device_register(&obj->dev);
+	if (ret < 0) {
+		pr_err("%s: registering adf object failed: %d\n", __func__,
+				ret);
+		goto err_device_register;
+	}
+
+	return 0;
+
+err_device_register:
+	idr_remove(&adf_minors, obj->minor);
+	return ret;
+}
+
+static char *adf_device_devnode(struct device *dev, umode_t *mode,
+		kuid_t *uid, kgid_t *gid)
+{
+	struct adf_obj *obj = container_of(dev, struct adf_obj, dev);
+	return kasprintf(GFP_KERNEL, "adf%d", obj->id);
+}
+
+static char *adf_interface_devnode(struct device *dev, umode_t *mode,
+		kuid_t *uid, kgid_t *gid)
+{
+	struct adf_obj *obj = container_of(dev, struct adf_obj, dev);
+	struct adf_interface *intf = adf_obj_to_interface(obj);
+	struct adf_device *parent = adf_interface_parent(intf);
+	return kasprintf(GFP_KERNEL, "adf-interface%d.%d",
+			parent->base.id, intf->base.id);
+}
+
+static char *adf_overlay_engine_devnode(struct device *dev, umode_t *mode,
+		kuid_t *uid, kgid_t *gid)
+{
+	struct adf_obj *obj = container_of(dev, struct adf_obj, dev);
+	struct adf_overlay_engine *eng = adf_obj_to_overlay_engine(obj);
+	struct adf_device *parent = adf_overlay_engine_parent(eng);
+	return kasprintf(GFP_KERNEL, "adf-overlay-engine%d.%d",
+			parent->base.id, eng->base.id);
+}
+
+static void adf_noop_release(struct device *dev)
+{
+}
+
+static struct device_type adf_device_type = {
+	.name = "adf_device",
+	.devnode = adf_device_devnode,
+	.release = adf_noop_release,
+};
+
+static struct device_type adf_interface_type = {
+	.name = "adf_interface",
+	.devnode = adf_interface_devnode,
+	.release = adf_noop_release,
+};
+
+static struct device_type adf_overlay_engine_type = {
+	.name = "adf_overlay_engine",
+	.devnode = adf_overlay_engine_devnode,
+	.release = adf_noop_release,
+};
+
+int adf_device_sysfs_init(struct adf_device *dev)
+{
+	dev->base.dev.type = &adf_device_type;
+	dev_set_name(&dev->base.dev, "%s", dev->base.name);
+	return adf_obj_sysfs_init(&dev->base, dev->dev);
+}
+
+int adf_interface_sysfs_init(struct adf_interface *intf)
+{
+	struct adf_device *parent = adf_interface_parent(intf);
+	size_t i, j;
+	int ret;
+
+	intf->base.dev.type = &adf_interface_type;
+	dev_set_name(&intf->base.dev, "%s-interface%d", parent->base.name,
+			intf->base.id);
+
+	ret = adf_obj_sysfs_init(&intf->base, &parent->base.dev);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < ARRAY_SIZE(adf_interface_attrs); i++) {
+		ret = device_create_file(&intf->base.dev,
+				&adf_interface_attrs[i]);
+		if (ret < 0) {
+			dev_err(&intf->base.dev, "creating sysfs attribute %s failed: %d\n",
+					adf_interface_attrs[i].attr.name, ret);
+			goto err;
+		}
+	}
+
+	return 0;
+
+err:
+	for (j = 0; j < i; j++)
+		device_remove_file(&intf->base.dev, &adf_interface_attrs[j]);
+	return ret;
+}
+
+int adf_overlay_engine_sysfs_init(struct adf_overlay_engine *eng)
+{
+	struct adf_device *parent = adf_overlay_engine_parent(eng);
+
+	eng->base.dev.type = &adf_overlay_engine_type;
+	dev_set_name(&eng->base.dev, "%s-overlay-engine%d", parent->base.name,
+			eng->base.id);
+
+	return adf_obj_sysfs_init(&eng->base, &parent->base.dev);
+}
+
+struct adf_obj *adf_obj_sysfs_find(int minor)
+{
+	return idr_find(&adf_minors, minor);
+}
+
+void adf_obj_sysfs_destroy(struct adf_obj *obj)
+{
+	idr_remove(&adf_minors, obj->minor);
+	device_unregister(&obj->dev);
+}
+
+void adf_device_sysfs_destroy(struct adf_device *dev)
+{
+	adf_obj_sysfs_destroy(&dev->base);
+}
+
+void adf_interface_sysfs_destroy(struct adf_interface *intf)
+{
+	size_t i;
+
+	for (i = 0; i < ARRAY_SIZE(adf_interface_attrs); i++)
+		device_remove_file(&intf->base.dev, &adf_interface_attrs[i]);
+	adf_obj_sysfs_destroy(&intf->base);
+}
+
+void adf_overlay_engine_sysfs_destroy(struct adf_overlay_engine *eng)
+{
+	adf_obj_sysfs_destroy(&eng->base);
+}
+
+int adf_sysfs_init(void)
+{
+	struct class *class;
+	int ret;
+
+	class = class_create(THIS_MODULE, "adf");
+	if (IS_ERR(class)) {
+		ret = PTR_ERR(class);
+		pr_err("%s: creating class failed: %d\n", __func__, ret);
+		return ret;
+	}
+
+	ret = register_chrdev(0, "adf", &adf_fops);
+	if (ret < 0) {
+		pr_err("%s: registering device failed: %d\n", __func__, ret);
+		goto err_chrdev;
+	}
+
+	adf_class = class;
+	adf_major = ret;
+	return 0;
+
+err_chrdev:
+	class_destroy(adf_class);
+	return ret;
+}
+
+void adf_sysfs_destroy(void)
+{
+	idr_destroy(&adf_minors);
+	class_destroy(adf_class);
+}
diff --git a/drivers/video/adf/adf_sysfs.h b/drivers/video/adf/adf_sysfs.h
new file mode 100644
index 000000000000..0613ac364f8d
--- /dev/null
+++ b/drivers/video/adf/adf_sysfs.h
@@ -0,0 +1,33 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __VIDEO_ADF_ADF_SYSFS_H
+#define __VIDEO_ADF_ADF_SYSFS_H
+
+struct adf_device;
+struct adf_interface;
+struct adf_overlay_engine;
+
+int adf_device_sysfs_init(struct adf_device *dev);
+void adf_device_sysfs_destroy(struct adf_device *dev);
+int adf_interface_sysfs_init(struct adf_interface *intf);
+void adf_interface_sysfs_destroy(struct adf_interface *intf);
+int adf_overlay_engine_sysfs_init(struct adf_overlay_engine *eng);
+void adf_overlay_engine_sysfs_destroy(struct adf_overlay_engine *eng);
+struct adf_obj *adf_obj_sysfs_find(int minor);
+
+int adf_sysfs_init(void);
+void adf_sysfs_destroy(void);
+
+#endif /* __VIDEO_ADF_ADF_SYSFS_H */
diff --git a/drivers/video/adf/adf_trace.h b/drivers/video/adf/adf_trace.h
new file mode 100644
index 000000000000..3cb2a84d728c
--- /dev/null
+++ b/drivers/video/adf/adf_trace.h
@@ -0,0 +1,93 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM adf
+
+#if !defined(__VIDEO_ADF_ADF_TRACE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define __VIDEO_ADF_ADF_TRACE_H
+
+#include <linux/tracepoint.h>
+#include <video/adf.h>
+
+TRACE_EVENT(adf_event,
+	TP_PROTO(struct adf_obj *obj, enum adf_event_type type),
+	TP_ARGS(obj, type),
+
+	TP_STRUCT__entry(
+		__string(name, obj->name)
+		__field(enum adf_event_type, type)
+		__array(char, type_str, 32)
+	),
+	TP_fast_assign(
+		__assign_str(name, obj->name);
+		__entry->type = type;
+		strlcpy(__entry->type_str, adf_event_type_str(obj, type),
+				sizeof(__entry->type_str));
+	),
+	TP_printk("obj=%s type=%u (%s)",
+			__get_str(name),
+			__entry->type,
+			__entry->type_str)
+);
+
+TRACE_EVENT(adf_event_enable,
+	TP_PROTO(struct adf_obj *obj, enum adf_event_type type),
+	TP_ARGS(obj, type),
+
+	TP_STRUCT__entry(
+		__string(name, obj->name)
+		__field(enum adf_event_type, type)
+		__array(char, type_str, 32)
+	),
+	TP_fast_assign(
+		__assign_str(name, obj->name);
+		__entry->type = type;
+		strlcpy(__entry->type_str, adf_event_type_str(obj, type),
+				sizeof(__entry->type_str));
+	),
+	TP_printk("obj=%s type=%u (%s)",
+			__get_str(name),
+			__entry->type,
+			__entry->type_str)
+);
+
+TRACE_EVENT(adf_event_disable,
+	TP_PROTO(struct adf_obj *obj, enum adf_event_type type),
+	TP_ARGS(obj, type),
+
+	TP_STRUCT__entry(
+		__string(name, obj->name)
+		__field(enum adf_event_type, type)
+		__array(char, type_str, 32)
+	),
+	TP_fast_assign(
+		__assign_str(name, obj->name);
+		__entry->type = type;
+		strlcpy(__entry->type_str, adf_event_type_str(obj, type),
+				sizeof(__entry->type_str));
+	),
+	TP_printk("obj=%s type=%u (%s)",
+			__get_str(name),
+			__entry->type,
+			__entry->type_str)
+);
+
+#endif /* __VIDEO_ADF_ADF_TRACE_H */
+
+#undef TRACE_INCLUDE_PATH
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_PATH .
+#define TRACE_INCLUDE_FILE adf_trace
+#include <trace/define_trace.h>
diff --git a/drivers/video/fbdev/goldfishfb.c b/drivers/video/fbdev/goldfishfb.c
index 7f6c9e6cfc6c..1e56b50e4082 100644
--- a/drivers/video/fbdev/goldfishfb.c
+++ b/drivers/video/fbdev/goldfishfb.c
@@ -26,6 +26,7 @@
 #include <linux/interrupt.h>
 #include <linux/ioport.h>
 #include <linux/platform_device.h>
+#include <linux/acpi.h>
 
 enum {
 	FB_GET_WIDTH        = 0x00,
@@ -234,7 +235,7 @@ static int goldfish_fb_probe(struct platform_device *pdev)
 	fb->fb.var.activate	= FB_ACTIVATE_NOW;
 	fb->fb.var.height	= readl(fb->reg_base + FB_GET_PHYS_HEIGHT);
 	fb->fb.var.width	= readl(fb->reg_base + FB_GET_PHYS_WIDTH);
-	fb->fb.var.pixclock	= 10000;
+	fb->fb.var.pixclock	= 0;
 
 	fb->fb.var.red.offset = 11;
 	fb->fb.var.red.length = 5;
@@ -304,12 +305,25 @@ static int goldfish_fb_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id goldfish_fb_of_match[] = {
+	{ .compatible = "google,goldfish-fb", },
+	{},
+};
+MODULE_DEVICE_TABLE(of, goldfish_fb_of_match);
+
+static const struct acpi_device_id goldfish_fb_acpi_match[] = {
+	{ "GFSH0004", 0 },
+	{ },
+};
+MODULE_DEVICE_TABLE(acpi, goldfish_fb_acpi_match);
 
 static struct platform_driver goldfish_fb_driver = {
 	.probe		= goldfish_fb_probe,
 	.remove		= goldfish_fb_remove,
 	.driver = {
-		.name = "goldfish_fb"
+		.name = "goldfish_fb",
+		.of_match_table = goldfish_fb_of_match,
+		.acpi_match_table = ACPI_PTR(goldfish_fb_acpi_match),
 	}
 };
 
diff --git a/drivers/w1/masters/ds2482.c b/drivers/w1/masters/ds2482.c
index b05e8fefbabd..03b2f8f41607 100644
--- a/drivers/w1/masters/ds2482.c
+++ b/drivers/w1/masters/ds2482.c
@@ -18,6 +18,8 @@
 #include <linux/slab.h>
 #include <linux/i2c.h>
 #include <linux/delay.h>
+#include <linux/gpio.h>
+#include <linux/platform_data/ds2482.h>
 #include <asm/delay.h>
 
 #include "../w1.h"
@@ -84,7 +86,8 @@ static const u8 ds2482_chan_rd[8] =
 static int ds2482_probe(struct i2c_client *client,
 			const struct i2c_device_id *id);
 static int ds2482_remove(struct i2c_client *client);
-
+static int ds2482_suspend(struct device *dev);
+static int ds2482_resume(struct device *dev);
 
 /**
  * Driver data (common to all clients)
@@ -95,9 +98,15 @@ static const struct i2c_device_id ds2482_id[] = {
 };
 MODULE_DEVICE_TABLE(i2c, ds2482_id);
 
+static const struct dev_pm_ops ds2482_pm_ops = {
+	.suspend = ds2482_suspend,
+	.resume = ds2482_resume,
+};
+
 static struct i2c_driver ds2482_driver = {
 	.driver = {
 		.name	= "ds2482",
+		.pm = &ds2482_pm_ops,
 	},
 	.probe		= ds2482_probe,
 	.remove		= ds2482_remove,
@@ -119,6 +128,7 @@ struct ds2482_w1_chan {
 struct ds2482_data {
 	struct i2c_client	*client;
 	struct mutex		access_lock;
+	int			slpz_gpio;
 
 	/* 1-wire interface(s) */
 	int			w1_count;	/* 1 or 8 */
@@ -444,11 +454,31 @@ static u8 ds2482_w1_set_pullup(void *data, int delay)
 	return retval;
 }
 
+static int ds2482_suspend(struct device *dev)
+{
+	struct i2c_client *client = to_i2c_client(dev);
+	struct ds2482_data *data = i2c_get_clientdata(client);
+
+	if (data->slpz_gpio >= 0)
+		gpio_set_value(data->slpz_gpio, 0);
+	return 0;
+}
+
+static int ds2482_resume(struct device *dev)
+{
+	struct i2c_client *client = to_i2c_client(dev);
+	struct ds2482_data *data = i2c_get_clientdata(client);
+
+	if (data->slpz_gpio >= 0)
+		gpio_set_value(data->slpz_gpio, 1);
+	return 0;
+}
 
 static int ds2482_probe(struct i2c_client *client,
 			const struct i2c_device_id *id)
 {
 	struct ds2482_data *data;
+	struct ds2482_platform_data *pdata;
 	int err = -ENODEV;
 	int temp1;
 	int idx;
@@ -515,6 +545,16 @@ static int ds2482_probe(struct i2c_client *client,
 		}
 	}
 
+	pdata = client->dev.platform_data;
+	data->slpz_gpio = pdata ? pdata->slpz_gpio : -1;
+
+	if (data->slpz_gpio >= 0) {
+		err = gpio_request_one(data->slpz_gpio, GPIOF_OUT_INIT_HIGH,
+				       "ds2482.slpz");
+		if (err < 0)
+			goto exit_w1_remove;
+	}
+
 	return 0;
 
 exit_w1_remove:
@@ -539,6 +579,11 @@ static int ds2482_remove(struct i2c_client *client)
 			w1_remove_master_device(&data->w1_ch[idx].w1_bm);
 	}
 
+	if (data->slpz_gpio >= 0) {
+		gpio_set_value(data->slpz_gpio, 0);
+		gpio_free(data->slpz_gpio);
+	}
+
 	/* Free the memory */
 	kfree(data);
 	return 0;
diff --git a/fs/Kconfig b/fs/Kconfig
index 6ce72d8d1ee1..80af05163579 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -73,6 +73,8 @@ config FILE_LOCKING
           for filesystems like NFS and for the flock() system
           call. Disabling this option saves about 11k.
 
+source "fs/crypto/Kconfig"
+
 source "fs/notify/Kconfig"
 
 source "fs/quota/Kconfig"
@@ -199,6 +201,7 @@ if MISC_FILESYSTEMS
 source "fs/adfs/Kconfig"
 source "fs/affs/Kconfig"
 source "fs/ecryptfs/Kconfig"
+source "fs/sdcardfs/Kconfig"
 source "fs/hfs/Kconfig"
 source "fs/hfsplus/Kconfig"
 source "fs/befs/Kconfig"
diff --git a/fs/Makefile b/fs/Makefile
index 79f522575cba..4644db462ba9 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -3,7 +3,7 @@
 #
 # 14 Sep 2000, Christoph Hellwig <hch@infradead.org>
 # Rewritten to use lists instead of if-statements.
-# 
+#
 
 obj-y :=	open.o read_write.o file_table.o super.o \
 		char_dev.o stat.o exec.o pipe.o namei.o fcntl.o \
@@ -30,6 +30,7 @@ obj-$(CONFIG_EVENTFD)		+= eventfd.o
 obj-$(CONFIG_USERFAULTFD)	+= userfaultfd.o
 obj-$(CONFIG_AIO)               += aio.o
 obj-$(CONFIG_FS_DAX)		+= dax.o
+obj-$(CONFIG_FS_ENCRYPTION)	+= crypto/
 obj-$(CONFIG_FILE_LOCKING)      += locks.o
 obj-$(CONFIG_COMPAT)		+= compat.o compat_ioctl.o
 obj-$(CONFIG_BINFMT_AOUT)	+= binfmt_aout.o
@@ -41,7 +42,7 @@ obj-$(CONFIG_COMPAT_BINFMT_ELF)	+= compat_binfmt_elf.o
 obj-$(CONFIG_BINFMT_ELF_FDPIC)	+= binfmt_elf_fdpic.o
 obj-$(CONFIG_BINFMT_FLAT)	+= binfmt_flat.o
 
-obj-$(CONFIG_FS_MBCACHE)	+= mbcache.o
+obj-$(CONFIG_FS_MBCACHE)	+= mbcache.o mbcache2.o
 obj-$(CONFIG_FS_POSIX_ACL)	+= posix_acl.o
 obj-$(CONFIG_NFS_COMMON)	+= nfs_common/
 obj-$(CONFIG_COREDUMP)		+= coredump.o
@@ -59,7 +60,7 @@ obj-y				+= devpts/
 
 obj-$(CONFIG_PROFILING)		+= dcookies.o
 obj-$(CONFIG_DLM)		+= dlm/
- 
+
 # Do not add any filesystems before this line
 obj-$(CONFIG_FSCACHE)		+= fscache/
 obj-$(CONFIG_REISERFS_FS)	+= reiserfs/
@@ -81,6 +82,7 @@ obj-$(CONFIG_ISO9660_FS)	+= isofs/
 obj-$(CONFIG_HFSPLUS_FS)	+= hfsplus/ # Before hfs to find wrapped HFS+
 obj-$(CONFIG_HFS_FS)		+= hfs/
 obj-$(CONFIG_ECRYPT_FS)		+= ecryptfs/
+obj-$(CONFIG_SDCARD_FS)		+= sdcardfs/
 obj-$(CONFIG_VXFS_FS)		+= freevxfs/
 obj-$(CONFIG_NFS_FS)		+= nfs/
 obj-$(CONFIG_EXPORTFS)		+= exportfs/
diff --git a/fs/attr.c b/fs/attr.c
index d62f674a605f..c86b37c38fb7 100644
--- a/fs/attr.c
+++ b/fs/attr.c
@@ -187,7 +187,7 @@ EXPORT_SYMBOL(setattr_copy);
  * the file open for write, as there can be no conflicting delegation in
  * that case.
  */
-int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **delegated_inode)
+int notify_change2(struct vfsmount *mnt, struct dentry * dentry, struct iattr * attr, struct inode **delegated_inode)
 {
 	struct inode *inode = dentry->d_inode;
 	umode_t mode = inode->i_mode;
@@ -211,7 +211,7 @@ int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **de
 			return -EPERM;
 
 		if (!inode_owner_or_capable(inode)) {
-			error = inode_permission(inode, MAY_WRITE);
+			error = inode_permission2(mnt, inode, MAY_WRITE);
 			if (error)
 				return error;
 		}
@@ -277,7 +277,9 @@ int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **de
 	if (error)
 		return error;
 
-	if (inode->i_op->setattr)
+	if (mnt && inode->i_op->setattr2)
+		error = inode->i_op->setattr2(mnt, dentry, attr);
+	else if (inode->i_op->setattr)
 		error = inode->i_op->setattr(dentry, attr);
 	else
 		error = simple_setattr(dentry, attr);
@@ -290,4 +292,10 @@ int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **de
 
 	return error;
 }
+EXPORT_SYMBOL(notify_change2);
+
+int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **delegated_inode)
+{
+	return notify_change2(NULL, dentry, attr, delegated_inode);
+}
 EXPORT_SYMBOL(notify_change);
diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
index f44e93d2650d..8a0243efd359 100644
--- a/fs/binfmt_elf.c
+++ b/fs/binfmt_elf.c
@@ -651,7 +651,7 @@ static unsigned long randomize_stack_top(unsigned long stack_top)
 
 	if ((current->flags & PF_RANDOMIZE) &&
 		!(current->personality & ADDR_NO_RANDOMIZE)) {
-		random_variable = (unsigned long) get_random_int();
+		random_variable = get_random_long();
 		random_variable &= STACK_RND_MASK;
 		random_variable <<= PAGE_SHIFT;
 	}
diff --git a/fs/btrfs/hash.c b/fs/btrfs/hash.c
index aae520b2aee5..6fc94fcba072 100644
--- a/fs/btrfs/hash.c
+++ b/fs/btrfs/hash.c
@@ -33,6 +33,7 @@ u32 btrfs_crc32c(u32 crc, const void *address, unsigned int length)
 {
 	SHASH_DESC_ON_STACK(shash, tfm);
 	u32 *ctx = (u32 *)shash_desc_ctx(shash);
+	u32 retval;
 	int err;
 
 	shash->tfm = tfm;
@@ -42,5 +43,7 @@ u32 btrfs_crc32c(u32 crc, const void *address, unsigned int length)
 	err = crypto_shash_update(shash, address, length);
 	BUG_ON(err);
 
-	return *ctx;
+	retval = *ctx;
+	barrier_data(ctx);
+	return retval;
 }
diff --git a/fs/coredump.c b/fs/coredump.c
index a8852293038a..2ce5ef429c48 100644
--- a/fs/coredump.c
+++ b/fs/coredump.c
@@ -720,7 +720,7 @@ void do_coredump(const siginfo_t *siginfo)
 			goto close_fail;
 		if (!(cprm.file->f_mode & FMODE_CAN_WRITE))
 			goto close_fail;
-		if (do_truncate(cprm.file->f_path.dentry, 0, 0, cprm.file))
+		if (do_truncate2(cprm.file->f_path.mnt, cprm.file->f_path.dentry, 0, 0, cprm.file))
 			goto close_fail;
 	}
 
diff --git a/fs/crypto/Kconfig b/fs/crypto/Kconfig
new file mode 100644
index 000000000000..92348faf9865
--- /dev/null
+++ b/fs/crypto/Kconfig
@@ -0,0 +1,18 @@
+config FS_ENCRYPTION
+	tristate "FS Encryption (Per-file encryption)"
+	depends on BLOCK
+	select CRYPTO
+	select CRYPTO_AES
+	select CRYPTO_CBC
+	select CRYPTO_ECB
+	select CRYPTO_XTS
+	select CRYPTO_CTS
+	select CRYPTO_CTR
+	select CRYPTO_SHA256
+	select KEYS
+	select ENCRYPTED_KEYS
+	help
+	  Enable encryption of files and directories.  This
+	  feature is similar to ecryptfs, but it is more memory
+	  efficient since it avoids caching the encrypted and
+	  decrypted pages in the page cache.
diff --git a/fs/crypto/Makefile b/fs/crypto/Makefile
new file mode 100644
index 000000000000..9f6607f17b53
--- /dev/null
+++ b/fs/crypto/Makefile
@@ -0,0 +1,4 @@
+obj-$(CONFIG_FS_ENCRYPTION)	+= fscrypto.o
+
+fscrypto-y := crypto.o fname.o policy.o keyinfo.o
+fscrypto-$(CONFIG_BLOCK) += bio.o
diff --git a/fs/crypto/bio.c b/fs/crypto/bio.c
new file mode 100644
index 000000000000..a91ed46fe503
--- /dev/null
+++ b/fs/crypto/bio.c
@@ -0,0 +1,143 @@
+/*
+ * This contains encryption functions for per-file encryption.
+ *
+ * Copyright (C) 2015, Google, Inc.
+ * Copyright (C) 2015, Motorola Mobility
+ *
+ * Written by Michael Halcrow, 2014.
+ *
+ * Filename encryption additions
+ *	Uday Savagaonkar, 2014
+ * Encryption policy handling additions
+ *	Ildar Muslukhov, 2014
+ * Add fscrypt_pullback_bio_page()
+ *	Jaegeuk Kim, 2015.
+ *
+ * This has not yet undergone a rigorous security audit.
+ *
+ * The usage of AES-XTS should conform to recommendations in NIST
+ * Special Publication 800-38E and IEEE P1619/D16.
+ */
+
+#include <linux/pagemap.h>
+#include <linux/module.h>
+#include <linux/bio.h>
+#include <linux/namei.h>
+#include "fscrypt_private.h"
+
+/*
+ * Call fscrypt_decrypt_page on every single page, reusing the encryption
+ * context.
+ */
+static void completion_pages(struct work_struct *work)
+{
+	struct fscrypt_ctx *ctx =
+		container_of(work, struct fscrypt_ctx, r.work);
+	struct bio *bio = ctx->r.bio;
+	struct bio_vec *bv;
+	int i;
+
+	bio_for_each_segment_all(bv, bio, i) {
+		struct page *page = bv->bv_page;
+		int ret = fscrypt_decrypt_page(page->mapping->host, page,
+				PAGE_SIZE, 0, page->index);
+
+		if (ret) {
+			WARN_ON_ONCE(1);
+			SetPageError(page);
+		} else {
+			SetPageUptodate(page);
+		}
+		unlock_page(page);
+	}
+	fscrypt_release_ctx(ctx);
+	bio_put(bio);
+}
+
+void fscrypt_decrypt_bio_pages(struct fscrypt_ctx *ctx, struct bio *bio)
+{
+	INIT_WORK(&ctx->r.work, completion_pages);
+	ctx->r.bio = bio;
+	queue_work(fscrypt_read_workqueue, &ctx->r.work);
+}
+EXPORT_SYMBOL(fscrypt_decrypt_bio_pages);
+
+void fscrypt_pullback_bio_page(struct page **page, bool restore)
+{
+	struct fscrypt_ctx *ctx;
+	struct page *bounce_page;
+
+	/* The bounce data pages are unmapped. */
+	if ((*page)->mapping)
+		return;
+
+	/* The bounce data page is unmapped. */
+	bounce_page = *page;
+	ctx = (struct fscrypt_ctx *)page_private(bounce_page);
+
+	/* restore control page */
+	*page = ctx->w.control_page;
+
+	if (restore)
+		fscrypt_restore_control_page(bounce_page);
+}
+EXPORT_SYMBOL(fscrypt_pullback_bio_page);
+
+int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
+				sector_t pblk, unsigned int len)
+{
+	struct fscrypt_ctx *ctx;
+	struct page *ciphertext_page = NULL;
+	struct bio *bio;
+	int ret, err = 0;
+
+	BUG_ON(inode->i_sb->s_blocksize != PAGE_SIZE);
+
+	ctx = fscrypt_get_ctx(inode, GFP_NOFS);
+	if (IS_ERR(ctx))
+		return PTR_ERR(ctx);
+
+	ciphertext_page = fscrypt_alloc_bounce_page(ctx, GFP_NOWAIT);
+	if (IS_ERR(ciphertext_page)) {
+		err = PTR_ERR(ciphertext_page);
+		goto errout;
+	}
+
+	while (len--) {
+		err = fscrypt_do_page_crypto(inode, FS_ENCRYPT, lblk,
+					     ZERO_PAGE(0), ciphertext_page,
+					     PAGE_SIZE, 0, GFP_NOFS);
+		if (err)
+			goto errout;
+
+		bio = bio_alloc(GFP_NOWAIT, 1);
+		if (!bio) {
+			err = -ENOMEM;
+			goto errout;
+		}
+		bio->bi_bdev = inode->i_sb->s_bdev;
+		bio->bi_iter.bi_sector =
+			pblk << (inode->i_sb->s_blocksize_bits - 9);
+		bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
+		ret = bio_add_page(bio, ciphertext_page,
+					inode->i_sb->s_blocksize, 0);
+		if (ret != inode->i_sb->s_blocksize) {
+			/* should never happen! */
+			WARN_ON(1);
+			bio_put(bio);
+			err = -EIO;
+			goto errout;
+		}
+		err = submit_bio_wait(0, bio);
+		bio_put(bio);
+		if (err)
+			goto errout;
+		lblk++;
+		pblk++;
+	}
+	err = 0;
+errout:
+	fscrypt_release_ctx(ctx);
+	return err;
+}
+EXPORT_SYMBOL(fscrypt_zeroout_range);
diff --git a/fs/crypto/crypto.c b/fs/crypto/crypto.c
new file mode 100644
index 000000000000..c7835df7e7b8
--- /dev/null
+++ b/fs/crypto/crypto.c
@@ -0,0 +1,492 @@
+/*
+ * This contains encryption functions for per-file encryption.
+ *
+ * Copyright (C) 2015, Google, Inc.
+ * Copyright (C) 2015, Motorola Mobility
+ *
+ * Written by Michael Halcrow, 2014.
+ *
+ * Filename encryption additions
+ *	Uday Savagaonkar, 2014
+ * Encryption policy handling additions
+ *	Ildar Muslukhov, 2014
+ * Add fscrypt_pullback_bio_page()
+ *	Jaegeuk Kim, 2015.
+ *
+ * This has not yet undergone a rigorous security audit.
+ *
+ * The usage of AES-XTS should conform to recommendations in NIST
+ * Special Publication 800-38E and IEEE P1619/D16.
+ */
+
+#include <linux/pagemap.h>
+#include <linux/mempool.h>
+#include <linux/module.h>
+#include <linux/scatterlist.h>
+#include <linux/ratelimit.h>
+#include <linux/dcache.h>
+#include <linux/namei.h>
+#include <crypto/aes.h>
+#include "fscrypt_private.h"
+
+static unsigned int num_prealloc_crypto_pages = 32;
+static unsigned int num_prealloc_crypto_ctxs = 128;
+
+module_param(num_prealloc_crypto_pages, uint, 0444);
+MODULE_PARM_DESC(num_prealloc_crypto_pages,
+		"Number of crypto pages to preallocate");
+module_param(num_prealloc_crypto_ctxs, uint, 0444);
+MODULE_PARM_DESC(num_prealloc_crypto_ctxs,
+		"Number of crypto contexts to preallocate");
+
+static mempool_t *fscrypt_bounce_page_pool = NULL;
+
+static LIST_HEAD(fscrypt_free_ctxs);
+static DEFINE_SPINLOCK(fscrypt_ctx_lock);
+
+struct workqueue_struct *fscrypt_read_workqueue;
+static DEFINE_MUTEX(fscrypt_init_mutex);
+
+static struct kmem_cache *fscrypt_ctx_cachep;
+struct kmem_cache *fscrypt_info_cachep;
+
+/**
+ * fscrypt_release_ctx() - Releases an encryption context
+ * @ctx: The encryption context to release.
+ *
+ * If the encryption context was allocated from the pre-allocated pool, returns
+ * it to that pool. Else, frees it.
+ *
+ * If there's a bounce page in the context, this frees that.
+ */
+void fscrypt_release_ctx(struct fscrypt_ctx *ctx)
+{
+	unsigned long flags;
+
+	if (ctx->flags & FS_CTX_HAS_BOUNCE_BUFFER_FL && ctx->w.bounce_page) {
+		mempool_free(ctx->w.bounce_page, fscrypt_bounce_page_pool);
+		ctx->w.bounce_page = NULL;
+	}
+	ctx->w.control_page = NULL;
+	if (ctx->flags & FS_CTX_REQUIRES_FREE_ENCRYPT_FL) {
+		kmem_cache_free(fscrypt_ctx_cachep, ctx);
+	} else {
+		spin_lock_irqsave(&fscrypt_ctx_lock, flags);
+		list_add(&ctx->free_list, &fscrypt_free_ctxs);
+		spin_unlock_irqrestore(&fscrypt_ctx_lock, flags);
+	}
+}
+EXPORT_SYMBOL(fscrypt_release_ctx);
+
+/**
+ * fscrypt_get_ctx() - Gets an encryption context
+ * @inode:       The inode for which we are doing the crypto
+ * @gfp_flags:   The gfp flag for memory allocation
+ *
+ * Allocates and initializes an encryption context.
+ *
+ * Return: An allocated and initialized encryption context on success; error
+ * value or NULL otherwise.
+ */
+struct fscrypt_ctx *fscrypt_get_ctx(const struct inode *inode, gfp_t gfp_flags)
+{
+	struct fscrypt_ctx *ctx = NULL;
+	struct fscrypt_info *ci = inode->i_crypt_info;
+	unsigned long flags;
+
+	if (ci == NULL)
+		return ERR_PTR(-ENOKEY);
+
+	/*
+	 * We first try getting the ctx from a free list because in
+	 * the common case the ctx will have an allocated and
+	 * initialized crypto tfm, so it's probably a worthwhile
+	 * optimization. For the bounce page, we first try getting it
+	 * from the kernel allocator because that's just about as fast
+	 * as getting it from a list and because a cache of free pages
+	 * should generally be a "last resort" option for a filesystem
+	 * to be able to do its job.
+	 */
+	spin_lock_irqsave(&fscrypt_ctx_lock, flags);
+	ctx = list_first_entry_or_null(&fscrypt_free_ctxs,
+					struct fscrypt_ctx, free_list);
+	if (ctx)
+		list_del(&ctx->free_list);
+	spin_unlock_irqrestore(&fscrypt_ctx_lock, flags);
+	if (!ctx) {
+		ctx = kmem_cache_zalloc(fscrypt_ctx_cachep, gfp_flags);
+		if (!ctx)
+			return ERR_PTR(-ENOMEM);
+		ctx->flags |= FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
+	} else {
+		ctx->flags &= ~FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
+	}
+	ctx->flags &= ~FS_CTX_HAS_BOUNCE_BUFFER_FL;
+	return ctx;
+}
+EXPORT_SYMBOL(fscrypt_get_ctx);
+
+/**
+ * page_crypt_complete() - completion callback for page crypto
+ * @req: The asynchronous cipher request context
+ * @res: The result of the cipher operation
+ */
+static void page_crypt_complete(struct crypto_async_request *req, int res)
+{
+	struct fscrypt_completion_result *ecr = req->data;
+
+	if (res == -EINPROGRESS)
+		return;
+	ecr->res = res;
+	complete(&ecr->completion);
+}
+
+int fscrypt_do_page_crypto(const struct inode *inode, fscrypt_direction_t rw,
+			   u64 lblk_num, struct page *src_page,
+			   struct page *dest_page, unsigned int len,
+			   unsigned int offs, gfp_t gfp_flags)
+{
+	struct {
+		__le64 index;
+		u8 padding[FS_IV_SIZE - sizeof(__le64)];
+	} iv;
+	struct skcipher_request *req = NULL;
+	DECLARE_FS_COMPLETION_RESULT(ecr);
+	struct scatterlist dst, src;
+	struct fscrypt_info *ci = inode->i_crypt_info;
+	struct crypto_skcipher *tfm = ci->ci_ctfm;
+	int res = 0;
+
+	BUG_ON(len == 0);
+
+	BUILD_BUG_ON(sizeof(iv) != FS_IV_SIZE);
+	BUILD_BUG_ON(AES_BLOCK_SIZE != FS_IV_SIZE);
+	iv.index = cpu_to_le64(lblk_num);
+	memset(iv.padding, 0, sizeof(iv.padding));
+
+	if (ci->ci_essiv_tfm != NULL) {
+		crypto_cipher_encrypt_one(ci->ci_essiv_tfm, (u8 *)&iv,
+					  (u8 *)&iv);
+	}
+
+	req = skcipher_request_alloc(tfm, gfp_flags);
+	if (!req) {
+		printk_ratelimited(KERN_ERR
+				"%s: crypto_request_alloc() failed\n",
+				__func__);
+		return -ENOMEM;
+	}
+
+	skcipher_request_set_callback(
+		req, CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
+		page_crypt_complete, &ecr);
+
+	sg_init_table(&dst, 1);
+	sg_set_page(&dst, dest_page, len, offs);
+	sg_init_table(&src, 1);
+	sg_set_page(&src, src_page, len, offs);
+	skcipher_request_set_crypt(req, &src, &dst, len, &iv);
+	if (rw == FS_DECRYPT)
+		res = crypto_skcipher_decrypt(req);
+	else
+		res = crypto_skcipher_encrypt(req);
+	if (res == -EINPROGRESS || res == -EBUSY) {
+		BUG_ON(req->base.data != &ecr);
+		wait_for_completion(&ecr.completion);
+		res = ecr.res;
+	}
+	skcipher_request_free(req);
+	if (res) {
+		printk_ratelimited(KERN_ERR
+			"%s: crypto_skcipher_encrypt() returned %d\n",
+			__func__, res);
+		return res;
+	}
+	return 0;
+}
+
+struct page *fscrypt_alloc_bounce_page(struct fscrypt_ctx *ctx,
+				       gfp_t gfp_flags)
+{
+	ctx->w.bounce_page = mempool_alloc(fscrypt_bounce_page_pool, gfp_flags);
+	if (ctx->w.bounce_page == NULL)
+		return ERR_PTR(-ENOMEM);
+	ctx->flags |= FS_CTX_HAS_BOUNCE_BUFFER_FL;
+	return ctx->w.bounce_page;
+}
+
+/**
+ * fscypt_encrypt_page() - Encrypts a page
+ * @inode:     The inode for which the encryption should take place
+ * @page:      The page to encrypt. Must be locked for bounce-page
+ *             encryption.
+ * @len:       Length of data to encrypt in @page and encrypted
+ *             data in returned page.
+ * @offs:      Offset of data within @page and returned
+ *             page holding encrypted data.
+ * @lblk_num:  Logical block number. This must be unique for multiple
+ *             calls with same inode, except when overwriting
+ *             previously written data.
+ * @gfp_flags: The gfp flag for memory allocation
+ *
+ * Encrypts @page using the ctx encryption context. Performs encryption
+ * either in-place or into a newly allocated bounce page.
+ * Called on the page write path.
+ *
+ * Bounce page allocation is the default.
+ * In this case, the contents of @page are encrypted and stored in an
+ * allocated bounce page. @page has to be locked and the caller must call
+ * fscrypt_restore_control_page() on the returned ciphertext page to
+ * release the bounce buffer and the encryption context.
+ *
+ * In-place encryption is used by setting the FS_CFLG_OWN_PAGES flag in
+ * fscrypt_operations. Here, the input-page is returned with its content
+ * encrypted.
+ *
+ * Return: A page with the encrypted content on success. Else, an
+ * error value or NULL.
+ */
+struct page *fscrypt_encrypt_page(const struct inode *inode,
+				struct page *page,
+				unsigned int len,
+				unsigned int offs,
+				u64 lblk_num, gfp_t gfp_flags)
+
+{
+	struct fscrypt_ctx *ctx;
+	struct page *ciphertext_page = page;
+	int err;
+
+	BUG_ON(len % FS_CRYPTO_BLOCK_SIZE != 0);
+
+	if (inode->i_sb->s_cop->flags & FS_CFLG_OWN_PAGES) {
+		/* with inplace-encryption we just encrypt the page */
+		err = fscrypt_do_page_crypto(inode, FS_ENCRYPT, lblk_num, page,
+					     ciphertext_page, len, offs,
+					     gfp_flags);
+		if (err)
+			return ERR_PTR(err);
+
+		return ciphertext_page;
+	}
+
+	BUG_ON(!PageLocked(page));
+
+	ctx = fscrypt_get_ctx(inode, gfp_flags);
+	if (IS_ERR(ctx))
+		return (struct page *)ctx;
+
+	/* The encryption operation will require a bounce page. */
+	ciphertext_page = fscrypt_alloc_bounce_page(ctx, gfp_flags);
+	if (IS_ERR(ciphertext_page))
+		goto errout;
+
+	ctx->w.control_page = page;
+	err = fscrypt_do_page_crypto(inode, FS_ENCRYPT, lblk_num,
+				     page, ciphertext_page, len, offs,
+				     gfp_flags);
+	if (err) {
+		ciphertext_page = ERR_PTR(err);
+		goto errout;
+	}
+	SetPagePrivate(ciphertext_page);
+	set_page_private(ciphertext_page, (unsigned long)ctx);
+	lock_page(ciphertext_page);
+	return ciphertext_page;
+
+errout:
+	fscrypt_release_ctx(ctx);
+	return ciphertext_page;
+}
+EXPORT_SYMBOL(fscrypt_encrypt_page);
+
+/**
+ * fscrypt_decrypt_page() - Decrypts a page in-place
+ * @inode:     The corresponding inode for the page to decrypt.
+ * @page:      The page to decrypt. Must be locked in case
+ *             it is a writeback page (FS_CFLG_OWN_PAGES unset).
+ * @len:       Number of bytes in @page to be decrypted.
+ * @offs:      Start of data in @page.
+ * @lblk_num:  Logical block number.
+ *
+ * Decrypts page in-place using the ctx encryption context.
+ *
+ * Called from the read completion callback.
+ *
+ * Return: Zero on success, non-zero otherwise.
+ */
+int fscrypt_decrypt_page(const struct inode *inode, struct page *page,
+			unsigned int len, unsigned int offs, u64 lblk_num)
+{
+	if (!(inode->i_sb->s_cop->flags & FS_CFLG_OWN_PAGES))
+		BUG_ON(!PageLocked(page));
+
+	return fscrypt_do_page_crypto(inode, FS_DECRYPT, lblk_num, page, page,
+				      len, offs, GFP_NOFS);
+}
+EXPORT_SYMBOL(fscrypt_decrypt_page);
+
+/*
+ * Validate dentries for encrypted directories to make sure we aren't
+ * potentially caching stale data after a key has been added or
+ * removed.
+ */
+static int fscrypt_d_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	struct dentry *dir;
+	int dir_has_key, cached_with_key;
+
+	if (flags & LOOKUP_RCU)
+		return -ECHILD;
+
+	dir = dget_parent(dentry);
+	if (!d_inode(dir)->i_sb->s_cop->is_encrypted(d_inode(dir))) {
+		dput(dir);
+		return 0;
+	}
+
+	/* this should eventually be an flag in d_flags */
+	spin_lock(&dentry->d_lock);
+	cached_with_key = dentry->d_flags & DCACHE_ENCRYPTED_WITH_KEY;
+	spin_unlock(&dentry->d_lock);
+	dir_has_key = (d_inode(dir)->i_crypt_info != NULL);
+	dput(dir);
+
+	/*
+	 * If the dentry was cached without the key, and it is a
+	 * negative dentry, it might be a valid name.  We can't check
+	 * if the key has since been made available due to locking
+	 * reasons, so we fail the validation so ext4_lookup() can do
+	 * this check.
+	 *
+	 * We also fail the validation if the dentry was created with
+	 * the key present, but we no longer have the key, or vice versa.
+	 */
+	if ((!cached_with_key && d_is_negative(dentry)) ||
+			(!cached_with_key && dir_has_key) ||
+			(cached_with_key && !dir_has_key))
+		return 0;
+	return 1;
+}
+
+const struct dentry_operations fscrypt_d_ops = {
+	.d_revalidate = fscrypt_d_revalidate,
+};
+EXPORT_SYMBOL(fscrypt_d_ops);
+
+void fscrypt_restore_control_page(struct page *page)
+{
+	struct fscrypt_ctx *ctx;
+
+	ctx = (struct fscrypt_ctx *)page_private(page);
+	set_page_private(page, (unsigned long)NULL);
+	ClearPagePrivate(page);
+	unlock_page(page);
+	fscrypt_release_ctx(ctx);
+}
+EXPORT_SYMBOL(fscrypt_restore_control_page);
+
+static void fscrypt_destroy(void)
+{
+	struct fscrypt_ctx *pos, *n;
+
+	list_for_each_entry_safe(pos, n, &fscrypt_free_ctxs, free_list)
+		kmem_cache_free(fscrypt_ctx_cachep, pos);
+	INIT_LIST_HEAD(&fscrypt_free_ctxs);
+	mempool_destroy(fscrypt_bounce_page_pool);
+	fscrypt_bounce_page_pool = NULL;
+}
+
+/**
+ * fscrypt_initialize() - allocate major buffers for fs encryption.
+ * @cop_flags:  fscrypt operations flags
+ *
+ * We only call this when we start accessing encrypted files, since it
+ * results in memory getting allocated that wouldn't otherwise be used.
+ *
+ * Return: Zero on success, non-zero otherwise.
+ */
+int fscrypt_initialize(unsigned int cop_flags)
+{
+	int i, res = -ENOMEM;
+
+	/*
+	 * No need to allocate a bounce page pool if there already is one or
+	 * this FS won't use it.
+	 */
+	if (cop_flags & FS_CFLG_OWN_PAGES || fscrypt_bounce_page_pool)
+		return 0;
+
+	mutex_lock(&fscrypt_init_mutex);
+	if (fscrypt_bounce_page_pool)
+		goto already_initialized;
+
+	for (i = 0; i < num_prealloc_crypto_ctxs; i++) {
+		struct fscrypt_ctx *ctx;
+
+		ctx = kmem_cache_zalloc(fscrypt_ctx_cachep, GFP_NOFS);
+		if (!ctx)
+			goto fail;
+		list_add(&ctx->free_list, &fscrypt_free_ctxs);
+	}
+
+	fscrypt_bounce_page_pool =
+		mempool_create_page_pool(num_prealloc_crypto_pages, 0);
+	if (!fscrypt_bounce_page_pool)
+		goto fail;
+
+already_initialized:
+	mutex_unlock(&fscrypt_init_mutex);
+	return 0;
+fail:
+	fscrypt_destroy();
+	mutex_unlock(&fscrypt_init_mutex);
+	return res;
+}
+
+/**
+ * fscrypt_init() - Set up for fs encryption.
+ */
+static int __init fscrypt_init(void)
+{
+	fscrypt_read_workqueue = alloc_workqueue("fscrypt_read_queue",
+							WQ_HIGHPRI, 0);
+	if (!fscrypt_read_workqueue)
+		goto fail;
+
+	fscrypt_ctx_cachep = KMEM_CACHE(fscrypt_ctx, SLAB_RECLAIM_ACCOUNT);
+	if (!fscrypt_ctx_cachep)
+		goto fail_free_queue;
+
+	fscrypt_info_cachep = KMEM_CACHE(fscrypt_info, SLAB_RECLAIM_ACCOUNT);
+	if (!fscrypt_info_cachep)
+		goto fail_free_ctx;
+
+	return 0;
+
+fail_free_ctx:
+	kmem_cache_destroy(fscrypt_ctx_cachep);
+fail_free_queue:
+	destroy_workqueue(fscrypt_read_workqueue);
+fail:
+	return -ENOMEM;
+}
+module_init(fscrypt_init)
+
+/**
+ * fscrypt_exit() - Shutdown the fs encryption system
+ */
+static void __exit fscrypt_exit(void)
+{
+	fscrypt_destroy();
+
+	if (fscrypt_read_workqueue)
+		destroy_workqueue(fscrypt_read_workqueue);
+	kmem_cache_destroy(fscrypt_ctx_cachep);
+	kmem_cache_destroy(fscrypt_info_cachep);
+
+	fscrypt_essiv_cleanup();
+}
+module_exit(fscrypt_exit);
+
+MODULE_LICENSE("GPL");
diff --git a/fs/crypto/fname.c b/fs/crypto/fname.c
new file mode 100644
index 000000000000..ad9f814fdead
--- /dev/null
+++ b/fs/crypto/fname.c
@@ -0,0 +1,455 @@
+/*
+ * This contains functions for filename crypto management
+ *
+ * Copyright (C) 2015, Google, Inc.
+ * Copyright (C) 2015, Motorola Mobility
+ *
+ * Written by Uday Savagaonkar, 2014.
+ * Modified by Jaegeuk Kim, 2015.
+ *
+ * This has not yet undergone a rigorous security audit.
+ */
+
+#include <linux/scatterlist.h>
+#include <linux/ratelimit.h>
+#include "fscrypt_private.h"
+
+/**
+ * fname_crypt_complete() - completion callback for filename crypto
+ * @req: The asynchronous cipher request context
+ * @res: The result of the cipher operation
+ */
+static void fname_crypt_complete(struct crypto_async_request *req, int res)
+{
+	struct fscrypt_completion_result *ecr = req->data;
+
+	if (res == -EINPROGRESS)
+		return;
+	ecr->res = res;
+	complete(&ecr->completion);
+}
+
+/**
+ * fname_encrypt() - encrypt a filename
+ *
+ * The caller must have allocated sufficient memory for the @oname string.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int fname_encrypt(struct inode *inode,
+			const struct qstr *iname, struct fscrypt_str *oname)
+{
+	struct skcipher_request *req = NULL;
+	DECLARE_FS_COMPLETION_RESULT(ecr);
+	struct fscrypt_info *ci = inode->i_crypt_info;
+	struct crypto_skcipher *tfm = ci->ci_ctfm;
+	int res = 0;
+	char iv[FS_CRYPTO_BLOCK_SIZE];
+	struct scatterlist sg;
+	int padding = 4 << (ci->ci_flags & FS_POLICY_FLAGS_PAD_MASK);
+	unsigned int lim;
+	unsigned int cryptlen;
+
+	lim = inode->i_sb->s_cop->max_namelen(inode);
+	if (iname->len <= 0 || iname->len > lim)
+		return -EIO;
+
+	/*
+	 * Copy the filename to the output buffer for encrypting in-place and
+	 * pad it with the needed number of NUL bytes.
+	 */
+	cryptlen = max_t(unsigned int, iname->len, FS_CRYPTO_BLOCK_SIZE);
+	cryptlen = round_up(cryptlen, padding);
+	cryptlen = min(cryptlen, lim);
+	memcpy(oname->name, iname->name, iname->len);
+	memset(oname->name + iname->len, 0, cryptlen - iname->len);
+
+	/* Initialize the IV */
+	memset(iv, 0, FS_CRYPTO_BLOCK_SIZE);
+
+	/* Set up the encryption request */
+	req = skcipher_request_alloc(tfm, GFP_NOFS);
+	if (!req) {
+		printk_ratelimited(KERN_ERR
+			"%s: skcipher_request_alloc() failed\n", __func__);
+		return -ENOMEM;
+	}
+	skcipher_request_set_callback(req,
+			CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
+			fname_crypt_complete, &ecr);
+	sg_init_one(&sg, oname->name, cryptlen);
+	skcipher_request_set_crypt(req, &sg, &sg, cryptlen, iv);
+
+	/* Do the encryption */
+	res = crypto_skcipher_encrypt(req);
+	if (res == -EINPROGRESS || res == -EBUSY) {
+		/* Request is being completed asynchronously; wait for it */
+		wait_for_completion(&ecr.completion);
+		res = ecr.res;
+	}
+	skcipher_request_free(req);
+	if (res < 0) {
+		printk_ratelimited(KERN_ERR
+				"%s: Error (error code %d)\n", __func__, res);
+		return res;
+	}
+
+	oname->len = cryptlen;
+	return 0;
+}
+
+/**
+ * fname_decrypt() - decrypt a filename
+ *
+ * The caller must have allocated sufficient memory for the @oname string.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int fname_decrypt(struct inode *inode,
+				const struct fscrypt_str *iname,
+				struct fscrypt_str *oname)
+{
+	struct skcipher_request *req = NULL;
+	DECLARE_FS_COMPLETION_RESULT(ecr);
+	struct scatterlist src_sg, dst_sg;
+	struct fscrypt_info *ci = inode->i_crypt_info;
+	struct crypto_skcipher *tfm = ci->ci_ctfm;
+	int res = 0;
+	char iv[FS_CRYPTO_BLOCK_SIZE];
+	unsigned lim;
+
+	lim = inode->i_sb->s_cop->max_namelen(inode);
+	if (iname->len <= 0 || iname->len > lim)
+		return -EIO;
+
+	/* Allocate request */
+	req = skcipher_request_alloc(tfm, GFP_NOFS);
+	if (!req) {
+		printk_ratelimited(KERN_ERR
+			"%s: crypto_request_alloc() failed\n",  __func__);
+		return -ENOMEM;
+	}
+	skcipher_request_set_callback(req,
+		CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
+		fname_crypt_complete, &ecr);
+
+	/* Initialize IV */
+	memset(iv, 0, FS_CRYPTO_BLOCK_SIZE);
+
+	/* Create decryption request */
+	sg_init_one(&src_sg, iname->name, iname->len);
+	sg_init_one(&dst_sg, oname->name, oname->len);
+	skcipher_request_set_crypt(req, &src_sg, &dst_sg, iname->len, iv);
+	res = crypto_skcipher_decrypt(req);
+	if (res == -EINPROGRESS || res == -EBUSY) {
+		wait_for_completion(&ecr.completion);
+		res = ecr.res;
+	}
+	skcipher_request_free(req);
+	if (res < 0) {
+		printk_ratelimited(KERN_ERR
+				"%s: Error (error code %d)\n", __func__, res);
+		return res;
+	}
+
+	oname->len = strnlen(oname->name, iname->len);
+	return 0;
+}
+
+static const char *lookup_table =
+	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+,";
+
+#define BASE64_CHARS(nbytes)	DIV_ROUND_UP((nbytes) * 4, 3)
+
+/**
+ * digest_encode() -
+ *
+ * Encodes the input digest using characters from the set [a-zA-Z0-9_+].
+ * The encoded string is roughly 4/3 times the size of the input string.
+ */
+static int digest_encode(const char *src, int len, char *dst)
+{
+	int i = 0, bits = 0, ac = 0;
+	char *cp = dst;
+
+	while (i < len) {
+		ac += (((unsigned char) src[i]) << bits);
+		bits += 8;
+		do {
+			*cp++ = lookup_table[ac & 0x3f];
+			ac >>= 6;
+			bits -= 6;
+		} while (bits >= 6);
+		i++;
+	}
+	if (bits)
+		*cp++ = lookup_table[ac & 0x3f];
+	return cp - dst;
+}
+
+static int digest_decode(const char *src, int len, char *dst)
+{
+	int i = 0, bits = 0, ac = 0;
+	const char *p;
+	char *cp = dst;
+
+	while (i < len) {
+		p = strchr(lookup_table, src[i]);
+		if (p == NULL || src[i] == 0)
+			return -2;
+		ac += (p - lookup_table) << bits;
+		bits += 6;
+		if (bits >= 8) {
+			*cp++ = ac & 0xff;
+			ac >>= 8;
+			bits -= 8;
+		}
+		i++;
+	}
+	if (ac)
+		return -1;
+	return cp - dst;
+}
+
+u32 fscrypt_fname_encrypted_size(const struct inode *inode, u32 ilen)
+{
+	int padding = 32;
+	struct fscrypt_info *ci = inode->i_crypt_info;
+
+	if (ci)
+		padding = 4 << (ci->ci_flags & FS_POLICY_FLAGS_PAD_MASK);
+	ilen = max(ilen, (u32)FS_CRYPTO_BLOCK_SIZE);
+	return round_up(ilen, padding);
+}
+EXPORT_SYMBOL(fscrypt_fname_encrypted_size);
+
+/**
+ * fscrypt_fname_crypto_alloc_obuff() -
+ *
+ * Allocates an output buffer that is sufficient for the crypto operation
+ * specified by the context and the direction.
+ */
+int fscrypt_fname_alloc_buffer(const struct inode *inode,
+				u32 ilen, struct fscrypt_str *crypto_str)
+{
+	u32 olen = fscrypt_fname_encrypted_size(inode, ilen);
+	const u32 max_encoded_len =
+		max_t(u32, BASE64_CHARS(FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE),
+		      1 + BASE64_CHARS(sizeof(struct fscrypt_digested_name)));
+
+	crypto_str->len = olen;
+	olen = max(olen, max_encoded_len);
+
+	/*
+	 * Allocated buffer can hold one more character to null-terminate the
+	 * string
+	 */
+	crypto_str->name = kmalloc(olen + 1, GFP_NOFS);
+	if (!(crypto_str->name))
+		return -ENOMEM;
+	return 0;
+}
+EXPORT_SYMBOL(fscrypt_fname_alloc_buffer);
+
+/**
+ * fscrypt_fname_crypto_free_buffer() -
+ *
+ * Frees the buffer allocated for crypto operation.
+ */
+void fscrypt_fname_free_buffer(struct fscrypt_str *crypto_str)
+{
+	if (!crypto_str)
+		return;
+	kfree(crypto_str->name);
+	crypto_str->name = NULL;
+}
+EXPORT_SYMBOL(fscrypt_fname_free_buffer);
+
+/**
+ * fscrypt_fname_disk_to_usr() - converts a filename from disk space to user
+ * space
+ *
+ * The caller must have allocated sufficient memory for the @oname string.
+ *
+ * If the key is available, we'll decrypt the disk name; otherwise, we'll encode
+ * it for presentation.  Short names are directly base64-encoded, while long
+ * names are encoded in fscrypt_digested_name format.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+int fscrypt_fname_disk_to_usr(struct inode *inode,
+			u32 hash, u32 minor_hash,
+			const struct fscrypt_str *iname,
+			struct fscrypt_str *oname)
+{
+	const struct qstr qname = FSTR_TO_QSTR(iname);
+	struct fscrypt_digested_name digested_name;
+
+	if (fscrypt_is_dot_dotdot(&qname)) {
+		oname->name[0] = '.';
+		oname->name[iname->len - 1] = '.';
+		oname->len = iname->len;
+		return 0;
+	}
+
+	if (iname->len < FS_CRYPTO_BLOCK_SIZE)
+		return -EUCLEAN;
+
+	if (inode->i_crypt_info)
+		return fname_decrypt(inode, iname, oname);
+
+	if (iname->len <= FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE) {
+		oname->len = digest_encode(iname->name, iname->len,
+					   oname->name);
+		return 0;
+	}
+	if (hash) {
+		digested_name.hash = hash;
+		digested_name.minor_hash = minor_hash;
+	} else {
+		digested_name.hash = 0;
+		digested_name.minor_hash = 0;
+	}
+	memcpy(digested_name.digest,
+	       FSCRYPT_FNAME_DIGEST(iname->name, iname->len),
+	       FSCRYPT_FNAME_DIGEST_SIZE);
+	oname->name[0] = '_';
+	oname->len = 1 + digest_encode((const char *)&digested_name,
+				       sizeof(digested_name), oname->name + 1);
+	return 0;
+}
+EXPORT_SYMBOL(fscrypt_fname_disk_to_usr);
+
+/**
+ * fscrypt_fname_usr_to_disk() - converts a filename from user space to disk
+ * space
+ *
+ * The caller must have allocated sufficient memory for the @oname string.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+int fscrypt_fname_usr_to_disk(struct inode *inode,
+			const struct qstr *iname,
+			struct fscrypt_str *oname)
+{
+	if (fscrypt_is_dot_dotdot(iname)) {
+		oname->name[0] = '.';
+		oname->name[iname->len - 1] = '.';
+		oname->len = iname->len;
+		return 0;
+	}
+	if (inode->i_crypt_info)
+		return fname_encrypt(inode, iname, oname);
+	/*
+	 * Without a proper key, a user is not allowed to modify the filenames
+	 * in a directory. Consequently, a user space name cannot be mapped to
+	 * a disk-space name
+	 */
+	return -ENOKEY;
+}
+EXPORT_SYMBOL(fscrypt_fname_usr_to_disk);
+
+/**
+ * fscrypt_setup_filename() - prepare to search a possibly encrypted directory
+ * @dir: the directory that will be searched
+ * @iname: the user-provided filename being searched for
+ * @lookup: 1 if we're allowed to proceed without the key because it's
+ *	->lookup() or we're finding the dir_entry for deletion; 0 if we cannot
+ *	proceed without the key because we're going to create the dir_entry.
+ * @fname: the filename information to be filled in
+ *
+ * Given a user-provided filename @iname, this function sets @fname->disk_name
+ * to the name that would be stored in the on-disk directory entry, if possible.
+ * If the directory is unencrypted this is simply @iname.  Else, if we have the
+ * directory's encryption key, then @iname is the plaintext, so we encrypt it to
+ * get the disk_name.
+ *
+ * Else, for keyless @lookup operations, @iname is the presented ciphertext, so
+ * we decode it to get either the ciphertext disk_name (for short names) or the
+ * fscrypt_digested_name (for long names).  Non-@lookup operations will be
+ * impossible in this case, so we fail them with ENOKEY.
+ *
+ * If successful, fscrypt_free_filename() must be called later to clean up.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+int fscrypt_setup_filename(struct inode *dir, const struct qstr *iname,
+			      int lookup, struct fscrypt_name *fname)
+{
+	int ret;
+	int digested;
+
+	memset(fname, 0, sizeof(struct fscrypt_name));
+	fname->usr_fname = iname;
+
+	if (!dir->i_sb->s_cop->is_encrypted(dir) ||
+				fscrypt_is_dot_dotdot(iname)) {
+		fname->disk_name.name = (unsigned char *)iname->name;
+		fname->disk_name.len = iname->len;
+		return 0;
+	}
+	ret = fscrypt_get_encryption_info(dir);
+	if (ret && ret != -EOPNOTSUPP)
+		return ret;
+
+	if (dir->i_crypt_info) {
+		ret = fscrypt_fname_alloc_buffer(dir, iname->len,
+							&fname->crypto_buf);
+		if (ret)
+			return ret;
+		ret = fname_encrypt(dir, iname, &fname->crypto_buf);
+		if (ret)
+			goto errout;
+		fname->disk_name.name = fname->crypto_buf.name;
+		fname->disk_name.len = fname->crypto_buf.len;
+		return 0;
+	}
+	if (!lookup)
+		return -ENOKEY;
+
+	/*
+	 * We don't have the key and we are doing a lookup; decode the
+	 * user-supplied name
+	 */
+	if (iname->name[0] == '_') {
+		if (iname->len !=
+		    1 + BASE64_CHARS(sizeof(struct fscrypt_digested_name)))
+			return -ENOENT;
+		digested = 1;
+	} else {
+		if (iname->len >
+		    BASE64_CHARS(FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE))
+			return -ENOENT;
+		digested = 0;
+	}
+
+	fname->crypto_buf.name =
+		kmalloc(max_t(size_t, FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE,
+			      sizeof(struct fscrypt_digested_name)),
+			GFP_KERNEL);
+	if (fname->crypto_buf.name == NULL)
+		return -ENOMEM;
+
+	ret = digest_decode(iname->name + digested, iname->len - digested,
+				fname->crypto_buf.name);
+	if (ret < 0) {
+		ret = -ENOENT;
+		goto errout;
+	}
+	fname->crypto_buf.len = ret;
+	if (digested) {
+		const struct fscrypt_digested_name *n =
+			(const void *)fname->crypto_buf.name;
+		fname->hash = n->hash;
+		fname->minor_hash = n->minor_hash;
+	} else {
+		fname->disk_name.name = fname->crypto_buf.name;
+		fname->disk_name.len = fname->crypto_buf.len;
+	}
+	return 0;
+
+errout:
+	fscrypt_fname_free_buffer(&fname->crypto_buf);
+	return ret;
+}
+EXPORT_SYMBOL(fscrypt_setup_filename);
diff --git a/fs/crypto/fscrypt_private.h b/fs/crypto/fscrypt_private.h
new file mode 100644
index 000000000000..79d79755d79b
--- /dev/null
+++ b/fs/crypto/fscrypt_private.h
@@ -0,0 +1,107 @@
+/*
+ * fscrypt_private.h
+ *
+ * Copyright (C) 2015, Google, Inc.
+ *
+ * This contains encryption key functions.
+ *
+ * Written by Michael Halcrow, Ildar Muslukhov, and Uday Savagaonkar, 2015.
+ */
+
+#ifndef _FSCRYPT_PRIVATE_H
+#define _FSCRYPT_PRIVATE_H
+
+#include <linux/fscrypt_supp.h>
+#include <crypto/hash.h>
+
+/* Encryption parameters */
+#define FS_IV_SIZE			16
+#define FS_AES_128_ECB_KEY_SIZE		16
+#define FS_AES_128_CBC_KEY_SIZE		16
+#define FS_AES_128_CTS_KEY_SIZE		16
+#define FS_AES_256_GCM_KEY_SIZE		32
+#define FS_AES_256_CBC_KEY_SIZE		32
+#define FS_AES_256_CTS_KEY_SIZE		32
+#define FS_AES_256_XTS_KEY_SIZE		64
+
+#define FS_KEY_DERIVATION_NONCE_SIZE		16
+
+/**
+ * Encryption context for inode
+ *
+ * Protector format:
+ *  1 byte: Protector format (1 = this version)
+ *  1 byte: File contents encryption mode
+ *  1 byte: File names encryption mode
+ *  1 byte: Flags
+ *  8 bytes: Master Key descriptor
+ *  16 bytes: Encryption Key derivation nonce
+ */
+struct fscrypt_context {
+	u8 format;
+	u8 contents_encryption_mode;
+	u8 filenames_encryption_mode;
+	u8 flags;
+	u8 master_key_descriptor[FS_KEY_DESCRIPTOR_SIZE];
+	u8 nonce[FS_KEY_DERIVATION_NONCE_SIZE];
+} __packed;
+
+#define FS_ENCRYPTION_CONTEXT_FORMAT_V1		1
+
+/*
+ * A pointer to this structure is stored in the file system's in-core
+ * representation of an inode.
+ */
+struct fscrypt_info {
+	u8 ci_data_mode;
+	u8 ci_filename_mode;
+	u8 ci_flags;
+	struct crypto_skcipher *ci_ctfm;
+	struct crypto_cipher *ci_essiv_tfm;
+	u8 ci_master_key[FS_KEY_DESCRIPTOR_SIZE];
+};
+
+typedef enum {
+	FS_DECRYPT = 0,
+	FS_ENCRYPT,
+} fscrypt_direction_t;
+
+#define FS_CTX_REQUIRES_FREE_ENCRYPT_FL		0x00000001
+#define FS_CTX_HAS_BOUNCE_BUFFER_FL		0x00000002
+
+struct fscrypt_completion_result {
+	struct completion completion;
+	int res;
+};
+
+#define DECLARE_FS_COMPLETION_RESULT(ecr) \
+	struct fscrypt_completion_result ecr = { \
+		COMPLETION_INITIALIZER_ONSTACK((ecr).completion), 0 }
+
+/* bio stuffs */
+#define REQ_OP_READ	READ
+#define REQ_OP_WRITE	WRITE
+#define bio_op(bio)	((bio)->bi_rw & 1)
+
+static inline void bio_set_op_attrs(struct bio *bio, unsigned op,
+		unsigned op_flags)
+{
+	bio->bi_rw = op | op_flags;
+}
+
+/* crypto.c */
+extern int fscrypt_initialize(unsigned int cop_flags);
+extern struct workqueue_struct *fscrypt_read_workqueue;
+extern int fscrypt_do_page_crypto(const struct inode *inode,
+				  fscrypt_direction_t rw, u64 lblk_num,
+				  struct page *src_page,
+				  struct page *dest_page,
+				  unsigned int len, unsigned int offs,
+				  gfp_t gfp_flags);
+extern struct page *fscrypt_alloc_bounce_page(struct fscrypt_ctx *ctx,
+					      gfp_t gfp_flags);
+
+/* keyinfo.c */
+extern void __exit fscrypt_essiv_cleanup(void);
+
+#endif /* _FSCRYPT_PRIVATE_H */
diff --git a/fs/crypto/keyinfo.c b/fs/crypto/keyinfo.c
new file mode 100644
index 000000000000..66e0728e9bbe
--- /dev/null
+++ b/fs/crypto/keyinfo.c
@@ -0,0 +1,381 @@
+/*
+ * key management facility for FS encryption support.
+ *
+ * Copyright (C) 2015, Google, Inc.
+ *
+ * This contains encryption key functions.
+ *
+ * Written by Michael Halcrow, Ildar Muslukhov, and Uday Savagaonkar, 2015.
+ */
+
+#include <keys/user-type.h>
+#include <linux/scatterlist.h>
+#include <linux/ratelimit.h>
+#include <crypto/aes.h>
+#include <crypto/sha.h>
+#include "fscrypt_private.h"
+
+static struct crypto_shash *essiv_hash_tfm;
+
+static void derive_crypt_complete(struct crypto_async_request *req, int rc)
+{
+	struct fscrypt_completion_result *ecr = req->data;
+
+	if (rc == -EINPROGRESS)
+		return;
+
+	ecr->res = rc;
+	complete(&ecr->completion);
+}
+
+/**
+ * derive_key_aes() - Derive a key using AES-128-ECB
+ * @deriving_key: Encryption key used for derivation.
+ * @source_key:   Source key to which to apply derivation.
+ * @derived_raw_key:  Derived raw key.
+ *
+ * Return: Zero on success; non-zero otherwise.
+ */
+static int derive_key_aes(u8 deriving_key[FS_AES_128_ECB_KEY_SIZE],
+				const struct fscrypt_key *source_key,
+				u8 derived_raw_key[FS_MAX_KEY_SIZE])
+{
+	int res = 0;
+	struct skcipher_request *req = NULL;
+	DECLARE_FS_COMPLETION_RESULT(ecr);
+	struct scatterlist src_sg, dst_sg;
+	struct crypto_skcipher *tfm = crypto_alloc_skcipher("ecb(aes)", 0, 0);
+
+	if (IS_ERR(tfm)) {
+		res = PTR_ERR(tfm);
+		tfm = NULL;
+		goto out;
+	}
+	crypto_skcipher_set_flags(tfm, CRYPTO_TFM_REQ_WEAK_KEY);
+	req = skcipher_request_alloc(tfm, GFP_NOFS);
+	if (!req) {
+		res = -ENOMEM;
+		goto out;
+	}
+	skcipher_request_set_callback(req,
+			CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
+			derive_crypt_complete, &ecr);
+	res = crypto_skcipher_setkey(tfm, deriving_key,
+					FS_AES_128_ECB_KEY_SIZE);
+	if (res < 0)
+		goto out;
+
+	sg_init_one(&src_sg, source_key->raw, source_key->size);
+	sg_init_one(&dst_sg, derived_raw_key, source_key->size);
+	skcipher_request_set_crypt(req, &src_sg, &dst_sg, source_key->size,
+				   NULL);
+	res = crypto_skcipher_encrypt(req);
+	if (res == -EINPROGRESS || res == -EBUSY) {
+		wait_for_completion(&ecr.completion);
+		res = ecr.res;
+	}
+out:
+	skcipher_request_free(req);
+	crypto_free_skcipher(tfm);
+	return res;
+}
+
+static int validate_user_key(struct fscrypt_info *crypt_info,
+			struct fscrypt_context *ctx, u8 *raw_key,
+			const char *prefix, int min_keysize)
+{
+	char *description;
+	struct key *keyring_key;
+	struct fscrypt_key *master_key;
+	const struct user_key_payload *ukp;
+	int res;
+
+	description = kasprintf(GFP_NOFS, "%s%*phN", prefix,
+				FS_KEY_DESCRIPTOR_SIZE,
+				ctx->master_key_descriptor);
+	if (!description)
+		return -ENOMEM;
+
+	keyring_key = request_key(&key_type_logon, description, NULL);
+	kfree(description);
+	if (IS_ERR(keyring_key))
+		return PTR_ERR(keyring_key);
+	down_read(&keyring_key->sem);
+
+	if (keyring_key->type != &key_type_logon) {
+		printk_once(KERN_WARNING
+				"%s: key type must be logon\n", __func__);
+		res = -ENOKEY;
+		goto out;
+	}
+	ukp = user_key_payload(keyring_key);
+	if (ukp->datalen != sizeof(struct fscrypt_key)) {
+		res = -EINVAL;
+		goto out;
+	}
+	master_key = (struct fscrypt_key *)ukp->data;
+	BUILD_BUG_ON(FS_AES_128_ECB_KEY_SIZE != FS_KEY_DERIVATION_NONCE_SIZE);
+
+	if (master_key->size < min_keysize || master_key->size > FS_MAX_KEY_SIZE
+	    || master_key->size % AES_BLOCK_SIZE != 0) {
+		printk_once(KERN_WARNING
+				"%s: key size incorrect: %d\n",
+				__func__, master_key->size);
+		res = -ENOKEY;
+		goto out;
+	}
+	res = derive_key_aes(ctx->nonce, master_key, raw_key);
+out:
+	up_read(&keyring_key->sem);
+	key_put(keyring_key);
+	return res;
+}
+
+static const struct {
+	const char *cipher_str;
+	int keysize;
+} available_modes[] = {
+	[FS_ENCRYPTION_MODE_AES_256_XTS] = { "xts(aes)",
+					     FS_AES_256_XTS_KEY_SIZE },
+	[FS_ENCRYPTION_MODE_AES_256_CTS] = { "cts(cbc(aes))",
+					     FS_AES_256_CTS_KEY_SIZE },
+	[FS_ENCRYPTION_MODE_AES_128_CBC] = { "cbc(aes)",
+					     FS_AES_128_CBC_KEY_SIZE },
+	[FS_ENCRYPTION_MODE_AES_128_CTS] = { "cts(cbc(aes))",
+					     FS_AES_128_CTS_KEY_SIZE },
+};
+
+static int determine_cipher_type(struct fscrypt_info *ci, struct inode *inode,
+				 const char **cipher_str_ret, int *keysize_ret)
+{
+	u32 mode;
+
+	if (!fscrypt_valid_enc_modes(ci->ci_data_mode, ci->ci_filename_mode)) {
+		pr_warn_ratelimited("fscrypt: inode %lu uses unsupported encryption modes (contents mode %d, filenames mode %d)\n",
+				    inode->i_ino,
+				    ci->ci_data_mode, ci->ci_filename_mode);
+		return -EINVAL;
+	}
+
+	if (S_ISREG(inode->i_mode)) {
+		mode = ci->ci_data_mode;
+	} else if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) {
+		mode = ci->ci_filename_mode;
+	} else {
+		WARN_ONCE(1, "fscrypt: filesystem tried to load encryption info for inode %lu, which is not encryptable (file type %d)\n",
+			  inode->i_ino, (inode->i_mode & S_IFMT));
+		return -EINVAL;
+	}
+
+	*cipher_str_ret = available_modes[mode].cipher_str;
+	*keysize_ret = available_modes[mode].keysize;
+	return 0;
+}
+
+static void put_crypt_info(struct fscrypt_info *ci)
+{
+	if (!ci)
+		return;
+
+	crypto_free_skcipher(ci->ci_ctfm);
+	crypto_free_cipher(ci->ci_essiv_tfm);
+	kmem_cache_free(fscrypt_info_cachep, ci);
+}
+
+static int derive_essiv_salt(const u8 *key, int keysize, u8 *salt)
+{
+	struct crypto_shash *tfm = READ_ONCE(essiv_hash_tfm);
+
+	/* init hash transform on demand */
+	if (unlikely(!tfm)) {
+		struct crypto_shash *prev_tfm;
+
+		tfm = crypto_alloc_shash("sha256", 0, 0);
+		if (IS_ERR(tfm)) {
+			pr_warn_ratelimited("fscrypt: error allocating SHA-256 transform: %ld\n",
+					    PTR_ERR(tfm));
+			return PTR_ERR(tfm);
+		}
+		prev_tfm = cmpxchg(&essiv_hash_tfm, NULL, tfm);
+		if (prev_tfm) {
+			crypto_free_shash(tfm);
+			tfm = prev_tfm;
+		}
+	}
+
+	{
+		SHASH_DESC_ON_STACK(desc, tfm);
+		desc->tfm = tfm;
+		desc->flags = 0;
+
+		return crypto_shash_digest(desc, key, keysize, salt);
+	}
+}
+
+static int init_essiv_generator(struct fscrypt_info *ci, const u8 *raw_key,
+				int keysize)
+{
+	int err;
+	struct crypto_cipher *essiv_tfm;
+	u8 salt[SHA256_DIGEST_SIZE];
+
+	essiv_tfm = crypto_alloc_cipher("aes", 0, 0);
+	if (IS_ERR(essiv_tfm))
+		return PTR_ERR(essiv_tfm);
+
+	ci->ci_essiv_tfm = essiv_tfm;
+
+	err = derive_essiv_salt(raw_key, keysize, salt);
+	if (err)
+		goto out;
+
+	/*
+	 * Using SHA256 to derive the salt/key will result in AES-256 being
+	 * used for IV generation. File contents encryption will still use the
+	 * configured keysize (AES-128) nevertheless.
+	 */
+	err = crypto_cipher_setkey(essiv_tfm, salt, sizeof(salt));
+	if (err)
+		goto out;
+
+out:
+	memzero_explicit(salt, sizeof(salt));
+	return err;
+}
+
+void __exit fscrypt_essiv_cleanup(void)
+{
+	crypto_free_shash(essiv_hash_tfm);
+}
+
+int fscrypt_get_encryption_info(struct inode *inode)
+{
+	struct fscrypt_info *crypt_info;
+	struct fscrypt_context ctx;
+	struct crypto_skcipher *ctfm;
+	const char *cipher_str;
+	int keysize;
+	u8 *raw_key = NULL;
+	int res;
+
+	if (inode->i_crypt_info)
+		return 0;
+
+	res = fscrypt_initialize(inode->i_sb->s_cop->flags);
+	if (res)
+		return res;
+
+	res = inode->i_sb->s_cop->get_context(inode, &ctx, sizeof(ctx));
+	if (res < 0) {
+		if (!fscrypt_dummy_context_enabled(inode) ||
+		    inode->i_sb->s_cop->is_encrypted(inode))
+			return res;
+		/* Fake up a context for an unencrypted directory */
+		memset(&ctx, 0, sizeof(ctx));
+		ctx.format = FS_ENCRYPTION_CONTEXT_FORMAT_V1;
+		ctx.contents_encryption_mode = FS_ENCRYPTION_MODE_AES_256_XTS;
+		ctx.filenames_encryption_mode = FS_ENCRYPTION_MODE_AES_256_CTS;
+		memset(ctx.master_key_descriptor, 0x42, FS_KEY_DESCRIPTOR_SIZE);
+	} else if (res != sizeof(ctx)) {
+		return -EINVAL;
+	}
+
+	if (ctx.format != FS_ENCRYPTION_CONTEXT_FORMAT_V1)
+		return -EINVAL;
+
+	if (ctx.flags & ~FS_POLICY_FLAGS_VALID)
+		return -EINVAL;
+
+	crypt_info = kmem_cache_alloc(fscrypt_info_cachep, GFP_NOFS);
+	if (!crypt_info)
+		return -ENOMEM;
+
+	crypt_info->ci_flags = ctx.flags;
+	crypt_info->ci_data_mode = ctx.contents_encryption_mode;
+	crypt_info->ci_filename_mode = ctx.filenames_encryption_mode;
+	crypt_info->ci_ctfm = NULL;
+	crypt_info->ci_essiv_tfm = NULL;
+	memcpy(crypt_info->ci_master_key, ctx.master_key_descriptor,
+				sizeof(crypt_info->ci_master_key));
+
+	res = determine_cipher_type(crypt_info, inode, &cipher_str, &keysize);
+	if (res)
+		goto out;
+
+	/*
+	 * This cannot be a stack buffer because it is passed to the scatterlist
+	 * crypto API as part of key derivation.
+	 */
+	res = -ENOMEM;
+	raw_key = kmalloc(FS_MAX_KEY_SIZE, GFP_NOFS);
+	if (!raw_key)
+		goto out;
+
+	res = validate_user_key(crypt_info, &ctx, raw_key, FS_KEY_DESC_PREFIX,
+				keysize);
+	if (res && inode->i_sb->s_cop->key_prefix) {
+		int res2 = validate_user_key(crypt_info, &ctx, raw_key,
+					     inode->i_sb->s_cop->key_prefix,
+					     keysize);
+		if (res2) {
+			if (res2 == -ENOKEY)
+				res = -ENOKEY;
+			goto out;
+		}
+	} else if (res) {
+		goto out;
+	}
+	ctfm = crypto_alloc_skcipher(cipher_str, 0, 0);
+	if (!ctfm || IS_ERR(ctfm)) {
+		res = ctfm ? PTR_ERR(ctfm) : -ENOMEM;
+		pr_debug("%s: error %d (inode %lu) allocating crypto tfm\n",
+			 __func__, res, inode->i_ino);
+		goto out;
+	}
+	crypt_info->ci_ctfm = ctfm;
+	crypto_skcipher_clear_flags(ctfm, ~0);
+	crypto_skcipher_set_flags(ctfm, CRYPTO_TFM_REQ_WEAK_KEY);
+	/*
+	 * if the provided key is longer than keysize, we use the first
+	 * keysize bytes of the derived key only
+	 */
+	res = crypto_skcipher_setkey(ctfm, raw_key, keysize);
+	if (res)
+		goto out;
+
+	if (S_ISREG(inode->i_mode) &&
+	    crypt_info->ci_data_mode == FS_ENCRYPTION_MODE_AES_128_CBC) {
+		res = init_essiv_generator(crypt_info, raw_key, keysize);
+		if (res) {
+			pr_debug("%s: error %d (inode %lu) allocating essiv tfm\n",
+				 __func__, res, inode->i_ino);
+			goto out;
+		}
+	}
+	if (cmpxchg(&inode->i_crypt_info, NULL, crypt_info) == NULL)
+		crypt_info = NULL;
+out:
+	if (res == -ENOKEY)
+		res = 0;
+	put_crypt_info(crypt_info);
+	kzfree(raw_key);
+	return res;
+}
+EXPORT_SYMBOL(fscrypt_get_encryption_info);
+
+void fscrypt_put_encryption_info(struct inode *inode, struct fscrypt_info *ci)
+{
+	struct fscrypt_info *prev;
+
+	if (ci == NULL)
+		ci = ACCESS_ONCE(inode->i_crypt_info);
+	if (ci == NULL)
+		return;
+
+	prev = cmpxchg(&inode->i_crypt_info, ci, NULL);
+	if (prev != ci)
+		return;
+
+	put_crypt_info(ci);
+}
+EXPORT_SYMBOL(fscrypt_put_encryption_info);
diff --git a/fs/crypto/policy.c b/fs/crypto/policy.c
new file mode 100644
index 000000000000..9914d51dff86
--- /dev/null
+++ b/fs/crypto/policy.c
@@ -0,0 +1,265 @@
+/*
+ * Encryption policy functions for per-file encryption support.
+ *
+ * Copyright (C) 2015, Google, Inc.
+ * Copyright (C) 2015, Motorola Mobility.
+ *
+ * Written by Michael Halcrow, 2015.
+ * Modified by Jaegeuk Kim, 2015.
+ */
+
+#include <linux/random.h>
+#include <linux/string.h>
+#include <linux/mount.h>
+#include "fscrypt_private.h"
+
+/*
+ * check whether an encryption policy is consistent with an encryption context
+ */
+static bool is_encryption_context_consistent_with_policy(
+				const struct fscrypt_context *ctx,
+				const struct fscrypt_policy *policy)
+{
+	return memcmp(ctx->master_key_descriptor, policy->master_key_descriptor,
+		      FS_KEY_DESCRIPTOR_SIZE) == 0 &&
+		(ctx->flags == policy->flags) &&
+		(ctx->contents_encryption_mode ==
+		 policy->contents_encryption_mode) &&
+		(ctx->filenames_encryption_mode ==
+		 policy->filenames_encryption_mode);
+}
+
+static int create_encryption_context_from_policy(struct inode *inode,
+				const struct fscrypt_policy *policy)
+{
+	struct fscrypt_context ctx;
+
+	ctx.format = FS_ENCRYPTION_CONTEXT_FORMAT_V1;
+	memcpy(ctx.master_key_descriptor, policy->master_key_descriptor,
+					FS_KEY_DESCRIPTOR_SIZE);
+
+	if (!fscrypt_valid_enc_modes(policy->contents_encryption_mode,
+				     policy->filenames_encryption_mode))
+		return -EINVAL;
+
+	if (policy->flags & ~FS_POLICY_FLAGS_VALID)
+		return -EINVAL;
+
+	ctx.contents_encryption_mode = policy->contents_encryption_mode;
+	ctx.filenames_encryption_mode = policy->filenames_encryption_mode;
+	ctx.flags = policy->flags;
+	BUILD_BUG_ON(sizeof(ctx.nonce) != FS_KEY_DERIVATION_NONCE_SIZE);
+	get_random_bytes(ctx.nonce, FS_KEY_DERIVATION_NONCE_SIZE);
+
+	return inode->i_sb->s_cop->set_context(inode, &ctx, sizeof(ctx), NULL);
+}
+
+int fscrypt_ioctl_set_policy(struct file *filp, const void __user *arg)
+{
+	struct fscrypt_policy policy;
+	struct inode *inode = file_inode(filp);
+	int ret;
+	struct fscrypt_context ctx;
+
+	if (copy_from_user(&policy, arg, sizeof(policy)))
+		return -EFAULT;
+
+	if (!inode_owner_or_capable(inode))
+		return -EACCES;
+
+	if (policy.version != 0)
+		return -EINVAL;
+
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	inode_lock(inode);
+
+	ret = inode->i_sb->s_cop->get_context(inode, &ctx, sizeof(ctx));
+	if (ret == -ENODATA) {
+		if (!S_ISDIR(inode->i_mode))
+			ret = -ENOTDIR;
+		else if (!inode->i_sb->s_cop->empty_dir(inode))
+			ret = -ENOTEMPTY;
+		else
+			ret = create_encryption_context_from_policy(inode,
+								    &policy);
+	} else if (ret == sizeof(ctx) &&
+		   is_encryption_context_consistent_with_policy(&ctx,
+								&policy)) {
+		/* The file already uses the same encryption policy. */
+		ret = 0;
+	} else if (ret >= 0 || ret == -ERANGE) {
+		/* The file already uses a different encryption policy. */
+		ret = -EEXIST;
+	}
+
+	inode_unlock(inode);
+
+	mnt_drop_write_file(filp);
+	return ret;
+}
+EXPORT_SYMBOL(fscrypt_ioctl_set_policy);
+
+int fscrypt_ioctl_get_policy(struct file *filp, void __user *arg)
+{
+	struct inode *inode = file_inode(filp);
+	struct fscrypt_context ctx;
+	struct fscrypt_policy policy;
+	int res;
+
+	if (!inode->i_sb->s_cop->is_encrypted(inode))
+		return -ENODATA;
+
+	res = inode->i_sb->s_cop->get_context(inode, &ctx, sizeof(ctx));
+	if (res < 0 && res != -ERANGE)
+		return res;
+	if (res != sizeof(ctx))
+		return -EINVAL;
+	if (ctx.format != FS_ENCRYPTION_CONTEXT_FORMAT_V1)
+		return -EINVAL;
+
+	policy.version = 0;
+	policy.contents_encryption_mode = ctx.contents_encryption_mode;
+	policy.filenames_encryption_mode = ctx.filenames_encryption_mode;
+	policy.flags = ctx.flags;
+	memcpy(policy.master_key_descriptor, ctx.master_key_descriptor,
+				FS_KEY_DESCRIPTOR_SIZE);
+
+	if (copy_to_user(arg, &policy, sizeof(policy)))
+		return -EFAULT;
+	return 0;
+}
+EXPORT_SYMBOL(fscrypt_ioctl_get_policy);
+
+/**
+ * fscrypt_has_permitted_context() - is a file's encryption policy permitted
+ *				     within its directory?
+ *
+ * @parent: inode for parent directory
+ * @child: inode for file being looked up, opened, or linked into @parent
+ *
+ * Filesystems must call this before permitting access to an inode in a
+ * situation where the parent directory is encrypted (either before allowing
+ * ->lookup() to succeed, or for a regular file before allowing it to be opened)
+ * and before any operation that involves linking an inode into an encrypted
+ * directory, including link, rename, and cross rename.  It enforces the
+ * constraint that within a given encrypted directory tree, all files use the
+ * same encryption policy.  The pre-access check is needed to detect potentially
+ * malicious offline violations of this constraint, while the link and rename
+ * checks are needed to prevent online violations of this constraint.
+ *
+ * Return: 1 if permitted, 0 if forbidden.  If forbidden, the caller must fail
+ * the filesystem operation with EPERM.
+ */
+int fscrypt_has_permitted_context(struct inode *parent, struct inode *child)
+{
+	const struct fscrypt_operations *cops = parent->i_sb->s_cop;
+	const struct fscrypt_info *parent_ci, *child_ci;
+	struct fscrypt_context parent_ctx, child_ctx;
+	int res;
+
+	/* No restrictions on file types which are never encrypted */
+	if (!S_ISREG(child->i_mode) && !S_ISDIR(child->i_mode) &&
+	    !S_ISLNK(child->i_mode))
+		return 1;
+
+	/* No restrictions if the parent directory is unencrypted */
+	if (!cops->is_encrypted(parent))
+		return 1;
+
+	/* Encrypted directories must not contain unencrypted files */
+	if (!cops->is_encrypted(child))
+		return 0;
+
+	/*
+	 * Both parent and child are encrypted, so verify they use the same
+	 * encryption policy.  Compare the fscrypt_info structs if the keys are
+	 * available, otherwise retrieve and compare the fscrypt_contexts.
+	 *
+	 * Note that the fscrypt_context retrieval will be required frequently
+	 * when accessing an encrypted directory tree without the key.
+	 * Performance-wise this is not a big deal because we already don't
+	 * really optimize for file access without the key (to the extent that
+	 * such access is even possible), given that any attempted access
+	 * already causes a fscrypt_context retrieval and keyring search.
+	 *
+	 * In any case, if an unexpected error occurs, fall back to "forbidden".
+	 */
+
+	res = fscrypt_get_encryption_info(parent);
+	if (res)
+		return 0;
+	res = fscrypt_get_encryption_info(child);
+	if (res)
+		return 0;
+	parent_ci = parent->i_crypt_info;
+	child_ci = child->i_crypt_info;
+
+	if (parent_ci && child_ci) {
+		return memcmp(parent_ci->ci_master_key, child_ci->ci_master_key,
+			      FS_KEY_DESCRIPTOR_SIZE) == 0 &&
+			(parent_ci->ci_data_mode == child_ci->ci_data_mode) &&
+			(parent_ci->ci_filename_mode ==
+			 child_ci->ci_filename_mode) &&
+			(parent_ci->ci_flags == child_ci->ci_flags);
+	}
+
+	res = cops->get_context(parent, &parent_ctx, sizeof(parent_ctx));
+	if (res != sizeof(parent_ctx))
+		return 0;
+
+	res = cops->get_context(child, &child_ctx, sizeof(child_ctx));
+	if (res != sizeof(child_ctx))
+		return 0;
+
+	return memcmp(parent_ctx.master_key_descriptor,
+		      child_ctx.master_key_descriptor,
+		      FS_KEY_DESCRIPTOR_SIZE) == 0 &&
+		(parent_ctx.contents_encryption_mode ==
+		 child_ctx.contents_encryption_mode) &&
+		(parent_ctx.filenames_encryption_mode ==
+		 child_ctx.filenames_encryption_mode) &&
+		(parent_ctx.flags == child_ctx.flags);
+}
+EXPORT_SYMBOL(fscrypt_has_permitted_context);
+
+/**
+ * fscrypt_inherit_context() - Sets a child context from its parent
+ * @parent: Parent inode from which the context is inherited.
+ * @child:  Child inode that inherits the context from @parent.
+ * @fs_data:  private data given by FS.
+ * @preload:  preload child i_crypt_info if true
+ *
+ * Return: 0 on success, -errno on failure
+ */
+int fscrypt_inherit_context(struct inode *parent, struct inode *child,
+						void *fs_data, bool preload)
+{
+	struct fscrypt_context ctx;
+	struct fscrypt_info *ci;
+	int res;
+
+	res = fscrypt_get_encryption_info(parent);
+	if (res < 0)
+		return res;
+
+	ci = parent->i_crypt_info;
+	if (ci == NULL)
+		return -ENOKEY;
+
+	ctx.format = FS_ENCRYPTION_CONTEXT_FORMAT_V1;
+	ctx.contents_encryption_mode = ci->ci_data_mode;
+	ctx.filenames_encryption_mode = ci->ci_filename_mode;
+	ctx.flags = ci->ci_flags;
+	memcpy(ctx.master_key_descriptor, ci->ci_master_key,
+	       FS_KEY_DESCRIPTOR_SIZE);
+	get_random_bytes(ctx.nonce, FS_KEY_DERIVATION_NONCE_SIZE);
+	res = parent->i_sb->s_cop->set_context(child, &ctx,
+						sizeof(ctx), fs_data);
+	if (res)
+		return res;
+	return preload ? fscrypt_get_encryption_info(child): 0;
+}
+EXPORT_SYMBOL(fscrypt_inherit_context);
diff --git a/fs/dcache.c b/fs/dcache.c
index 3ed642e0a0c2..5bf7b4a188e9 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -3054,6 +3054,7 @@ char *d_absolute_path(const struct path *path,
 		return ERR_PTR(error);
 	return res;
 }
+EXPORT_SYMBOL(d_absolute_path);
 
 /*
  * same as __d_path but appends "(deleted)" for unlinked files.
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 1b08556776ce..066df649a6b0 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -34,6 +34,7 @@
 #include <linux/mutex.h>
 #include <linux/anon_inodes.h>
 #include <linux/device.h>
+#include <linux/freezer.h>
 #include <asm/uaccess.h>
 #include <asm/io.h>
 #include <asm/mman.h>
@@ -1598,7 +1599,7 @@ static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
 {
 	int res = 0, eavail, timed_out = 0;
 	unsigned long flags;
-	long slack = 0;
+	u64 slack = 0;
 	wait_queue_t wait;
 	ktime_t expires, *to = NULL;
 
@@ -1645,7 +1646,8 @@ fetch_events:
 			}
 
 			spin_unlock_irqrestore(&ep->lock, flags);
-			if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS))
+			if (!freezable_schedule_hrtimeout_range(to, slack,
+								HRTIMER_MODE_ABS))
 				timed_out = 1;
 
 			spin_lock_irqsave(&ep->lock, flags);
diff --git a/fs/exec.c b/fs/exec.c
index 9c5ee2a880aa..dd3a59420506 100644
--- a/fs/exec.c
+++ b/fs/exec.c
@@ -1153,7 +1153,7 @@ EXPORT_SYMBOL(flush_old_exec);
 void would_dump(struct linux_binprm *bprm, struct file *file)
 {
 	struct inode *inode = file_inode(file);
-	if (inode_permission(inode, MAY_READ) < 0) {
+	if (inode_permission2(file->f_path.mnt, inode, MAY_READ) < 0) {
 		struct user_namespace *old, *user_ns;
 		bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP;
 
diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h
index 4c69c94cafd8..f98ce7e60a0f 100644
--- a/fs/ext2/ext2.h
+++ b/fs/ext2/ext2.h
@@ -61,6 +61,8 @@ struct ext2_block_alloc_info {
 #define rsv_start rsv_window._rsv_start
 #define rsv_end rsv_window._rsv_end
 
+struct mb2_cache;
+
 /*
  * second extended-fs super-block data in memory
  */
@@ -111,6 +113,7 @@ struct ext2_sb_info {
 	 * of the mount options.
 	 */
 	spinlock_t s_lock;
+	struct mb2_cache *s_mb_cache;
 };
 
 static inline spinlock_t *
diff --git a/fs/ext2/super.c b/fs/ext2/super.c
index 748d35afc902..111a31761ffa 100644
--- a/fs/ext2/super.c
+++ b/fs/ext2/super.c
@@ -131,7 +131,10 @@ static void ext2_put_super (struct super_block * sb)
 
 	dquot_disable(sb, -1, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED);
 
-	ext2_xattr_put_super(sb);
+	if (sbi->s_mb_cache) {
+		ext2_xattr_destroy_cache(sbi->s_mb_cache);
+		sbi->s_mb_cache = NULL;
+	}
 	if (!(sb->s_flags & MS_RDONLY)) {
 		struct ext2_super_block *es = sbi->s_es;
 
@@ -1104,6 +1107,14 @@ static int ext2_fill_super(struct super_block *sb, void *data, int silent)
 		ext2_msg(sb, KERN_ERR, "error: insufficient memory");
 		goto failed_mount3;
 	}
+
+#ifdef CONFIG_EXT2_FS_XATTR
+	sbi->s_mb_cache = ext2_xattr_create_cache();
+	if (!sbi->s_mb_cache) {
+		ext2_msg(sb, KERN_ERR, "Failed to create an mb_cache");
+		goto failed_mount3;
+	}
+#endif
 	/*
 	 * set up enough so that it can read an inode
 	 */
@@ -1149,6 +1160,8 @@ cantfind_ext2:
 			sb->s_id);
 	goto failed_mount;
 failed_mount3:
+	if (sbi->s_mb_cache)
+		ext2_xattr_destroy_cache(sbi->s_mb_cache);
 	percpu_counter_destroy(&sbi->s_freeblocks_counter);
 	percpu_counter_destroy(&sbi->s_freeinodes_counter);
 	percpu_counter_destroy(&sbi->s_dirs_counter);
@@ -1555,20 +1568,17 @@ MODULE_ALIAS_FS("ext2");
 
 static int __init init_ext2_fs(void)
 {
-	int err = init_ext2_xattr();
-	if (err)
-		return err;
+	int err;
+
 	err = init_inodecache();
 	if (err)
-		goto out1;
+		return err;
         err = register_filesystem(&ext2_fs_type);
 	if (err)
 		goto out;
 	return 0;
 out:
 	destroy_inodecache();
-out1:
-	exit_ext2_xattr();
 	return err;
 }
 
@@ -1576,7 +1586,6 @@ static void __exit exit_ext2_fs(void)
 {
 	unregister_filesystem(&ext2_fs_type);
 	destroy_inodecache();
-	exit_ext2_xattr();
 }
 
 MODULE_AUTHOR("Remy Card and others");
diff --git a/fs/ext2/xattr.c b/fs/ext2/xattr.c
index fa70848afa8f..24736c8b3d51 100644
--- a/fs/ext2/xattr.c
+++ b/fs/ext2/xattr.c
@@ -56,7 +56,7 @@
 #include <linux/buffer_head.h>
 #include <linux/init.h>
 #include <linux/slab.h>
-#include <linux/mbcache.h>
+#include <linux/mbcache2.h>
 #include <linux/quotaops.h>
 #include <linux/rwsem.h>
 #include <linux/security.h>
@@ -92,14 +92,12 @@
 static int ext2_xattr_set2(struct inode *, struct buffer_head *,
 			   struct ext2_xattr_header *);
 
-static int ext2_xattr_cache_insert(struct buffer_head *);
+static int ext2_xattr_cache_insert(struct mb2_cache *, struct buffer_head *);
 static struct buffer_head *ext2_xattr_cache_find(struct inode *,
 						 struct ext2_xattr_header *);
 static void ext2_xattr_rehash(struct ext2_xattr_header *,
 			      struct ext2_xattr_entry *);
 
-static struct mb_cache *ext2_xattr_cache;
-
 static const struct xattr_handler *ext2_xattr_handler_map[] = {
 	[EXT2_XATTR_INDEX_USER]		     = &ext2_xattr_user_handler,
 #ifdef CONFIG_EXT2_FS_POSIX_ACL
@@ -154,6 +152,7 @@ ext2_xattr_get(struct inode *inode, int name_index, const char *name,
 	size_t name_len, size;
 	char *end;
 	int error;
+	struct mb2_cache *ext2_mb_cache = EXT2_SB(inode->i_sb)->s_mb_cache;
 
 	ea_idebug(inode, "name=%d.%s, buffer=%p, buffer_size=%ld",
 		  name_index, name, buffer, (long)buffer_size);
@@ -198,7 +197,7 @@ bad_block:	ext2_error(inode->i_sb, "ext2_xattr_get",
 			goto found;
 		entry = next;
 	}
-	if (ext2_xattr_cache_insert(bh))
+	if (ext2_xattr_cache_insert(ext2_mb_cache, bh))
 		ea_idebug(inode, "cache insert failed");
 	error = -ENODATA;
 	goto cleanup;
@@ -211,7 +210,7 @@ found:
 	    le16_to_cpu(entry->e_value_offs) + size > inode->i_sb->s_blocksize)
 		goto bad_block;
 
-	if (ext2_xattr_cache_insert(bh))
+	if (ext2_xattr_cache_insert(ext2_mb_cache, bh))
 		ea_idebug(inode, "cache insert failed");
 	if (buffer) {
 		error = -ERANGE;
@@ -249,6 +248,7 @@ ext2_xattr_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 	char *end;
 	size_t rest = buffer_size;
 	int error;
+	struct mb2_cache *ext2_mb_cache = EXT2_SB(inode->i_sb)->s_mb_cache;
 
 	ea_idebug(inode, "buffer=%p, buffer_size=%ld",
 		  buffer, (long)buffer_size);
@@ -283,7 +283,7 @@ bad_block:	ext2_error(inode->i_sb, "ext2_xattr_list",
 			goto bad_block;
 		entry = next;
 	}
-	if (ext2_xattr_cache_insert(bh))
+	if (ext2_xattr_cache_insert(ext2_mb_cache, bh))
 		ea_idebug(inode, "cache insert failed");
 
 	/* list the attribute names */
@@ -480,22 +480,23 @@ bad_block:		ext2_error(sb, "ext2_xattr_set",
 	/* Here we know that we can set the new attribute. */
 
 	if (header) {
-		struct mb_cache_entry *ce;
-
 		/* assert(header == HDR(bh)); */
-		ce = mb_cache_entry_get(ext2_xattr_cache, bh->b_bdev,
-					bh->b_blocknr);
 		lock_buffer(bh);
 		if (header->h_refcount == cpu_to_le32(1)) {
+			__u32 hash = le32_to_cpu(header->h_hash);
+
 			ea_bdebug(bh, "modifying in-place");
-			if (ce)
-				mb_cache_entry_free(ce);
+			/*
+			 * This must happen under buffer lock for
+			 * ext2_xattr_set2() to reliably detect modified block
+			 */
+			mb2_cache_entry_delete_block(EXT2_SB(sb)->s_mb_cache,
+						     hash, bh->b_blocknr);
+
 			/* keep the buffer locked while modifying it. */
 		} else {
 			int offset;
 
-			if (ce)
-				mb_cache_entry_release(ce);
 			unlock_buffer(bh);
 			ea_bdebug(bh, "cloning");
 			header = kmalloc(bh->b_size, GFP_KERNEL);
@@ -623,6 +624,7 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh,
 	struct super_block *sb = inode->i_sb;
 	struct buffer_head *new_bh = NULL;
 	int error;
+	struct mb2_cache *ext2_mb_cache = EXT2_SB(sb)->s_mb_cache;
 
 	if (header) {
 		new_bh = ext2_xattr_cache_find(inode, header);
@@ -650,7 +652,7 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh,
 			   don't need to change the reference count. */
 			new_bh = old_bh;
 			get_bh(new_bh);
-			ext2_xattr_cache_insert(new_bh);
+			ext2_xattr_cache_insert(ext2_mb_cache, new_bh);
 		} else {
 			/* We need to allocate a new block */
 			ext2_fsblk_t goal = ext2_group_first_block_no(sb,
@@ -671,7 +673,7 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh,
 			memcpy(new_bh->b_data, header, new_bh->b_size);
 			set_buffer_uptodate(new_bh);
 			unlock_buffer(new_bh);
-			ext2_xattr_cache_insert(new_bh);
+			ext2_xattr_cache_insert(ext2_mb_cache, new_bh);
 			
 			ext2_xattr_update_super_block(sb);
 		}
@@ -704,19 +706,21 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh,
 
 	error = 0;
 	if (old_bh && old_bh != new_bh) {
-		struct mb_cache_entry *ce;
-
 		/*
 		 * If there was an old block and we are no longer using it,
 		 * release the old block.
 		 */
-		ce = mb_cache_entry_get(ext2_xattr_cache, old_bh->b_bdev,
-					old_bh->b_blocknr);
 		lock_buffer(old_bh);
 		if (HDR(old_bh)->h_refcount == cpu_to_le32(1)) {
+			__u32 hash = le32_to_cpu(HDR(old_bh)->h_hash);
+
+			/*
+			 * This must happen under buffer lock for
+			 * ext2_xattr_set2() to reliably detect freed block
+			 */
+			mb2_cache_entry_delete_block(ext2_mb_cache,
+						     hash, old_bh->b_blocknr);
 			/* Free the old block. */
-			if (ce)
-				mb_cache_entry_free(ce);
 			ea_bdebug(old_bh, "freeing");
 			ext2_free_blocks(inode, old_bh->b_blocknr, 1);
 			mark_inode_dirty(inode);
@@ -727,8 +731,6 @@ ext2_xattr_set2(struct inode *inode, struct buffer_head *old_bh,
 		} else {
 			/* Decrement the refcount only. */
 			le32_add_cpu(&HDR(old_bh)->h_refcount, -1);
-			if (ce)
-				mb_cache_entry_release(ce);
 			dquot_free_block_nodirty(inode, 1);
 			mark_inode_dirty(inode);
 			mark_buffer_dirty(old_bh);
@@ -754,7 +756,6 @@ void
 ext2_xattr_delete_inode(struct inode *inode)
 {
 	struct buffer_head *bh = NULL;
-	struct mb_cache_entry *ce;
 
 	down_write(&EXT2_I(inode)->xattr_sem);
 	if (!EXT2_I(inode)->i_file_acl)
@@ -774,19 +775,22 @@ ext2_xattr_delete_inode(struct inode *inode)
 			EXT2_I(inode)->i_file_acl);
 		goto cleanup;
 	}
-	ce = mb_cache_entry_get(ext2_xattr_cache, bh->b_bdev, bh->b_blocknr);
 	lock_buffer(bh);
 	if (HDR(bh)->h_refcount == cpu_to_le32(1)) {
-		if (ce)
-			mb_cache_entry_free(ce);
+		__u32 hash = le32_to_cpu(HDR(bh)->h_hash);
+
+		/*
+		 * This must happen under buffer lock for ext2_xattr_set2() to
+		 * reliably detect freed block
+		 */
+		mb2_cache_entry_delete_block(EXT2_SB(inode->i_sb)->s_mb_cache,
+					     hash, bh->b_blocknr);
 		ext2_free_blocks(inode, EXT2_I(inode)->i_file_acl, 1);
 		get_bh(bh);
 		bforget(bh);
 		unlock_buffer(bh);
 	} else {
 		le32_add_cpu(&HDR(bh)->h_refcount, -1);
-		if (ce)
-			mb_cache_entry_release(ce);
 		ea_bdebug(bh, "refcount now=%d",
 			le32_to_cpu(HDR(bh)->h_refcount));
 		unlock_buffer(bh);
@@ -803,18 +807,6 @@ cleanup:
 }
 
 /*
- * ext2_xattr_put_super()
- *
- * This is called when a file system is unmounted.
- */
-void
-ext2_xattr_put_super(struct super_block *sb)
-{
-	mb_cache_shrink(sb->s_bdev);
-}
-
-
-/*
  * ext2_xattr_cache_insert()
  *
  * Create a new entry in the extended attribute cache, and insert
@@ -823,28 +815,20 @@ ext2_xattr_put_super(struct super_block *sb)
  * Returns 0, or a negative error number on failure.
  */
 static int
-ext2_xattr_cache_insert(struct buffer_head *bh)
+ext2_xattr_cache_insert(struct mb2_cache *cache, struct buffer_head *bh)
 {
 	__u32 hash = le32_to_cpu(HDR(bh)->h_hash);
-	struct mb_cache_entry *ce;
 	int error;
 
-	ce = mb_cache_entry_alloc(ext2_xattr_cache, GFP_NOFS);
-	if (!ce)
-		return -ENOMEM;
-	error = mb_cache_entry_insert(ce, bh->b_bdev, bh->b_blocknr, hash);
+	error = mb2_cache_entry_create(cache, GFP_NOFS, hash, bh->b_blocknr);
 	if (error) {
-		mb_cache_entry_free(ce);
 		if (error == -EBUSY) {
 			ea_bdebug(bh, "already in cache (%d cache entries)",
 				atomic_read(&ext2_xattr_cache->c_entry_count));
 			error = 0;
 		}
-	} else {
-		ea_bdebug(bh, "inserting [%x] (%d cache entries)", (int)hash,
-			  atomic_read(&ext2_xattr_cache->c_entry_count));
-		mb_cache_entry_release(ce);
-	}
+	} else
+		ea_bdebug(bh, "inserting [%x]", (int)hash);
 	return error;
 }
 
@@ -900,23 +884,17 @@ static struct buffer_head *
 ext2_xattr_cache_find(struct inode *inode, struct ext2_xattr_header *header)
 {
 	__u32 hash = le32_to_cpu(header->h_hash);
-	struct mb_cache_entry *ce;
+	struct mb2_cache_entry *ce;
+	struct mb2_cache *ext2_mb_cache = EXT2_SB(inode->i_sb)->s_mb_cache;
 
 	if (!header->h_hash)
 		return NULL;  /* never share */
 	ea_idebug(inode, "looking for cached blocks [%x]", (int)hash);
 again:
-	ce = mb_cache_entry_find_first(ext2_xattr_cache, inode->i_sb->s_bdev,
-				       hash);
+	ce = mb2_cache_entry_find_first(ext2_mb_cache, hash);
 	while (ce) {
 		struct buffer_head *bh;
 
-		if (IS_ERR(ce)) {
-			if (PTR_ERR(ce) == -EAGAIN)
-				goto again;
-			break;
-		}
-
 		bh = sb_bread(inode->i_sb, ce->e_block);
 		if (!bh) {
 			ext2_error(inode->i_sb, "ext2_xattr_cache_find",
@@ -924,7 +902,21 @@ again:
 				inode->i_ino, (unsigned long) ce->e_block);
 		} else {
 			lock_buffer(bh);
-			if (le32_to_cpu(HDR(bh)->h_refcount) >
+			/*
+			 * We have to be careful about races with freeing or
+			 * rehashing of xattr block. Once we hold buffer lock
+			 * xattr block's state is stable so we can check
+			 * whether the block got freed / rehashed or not.
+			 * Since we unhash mbcache entry under buffer lock when
+			 * freeing / rehashing xattr block, checking whether
+			 * entry is still hashed is reliable.
+			 */
+			if (hlist_bl_unhashed(&ce->e_hash_list)) {
+				mb2_cache_entry_put(ext2_mb_cache, ce);
+				unlock_buffer(bh);
+				brelse(bh);
+				goto again;
+			} else if (le32_to_cpu(HDR(bh)->h_refcount) >
 				   EXT2_XATTR_REFCOUNT_MAX) {
 				ea_idebug(inode, "block %ld refcount %d>%d",
 					  (unsigned long) ce->e_block,
@@ -933,13 +925,14 @@ again:
 			} else if (!ext2_xattr_cmp(header, HDR(bh))) {
 				ea_bdebug(bh, "b_count=%d",
 					  atomic_read(&(bh->b_count)));
-				mb_cache_entry_release(ce);
+				mb2_cache_entry_touch(ext2_mb_cache, ce);
+				mb2_cache_entry_put(ext2_mb_cache, ce);
 				return bh;
 			}
 			unlock_buffer(bh);
 			brelse(bh);
 		}
-		ce = mb_cache_entry_find_next(ce, inode->i_sb->s_bdev, hash);
+		ce = mb2_cache_entry_find_next(ext2_mb_cache, ce);
 	}
 	return NULL;
 }
@@ -1012,17 +1005,15 @@ static void ext2_xattr_rehash(struct ext2_xattr_header *header,
 
 #undef BLOCK_HASH_SHIFT
 
-int __init
-init_ext2_xattr(void)
+#define HASH_BUCKET_BITS 10
+
+struct mb2_cache *ext2_xattr_create_cache(void)
 {
-	ext2_xattr_cache = mb_cache_create("ext2_xattr", 6);
-	if (!ext2_xattr_cache)
-		return -ENOMEM;
-	return 0;
+	return mb2_cache_create(HASH_BUCKET_BITS);
 }
 
-void
-exit_ext2_xattr(void)
+void ext2_xattr_destroy_cache(struct mb2_cache *cache)
 {
-	mb_cache_destroy(ext2_xattr_cache);
+	if (cache)
+		mb2_cache_destroy(cache);
 }
diff --git a/fs/ext2/xattr.h b/fs/ext2/xattr.h
index 60edf298644e..6ea38aa9563a 100644
--- a/fs/ext2/xattr.h
+++ b/fs/ext2/xattr.h
@@ -53,6 +53,8 @@ struct ext2_xattr_entry {
 #define EXT2_XATTR_SIZE(size) \
 	(((size) + EXT2_XATTR_ROUND) & ~EXT2_XATTR_ROUND)
 
+struct mb2_cache;
+
 # ifdef CONFIG_EXT2_FS_XATTR
 
 extern const struct xattr_handler ext2_xattr_user_handler;
@@ -65,10 +67,9 @@ extern int ext2_xattr_get(struct inode *, int, const char *, void *, size_t);
 extern int ext2_xattr_set(struct inode *, int, const char *, const void *, size_t, int);
 
 extern void ext2_xattr_delete_inode(struct inode *);
-extern void ext2_xattr_put_super(struct super_block *);
 
-extern int init_ext2_xattr(void);
-extern void exit_ext2_xattr(void);
+extern struct mb2_cache *ext2_xattr_create_cache(void);
+extern void ext2_xattr_destroy_cache(struct mb2_cache *cache);
 
 extern const struct xattr_handler *ext2_xattr_handlers[];
 
@@ -93,19 +94,7 @@ ext2_xattr_delete_inode(struct inode *inode)
 {
 }
 
-static inline void
-ext2_xattr_put_super(struct super_block *sb)
-{
-}
-
-static inline int
-init_ext2_xattr(void)
-{
-	return 0;
-}
-
-static inline void
-exit_ext2_xattr(void)
+static inline void ext2_xattr_destroy_cache(struct mb2_cache *cache)
 {
 }
 
diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig
index b46e9fc64196..3c8293215603 100644
--- a/fs/ext4/Kconfig
+++ b/fs/ext4/Kconfig
@@ -106,6 +106,7 @@ config EXT4_ENCRYPTION
 	select CRYPTO_ECB
 	select CRYPTO_XTS
 	select CRYPTO_CTS
+	select CRYPTO_HEH
 	select CRYPTO_CTR
 	select CRYPTO_SHA256
 	select KEYS
diff --git a/fs/ext4/crypto_fname.c b/fs/ext4/crypto_fname.c
index 2cfe3ffc276f..026716bdbbfc 100644
--- a/fs/ext4/crypto_fname.c
+++ b/fs/ext4/crypto_fname.c
@@ -44,7 +44,8 @@ static void ext4_dir_crypt_complete(struct crypto_async_request *req, int res)
 
 bool ext4_valid_filenames_enc_mode(uint32_t mode)
 {
-	return (mode == EXT4_ENCRYPTION_MODE_AES_256_CTS);
+	return (mode == EXT4_ENCRYPTION_MODE_AES_256_CTS ||
+		mode == EXT4_ENCRYPTION_MODE_AES_256_HEH);
 }
 
 static unsigned max_name_len(struct inode *inode)
diff --git a/fs/ext4/crypto_key.c b/fs/ext4/crypto_key.c
index 9a1bc638abce..b3550c716452 100644
--- a/fs/ext4/crypto_key.c
+++ b/fs/ext4/crypto_key.c
@@ -29,16 +29,16 @@ static void derive_crypt_complete(struct crypto_async_request *req, int rc)
 }
 
 /**
- * ext4_derive_key_aes() - Derive a key using AES-128-ECB
+ * ext4_derive_key_v1() - Derive a key using AES-128-ECB
  * @deriving_key: Encryption key used for derivation.
  * @source_key:   Source key to which to apply derivation.
  * @derived_key:  Derived key.
  *
- * Return: Zero on success; non-zero otherwise.
+ * Return: 0 on success, -errno on failure
  */
-static int ext4_derive_key_aes(char deriving_key[EXT4_AES_128_ECB_KEY_SIZE],
-			       char source_key[EXT4_AES_256_XTS_KEY_SIZE],
-			       char derived_key[EXT4_AES_256_XTS_KEY_SIZE])
+static int ext4_derive_key_v1(const char deriving_key[EXT4_AES_128_ECB_KEY_SIZE],
+			      const char source_key[EXT4_AES_256_XTS_KEY_SIZE],
+			      char derived_key[EXT4_AES_256_XTS_KEY_SIZE])
 {
 	int res = 0;
 	struct ablkcipher_request *req = NULL;
@@ -83,6 +83,91 @@ out:
 	return res;
 }
 
+/**
+ * ext4_derive_key_v2() - Derive a key non-reversibly
+ * @nonce: the nonce associated with the file
+ * @master_key: the master key referenced by the file
+ * @derived_key: (output) the resulting derived key
+ *
+ * This function computes the following:
+ *	 derived_key[0:127]   = AES-256-ENCRYPT(master_key[0:255], nonce)
+ *	 derived_key[128:255] = AES-256-ENCRYPT(master_key[0:255], nonce ^ 0x01)
+ *	 derived_key[256:383] = AES-256-ENCRYPT(master_key[256:511], nonce)
+ *	 derived_key[384:511] = AES-256-ENCRYPT(master_key[256:511], nonce ^ 0x01)
+ *
+ * 'nonce ^ 0x01' denotes flipping the low order bit of the last byte.
+ *
+ * Unlike the v1 algorithm, the v2 algorithm is "non-reversible", meaning that
+ * compromising a derived key does not also compromise the master key.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int ext4_derive_key_v2(const char nonce[EXT4_KEY_DERIVATION_NONCE_SIZE],
+			      const char master_key[EXT4_MAX_KEY_SIZE],
+			      char derived_key[EXT4_MAX_KEY_SIZE])
+{
+	const int noncelen = EXT4_KEY_DERIVATION_NONCE_SIZE;
+	struct crypto_cipher *tfm;
+	int err;
+	int i;
+
+	/*
+	 * Since we only use each transform for a small number of encryptions,
+	 * requesting just "aes" turns out to be significantly faster than
+	 * "ecb(aes)", by about a factor of two.
+	 */
+	tfm = crypto_alloc_cipher("aes", 0, 0);
+	if (IS_ERR(tfm))
+		return PTR_ERR(tfm);
+
+	BUILD_BUG_ON(4 * EXT4_KEY_DERIVATION_NONCE_SIZE != EXT4_MAX_KEY_SIZE);
+	BUILD_BUG_ON(2 * EXT4_AES_256_ECB_KEY_SIZE != EXT4_MAX_KEY_SIZE);
+	for (i = 0; i < 2; i++) {
+		memcpy(derived_key, nonce, noncelen);
+		memcpy(derived_key + noncelen, nonce, noncelen);
+		derived_key[2 * noncelen - 1] ^= 0x01;
+		err = crypto_cipher_setkey(tfm, master_key,
+					   EXT4_AES_256_ECB_KEY_SIZE);
+		if (err)
+			break;
+		crypto_cipher_encrypt_one(tfm, derived_key, derived_key);
+		crypto_cipher_encrypt_one(tfm, derived_key + noncelen,
+					  derived_key + noncelen);
+		master_key += EXT4_AES_256_ECB_KEY_SIZE;
+		derived_key += 2 * noncelen;
+	}
+	crypto_free_cipher(tfm);
+	return err;
+}
+
+/**
+ * ext4_derive_key() - Derive a per-file key from a nonce and master key
+ * @ctx: the encryption context associated with the file
+ * @master_key: the master key referenced by the file
+ * @derived_key: (output) the resulting derived key
+ *
+ * Return: 0 on success, -errno on failure
+ */
+static int ext4_derive_key(const struct ext4_encryption_context *ctx,
+			   const char master_key[EXT4_MAX_KEY_SIZE],
+			   char derived_key[EXT4_MAX_KEY_SIZE])
+{
+	BUILD_BUG_ON(EXT4_AES_128_ECB_KEY_SIZE != EXT4_KEY_DERIVATION_NONCE_SIZE);
+	BUILD_BUG_ON(EXT4_AES_256_XTS_KEY_SIZE != EXT4_MAX_KEY_SIZE);
+
+	/*
+	 * Although the key derivation algorithm is logically independent of the
+	 * choice of encryption modes, in this kernel it is bundled with HEH
+	 * encryption of filenames, which is another crypto improvement that
+	 * requires an on-disk format change and requires userspace to specify
+	 * different encryption policies.
+	 */
+	if (ctx->filenames_encryption_mode == EXT4_ENCRYPTION_MODE_AES_256_HEH)
+		return ext4_derive_key_v2(ctx->nonce, master_key, derived_key);
+	else
+		return ext4_derive_key_v1(ctx->nonce, master_key, derived_key);
+}
+
 void ext4_free_crypt_info(struct ext4_crypt_info *ci)
 {
 	if (!ci)
@@ -172,6 +257,9 @@ int ext4_get_encryption_info(struct inode *inode)
 	case EXT4_ENCRYPTION_MODE_AES_256_CTS:
 		cipher_str = "cts(cbc(aes))";
 		break;
+	case EXT4_ENCRYPTION_MODE_AES_256_HEH:
+		cipher_str = "heh(aes)";
+		break;
 	default:
 		printk_once(KERN_WARNING
 			    "ext4: unsupported key mode %d (ino %u)\n",
@@ -226,8 +314,7 @@ int ext4_get_encryption_info(struct inode *inode)
 		up_read(&keyring_key->sem);
 		goto out;
 	}
-	res = ext4_derive_key_aes(ctx.nonce, master_key->raw,
-				  raw_key);
+	res = ext4_derive_key(&ctx, master_key->raw, raw_key);
 	up_read(&keyring_key->sem);
 	if (res)
 		goto out;
diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index c8ad14c697c4..6edacb849e48 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -589,6 +589,7 @@ enum {
 #define EXT4_ENCRYPTION_MODE_AES_256_GCM	2
 #define EXT4_ENCRYPTION_MODE_AES_256_CBC	3
 #define EXT4_ENCRYPTION_MODE_AES_256_CTS	4
+#define EXT4_ENCRYPTION_MODE_AES_256_HEH	126
 
 #include "ext4_crypto.h"
 
@@ -1441,7 +1442,7 @@ struct ext4_sb_info {
 	struct list_head s_es_list;	/* List of inodes with reclaimable extents */
 	long s_es_nr_inode;
 	struct ext4_es_stats s_es_stats;
-	struct mb_cache *s_mb_cache;
+	struct mb2_cache *s_mb_cache;
 	spinlock_t s_es_lock ____cacheline_aligned_in_smp;
 
 	/* Ratelimit ext4 messages. */
@@ -2456,7 +2457,8 @@ extern int ext4_mb_add_groupinfo(struct super_block *sb,
 		ext4_group_t i, struct ext4_group_desc *desc);
 extern int ext4_group_add_blocks(handle_t *handle, struct super_block *sb,
 				ext4_fsblk_t block, unsigned long count);
-extern int ext4_trim_fs(struct super_block *, struct fstrim_range *);
+extern int ext4_trim_fs(struct super_block *, struct fstrim_range *,
+				unsigned long blkdev_flags);
 
 /* inode.c */
 int ext4_inode_is_fast_symlink(struct inode *inode);
diff --git a/fs/ext4/ext4_crypto.h b/fs/ext4/ext4_crypto.h
index 1b17b05b9f4d..e52637d969db 100644
--- a/fs/ext4/ext4_crypto.h
+++ b/fs/ext4/ext4_crypto.h
@@ -58,8 +58,10 @@ struct ext4_encryption_context {
 #define EXT4_XTS_TWEAK_SIZE 16
 #define EXT4_AES_128_ECB_KEY_SIZE 16
 #define EXT4_AES_256_GCM_KEY_SIZE 32
+#define EXT4_AES_256_ECB_KEY_SIZE 32
 #define EXT4_AES_256_CBC_KEY_SIZE 32
 #define EXT4_AES_256_CTS_KEY_SIZE 32
+#define EXT4_AES_256_HEH_KEY_SIZE 32
 #define EXT4_AES_256_XTS_KEY_SIZE 64
 #define EXT4_MAX_KEY_SIZE 64
 
@@ -120,6 +122,8 @@ static inline int ext4_encryption_key_size(int mode)
 		return EXT4_AES_256_CBC_KEY_SIZE;
 	case EXT4_ENCRYPTION_MODE_AES_256_CTS:
 		return EXT4_AES_256_CTS_KEY_SIZE;
+	case EXT4_ENCRYPTION_MODE_AES_256_HEH:
+		return EXT4_AES_256_HEH_KEY_SIZE;
 	default:
 		BUG();
 	}
diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c
index dad8e7bdf0a6..bc7c082b7913 100644
--- a/fs/ext4/inline.c
+++ b/fs/ext4/inline.c
@@ -18,6 +18,7 @@
 #include "ext4.h"
 #include "xattr.h"
 #include "truncate.h"
+#include <trace/events/android_fs.h>
 
 #define EXT4_XATTR_SYSTEM_DATA	"data"
 #define EXT4_MIN_INLINE_DATA_SIZE	((sizeof(__le32) * EXT4_N_BLOCKS))
@@ -502,6 +503,17 @@ int ext4_readpage_inline(struct inode *inode, struct page *page)
 		return -EAGAIN;
 	}
 
+	if (trace_android_fs_dataread_start_enabled()) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_dataread_start(inode, page_offset(page),
+						PAGE_SIZE, current->pid,
+						path, current->comm);
+	}
+
 	/*
 	 * Current inline data can only exist in the 1st page,
 	 * So for all the other pages, just set them uptodate.
@@ -513,6 +525,8 @@ int ext4_readpage_inline(struct inode *inode, struct page *page)
 		SetPageUptodate(page);
 	}
 
+	trace_android_fs_dataread_end(inode, page_offset(page), PAGE_SIZE);
+
 	up_read(&EXT4_I(inode)->xattr_sem);
 
 	unlock_page(page);
diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c
index 4df1cb19a243..530e790f22e0 100644
--- a/fs/ext4/inode.c
+++ b/fs/ext4/inode.c
@@ -44,6 +44,7 @@
 #include "truncate.h"
 
 #include <trace/events/ext4.h>
+#include <trace/events/android_fs.h>
 
 #define MPAGE_DA_EXTENT_TAIL 0x01
 
@@ -1029,6 +1030,16 @@ static int ext4_write_begin(struct file *file, struct address_space *mapping,
 	pgoff_t index;
 	unsigned from, to;
 
+	if (trace_android_fs_datawrite_start_enabled()) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_datawrite_start(inode, pos, len,
+						 current->pid, path,
+						 current->comm);
+	}
 	trace_ext4_write_begin(inode, pos, len, flags);
 	/*
 	 * Reserve one block more for addition to orphan list in case
@@ -1165,6 +1176,7 @@ static int ext4_write_end(struct file *file,
 	int ret = 0, ret2;
 	int i_size_changed = 0;
 
+	trace_android_fs_datawrite_end(inode, pos, len);
 	trace_ext4_write_end(inode, pos, len, copied);
 	if (ext4_has_inline_data(inode)) {
 		ret = ext4_write_inline_data_end(inode, pos, len,
@@ -1269,6 +1281,7 @@ static int ext4_journalled_write_end(struct file *file,
 	unsigned from, to;
 	int size_changed = 0;
 
+	trace_android_fs_datawrite_end(inode, pos, len);
 	trace_ext4_journalled_write_end(inode, pos, len, copied);
 	from = pos & (PAGE_CACHE_SIZE - 1);
 	to = from + len;
@@ -2758,6 +2771,16 @@ static int ext4_da_write_begin(struct file *file, struct address_space *mapping,
 					len, flags, pagep, fsdata);
 	}
 	*fsdata = (void *)0;
+	if (trace_android_fs_datawrite_start_enabled()) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_datawrite_start(inode, pos, len,
+						 current->pid,
+						 path, current->comm);
+	}
 	trace_ext4_da_write_begin(inode, pos, len, flags);
 
 	if (ext4_test_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA)) {
@@ -2876,6 +2899,7 @@ static int ext4_da_write_end(struct file *file,
 		return ext4_write_end(file, mapping, pos,
 				      len, copied, page, fsdata);
 
+	trace_android_fs_datawrite_end(inode, pos, len);
 	trace_ext4_da_write_end(inode, pos, len, copied);
 	start = pos & (PAGE_CACHE_SIZE - 1);
 	end = start + copied - 1;
@@ -3364,12 +3388,42 @@ static ssize_t ext4_direct_IO(struct kiocb *iocb, struct iov_iter *iter,
 	if (ext4_has_inline_data(inode))
 		return 0;
 
+	if (trace_android_fs_dataread_start_enabled() &&
+	    (iov_iter_rw(iter) == READ)) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_dataread_start(inode, offset, count,
+						current->pid, path,
+						current->comm);
+	}
+	if (trace_android_fs_datawrite_start_enabled() &&
+	    (iov_iter_rw(iter) == WRITE)) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_datawrite_start(inode, offset, count,
+						 current->pid, path,
+						 current->comm);
+	}
 	trace_ext4_direct_IO_enter(inode, offset, count, iov_iter_rw(iter));
 	if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))
 		ret = ext4_ext_direct_IO(iocb, iter, offset);
 	else
 		ret = ext4_ind_direct_IO(iocb, iter, offset);
 	trace_ext4_direct_IO_exit(inode, offset, count, iov_iter_rw(iter), ret);
+
+	if (trace_android_fs_dataread_start_enabled() &&
+	    (iov_iter_rw(iter) == READ))
+		trace_android_fs_dataread_end(inode, offset, count);
+	if (trace_android_fs_datawrite_start_enabled() &&
+	    (iov_iter_rw(iter) == WRITE))
+		trace_android_fs_datawrite_end(inode, offset, count);
+
 	return ret;
 }
 
diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c
index bcd7c4788903..3a2594665b44 100644
--- a/fs/ext4/ioctl.c
+++ b/fs/ext4/ioctl.c
@@ -587,11 +587,13 @@ resizefs_out:
 		return err;
 	}
 
+	case FIDTRIM:
 	case FITRIM:
 	{
 		struct request_queue *q = bdev_get_queue(sb->s_bdev);
 		struct fstrim_range range;
 		int ret = 0;
+		int flags  = cmd == FIDTRIM ? BLKDEV_DISCARD_SECURE : 0;
 
 		if (!capable(CAP_SYS_ADMIN))
 			return -EPERM;
@@ -599,13 +601,15 @@ resizefs_out:
 		if (!blk_queue_discard(q))
 			return -EOPNOTSUPP;
 
+		if ((flags & BLKDEV_DISCARD_SECURE) && !blk_queue_secdiscard(q))
+			return -EOPNOTSUPP;
 		if (copy_from_user(&range, (struct fstrim_range __user *)arg,
 		    sizeof(range)))
 			return -EFAULT;
 
 		range.minlen = max((unsigned int)range.minlen,
 				   q->limits.discard_granularity);
-		ret = ext4_trim_fs(sb, &range);
+		ret = ext4_trim_fs(sb, &range, flags);
 		if (ret < 0)
 			return ret;
 
@@ -622,9 +626,6 @@ resizefs_out:
 		struct ext4_encryption_policy policy;
 		int err = 0;
 
-		if (!ext4_has_feature_encrypt(sb))
-			return -EOPNOTSUPP;
-
 		if (copy_from_user(&policy,
 				   (struct ext4_encryption_policy __user *)arg,
 				   sizeof(policy))) {
diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c
index 1ba82dc5afa3..27ff3706d632 100644
--- a/fs/ext4/mballoc.c
+++ b/fs/ext4/mballoc.c
@@ -2772,7 +2772,8 @@ int ext4_mb_release(struct super_block *sb)
 }
 
 static inline int ext4_issue_discard(struct super_block *sb,
-		ext4_group_t block_group, ext4_grpblk_t cluster, int count)
+		ext4_group_t block_group, ext4_grpblk_t cluster, int count,
+		unsigned long flags)
 {
 	ext4_fsblk_t discard_block;
 
@@ -2781,7 +2782,7 @@ static inline int ext4_issue_discard(struct super_block *sb,
 	count = EXT4_C2B(EXT4_SB(sb), count);
 	trace_ext4_discard_blocks(sb,
 			(unsigned long long) discard_block, count);
-	return sb_issue_discard(sb, discard_block, count, GFP_NOFS, 0);
+	return sb_issue_discard(sb, discard_block, count, GFP_NOFS, flags);
 }
 
 /*
@@ -2803,7 +2804,7 @@ static void ext4_free_data_callback(struct super_block *sb,
 	if (test_opt(sb, DISCARD)) {
 		err = ext4_issue_discard(sb, entry->efd_group,
 					 entry->efd_start_cluster,
-					 entry->efd_count);
+					 entry->efd_count, 0);
 		if (err && err != -EOPNOTSUPP)
 			ext4_msg(sb, KERN_WARNING, "discard request in"
 				 " group:%d block:%d count:%d failed"
@@ -4855,7 +4856,8 @@ do_more:
 		 * them with group lock_held
 		 */
 		if (test_opt(sb, DISCARD)) {
-			err = ext4_issue_discard(sb, block_group, bit, count);
+			err = ext4_issue_discard(sb, block_group, bit, count,
+						 0);
 			if (err && err != -EOPNOTSUPP)
 				ext4_msg(sb, KERN_WARNING, "discard request in"
 					 " group:%d block:%d count:%lu failed"
@@ -5051,13 +5053,15 @@ error_return:
  * @count:	number of blocks to TRIM
  * @group:	alloc. group we are working with
  * @e4b:	ext4 buddy for the group
+ * @blkdev_flags: flags for the block device
  *
  * Trim "count" blocks starting at "start" in the "group". To assure that no
  * one will allocate those blocks, mark it as used in buddy bitmap. This must
  * be called with under the group lock.
  */
 static int ext4_trim_extent(struct super_block *sb, int start, int count,
-			     ext4_group_t group, struct ext4_buddy *e4b)
+			    ext4_group_t group, struct ext4_buddy *e4b,
+			    unsigned long blkdev_flags)
 __releases(bitlock)
 __acquires(bitlock)
 {
@@ -5078,7 +5082,7 @@ __acquires(bitlock)
 	 */
 	mb_mark_used(e4b, &ex);
 	ext4_unlock_group(sb, group);
-	ret = ext4_issue_discard(sb, group, start, count);
+	ret = ext4_issue_discard(sb, group, start, count, blkdev_flags);
 	ext4_lock_group(sb, group);
 	mb_free_blocks(NULL, e4b, start, ex.fe_len);
 	return ret;
@@ -5091,6 +5095,7 @@ __acquires(bitlock)
  * @start:		first group block to examine
  * @max:		last group block to examine
  * @minblocks:		minimum extent block count
+ * @blkdev_flags:	flags for the block device
  *
  * ext4_trim_all_free walks through group's buddy bitmap searching for free
  * extents. When the free block is found, ext4_trim_extent is called to TRIM
@@ -5105,7 +5110,7 @@ __acquires(bitlock)
 static ext4_grpblk_t
 ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
 		   ext4_grpblk_t start, ext4_grpblk_t max,
-		   ext4_grpblk_t minblocks)
+		   ext4_grpblk_t minblocks, unsigned long blkdev_flags)
 {
 	void *bitmap;
 	ext4_grpblk_t next, count = 0, free_count = 0;
@@ -5138,7 +5143,8 @@ ext4_trim_all_free(struct super_block *sb, ext4_group_t group,
 
 		if ((next - start) >= minblocks) {
 			ret = ext4_trim_extent(sb, start,
-					       next - start, group, &e4b);
+					       next - start, group, &e4b,
+					       blkdev_flags);
 			if (ret && ret != -EOPNOTSUPP)
 				break;
 			ret = 0;
@@ -5180,6 +5186,7 @@ out:
  * ext4_trim_fs() -- trim ioctl handle function
  * @sb:			superblock for filesystem
  * @range:		fstrim_range structure
+ * @blkdev_flags:	flags for the block device
  *
  * start:	First Byte to trim
  * len:		number of Bytes to trim from start
@@ -5188,7 +5195,8 @@ out:
  * start to start+len. For each such a group ext4_trim_all_free function
  * is invoked to trim all free space.
  */
-int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
+int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range,
+			unsigned long blkdev_flags)
 {
 	struct ext4_group_info *grp;
 	ext4_group_t group, first_group, last_group;
@@ -5244,7 +5252,7 @@ int ext4_trim_fs(struct super_block *sb, struct fstrim_range *range)
 
 		if (grp->bb_free >= minlen) {
 			cnt = ext4_trim_all_free(sb, group, first_cluster,
-						end, minlen);
+						end, minlen, blkdev_flags);
 			if (cnt < 0) {
 				ret = cnt;
 				break;
diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c
index bc7642f57dc8..783e33d839cf 100644
--- a/fs/ext4/readpage.c
+++ b/fs/ext4/readpage.c
@@ -45,6 +45,7 @@
 #include <linux/cleancache.h>
 
 #include "ext4.h"
+#include <trace/events/android_fs.h>
 
 /*
  * Call ext4_decrypt on every single page, reusing the encryption
@@ -86,6 +87,17 @@ static inline bool ext4_bio_encrypted(struct bio *bio)
 #endif
 }
 
+static void
+ext4_trace_read_completion(struct bio *bio)
+{
+	struct page *first_page = bio->bi_io_vec[0].bv_page;
+
+	if (first_page != NULL)
+		trace_android_fs_dataread_end(first_page->mapping->host,
+					      page_offset(first_page),
+					      bio->bi_iter.bi_size);
+}
+
 /*
  * I/O completion handler for multipage BIOs.
  *
@@ -103,6 +115,9 @@ static void mpage_end_io(struct bio *bio)
 	struct bio_vec *bv;
 	int i;
 
+	if (trace_android_fs_dataread_start_enabled())
+		ext4_trace_read_completion(bio);
+
 	if (ext4_bio_encrypted(bio)) {
 		struct ext4_crypto_ctx *ctx = bio->bi_private;
 
@@ -130,6 +145,30 @@ static void mpage_end_io(struct bio *bio)
 	bio_put(bio);
 }
 
+static void
+ext4_submit_bio_read(struct bio *bio)
+{
+	if (trace_android_fs_dataread_start_enabled()) {
+		struct page *first_page = bio->bi_io_vec[0].bv_page;
+
+		if (first_page != NULL) {
+			char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+			path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    first_page->mapping->host);
+			trace_android_fs_dataread_start(
+				first_page->mapping->host,
+				page_offset(first_page),
+				bio->bi_iter.bi_size,
+				current->pid,
+				path,
+				current->comm);
+		}
+	}
+	submit_bio(READ, bio);
+}
+
 int ext4_mpage_readpages(struct address_space *mapping,
 			 struct list_head *pages, struct page *page,
 			 unsigned nr_pages)
@@ -271,7 +310,7 @@ int ext4_mpage_readpages(struct address_space *mapping,
 		 */
 		if (bio && (last_block_in_bio != blocks[0] - 1)) {
 		submit_and_realloc:
-			submit_bio(READ, bio);
+			ext4_submit_bio_read(bio);
 			bio = NULL;
 		}
 		if (bio == NULL) {
@@ -303,14 +342,14 @@ int ext4_mpage_readpages(struct address_space *mapping,
 		if (((map.m_flags & EXT4_MAP_BOUNDARY) &&
 		     (relative_block == map.m_len)) ||
 		    (first_hole != blocks_per_page)) {
-			submit_bio(READ, bio);
+			ext4_submit_bio_read(bio);
 			bio = NULL;
 		} else
 			last_block_in_bio = blocks[blocks_per_page - 1];
 		goto next_page;
 	confused:
 		if (bio) {
-			submit_bio(READ, bio);
+			ext4_submit_bio_read(bio);
 			bio = NULL;
 		}
 		if (!PageUptodate(page))
@@ -323,6 +362,6 @@ int ext4_mpage_readpages(struct address_space *mapping,
 	}
 	BUG_ON(pages && !list_empty(pages));
 	if (bio)
-		submit_bio(READ, bio);
+		ext4_submit_bio_read(bio);
 	return 0;
 }
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 8bdb0cc2722f..d3cbdbc8ad33 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -816,7 +816,6 @@ static void ext4_put_super(struct super_block *sb)
 	ext4_release_system_zone(sb);
 	ext4_mb_release(sb);
 	ext4_ext_release(sb);
-	ext4_xattr_put_super(sb);
 
 	if (!(sb->s_flags & MS_RDONLY) && !aborted) {
 		ext4_clear_feature_journal_needs_recovery(sb);
@@ -3854,7 +3853,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent)
 
 no_journal:
 	if (ext4_mballoc_ready) {
-		sbi->s_mb_cache = ext4_xattr_create_cache(sb->s_id);
+		sbi->s_mb_cache = ext4_xattr_create_cache();
 		if (!sbi->s_mb_cache) {
 			ext4_msg(sb, KERN_ERR, "Failed to create an mb_cache");
 			goto failed_mount_wq;
@@ -4086,6 +4085,10 @@ failed_mount4:
 	if (EXT4_SB(sb)->rsv_conversion_wq)
 		destroy_workqueue(EXT4_SB(sb)->rsv_conversion_wq);
 failed_mount_wq:
+	if (sbi->s_mb_cache) {
+		ext4_xattr_destroy_cache(sbi->s_mb_cache);
+		sbi->s_mb_cache = NULL;
+	}
 	if (sbi->s_journal) {
 		jbd2_journal_destroy(sbi->s_journal);
 		sbi->s_journal = NULL;
diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c
index 7c23363ecf19..b310ed81c10e 100644
--- a/fs/ext4/xattr.c
+++ b/fs/ext4/xattr.c
@@ -53,7 +53,7 @@
 #include <linux/init.h>
 #include <linux/fs.h>
 #include <linux/slab.h>
-#include <linux/mbcache.h>
+#include <linux/mbcache2.h>
 #include <linux/quotaops.h>
 #include "ext4_jbd2.h"
 #include "ext4.h"
@@ -80,10 +80,10 @@
 # define ea_bdebug(bh, fmt, ...)	no_printk(fmt, ##__VA_ARGS__)
 #endif
 
-static void ext4_xattr_cache_insert(struct mb_cache *, struct buffer_head *);
+static void ext4_xattr_cache_insert(struct mb2_cache *, struct buffer_head *);
 static struct buffer_head *ext4_xattr_cache_find(struct inode *,
 						 struct ext4_xattr_header *,
-						 struct mb_cache_entry **);
+						 struct mb2_cache_entry **);
 static void ext4_xattr_rehash(struct ext4_xattr_header *,
 			      struct ext4_xattr_entry *);
 static int ext4_xattr_list(struct dentry *dentry, char *buffer,
@@ -300,7 +300,7 @@ ext4_xattr_block_get(struct inode *inode, int name_index, const char *name,
 	struct ext4_xattr_entry *entry;
 	size_t size;
 	int error;
-	struct mb_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
+	struct mb2_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
 
 	ea_idebug(inode, "name=%d.%s, buffer=%p, buffer_size=%ld",
 		  name_index, name, buffer, (long)buffer_size);
@@ -447,7 +447,7 @@ ext4_xattr_block_list(struct dentry *dentry, char *buffer, size_t buffer_size)
 	struct inode *inode = d_inode(dentry);
 	struct buffer_head *bh = NULL;
 	int error;
-	struct mb_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
+	struct mb2_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
 
 	ea_idebug(inode, "buffer=%p, buffer_size=%ld",
 		  buffer, (long)buffer_size);
@@ -564,11 +564,8 @@ static void
 ext4_xattr_release_block(handle_t *handle, struct inode *inode,
 			 struct buffer_head *bh)
 {
-	struct mb_cache_entry *ce = NULL;
 	int error = 0;
-	struct mb_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
 
-	ce = mb_cache_entry_get(ext4_mb_cache, bh->b_bdev, bh->b_blocknr);
 	BUFFER_TRACE(bh, "get_write_access");
 	error = ext4_journal_get_write_access(handle, bh);
 	if (error)
@@ -576,9 +573,15 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode,
 
 	lock_buffer(bh);
 	if (BHDR(bh)->h_refcount == cpu_to_le32(1)) {
+		__u32 hash = le32_to_cpu(BHDR(bh)->h_hash);
+
 		ea_bdebug(bh, "refcount now=0; freeing");
-		if (ce)
-			mb_cache_entry_free(ce);
+		/*
+		 * This must happen under buffer lock for
+		 * ext4_xattr_block_set() to reliably detect freed block
+		 */
+		mb2_cache_entry_delete_block(EXT4_GET_MB_CACHE(inode), hash,
+					     bh->b_blocknr);
 		get_bh(bh);
 		unlock_buffer(bh);
 		ext4_free_blocks(handle, inode, bh, 0, 1,
@@ -586,8 +589,6 @@ ext4_xattr_release_block(handle_t *handle, struct inode *inode,
 				 EXT4_FREE_BLOCKS_FORGET);
 	} else {
 		le32_add_cpu(&BHDR(bh)->h_refcount, -1);
-		if (ce)
-			mb_cache_entry_release(ce);
 		/*
 		 * Beware of this ugliness: Releasing of xattr block references
 		 * from different inodes can race and so we have to protect
@@ -800,17 +801,15 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode,
 	struct super_block *sb = inode->i_sb;
 	struct buffer_head *new_bh = NULL;
 	struct ext4_xattr_search *s = &bs->s;
-	struct mb_cache_entry *ce = NULL;
+	struct mb2_cache_entry *ce = NULL;
 	int error = 0;
-	struct mb_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
+	struct mb2_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
 
 #define header(x) ((struct ext4_xattr_header *)(x))
 
 	if (i->value && i->value_len > sb->s_blocksize)
 		return -ENOSPC;
 	if (s->base) {
-		ce = mb_cache_entry_get(ext4_mb_cache, bs->bh->b_bdev,
-					bs->bh->b_blocknr);
 		BUFFER_TRACE(bs->bh, "get_write_access");
 		error = ext4_journal_get_write_access(handle, bs->bh);
 		if (error)
@@ -818,10 +817,15 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode,
 		lock_buffer(bs->bh);
 
 		if (header(s->base)->h_refcount == cpu_to_le32(1)) {
-			if (ce) {
-				mb_cache_entry_free(ce);
-				ce = NULL;
-			}
+			__u32 hash = le32_to_cpu(BHDR(bs->bh)->h_hash);
+
+			/*
+			 * This must happen under buffer lock for
+			 * ext4_xattr_block_set() to reliably detect modified
+			 * block
+			 */
+			mb2_cache_entry_delete_block(ext4_mb_cache, hash,
+						     bs->bh->b_blocknr);
 			ea_bdebug(bs->bh, "modifying in-place");
 			error = ext4_xattr_set_entry(i, s);
 			if (!error) {
@@ -845,10 +849,6 @@ ext4_xattr_block_set(handle_t *handle, struct inode *inode,
 			int offset = (char *)s->here - bs->bh->b_data;
 
 			unlock_buffer(bs->bh);
-			if (ce) {
-				mb_cache_entry_release(ce);
-				ce = NULL;
-			}
 			ea_bdebug(bs->bh, "cloning");
 			s->base = kmalloc(bs->bh->b_size, GFP_NOFS);
 			error = -ENOMEM;
@@ -903,6 +903,31 @@ inserted:
 				if (error)
 					goto cleanup_dquot;
 				lock_buffer(new_bh);
+				/*
+				 * We have to be careful about races with
+				 * freeing or rehashing of xattr block. Once we
+				 * hold buffer lock xattr block's state is
+				 * stable so we can check whether the block got
+				 * freed / rehashed or not.  Since we unhash
+				 * mbcache entry under buffer lock when freeing
+				 * / rehashing xattr block, checking whether
+				 * entry is still hashed is reliable.
+				 */
+				if (hlist_bl_unhashed(&ce->e_hash_list)) {
+					/*
+					 * Undo everything and check mbcache
+					 * again.
+					 */
+					unlock_buffer(new_bh);
+					dquot_free_block(inode,
+							 EXT4_C2B(EXT4_SB(sb),
+								  1));
+					brelse(new_bh);
+					mb2_cache_entry_put(ext4_mb_cache, ce);
+					ce = NULL;
+					new_bh = NULL;
+					goto inserted;
+				}
 				le32_add_cpu(&BHDR(new_bh)->h_refcount, 1);
 				ea_bdebug(new_bh, "reusing; refcount now=%d",
 					le32_to_cpu(BHDR(new_bh)->h_refcount));
@@ -913,7 +938,8 @@ inserted:
 				if (error)
 					goto cleanup_dquot;
 			}
-			mb_cache_entry_release(ce);
+			mb2_cache_entry_touch(ext4_mb_cache, ce);
+			mb2_cache_entry_put(ext4_mb_cache, ce);
 			ce = NULL;
 		} else if (bs->bh && s->base == bs->bh->b_data) {
 			/* We were modifying this block in-place. */
@@ -978,7 +1004,7 @@ getblk_failed:
 
 cleanup:
 	if (ce)
-		mb_cache_entry_release(ce);
+		mb2_cache_entry_put(ext4_mb_cache, ce);
 	brelse(new_bh);
 	if (!(bs->bh && s->base == bs->bh->b_data))
 		kfree(s->base);
@@ -1543,17 +1569,6 @@ cleanup:
 }
 
 /*
- * ext4_xattr_put_super()
- *
- * This is called when a file system is unmounted.
- */
-void
-ext4_xattr_put_super(struct super_block *sb)
-{
-	mb_cache_shrink(sb->s_bdev);
-}
-
-/*
  * ext4_xattr_cache_insert()
  *
  * Create a new entry in the extended attribute cache, and insert
@@ -1562,28 +1577,18 @@ ext4_xattr_put_super(struct super_block *sb)
  * Returns 0, or a negative error number on failure.
  */
 static void
-ext4_xattr_cache_insert(struct mb_cache *ext4_mb_cache, struct buffer_head *bh)
+ext4_xattr_cache_insert(struct mb2_cache *ext4_mb_cache, struct buffer_head *bh)
 {
 	__u32 hash = le32_to_cpu(BHDR(bh)->h_hash);
-	struct mb_cache_entry *ce;
 	int error;
 
-	ce = mb_cache_entry_alloc(ext4_mb_cache, GFP_NOFS);
-	if (!ce) {
-		ea_bdebug(bh, "out of memory");
-		return;
-	}
-	error = mb_cache_entry_insert(ce, bh->b_bdev, bh->b_blocknr, hash);
+	error = mb2_cache_entry_create(ext4_mb_cache, GFP_NOFS, hash,
+				       bh->b_blocknr);
 	if (error) {
-		mb_cache_entry_free(ce);
-		if (error == -EBUSY) {
+		if (error == -EBUSY)
 			ea_bdebug(bh, "already in cache");
-			error = 0;
-		}
-	} else {
+	} else
 		ea_bdebug(bh, "inserting [%x]", (int)hash);
-		mb_cache_entry_release(ce);
-	}
 }
 
 /*
@@ -1636,26 +1641,19 @@ ext4_xattr_cmp(struct ext4_xattr_header *header1,
  */
 static struct buffer_head *
 ext4_xattr_cache_find(struct inode *inode, struct ext4_xattr_header *header,
-		      struct mb_cache_entry **pce)
+		      struct mb2_cache_entry **pce)
 {
 	__u32 hash = le32_to_cpu(header->h_hash);
-	struct mb_cache_entry *ce;
-	struct mb_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
+	struct mb2_cache_entry *ce;
+	struct mb2_cache *ext4_mb_cache = EXT4_GET_MB_CACHE(inode);
 
 	if (!header->h_hash)
 		return NULL;  /* never share */
 	ea_idebug(inode, "looking for cached blocks [%x]", (int)hash);
-again:
-	ce = mb_cache_entry_find_first(ext4_mb_cache, inode->i_sb->s_bdev,
-				       hash);
+	ce = mb2_cache_entry_find_first(ext4_mb_cache, hash);
 	while (ce) {
 		struct buffer_head *bh;
 
-		if (IS_ERR(ce)) {
-			if (PTR_ERR(ce) == -EAGAIN)
-				goto again;
-			break;
-		}
 		bh = sb_bread(inode->i_sb, ce->e_block);
 		if (!bh) {
 			EXT4_ERROR_INODE(inode, "block %lu read error",
@@ -1671,7 +1669,7 @@ again:
 			return bh;
 		}
 		brelse(bh);
-		ce = mb_cache_entry_find_next(ce, inode->i_sb->s_bdev, hash);
+		ce = mb2_cache_entry_find_next(ext4_mb_cache, ce);
 	}
 	return NULL;
 }
@@ -1746,15 +1744,15 @@ static void ext4_xattr_rehash(struct ext4_xattr_header *header,
 
 #define	HASH_BUCKET_BITS	10
 
-struct mb_cache *
-ext4_xattr_create_cache(char *name)
+struct mb2_cache *
+ext4_xattr_create_cache(void)
 {
-	return mb_cache_create(name, HASH_BUCKET_BITS);
+	return mb2_cache_create(HASH_BUCKET_BITS);
 }
 
-void ext4_xattr_destroy_cache(struct mb_cache *cache)
+void ext4_xattr_destroy_cache(struct mb2_cache *cache)
 {
 	if (cache)
-		mb_cache_destroy(cache);
+		mb2_cache_destroy(cache);
 }
 
diff --git a/fs/ext4/xattr.h b/fs/ext4/xattr.h
index ddc0957760ba..10b0f7323ed6 100644
--- a/fs/ext4/xattr.h
+++ b/fs/ext4/xattr.h
@@ -108,7 +108,6 @@ extern int ext4_xattr_set(struct inode *, int, const char *, const void *, size_
 extern int ext4_xattr_set_handle(handle_t *, struct inode *, int, const char *, const void *, size_t, int);
 
 extern void ext4_xattr_delete_inode(handle_t *, struct inode *);
-extern void ext4_xattr_put_super(struct super_block *);
 
 extern int ext4_expand_extra_isize_ea(struct inode *inode, int new_extra_isize,
 			    struct ext4_inode *raw_inode, handle_t *handle);
@@ -124,8 +123,8 @@ extern int ext4_xattr_ibody_inline_set(handle_t *handle, struct inode *inode,
 				       struct ext4_xattr_info *i,
 				       struct ext4_xattr_ibody_find *is);
 
-extern struct mb_cache *ext4_xattr_create_cache(char *name);
-extern void ext4_xattr_destroy_cache(struct mb_cache *);
+extern struct mb2_cache *ext4_xattr_create_cache(void);
+extern void ext4_xattr_destroy_cache(struct mb2_cache *);
 
 #ifdef CONFIG_EXT4_FS_SECURITY
 extern int ext4_init_security(handle_t *handle, struct inode *inode,
diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig
index b0a9dc929f88..378c221d68a9 100644
--- a/fs/f2fs/Kconfig
+++ b/fs/f2fs/Kconfig
@@ -1,6 +1,8 @@
 config F2FS_FS
 	tristate "F2FS filesystem support"
 	depends on BLOCK
+	select CRYPTO
+	select CRYPTO_CRC32
 	help
 	  F2FS is based on Log-structured File System (LFS), which supports
 	  versatile "flash-friendly" features. The design has been focused on
@@ -76,15 +78,7 @@ config F2FS_FS_ENCRYPTION
 	bool "F2FS Encryption"
 	depends on F2FS_FS
 	depends on F2FS_FS_XATTR
-	select CRYPTO_AES
-	select CRYPTO_CBC
-	select CRYPTO_ECB
-	select CRYPTO_XTS
-	select CRYPTO_CTS
-	select CRYPTO_CTR
-	select CRYPTO_SHA256
-	select KEYS
-	select ENCRYPTED_KEYS
+	select FS_ENCRYPTION
 	help
 	  Enable encryption of f2fs files and directories.  This
 	  feature is similar to ecryptfs, but it is more memory
@@ -100,3 +94,11 @@ config F2FS_IO_TRACE
 	  information and block IO patterns in the filesystem level.
 
 	  If unsure, say N.
+
+config F2FS_FAULT_INJECTION
+	bool "F2FS fault injection facility"
+	depends on F2FS_FS
+	help
+	  Test F2FS to inject faults such as ENOMEM, ENOSPC, and so on.
+
+	  If unsure, say N.
diff --git a/fs/f2fs/Makefile b/fs/f2fs/Makefile
index 08e101ed914c..a0dc559b1b47 100644
--- a/fs/f2fs/Makefile
+++ b/fs/f2fs/Makefile
@@ -2,10 +2,8 @@ obj-$(CONFIG_F2FS_FS) += f2fs.o
 
 f2fs-y		:= dir.o file.o inode.o namei.o hash.o super.o inline.o
 f2fs-y		+= checkpoint.o gc.o data.o node.o segment.o recovery.o
-f2fs-y		+= shrinker.o extent_cache.o
+f2fs-y		+= shrinker.o extent_cache.o sysfs.o
 f2fs-$(CONFIG_F2FS_STAT_FS) += debug.o
 f2fs-$(CONFIG_F2FS_FS_XATTR) += xattr.o
 f2fs-$(CONFIG_F2FS_FS_POSIX_ACL) += acl.o
 f2fs-$(CONFIG_F2FS_IO_TRACE) += trace.o
-f2fs-$(CONFIG_F2FS_FS_ENCRYPTION) += crypto_policy.o crypto.o \
-		crypto_key.o crypto_fname.o
diff --git a/fs/f2fs/acl.c b/fs/f2fs/acl.c
index 83dcf7bfd7b8..112f8e04c549 100644
--- a/fs/f2fs/acl.c
+++ b/fs/f2fs/acl.c
@@ -109,14 +109,16 @@ fail:
 	return ERR_PTR(-EINVAL);
 }
 
-static void *f2fs_acl_to_disk(const struct posix_acl *acl, size_t *size)
+static void *f2fs_acl_to_disk(struct f2fs_sb_info *sbi,
+				const struct posix_acl *acl, size_t *size)
 {
 	struct f2fs_acl_header *f2fs_acl;
 	struct f2fs_acl_entry *entry;
 	int i;
 
-	f2fs_acl = kmalloc(sizeof(struct f2fs_acl_header) + acl->a_count *
-			sizeof(struct f2fs_acl_entry), GFP_NOFS);
+	f2fs_acl = f2fs_kmalloc(sbi, sizeof(struct f2fs_acl_header) +
+			acl->a_count * sizeof(struct f2fs_acl_entry),
+			GFP_NOFS);
 	if (!f2fs_acl)
 		return ERR_PTR(-ENOMEM);
 
@@ -175,7 +177,7 @@ static struct posix_acl *__f2fs_get_acl(struct inode *inode, int type,
 
 	retval = f2fs_getxattr(inode, name_index, "", NULL, 0, dpage);
 	if (retval > 0) {
-		value = kmalloc(retval, GFP_F2FS_ZERO);
+		value = f2fs_kmalloc(F2FS_I_SB(inode), retval, GFP_F2FS_ZERO);
 		if (!value)
 			return ERR_PTR(-ENOMEM);
 		retval = f2fs_getxattr(inode, name_index, "", value,
@@ -204,20 +206,20 @@ struct posix_acl *f2fs_get_acl(struct inode *inode, int type)
 static int __f2fs_set_acl(struct inode *inode, int type,
 			struct posix_acl *acl, struct page *ipage)
 {
-	struct f2fs_inode_info *fi = F2FS_I(inode);
 	int name_index;
 	void *value = NULL;
 	size_t size = 0;
 	int error;
+	umode_t mode = inode->i_mode;
 
 	switch (type) {
 	case ACL_TYPE_ACCESS:
 		name_index = F2FS_XATTR_INDEX_POSIX_ACL_ACCESS;
 		if (acl && !ipage) {
-			error = posix_acl_update_mode(inode, &inode->i_mode, &acl);
+			error = posix_acl_update_mode(inode, &mode, &acl);
 			if (error)
 				return error;
-			set_acl_inode(fi, inode->i_mode);
+			set_acl_inode(inode, mode);
 		}
 		break;
 
@@ -232,10 +234,10 @@ static int __f2fs_set_acl(struct inode *inode, int type,
 	}
 
 	if (acl) {
-		value = f2fs_acl_to_disk(acl, &size);
+		value = f2fs_acl_to_disk(F2FS_I_SB(inode), acl, &size);
 		if (IS_ERR(value)) {
-			clear_inode_flag(fi, FI_ACL_MODE);
-			return (int)PTR_ERR(value);
+			clear_inode_flag(inode, FI_ACL_MODE);
+			return PTR_ERR(value);
 		}
 	}
 
@@ -245,7 +247,7 @@ static int __f2fs_set_acl(struct inode *inode, int type,
 	if (!error)
 		set_cached_acl(inode, type, acl);
 
-	clear_inode_flag(fi, FI_ACL_MODE);
+	clear_inode_flag(inode, FI_ACL_MODE);
 	return error;
 }
 
@@ -386,6 +388,8 @@ int f2fs_init_acl(struct inode *inode, struct inode *dir, struct page *ipage,
 	if (error)
 		return error;
 
+	f2fs_mark_inode_dirty_sync(inode, true);
+
 	if (default_acl) {
 		error = __f2fs_set_acl(inode, ACL_TYPE_DEFAULT, default_acl,
 				       ipage);
diff --git a/fs/f2fs/acl.h b/fs/f2fs/acl.h
index 997ca8edb6cb..2c685185c24d 100644
--- a/fs/f2fs/acl.h
+++ b/fs/f2fs/acl.h
@@ -37,11 +37,10 @@ struct f2fs_acl_header {
 #ifdef CONFIG_F2FS_FS_POSIX_ACL
 
 extern struct posix_acl *f2fs_get_acl(struct inode *, int);
-extern int f2fs_set_acl(struct inode *inode, struct posix_acl *acl, int type);
+extern int f2fs_set_acl(struct inode *, struct posix_acl *, int);
 extern int f2fs_init_acl(struct inode *, struct inode *, struct page *,
 							struct page *);
 #else
-#define f2fs_check_acl	NULL
 #define f2fs_get_acl	NULL
 #define f2fs_set_acl	NULL
 
diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c
index f661d80474be..e86f67ac96c6 100644
--- a/fs/f2fs/checkpoint.c
+++ b/fs/f2fs/checkpoint.c
@@ -26,6 +26,14 @@
 static struct kmem_cache *ino_entry_slab;
 struct kmem_cache *inode_entry_slab;
 
+void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi, bool end_io)
+{
+	set_ckpt_flags(sbi, CP_ERROR_FLAG);
+	sbi->sb->s_flags |= MS_RDONLY;
+	if (!end_io)
+		f2fs_flush_merged_writes(sbi);
+}
+
 /*
  * We guarantee no failure on the returned page.
  */
@@ -34,13 +42,14 @@ struct page *grab_meta_page(struct f2fs_sb_info *sbi, pgoff_t index)
 	struct address_space *mapping = META_MAPPING(sbi);
 	struct page *page = NULL;
 repeat:
-	page = grab_cache_page(mapping, index);
+	page = f2fs_grab_cache_page(mapping, index, false);
 	if (!page) {
 		cond_resched();
 		goto repeat;
 	}
-	f2fs_wait_on_page_writeback(page, META);
-	SetPageUptodate(page);
+	f2fs_wait_on_page_writeback(page, META, true);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
 	return page;
 }
 
@@ -55,15 +64,17 @@ static struct page *__get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index,
 	struct f2fs_io_info fio = {
 		.sbi = sbi,
 		.type = META,
-		.rw = READ_SYNC | REQ_META | REQ_PRIO,
-		.blk_addr = index,
+		.op = REQ_OP_READ,
+		.op_flags = REQ_SYNC | REQ_META | REQ_PRIO,
+		.old_blkaddr = index,
+		.new_blkaddr = index,
 		.encrypted_page = NULL,
 	};
 
 	if (unlikely(!is_meta))
-		fio.rw &= ~REQ_META;
+		fio.op_flags &= ~REQ_META;
 repeat:
-	page = grab_cache_page(mapping, index);
+	page = f2fs_grab_cache_page(mapping, index, false);
 	if (!page) {
 		cond_resched();
 		goto repeat;
@@ -90,7 +101,7 @@ repeat:
 	 * meta page.
 	 */
 	if (unlikely(!PageUptodate(page)))
-		f2fs_stop_checkpoint(sbi);
+		f2fs_stop_checkpoint(sbi, false);
 out:
 	return page;
 }
@@ -143,19 +154,23 @@ bool is_valid_blkaddr(struct f2fs_sb_info *sbi, block_t blkaddr, int type)
 int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages,
 							int type, bool sync)
 {
-	block_t prev_blk_addr = 0;
 	struct page *page;
 	block_t blkno = start;
 	struct f2fs_io_info fio = {
 		.sbi = sbi,
 		.type = META,
-		.rw = sync ? (READ_SYNC | REQ_META | REQ_PRIO) : READA,
+		.op = REQ_OP_READ,
+		.op_flags = sync ? (REQ_SYNC | REQ_META | REQ_PRIO) :
+						REQ_RAHEAD,
 		.encrypted_page = NULL,
+		.in_list = false,
 	};
+	struct blk_plug plug;
 
 	if (unlikely(type == META_POR))
-		fio.rw &= ~REQ_META;
+		fio.op_flags &= ~REQ_META;
 
+	blk_start_plug(&plug);
 	for (; nrpages-- > 0; blkno++) {
 
 		if (!is_valid_blkaddr(sbi, blkno, type))
@@ -167,27 +182,25 @@ int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages,
 					NAT_BLOCK_OFFSET(NM_I(sbi)->max_nid)))
 				blkno = 0;
 			/* get nat block addr */
-			fio.blk_addr = current_nat_addr(sbi,
+			fio.new_blkaddr = current_nat_addr(sbi,
 					blkno * NAT_ENTRY_PER_BLOCK);
 			break;
 		case META_SIT:
 			/* get sit block addr */
-			fio.blk_addr = current_sit_addr(sbi,
+			fio.new_blkaddr = current_sit_addr(sbi,
 					blkno * SIT_ENTRY_PER_BLOCK);
-			if (blkno != start && prev_blk_addr + 1 != fio.blk_addr)
-				goto out;
-			prev_blk_addr = fio.blk_addr;
 			break;
 		case META_SSA:
 		case META_CP:
 		case META_POR:
-			fio.blk_addr = blkno;
+			fio.new_blkaddr = blkno;
 			break;
 		default:
 			BUG();
 		}
 
-		page = grab_cache_page(META_MAPPING(sbi), fio.blk_addr);
+		page = f2fs_grab_cache_page(META_MAPPING(sbi),
+						fio.new_blkaddr, false);
 		if (!page)
 			continue;
 		if (PageUptodate(page)) {
@@ -196,11 +209,11 @@ int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages,
 		}
 
 		fio.page = page;
-		f2fs_submit_page_mbio(&fio);
+		f2fs_submit_page_bio(&fio);
 		f2fs_put_page(page, 0);
 	}
 out:
-	f2fs_submit_merged_bio(sbi, META, READ);
+	blk_finish_plug(&plug);
 	return blkno - start;
 }
 
@@ -210,16 +223,17 @@ void ra_meta_pages_cond(struct f2fs_sb_info *sbi, pgoff_t index)
 	bool readahead = false;
 
 	page = find_get_page(META_MAPPING(sbi), index);
-	if (!page || (page && !PageUptodate(page)))
+	if (!page || !PageUptodate(page))
 		readahead = true;
 	f2fs_put_page(page, 0);
 
 	if (readahead)
-		ra_meta_pages(sbi, index, MAX_BIO_BLOCKS(sbi), META_POR, true);
+		ra_meta_pages(sbi, index, BIO_MAX_PAGES, META_POR, true);
 }
 
-static int f2fs_write_meta_page(struct page *page,
-				struct writeback_control *wbc)
+static int __f2fs_write_meta_page(struct page *page,
+				struct writeback_control *wbc,
+				enum iostat_type io_type)
 {
 	struct f2fs_sb_info *sbi = F2FS_P_SB(page);
 
@@ -232,13 +246,18 @@ static int f2fs_write_meta_page(struct page *page,
 	if (unlikely(f2fs_cp_error(sbi)))
 		goto redirty_out;
 
-	f2fs_wait_on_page_writeback(page, META);
-	write_meta_page(sbi, page);
+	write_meta_page(sbi, page, io_type);
 	dec_page_count(sbi, F2FS_DIRTY_META);
-	unlock_page(page);
 
 	if (wbc->for_reclaim)
-		f2fs_submit_merged_bio(sbi, META, WRITE);
+		f2fs_submit_merged_write_cond(sbi, page->mapping->host,
+						0, page->index, META);
+
+	unlock_page(page);
+
+	if (unlikely(f2fs_cp_error(sbi)))
+		f2fs_submit_merged_write(sbi, META);
+
 	return 0;
 
 redirty_out:
@@ -246,45 +265,59 @@ redirty_out:
 	return AOP_WRITEPAGE_ACTIVATE;
 }
 
+static int f2fs_write_meta_page(struct page *page,
+				struct writeback_control *wbc)
+{
+	return __f2fs_write_meta_page(page, wbc, FS_META_IO);
+}
+
 static int f2fs_write_meta_pages(struct address_space *mapping,
 				struct writeback_control *wbc)
 {
 	struct f2fs_sb_info *sbi = F2FS_M_SB(mapping);
 	long diff, written;
 
-	trace_f2fs_writepages(mapping->host, wbc, META);
+	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+		goto skip_write;
 
 	/* collect a number of dirty meta pages and write together */
 	if (wbc->for_kupdate ||
 		get_pages(sbi, F2FS_DIRTY_META) < nr_pages_to_skip(sbi, META))
 		goto skip_write;
 
-	/* if mounting is failed, skip writing node pages */
-	mutex_lock(&sbi->cp_mutex);
+	/* if locked failed, cp will flush dirty pages instead */
+	if (!mutex_trylock(&sbi->cp_mutex))
+		goto skip_write;
+
+	trace_f2fs_writepages(mapping->host, wbc, META);
 	diff = nr_pages_to_write(sbi, META, wbc);
-	written = sync_meta_pages(sbi, META, wbc->nr_to_write);
+	written = sync_meta_pages(sbi, META, wbc->nr_to_write, FS_META_IO);
 	mutex_unlock(&sbi->cp_mutex);
 	wbc->nr_to_write = max((long)0, wbc->nr_to_write - written - diff);
 	return 0;
 
 skip_write:
 	wbc->pages_skipped += get_pages(sbi, F2FS_DIRTY_META);
+	trace_f2fs_writepages(mapping->host, wbc, META);
 	return 0;
 }
 
 long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,
-						long nr_to_write)
+				long nr_to_write, enum iostat_type io_type)
 {
 	struct address_space *mapping = META_MAPPING(sbi);
-	pgoff_t index = 0, end = LONG_MAX, prev = LONG_MAX;
+	pgoff_t index = 0, end = ULONG_MAX, prev = ULONG_MAX;
 	struct pagevec pvec;
 	long nwritten = 0;
 	struct writeback_control wbc = {
 		.for_reclaim = 0,
 	};
+	struct blk_plug plug;
 
 	pagevec_init(&pvec, 0);
 
+	blk_start_plug(&plug);
+
 	while (index <= end) {
 		int i, nr_pages;
 		nr_pages = pagevec_lookup_tag(&pvec, mapping, &index,
@@ -296,7 +329,7 @@ long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
 
-			if (prev == LONG_MAX)
+			if (prev == ULONG_MAX)
 				prev = page->index - 1;
 			if (nr_to_write != LONG_MAX && page->index != prev + 1) {
 				pagevec_release(&pvec);
@@ -315,10 +348,13 @@ continue_unlock:
 				goto continue_unlock;
 			}
 
+			f2fs_wait_on_page_writeback(page, META, true);
+
+			BUG_ON(PageWriteback(page));
 			if (!clear_page_dirty_for_io(page))
 				goto continue_unlock;
 
-			if (mapping->a_ops->writepage(page, &wbc)) {
+			if (__f2fs_write_meta_page(page, &wbc, io_type)) {
 				unlock_page(page);
 				break;
 			}
@@ -332,7 +368,9 @@ continue_unlock:
 	}
 stop:
 	if (nwritten)
-		f2fs_submit_merged_bio(sbi, type, WRITE);
+		f2fs_submit_merged_write(sbi, type);
+
+	blk_finish_plug(&plug);
 
 	return nwritten;
 }
@@ -341,9 +379,10 @@ static int f2fs_set_meta_page_dirty(struct page *page)
 {
 	trace_f2fs_set_page_dirty(page, META);
 
-	SetPageUptodate(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
 	if (!PageDirty(page)) {
-		__set_page_dirty_nobuffers(page);
+		f2fs_set_page_dirty_nobuffers(page);
 		inc_page_count(F2FS_P_SB(page), F2FS_DIRTY_META);
 		SetPagePrivate(page);
 		f2fs_trace_pid(page);
@@ -358,6 +397,9 @@ const struct address_space_operations f2fs_meta_aops = {
 	.set_page_dirty	= f2fs_set_meta_page_dirty,
 	.invalidatepage = f2fs_invalidate_page,
 	.releasepage	= f2fs_release_page,
+#ifdef CONFIG_MIGRATION
+	.migratepage    = f2fs_migrate_page,
+#endif
 };
 
 static void __add_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type)
@@ -410,13 +452,13 @@ static void __remove_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type)
 	spin_unlock(&im->ino_lock);
 }
 
-void add_dirty_inode(struct f2fs_sb_info *sbi, nid_t ino, int type)
+void add_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type)
 {
 	/* add new dirty ino entry into list */
 	__add_ino_entry(sbi, ino, type);
 }
 
-void remove_dirty_inode(struct f2fs_sb_info *sbi, nid_t ino, int type)
+void remove_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type)
 {
 	/* remove dirty ino entry from list */
 	__remove_ino_entry(sbi, ino, type);
@@ -434,12 +476,12 @@ bool exist_written_data(struct f2fs_sb_info *sbi, nid_t ino, int mode)
 	return e ? true : false;
 }
 
-void release_dirty_inode(struct f2fs_sb_info *sbi)
+void release_ino_entry(struct f2fs_sb_info *sbi, bool all)
 {
 	struct ino_entry *e, *tmp;
 	int i;
 
-	for (i = APPEND_INO; i <= UPDATE_INO; i++) {
+	for (i = all ? ORPHAN_INO: APPEND_INO; i <= UPDATE_INO; i++) {
 		struct inode_management *im = &sbi->im[i];
 
 		spin_lock(&im->ino_lock);
@@ -459,6 +501,14 @@ int acquire_orphan_inode(struct f2fs_sb_info *sbi)
 	int err = 0;
 
 	spin_lock(&im->ino_lock);
+
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_ORPHAN)) {
+		spin_unlock(&im->ino_lock);
+		f2fs_show_injection_info(FAULT_ORPHAN);
+		return -ENOSPC;
+	}
+#endif
 	if (unlikely(im->ino_num >= sbi->max_orphans))
 		err = -ENOSPC;
 	else
@@ -478,10 +528,11 @@ void release_orphan_inode(struct f2fs_sb_info *sbi)
 	spin_unlock(&im->ino_lock);
 }
 
-void add_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
+void add_orphan_inode(struct inode *inode)
 {
 	/* add new orphan ino entry into list */
-	__add_ino_entry(sbi, ino, ORPHAN_INO);
+	__add_ino_entry(F2FS_I_SB(inode), inode->i_ino, ORPHAN_INO);
+	update_inode_page(inode);
 }
 
 void remove_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
@@ -493,8 +544,20 @@ void remove_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 static int recover_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 {
 	struct inode *inode;
+	struct node_info ni;
+	int err = acquire_orphan_inode(sbi);
+
+	if (err) {
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
+		f2fs_msg(sbi->sb, KERN_WARNING,
+				"%s: orphan failed (ino=%x), run fsck to fix.",
+				__func__, ino);
+		return err;
+	}
 
-	inode = f2fs_iget(sbi->sb, ino);
+	__add_ino_entry(sbi, ino, ORPHAN_INO);
+
+	inode = f2fs_iget_retry(sbi->sb, ino);
 	if (IS_ERR(inode)) {
 		/*
 		 * there should be a bug that we can't find the entry
@@ -508,17 +571,42 @@ static int recover_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino)
 
 	/* truncate all the data during iput */
 	iput(inode);
+
+	get_node_info(sbi, ino, &ni);
+
+	/* ENOMEM was fully retried in f2fs_evict_inode. */
+	if (ni.blk_addr != NULL_ADDR) {
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
+		f2fs_msg(sbi->sb, KERN_WARNING,
+			"%s: orphan failed (ino=%x) by kernel, retry mount.",
+				__func__, ino);
+		return -EIO;
+	}
+	__remove_ino_entry(sbi, ino, ORPHAN_INO);
 	return 0;
 }
 
 int recover_orphan_inodes(struct f2fs_sb_info *sbi)
 {
 	block_t start_blk, orphan_blocks, i, j;
-	int err;
+	unsigned int s_flags = sbi->sb->s_flags;
+	int err = 0;
 
-	if (!is_set_ckpt_flags(F2FS_CKPT(sbi), CP_ORPHAN_PRESENT_FLAG))
+	if (!is_set_ckpt_flags(sbi, CP_ORPHAN_PRESENT_FLAG))
 		return 0;
 
+	if (s_flags & MS_RDONLY) {
+		f2fs_msg(sbi->sb, KERN_INFO, "orphan cleanup on readonly fs");
+		sbi->sb->s_flags &= ~MS_RDONLY;
+	}
+
+#ifdef CONFIG_QUOTA
+	/* Needed for iput() to work correctly and not trash data */
+	sbi->sb->s_flags |= MS_ACTIVE;
+	/* Turn on quotas so that they are updated correctly */
+	f2fs_enable_quota_files(sbi);
+#endif
+
 	start_blk = __start_cp_addr(sbi) + 1 + __cp_payload(sbi);
 	orphan_blocks = __start_sum_addr(sbi) - 1 - __cp_payload(sbi);
 
@@ -534,14 +622,21 @@ int recover_orphan_inodes(struct f2fs_sb_info *sbi)
 			err = recover_orphan_inode(sbi, ino);
 			if (err) {
 				f2fs_put_page(page, 1);
-				return err;
+				goto out;
 			}
 		}
 		f2fs_put_page(page, 1);
 	}
 	/* clear Orphan Flag */
-	clear_ckpt_flags(F2FS_CKPT(sbi), CP_ORPHAN_PRESENT_FLAG);
-	return 0;
+	clear_ckpt_flags(sbi, CP_ORPHAN_PRESENT_FLAG);
+out:
+#ifdef CONFIG_QUOTA
+	/* Turn quotas off */
+	f2fs_quota_off_umount(sbi->sb);
+#endif
+	sbi->sb->s_flags = s_flags; /* Restore MS_RDONLY status */
+
+	return err;
 }
 
 static void write_orphan_inodes(struct f2fs_sb_info *sbi, block_t start_blk)
@@ -601,45 +696,54 @@ static void write_orphan_inodes(struct f2fs_sb_info *sbi, block_t start_blk)
 	}
 }
 
-static struct page *validate_checkpoint(struct f2fs_sb_info *sbi,
-				block_t cp_addr, unsigned long long *version)
+static int get_checkpoint_version(struct f2fs_sb_info *sbi, block_t cp_addr,
+		struct f2fs_checkpoint **cp_block, struct page **cp_page,
+		unsigned long long *version)
 {
-	struct page *cp_page_1, *cp_page_2 = NULL;
 	unsigned long blk_size = sbi->blocksize;
-	struct f2fs_checkpoint *cp_block;
-	unsigned long long cur_version = 0, pre_version = 0;
-	size_t crc_offset;
+	size_t crc_offset = 0;
 	__u32 crc = 0;
 
-	/* Read the 1st cp block in this CP pack */
-	cp_page_1 = get_meta_page(sbi, cp_addr);
+	*cp_page = get_meta_page(sbi, cp_addr);
+	*cp_block = (struct f2fs_checkpoint *)page_address(*cp_page);
 
-	/* get the version number */
-	cp_block = (struct f2fs_checkpoint *)page_address(cp_page_1);
-	crc_offset = le32_to_cpu(cp_block->checksum_offset);
-	if (crc_offset >= blk_size)
-		goto invalid_cp1;
+	crc_offset = le32_to_cpu((*cp_block)->checksum_offset);
+	if (crc_offset > (blk_size - sizeof(__le32))) {
+		f2fs_msg(sbi->sb, KERN_WARNING,
+			"invalid crc_offset: %zu", crc_offset);
+		return -EINVAL;
+	}
 
-	crc = le32_to_cpu(*((__le32 *)((unsigned char *)cp_block + crc_offset)));
-	if (!f2fs_crc_valid(crc, cp_block, crc_offset))
-		goto invalid_cp1;
+	crc = cur_cp_crc(*cp_block);
+	if (!f2fs_crc_valid(sbi, crc, *cp_block, crc_offset)) {
+		f2fs_msg(sbi->sb, KERN_WARNING, "invalid crc value");
+		return -EINVAL;
+	}
 
-	pre_version = cur_cp_version(cp_block);
+	*version = cur_cp_version(*cp_block);
+	return 0;
+}
 
-	/* Read the 2nd cp block in this CP pack */
-	cp_addr += le32_to_cpu(cp_block->cp_pack_total_block_count) - 1;
-	cp_page_2 = get_meta_page(sbi, cp_addr);
+static struct page *validate_checkpoint(struct f2fs_sb_info *sbi,
+				block_t cp_addr, unsigned long long *version)
+{
+	struct page *cp_page_1 = NULL, *cp_page_2 = NULL;
+	struct f2fs_checkpoint *cp_block = NULL;
+	unsigned long long cur_version = 0, pre_version = 0;
+	int err;
 
-	cp_block = (struct f2fs_checkpoint *)page_address(cp_page_2);
-	crc_offset = le32_to_cpu(cp_block->checksum_offset);
-	if (crc_offset >= blk_size)
-		goto invalid_cp2;
+	err = get_checkpoint_version(sbi, cp_addr, &cp_block,
+					&cp_page_1, version);
+	if (err)
+		goto invalid_cp1;
+	pre_version = *version;
 
-	crc = le32_to_cpu(*((__le32 *)((unsigned char *)cp_block + crc_offset)));
-	if (!f2fs_crc_valid(crc, cp_block, crc_offset))
+	cp_addr += le32_to_cpu(cp_block->cp_pack_total_block_count) - 1;
+	err = get_checkpoint_version(sbi, cp_addr, &cp_block,
+					&cp_page_2, version);
+	if (err)
 		goto invalid_cp2;
-
-	cur_version = cur_cp_version(cp_block);
+	cur_version = *version;
 
 	if (cur_version == pre_version) {
 		*version = cur_version;
@@ -696,6 +800,15 @@ int get_valid_checkpoint(struct f2fs_sb_info *sbi)
 	cp_block = (struct f2fs_checkpoint *)page_address(cur_page);
 	memcpy(sbi->ckpt, cp_block, blk_size);
 
+	/* Sanity checking of checkpoint */
+	if (sanity_check_ckpt(sbi))
+		goto free_fail_no_cp;
+
+	if (cur_page == cp1)
+		sbi->cur_cp_pack = 1;
+	else
+		sbi->cur_cp_pack = 2;
+
 	if (cp_blks <= 1)
 		goto done;
 
@@ -717,137 +830,180 @@ done:
 	f2fs_put_page(cp2, 1);
 	return 0;
 
+free_fail_no_cp:
+	f2fs_put_page(cp1, 1);
+	f2fs_put_page(cp2, 1);
 fail_no_cp:
 	kfree(sbi->ckpt);
 	return -EINVAL;
 }
 
-static int __add_dirty_inode(struct inode *inode, struct inode_entry *new)
+static void __add_dirty_inode(struct inode *inode, enum inode_type type)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	int flag = (type == DIR_INODE) ? FI_DIRTY_DIR : FI_DIRTY_FILE;
 
-	if (is_inode_flag_set(F2FS_I(inode), FI_DIRTY_DIR))
-		return -EEXIST;
+	if (is_inode_flag_set(inode, flag))
+		return;
 
-	set_inode_flag(F2FS_I(inode), FI_DIRTY_DIR);
-	F2FS_I(inode)->dirty_dir = new;
-	list_add_tail(&new->list, &sbi->dir_inode_list);
-	stat_inc_dirty_dir(sbi);
-	return 0;
+	set_inode_flag(inode, flag);
+	if (!f2fs_is_volatile_file(inode))
+		list_add_tail(&F2FS_I(inode)->dirty_list,
+						&sbi->inode_list[type]);
+	stat_inc_dirty_inode(sbi, type);
+}
+
+static void __remove_dirty_inode(struct inode *inode, enum inode_type type)
+{
+	int flag = (type == DIR_INODE) ? FI_DIRTY_DIR : FI_DIRTY_FILE;
+
+	if (get_dirty_pages(inode) || !is_inode_flag_set(inode, flag))
+		return;
+
+	list_del_init(&F2FS_I(inode)->dirty_list);
+	clear_inode_flag(inode, flag);
+	stat_dec_dirty_inode(F2FS_I_SB(inode), type);
 }
 
 void update_dirty_page(struct inode *inode, struct page *page)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct inode_entry *new;
-	int ret = 0;
+	enum inode_type type = S_ISDIR(inode->i_mode) ? DIR_INODE : FILE_INODE;
 
 	if (!S_ISDIR(inode->i_mode) && !S_ISREG(inode->i_mode) &&
 			!S_ISLNK(inode->i_mode))
 		return;
 
-	if (!S_ISDIR(inode->i_mode)) {
-		inode_inc_dirty_pages(inode);
-		goto out;
-	}
-
-	new = f2fs_kmem_cache_alloc(inode_entry_slab, GFP_NOFS);
-	new->inode = inode;
-	INIT_LIST_HEAD(&new->list);
-
-	spin_lock(&sbi->dir_inode_lock);
-	ret = __add_dirty_inode(inode, new);
+	spin_lock(&sbi->inode_lock[type]);
+	if (type != FILE_INODE || test_opt(sbi, DATA_FLUSH))
+		__add_dirty_inode(inode, type);
 	inode_inc_dirty_pages(inode);
-	spin_unlock(&sbi->dir_inode_lock);
+	spin_unlock(&sbi->inode_lock[type]);
 
-	if (ret)
-		kmem_cache_free(inode_entry_slab, new);
-out:
 	SetPagePrivate(page);
 	f2fs_trace_pid(page);
 }
 
-void add_dirty_dir_inode(struct inode *inode)
-{
-	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct inode_entry *new =
-			f2fs_kmem_cache_alloc(inode_entry_slab, GFP_NOFS);
-	int ret = 0;
-
-	new->inode = inode;
-	INIT_LIST_HEAD(&new->list);
-
-	spin_lock(&sbi->dir_inode_lock);
-	ret = __add_dirty_inode(inode, new);
-	spin_unlock(&sbi->dir_inode_lock);
-
-	if (ret)
-		kmem_cache_free(inode_entry_slab, new);
-}
-
-void remove_dirty_dir_inode(struct inode *inode)
+void remove_dirty_inode(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct inode_entry *entry;
+	enum inode_type type = S_ISDIR(inode->i_mode) ? DIR_INODE : FILE_INODE;
 
-	if (!S_ISDIR(inode->i_mode))
+	if (!S_ISDIR(inode->i_mode) && !S_ISREG(inode->i_mode) &&
+			!S_ISLNK(inode->i_mode))
 		return;
 
-	spin_lock(&sbi->dir_inode_lock);
-	if (get_dirty_pages(inode) ||
-			!is_inode_flag_set(F2FS_I(inode), FI_DIRTY_DIR)) {
-		spin_unlock(&sbi->dir_inode_lock);
+	if (type == FILE_INODE && !test_opt(sbi, DATA_FLUSH))
 		return;
-	}
 
-	entry = F2FS_I(inode)->dirty_dir;
-	list_del(&entry->list);
-	F2FS_I(inode)->dirty_dir = NULL;
-	clear_inode_flag(F2FS_I(inode), FI_DIRTY_DIR);
-	stat_dec_dirty_dir(sbi);
-	spin_unlock(&sbi->dir_inode_lock);
-	kmem_cache_free(inode_entry_slab, entry);
-
-	/* Only from the recovery routine */
-	if (is_inode_flag_set(F2FS_I(inode), FI_DELAY_IPUT)) {
-		clear_inode_flag(F2FS_I(inode), FI_DELAY_IPUT);
-		iput(inode);
-	}
+	spin_lock(&sbi->inode_lock[type]);
+	__remove_dirty_inode(inode, type);
+	spin_unlock(&sbi->inode_lock[type]);
 }
 
-void sync_dirty_dir_inodes(struct f2fs_sb_info *sbi)
+int sync_dirty_inodes(struct f2fs_sb_info *sbi, enum inode_type type)
 {
 	struct list_head *head;
-	struct inode_entry *entry;
 	struct inode *inode;
+	struct f2fs_inode_info *fi;
+	bool is_dir = (type == DIR_INODE);
+	unsigned long ino = 0;
+
+	trace_f2fs_sync_dirty_inodes_enter(sbi->sb, is_dir,
+				get_pages(sbi, is_dir ?
+				F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA));
 retry:
 	if (unlikely(f2fs_cp_error(sbi)))
-		return;
+		return -EIO;
 
-	spin_lock(&sbi->dir_inode_lock);
+	spin_lock(&sbi->inode_lock[type]);
 
-	head = &sbi->dir_inode_list;
+	head = &sbi->inode_list[type];
 	if (list_empty(head)) {
-		spin_unlock(&sbi->dir_inode_lock);
-		return;
+		spin_unlock(&sbi->inode_lock[type]);
+		trace_f2fs_sync_dirty_inodes_exit(sbi->sb, is_dir,
+				get_pages(sbi, is_dir ?
+				F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA));
+		return 0;
 	}
-	entry = list_entry(head->next, struct inode_entry, list);
-	inode = igrab(entry->inode);
-	spin_unlock(&sbi->dir_inode_lock);
+	fi = list_first_entry(head, struct f2fs_inode_info, dirty_list);
+	inode = igrab(&fi->vfs_inode);
+	spin_unlock(&sbi->inode_lock[type]);
 	if (inode) {
+		unsigned long cur_ino = inode->i_ino;
+
+		if (is_dir)
+			F2FS_I(inode)->cp_task = current;
+
 		filemap_fdatawrite(inode->i_mapping);
+
+		if (is_dir)
+			F2FS_I(inode)->cp_task = NULL;
+
 		iput(inode);
+		/* We need to give cpu to another writers. */
+		if (ino == cur_ino) {
+			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			cond_resched();
+		} else {
+			ino = cur_ino;
+		}
 	} else {
 		/*
 		 * We should submit bio, since it exists several
 		 * wribacking dentry pages in the freeing inode.
 		 */
-		f2fs_submit_merged_bio(sbi, DATA, WRITE);
+		f2fs_submit_merged_write(sbi, DATA);
 		cond_resched();
 	}
 	goto retry;
 }
 
+int f2fs_sync_inode_meta(struct f2fs_sb_info *sbi)
+{
+	struct list_head *head = &sbi->inode_list[DIRTY_META];
+	struct inode *inode;
+	struct f2fs_inode_info *fi;
+	s64 total = get_pages(sbi, F2FS_DIRTY_IMETA);
+
+	while (total--) {
+		if (unlikely(f2fs_cp_error(sbi)))
+			return -EIO;
+
+		spin_lock(&sbi->inode_lock[DIRTY_META]);
+		if (list_empty(head)) {
+			spin_unlock(&sbi->inode_lock[DIRTY_META]);
+			return 0;
+		}
+		fi = list_first_entry(head, struct f2fs_inode_info,
+							gdirty_list);
+		inode = igrab(&fi->vfs_inode);
+		spin_unlock(&sbi->inode_lock[DIRTY_META]);
+		if (inode) {
+			sync_inode_metadata(inode, 0);
+
+			/* it's on eviction */
+			if (is_inode_flag_set(inode, FI_DIRTY_INODE))
+				update_inode_page(inode);
+			iput(inode);
+		}
+	};
+	return 0;
+}
+
+static void __prepare_cp_block(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	nid_t last_nid = nm_i->next_scan_nid;
+
+	next_free_nid(sbi, &last_nid);
+	ckpt->valid_block_count = cpu_to_le64(valid_user_blocks(sbi));
+	ckpt->valid_node_count = cpu_to_le32(valid_node_count(sbi));
+	ckpt->valid_inode_count = cpu_to_le32(valid_inode_count(sbi));
+	ckpt->next_free_nid = cpu_to_le32(last_nid);
+}
+
 /*
  * Freeze all the FS-operations for checkpoint.
  */
@@ -868,31 +1024,50 @@ retry_flush_dents:
 	/* write all the dirty dentry pages */
 	if (get_pages(sbi, F2FS_DIRTY_DENTS)) {
 		f2fs_unlock_all(sbi);
-		sync_dirty_dir_inodes(sbi);
-		if (unlikely(f2fs_cp_error(sbi))) {
-			err = -EIO;
+		err = sync_dirty_inodes(sbi, DIR_INODE);
+		if (err)
 			goto out;
-		}
+		cond_resched();
 		goto retry_flush_dents;
 	}
 
 	/*
 	 * POR: we should ensure that there are no dirty node pages
-	 * until finishing nat/sit flush.
+	 * until finishing nat/sit flush. inode->i_blocks can be updated.
 	 */
+	down_write(&sbi->node_change);
+
+	if (get_pages(sbi, F2FS_DIRTY_IMETA)) {
+		up_write(&sbi->node_change);
+		f2fs_unlock_all(sbi);
+		err = f2fs_sync_inode_meta(sbi);
+		if (err)
+			goto out;
+		cond_resched();
+		goto retry_flush_dents;
+	}
+
 retry_flush_nodes:
 	down_write(&sbi->node_write);
 
 	if (get_pages(sbi, F2FS_DIRTY_NODES)) {
 		up_write(&sbi->node_write);
-		sync_node_pages(sbi, 0, &wbc);
-		if (unlikely(f2fs_cp_error(sbi))) {
+		err = sync_node_pages(sbi, &wbc, false, FS_CP_NODE_IO);
+		if (err) {
+			up_write(&sbi->node_change);
 			f2fs_unlock_all(sbi);
-			err = -EIO;
 			goto out;
 		}
+		cond_resched();
 		goto retry_flush_nodes;
 	}
+
+	/*
+	 * sbi->node_change is used only for AIO write_begin path which produces
+	 * dirty node blocks and some checkpoint values by block allocation.
+	 */
+	__prepare_cp_block(sbi);
+	up_write(&sbi->node_change);
 out:
 	blk_finish_plug(&plug);
 	return err;
@@ -911,51 +1086,80 @@ static void wait_on_all_pages_writeback(struct f2fs_sb_info *sbi)
 	for (;;) {
 		prepare_to_wait(&sbi->cp_wait, &wait, TASK_UNINTERRUPTIBLE);
 
-		if (!get_pages(sbi, F2FS_WRITEBACK))
+		if (!get_pages(sbi, F2FS_WB_CP_DATA))
 			break;
 
-		io_schedule();
+		io_schedule_timeout(5*HZ);
 	}
 	finish_wait(&sbi->cp_wait, &wait);
 }
 
-static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
+static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc)
+{
+	unsigned long orphan_num = sbi->im[ORPHAN_INO].ino_num;
+	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
+	unsigned long flags;
+
+	spin_lock_irqsave(&sbi->cp_lock, flags);
+
+	if ((cpc->reason & CP_UMOUNT) &&
+			le32_to_cpu(ckpt->cp_pack_total_block_count) >
+			sbi->blocks_per_seg - NM_I(sbi)->nat_bits_blocks)
+		disable_nat_bits(sbi, false);
+
+	if (cpc->reason & CP_TRIMMED)
+		__set_ckpt_flags(ckpt, CP_TRIMMED_FLAG);
+
+	if (cpc->reason & CP_UMOUNT)
+		__set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
+	else
+		__clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
+
+	if (cpc->reason & CP_FASTBOOT)
+		__set_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
+	else
+		__clear_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
+
+	if (orphan_num)
+		__set_ckpt_flags(ckpt, CP_ORPHAN_PRESENT_FLAG);
+	else
+		__clear_ckpt_flags(ckpt, CP_ORPHAN_PRESENT_FLAG);
+
+	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK))
+		__set_ckpt_flags(ckpt, CP_FSCK_FLAG);
+
+	/* set this flag to activate crc|cp_ver for recovery */
+	__set_ckpt_flags(ckpt, CP_CRC_RECOVERY_FLAG);
+
+	spin_unlock_irqrestore(&sbi->cp_lock, flags);
+}
+
+static int do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 {
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
-	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
-	unsigned long orphan_num = sbi->im[ORPHAN_INO].ino_num;
-	nid_t last_nid = nm_i->next_scan_nid;
+	unsigned long orphan_num = sbi->im[ORPHAN_INO].ino_num, flags;
 	block_t start_blk;
 	unsigned int data_sum_blocks, orphan_blocks;
 	__u32 crc32 = 0;
 	int i;
 	int cp_payload_blks = __cp_payload(sbi);
-	block_t discard_blk = NEXT_FREE_BLKADDR(sbi, curseg);
-	bool invalidate = false;
-
-	/*
-	 * This avoids to conduct wrong roll-forward operations and uses
-	 * metapages, so should be called prior to sync_meta_pages below.
-	 */
-	if (discard_next_dnode(sbi, discard_blk))
-		invalidate = true;
+	struct super_block *sb = sbi->sb;
+	struct curseg_info *seg_i = CURSEG_I(sbi, CURSEG_HOT_NODE);
+	u64 kbytes_written;
 
 	/* Flush all the NAT/SIT pages */
 	while (get_pages(sbi, F2FS_DIRTY_META)) {
-		sync_meta_pages(sbi, META, LONG_MAX);
+		sync_meta_pages(sbi, META, LONG_MAX, FS_CP_META_IO);
 		if (unlikely(f2fs_cp_error(sbi)))
-			return;
+			return -EIO;
 	}
 
-	next_free_nid(sbi, &last_nid);
-
 	/*
 	 * modify checkpoint
 	 * version number is already updated
 	 */
 	ckpt->elapsed_time = cpu_to_le64(get_mtime(sbi));
-	ckpt->valid_block_count = cpu_to_le64(valid_user_blocks(sbi));
 	ckpt->free_segment_count = cpu_to_le32(free_segments(sbi));
 	for (i = 0; i < NR_CURSEG_NODE_TYPE; i++) {
 		ckpt->cur_node_segno[i] =
@@ -974,16 +1178,14 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 				curseg_alloc_type(sbi, i + CURSEG_HOT_DATA);
 	}
 
-	ckpt->valid_node_count = cpu_to_le32(valid_node_count(sbi));
-	ckpt->valid_inode_count = cpu_to_le32(valid_inode_count(sbi));
-	ckpt->next_free_nid = cpu_to_le32(last_nid);
-
 	/* 2 cp  + n data seg summary + orphan inode blocks */
 	data_sum_blocks = npages_for_summary_flush(sbi, false);
+	spin_lock_irqsave(&sbi->cp_lock, flags);
 	if (data_sum_blocks < NR_CURSEG_DATA_TYPE)
-		set_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
+		__set_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
 	else
-		clear_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
+		__clear_ckpt_flags(ckpt, CP_COMPACT_SUM_FLAG);
+	spin_unlock_irqrestore(&sbi->cp_lock, flags);
 
 	orphan_blocks = GET_ORPHAN_BLOCKS(orphan_num);
 	ckpt->cp_pack_start_sum = cpu_to_le32(1 + cp_payload_blks +
@@ -998,39 +1200,45 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 				cp_payload_blks + data_sum_blocks +
 				orphan_blocks);
 
-	if (cpc->reason == CP_UMOUNT)
-		set_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
-	else
-		clear_ckpt_flags(ckpt, CP_UMOUNT_FLAG);
-
-	if (cpc->reason == CP_FASTBOOT)
-		set_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
-	else
-		clear_ckpt_flags(ckpt, CP_FASTBOOT_FLAG);
-
-	if (orphan_num)
-		set_ckpt_flags(ckpt, CP_ORPHAN_PRESENT_FLAG);
-	else
-		clear_ckpt_flags(ckpt, CP_ORPHAN_PRESENT_FLAG);
-
-	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK))
-		set_ckpt_flags(ckpt, CP_FSCK_FLAG);
+	/* update ckpt flag for checkpoint */
+	update_ckpt_flags(sbi, cpc);
 
 	/* update SIT/NAT bitmap */
 	get_sit_bitmap(sbi, __bitmap_ptr(sbi, SIT_BITMAP));
 	get_nat_bitmap(sbi, __bitmap_ptr(sbi, NAT_BITMAP));
 
-	crc32 = f2fs_crc32(ckpt, le32_to_cpu(ckpt->checksum_offset));
+	crc32 = f2fs_crc32(sbi, ckpt, le32_to_cpu(ckpt->checksum_offset));
 	*((__le32 *)((unsigned char *)ckpt +
 				le32_to_cpu(ckpt->checksum_offset)))
 				= cpu_to_le32(crc32);
 
-	start_blk = __start_cp_addr(sbi);
+	start_blk = __start_cp_next_addr(sbi);
+
+	/* write nat bits */
+	if (enabled_nat_bits(sbi, cpc)) {
+		__u64 cp_ver = cur_cp_version(ckpt);
+		block_t blk;
+
+		cp_ver |= ((__u64)crc32 << 32);
+		*(__le64 *)nm_i->nat_bits = cpu_to_le64(cp_ver);
+
+		blk = start_blk + sbi->blocks_per_seg - nm_i->nat_bits_blocks;
+		for (i = 0; i < nm_i->nat_bits_blocks; i++)
+			update_meta_page(sbi, nm_i->nat_bits +
+					(i << F2FS_BLKSIZE_BITS), blk + i);
+
+		/* Flush all the NAT BITS pages */
+		while (get_pages(sbi, F2FS_DIRTY_META)) {
+			sync_meta_pages(sbi, META, LONG_MAX, FS_CP_META_IO);
+			if (unlikely(f2fs_cp_error(sbi)))
+				return -EIO;
+		}
+	}
 
 	/* need to wait for end_io results */
 	wait_on_all_pages_writeback(sbi);
 	if (unlikely(f2fs_cp_error(sbi)))
-		return;
+		return -EIO;
 
 	/* write out checkpoint buffer at block 0 */
 	update_meta_page(sbi, ckpt, start_blk++);
@@ -1046,6 +1254,14 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 
 	write_data_summaries(sbi, start_blk);
 	start_blk += data_sum_blocks;
+
+	/* Record write statistics in the hot node summary */
+	kbytes_written = sbi->kbytes_written;
+	if (sb->s_bdev->bd_part)
+		kbytes_written += BD_PART_WRITTEN(sbi);
+
+	seg_i->journal->info.kbytes_written = cpu_to_le64(kbytes_written);
+
 	if (__remain_node_summaries(cpc->reason)) {
 		write_node_summaries(sbi, start_blk);
 		start_blk += NR_CURSEG_NODE_TYPE;
@@ -1058,67 +1274,93 @@ static void do_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	wait_on_all_pages_writeback(sbi);
 
 	if (unlikely(f2fs_cp_error(sbi)))
-		return;
+		return -EIO;
 
-	filemap_fdatawait_range(NODE_MAPPING(sbi), 0, LONG_MAX);
-	filemap_fdatawait_range(META_MAPPING(sbi), 0, LONG_MAX);
+	filemap_fdatawait_range(NODE_MAPPING(sbi), 0, LLONG_MAX);
+	filemap_fdatawait_range(META_MAPPING(sbi), 0, LLONG_MAX);
 
 	/* update user_block_counts */
 	sbi->last_valid_block_count = sbi->total_valid_block_count;
-	sbi->alloc_valid_block_count = 0;
+	percpu_counter_set(&sbi->alloc_valid_block_count, 0);
 
 	/* Here, we only have one bio having CP pack */
-	sync_meta_pages(sbi, META_FLUSH, LONG_MAX);
+	sync_meta_pages(sbi, META_FLUSH, LONG_MAX, FS_CP_META_IO);
 
 	/* wait for previous submitted meta pages writeback */
 	wait_on_all_pages_writeback(sbi);
 
-	/*
-	 * invalidate meta page which is used temporarily for zeroing out
-	 * block at the end of warm node chain.
-	 */
-	if (invalidate)
-		invalidate_mapping_pages(META_MAPPING(sbi), discard_blk,
-								discard_blk);
-
-	release_dirty_inode(sbi);
+	release_ino_entry(sbi, false);
 
 	if (unlikely(f2fs_cp_error(sbi)))
-		return;
+		return -EIO;
 
-	clear_prefree_segments(sbi, cpc);
 	clear_sbi_flag(sbi, SBI_IS_DIRTY);
+	clear_sbi_flag(sbi, SBI_NEED_CP);
+	__set_cp_next_pack(sbi);
+
+	/*
+	 * redirty superblock if metadata like node page or inode cache is
+	 * updated during writing checkpoint.
+	 */
+	if (get_pages(sbi, F2FS_DIRTY_NODES) ||
+			get_pages(sbi, F2FS_DIRTY_IMETA))
+		set_sbi_flag(sbi, SBI_IS_DIRTY);
+
+	f2fs_bug_on(sbi, get_pages(sbi, F2FS_DIRTY_DENTS));
+
+	return 0;
 }
 
 /*
  * We guarantee that this checkpoint procedure will not fail.
  */
-void write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
+int write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 {
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
 	unsigned long long ckpt_ver;
+	int err = 0;
 
 	mutex_lock(&sbi->cp_mutex);
 
 	if (!is_sbi_flag_set(sbi, SBI_IS_DIRTY) &&
-		(cpc->reason == CP_FASTBOOT || cpc->reason == CP_SYNC ||
-		(cpc->reason == CP_DISCARD && !sbi->discard_blks)))
+		((cpc->reason & CP_FASTBOOT) || (cpc->reason & CP_SYNC) ||
+		((cpc->reason & CP_DISCARD) && !sbi->discard_blks)))
 		goto out;
-	if (unlikely(f2fs_cp_error(sbi)))
+	if (unlikely(f2fs_cp_error(sbi))) {
+		err = -EIO;
 		goto out;
-	if (f2fs_readonly(sbi->sb))
+	}
+	if (f2fs_readonly(sbi->sb)) {
+		err = -EROFS;
 		goto out;
+	}
 
 	trace_f2fs_write_checkpoint(sbi->sb, cpc->reason, "start block_ops");
 
-	if (block_operations(sbi))
+	err = block_operations(sbi);
+	if (err)
 		goto out;
 
 	trace_f2fs_write_checkpoint(sbi->sb, cpc->reason, "finish block_ops");
 
-	f2fs_submit_merged_bio(sbi, DATA, WRITE);
-	f2fs_submit_merged_bio(sbi, NODE, WRITE);
-	f2fs_submit_merged_bio(sbi, META, WRITE);
+	f2fs_flush_merged_writes(sbi);
+
+	/* this is the case of multiple fstrims without any changes */
+	if (cpc->reason & CP_DISCARD) {
+		if (!exist_trim_candidates(sbi, cpc)) {
+			unblock_operations(sbi);
+			goto out;
+		}
+
+		if (NM_I(sbi)->dirty_nat_cnt == 0 &&
+				SIT_I(sbi)->dirty_sentries == 0 &&
+				prefree_segments(sbi) == 0) {
+			flush_sit_entries(sbi, cpc);
+			clear_prefree_segments(sbi, cpc);
+			unblock_operations(sbi);
+			goto out;
+		}
+	}
 
 	/*
 	 * update checkpoint pack index
@@ -1129,24 +1371,29 @@ void write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	ckpt->checkpoint_ver = cpu_to_le64(++ckpt_ver);
 
 	/* write cached NAT/SIT entries to NAT/SIT area */
-	flush_nat_entries(sbi);
+	flush_nat_entries(sbi, cpc);
 	flush_sit_entries(sbi, cpc);
 
 	/* unlock all the fs_lock[] in do_checkpoint() */
-	do_checkpoint(sbi, cpc);
+	err = do_checkpoint(sbi, cpc);
+	if (err)
+		release_discard_addrs(sbi);
+	else
+		clear_prefree_segments(sbi, cpc);
 
 	unblock_operations(sbi);
 	stat_inc_cp_count(sbi->stat_info);
 
-	if (cpc->reason == CP_RECOVERY)
+	if (cpc->reason & CP_RECOVERY)
 		f2fs_msg(sbi->sb, KERN_NOTICE,
 			"checkpoint: version = %llx", ckpt_ver);
 
 	/* do checkpoint periodically */
-	sbi->cp_expires = round_jiffies_up(jiffies + HZ * sbi->cp_interval);
+	f2fs_update_time(sbi, CP_TIME);
+	trace_f2fs_write_checkpoint(sbi->sb, cpc->reason, "finish checkpoint");
 out:
 	mutex_unlock(&sbi->cp_mutex);
-	trace_f2fs_write_checkpoint(sbi->sb, cpc->reason, "finish checkpoint");
+	return err;
 }
 
 void init_ino_entry_info(struct f2fs_sb_info *sbi)
diff --git a/fs/f2fs/crypto.c b/fs/f2fs/crypto.c
deleted file mode 100644
index d879c6c846b7..000000000000
--- a/fs/f2fs/crypto.c
+++ /dev/null
@@ -1,490 +0,0 @@
-/*
- * linux/fs/f2fs/crypto.c
- *
- * Copied from linux/fs/ext4/crypto.c
- *
- * Copyright (C) 2015, Google, Inc.
- * Copyright (C) 2015, Motorola Mobility
- *
- * This contains encryption functions for f2fs
- *
- * Written by Michael Halcrow, 2014.
- *
- * Filename encryption additions
- *	Uday Savagaonkar, 2014
- * Encryption policy handling additions
- *	Ildar Muslukhov, 2014
- * Remove ext4_encrypted_zeroout(),
- *   add f2fs_restore_and_release_control_page()
- *	Jaegeuk Kim, 2015.
- *
- * This has not yet undergone a rigorous security audit.
- *
- * The usage of AES-XTS should conform to recommendations in NIST
- * Special Publication 800-38E and IEEE P1619/D16.
- */
-#include <crypto/hash.h>
-#include <crypto/sha.h>
-#include <keys/user-type.h>
-#include <keys/encrypted-type.h>
-#include <linux/crypto.h>
-#include <linux/ecryptfs.h>
-#include <linux/gfp.h>
-#include <linux/kernel.h>
-#include <linux/key.h>
-#include <linux/list.h>
-#include <linux/mempool.h>
-#include <linux/module.h>
-#include <linux/mutex.h>
-#include <linux/random.h>
-#include <linux/scatterlist.h>
-#include <linux/spinlock_types.h>
-#include <linux/f2fs_fs.h>
-#include <linux/ratelimit.h>
-#include <linux/bio.h>
-
-#include "f2fs.h"
-#include "xattr.h"
-
-/* Encryption added and removed here! (L: */
-
-static unsigned int num_prealloc_crypto_pages = 32;
-static unsigned int num_prealloc_crypto_ctxs = 128;
-
-module_param(num_prealloc_crypto_pages, uint, 0444);
-MODULE_PARM_DESC(num_prealloc_crypto_pages,
-		"Number of crypto pages to preallocate");
-module_param(num_prealloc_crypto_ctxs, uint, 0444);
-MODULE_PARM_DESC(num_prealloc_crypto_ctxs,
-		"Number of crypto contexts to preallocate");
-
-static mempool_t *f2fs_bounce_page_pool;
-
-static LIST_HEAD(f2fs_free_crypto_ctxs);
-static DEFINE_SPINLOCK(f2fs_crypto_ctx_lock);
-
-static struct workqueue_struct *f2fs_read_workqueue;
-static DEFINE_MUTEX(crypto_init);
-
-static struct kmem_cache *f2fs_crypto_ctx_cachep;
-struct kmem_cache *f2fs_crypt_info_cachep;
-
-/**
- * f2fs_release_crypto_ctx() - Releases an encryption context
- * @ctx: The encryption context to release.
- *
- * If the encryption context was allocated from the pre-allocated pool, returns
- * it to that pool. Else, frees it.
- *
- * If there's a bounce page in the context, this frees that.
- */
-void f2fs_release_crypto_ctx(struct f2fs_crypto_ctx *ctx)
-{
-	unsigned long flags;
-
-	if (ctx->flags & F2FS_WRITE_PATH_FL && ctx->w.bounce_page) {
-		mempool_free(ctx->w.bounce_page, f2fs_bounce_page_pool);
-		ctx->w.bounce_page = NULL;
-	}
-	ctx->w.control_page = NULL;
-	if (ctx->flags & F2FS_CTX_REQUIRES_FREE_ENCRYPT_FL) {
-		kmem_cache_free(f2fs_crypto_ctx_cachep, ctx);
-	} else {
-		spin_lock_irqsave(&f2fs_crypto_ctx_lock, flags);
-		list_add(&ctx->free_list, &f2fs_free_crypto_ctxs);
-		spin_unlock_irqrestore(&f2fs_crypto_ctx_lock, flags);
-	}
-}
-
-/**
- * f2fs_get_crypto_ctx() - Gets an encryption context
- * @inode:       The inode for which we are doing the crypto
- *
- * Allocates and initializes an encryption context.
- *
- * Return: An allocated and initialized encryption context on success; error
- * value or NULL otherwise.
- */
-struct f2fs_crypto_ctx *f2fs_get_crypto_ctx(struct inode *inode)
-{
-	struct f2fs_crypto_ctx *ctx = NULL;
-	unsigned long flags;
-	struct f2fs_crypt_info *ci = F2FS_I(inode)->i_crypt_info;
-
-	if (ci == NULL)
-		return ERR_PTR(-ENOKEY);
-
-	/*
-	 * We first try getting the ctx from a free list because in
-	 * the common case the ctx will have an allocated and
-	 * initialized crypto tfm, so it's probably a worthwhile
-	 * optimization. For the bounce page, we first try getting it
-	 * from the kernel allocator because that's just about as fast
-	 * as getting it from a list and because a cache of free pages
-	 * should generally be a "last resort" option for a filesystem
-	 * to be able to do its job.
-	 */
-	spin_lock_irqsave(&f2fs_crypto_ctx_lock, flags);
-	ctx = list_first_entry_or_null(&f2fs_free_crypto_ctxs,
-					struct f2fs_crypto_ctx, free_list);
-	if (ctx)
-		list_del(&ctx->free_list);
-	spin_unlock_irqrestore(&f2fs_crypto_ctx_lock, flags);
-	if (!ctx) {
-		ctx = kmem_cache_zalloc(f2fs_crypto_ctx_cachep, GFP_NOFS);
-		if (!ctx)
-			return ERR_PTR(-ENOMEM);
-		ctx->flags |= F2FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	} else {
-		ctx->flags &= ~F2FS_CTX_REQUIRES_FREE_ENCRYPT_FL;
-	}
-	ctx->flags &= ~F2FS_WRITE_PATH_FL;
-	return ctx;
-}
-
-/*
- * Call f2fs_decrypt on every single page, reusing the encryption
- * context.
- */
-static void completion_pages(struct work_struct *work)
-{
-	struct f2fs_crypto_ctx *ctx =
-		container_of(work, struct f2fs_crypto_ctx, r.work);
-	struct bio *bio = ctx->r.bio;
-	struct bio_vec *bv;
-	int i;
-
-	bio_for_each_segment_all(bv, bio, i) {
-		struct page *page = bv->bv_page;
-		int ret = f2fs_decrypt(ctx, page);
-
-		if (ret) {
-			WARN_ON_ONCE(1);
-			SetPageError(page);
-		} else
-			SetPageUptodate(page);
-		unlock_page(page);
-	}
-	f2fs_release_crypto_ctx(ctx);
-	bio_put(bio);
-}
-
-void f2fs_end_io_crypto_work(struct f2fs_crypto_ctx *ctx, struct bio *bio)
-{
-	INIT_WORK(&ctx->r.work, completion_pages);
-	ctx->r.bio = bio;
-	queue_work(f2fs_read_workqueue, &ctx->r.work);
-}
-
-static void f2fs_crypto_destroy(void)
-{
-	struct f2fs_crypto_ctx *pos, *n;
-
-	list_for_each_entry_safe(pos, n, &f2fs_free_crypto_ctxs, free_list)
-		kmem_cache_free(f2fs_crypto_ctx_cachep, pos);
-	INIT_LIST_HEAD(&f2fs_free_crypto_ctxs);
-	if (f2fs_bounce_page_pool)
-		mempool_destroy(f2fs_bounce_page_pool);
-	f2fs_bounce_page_pool = NULL;
-}
-
-/**
- * f2fs_crypto_initialize() - Set up for f2fs encryption.
- *
- * We only call this when we start accessing encrypted files, since it
- * results in memory getting allocated that wouldn't otherwise be used.
- *
- * Return: Zero on success, non-zero otherwise.
- */
-int f2fs_crypto_initialize(void)
-{
-	int i, res = -ENOMEM;
-
-	if (f2fs_bounce_page_pool)
-		return 0;
-
-	mutex_lock(&crypto_init);
-	if (f2fs_bounce_page_pool)
-		goto already_initialized;
-
-	for (i = 0; i < num_prealloc_crypto_ctxs; i++) {
-		struct f2fs_crypto_ctx *ctx;
-
-		ctx = kmem_cache_zalloc(f2fs_crypto_ctx_cachep, GFP_KERNEL);
-		if (!ctx)
-			goto fail;
-		list_add(&ctx->free_list, &f2fs_free_crypto_ctxs);
-	}
-
-	/* must be allocated at the last step to avoid race condition above */
-	f2fs_bounce_page_pool =
-		mempool_create_page_pool(num_prealloc_crypto_pages, 0);
-	if (!f2fs_bounce_page_pool)
-		goto fail;
-
-already_initialized:
-	mutex_unlock(&crypto_init);
-	return 0;
-fail:
-	f2fs_crypto_destroy();
-	mutex_unlock(&crypto_init);
-	return res;
-}
-
-/**
- * f2fs_exit_crypto() - Shutdown the f2fs encryption system
- */
-void f2fs_exit_crypto(void)
-{
-	f2fs_crypto_destroy();
-
-	if (f2fs_read_workqueue)
-		destroy_workqueue(f2fs_read_workqueue);
-	if (f2fs_crypto_ctx_cachep)
-		kmem_cache_destroy(f2fs_crypto_ctx_cachep);
-	if (f2fs_crypt_info_cachep)
-		kmem_cache_destroy(f2fs_crypt_info_cachep);
-}
-
-int __init f2fs_init_crypto(void)
-{
-	int res = -ENOMEM;
-
-	f2fs_read_workqueue = alloc_workqueue("f2fs_crypto", WQ_HIGHPRI, 0);
-	if (!f2fs_read_workqueue)
-		goto fail;
-
-	f2fs_crypto_ctx_cachep = KMEM_CACHE(f2fs_crypto_ctx,
-						SLAB_RECLAIM_ACCOUNT);
-	if (!f2fs_crypto_ctx_cachep)
-		goto fail;
-
-	f2fs_crypt_info_cachep = KMEM_CACHE(f2fs_crypt_info,
-						SLAB_RECLAIM_ACCOUNT);
-	if (!f2fs_crypt_info_cachep)
-		goto fail;
-
-	return 0;
-fail:
-	f2fs_exit_crypto();
-	return res;
-}
-
-void f2fs_restore_and_release_control_page(struct page **page)
-{
-	struct f2fs_crypto_ctx *ctx;
-	struct page *bounce_page;
-
-	/* The bounce data pages are unmapped. */
-	if ((*page)->mapping)
-		return;
-
-	/* The bounce data page is unmapped. */
-	bounce_page = *page;
-	ctx = (struct f2fs_crypto_ctx *)page_private(bounce_page);
-
-	/* restore control page */
-	*page = ctx->w.control_page;
-
-	f2fs_restore_control_page(bounce_page);
-}
-
-void f2fs_restore_control_page(struct page *data_page)
-{
-	struct f2fs_crypto_ctx *ctx =
-		(struct f2fs_crypto_ctx *)page_private(data_page);
-
-	set_page_private(data_page, (unsigned long)NULL);
-	ClearPagePrivate(data_page);
-	unlock_page(data_page);
-	f2fs_release_crypto_ctx(ctx);
-}
-
-/**
- * f2fs_crypt_complete() - The completion callback for page encryption
- * @req: The asynchronous encryption request context
- * @res: The result of the encryption operation
- */
-static void f2fs_crypt_complete(struct crypto_async_request *req, int res)
-{
-	struct f2fs_completion_result *ecr = req->data;
-
-	if (res == -EINPROGRESS)
-		return;
-	ecr->res = res;
-	complete(&ecr->completion);
-}
-
-typedef enum {
-	F2FS_DECRYPT = 0,
-	F2FS_ENCRYPT,
-} f2fs_direction_t;
-
-static int f2fs_page_crypto(struct f2fs_crypto_ctx *ctx,
-				struct inode *inode,
-				f2fs_direction_t rw,
-				pgoff_t index,
-				struct page *src_page,
-				struct page *dest_page)
-{
-	u8 xts_tweak[F2FS_XTS_TWEAK_SIZE];
-	struct ablkcipher_request *req = NULL;
-	DECLARE_F2FS_COMPLETION_RESULT(ecr);
-	struct scatterlist dst, src;
-	struct f2fs_crypt_info *ci = F2FS_I(inode)->i_crypt_info;
-	struct crypto_ablkcipher *tfm = ci->ci_ctfm;
-	int res = 0;
-
-	req = ablkcipher_request_alloc(tfm, GFP_NOFS);
-	if (!req) {
-		printk_ratelimited(KERN_ERR
-				"%s: crypto_request_alloc() failed\n",
-				__func__);
-		return -ENOMEM;
-	}
-	ablkcipher_request_set_callback(
-		req, CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
-		f2fs_crypt_complete, &ecr);
-
-	BUILD_BUG_ON(F2FS_XTS_TWEAK_SIZE < sizeof(index));
-	memcpy(xts_tweak, &index, sizeof(index));
-	memset(&xts_tweak[sizeof(index)], 0,
-			F2FS_XTS_TWEAK_SIZE - sizeof(index));
-
-	sg_init_table(&dst, 1);
-	sg_set_page(&dst, dest_page, PAGE_CACHE_SIZE, 0);
-	sg_init_table(&src, 1);
-	sg_set_page(&src, src_page, PAGE_CACHE_SIZE, 0);
-	ablkcipher_request_set_crypt(req, &src, &dst, PAGE_CACHE_SIZE,
-					xts_tweak);
-	if (rw == F2FS_DECRYPT)
-		res = crypto_ablkcipher_decrypt(req);
-	else
-		res = crypto_ablkcipher_encrypt(req);
-	if (res == -EINPROGRESS || res == -EBUSY) {
-		wait_for_completion(&ecr.completion);
-		res = ecr.res;
-	}
-	ablkcipher_request_free(req);
-	if (res) {
-		printk_ratelimited(KERN_ERR
-			"%s: crypto_ablkcipher_encrypt() returned %d\n",
-			__func__, res);
-		return res;
-	}
-	return 0;
-}
-
-static struct page *alloc_bounce_page(struct f2fs_crypto_ctx *ctx)
-{
-	ctx->w.bounce_page = mempool_alloc(f2fs_bounce_page_pool, GFP_NOWAIT);
-	if (ctx->w.bounce_page == NULL)
-		return ERR_PTR(-ENOMEM);
-	ctx->flags |= F2FS_WRITE_PATH_FL;
-	return ctx->w.bounce_page;
-}
-
-/**
- * f2fs_encrypt() - Encrypts a page
- * @inode:          The inode for which the encryption should take place
- * @plaintext_page: The page to encrypt. Must be locked.
- *
- * Allocates a ciphertext page and encrypts plaintext_page into it using the ctx
- * encryption context.
- *
- * Called on the page write path.  The caller must call
- * f2fs_restore_control_page() on the returned ciphertext page to
- * release the bounce buffer and the encryption context.
- *
- * Return: An allocated page with the encrypted content on success. Else, an
- * error value or NULL.
- */
-struct page *f2fs_encrypt(struct inode *inode,
-			  struct page *plaintext_page)
-{
-	struct f2fs_crypto_ctx *ctx;
-	struct page *ciphertext_page = NULL;
-	int err;
-
-	BUG_ON(!PageLocked(plaintext_page));
-
-	ctx = f2fs_get_crypto_ctx(inode);
-	if (IS_ERR(ctx))
-		return (struct page *)ctx;
-
-	/* The encryption operation will require a bounce page. */
-	ciphertext_page = alloc_bounce_page(ctx);
-	if (IS_ERR(ciphertext_page))
-		goto err_out;
-
-	ctx->w.control_page = plaintext_page;
-	err = f2fs_page_crypto(ctx, inode, F2FS_ENCRYPT, plaintext_page->index,
-					plaintext_page, ciphertext_page);
-	if (err) {
-		ciphertext_page = ERR_PTR(err);
-		goto err_out;
-	}
-
-	SetPagePrivate(ciphertext_page);
-	set_page_private(ciphertext_page, (unsigned long)ctx);
-	lock_page(ciphertext_page);
-	return ciphertext_page;
-
-err_out:
-	f2fs_release_crypto_ctx(ctx);
-	return ciphertext_page;
-}
-
-/**
- * f2fs_decrypt() - Decrypts a page in-place
- * @ctx:  The encryption context.
- * @page: The page to decrypt. Must be locked.
- *
- * Decrypts page in-place using the ctx encryption context.
- *
- * Called from the read completion callback.
- *
- * Return: Zero on success, non-zero otherwise.
- */
-int f2fs_decrypt(struct f2fs_crypto_ctx *ctx, struct page *page)
-{
-	BUG_ON(!PageLocked(page));
-
-	return f2fs_page_crypto(ctx, page->mapping->host,
-				F2FS_DECRYPT, page->index, page, page);
-}
-
-/*
- * Convenience function which takes care of allocating and
- * deallocating the encryption context
- */
-int f2fs_decrypt_one(struct inode *inode, struct page *page)
-{
-	struct f2fs_crypto_ctx *ctx = f2fs_get_crypto_ctx(inode);
-	int ret;
-
-	if (IS_ERR(ctx))
-		return PTR_ERR(ctx);
-	ret = f2fs_decrypt(ctx, page);
-	f2fs_release_crypto_ctx(ctx);
-	return ret;
-}
-
-bool f2fs_valid_contents_enc_mode(uint32_t mode)
-{
-	return (mode == F2FS_ENCRYPTION_MODE_AES_256_XTS);
-}
-
-/**
- * f2fs_validate_encryption_key_size() - Validate the encryption key size
- * @mode: The key mode.
- * @size: The key size to validate.
- *
- * Return: The validated key size for @mode. Zero if invalid.
- */
-uint32_t f2fs_validate_encryption_key_size(uint32_t mode, uint32_t size)
-{
-	if (size == f2fs_encryption_key_size(mode))
-		return size;
-	return 0;
-}
diff --git a/fs/f2fs/crypto_fname.c b/fs/f2fs/crypto_fname.c
deleted file mode 100644
index 0fce444dd5ae..000000000000
--- a/fs/f2fs/crypto_fname.c
+++ /dev/null
@@ -1,438 +0,0 @@
-/*
- * linux/fs/f2fs/crypto_fname.c
- *
- * Copied from linux/fs/ext4/crypto.c
- *
- * Copyright (C) 2015, Google, Inc.
- * Copyright (C) 2015, Motorola Mobility
- *
- * This contains functions for filename crypto management in f2fs
- *
- * Written by Uday Savagaonkar, 2014.
- *
- * Adjust f2fs dentry structure
- *	Jaegeuk Kim, 2015.
- *
- * This has not yet undergone a rigorous security audit.
- */
-#include <crypto/hash.h>
-#include <crypto/sha.h>
-#include <keys/encrypted-type.h>
-#include <keys/user-type.h>
-#include <linux/crypto.h>
-#include <linux/gfp.h>
-#include <linux/kernel.h>
-#include <linux/key.h>
-#include <linux/list.h>
-#include <linux/mempool.h>
-#include <linux/random.h>
-#include <linux/scatterlist.h>
-#include <linux/spinlock_types.h>
-#include <linux/f2fs_fs.h>
-#include <linux/ratelimit.h>
-
-#include "f2fs.h"
-#include "f2fs_crypto.h"
-#include "xattr.h"
-
-/**
- * f2fs_dir_crypt_complete() -
- */
-static void f2fs_dir_crypt_complete(struct crypto_async_request *req, int res)
-{
-	struct f2fs_completion_result *ecr = req->data;
-
-	if (res == -EINPROGRESS)
-		return;
-	ecr->res = res;
-	complete(&ecr->completion);
-}
-
-bool f2fs_valid_filenames_enc_mode(uint32_t mode)
-{
-	return (mode == F2FS_ENCRYPTION_MODE_AES_256_CTS);
-}
-
-static unsigned max_name_len(struct inode *inode)
-{
-	return S_ISLNK(inode->i_mode) ? inode->i_sb->s_blocksize :
-					F2FS_NAME_LEN;
-}
-
-/**
- * f2fs_fname_encrypt() -
- *
- * This function encrypts the input filename, and returns the length of the
- * ciphertext. Errors are returned as negative numbers.  We trust the caller to
- * allocate sufficient memory to oname string.
- */
-static int f2fs_fname_encrypt(struct inode *inode,
-			const struct qstr *iname, struct f2fs_str *oname)
-{
-	u32 ciphertext_len;
-	struct ablkcipher_request *req = NULL;
-	DECLARE_F2FS_COMPLETION_RESULT(ecr);
-	struct f2fs_crypt_info *ci = F2FS_I(inode)->i_crypt_info;
-	struct crypto_ablkcipher *tfm = ci->ci_ctfm;
-	int res = 0;
-	char iv[F2FS_CRYPTO_BLOCK_SIZE];
-	struct scatterlist src_sg, dst_sg;
-	int padding = 4 << (ci->ci_flags & F2FS_POLICY_FLAGS_PAD_MASK);
-	char *workbuf, buf[32], *alloc_buf = NULL;
-	unsigned lim = max_name_len(inode);
-
-	if (iname->len <= 0 || iname->len > lim)
-		return -EIO;
-
-	ciphertext_len = (iname->len < F2FS_CRYPTO_BLOCK_SIZE) ?
-		F2FS_CRYPTO_BLOCK_SIZE : iname->len;
-	ciphertext_len = f2fs_fname_crypto_round_up(ciphertext_len, padding);
-	ciphertext_len = (ciphertext_len > lim) ? lim : ciphertext_len;
-
-	if (ciphertext_len <= sizeof(buf)) {
-		workbuf = buf;
-	} else {
-		alloc_buf = kmalloc(ciphertext_len, GFP_NOFS);
-		if (!alloc_buf)
-			return -ENOMEM;
-		workbuf = alloc_buf;
-	}
-
-	/* Allocate request */
-	req = ablkcipher_request_alloc(tfm, GFP_NOFS);
-	if (!req) {
-		printk_ratelimited(KERN_ERR
-			"%s: crypto_request_alloc() failed\n", __func__);
-		kfree(alloc_buf);
-		return -ENOMEM;
-	}
-	ablkcipher_request_set_callback(req,
-			CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
-			f2fs_dir_crypt_complete, &ecr);
-
-	/* Copy the input */
-	memcpy(workbuf, iname->name, iname->len);
-	if (iname->len < ciphertext_len)
-		memset(workbuf + iname->len, 0, ciphertext_len - iname->len);
-
-	/* Initialize IV */
-	memset(iv, 0, F2FS_CRYPTO_BLOCK_SIZE);
-
-	/* Create encryption request */
-	sg_init_one(&src_sg, workbuf, ciphertext_len);
-	sg_init_one(&dst_sg, oname->name, ciphertext_len);
-	ablkcipher_request_set_crypt(req, &src_sg, &dst_sg, ciphertext_len, iv);
-	res = crypto_ablkcipher_encrypt(req);
-	if (res == -EINPROGRESS || res == -EBUSY) {
-		wait_for_completion(&ecr.completion);
-		res = ecr.res;
-	}
-	kfree(alloc_buf);
-	ablkcipher_request_free(req);
-	if (res < 0) {
-		printk_ratelimited(KERN_ERR
-				"%s: Error (error code %d)\n", __func__, res);
-	}
-	oname->len = ciphertext_len;
-	return res;
-}
-
-/*
- * f2fs_fname_decrypt()
- *	This function decrypts the input filename, and returns
- *	the length of the plaintext.
- *	Errors are returned as negative numbers.
- *	We trust the caller to allocate sufficient memory to oname string.
- */
-static int f2fs_fname_decrypt(struct inode *inode,
-			const struct f2fs_str *iname, struct f2fs_str *oname)
-{
-	struct ablkcipher_request *req = NULL;
-	DECLARE_F2FS_COMPLETION_RESULT(ecr);
-	struct scatterlist src_sg, dst_sg;
-	struct f2fs_crypt_info *ci = F2FS_I(inode)->i_crypt_info;
-	struct crypto_ablkcipher *tfm = ci->ci_ctfm;
-	int res = 0;
-	char iv[F2FS_CRYPTO_BLOCK_SIZE];
-	unsigned lim = max_name_len(inode);
-
-	if (iname->len <= 0 || iname->len > lim)
-		return -EIO;
-
-	/* Allocate request */
-	req = ablkcipher_request_alloc(tfm, GFP_NOFS);
-	if (!req) {
-		printk_ratelimited(KERN_ERR
-			"%s: crypto_request_alloc() failed\n",  __func__);
-		return -ENOMEM;
-	}
-	ablkcipher_request_set_callback(req,
-		CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
-		f2fs_dir_crypt_complete, &ecr);
-
-	/* Initialize IV */
-	memset(iv, 0, F2FS_CRYPTO_BLOCK_SIZE);
-
-	/* Create decryption request */
-	sg_init_one(&src_sg, iname->name, iname->len);
-	sg_init_one(&dst_sg, oname->name, oname->len);
-	ablkcipher_request_set_crypt(req, &src_sg, &dst_sg, iname->len, iv);
-	res = crypto_ablkcipher_decrypt(req);
-	if (res == -EINPROGRESS || res == -EBUSY) {
-		wait_for_completion(&ecr.completion);
-		res = ecr.res;
-	}
-	ablkcipher_request_free(req);
-	if (res < 0) {
-		printk_ratelimited(KERN_ERR
-			"%s: Error in f2fs_fname_decrypt (error code %d)\n",
-			__func__, res);
-		return res;
-	}
-
-	oname->len = strnlen(oname->name, iname->len);
-	return oname->len;
-}
-
-static const char *lookup_table =
-	"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+,";
-
-/**
- * f2fs_fname_encode_digest() -
- *
- * Encodes the input digest using characters from the set [a-zA-Z0-9_+].
- * The encoded string is roughly 4/3 times the size of the input string.
- */
-static int digest_encode(const char *src, int len, char *dst)
-{
-	int i = 0, bits = 0, ac = 0;
-	char *cp = dst;
-
-	while (i < len) {
-		ac += (((unsigned char) src[i]) << bits);
-		bits += 8;
-		do {
-			*cp++ = lookup_table[ac & 0x3f];
-			ac >>= 6;
-			bits -= 6;
-		} while (bits >= 6);
-		i++;
-	}
-	if (bits)
-		*cp++ = lookup_table[ac & 0x3f];
-	return cp - dst;
-}
-
-static int digest_decode(const char *src, int len, char *dst)
-{
-	int i = 0, bits = 0, ac = 0;
-	const char *p;
-	char *cp = dst;
-
-	while (i < len) {
-		p = strchr(lookup_table, src[i]);
-		if (p == NULL || src[i] == 0)
-			return -2;
-		ac += (p - lookup_table) << bits;
-		bits += 6;
-		if (bits >= 8) {
-			*cp++ = ac & 0xff;
-			ac >>= 8;
-			bits -= 8;
-		}
-		i++;
-	}
-	if (ac)
-		return -1;
-	return cp - dst;
-}
-
-/**
- * f2fs_fname_crypto_round_up() -
- *
- * Return: The next multiple of block size
- */
-u32 f2fs_fname_crypto_round_up(u32 size, u32 blksize)
-{
-	return ((size + blksize - 1) / blksize) * blksize;
-}
-
-/**
- * f2fs_fname_crypto_alloc_obuff() -
- *
- * Allocates an output buffer that is sufficient for the crypto operation
- * specified by the context and the direction.
- */
-int f2fs_fname_crypto_alloc_buffer(struct inode *inode,
-				   u32 ilen, struct f2fs_str *crypto_str)
-{
-	unsigned int olen;
-	int padding = 16;
-	struct f2fs_crypt_info *ci = F2FS_I(inode)->i_crypt_info;
-
-	if (ci)
-		padding = 4 << (ci->ci_flags & F2FS_POLICY_FLAGS_PAD_MASK);
-	if (padding < F2FS_CRYPTO_BLOCK_SIZE)
-		padding = F2FS_CRYPTO_BLOCK_SIZE;
-	olen = f2fs_fname_crypto_round_up(ilen, padding);
-	crypto_str->len = olen;
-	if (olen < F2FS_FNAME_CRYPTO_DIGEST_SIZE * 2)
-		olen = F2FS_FNAME_CRYPTO_DIGEST_SIZE * 2;
-	/* Allocated buffer can hold one more character to null-terminate the
-	 * string */
-	crypto_str->name = kmalloc(olen + 1, GFP_NOFS);
-	if (!(crypto_str->name))
-		return -ENOMEM;
-	return 0;
-}
-
-/**
- * f2fs_fname_crypto_free_buffer() -
- *
- * Frees the buffer allocated for crypto operation.
- */
-void f2fs_fname_crypto_free_buffer(struct f2fs_str *crypto_str)
-{
-	if (!crypto_str)
-		return;
-	kfree(crypto_str->name);
-	crypto_str->name = NULL;
-}
-
-/**
- * f2fs_fname_disk_to_usr() - converts a filename from disk space to user space
- */
-int f2fs_fname_disk_to_usr(struct inode *inode,
-			f2fs_hash_t *hash,
-			const struct f2fs_str *iname,
-			struct f2fs_str *oname)
-{
-	const struct qstr qname = FSTR_TO_QSTR(iname);
-	char buf[24];
-	int ret;
-
-	if (is_dot_dotdot(&qname)) {
-		oname->name[0] = '.';
-		oname->name[iname->len - 1] = '.';
-		oname->len = iname->len;
-		return oname->len;
-	}
-
-	if (F2FS_I(inode)->i_crypt_info)
-		return f2fs_fname_decrypt(inode, iname, oname);
-
-	if (iname->len <= F2FS_FNAME_CRYPTO_DIGEST_SIZE) {
-		ret = digest_encode(iname->name, iname->len, oname->name);
-		oname->len = ret;
-		return ret;
-	}
-	if (hash) {
-		memcpy(buf, hash, 4);
-		memset(buf + 4, 0, 4);
-	} else
-		memset(buf, 0, 8);
-	memcpy(buf + 8, iname->name + ((iname->len - 17) & ~15), 16);
-	oname->name[0] = '_';
-	ret = digest_encode(buf, 24, oname->name + 1);
-	oname->len = ret + 1;
-	return ret + 1;
-}
-
-/**
- * f2fs_fname_usr_to_disk() - converts a filename from user space to disk space
- */
-int f2fs_fname_usr_to_disk(struct inode *inode,
-			const struct qstr *iname,
-			struct f2fs_str *oname)
-{
-	int res;
-	struct f2fs_crypt_info *ci = F2FS_I(inode)->i_crypt_info;
-
-	if (is_dot_dotdot(iname)) {
-		oname->name[0] = '.';
-		oname->name[iname->len - 1] = '.';
-		oname->len = iname->len;
-		return oname->len;
-	}
-
-	if (ci) {
-		res = f2fs_fname_encrypt(inode, iname, oname);
-		return res;
-	}
-	/* Without a proper key, a user is not allowed to modify the filenames
-	 * in a directory. Consequently, a user space name cannot be mapped to
-	 * a disk-space name */
-	return -EACCES;
-}
-
-int f2fs_fname_setup_filename(struct inode *dir, const struct qstr *iname,
-			      int lookup, struct f2fs_filename *fname)
-{
-	struct f2fs_crypt_info *ci;
-	int ret = 0, bigname = 0;
-
-	memset(fname, 0, sizeof(struct f2fs_filename));
-	fname->usr_fname = iname;
-
-	if (!f2fs_encrypted_inode(dir) || is_dot_dotdot(iname)) {
-		fname->disk_name.name = (unsigned char *)iname->name;
-		fname->disk_name.len = iname->len;
-		return 0;
-	}
-	ret = f2fs_get_encryption_info(dir);
-	if (ret)
-		return ret;
-	ci = F2FS_I(dir)->i_crypt_info;
-	if (ci) {
-		ret = f2fs_fname_crypto_alloc_buffer(dir, iname->len,
-						     &fname->crypto_buf);
-		if (ret < 0)
-			return ret;
-		ret = f2fs_fname_encrypt(dir, iname, &fname->crypto_buf);
-		if (ret < 0)
-			goto errout;
-		fname->disk_name.name = fname->crypto_buf.name;
-		fname->disk_name.len = fname->crypto_buf.len;
-		return 0;
-	}
-	if (!lookup)
-		return -EACCES;
-
-	/* We don't have the key and we are doing a lookup; decode the
-	 * user-supplied name
-	 */
-	if (iname->name[0] == '_')
-		bigname = 1;
-	if ((bigname && (iname->len != 33)) ||
-	    (!bigname && (iname->len > 43)))
-		return -ENOENT;
-
-	fname->crypto_buf.name = kmalloc(32, GFP_KERNEL);
-	if (fname->crypto_buf.name == NULL)
-		return -ENOMEM;
-	ret = digest_decode(iname->name + bigname, iname->len - bigname,
-				fname->crypto_buf.name);
-	if (ret < 0) {
-		ret = -ENOENT;
-		goto errout;
-	}
-	fname->crypto_buf.len = ret;
-	if (bigname) {
-		memcpy(&fname->hash, fname->crypto_buf.name, 4);
-	} else {
-		fname->disk_name.name = fname->crypto_buf.name;
-		fname->disk_name.len = fname->crypto_buf.len;
-	}
-	return 0;
-errout:
-	f2fs_fname_crypto_free_buffer(&fname->crypto_buf);
-	return ret;
-}
-
-void f2fs_fname_free_filename(struct f2fs_filename *fname)
-{
-	kfree(fname->crypto_buf.name);
-	fname->crypto_buf.name = NULL;
-	fname->usr_fname = NULL;
-	fname->disk_name.name = NULL;
-}
diff --git a/fs/f2fs/crypto_key.c b/fs/f2fs/crypto_key.c
deleted file mode 100644
index 7e62889a1d3d..000000000000
--- a/fs/f2fs/crypto_key.c
+++ /dev/null
@@ -1,259 +0,0 @@
-/*
- * linux/fs/f2fs/crypto_key.c
- *
- * Copied from linux/fs/f2fs/crypto_key.c
- *
- * Copyright (C) 2015, Google, Inc.
- *
- * This contains encryption key functions for f2fs
- *
- * Written by Michael Halcrow, Ildar Muslukhov, and Uday Savagaonkar, 2015.
- */
-#include <keys/encrypted-type.h>
-#include <keys/user-type.h>
-#include <linux/random.h>
-#include <linux/scatterlist.h>
-#include <uapi/linux/keyctl.h>
-#include <crypto/hash.h>
-#include <linux/f2fs_fs.h>
-
-#include "f2fs.h"
-#include "xattr.h"
-
-static void derive_crypt_complete(struct crypto_async_request *req, int rc)
-{
-	struct f2fs_completion_result *ecr = req->data;
-
-	if (rc == -EINPROGRESS)
-		return;
-
-	ecr->res = rc;
-	complete(&ecr->completion);
-}
-
-/**
- * f2fs_derive_key_aes() - Derive a key using AES-128-ECB
- * @deriving_key: Encryption key used for derivatio.
- * @source_key:   Source key to which to apply derivation.
- * @derived_key:  Derived key.
- *
- * Return: Zero on success; non-zero otherwise.
- */
-static int f2fs_derive_key_aes(char deriving_key[F2FS_AES_128_ECB_KEY_SIZE],
-				char source_key[F2FS_AES_256_XTS_KEY_SIZE],
-				char derived_key[F2FS_AES_256_XTS_KEY_SIZE])
-{
-	int res = 0;
-	struct ablkcipher_request *req = NULL;
-	DECLARE_F2FS_COMPLETION_RESULT(ecr);
-	struct scatterlist src_sg, dst_sg;
-	struct crypto_ablkcipher *tfm = crypto_alloc_ablkcipher("ecb(aes)", 0,
-								0);
-
-	if (IS_ERR(tfm)) {
-		res = PTR_ERR(tfm);
-		tfm = NULL;
-		goto out;
-	}
-	crypto_ablkcipher_set_flags(tfm, CRYPTO_TFM_REQ_WEAK_KEY);
-	req = ablkcipher_request_alloc(tfm, GFP_NOFS);
-	if (!req) {
-		res = -ENOMEM;
-		goto out;
-	}
-	ablkcipher_request_set_callback(req,
-			CRYPTO_TFM_REQ_MAY_BACKLOG | CRYPTO_TFM_REQ_MAY_SLEEP,
-			derive_crypt_complete, &ecr);
-	res = crypto_ablkcipher_setkey(tfm, deriving_key,
-				F2FS_AES_128_ECB_KEY_SIZE);
-	if (res < 0)
-		goto out;
-
-	sg_init_one(&src_sg, source_key, F2FS_AES_256_XTS_KEY_SIZE);
-	sg_init_one(&dst_sg, derived_key, F2FS_AES_256_XTS_KEY_SIZE);
-	ablkcipher_request_set_crypt(req, &src_sg, &dst_sg,
-					F2FS_AES_256_XTS_KEY_SIZE, NULL);
-	res = crypto_ablkcipher_encrypt(req);
-	if (res == -EINPROGRESS || res == -EBUSY) {
-		wait_for_completion(&ecr.completion);
-		res = ecr.res;
-	}
-out:
-	if (req)
-		ablkcipher_request_free(req);
-	if (tfm)
-		crypto_free_ablkcipher(tfm);
-	return res;
-}
-
-static void f2fs_free_crypt_info(struct f2fs_crypt_info *ci)
-{
-	if (!ci)
-		return;
-
-	crypto_free_ablkcipher(ci->ci_ctfm);
-	kmem_cache_free(f2fs_crypt_info_cachep, ci);
-}
-
-void f2fs_free_encryption_info(struct inode *inode, struct f2fs_crypt_info *ci)
-{
-	struct f2fs_inode_info *fi = F2FS_I(inode);
-	struct f2fs_crypt_info *prev;
-
-	if (ci == NULL)
-		ci = ACCESS_ONCE(fi->i_crypt_info);
-	if (ci == NULL)
-		return;
-	prev = cmpxchg(&fi->i_crypt_info, ci, NULL);
-	if (prev != ci)
-		return;
-
-	f2fs_free_crypt_info(ci);
-}
-
-int f2fs_get_encryption_info(struct inode *inode)
-{
-	struct f2fs_inode_info *fi = F2FS_I(inode);
-	struct f2fs_crypt_info *crypt_info;
-	char full_key_descriptor[F2FS_KEY_DESC_PREFIX_SIZE +
-				(F2FS_KEY_DESCRIPTOR_SIZE * 2) + 1];
-	struct key *keyring_key = NULL;
-	struct f2fs_encryption_key *master_key;
-	struct f2fs_encryption_context ctx;
-	const struct user_key_payload *ukp;
-	struct crypto_ablkcipher *ctfm;
-	const char *cipher_str;
-	char raw_key[F2FS_MAX_KEY_SIZE];
-	char mode;
-	int res;
-
-	if (fi->i_crypt_info)
-		return 0;
-
-	res = f2fs_crypto_initialize();
-	if (res)
-		return res;
-
-	res = f2fs_getxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
-				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
-				&ctx, sizeof(ctx), NULL);
-	if (res < 0)
-		return res;
-	else if (res != sizeof(ctx))
-		return -EINVAL;
-	res = 0;
-
-	crypt_info = kmem_cache_alloc(f2fs_crypt_info_cachep, GFP_NOFS);
-	if (!crypt_info)
-		return -ENOMEM;
-
-	crypt_info->ci_flags = ctx.flags;
-	crypt_info->ci_data_mode = ctx.contents_encryption_mode;
-	crypt_info->ci_filename_mode = ctx.filenames_encryption_mode;
-	crypt_info->ci_ctfm = NULL;
-	memcpy(crypt_info->ci_master_key, ctx.master_key_descriptor,
-				sizeof(crypt_info->ci_master_key));
-	if (S_ISREG(inode->i_mode))
-		mode = crypt_info->ci_data_mode;
-	else if (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))
-		mode = crypt_info->ci_filename_mode;
-	else
-		BUG();
-
-	switch (mode) {
-	case F2FS_ENCRYPTION_MODE_AES_256_XTS:
-		cipher_str = "xts(aes)";
-		break;
-	case F2FS_ENCRYPTION_MODE_AES_256_CTS:
-		cipher_str = "cts(cbc(aes))";
-		break;
-	default:
-		printk_once(KERN_WARNING
-			    "f2fs: unsupported key mode %d (ino %u)\n",
-			    mode, (unsigned) inode->i_ino);
-		res = -ENOKEY;
-		goto out;
-	}
-
-	memcpy(full_key_descriptor, F2FS_KEY_DESC_PREFIX,
-					F2FS_KEY_DESC_PREFIX_SIZE);
-	sprintf(full_key_descriptor + F2FS_KEY_DESC_PREFIX_SIZE,
-					"%*phN", F2FS_KEY_DESCRIPTOR_SIZE,
-					ctx.master_key_descriptor);
-	full_key_descriptor[F2FS_KEY_DESC_PREFIX_SIZE +
-					(2 * F2FS_KEY_DESCRIPTOR_SIZE)] = '\0';
-	keyring_key = request_key(&key_type_logon, full_key_descriptor, NULL);
-	if (IS_ERR(keyring_key)) {
-		res = PTR_ERR(keyring_key);
-		keyring_key = NULL;
-		goto out;
-	}
-	if (keyring_key->type != &key_type_logon) {
-		printk_once(KERN_WARNING "f2fs: key type must be logon\n");
-		res = -ENOKEY;
-		goto out;
-	}
-	down_read(&keyring_key->sem);
-	ukp = user_key_payload(keyring_key);
-	if (!ukp) {
-		/* key was revoked before we acquired its semaphore */
-		res = -EKEYREVOKED;
-		up_read(&keyring_key->sem);
-		goto out;
-	}
-	if (ukp->datalen != sizeof(struct f2fs_encryption_key)) {
-		res = -EINVAL;
-		up_read(&keyring_key->sem);
-		goto out;
-	}
-	master_key = (struct f2fs_encryption_key *)ukp->data;
-	BUILD_BUG_ON(F2FS_AES_128_ECB_KEY_SIZE !=
-				F2FS_KEY_DERIVATION_NONCE_SIZE);
-	if (master_key->size != F2FS_AES_256_XTS_KEY_SIZE) {
-		printk_once(KERN_WARNING
-				"f2fs: key size incorrect: %d\n",
-				master_key->size);
-		res = -ENOKEY;
-		up_read(&keyring_key->sem);
-		goto out;
-	}
-	res = f2fs_derive_key_aes(ctx.nonce, master_key->raw,
-				  raw_key);
-	up_read(&keyring_key->sem);
-	if (res)
-		goto out;
-
-	ctfm = crypto_alloc_ablkcipher(cipher_str, 0, 0);
-	if (!ctfm || IS_ERR(ctfm)) {
-		res = ctfm ? PTR_ERR(ctfm) : -ENOMEM;
-		printk(KERN_DEBUG
-		       "%s: error %d (inode %u) allocating crypto tfm\n",
-		       __func__, res, (unsigned) inode->i_ino);
-		goto out;
-	}
-	crypt_info->ci_ctfm = ctfm;
-	crypto_ablkcipher_clear_flags(ctfm, ~0);
-	crypto_tfm_set_flags(crypto_ablkcipher_tfm(ctfm),
-			     CRYPTO_TFM_REQ_WEAK_KEY);
-	res = crypto_ablkcipher_setkey(ctfm, raw_key,
-					f2fs_encryption_key_size(mode));
-	if (res)
-		goto out;
-
-	if (cmpxchg(&fi->i_crypt_info, NULL, crypt_info) == NULL)
-		crypt_info = NULL;
-out:
-	if (res == -ENOKEY && !S_ISREG(inode->i_mode))
-		res = 0;
-	key_put(keyring_key);
-	f2fs_free_crypt_info(crypt_info);
-	memzero_explicit(raw_key, sizeof(raw_key));
-	return res;
-}
-
-int f2fs_has_encryption_key(struct inode *inode)
-{
-	struct f2fs_inode_info *fi = F2FS_I(inode);
-
-	return (fi->i_crypt_info != NULL);
-}
diff --git a/fs/f2fs/crypto_policy.c b/fs/f2fs/crypto_policy.c
deleted file mode 100644
index 884f3f0fe29d..000000000000
--- a/fs/f2fs/crypto_policy.c
+++ /dev/null
@@ -1,248 +0,0 @@
-/*
- * copied from linux/fs/ext4/crypto_policy.c
- *
- * Copyright (C) 2015, Google, Inc.
- * Copyright (C) 2015, Motorola Mobility.
- *
- * This contains encryption policy functions for f2fs with some modifications
- * to support f2fs-specific xattr APIs.
- *
- * Written by Michael Halcrow, 2015.
- * Modified by Jaegeuk Kim, 2015.
- */
-#include <linux/random.h>
-#include <linux/string.h>
-#include <linux/types.h>
-#include <linux/f2fs_fs.h>
-
-#include "f2fs.h"
-#include "xattr.h"
-
-static int f2fs_inode_has_encryption_context(struct inode *inode)
-{
-	int res = f2fs_getxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
-			F2FS_XATTR_NAME_ENCRYPTION_CONTEXT, NULL, 0, NULL);
-	return (res > 0);
-}
-
-/*
- * check whether the policy is consistent with the encryption context
- * for the inode
- */
-static int f2fs_is_encryption_context_consistent_with_policy(
-	struct inode *inode, const struct f2fs_encryption_policy *policy)
-{
-	struct f2fs_encryption_context ctx;
-	int res = f2fs_getxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
-				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT, &ctx,
-				sizeof(ctx), NULL);
-
-	if (res != sizeof(ctx))
-		return 0;
-
-	return (memcmp(ctx.master_key_descriptor, policy->master_key_descriptor,
-				F2FS_KEY_DESCRIPTOR_SIZE) == 0 &&
-			(ctx.flags == policy->flags) &&
-			(ctx.contents_encryption_mode ==
-			 policy->contents_encryption_mode) &&
-			(ctx.filenames_encryption_mode ==
-			 policy->filenames_encryption_mode));
-}
-
-static int f2fs_create_encryption_context_from_policy(
-	struct inode *inode, const struct f2fs_encryption_policy *policy)
-{
-	struct f2fs_encryption_context ctx;
-
-	ctx.format = F2FS_ENCRYPTION_CONTEXT_FORMAT_V1;
-	memcpy(ctx.master_key_descriptor, policy->master_key_descriptor,
-			F2FS_KEY_DESCRIPTOR_SIZE);
-
-	if (!f2fs_valid_contents_enc_mode(policy->contents_encryption_mode)) {
-		printk(KERN_WARNING
-		       "%s: Invalid contents encryption mode %d\n", __func__,
-			policy->contents_encryption_mode);
-		return -EINVAL;
-	}
-
-	if (!f2fs_valid_filenames_enc_mode(policy->filenames_encryption_mode)) {
-		printk(KERN_WARNING
-		       "%s: Invalid filenames encryption mode %d\n", __func__,
-			policy->filenames_encryption_mode);
-		return -EINVAL;
-	}
-
-	if (policy->flags & ~F2FS_POLICY_FLAGS_VALID)
-		return -EINVAL;
-
-	ctx.contents_encryption_mode = policy->contents_encryption_mode;
-	ctx.filenames_encryption_mode = policy->filenames_encryption_mode;
-	ctx.flags = policy->flags;
-	BUILD_BUG_ON(sizeof(ctx.nonce) != F2FS_KEY_DERIVATION_NONCE_SIZE);
-	get_random_bytes(ctx.nonce, F2FS_KEY_DERIVATION_NONCE_SIZE);
-
-	return f2fs_setxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
-			F2FS_XATTR_NAME_ENCRYPTION_CONTEXT, &ctx,
-			sizeof(ctx), NULL, XATTR_CREATE);
-}
-
-int f2fs_process_policy(const struct f2fs_encryption_policy *policy,
-			struct inode *inode)
-{
-	if (!inode_owner_or_capable(inode))
-		return -EACCES;
-
-	if (policy->version != 0)
-		return -EINVAL;
-
-	if (!S_ISDIR(inode->i_mode))
-		return -EINVAL;
-
-	if (!f2fs_inode_has_encryption_context(inode)) {
-		if (!f2fs_empty_dir(inode))
-			return -ENOTEMPTY;
-		return f2fs_create_encryption_context_from_policy(inode,
-								  policy);
-	}
-
-	if (f2fs_is_encryption_context_consistent_with_policy(inode, policy))
-		return 0;
-
-	printk(KERN_WARNING "%s: Policy inconsistent with encryption context\n",
-	       __func__);
-	return -EINVAL;
-}
-
-int f2fs_get_policy(struct inode *inode, struct f2fs_encryption_policy *policy)
-{
-	struct f2fs_encryption_context ctx;
-	int res;
-
-	if (!f2fs_encrypted_inode(inode))
-		return -ENODATA;
-
-	res = f2fs_getxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
-				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
-				&ctx, sizeof(ctx), NULL);
-	if (res != sizeof(ctx))
-		return -ENODATA;
-	if (ctx.format != F2FS_ENCRYPTION_CONTEXT_FORMAT_V1)
-		return -EINVAL;
-
-	policy->version = 0;
-	policy->contents_encryption_mode = ctx.contents_encryption_mode;
-	policy->filenames_encryption_mode = ctx.filenames_encryption_mode;
-	policy->flags = ctx.flags;
-	memcpy(&policy->master_key_descriptor, ctx.master_key_descriptor,
-			F2FS_KEY_DESCRIPTOR_SIZE);
-	return 0;
-}
-
-int f2fs_is_child_context_consistent_with_parent(struct inode *parent,
-						struct inode *child)
-{
-	const struct f2fs_crypt_info *parent_ci, *child_ci;
-	struct f2fs_encryption_context parent_ctx, child_ctx;
-	int res;
-
-	/* No restrictions on file types which are never encrypted */
-	if (!S_ISREG(child->i_mode) && !S_ISDIR(child->i_mode) &&
-	    !S_ISLNK(child->i_mode))
-		return 1;
-
-	/* No restrictions if the parent directory is unencrypted */
-	if (!f2fs_encrypted_inode(parent))
-		return 1;
-
-	/* Encrypted directories must not contain unencrypted files */
-	if (!f2fs_encrypted_inode(child))
-		return 0;
-
-	/*
-	 * Both parent and child are encrypted, so verify they use the same
-	 * encryption policy.  Compare the fscrypt_info structs if the keys are
-	 * available, otherwise retrieve and compare the fscrypt_contexts.
-	 *
-	 * Note that the fscrypt_context retrieval will be required frequently
-	 * when accessing an encrypted directory tree without the key.
-	 * Performance-wise this is not a big deal because we already don't
-	 * really optimize for file access without the key (to the extent that
-	 * such access is even possible), given that any attempted access
-	 * already causes a fscrypt_context retrieval and keyring search.
-	 *
-	 * In any case, if an unexpected error occurs, fall back to "forbidden".
-	 */
-
-	res = f2fs_get_encryption_info(parent);
-	if (res)
-		return 0;
-	res = f2fs_get_encryption_info(child);
-	if (res)
-		return 0;
-	parent_ci = F2FS_I(parent)->i_crypt_info;
-	child_ci = F2FS_I(child)->i_crypt_info;
-	if (parent_ci && child_ci) {
-		return memcmp(parent_ci->ci_master_key, child_ci->ci_master_key,
-			      F2FS_KEY_DESCRIPTOR_SIZE) == 0 &&
-			(parent_ci->ci_data_mode == child_ci->ci_data_mode) &&
-			(parent_ci->ci_filename_mode ==
-			 child_ci->ci_filename_mode) &&
-			(parent_ci->ci_flags == child_ci->ci_flags);
-	}
-
-	res = f2fs_getxattr(parent, F2FS_XATTR_INDEX_ENCRYPTION,
-			    F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
-			    &parent_ctx, sizeof(parent_ctx), NULL);
-	if (res != sizeof(parent_ctx))
-		return 0;
-
-	res = f2fs_getxattr(child, F2FS_XATTR_INDEX_ENCRYPTION,
-			    F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
-			    &child_ctx, sizeof(child_ctx), NULL);
-	if (res != sizeof(child_ctx))
-		return 0;
-
-	return memcmp(parent_ctx.master_key_descriptor,
-		      child_ctx.master_key_descriptor,
-		      F2FS_KEY_DESCRIPTOR_SIZE) == 0 &&
-		(parent_ctx.contents_encryption_mode ==
-		 child_ctx.contents_encryption_mode) &&
-		(parent_ctx.filenames_encryption_mode ==
-		 child_ctx.filenames_encryption_mode) &&
-		(parent_ctx.flags == child_ctx.flags);
-}
-
-/**
- * f2fs_inherit_context() - Sets a child context from its parent
- * @parent: Parent inode from which the context is inherited.
- * @child:  Child inode that inherits the context from @parent.
- *
- * Return: Zero on success, non-zero otherwise
- */
-int f2fs_inherit_context(struct inode *parent, struct inode *child,
-						struct page *ipage)
-{
-	struct f2fs_encryption_context ctx;
-	struct f2fs_crypt_info *ci;
-	int res;
-
-	res = f2fs_get_encryption_info(parent);
-	if (res < 0)
-		return res;
-
-	ci = F2FS_I(parent)->i_crypt_info;
-	BUG_ON(ci == NULL);
-
-	ctx.format = F2FS_ENCRYPTION_CONTEXT_FORMAT_V1;
-
-	ctx.contents_encryption_mode = ci->ci_data_mode;
-	ctx.filenames_encryption_mode = ci->ci_filename_mode;
-	ctx.flags = ci->ci_flags;
-	memcpy(ctx.master_key_descriptor, ci->ci_master_key,
-			F2FS_KEY_DESCRIPTOR_SIZE);
-
-	get_random_bytes(ctx.nonce, F2FS_KEY_DERIVATION_NONCE_SIZE);
-	return f2fs_setxattr(child, F2FS_XATTR_INDEX_ENCRYPTION,
-				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT, &ctx,
-				sizeof(ctx), ipage, XATTR_CREATE);
-}
diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c
index 98b2fc2678ff..c8583d7a1845 100644
--- a/fs/f2fs/data.c
+++ b/fs/f2fs/data.c
@@ -19,6 +19,8 @@
 #include <linux/bio.h>
 #include <linux/prefetch.h>
 #include <linux/uio.h>
+#include <linux/mm.h>
+#include <linux/memcontrol.h>
 #include <linux/cleancache.h>
 
 #include "f2fs.h"
@@ -26,17 +28,45 @@
 #include "segment.h"
 #include "trace.h"
 #include <trace/events/f2fs.h>
+#include <trace/events/android_fs.h>
+
+static bool __is_cp_guaranteed(struct page *page)
+{
+	struct address_space *mapping = page->mapping;
+	struct inode *inode;
+	struct f2fs_sb_info *sbi;
+
+	if (!mapping)
+		return false;
+
+	inode = mapping->host;
+	sbi = F2FS_I_SB(inode);
+
+	if (inode->i_ino == F2FS_META_INO(sbi) ||
+			inode->i_ino ==  F2FS_NODE_INO(sbi) ||
+			S_ISDIR(inode->i_mode) ||
+			is_cold_data(page))
+		return true;
+	return false;
+}
 
 static void f2fs_read_end_io(struct bio *bio)
 {
 	struct bio_vec *bvec;
 	int i;
 
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(F2FS_P_SB(bio->bi_io_vec->bv_page), FAULT_IO)) {
+		f2fs_show_injection_info(FAULT_IO);
+		bio->bi_error = -EIO;
+	}
+#endif
+
 	if (f2fs_bio_encrypted(bio)) {
 		if (bio->bi_error) {
-			f2fs_release_crypto_ctx(bio->bi_private);
+			fscrypt_release_ctx(bio->bi_private);
 		} else {
-			f2fs_end_io_crypto_work(bio->bi_private, bio);
+			fscrypt_decrypt_bio_pages(bio->bi_private, bio);
 			return;
 		}
 	}
@@ -45,7 +75,8 @@ static void f2fs_read_end_io(struct bio *bio)
 		struct page *page = bvec->bv_page;
 
 		if (!bio->bi_error) {
-			SetPageUptodate(page);
+			if (!PageUptodate(page))
+				SetPageUptodate(page);
 		} else {
 			ClearPageUptodate(page);
 			SetPageError(page);
@@ -63,26 +94,77 @@ static void f2fs_write_end_io(struct bio *bio)
 
 	bio_for_each_segment_all(bvec, bio, i) {
 		struct page *page = bvec->bv_page;
+		enum count_type type = WB_DATA_TYPE(page);
+
+		if (IS_DUMMY_WRITTEN_PAGE(page)) {
+			set_page_private(page, (unsigned long)NULL);
+			ClearPagePrivate(page);
+			unlock_page(page);
+			mempool_free(page, sbi->write_io_dummy);
+
+			if (unlikely(bio->bi_error))
+				f2fs_stop_checkpoint(sbi, true);
+			continue;
+		}
 
-		f2fs_restore_and_release_control_page(&page);
+		fscrypt_pullback_bio_page(&page, true);
 
 		if (unlikely(bio->bi_error)) {
-			set_page_dirty(page);
 			set_bit(AS_EIO, &page->mapping->flags);
-			f2fs_stop_checkpoint(sbi);
+			f2fs_stop_checkpoint(sbi, true);
 		}
+		dec_page_count(sbi, type);
+		clear_cold_data(page);
 		end_page_writeback(page);
-		dec_page_count(sbi, F2FS_WRITEBACK);
 	}
-
-	if (!get_pages(sbi, F2FS_WRITEBACK) &&
-			!list_empty(&sbi->cp_wait.task_list))
+	if (!get_pages(sbi, F2FS_WB_CP_DATA) &&
+				wq_has_sleeper(&sbi->cp_wait))
 		wake_up(&sbi->cp_wait);
 
 	bio_put(bio);
 }
 
 /*
+ * Return true, if pre_bio's bdev is same as its target device.
+ */
+struct block_device *f2fs_target_device(struct f2fs_sb_info *sbi,
+				block_t blk_addr, struct bio *bio)
+{
+	struct block_device *bdev = sbi->sb->s_bdev;
+	int i;
+
+	for (i = 0; i < sbi->s_ndevs; i++) {
+		if (FDEV(i).start_blk <= blk_addr &&
+					FDEV(i).end_blk >= blk_addr) {
+			blk_addr -= FDEV(i).start_blk;
+			bdev = FDEV(i).bdev;
+			break;
+		}
+	}
+	if (bio) {
+		bio->bi_bdev = bdev;
+		bio->bi_iter.bi_sector = SECTOR_FROM_BLOCK(blk_addr);
+	}
+	return bdev;
+}
+
+int f2fs_target_device_index(struct f2fs_sb_info *sbi, block_t blkaddr)
+{
+	int i;
+
+	for (i = 0; i < sbi->s_ndevs; i++)
+		if (FDEV(i).start_blk <= blkaddr && FDEV(i).end_blk >= blkaddr)
+			return i;
+	return 0;
+}
+
+static bool __same_bdev(struct f2fs_sb_info *sbi,
+				block_t blk_addr, struct bio *bio)
+{
+	return f2fs_target_device(sbi, blk_addr, NULL) == bio->bi_bdev;
+}
+
+/*
  * Low-level block read/write IO operations.
  */
 static struct bio *__bio_alloc(struct f2fs_sb_info *sbi, block_t blk_addr,
@@ -92,14 +174,60 @@ static struct bio *__bio_alloc(struct f2fs_sb_info *sbi, block_t blk_addr,
 
 	bio = f2fs_bio_alloc(npages);
 
-	bio->bi_bdev = sbi->sb->s_bdev;
-	bio->bi_iter.bi_sector = SECTOR_FROM_BLOCK(blk_addr);
+	f2fs_target_device(sbi, blk_addr, bio);
 	bio->bi_end_io = is_read ? f2fs_read_end_io : f2fs_write_end_io;
 	bio->bi_private = is_read ? NULL : sbi;
 
 	return bio;
 }
 
+static inline void __submit_bio(struct f2fs_sb_info *sbi,
+				struct bio *bio, enum page_type type)
+{
+	if (!is_read_io(bio_op(bio))) {
+		unsigned int start;
+
+		if (f2fs_sb_mounted_blkzoned(sbi->sb) &&
+			current->plug && (type == DATA || type == NODE))
+			blk_finish_plug(current->plug);
+
+		if (type != DATA && type != NODE)
+			goto submit_io;
+
+		start = bio->bi_iter.bi_size >> F2FS_BLKSIZE_BITS;
+		start %= F2FS_IO_SIZE(sbi);
+
+		if (start == 0)
+			goto submit_io;
+
+		/* fill dummy pages */
+		for (; start < F2FS_IO_SIZE(sbi); start++) {
+			struct page *page =
+				mempool_alloc(sbi->write_io_dummy,
+					GFP_NOIO | __GFP_ZERO | __GFP_NOFAIL);
+			f2fs_bug_on(sbi, !page);
+
+			SetPagePrivate(page);
+			set_page_private(page, (unsigned long)DUMMY_WRITTEN_PAGE);
+			lock_page(page);
+			if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE)
+				f2fs_bug_on(sbi, 1);
+		}
+		/*
+		 * In the NODE case, we lose next block address chain. So, we
+		 * need to do checkpoint in f2fs_sync_file.
+		 */
+		if (type == NODE)
+			set_sbi_flag(sbi, SBI_NEED_CP);
+	}
+submit_io:
+	if (is_read_io(bio_op(bio)))
+		trace_f2fs_submit_read_bio(sbi->sb, type, bio);
+	else
+		trace_f2fs_submit_write_bio(sbi->sb, type, bio);
+	submit_bio(bio_op(bio), bio);
+}
+
 static void __submit_merged_bio(struct f2fs_bio_info *io)
 {
 	struct f2fs_io_info *fio = &io->fio;
@@ -107,102 +235,286 @@ static void __submit_merged_bio(struct f2fs_bio_info *io)
 	if (!io->bio)
 		return;
 
-	if (is_read_io(fio->rw))
-		trace_f2fs_submit_read_bio(io->sbi->sb, fio, io->bio);
+	bio_set_op_attrs(io->bio, fio->op, fio->op_flags);
+
+	if (is_read_io(fio->op))
+		trace_f2fs_prepare_read_bio(io->sbi->sb, fio->type, io->bio);
 	else
-		trace_f2fs_submit_write_bio(io->sbi->sb, fio, io->bio);
+		trace_f2fs_prepare_write_bio(io->sbi->sb, fio->type, io->bio);
 
-	submit_bio(fio->rw, io->bio);
+	__submit_bio(io->sbi, io->bio, fio->type);
 	io->bio = NULL;
 }
 
-void f2fs_submit_merged_bio(struct f2fs_sb_info *sbi,
-				enum page_type type, int rw)
+static bool __has_merged_page(struct f2fs_bio_info *io,
+				struct inode *inode, nid_t ino, pgoff_t idx)
+{
+	struct bio_vec *bvec;
+	struct page *target;
+	int i;
+
+	if (!io->bio)
+		return false;
+
+	if (!inode && !ino)
+		return true;
+
+	bio_for_each_segment_all(bvec, io->bio, i) {
+
+		if (bvec->bv_page->mapping)
+			target = bvec->bv_page;
+		else
+			target = fscrypt_control_page(bvec->bv_page);
+
+		if (idx != target->index)
+			continue;
+
+		if (inode && inode == target->mapping->host)
+			return true;
+		if (ino && ino == ino_of_node(target))
+			return true;
+	}
+
+	return false;
+}
+
+static bool has_merged_page(struct f2fs_sb_info *sbi, struct inode *inode,
+				nid_t ino, pgoff_t idx, enum page_type type)
 {
 	enum page_type btype = PAGE_TYPE_OF_BIO(type);
+	enum temp_type temp;
 	struct f2fs_bio_info *io;
+	bool ret = false;
+
+	for (temp = HOT; temp < NR_TEMP_TYPE; temp++) {
+		io = sbi->write_io[btype] + temp;
+
+		down_read(&io->io_rwsem);
+		ret = __has_merged_page(io, inode, ino, idx);
+		up_read(&io->io_rwsem);
 
-	io = is_read_io(rw) ? &sbi->read_io : &sbi->write_io[btype];
+		/* TODO: use HOT temp only for meta pages now. */
+		if (ret || btype == META)
+			break;
+	}
+	return ret;
+}
+
+static void __f2fs_submit_merged_write(struct f2fs_sb_info *sbi,
+				enum page_type type, enum temp_type temp)
+{
+	enum page_type btype = PAGE_TYPE_OF_BIO(type);
+	struct f2fs_bio_info *io = sbi->write_io[btype] + temp;
 
 	down_write(&io->io_rwsem);
 
 	/* change META to META_FLUSH in the checkpoint procedure */
 	if (type >= META_FLUSH) {
 		io->fio.type = META_FLUSH;
-		if (test_opt(sbi, NOBARRIER))
-			io->fio.rw = WRITE_FLUSH | REQ_META | REQ_PRIO;
-		else
-			io->fio.rw = WRITE_FLUSH_FUA | REQ_META | REQ_PRIO;
+		io->fio.op = REQ_OP_WRITE;
+		io->fio.op_flags = REQ_META | REQ_PRIO;
+		if (!test_opt(sbi, NOBARRIER))
+			io->fio.op_flags |= WRITE_FLUSH | REQ_FUA;
 	}
 	__submit_merged_bio(io);
 	up_write(&io->io_rwsem);
 }
 
+static void __submit_merged_write_cond(struct f2fs_sb_info *sbi,
+				struct inode *inode, nid_t ino, pgoff_t idx,
+				enum page_type type, bool force)
+{
+	enum temp_type temp;
+
+	if (!force && !has_merged_page(sbi, inode, ino, idx, type))
+		return;
+
+	for (temp = HOT; temp < NR_TEMP_TYPE; temp++) {
+
+		__f2fs_submit_merged_write(sbi, type, temp);
+
+		/* TODO: use HOT temp only for meta pages now. */
+		if (type >= META)
+			break;
+	}
+}
+
+void f2fs_submit_merged_write(struct f2fs_sb_info *sbi, enum page_type type)
+{
+	__submit_merged_write_cond(sbi, NULL, 0, 0, type, true);
+}
+
+void f2fs_submit_merged_write_cond(struct f2fs_sb_info *sbi,
+				struct inode *inode, nid_t ino, pgoff_t idx,
+				enum page_type type)
+{
+	__submit_merged_write_cond(sbi, inode, ino, idx, type, false);
+}
+
+void f2fs_flush_merged_writes(struct f2fs_sb_info *sbi)
+{
+	f2fs_submit_merged_write(sbi, DATA);
+	f2fs_submit_merged_write(sbi, NODE);
+	f2fs_submit_merged_write(sbi, META);
+}
+
 /*
  * Fill the locked page with data located in the block address.
- * Return unlocked page.
+ * A caller needs to unlock the page on failure.
  */
 int f2fs_submit_page_bio(struct f2fs_io_info *fio)
 {
 	struct bio *bio;
-	struct page *page = fio->encrypted_page ? fio->encrypted_page : fio->page;
+	struct page *page = fio->encrypted_page ?
+			fio->encrypted_page : fio->page;
 
 	trace_f2fs_submit_page_bio(page, fio);
 	f2fs_trace_ios(fio, 0);
 
 	/* Allocate a new bio */
-	bio = __bio_alloc(fio->sbi, fio->blk_addr, 1, is_read_io(fio->rw));
+	bio = __bio_alloc(fio->sbi, fio->new_blkaddr, 1, is_read_io(fio->op));
 
-	if (bio_add_page(bio, page, PAGE_CACHE_SIZE, 0) < PAGE_CACHE_SIZE) {
+	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
 		bio_put(bio);
 		return -EFAULT;
 	}
+	bio_set_op_attrs(bio, fio->op, fio->op_flags);
 
-	submit_bio(fio->rw, bio);
+	__submit_bio(fio->sbi, bio, fio->type);
+
+	if (!is_read_io(fio->op))
+		inc_page_count(fio->sbi, WB_DATA_TYPE(fio->page));
 	return 0;
 }
 
-void f2fs_submit_page_mbio(struct f2fs_io_info *fio)
+int f2fs_submit_page_write(struct f2fs_io_info *fio)
 {
 	struct f2fs_sb_info *sbi = fio->sbi;
 	enum page_type btype = PAGE_TYPE_OF_BIO(fio->type);
-	struct f2fs_bio_info *io;
-	bool is_read = is_read_io(fio->rw);
+	struct f2fs_bio_info *io = sbi->write_io[btype] + fio->temp;
 	struct page *bio_page;
+	int err = 0;
 
-	io = is_read ? &sbi->read_io : &sbi->write_io[btype];
-
-	verify_block_addr(sbi, fio->blk_addr);
+	f2fs_bug_on(sbi, is_read_io(fio->op));
 
 	down_write(&io->io_rwsem);
+next:
+	if (fio->in_list) {
+		spin_lock(&io->io_lock);
+		if (list_empty(&io->io_list)) {
+			spin_unlock(&io->io_lock);
+			goto out_fail;
+		}
+		fio = list_first_entry(&io->io_list,
+						struct f2fs_io_info, list);
+		list_del(&fio->list);
+		spin_unlock(&io->io_lock);
+	}
 
-	if (!is_read)
-		inc_page_count(sbi, F2FS_WRITEBACK);
+	if (fio->old_blkaddr != NEW_ADDR)
+		verify_block_addr(sbi, fio->old_blkaddr);
+	verify_block_addr(sbi, fio->new_blkaddr);
+
+	bio_page = fio->encrypted_page ? fio->encrypted_page : fio->page;
 
-	if (io->bio && (io->last_block_in_bio != fio->blk_addr - 1 ||
-						io->fio.rw != fio->rw))
+	/* set submitted = 1 as a return value */
+	fio->submitted = 1;
+
+	inc_page_count(sbi, WB_DATA_TYPE(bio_page));
+
+	if (io->bio && (io->last_block_in_bio != fio->new_blkaddr - 1 ||
+	    (io->fio.op != fio->op || io->fio.op_flags != fio->op_flags) ||
+			!__same_bdev(sbi, fio->new_blkaddr, io->bio)))
 		__submit_merged_bio(io);
 alloc_new:
 	if (io->bio == NULL) {
-		int bio_blocks = MAX_BIO_BLOCKS(sbi);
-
-		io->bio = __bio_alloc(sbi, fio->blk_addr, bio_blocks, is_read);
+		if ((fio->type == DATA || fio->type == NODE) &&
+				fio->new_blkaddr & F2FS_IO_SIZE_MASK(sbi)) {
+			err = -EAGAIN;
+			dec_page_count(sbi, WB_DATA_TYPE(bio_page));
+			goto out_fail;
+		}
+		io->bio = __bio_alloc(sbi, fio->new_blkaddr,
+						BIO_MAX_PAGES, false);
 		io->fio = *fio;
 	}
 
-	bio_page = fio->encrypted_page ? fio->encrypted_page : fio->page;
-
-	if (bio_add_page(io->bio, bio_page, PAGE_CACHE_SIZE, 0) <
-							PAGE_CACHE_SIZE) {
+	if (bio_add_page(io->bio, bio_page, PAGE_SIZE, 0) < PAGE_SIZE) {
 		__submit_merged_bio(io);
 		goto alloc_new;
 	}
 
-	io->last_block_in_bio = fio->blk_addr;
+	io->last_block_in_bio = fio->new_blkaddr;
 	f2fs_trace_ios(fio, 0);
 
+	trace_f2fs_submit_page_write(fio->page, fio);
+
+	if (fio->in_list)
+		goto next;
+out_fail:
 	up_write(&io->io_rwsem);
-	trace_f2fs_submit_page_mbio(fio->page, fio);
+	return err;
+}
+
+static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr,
+							 unsigned nr_pages)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct fscrypt_ctx *ctx = NULL;
+	struct bio *bio;
+
+	if (f2fs_encrypted_file(inode)) {
+		ctx = fscrypt_get_ctx(inode, GFP_NOFS);
+		if (IS_ERR(ctx))
+			return ERR_CAST(ctx);
+
+		/* wait the page to be moved by cleaning */
+		f2fs_wait_on_block_writeback(sbi, blkaddr);
+	}
+
+	bio = bio_alloc(GFP_KERNEL, min_t(int, nr_pages, BIO_MAX_PAGES));
+	if (!bio) {
+		if (ctx)
+			fscrypt_release_ctx(ctx);
+		return ERR_PTR(-ENOMEM);
+	}
+	f2fs_target_device(sbi, blkaddr, bio);
+	bio->bi_end_io = f2fs_read_end_io;
+	bio->bi_private = ctx;
+	bio_set_op_attrs(bio, REQ_OP_READ, 0);
+
+	return bio;
+}
+
+/* This can handle encryption stuffs */
+static int f2fs_submit_page_read(struct inode *inode, struct page *page,
+							block_t blkaddr)
+{
+	struct bio *bio = f2fs_grab_read_bio(inode, blkaddr, 1);
+
+	if (IS_ERR(bio))
+		return PTR_ERR(bio);
+
+	if (bio_add_page(bio, page, PAGE_SIZE, 0) < PAGE_SIZE) {
+		bio_put(bio);
+		return -EFAULT;
+	}
+	__submit_bio(F2FS_I_SB(inode), bio, DATA);
+	return 0;
+}
+
+static void __set_data_blkaddr(struct dnode_of_data *dn)
+{
+	struct f2fs_node *rn = F2FS_NODE(dn->node_page);
+	__le32 *addr_array;
+	int base = 0;
+
+	if (IS_INODE(dn->node_page) && f2fs_has_extra_attr(dn->inode))
+		base = get_extra_isize(dn->inode);
+
+	/* Get physical address of data block */
+	addr_array = blkaddr_in_node(rn);
+	addr_array[base + dn->ofs_in_node] = cpu_to_le32(dn->data_blkaddr);
 }
 
 /*
@@ -213,39 +525,64 @@ alloc_new:
  */
 void set_data_blkaddr(struct dnode_of_data *dn)
 {
-	struct f2fs_node *rn;
-	__le32 *addr_array;
-	struct page *node_page = dn->node_page;
-	unsigned int ofs_in_node = dn->ofs_in_node;
-
-	f2fs_wait_on_page_writeback(node_page, NODE);
-
-	rn = F2FS_NODE(node_page);
+	f2fs_wait_on_page_writeback(dn->node_page, NODE, true);
+	__set_data_blkaddr(dn);
+	if (set_page_dirty(dn->node_page))
+		dn->node_changed = true;
+}
 
-	/* Get physical address of data block */
-	addr_array = blkaddr_in_node(rn);
-	addr_array[ofs_in_node] = cpu_to_le32(dn->data_blkaddr);
-	set_page_dirty(node_page);
+void f2fs_update_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr)
+{
+	dn->data_blkaddr = blkaddr;
+	set_data_blkaddr(dn);
+	f2fs_update_extent_cache(dn);
 }
 
-int reserve_new_block(struct dnode_of_data *dn)
+/* dn->ofs_in_node will be returned with up-to-date last block pointer */
+int reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
+	int err;
 
-	if (unlikely(is_inode_flag_set(F2FS_I(dn->inode), FI_NO_ALLOC)))
+	if (!count)
+		return 0;
+
+	if (unlikely(is_inode_flag_set(dn->inode, FI_NO_ALLOC)))
 		return -EPERM;
-	if (unlikely(!inc_valid_block_count(sbi, dn->inode, 1)))
-		return -ENOSPC;
+	if (unlikely((err = inc_valid_block_count(sbi, dn->inode, &count))))
+		return err;
 
-	trace_f2fs_reserve_new_block(dn->inode, dn->nid, dn->ofs_in_node);
+	trace_f2fs_reserve_new_blocks(dn->inode, dn->nid,
+						dn->ofs_in_node, count);
 
-	dn->data_blkaddr = NEW_ADDR;
-	set_data_blkaddr(dn);
-	mark_inode_dirty(dn->inode);
-	sync_inode_page(dn);
+	f2fs_wait_on_page_writeback(dn->node_page, NODE, true);
+
+	for (; count > 0; dn->ofs_in_node++) {
+		block_t blkaddr = datablock_addr(dn->inode,
+					dn->node_page, dn->ofs_in_node);
+		if (blkaddr == NULL_ADDR) {
+			dn->data_blkaddr = NEW_ADDR;
+			__set_data_blkaddr(dn);
+			count--;
+		}
+	}
+
+	if (set_page_dirty(dn->node_page))
+		dn->node_changed = true;
 	return 0;
 }
 
+/* Should keep dn->ofs_in_node unchanged */
+int reserve_new_block(struct dnode_of_data *dn)
+{
+	unsigned int ofs_in_node = dn->ofs_in_node;
+	int ret;
+
+	ret = reserve_new_blocks(dn, 1);
+	dn->ofs_in_node = ofs_in_node;
+	return ret;
+}
+
 int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index)
 {
 	bool need_put = dn->inode_page ? false : true;
@@ -264,7 +601,7 @@ int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index)
 
 int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index)
 {
-	struct extent_info ei;
+	struct extent_info ei  = {0,0,0};
 	struct inode *inode = dn->inode;
 
 	if (f2fs_lookup_extent_cache(inode, index, &ei)) {
@@ -276,22 +613,13 @@ int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index)
 }
 
 struct page *get_read_data_page(struct inode *inode, pgoff_t index,
-						int rw, bool for_write)
+						int op_flags, bool for_write)
 {
 	struct address_space *mapping = inode->i_mapping;
 	struct dnode_of_data dn;
 	struct page *page;
-	struct extent_info ei;
+	struct extent_info ei = {0,0,0};
 	int err;
-	struct f2fs_io_info fio = {
-		.sbi = F2FS_I_SB(inode),
-		.type = DATA,
-		.rw = rw,
-		.encrypted_page = NULL,
-	};
-
-	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
-		return read_mapping_page(mapping, index, NULL);
 
 	page = f2fs_grab_cache_page(mapping, index, for_write);
 	if (!page)
@@ -325,15 +653,14 @@ got_it:
 	 * see, f2fs_add_link -> get_new_data_page -> init_inode_metadata.
 	 */
 	if (dn.data_blkaddr == NEW_ADDR) {
-		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
-		SetPageUptodate(page);
+		zero_user_segment(page, 0, PAGE_SIZE);
+		if (!PageUptodate(page))
+			SetPageUptodate(page);
 		unlock_page(page);
 		return page;
 	}
 
-	fio.blk_addr = dn.data_blkaddr;
-	fio.page = page;
-	err = f2fs_submit_page_bio(&fio);
+	err = f2fs_submit_page_read(inode, page, dn.data_blkaddr);
 	if (err)
 		goto put_err;
 	return page;
@@ -353,7 +680,7 @@ struct page *find_data_page(struct inode *inode, pgoff_t index)
 		return page;
 	f2fs_put_page(page, 0);
 
-	page = get_read_data_page(inode, index, READ_SYNC, false);
+	page = get_read_data_page(inode, index, REQ_SYNC, false);
 	if (IS_ERR(page))
 		return page;
 
@@ -379,20 +706,20 @@ struct page *get_lock_data_page(struct inode *inode, pgoff_t index,
 	struct address_space *mapping = inode->i_mapping;
 	struct page *page;
 repeat:
-	page = get_read_data_page(inode, index, READ_SYNC, for_write);
+	page = get_read_data_page(inode, index, REQ_SYNC, for_write);
 	if (IS_ERR(page))
 		return page;
 
 	/* wait for read completion */
 	lock_page(page);
-	if (unlikely(!PageUptodate(page))) {
-		f2fs_put_page(page, 1);
-		return ERR_PTR(-EIO);
-	}
 	if (unlikely(page->mapping != mapping)) {
 		f2fs_put_page(page, 1);
 		goto repeat;
 	}
+	if (unlikely(!PageUptodate(page))) {
+		f2fs_put_page(page, 1);
+		return ERR_PTR(-EIO);
+	}
 	return page;
 }
 
@@ -412,7 +739,7 @@ struct page *get_new_data_page(struct inode *inode,
 	struct page *page;
 	struct dnode_of_data dn;
 	int err;
-repeat:
+
 	page = f2fs_grab_cache_page(mapping, index, true);
 	if (!page) {
 		/*
@@ -436,125 +763,119 @@ repeat:
 		goto got_it;
 
 	if (dn.data_blkaddr == NEW_ADDR) {
-		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
-		SetPageUptodate(page);
+		zero_user_segment(page, 0, PAGE_SIZE);
+		if (!PageUptodate(page))
+			SetPageUptodate(page);
 	} else {
 		f2fs_put_page(page, 1);
 
-		page = get_read_data_page(inode, index, READ_SYNC, true);
+		/* if ipage exists, blkaddr should be NEW_ADDR */
+		f2fs_bug_on(F2FS_I_SB(inode), ipage);
+		page = get_lock_data_page(inode, index, true);
 		if (IS_ERR(page))
-			goto repeat;
-
-		/* wait for read completion */
-		lock_page(page);
+			return page;
 	}
 got_it:
 	if (new_i_size && i_size_read(inode) <
-				((loff_t)(index + 1) << PAGE_CACHE_SHIFT)) {
-		i_size_write(inode, ((loff_t)(index + 1) << PAGE_CACHE_SHIFT));
-		/* Only the directory inode sets new_i_size */
-		set_inode_flag(F2FS_I(inode), FI_UPDATE_DIR);
-	}
+				((loff_t)(index + 1) << PAGE_SHIFT))
+		f2fs_i_size_write(inode, ((loff_t)(index + 1) << PAGE_SHIFT));
 	return page;
 }
 
 static int __allocate_data_block(struct dnode_of_data *dn)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
-	struct f2fs_inode_info *fi = F2FS_I(dn->inode);
 	struct f2fs_summary sum;
 	struct node_info ni;
-	int seg = CURSEG_WARM_DATA;
 	pgoff_t fofs;
+	blkcnt_t count = 1;
+	int err;
 
-	if (unlikely(is_inode_flag_set(F2FS_I(dn->inode), FI_NO_ALLOC)))
+	if (unlikely(is_inode_flag_set(dn->inode, FI_NO_ALLOC)))
 		return -EPERM;
 
-	dn->data_blkaddr = datablock_addr(dn->node_page, dn->ofs_in_node);
+	dn->data_blkaddr = datablock_addr(dn->inode,
+				dn->node_page, dn->ofs_in_node);
 	if (dn->data_blkaddr == NEW_ADDR)
 		goto alloc;
 
-	if (unlikely(!inc_valid_block_count(sbi, dn->inode, 1)))
-		return -ENOSPC;
+	if (unlikely((err = inc_valid_block_count(sbi, dn->inode, &count))))
+		return err;
 
 alloc:
 	get_node_info(sbi, dn->nid, &ni);
 	set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
 
-	if (dn->ofs_in_node == 0 && dn->inode_page == dn->node_page)
-		seg = CURSEG_DIRECT_IO;
-
 	allocate_data_block(sbi, NULL, dn->data_blkaddr, &dn->data_blkaddr,
-								&sum, seg);
+					&sum, CURSEG_WARM_DATA, NULL, false);
 	set_data_blkaddr(dn);
 
 	/* update i_size */
-	fofs = start_bidx_of_node(ofs_of_node(dn->node_page), fi) +
+	fofs = start_bidx_of_node(ofs_of_node(dn->node_page), dn->inode) +
 							dn->ofs_in_node;
-	if (i_size_read(dn->inode) < ((loff_t)(fofs + 1) << PAGE_CACHE_SHIFT))
-		i_size_write(dn->inode,
-				((loff_t)(fofs + 1) << PAGE_CACHE_SHIFT));
-
-	/* direct IO doesn't use extent cache to maximize the performance */
-	f2fs_drop_largest_extent(dn->inode, fofs);
-
+	if (i_size_read(dn->inode) < ((loff_t)(fofs + 1) << PAGE_SHIFT))
+		f2fs_i_size_write(dn->inode,
+				((loff_t)(fofs + 1) << PAGE_SHIFT));
 	return 0;
 }
 
-static void __allocate_data_blocks(struct inode *inode, loff_t offset,
-							size_t count)
+static inline bool __force_buffered_io(struct inode *inode, int rw)
 {
-	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct dnode_of_data dn;
-	u64 start = F2FS_BYTES_TO_BLK(offset);
-	u64 len = F2FS_BYTES_TO_BLK(count);
-	bool allocated;
-	u64 end_offset;
-
-	while (len) {
-		f2fs_balance_fs(sbi);
-		f2fs_lock_op(sbi);
-
-		/* When reading holes, we need its node page */
-		set_new_dnode(&dn, inode, NULL, NULL, 0);
-		if (get_dnode_of_data(&dn, start, ALLOC_NODE))
-			goto out;
-
-		allocated = false;
-		end_offset = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
+	return (f2fs_encrypted_file(inode) ||
+			(rw == WRITE && test_opt(F2FS_I_SB(inode), LFS)) ||
+			F2FS_I_SB(inode)->s_ndevs);
+}
 
-		while (dn.ofs_in_node < end_offset && len) {
-			block_t blkaddr;
+int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from)
+{
+	struct inode *inode = file_inode(iocb->ki_filp);
+	struct f2fs_map_blocks map;
+	int err = 0;
 
-			if (unlikely(f2fs_cp_error(sbi)))
-				goto sync_out;
+	if (is_inode_flag_set(inode, FI_NO_PREALLOC))
+		return 0;
 
-			blkaddr = datablock_addr(dn.node_page, dn.ofs_in_node);
-			if (blkaddr == NULL_ADDR || blkaddr == NEW_ADDR) {
-				if (__allocate_data_block(&dn))
-					goto sync_out;
-				allocated = true;
-			}
-			len--;
-			start++;
-			dn.ofs_in_node++;
-		}
+	map.m_lblk = F2FS_BLK_ALIGN(iocb->ki_pos);
+	map.m_len = F2FS_BYTES_TO_BLK(iocb->ki_pos + iov_iter_count(from));
+	if (map.m_len > map.m_lblk)
+		map.m_len -= map.m_lblk;
+	else
+		map.m_len = 0;
 
-		if (allocated)
-			sync_inode_page(&dn);
+	map.m_next_pgofs = NULL;
 
-		f2fs_put_dnode(&dn);
-		f2fs_unlock_op(sbi);
+	if (iocb->ki_flags & IOCB_DIRECT) {
+		err = f2fs_convert_inline_inode(inode);
+		if (err)
+			return err;
+		return f2fs_map_blocks(inode, &map, 1,
+			__force_buffered_io(inode, WRITE) ?
+				F2FS_GET_BLOCK_PRE_AIO :
+				F2FS_GET_BLOCK_PRE_DIO);
 	}
-	return;
+	if (iocb->ki_pos + iov_iter_count(from) > MAX_INLINE_DATA(inode)) {
+		err = f2fs_convert_inline_inode(inode);
+		if (err)
+			return err;
+	}
+	if (!f2fs_has_inline_data(inode))
+		return f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_AIO);
+	return err;
+}
 
-sync_out:
-	if (allocated)
-		sync_inode_page(&dn);
-	f2fs_put_dnode(&dn);
-out:
-	f2fs_unlock_op(sbi);
-	return;
+static inline void __do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock)
+{
+	if (flag == F2FS_GET_BLOCK_PRE_AIO) {
+		if (lock)
+			down_read(&sbi->node_change);
+		else
+			up_read(&sbi->node_change);
+	} else {
+		if (lock)
+			f2fs_lock_op(sbi);
+		else
+			f2fs_unlock_op(sbi);
+	}
 }
 
 /*
@@ -566,184 +887,210 @@ out:
  *     b. do not use extent cache for better performance
  *     c. give the block addresses to blockdev
  */
-static int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
+int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
 						int create, int flag)
 {
 	unsigned int maxblocks = map->m_len;
 	struct dnode_of_data dn;
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	int mode = create ? ALLOC_NODE : LOOKUP_NODE_RA;
-	pgoff_t pgofs, end_offset;
+	int mode = create ? ALLOC_NODE : LOOKUP_NODE;
+	pgoff_t pgofs, end_offset, end;
 	int err = 0, ofs = 1;
-	struct extent_info ei;
-	bool allocated = false;
+	unsigned int ofs_in_node, last_ofs_in_node;
+	blkcnt_t prealloc;
+	struct extent_info ei = {0,0,0};
+	block_t blkaddr;
+
+	if (!maxblocks)
+		return 0;
 
 	map->m_len = 0;
 	map->m_flags = 0;
 
 	/* it only supports block size == page size */
 	pgofs =	(pgoff_t)map->m_lblk;
+	end = pgofs + maxblocks;
 
-	if (f2fs_lookup_extent_cache(inode, pgofs, &ei)) {
+	if (!create && f2fs_lookup_extent_cache(inode, pgofs, &ei)) {
 		map->m_pblk = ei.blk + pgofs - ei.fofs;
 		map->m_len = min((pgoff_t)maxblocks, ei.fofs + ei.len - pgofs);
 		map->m_flags = F2FS_MAP_MAPPED;
 		goto out;
 	}
 
+next_dnode:
 	if (create)
-		f2fs_lock_op(F2FS_I_SB(inode));
+		__do_map_lock(sbi, flag, true);
 
 	/* When reading holes, we need its node page */
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = get_dnode_of_data(&dn, pgofs, mode);
 	if (err) {
-		if (err == -ENOENT)
+		if (flag == F2FS_GET_BLOCK_BMAP)
+			map->m_pblk = 0;
+		if (err == -ENOENT) {
 			err = 0;
+			if (map->m_next_pgofs)
+				*map->m_next_pgofs =
+					get_next_page_offset(&dn, pgofs);
+		}
 		goto unlock_out;
 	}
 
-	if (dn.data_blkaddr == NEW_ADDR || dn.data_blkaddr == NULL_ADDR) {
+	prealloc = 0;
+	last_ofs_in_node = ofs_in_node = dn.ofs_in_node;
+	end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
+
+next_block:
+	blkaddr = datablock_addr(dn.inode, dn.node_page, dn.ofs_in_node);
+
+	if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR) {
 		if (create) {
 			if (unlikely(f2fs_cp_error(sbi))) {
 				err = -EIO;
-				goto put_out;
+				goto sync_out;
+			}
+			if (flag == F2FS_GET_BLOCK_PRE_AIO) {
+				if (blkaddr == NULL_ADDR) {
+					prealloc++;
+					last_ofs_in_node = dn.ofs_in_node;
+				}
+			} else {
+				err = __allocate_data_block(&dn);
+				if (!err)
+					set_inode_flag(inode, FI_APPEND_WRITE);
 			}
-			err = __allocate_data_block(&dn);
 			if (err)
-				goto put_out;
-			allocated = true;
-			map->m_flags = F2FS_MAP_NEW;
+				goto sync_out;
+			map->m_flags |= F2FS_MAP_NEW;
+			blkaddr = dn.data_blkaddr;
 		} else {
-			if (flag != F2FS_GET_BLOCK_FIEMAP ||
-						dn.data_blkaddr != NEW_ADDR) {
-				if (flag == F2FS_GET_BLOCK_BMAP)
-					err = -ENOENT;
-				goto put_out;
+			if (flag == F2FS_GET_BLOCK_BMAP) {
+				map->m_pblk = 0;
+				goto sync_out;
 			}
-
-			/*
-			 * preallocated unwritten block should be mapped
-			 * for fiemap.
-			 */
-			if (dn.data_blkaddr == NEW_ADDR)
-				map->m_flags = F2FS_MAP_UNWRITTEN;
+			if (flag == F2FS_GET_BLOCK_FIEMAP &&
+						blkaddr == NULL_ADDR) {
+				if (map->m_next_pgofs)
+					*map->m_next_pgofs = pgofs + 1;
+			}
+			if (flag != F2FS_GET_BLOCK_FIEMAP ||
+						blkaddr != NEW_ADDR)
+				goto sync_out;
 		}
 	}
 
-	map->m_flags |= F2FS_MAP_MAPPED;
-	map->m_pblk = dn.data_blkaddr;
-	map->m_len = 1;
+	if (flag == F2FS_GET_BLOCK_PRE_AIO)
+		goto skip;
+
+	if (map->m_len == 0) {
+		/* preallocated unwritten block should be mapped for fiemap. */
+		if (blkaddr == NEW_ADDR)
+			map->m_flags |= F2FS_MAP_UNWRITTEN;
+		map->m_flags |= F2FS_MAP_MAPPED;
+
+		map->m_pblk = blkaddr;
+		map->m_len = 1;
+	} else if ((map->m_pblk != NEW_ADDR &&
+			blkaddr == (map->m_pblk + ofs)) ||
+			(map->m_pblk == NEW_ADDR && blkaddr == NEW_ADDR) ||
+			flag == F2FS_GET_BLOCK_PRE_DIO) {
+		ofs++;
+		map->m_len++;
+	} else {
+		goto sync_out;
+	}
 
-	end_offset = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
+skip:
 	dn.ofs_in_node++;
 	pgofs++;
 
-get_next:
-	if (dn.ofs_in_node >= end_offset) {
-		if (allocated)
-			sync_inode_page(&dn);
-		allocated = false;
-		f2fs_put_dnode(&dn);
+	/* preallocate blocks in batch for one dnode page */
+	if (flag == F2FS_GET_BLOCK_PRE_AIO &&
+			(pgofs == end || dn.ofs_in_node == end_offset)) {
 
-		set_new_dnode(&dn, inode, NULL, NULL, 0);
-		err = get_dnode_of_data(&dn, pgofs, mode);
-		if (err) {
-			if (err == -ENOENT)
-				err = 0;
-			goto unlock_out;
-		}
+		dn.ofs_in_node = ofs_in_node;
+		err = reserve_new_blocks(&dn, prealloc);
+		if (err)
+			goto sync_out;
 
-		end_offset = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
+		map->m_len += dn.ofs_in_node - ofs_in_node;
+		if (prealloc && dn.ofs_in_node != last_ofs_in_node + 1) {
+			err = -ENOSPC;
+			goto sync_out;
+		}
+		dn.ofs_in_node = end_offset;
 	}
 
-	if (maxblocks > map->m_len) {
-		block_t blkaddr = datablock_addr(dn.node_page, dn.ofs_in_node);
+	if (pgofs >= end)
+		goto sync_out;
+	else if (dn.ofs_in_node < end_offset)
+		goto next_block;
 
-		if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR) {
-			if (create) {
-				if (unlikely(f2fs_cp_error(sbi))) {
-					err = -EIO;
-					goto sync_out;
-				}
-				err = __allocate_data_block(&dn);
-				if (err)
-					goto sync_out;
-				allocated = true;
-				map->m_flags |= F2FS_MAP_NEW;
-				blkaddr = dn.data_blkaddr;
-			} else {
-				/*
-				 * we only merge preallocated unwritten blocks
-				 * for fiemap.
-				 */
-				if (flag != F2FS_GET_BLOCK_FIEMAP ||
-						blkaddr != NEW_ADDR)
-					goto sync_out;
-			}
-		}
+	f2fs_put_dnode(&dn);
 
-		/* Give more consecutive addresses for the readahead */
-		if ((map->m_pblk != NEW_ADDR &&
-				blkaddr == (map->m_pblk + ofs)) ||
-				(map->m_pblk == NEW_ADDR &&
-				blkaddr == NEW_ADDR)) {
-			ofs++;
-			dn.ofs_in_node++;
-			pgofs++;
-			map->m_len++;
-			goto get_next;
-		}
+	if (create) {
+		__do_map_lock(sbi, flag, false);
+		f2fs_balance_fs(sbi, dn.node_changed);
 	}
+	goto next_dnode;
+
 sync_out:
-	if (allocated)
-		sync_inode_page(&dn);
-put_out:
 	f2fs_put_dnode(&dn);
 unlock_out:
-	if (create)
-		f2fs_unlock_op(F2FS_I_SB(inode));
+	if (create) {
+		__do_map_lock(sbi, flag, false);
+		f2fs_balance_fs(sbi, dn.node_changed);
+	}
 out:
 	trace_f2fs_map_blocks(inode, map, err);
 	return err;
 }
 
 static int __get_data_block(struct inode *inode, sector_t iblock,
-			struct buffer_head *bh, int create, int flag)
+			struct buffer_head *bh, int create, int flag,
+			pgoff_t *next_pgofs)
 {
 	struct f2fs_map_blocks map;
-	int ret;
+	int err;
 
 	map.m_lblk = iblock;
 	map.m_len = bh->b_size >> inode->i_blkbits;
+	map.m_next_pgofs = next_pgofs;
 
-	ret = f2fs_map_blocks(inode, &map, create, flag);
-	if (!ret) {
+	err = f2fs_map_blocks(inode, &map, create, flag);
+	if (!err) {
 		map_bh(bh, inode->i_sb, map.m_pblk);
 		bh->b_state = (bh->b_state & ~F2FS_MAP_FLAGS) | map.m_flags;
-		bh->b_size = map.m_len << inode->i_blkbits;
+		bh->b_size = (u64)map.m_len << inode->i_blkbits;
 	}
-	return ret;
+	return err;
 }
 
 static int get_data_block(struct inode *inode, sector_t iblock,
-			struct buffer_head *bh_result, int create, int flag)
+			struct buffer_head *bh_result, int create, int flag,
+			pgoff_t *next_pgofs)
 {
-	return __get_data_block(inode, iblock, bh_result, create, flag);
+	return __get_data_block(inode, iblock, bh_result, create,
+							flag, next_pgofs);
 }
 
 static int get_data_block_dio(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh_result, int create)
 {
 	return __get_data_block(inode, iblock, bh_result, create,
-						F2FS_GET_BLOCK_DIO);
+						F2FS_GET_BLOCK_DEFAULT, NULL);
 }
 
 static int get_data_block_bmap(struct inode *inode, sector_t iblock,
 			struct buffer_head *bh_result, int create)
 {
+	/* Block number less than F2FS MAX BLOCKS */
+	if (unlikely(iblock >= F2FS_I_SB(inode)->max_file_blocks))
+		return -EFBIG;
+
 	return __get_data_block(inode, iblock, bh_result, create,
-						F2FS_GET_BLOCK_BMAP);
+						F2FS_GET_BLOCK_BMAP, NULL);
 }
 
 static inline sector_t logical_to_blk(struct inode *inode, loff_t offset)
@@ -761,10 +1108,9 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 {
 	struct buffer_head map_bh;
 	sector_t start_blk, last_blk;
-	loff_t isize = i_size_read(inode);
+	pgoff_t next_pgofs;
 	u64 logical = 0, phys = 0, size = 0;
 	u32 flags = 0;
-	bool past_eof = false, whole_file = false;
 	int ret = 0;
 
 	ret = fiemap_check_flags(fieinfo, FIEMAP_FLAG_SYNC);
@@ -777,82 +1123,55 @@ int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
 			return ret;
 	}
 
-	mutex_lock(&inode->i_mutex);
-
-	if (len >= isize) {
-		whole_file = true;
-		len = isize;
-	}
+	inode_lock(inode);
 
 	if (logical_to_blk(inode, len) == 0)
 		len = blk_to_logical(inode, 1);
 
 	start_blk = logical_to_blk(inode, start);
 	last_blk = logical_to_blk(inode, start + len - 1);
+
 next:
 	memset(&map_bh, 0, sizeof(struct buffer_head));
 	map_bh.b_size = len;
 
 	ret = get_data_block(inode, start_blk, &map_bh, 0,
-					F2FS_GET_BLOCK_FIEMAP);
+					F2FS_GET_BLOCK_FIEMAP, &next_pgofs);
 	if (ret)
 		goto out;
 
 	/* HOLE */
 	if (!buffer_mapped(&map_bh)) {
-		start_blk++;
-
-		if (!past_eof && blk_to_logical(inode, start_blk) >= isize)
-			past_eof = 1;
-
-		if (past_eof && size) {
-			flags |= FIEMAP_EXTENT_LAST;
-			ret = fiemap_fill_next_extent(fieinfo, logical,
-					phys, size, flags);
-		} else if (size) {
-			ret = fiemap_fill_next_extent(fieinfo, logical,
-					phys, size, flags);
-			size = 0;
-		}
-
-		/* if we have holes up to/past EOF then we're done */
-		if (start_blk > last_blk || past_eof || ret)
-			goto out;
-	} else {
-		if (start_blk > last_blk && !whole_file) {
-			ret = fiemap_fill_next_extent(fieinfo, logical,
-					phys, size, flags);
-			goto out;
-		}
+		start_blk = next_pgofs;
 
-		/*
-		 * if size != 0 then we know we already have an extent
-		 * to add, so add it.
-		 */
-		if (size) {
-			ret = fiemap_fill_next_extent(fieinfo, logical,
-					phys, size, flags);
-			if (ret)
-				goto out;
-		}
+		if (blk_to_logical(inode, start_blk) < blk_to_logical(inode,
+					F2FS_I_SB(inode)->max_file_blocks))
+			goto prep_next;
 
-		logical = blk_to_logical(inode, start_blk);
-		phys = blk_to_logical(inode, map_bh.b_blocknr);
-		size = map_bh.b_size;
-		flags = 0;
-		if (buffer_unwritten(&map_bh))
-			flags = FIEMAP_EXTENT_UNWRITTEN;
+		flags |= FIEMAP_EXTENT_LAST;
+	}
 
-		start_blk += logical_to_blk(inode, size);
+	if (size) {
+		if (f2fs_encrypted_inode(inode))
+			flags |= FIEMAP_EXTENT_DATA_ENCRYPTED;
 
-		/*
-		 * If we are past the EOF, then we need to make sure as
-		 * soon as we find a hole that the last extent we found
-		 * is marked with FIEMAP_EXTENT_LAST
-		 */
-		if (!past_eof && logical + size >= isize)
-			past_eof = true;
+		ret = fiemap_fill_next_extent(fieinfo, logical,
+				phys, size, flags);
 	}
+
+	if (start_blk > last_blk || ret)
+		goto out;
+
+	logical = blk_to_logical(inode, start_blk);
+	phys = blk_to_logical(inode, map_bh.b_blocknr);
+	size = map_bh.b_size;
+	flags = 0;
+	if (buffer_unwritten(&map_bh))
+		flags = FIEMAP_EXTENT_UNWRITTEN;
+
+	start_blk += logical_to_blk(inode, size);
+
+prep_next:
 	cond_resched();
 	if (fatal_signal_pending(current))
 		ret = -EINTR;
@@ -862,7 +1181,7 @@ out:
 	if (ret == 1)
 		ret = 0;
 
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return ret;
 }
 
@@ -884,19 +1203,20 @@ static int f2fs_mpage_readpages(struct address_space *mapping,
 	sector_t last_block;
 	sector_t last_block_in_file;
 	sector_t block_nr;
-	struct block_device *bdev = inode->i_sb->s_bdev;
 	struct f2fs_map_blocks map;
 
 	map.m_pblk = 0;
 	map.m_lblk = 0;
 	map.m_len = 0;
 	map.m_flags = 0;
+	map.m_next_pgofs = NULL;
 
 	for (page_idx = 0; nr_pages; page_idx++, nr_pages--) {
 
-		prefetchw(&page->flags);
 		if (pages) {
-			page = list_entry(pages->prev, struct page, lru);
+			page = list_last_entry(pages, struct page, lru);
+
+			prefetchw(&page->flags);
 			list_del(&page->lru);
 			if (add_to_page_cache_lru(page, mapping,
 						  page->index, GFP_KERNEL))
@@ -929,7 +1249,7 @@ static int f2fs_mpage_readpages(struct address_space *mapping,
 			map.m_len = last_block - block_in_file;
 
 			if (f2fs_map_blocks(inode, &map, 0,
-							F2FS_GET_BLOCK_READ))
+						F2FS_GET_BLOCK_DEFAULT))
 				goto set_error_page;
 		}
 got_it:
@@ -942,8 +1262,9 @@ got_it:
 				goto confused;
 			}
 		} else {
-			zero_user_segment(page, 0, PAGE_CACHE_SIZE);
-			SetPageUptodate(page);
+			zero_user_segment(page, 0, PAGE_SIZE);
+			if (!PageUptodate(page))
+				SetPageUptodate(page);
 			unlock_page(page);
 			goto next_page;
 		}
@@ -952,37 +1273,18 @@ got_it:
 		 * This page will go to BIO.  Do we need to send this
 		 * BIO off first?
 		 */
-		if (bio && (last_block_in_bio != block_nr - 1)) {
+		if (bio && (last_block_in_bio != block_nr - 1 ||
+			!__same_bdev(F2FS_I_SB(inode), block_nr, bio))) {
 submit_and_realloc:
-			submit_bio(READ, bio);
+			__submit_bio(F2FS_I_SB(inode), bio, DATA);
 			bio = NULL;
 		}
 		if (bio == NULL) {
-			struct f2fs_crypto_ctx *ctx = NULL;
-
-			if (f2fs_encrypted_inode(inode) &&
-					S_ISREG(inode->i_mode)) {
-
-				ctx = f2fs_get_crypto_ctx(inode);
-				if (IS_ERR(ctx))
-					goto set_error_page;
-
-				/* wait the page to be moved by cleaning */
-				f2fs_wait_on_encrypted_page_writeback(
-						F2FS_I_SB(inode), block_nr);
-			}
-
-			bio = bio_alloc(GFP_KERNEL,
-				min_t(int, nr_pages, BIO_MAX_PAGES));
-			if (!bio) {
-				if (ctx)
-					f2fs_release_crypto_ctx(ctx);
+			bio = f2fs_grab_read_bio(inode, block_nr, nr_pages);
+			if (IS_ERR(bio)) {
+				bio = NULL;
 				goto set_error_page;
 			}
-			bio->bi_bdev = bdev;
-			bio->bi_iter.bi_sector = SECTOR_FROM_BLOCK(block_nr);
-			bio->bi_end_io = f2fs_read_end_io;
-			bio->bi_private = ctx;
 		}
 
 		if (bio_add_page(bio, page, blocksize, 0) < blocksize)
@@ -992,22 +1294,22 @@ submit_and_realloc:
 		goto next_page;
 set_error_page:
 		SetPageError(page);
-		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
+		zero_user_segment(page, 0, PAGE_SIZE);
 		unlock_page(page);
 		goto next_page;
 confused:
 		if (bio) {
-			submit_bio(READ, bio);
+			__submit_bio(F2FS_I_SB(inode), bio, DATA);
 			bio = NULL;
 		}
 		unlock_page(page);
 next_page:
 		if (pages)
-			page_cache_release(page);
+			put_page(page);
 	}
 	BUG_ON(pages && !list_empty(pages));
 	if (bio)
-		submit_bio(READ, bio);
+		__submit_bio(F2FS_I_SB(inode), bio, DATA);
 	return 0;
 }
 
@@ -1031,7 +1333,7 @@ static int f2fs_read_data_pages(struct file *file,
 			struct list_head *pages, unsigned nr_pages)
 {
 	struct inode *inode = file->f_mapping->host;
-	struct page *page = list_entry(pages->prev, struct page, lru);
+	struct page *page = list_last_entry(pages, struct page, lru);
 
 	trace_f2fs_readpages(inode, page, nr_pages);
 
@@ -1042,86 +1344,171 @@ static int f2fs_read_data_pages(struct file *file,
 	return f2fs_mpage_readpages(mapping, pages, NULL, nr_pages);
 }
 
+static int encrypt_one_page(struct f2fs_io_info *fio)
+{
+	struct inode *inode = fio->page->mapping->host;
+	gfp_t gfp_flags = GFP_NOFS;
+
+	if (!f2fs_encrypted_file(inode))
+		return 0;
+
+	/* wait for GCed encrypted page writeback */
+	f2fs_wait_on_block_writeback(fio->sbi, fio->old_blkaddr);
+
+retry_encrypt:
+	fio->encrypted_page = fscrypt_encrypt_page(inode, fio->page,
+			PAGE_SIZE, 0, fio->page->index, gfp_flags);
+	if (!IS_ERR(fio->encrypted_page))
+		return 0;
+
+	/* flush pending IOs and wait for a while in the ENOMEM case */
+	if (PTR_ERR(fio->encrypted_page) == -ENOMEM) {
+		f2fs_flush_merged_writes(fio->sbi);
+		congestion_wait(BLK_RW_ASYNC, HZ/50);
+		gfp_flags |= __GFP_NOFAIL;
+		goto retry_encrypt;
+	}
+	return PTR_ERR(fio->encrypted_page);
+}
+
+static inline bool need_inplace_update(struct f2fs_io_info *fio)
+{
+	struct inode *inode = fio->page->mapping->host;
+
+	if (S_ISDIR(inode->i_mode) || f2fs_is_atomic_file(inode))
+		return false;
+	if (is_cold_data(fio->page))
+		return false;
+	if (IS_ATOMIC_WRITTEN_PAGE(fio->page))
+		return false;
+
+	return need_inplace_update_policy(inode, fio);
+}
+
+static inline bool valid_ipu_blkaddr(struct f2fs_io_info *fio)
+{
+	if (fio->old_blkaddr == NEW_ADDR)
+		return false;
+	if (fio->old_blkaddr == NULL_ADDR)
+		return false;
+	return true;
+}
+
 int do_write_data_page(struct f2fs_io_info *fio)
 {
 	struct page *page = fio->page;
 	struct inode *inode = page->mapping->host;
 	struct dnode_of_data dn;
+	struct extent_info ei = {0,0,0};
+	bool ipu_force = false;
 	int err = 0;
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
+	if (need_inplace_update(fio) &&
+			f2fs_lookup_extent_cache(inode, page->index, &ei)) {
+		fio->old_blkaddr = ei.blk + page->index - ei.fofs;
+
+		if (valid_ipu_blkaddr(fio)) {
+			ipu_force = true;
+			fio->need_lock = LOCK_DONE;
+			goto got_it;
+		}
+	}
+
+	/* Deadlock due to between page->lock and f2fs_lock_op */
+	if (fio->need_lock == LOCK_REQ && !f2fs_trylock_op(fio->sbi))
+		return -EAGAIN;
+
 	err = get_dnode_of_data(&dn, page->index, LOOKUP_NODE);
 	if (err)
-		return err;
+		goto out;
 
-	fio->blk_addr = dn.data_blkaddr;
+	fio->old_blkaddr = dn.data_blkaddr;
 
 	/* This page is already truncated */
-	if (fio->blk_addr == NULL_ADDR) {
+	if (fio->old_blkaddr == NULL_ADDR) {
 		ClearPageUptodate(page);
 		goto out_writepage;
 	}
+got_it:
+	/*
+	 * If current allocation needs SSR,
+	 * it had better in-place writes for updated data.
+	 */
+	if (ipu_force || (valid_ipu_blkaddr(fio) && need_inplace_update(fio))) {
+		err = encrypt_one_page(fio);
+		if (err)
+			goto out_writepage;
 
-	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode)) {
-
-		/* wait for GCed encrypted page writeback */
-		f2fs_wait_on_encrypted_page_writeback(F2FS_I_SB(inode),
-							fio->blk_addr);
+		set_page_writeback(page);
+		f2fs_put_dnode(&dn);
+		if (fio->need_lock == LOCK_REQ)
+			f2fs_unlock_op(fio->sbi);
+		err = rewrite_data_page(fio);
+		trace_f2fs_do_write_data_page(fio->page, IPU);
+		set_inode_flag(inode, FI_UPDATE_WRITE);
+		return err;
+	}
 
-		fio->encrypted_page = f2fs_encrypt(inode, fio->page);
-		if (IS_ERR(fio->encrypted_page)) {
-			err = PTR_ERR(fio->encrypted_page);
+	if (fio->need_lock == LOCK_RETRY) {
+		if (!f2fs_trylock_op(fio->sbi)) {
+			err = -EAGAIN;
 			goto out_writepage;
 		}
+		fio->need_lock = LOCK_REQ;
 	}
 
+	err = encrypt_one_page(fio);
+	if (err)
+		goto out_writepage;
+
 	set_page_writeback(page);
 
-	/*
-	 * If current allocation needs SSR,
-	 * it had better in-place writes for updated data.
-	 */
-	if (unlikely(fio->blk_addr != NEW_ADDR &&
-			!is_cold_data(page) &&
-			need_inplace_update(inode))) {
-		rewrite_data_page(fio);
-		set_inode_flag(F2FS_I(inode), FI_UPDATE_WRITE);
-		trace_f2fs_do_write_data_page(page, IPU);
-	} else {
-		write_data_page(&dn, fio);
-		set_data_blkaddr(&dn);
-		f2fs_update_extent_cache(&dn);
-		trace_f2fs_do_write_data_page(page, OPU);
-		set_inode_flag(F2FS_I(inode), FI_APPEND_WRITE);
-		if (page->index == 0)
-			set_inode_flag(F2FS_I(inode), FI_FIRST_BLOCK_WRITTEN);
-	}
+	/* LFS mode write path */
+	write_data_page(&dn, fio);
+	trace_f2fs_do_write_data_page(page, OPU);
+	set_inode_flag(inode, FI_APPEND_WRITE);
+	if (page->index == 0)
+		set_inode_flag(inode, FI_FIRST_BLOCK_WRITTEN);
 out_writepage:
 	f2fs_put_dnode(&dn);
+out:
+	if (fio->need_lock == LOCK_REQ)
+		f2fs_unlock_op(fio->sbi);
 	return err;
 }
 
-static int f2fs_write_data_page(struct page *page,
-					struct writeback_control *wbc)
+static int __write_data_page(struct page *page, bool *submitted,
+				struct writeback_control *wbc,
+				enum iostat_type io_type)
 {
 	struct inode *inode = page->mapping->host;
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	loff_t i_size = i_size_read(inode);
 	const pgoff_t end_index = ((unsigned long long) i_size)
-							>> PAGE_CACHE_SHIFT;
+							>> PAGE_SHIFT;
+	loff_t psize = (page->index + 1) << PAGE_SHIFT;
 	unsigned offset = 0;
 	bool need_balance_fs = false;
 	int err = 0;
 	struct f2fs_io_info fio = {
 		.sbi = sbi,
 		.type = DATA,
-		.rw = (wbc->sync_mode == WB_SYNC_ALL) ? WRITE_SYNC : WRITE,
+		.op = REQ_OP_WRITE,
+		.op_flags = wbc_to_write_flags(wbc),
+		.old_blkaddr = NULL_ADDR,
 		.page = page,
 		.encrypted_page = NULL,
+		.submitted = false,
+		.need_lock = LOCK_RETRY,
+		.io_type = io_type,
 	};
 
 	trace_f2fs_writepage(page, DATA);
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+		goto redirty_out;
+
 	if (page->index < end_index)
 		goto write;
 
@@ -1129,74 +1516,99 @@ static int f2fs_write_data_page(struct page *page,
 	 * If the offset is out-of-range of file size,
 	 * this page does not have to be written to disk.
 	 */
-	offset = i_size & (PAGE_CACHE_SIZE - 1);
+	offset = i_size & (PAGE_SIZE - 1);
 	if ((page->index >= end_index + 1) || !offset)
 		goto out;
 
-	zero_user_segment(page, offset, PAGE_CACHE_SIZE);
+	zero_user_segment(page, offset, PAGE_SIZE);
 write:
-	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
-		goto redirty_out;
 	if (f2fs_is_drop_cache(inode))
 		goto out;
-	if (f2fs_is_volatile_file(inode) && !wbc->for_reclaim &&
-			available_free_memory(sbi, BASE_CHECK))
+	/* we should not write 0'th page having journal header */
+	if (f2fs_is_volatile_file(inode) && (!page->index ||
+			(!wbc->for_reclaim &&
+			available_free_memory(sbi, BASE_CHECK))))
 		goto redirty_out;
 
+	/* we should bypass data pages to proceed the kworkder jobs */
+	if (unlikely(f2fs_cp_error(sbi))) {
+		mapping_set_error(page->mapping, -EIO);
+		goto out;
+	}
+
 	/* Dentry blocks are controlled by checkpoint */
 	if (S_ISDIR(inode->i_mode)) {
-		if (unlikely(f2fs_cp_error(sbi)))
-			goto redirty_out;
+		fio.need_lock = LOCK_DONE;
 		err = do_write_data_page(&fio);
 		goto done;
 	}
 
-	/* we should bypass data pages to proceed the kworkder jobs */
-	if (unlikely(f2fs_cp_error(sbi))) {
-		SetPageError(page);
-		goto out;
-	}
-
 	if (!wbc->for_reclaim)
 		need_balance_fs = true;
-	else if (has_not_enough_free_secs(sbi, 0))
+	else if (has_not_enough_free_secs(sbi, 0, 0))
 		goto redirty_out;
+	else
+		set_inode_flag(inode, FI_HOT_DATA);
 
 	err = -EAGAIN;
-	f2fs_lock_op(sbi);
-	if (f2fs_has_inline_data(inode))
+	if (f2fs_has_inline_data(inode)) {
 		err = f2fs_write_inline_data(inode, page);
-	if (err == -EAGAIN)
+		if (!err)
+			goto out;
+	}
+
+	if (err == -EAGAIN) {
 		err = do_write_data_page(&fio);
-	f2fs_unlock_op(sbi);
+		if (err == -EAGAIN) {
+			fio.need_lock = LOCK_REQ;
+			err = do_write_data_page(&fio);
+		}
+	}
+	if (F2FS_I(inode)->last_disk_size < psize)
+		F2FS_I(inode)->last_disk_size = psize;
+
 done:
 	if (err && err != -ENOENT)
 		goto redirty_out;
 
-	clear_cold_data(page);
 out:
 	inode_dec_dirty_pages(inode);
 	if (err)
 		ClearPageUptodate(page);
+
+	if (wbc->for_reclaim) {
+		f2fs_submit_merged_write_cond(sbi, inode, 0, page->index, DATA);
+		clear_inode_flag(inode, FI_HOT_DATA);
+		remove_dirty_inode(inode);
+		submitted = NULL;
+	}
+
 	unlock_page(page);
-	if (need_balance_fs)
-		f2fs_balance_fs(sbi);
-	if (wbc->for_reclaim)
-		f2fs_submit_merged_bio(sbi, DATA, WRITE);
+	if (!S_ISDIR(inode->i_mode))
+		f2fs_balance_fs(sbi, need_balance_fs);
+
+	if (unlikely(f2fs_cp_error(sbi))) {
+		f2fs_submit_merged_write(sbi, DATA);
+		submitted = NULL;
+	}
+
+	if (submitted)
+		*submitted = fio.submitted;
+
 	return 0;
 
 redirty_out:
 	redirty_page_for_writepage(wbc, page);
-	return AOP_WRITEPAGE_ACTIVATE;
+	if (!err)
+		return AOP_WRITEPAGE_ACTIVATE;
+	unlock_page(page);
+	return err;
 }
 
-static int __f2fs_writepage(struct page *page, struct writeback_control *wbc,
-			void *data)
+static int f2fs_write_data_page(struct page *page,
+					struct writeback_control *wbc)
 {
-	struct address_space *mapping = data;
-	int ret = mapping->a_ops->writepage(page, wbc);
-	mapping_set_error(mapping, ret);
-	return ret;
+	return __write_data_page(page, NULL, wbc, FS_DATA_IO);
 }
 
 /*
@@ -1205,8 +1617,8 @@ static int __f2fs_writepage(struct page *page, struct writeback_control *wbc,
  * warm/hot data page.
  */
 static int f2fs_write_cache_pages(struct address_space *mapping,
-			struct writeback_control *wbc, writepage_t writepage,
-			void *data)
+					struct writeback_control *wbc,
+					enum iostat_type io_type)
 {
 	int ret = 0;
 	int done = 0;
@@ -1216,13 +1628,19 @@ static int f2fs_write_cache_pages(struct address_space *mapping,
 	pgoff_t index;
 	pgoff_t end;		/* Inclusive */
 	pgoff_t done_index;
+	pgoff_t last_idx = ULONG_MAX;
 	int cycled;
 	int range_whole = 0;
 	int tag;
-	int step = 0;
 
 	pagevec_init(&pvec, 0);
-next:
+
+	if (get_dirty_pages(mapping->host) <=
+				SM_I(F2FS_M_SB(mapping))->min_hot_blocks)
+		set_inode_flag(mapping->host, FI_HOT_DATA);
+	else
+		clear_inode_flag(mapping->host, FI_HOT_DATA);
+
 	if (wbc->range_cyclic) {
 		writeback_index = mapping->writeback_index; /* prev offset */
 		index = writeback_index;
@@ -1232,8 +1650,8 @@ next:
 			cycled = 0;
 		end = -1;
 	} else {
-		index = wbc->range_start >> PAGE_CACHE_SHIFT;
-		end = wbc->range_end >> PAGE_CACHE_SHIFT;
+		index = wbc->range_start >> PAGE_SHIFT;
+		end = wbc->range_end >> PAGE_SHIFT;
 		if (wbc->range_start == 0 && wbc->range_end == LLONG_MAX)
 			range_whole = 1;
 		cycled = 1; /* ignore range_cyclic tests */
@@ -1256,6 +1674,7 @@ retry:
 
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
+			bool submitted = false;
 
 			if (page->index > end) {
 				done = 1;
@@ -1263,7 +1682,7 @@ retry:
 			}
 
 			done_index = page->index;
-
+retry_write:
 			lock_page(page);
 
 			if (unlikely(page->mapping != mapping)) {
@@ -1277,12 +1696,10 @@ continue_unlock:
 				goto continue_unlock;
 			}
 
-			if (step == is_cold_data(page))
-				goto continue_unlock;
-
 			if (PageWriteback(page)) {
 				if (wbc->sync_mode != WB_SYNC_NONE)
-					f2fs_wait_on_page_writeback(page, DATA);
+					f2fs_wait_on_page_writeback(page,
+								DATA, true);
 				else
 					goto continue_unlock;
 			}
@@ -1291,20 +1708,37 @@ continue_unlock:
 			if (!clear_page_dirty_for_io(page))
 				goto continue_unlock;
 
-			ret = (*writepage)(page, wbc, data);
+			ret = __write_data_page(page, &submitted, wbc, io_type);
 			if (unlikely(ret)) {
+				/*
+				 * keep nr_to_write, since vfs uses this to
+				 * get # of written pages.
+				 */
 				if (ret == AOP_WRITEPAGE_ACTIVATE) {
 					unlock_page(page);
 					ret = 0;
-				} else {
-					done_index = page->index + 1;
-					done = 1;
-					break;
+					continue;
+				} else if (ret == -EAGAIN) {
+					ret = 0;
+					if (wbc->sync_mode == WB_SYNC_ALL) {
+						cond_resched();
+						congestion_wait(BLK_RW_ASYNC,
+									HZ/50);
+						goto retry_write;
+					}
+					continue;
 				}
+				done_index = page->index + 1;
+				done = 1;
+				break;
+			} else if (submitted) {
+				last_idx = page->index;
 			}
 
-			if (--wbc->nr_to_write <= 0 &&
-			    wbc->sync_mode == WB_SYNC_NONE) {
+			/* give a priority to WB_SYNC threads */
+			if ((atomic_read(&F2FS_M_SB(mapping)->wb_sync_req) ||
+					--wbc->nr_to_write <= 0) &&
+					wbc->sync_mode == WB_SYNC_NONE) {
 				done = 1;
 				break;
 			}
@@ -1313,11 +1747,6 @@ continue_unlock:
 		cond_resched();
 	}
 
-	if (step < 1) {
-		step++;
-		goto next;
-	}
-
 	if (!cycled && !done) {
 		cycled = 1;
 		index = 0;
@@ -1327,19 +1756,21 @@ continue_unlock:
 	if (wbc->range_cyclic || (range_whole && wbc->nr_to_write > 0))
 		mapping->writeback_index = done_index;
 
+	if (last_idx != ULONG_MAX)
+		f2fs_submit_merged_write_cond(F2FS_M_SB(mapping), mapping->host,
+						0, last_idx, DATA);
+
 	return ret;
 }
 
-static int f2fs_write_data_pages(struct address_space *mapping,
-			    struct writeback_control *wbc)
+int __f2fs_write_data_pages(struct address_space *mapping,
+						struct writeback_control *wbc,
+						enum iostat_type io_type)
 {
 	struct inode *inode = mapping->host;
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	bool locked = false;
+	struct blk_plug plug;
 	int ret;
-	long diff;
-
-	trace_f2fs_writepages(mapping->host, wbc, DATA);
 
 	/* deal with chardevs and other special file */
 	if (!mapping->a_ops->writepage)
@@ -1349,46 +1780,147 @@ static int f2fs_write_data_pages(struct address_space *mapping,
 	if (!get_dirty_pages(inode) && wbc->sync_mode == WB_SYNC_NONE)
 		return 0;
 
+	/* during POR, we don't need to trigger writepage at all. */
+	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+		goto skip_write;
+
 	if (S_ISDIR(inode->i_mode) && wbc->sync_mode == WB_SYNC_NONE &&
 			get_dirty_pages(inode) < nr_pages_to_skip(sbi, DATA) &&
 			available_free_memory(sbi, DIRTY_DENTS))
 		goto skip_write;
 
-	/* during POR, we don't need to trigger writepage at all. */
-	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+	/* skip writing during file defragment */
+	if (is_inode_flag_set(inode, FI_DO_DEFRAG))
 		goto skip_write;
 
-	diff = nr_pages_to_write(sbi, DATA, wbc);
+	trace_f2fs_writepages(mapping->host, wbc, DATA);
 
-	if (!S_ISDIR(inode->i_mode)) {
-		mutex_lock(&sbi->writepages);
-		locked = true;
-	}
-	ret = f2fs_write_cache_pages(mapping, wbc, __f2fs_writepage, mapping);
-	f2fs_submit_merged_bio(sbi, DATA, WRITE);
-	if (locked)
-		mutex_unlock(&sbi->writepages);
+	/* to avoid spliting IOs due to mixed WB_SYNC_ALL and WB_SYNC_NONE */
+	if (wbc->sync_mode == WB_SYNC_ALL)
+		atomic_inc(&sbi->wb_sync_req);
+	else if (atomic_read(&sbi->wb_sync_req))
+		goto skip_write;
 
-	remove_dirty_dir_inode(inode);
+	blk_start_plug(&plug);
+	ret = f2fs_write_cache_pages(mapping, wbc, io_type);
+	blk_finish_plug(&plug);
+
+	if (wbc->sync_mode == WB_SYNC_ALL)
+		atomic_dec(&sbi->wb_sync_req);
+	/*
+	 * if some pages were truncated, we cannot guarantee its mapping->host
+	 * to detect pending bios.
+	 */
 
-	wbc->nr_to_write = max((long)0, wbc->nr_to_write - diff);
+	remove_dirty_inode(inode);
 	return ret;
 
 skip_write:
 	wbc->pages_skipped += get_dirty_pages(inode);
+	trace_f2fs_writepages(mapping->host, wbc, DATA);
 	return 0;
 }
 
+static int f2fs_write_data_pages(struct address_space *mapping,
+			    struct writeback_control *wbc)
+{
+	struct inode *inode = mapping->host;
+
+	return __f2fs_write_data_pages(mapping, wbc,
+			F2FS_I(inode)->cp_task == current ?
+			FS_CP_DATA_IO : FS_DATA_IO);
+}
+
 static void f2fs_write_failed(struct address_space *mapping, loff_t to)
 {
 	struct inode *inode = mapping->host;
+	loff_t i_size = i_size_read(inode);
 
-	if (to > inode->i_size) {
-		truncate_pagecache(inode, inode->i_size);
-		truncate_blocks(inode, inode->i_size, true);
+	if (to > i_size) {
+		down_write(&F2FS_I(inode)->i_mmap_sem);
+		truncate_pagecache(inode, i_size);
+		truncate_blocks(inode, i_size, true);
+		up_write(&F2FS_I(inode)->i_mmap_sem);
 	}
 }
 
+static int prepare_write_begin(struct f2fs_sb_info *sbi,
+			struct page *page, loff_t pos, unsigned len,
+			block_t *blk_addr, bool *node_changed)
+{
+	struct inode *inode = page->mapping->host;
+	pgoff_t index = page->index;
+	struct dnode_of_data dn;
+	struct page *ipage;
+	bool locked = false;
+	struct extent_info ei = {0,0,0};
+	int err = 0;
+
+	/*
+	 * we already allocated all the blocks, so we don't need to get
+	 * the block addresses when there is no need to fill the page.
+	 */
+	if (!f2fs_has_inline_data(inode) && len == PAGE_SIZE &&
+			!is_inode_flag_set(inode, FI_NO_PREALLOC))
+		return 0;
+
+	if (f2fs_has_inline_data(inode) ||
+			(pos & PAGE_MASK) >= i_size_read(inode)) {
+		__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, true);
+		locked = true;
+	}
+restart:
+	/* check inline_data */
+	ipage = get_node_page(sbi, inode->i_ino);
+	if (IS_ERR(ipage)) {
+		err = PTR_ERR(ipage);
+		goto unlock_out;
+	}
+
+	set_new_dnode(&dn, inode, ipage, ipage, 0);
+
+	if (f2fs_has_inline_data(inode)) {
+		if (pos + len <= MAX_INLINE_DATA(inode)) {
+			read_inline_data(page, ipage);
+			set_inode_flag(inode, FI_DATA_EXIST);
+			if (inode->i_nlink)
+				set_inline_node(ipage);
+		} else {
+			err = f2fs_convert_inline_page(&dn, page);
+			if (err)
+				goto out;
+			if (dn.data_blkaddr == NULL_ADDR)
+				err = f2fs_get_block(&dn, index);
+		}
+	} else if (locked) {
+		err = f2fs_get_block(&dn, index);
+	} else {
+		if (f2fs_lookup_extent_cache(inode, index, &ei)) {
+			dn.data_blkaddr = ei.blk + index - ei.fofs;
+		} else {
+			/* hole case */
+			err = get_dnode_of_data(&dn, index, LOOKUP_NODE);
+			if (err || dn.data_blkaddr == NULL_ADDR) {
+				f2fs_put_dnode(&dn);
+				__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO,
+								true);
+				locked = true;
+				goto restart;
+			}
+		}
+	}
+
+	/* convert_inline_page can make node_changed */
+	*blk_addr = dn.data_blkaddr;
+	*node_changed = dn.node_changed;
+out:
+	f2fs_put_dnode(&dn);
+unlock_out:
+	if (locked)
+		__do_map_lock(sbi, F2FS_GET_BLOCK_PRE_AIO, false);
+	return err;
+}
+
 static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 		loff_t pos, unsigned len, unsigned flags,
 		struct page **pagep, void **fsdata)
@@ -1396,14 +1928,22 @@ static int f2fs_write_begin(struct file *file, struct address_space *mapping,
 	struct inode *inode = mapping->host;
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct page *page = NULL;
-	struct page *ipage;
-	pgoff_t index = ((unsigned long long) pos) >> PAGE_CACHE_SHIFT;
-	struct dnode_of_data dn;
+	pgoff_t index = ((unsigned long long) pos) >> PAGE_SHIFT;
+	bool need_balance = false;
+	block_t blkaddr = NULL_ADDR;
 	int err = 0;
 
-	trace_f2fs_write_begin(inode, pos, len, flags);
+	if (trace_android_fs_datawrite_start_enabled()) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
 
-	f2fs_balance_fs(sbi);
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_datawrite_start(inode, pos, len,
+						 current->pid, path,
+						 current->comm);
+	}
+	trace_f2fs_write_begin(inode, pos, len, flags);
 
 	/*
 	 * We should check this at this moment to avoid deadlock on inode page
@@ -1420,8 +1960,7 @@ repeat:
 	 * Do not use grab_cache_page_write_begin() to avoid deadlock due to
 	 * wait_for_stable_page. Will wait that below with our IO control.
 	 */
-	page = pagecache_get_page(mapping, index,
-				FGP_LOCK | FGP_WRITE | FGP_CREAT, GFP_NOFS);
+	page = grab_cache_page(mapping, index);
 	if (!page) {
 		err = -ENOMEM;
 		goto fail;
@@ -1429,98 +1968,56 @@ repeat:
 
 	*pagep = page;
 
-	f2fs_lock_op(sbi);
-
-	/* check inline_data */
-	ipage = get_node_page(sbi, inode->i_ino);
-	if (IS_ERR(ipage)) {
-		err = PTR_ERR(ipage);
-		goto unlock_fail;
-	}
-
-	set_new_dnode(&dn, inode, ipage, ipage, 0);
+	err = prepare_write_begin(sbi, page, pos, len,
+					&blkaddr, &need_balance);
+	if (err)
+		goto fail;
 
-	if (f2fs_has_inline_data(inode)) {
-		if (pos + len <= MAX_INLINE_DATA) {
-			read_inline_data(page, ipage);
-			set_inode_flag(F2FS_I(inode), FI_DATA_EXIST);
-			sync_inode_page(&dn);
-			goto put_next;
+	if (need_balance && has_not_enough_free_secs(sbi, 0, 0)) {
+		unlock_page(page);
+		f2fs_balance_fs(sbi, true);
+		lock_page(page);
+		if (page->mapping != mapping) {
+			/* The page got truncated from under us */
+			f2fs_put_page(page, 1);
+			goto repeat;
 		}
-		err = f2fs_convert_inline_page(&dn, page);
-		if (err)
-			goto put_fail;
 	}
 
-	err = f2fs_get_block(&dn, index);
-	if (err)
-		goto put_fail;
-put_next:
-	f2fs_put_dnode(&dn);
-	f2fs_unlock_op(sbi);
-
-	f2fs_wait_on_page_writeback(page, DATA);
+	f2fs_wait_on_page_writeback(page, DATA, false);
 
 	/* wait for GCed encrypted page writeback */
-	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
-		f2fs_wait_on_encrypted_page_writeback(sbi, dn.data_blkaddr);
-
-	if (len == PAGE_CACHE_SIZE)
-		goto out_update;
-	if (PageUptodate(page))
-		goto out_clear;
+	if (f2fs_encrypted_file(inode))
+		f2fs_wait_on_block_writeback(sbi, blkaddr);
 
-	if ((pos & PAGE_CACHE_MASK) >= i_size_read(inode)) {
-		unsigned start = pos & (PAGE_CACHE_SIZE - 1);
-		unsigned end = start + len;
+	if (len == PAGE_SIZE || PageUptodate(page))
+		return 0;
 
-		/* Reading beyond i_size is simple: memset to zero */
-		zero_user_segments(page, 0, start, end, PAGE_CACHE_SIZE);
-		goto out_update;
+	if (!(pos & (PAGE_SIZE - 1)) && (pos + len) >= i_size_read(inode)) {
+		zero_user_segment(page, len, PAGE_SIZE);
+		return 0;
 	}
 
-	if (dn.data_blkaddr == NEW_ADDR) {
-		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
+	if (blkaddr == NEW_ADDR) {
+		zero_user_segment(page, 0, PAGE_SIZE);
+		SetPageUptodate(page);
 	} else {
-		struct f2fs_io_info fio = {
-			.sbi = sbi,
-			.type = DATA,
-			.rw = READ_SYNC,
-			.blk_addr = dn.data_blkaddr,
-			.page = page,
-			.encrypted_page = NULL,
-		};
-		err = f2fs_submit_page_bio(&fio);
+		err = f2fs_submit_page_read(inode, page, blkaddr);
 		if (err)
 			goto fail;
 
 		lock_page(page);
-		if (unlikely(!PageUptodate(page))) {
-			err = -EIO;
-			goto fail;
-		}
 		if (unlikely(page->mapping != mapping)) {
 			f2fs_put_page(page, 1);
 			goto repeat;
 		}
-
-		/* avoid symlink page */
-		if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode)) {
-			err = f2fs_decrypt_one(inode, page);
-			if (err)
-				goto fail;
+		if (unlikely(!PageUptodate(page))) {
+			err = -EIO;
+			goto fail;
 		}
 	}
-out_update:
-	SetPageUptodate(page);
-out_clear:
-	clear_cold_data(page);
 	return 0;
 
-put_fail:
-	f2fs_put_dnode(&dn);
-unlock_fail:
-	f2fs_unlock_op(sbi);
 fail:
 	f2fs_put_page(page, 1);
 	f2fs_write_failed(mapping, pos + len);
@@ -1534,17 +2031,30 @@ static int f2fs_write_end(struct file *file,
 {
 	struct inode *inode = page->mapping->host;
 
+	trace_android_fs_datawrite_end(inode, pos, len);
 	trace_f2fs_write_end(inode, pos, len, copied);
 
-	set_page_dirty(page);
-
-	if (pos + copied > i_size_read(inode)) {
-		i_size_write(inode, pos + copied);
-		mark_inode_dirty(inode);
-		update_inode_page(inode);
+	/*
+	 * This should be come from len == PAGE_SIZE, and we expect copied
+	 * should be PAGE_SIZE. Otherwise, we treat it with zero copied and
+	 * let generic_perform_write() try to copy data again through copied=0.
+	 */
+	if (!PageUptodate(page)) {
+		if (unlikely(copied != len))
+			copied = 0;
+		else
+			SetPageUptodate(page);
 	}
+	if (!copied)
+		goto unlock_out;
+
+	set_page_dirty(page);
 
+	if (pos + copied > i_size_read(inode))
+		f2fs_i_size_write(inode, pos + copied);
+unlock_out:
 	f2fs_put_page(page, 1);
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
 	return copied;
 }
 
@@ -1563,44 +2073,67 @@ static int check_direct_IO(struct inode *inode, struct iov_iter *iter,
 }
 
 static ssize_t f2fs_direct_IO(struct kiocb *iocb, struct iov_iter *iter,
-			      loff_t offset)
+				loff_t offset)
 {
-	struct file *file = iocb->ki_filp;
-	struct address_space *mapping = file->f_mapping;
+	struct address_space *mapping = iocb->ki_filp->f_mapping;
 	struct inode *inode = mapping->host;
 	size_t count = iov_iter_count(iter);
+	int rw = iov_iter_rw(iter);
 	int err;
 
-	/* we don't need to use inline_data strictly */
-	if (f2fs_has_inline_data(inode)) {
-		err = f2fs_convert_inline_inode(inode);
-		if (err)
-			return err;
-	}
-
-	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
-		return 0;
-
 	err = check_direct_IO(inode, iter, offset);
 	if (err)
 		return err;
 
-	trace_f2fs_direct_IO_enter(inode, offset, count, iov_iter_rw(iter));
+	if (__force_buffered_io(inode, rw))
+		return 0;
 
-	if (iov_iter_rw(iter) == WRITE) {
-		__allocate_data_blocks(inode, offset, count);
-		if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) {
-			err = -EIO;
-			goto out;
-		}
+	if (trace_android_fs_dataread_start_enabled() &&
+	    (iov_iter_rw(iter) == READ)) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_dataread_start(inode, offset,
+						count, current->pid, path,
+						current->comm);
+	}
+	if (trace_android_fs_datawrite_start_enabled() &&
+	    (iov_iter_rw(iter) == WRITE)) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_datawrite_start(inode, offset, count,
+						 current->pid, path,
+						 current->comm);
 	}
+	trace_f2fs_direct_IO_enter(inode, offset, count, rw);
 
+	down_read(&F2FS_I(inode)->dio_rwsem[rw]);
 	err = blockdev_direct_IO(iocb, inode, iter, offset, get_data_block_dio);
-out:
-	if (err < 0 && iov_iter_rw(iter) == WRITE)
-		f2fs_write_failed(mapping, offset + count);
+	up_read(&F2FS_I(inode)->dio_rwsem[rw]);
+
+	if (rw == WRITE) {
+		if (err > 0) {
+			f2fs_update_iostat(F2FS_I_SB(inode), APP_DIRECT_IO,
+									err);
+			set_inode_flag(inode, FI_UPDATE_WRITE);
+		} else if (err < 0) {
+			f2fs_write_failed(mapping, offset + count);
+		}
+	}
+
+	if (trace_android_fs_dataread_start_enabled() &&
+	    (iov_iter_rw(iter) == READ))
+		trace_android_fs_dataread_end(inode, offset, count);
+	if (trace_android_fs_datawrite_start_enabled() &&
+	    (iov_iter_rw(iter) == WRITE))
+		trace_android_fs_datawrite_end(inode, offset, count);
 
-	trace_f2fs_direct_IO_exit(inode, offset, count, iov_iter_rw(iter), err);
+	trace_f2fs_direct_IO_exit(inode, offset, count, rw, err);
 
 	return err;
 }
@@ -1612,22 +2145,25 @@ void f2fs_invalidate_page(struct page *page, unsigned int offset,
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 
 	if (inode->i_ino >= F2FS_ROOT_INO(sbi) &&
-		(offset % PAGE_CACHE_SIZE || length != PAGE_CACHE_SIZE))
+		(offset % PAGE_SIZE || length != PAGE_SIZE))
 		return;
 
 	if (PageDirty(page)) {
-		if (inode->i_ino == F2FS_META_INO(sbi))
+		if (inode->i_ino == F2FS_META_INO(sbi)) {
 			dec_page_count(sbi, F2FS_DIRTY_META);
-		else if (inode->i_ino == F2FS_NODE_INO(sbi))
+		} else if (inode->i_ino == F2FS_NODE_INO(sbi)) {
 			dec_page_count(sbi, F2FS_DIRTY_NODES);
-		else
+		} else {
 			inode_dec_dirty_pages(inode);
+			remove_dirty_inode(inode);
+		}
 	}
 
 	/* This is atomic written page, keep Private */
 	if (IS_ATOMIC_WRITTEN_PAGE(page))
-		return;
+		return drop_inmem_page(inode, page);
 
+	set_page_private(page, 0);
 	ClearPagePrivate(page);
 }
 
@@ -1641,10 +2177,42 @@ int f2fs_release_page(struct page *page, gfp_t wait)
 	if (IS_ATOMIC_WRITTEN_PAGE(page))
 		return 0;
 
+	set_page_private(page, 0);
 	ClearPagePrivate(page);
 	return 1;
 }
 
+/*
+ * This was copied from __set_page_dirty_buffers which gives higher performance
+ * in very high speed storages. (e.g., pmem)
+ */
+void f2fs_set_page_dirty_nobuffers(struct page *page)
+{
+	struct address_space *mapping = page->mapping;
+	struct mem_cgroup *memcg;
+	unsigned long flags;
+
+	if (unlikely(!mapping))
+		return;
+
+	spin_lock(&mapping->private_lock);
+	memcg = mem_cgroup_begin_page_stat(page);
+	SetPageDirty(page);
+	spin_unlock(&mapping->private_lock);
+
+	spin_lock_irqsave(&mapping->tree_lock, flags);
+	WARN_ON_ONCE(!PageUptodate(page));
+	account_page_dirtied(page, mapping, memcg);
+	radix_tree_tag_set(&mapping->page_tree,
+			page_index(page), PAGECACHE_TAG_DIRTY);
+	spin_unlock_irqrestore(&mapping->tree_lock, flags);
+
+	mem_cgroup_end_page_stat(memcg);
+
+	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+	return;
+}
+
 static int f2fs_set_data_page_dirty(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
@@ -1652,9 +2220,10 @@ static int f2fs_set_data_page_dirty(struct page *page)
 
 	trace_f2fs_set_page_dirty(page, DATA);
 
-	SetPageUptodate(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
 
-	if (f2fs_is_atomic_file(inode)) {
+	if (f2fs_is_atomic_file(inode) && !f2fs_is_commit_atomic_write(inode)) {
 		if (!IS_ATOMIC_WRITTEN_PAGE(page)) {
 			register_inmem_page(inode, page);
 			return 1;
@@ -1667,7 +2236,7 @@ static int f2fs_set_data_page_dirty(struct page *page)
 	}
 
 	if (!PageDirty(page)) {
-		__set_page_dirty_nobuffers(page);
+		f2fs_set_page_dirty_nobuffers(page);
 		update_dirty_page(inode, page);
 		return 1;
 	}
@@ -1688,6 +2257,62 @@ static sector_t f2fs_bmap(struct address_space *mapping, sector_t block)
 	return generic_block_bmap(mapping, block, get_data_block_bmap);
 }
 
+#ifdef CONFIG_MIGRATION
+#include <linux/migrate.h>
+
+int f2fs_migrate_page(struct address_space *mapping,
+		struct page *newpage, struct page *page, enum migrate_mode mode)
+{
+	int rc, extra_count;
+	struct f2fs_inode_info *fi = F2FS_I(mapping->host);
+	bool atomic_written = IS_ATOMIC_WRITTEN_PAGE(page);
+
+	BUG_ON(PageWriteback(page));
+
+	/* migrating an atomic written page is safe with the inmem_lock hold */
+	if (atomic_written) {
+		if (mode != MIGRATE_SYNC)
+			return -EBUSY;
+		if (!mutex_trylock(&fi->inmem_lock))
+			return -EAGAIN;
+	}
+
+	/*
+	 * A reference is expected if PagePrivate set when move mapping,
+	 * however F2FS breaks this for maintaining dirty page counts when
+	 * truncating pages. So here adjusting the 'extra_count' make it work.
+	 */
+	extra_count = (atomic_written ? 1 : 0) - page_has_private(page);
+	rc = migrate_page_move_mapping(mapping, newpage,
+				page, NULL, mode, extra_count);
+	if (rc != MIGRATEPAGE_SUCCESS) {
+		if (atomic_written)
+			mutex_unlock(&fi->inmem_lock);
+		return rc;
+	}
+
+	if (atomic_written) {
+		struct inmem_pages *cur;
+		list_for_each_entry(cur, &fi->inmem_pages, list)
+			if (cur->page == page) {
+				cur->page = newpage;
+				break;
+			}
+		mutex_unlock(&fi->inmem_lock);
+		put_page(page);
+		get_page(newpage);
+	}
+
+	if (PagePrivate(page))
+		SetPagePrivate(newpage);
+	set_page_private(newpage, page_private(page));
+
+	migrate_page_copy(newpage, page);
+
+	return MIGRATEPAGE_SUCCESS;
+}
+#endif
+
 const struct address_space_operations f2fs_dblock_aops = {
 	.readpage	= f2fs_read_data_page,
 	.readpages	= f2fs_read_data_pages,
@@ -1700,4 +2325,7 @@ const struct address_space_operations f2fs_dblock_aops = {
 	.releasepage	= f2fs_release_page,
 	.direct_IO	= f2fs_direct_IO,
 	.bmap		= f2fs_bmap,
+#ifdef CONFIG_MIGRATION
+	.migratepage    = f2fs_migrate_page,
+#endif
 };
diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c
index 24d6a51b48d1..87f449845f5f 100644
--- a/fs/f2fs/debug.c
+++ b/fs/f2fs/debug.c
@@ -38,23 +38,52 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->hit_rbtree = atomic64_read(&sbi->read_hit_rbtree);
 	si->hit_total = si->hit_largest + si->hit_cached + si->hit_rbtree;
 	si->total_ext = atomic64_read(&sbi->total_hit_ext);
-	si->ext_tree = sbi->total_ext_tree;
+	si->ext_tree = atomic_read(&sbi->total_ext_tree);
+	si->zombie_tree = atomic_read(&sbi->total_zombie_tree);
 	si->ext_node = atomic_read(&sbi->total_ext_node);
 	si->ndirty_node = get_pages(sbi, F2FS_DIRTY_NODES);
 	si->ndirty_dent = get_pages(sbi, F2FS_DIRTY_DENTS);
-	si->ndirty_dirs = sbi->n_dirty_dirs;
 	si->ndirty_meta = get_pages(sbi, F2FS_DIRTY_META);
+	si->ndirty_data = get_pages(sbi, F2FS_DIRTY_DATA);
+	si->ndirty_imeta = get_pages(sbi, F2FS_DIRTY_IMETA);
+	si->ndirty_dirs = sbi->ndirty_inode[DIR_INODE];
+	si->ndirty_files = sbi->ndirty_inode[FILE_INODE];
+	si->ndirty_all = sbi->ndirty_inode[DIRTY_META];
 	si->inmem_pages = get_pages(sbi, F2FS_INMEM_PAGES);
-	si->wb_pages = get_pages(sbi, F2FS_WRITEBACK);
+	si->aw_cnt = atomic_read(&sbi->aw_cnt);
+	si->vw_cnt = atomic_read(&sbi->vw_cnt);
+	si->max_aw_cnt = atomic_read(&sbi->max_aw_cnt);
+	si->max_vw_cnt = atomic_read(&sbi->max_vw_cnt);
+	si->nr_wb_cp_data = get_pages(sbi, F2FS_WB_CP_DATA);
+	si->nr_wb_data = get_pages(sbi, F2FS_WB_DATA);
+	if (SM_I(sbi) && SM_I(sbi)->fcc_info) {
+		si->nr_flushed =
+			atomic_read(&SM_I(sbi)->fcc_info->issued_flush);
+		si->nr_flushing =
+			atomic_read(&SM_I(sbi)->fcc_info->issing_flush);
+	}
+	if (SM_I(sbi) && SM_I(sbi)->dcc_info) {
+		si->nr_discarded =
+			atomic_read(&SM_I(sbi)->dcc_info->issued_discard);
+		si->nr_discarding =
+			atomic_read(&SM_I(sbi)->dcc_info->issing_discard);
+		si->nr_discard_cmd =
+			atomic_read(&SM_I(sbi)->dcc_info->discard_cmd_cnt);
+		si->undiscard_blks = SM_I(sbi)->dcc_info->undiscard_blks;
+	}
 	si->total_count = (int)sbi->user_block_count / sbi->blocks_per_seg;
 	si->rsvd_segs = reserved_segments(sbi);
 	si->overp_segs = overprovision_segments(sbi);
 	si->valid_count = valid_user_blocks(sbi);
+	si->discard_blks = discard_blocks(sbi);
 	si->valid_node_count = valid_node_count(sbi);
 	si->valid_inode_count = valid_inode_count(sbi);
 	si->inline_xattr = atomic_read(&sbi->inline_xattr);
 	si->inline_inode = atomic_read(&sbi->inline_inode);
 	si->inline_dir = atomic_read(&sbi->inline_dir);
+	si->append = sbi->im[APPEND_INO].ino_num;
+	si->update = sbi->im[UPDATE_INO].ino_num;
+	si->orphans = sbi->im[ORPHAN_INO].ino_num;
 	si->utilization = utilization(sbi);
 
 	si->free_segs = free_segments(sbi);
@@ -67,7 +96,9 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	si->dirty_nats = NM_I(sbi)->dirty_nat_cnt;
 	si->sits = MAIN_SEGS(sbi);
 	si->dirty_sits = SIT_I(sbi)->dirty_sentries;
-	si->fnids = NM_I(sbi)->fcnt;
+	si->free_nids = NM_I(sbi)->nid_cnt[FREE_NID_LIST];
+	si->avail_nids = NM_I(sbi)->available_nids;
+	si->alloc_nids = NM_I(sbi)->nid_cnt[ALLOC_NID_LIST];
 	si->bg_gc = sbi->bg_gc;
 	si->util_free = (int)(free_user_blocks(sbi) >> sbi->log_blocks_per_seg)
 		* 100 / (int)(sbi->user_block_count >> sbi->log_blocks_per_seg)
@@ -80,8 +111,8 @@ static void update_general_status(struct f2fs_sb_info *sbi)
 	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_NODE; i++) {
 		struct curseg_info *curseg = CURSEG_I(sbi, i);
 		si->curseg[i] = curseg->segno;
-		si->cursec[i] = curseg->segno / sbi->segs_per_sec;
-		si->curzone[i] = si->cursec[i] / sbi->secs_per_zone;
+		si->cursec[i] = GET_SEC_FROM_SEG(sbi, curseg->segno);
+		si->curzone[i] = GET_ZONE_FROM_SEC(sbi, si->cursec[i]);
 	}
 
 	for (i = 0; i < 2; i++) {
@@ -105,10 +136,10 @@ static void update_sit_info(struct f2fs_sb_info *sbi)
 
 	bimodal = 0;
 	total_vblocks = 0;
-	blks_per_sec = sbi->segs_per_sec * (1 << sbi->log_blocks_per_seg);
+	blks_per_sec = BLKS_PER_SEC(sbi);
 	hblks_per_sec = blks_per_sec / 2;
 	for (segno = 0; segno < MAIN_SEGS(sbi); segno += sbi->segs_per_sec) {
-		vblocks = get_valid_blocks(sbi, segno, sbi->segs_per_sec);
+		vblocks = get_valid_blocks(sbi, segno, true);
 		dist = abs(vblocks - hblks_per_sec);
 		bimodal += dist * dist;
 
@@ -137,9 +168,14 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	if (si->base_mem)
 		goto get_cache;
 
-	si->base_mem = sizeof(struct f2fs_sb_info) + sbi->sb->s_blocksize;
+	/* build stat */
+	si->base_mem = sizeof(struct f2fs_stat_info);
+
+	/* build superblock */
+	si->base_mem += sizeof(struct f2fs_sb_info) + sbi->sb->s_blocksize;
 	si->base_mem += 2 * sizeof(struct f2fs_inode_info);
 	si->base_mem += sizeof(*sbi->ckpt);
+	si->base_mem += sizeof(struct percpu_counter) * NR_COUNT_TYPE;
 
 	/* build sm */
 	si->base_mem += sizeof(struct f2fs_sm_info);
@@ -148,7 +184,9 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	si->base_mem += sizeof(struct sit_info);
 	si->base_mem += MAIN_SEGS(sbi) * sizeof(struct seg_entry);
 	si->base_mem += f2fs_bitmap_size(MAIN_SEGS(sbi));
-	si->base_mem += 3 * SIT_VBLOCK_MAP_SIZE * MAIN_SEGS(sbi);
+	si->base_mem += 2 * SIT_VBLOCK_MAP_SIZE * MAIN_SEGS(sbi);
+	if (f2fs_discard_en(sbi))
+		si->base_mem += SIT_VBLOCK_MAP_SIZE * MAIN_SEGS(sbi);
 	si->base_mem += SIT_VBLOCK_MAP_SIZE;
 	if (sbi->segs_per_sec > 1)
 		si->base_mem += MAIN_SECS(sbi) * sizeof(struct sec_entry);
@@ -161,7 +199,7 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 
 	/* build curseg */
 	si->base_mem += sizeof(struct curseg_info) * NR_CURSEG_TYPE;
-	si->base_mem += PAGE_CACHE_SIZE * NR_CURSEG_TYPE;
+	si->base_mem += PAGE_SIZE * NR_CURSEG_TYPE;
 
 	/* build dirty segmap */
 	si->base_mem += sizeof(struct dirty_seglist_info);
@@ -171,6 +209,10 @@ static void update_mem_info(struct f2fs_sb_info *sbi)
 	/* build nm */
 	si->base_mem += sizeof(struct f2fs_nm_info);
 	si->base_mem += __bitmap_size(sbi, NAT_BITMAP);
+	si->base_mem += (NM_I(sbi)->nat_bits_blocks << F2FS_BLKSIZE_BITS);
+	si->base_mem += NM_I(sbi)->nat_blocks * NAT_ENTRY_BITMAP_SIZE;
+	si->base_mem += NM_I(sbi)->nat_blocks / 8;
+	si->base_mem += NM_I(sbi)->nat_blocks * sizeof(unsigned short);
 
 get_cache:
 	si->cache_mem = 0;
@@ -180,27 +222,34 @@ get_cache:
 		si->cache_mem += sizeof(struct f2fs_gc_kthread);
 
 	/* build merge flush thread */
-	if (SM_I(sbi)->cmd_control_info)
+	if (SM_I(sbi)->fcc_info)
 		si->cache_mem += sizeof(struct flush_cmd_control);
+	if (SM_I(sbi)->dcc_info) {
+		si->cache_mem += sizeof(struct discard_cmd_control);
+		si->cache_mem += sizeof(struct discard_cmd) *
+			atomic_read(&SM_I(sbi)->dcc_info->discard_cmd_cnt);
+	}
 
 	/* free nids */
-	si->cache_mem += NM_I(sbi)->fcnt * sizeof(struct free_nid);
+	si->cache_mem += (NM_I(sbi)->nid_cnt[FREE_NID_LIST] +
+				NM_I(sbi)->nid_cnt[ALLOC_NID_LIST]) *
+				sizeof(struct free_nid);
 	si->cache_mem += NM_I(sbi)->nat_cnt * sizeof(struct nat_entry);
 	si->cache_mem += NM_I(sbi)->dirty_nat_cnt *
 					sizeof(struct nat_entry_set);
 	si->cache_mem += si->inmem_pages * sizeof(struct inmem_pages);
-	si->cache_mem += sbi->n_dirty_dirs * sizeof(struct inode_entry);
-	for (i = 0; i <= UPDATE_INO; i++)
+	for (i = 0; i <= ORPHAN_INO; i++)
 		si->cache_mem += sbi->im[i].ino_num * sizeof(struct ino_entry);
-	si->cache_mem += sbi->total_ext_tree * sizeof(struct extent_tree);
+	si->cache_mem += atomic_read(&sbi->total_ext_tree) *
+						sizeof(struct extent_tree);
 	si->cache_mem += atomic_read(&sbi->total_ext_node) *
 						sizeof(struct extent_node);
 
 	si->page_mem = 0;
 	npages = NODE_MAPPING(sbi)->nrpages;
-	si->page_mem += (unsigned long long)npages << PAGE_CACHE_SHIFT;
+	si->page_mem += (unsigned long long)npages << PAGE_SHIFT;
 	npages = META_MAPPING(sbi)->nrpages;
-	si->page_mem += (unsigned long long)npages << PAGE_CACHE_SHIFT;
+	si->page_mem += (unsigned long long)npages << PAGE_SHIFT;
 }
 
 static int stat_show(struct seq_file *s, void *v)
@@ -211,20 +260,24 @@ static int stat_show(struct seq_file *s, void *v)
 
 	mutex_lock(&f2fs_stat_mutex);
 	list_for_each_entry(si, &f2fs_stat_list, stat_list) {
-		char devname[BDEVNAME_SIZE];
-
 		update_general_status(si->sbi);
 
-		seq_printf(s, "\n=====[ partition info(%s). #%d ]=====\n",
-			bdevname(si->sbi->sb->s_bdev, devname), i++);
+		seq_printf(s, "\n=====[ partition info(%pg). #%d, %s]=====\n",
+			si->sbi->sb->s_bdev, i++,
+			f2fs_readonly(si->sbi->sb) ? "RO": "RW");
 		seq_printf(s, "[SB: 1] [CP: 2] [SIT: %d] [NAT: %d] ",
 			   si->sit_area_segs, si->nat_area_segs);
 		seq_printf(s, "[SSA: %d] [MAIN: %d",
 			   si->ssa_area_segs, si->main_area_segs);
 		seq_printf(s, "(OverProv:%d Resv:%d)]\n\n",
 			   si->overp_segs, si->rsvd_segs);
-		seq_printf(s, "Utilization: %d%% (%d valid blocks)\n",
-			   si->utilization, si->valid_count);
+		if (test_opt(si->sbi, DISCARD))
+			seq_printf(s, "Utilization: %u%% (%u valid blocks, %u discard blocks)\n",
+				si->utilization, si->valid_count, si->discard_blks);
+		else
+			seq_printf(s, "Utilization: %u%% (%u valid blocks)\n",
+				si->utilization, si->valid_count);
+
 		seq_printf(s, "  - Node: %u (Inode: %u, ",
 			   si->valid_node_count, si->valid_inode_count);
 		seq_printf(s, "Other: %u)\n  - Data: %u\n",
@@ -236,6 +289,8 @@ static int stat_show(struct seq_file *s, void *v)
 			   si->inline_inode);
 		seq_printf(s, "  - Inline_dentry Inode: %u\n",
 			   si->inline_dir);
+		seq_printf(s, "  - Orphan/Append/Update Inode: %u, %u, %u\n",
+			   si->orphans, si->append, si->update);
 		seq_printf(s, "\nMain area: %d segs, %d secs %d zones\n",
 			   si->main_area_segs, si->main_area_sections,
 			   si->main_area_zones);
@@ -269,7 +324,8 @@ static int stat_show(struct seq_file *s, void *v)
 			   si->dirty_count);
 		seq_printf(s, "  - Prefree: %d\n  - Free: %d (%d)\n\n",
 			   si->prefree_count, si->free_segs, si->free_secs);
-		seq_printf(s, "CP calls: %d\n", si->cp_count);
+		seq_printf(s, "CP calls: %d (BG: %d)\n",
+				si->cp_count, si->bg_cp_count);
 		seq_printf(s, "GC calls: %d (BG: %d)\n",
 			   si->call_count, si->bg_gc);
 		seq_printf(s, "  - data segments : %d (%d)\n",
@@ -290,21 +346,33 @@ static int stat_show(struct seq_file *s, void *v)
 				!si->total_ext ? 0 :
 				div64_u64(si->hit_total * 100, si->total_ext),
 				si->hit_total, si->total_ext);
-		seq_printf(s, "  - Inner Struct Count: tree: %d, node: %d\n",
-				si->ext_tree, si->ext_node);
+		seq_printf(s, "  - Inner Struct Count: tree: %d(%d), node: %d\n",
+				si->ext_tree, si->zombie_tree, si->ext_node);
 		seq_puts(s, "\nBalancing F2FS Async:\n");
-		seq_printf(s, "  - inmem: %4d, wb: %4d\n",
-			   si->inmem_pages, si->wb_pages);
+		seq_printf(s, "  - IO (CP: %4d, Data: %4d, Flush: (%4d %4d), "
+			"Discard: (%4d %4d)) cmd: %4d undiscard:%4u\n",
+			   si->nr_wb_cp_data, si->nr_wb_data,
+			   si->nr_flushing, si->nr_flushed,
+			   si->nr_discarding, si->nr_discarded,
+			   si->nr_discard_cmd, si->undiscard_blks);
+		seq_printf(s, "  - inmem: %4d, atomic IO: %4d (Max. %4d), "
+			"volatile IO: %4d (Max. %4d)\n",
+			   si->inmem_pages, si->aw_cnt, si->max_aw_cnt,
+			   si->vw_cnt, si->max_vw_cnt);
 		seq_printf(s, "  - nodes: %4d in %4d\n",
 			   si->ndirty_node, si->node_pages);
-		seq_printf(s, "  - dents: %4d in dirs:%4d\n",
-			   si->ndirty_dent, si->ndirty_dirs);
+		seq_printf(s, "  - dents: %4d in dirs:%4d (%4d)\n",
+			   si->ndirty_dent, si->ndirty_dirs, si->ndirty_all);
+		seq_printf(s, "  - datas: %4d in files:%4d\n",
+			   si->ndirty_data, si->ndirty_files);
 		seq_printf(s, "  - meta: %4d in %4d\n",
 			   si->ndirty_meta, si->meta_pages);
+		seq_printf(s, "  - imeta: %4d\n",
+			   si->ndirty_imeta);
 		seq_printf(s, "  - NATs: %9d/%9d\n  - SITs: %9d/%9d\n",
 			   si->dirty_nats, si->nats, si->dirty_sits, si->sits);
-		seq_printf(s, "  - free_nids: %9d\n",
-			   si->fnids);
+		seq_printf(s, "  - free_nids: %9d/%9d\n  - alloc_nids: %9d\n",
+			   si->free_nids, si->avail_nids, si->alloc_nids);
 		seq_puts(s, "\nDistribution of User Blocks:");
 		seq_puts(s, " [ valid | invalid | free ]\n");
 		seq_puts(s, "  [");
@@ -389,6 +457,11 @@ int f2fs_build_stats(struct f2fs_sb_info *sbi)
 	atomic_set(&sbi->inline_dir, 0);
 	atomic_set(&sbi->inplace_count, 0);
 
+	atomic_set(&sbi->aw_cnt, 0);
+	atomic_set(&sbi->vw_cnt, 0);
+	atomic_set(&sbi->max_aw_cnt, 0);
+	atomic_set(&sbi->max_vw_cnt, 0);
+
 	mutex_lock(&f2fs_stat_mutex);
 	list_add_tail(&si->stat_list, &f2fs_stat_list);
 	mutex_unlock(&f2fs_stat_mutex);
@@ -407,20 +480,23 @@ void f2fs_destroy_stats(struct f2fs_sb_info *sbi)
 	kfree(si);
 }
 
-void __init f2fs_create_root_stats(void)
+int __init f2fs_create_root_stats(void)
 {
 	struct dentry *file;
 
 	f2fs_debugfs_root = debugfs_create_dir("f2fs", NULL);
 	if (!f2fs_debugfs_root)
-		return;
+		return -ENOMEM;
 
 	file = debugfs_create_file("status", S_IRUGO, f2fs_debugfs_root,
 			NULL, &stat_fops);
 	if (!file) {
 		debugfs_remove(f2fs_debugfs_root);
 		f2fs_debugfs_root = NULL;
+		return -ENOMEM;
 	}
+
+	return 0;
 }
 
 void f2fs_destroy_root_stats(void)
diff --git a/fs/f2fs/dir.c b/fs/f2fs/dir.c
index 60972a559685..4f2a8fedb313 100644
--- a/fs/f2fs/dir.c
+++ b/fs/f2fs/dir.c
@@ -17,8 +17,8 @@
 
 static unsigned long dir_blocks(struct inode *inode)
 {
-	return ((unsigned long long) (i_size_read(inode) + PAGE_CACHE_SIZE - 1))
-							>> PAGE_CACHE_SHIFT;
+	return ((unsigned long long) (i_size_read(inode) + PAGE_SIZE - 1))
+							>> PAGE_SHIFT;
 }
 
 static unsigned int dir_buckets(unsigned int level, int dir_level)
@@ -37,7 +37,7 @@ static unsigned int bucket_blocks(unsigned int level)
 		return 4;
 }
 
-unsigned char f2fs_filetype_table[F2FS_FT_MAX] = {
+static unsigned char f2fs_filetype_table[F2FS_FT_MAX] = {
 	[F2FS_FT_UNKNOWN]	= DT_UNKNOWN,
 	[F2FS_FT_REG_FILE]	= DT_REG,
 	[F2FS_FT_DIR]		= DT_DIR,
@@ -48,7 +48,6 @@ unsigned char f2fs_filetype_table[F2FS_FT_MAX] = {
 	[F2FS_FT_SYMLINK]	= DT_LNK,
 };
 
-#define S_SHIFT 12
 static unsigned char f2fs_type_by_mode[S_IFMT >> S_SHIFT] = {
 	[S_IFREG >> S_SHIFT]	= F2FS_FT_REG_FILE,
 	[S_IFDIR >> S_SHIFT]	= F2FS_FT_DIR,
@@ -64,6 +63,13 @@ void set_de_type(struct f2fs_dir_entry *de, umode_t mode)
 	de->file_type = f2fs_type_by_mode[(mode & S_IFMT) >> S_SHIFT];
 }
 
+unsigned char get_de_type(struct f2fs_dir_entry *de)
+{
+	if (de->file_type < F2FS_FT_MAX)
+		return f2fs_filetype_table[de->file_type];
+	return DT_UNKNOWN;
+}
+
 static unsigned long dir_block_index(unsigned int level,
 				int dir_level, unsigned int idx)
 {
@@ -77,7 +83,7 @@ static unsigned long dir_block_index(unsigned int level,
 }
 
 static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
-				struct f2fs_filename *fname,
+				struct fscrypt_name *fname,
 				f2fs_hash_t namehash,
 				int *max_slots,
 				struct page **res_page)
@@ -88,30 +94,23 @@ static struct f2fs_dir_entry *find_in_block(struct page *dentry_page,
 
 	dentry_blk = (struct f2fs_dentry_block *)kmap(dentry_page);
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 1);
+	make_dentry_ptr_block(NULL, &d, dentry_blk);
 	de = find_target_dentry(fname, namehash, max_slots, &d);
 	if (de)
 		*res_page = dentry_page;
 	else
 		kunmap(dentry_page);
 
-	/*
-	 * For the most part, it should be a bug when name_len is zero.
-	 * We stop here for figuring out where the bugs has occurred.
-	 */
-	f2fs_bug_on(F2FS_P_SB(dentry_page), d.max < 0);
 	return de;
 }
 
-struct f2fs_dir_entry *find_target_dentry(struct f2fs_filename *fname,
+struct f2fs_dir_entry *find_target_dentry(struct fscrypt_name *fname,
 			f2fs_hash_t namehash, int *max_slots,
 			struct f2fs_dentry_ptr *d)
 {
 	struct f2fs_dir_entry *de;
 	unsigned long bit_pos = 0;
 	int max_len = 0;
-	struct f2fs_str de_name = FSTR_INIT(NULL, 0);
-	struct f2fs_str *name = &fname->disk_name;
 
 	if (max_slots)
 		*max_slots = 0;
@@ -124,37 +123,20 @@ struct f2fs_dir_entry *find_target_dentry(struct f2fs_filename *fname,
 
 		de = &d->dentry[bit_pos];
 
-		if (de->hash_code != namehash)
-			goto not_match;
-
-		de_name.name = d->filename[bit_pos];
-		de_name.len = le16_to_cpu(de->name_len);
-
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-		if (unlikely(!name->name)) {
-			if (fname->usr_fname->name[0] == '_') {
-				if (de_name.len > 32 &&
-					!memcmp(de_name.name + ((de_name.len - 17) & ~15),
-						fname->crypto_buf.name + 8, 16))
-					goto found;
-				goto not_match;
-			}
-			name->name = fname->crypto_buf.name;
-			name->len = fname->crypto_buf.len;
+		if (unlikely(!de->name_len)) {
+			bit_pos++;
+			continue;
 		}
-#endif
-		if (de_name.len == name->len &&
-				!memcmp(de_name.name, name->name, name->len))
+
+		if (de->hash_code == namehash &&
+		    fscrypt_match_name(fname, d->filename[bit_pos],
+				       le16_to_cpu(de->name_len)))
 			goto found;
-not_match:
+
 		if (max_slots && max_len > *max_slots)
 			*max_slots = max_len;
 		max_len = 0;
 
-		/* remain bug on condition */
-		if (unlikely(!de->name_len))
-			d->max = -1;
-
 		bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len));
 	}
 
@@ -167,7 +149,7 @@ found:
 
 static struct f2fs_dir_entry *find_in_level(struct inode *dir,
 					unsigned int level,
-					struct f2fs_filename *fname,
+					struct fscrypt_name *fname,
 					struct page **res_page)
 {
 	struct qstr name = FSTR_TO_QSTR(&fname->disk_name);
@@ -178,11 +160,7 @@ static struct f2fs_dir_entry *find_in_level(struct inode *dir,
 	struct f2fs_dir_entry *de = NULL;
 	bool room = false;
 	int max_slots;
-	f2fs_hash_t namehash;
-
-	namehash = f2fs_dentry_hash(&name, fname);
-
-	f2fs_bug_on(F2FS_I_SB(dir), level > MAX_DIR_HASH_DEPTH);
+	f2fs_hash_t namehash = f2fs_dentry_hash(&name, fname);
 
 	nbucket = dir_buckets(level, F2FS_I(dir)->i_dir_level);
 	nblock = bucket_blocks(level);
@@ -195,8 +173,13 @@ static struct f2fs_dir_entry *find_in_level(struct inode *dir,
 		/* no need to allocate new dentry pages to all the indices */
 		dentry_page = find_data_page(dir, bidx);
 		if (IS_ERR(dentry_page)) {
-			room = true;
-			continue;
+			if (PTR_ERR(dentry_page) == -ENOENT) {
+				room = true;
+				continue;
+			} else {
+				*res_page = dentry_page;
+				break;
+			}
 		}
 
 		de = find_in_block(dentry_page, fname, namehash, &max_slots,
@@ -217,79 +200,93 @@ static struct f2fs_dir_entry *find_in_level(struct inode *dir,
 	return de;
 }
 
-/*
- * Find an entry in the specified directory with the wanted name.
- * It returns the page where the entry was found (as a parameter - res_page),
- * and the entry itself. Page is returned mapped and unlocked.
- * Entry is guaranteed to be valid.
- */
-struct f2fs_dir_entry *f2fs_find_entry(struct inode *dir,
-			struct qstr *child, struct page **res_page)
+struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir,
+			struct fscrypt_name *fname, struct page **res_page)
 {
 	unsigned long npages = dir_blocks(dir);
 	struct f2fs_dir_entry *de = NULL;
 	unsigned int max_depth;
 	unsigned int level;
-	struct f2fs_filename fname;
-	int err;
-
-	*res_page = NULL;
-
-	err = f2fs_fname_setup_filename(dir, child, 1, &fname);
-	if (err)
-		return NULL;
 
 	if (f2fs_has_inline_dentry(dir)) {
-		de = find_in_inline_dir(dir, &fname, res_page);
+		*res_page = NULL;
+		de = find_in_inline_dir(dir, fname, res_page);
 		goto out;
 	}
 
-	if (npages == 0)
+	if (npages == 0) {
+		*res_page = NULL;
 		goto out;
+	}
 
 	max_depth = F2FS_I(dir)->i_current_depth;
+	if (unlikely(max_depth > MAX_DIR_HASH_DEPTH)) {
+		f2fs_msg(F2FS_I_SB(dir)->sb, KERN_WARNING,
+				"Corrupted max_depth of %lu: %u",
+				dir->i_ino, max_depth);
+		max_depth = MAX_DIR_HASH_DEPTH;
+		f2fs_i_depth_write(dir, max_depth);
+	}
 
 	for (level = 0; level < max_depth; level++) {
-		de = find_in_level(dir, level, &fname, res_page);
-		if (de)
+		*res_page = NULL;
+		de = find_in_level(dir, level, fname, res_page);
+		if (de || IS_ERR(*res_page))
 			break;
 	}
 out:
-	f2fs_fname_free_filename(&fname);
+	/* This is to increase the speed of f2fs_create */
+	if (!de)
+		F2FS_I(dir)->task = current;
 	return de;
 }
 
-struct f2fs_dir_entry *f2fs_parent_dir(struct inode *dir, struct page **p)
+/*
+ * Find an entry in the specified directory with the wanted name.
+ * It returns the page where the entry was found (as a parameter - res_page),
+ * and the entry itself. Page is returned mapped and unlocked.
+ * Entry is guaranteed to be valid.
+ */
+struct f2fs_dir_entry *f2fs_find_entry(struct inode *dir,
+			const struct qstr *child, struct page **res_page)
 {
-	struct page *page;
-	struct f2fs_dir_entry *de;
-	struct f2fs_dentry_block *dentry_blk;
-
-	if (f2fs_has_inline_dentry(dir))
-		return f2fs_parent_inline_dir(dir, p);
+	struct f2fs_dir_entry *de = NULL;
+	struct fscrypt_name fname;
+	int err;
 
-	page = get_lock_data_page(dir, 0, false);
-	if (IS_ERR(page))
+	err = fscrypt_setup_filename(dir, child, 1, &fname);
+	if (err) {
+		if (err == -ENOENT)
+			*res_page = NULL;
+		else
+			*res_page = ERR_PTR(err);
 		return NULL;
+	}
+
+	de = __f2fs_find_entry(dir, &fname, res_page);
 
-	dentry_blk = kmap(page);
-	de = &dentry_blk->dentry[1];
-	*p = page;
-	unlock_page(page);
+	fscrypt_free_filename(&fname);
 	return de;
 }
 
-ino_t f2fs_inode_by_name(struct inode *dir, struct qstr *qstr)
+struct f2fs_dir_entry *f2fs_parent_dir(struct inode *dir, struct page **p)
+{
+	struct qstr dotdot = QSTR_INIT("..", 2);
+
+	return f2fs_find_entry(dir, &dotdot, p);
+}
+
+ino_t f2fs_inode_by_name(struct inode *dir, const struct qstr *qstr,
+							struct page **page)
 {
 	ino_t res = 0;
 	struct f2fs_dir_entry *de;
-	struct page *page;
 
-	de = f2fs_find_entry(dir, qstr, &page);
+	de = f2fs_find_entry(dir, qstr, page);
 	if (de) {
 		res = le32_to_cpu(de->ino);
-		f2fs_dentry_kunmap(dir, page);
-		f2fs_put_page(page, 0);
+		f2fs_dentry_kunmap(dir, *page);
+		f2fs_put_page(*page, 0);
 	}
 
 	return res;
@@ -300,14 +297,14 @@ void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de,
 {
 	enum page_type type = f2fs_has_inline_dentry(dir) ? NODE : DATA;
 	lock_page(page);
-	f2fs_wait_on_page_writeback(page, type);
+	f2fs_wait_on_page_writeback(page, type, true);
 	de->ino = cpu_to_le32(inode->i_ino);
 	set_de_type(de, inode->i_mode);
 	f2fs_dentry_kunmap(dir, page);
 	set_page_dirty(page);
-	dir->i_mtime = dir->i_ctime = CURRENT_TIME;
-	mark_inode_dirty(dir);
 
+	dir->i_mtime = dir->i_ctime = current_time(dir);
+	f2fs_mark_inode_dirty_sync(dir, false);
 	f2fs_put_page(page, 1);
 }
 
@@ -315,7 +312,7 @@ static void init_dent_inode(const struct qstr *name, struct page *ipage)
 {
 	struct f2fs_inode *ri;
 
-	f2fs_wait_on_page_writeback(ipage, NODE);
+	f2fs_wait_on_page_writeback(ipage, NODE, true);
 
 	/* copy name info. to this inode page */
 	ri = F2FS_INODE(ipage);
@@ -324,45 +321,17 @@ static void init_dent_inode(const struct qstr *name, struct page *ipage)
 	set_page_dirty(ipage);
 }
 
-int update_dent_inode(struct inode *inode, struct inode *to,
-					const struct qstr *name)
-{
-	struct page *page;
-
-	if (file_enc_name(to))
-		return 0;
-
-	page = get_node_page(F2FS_I_SB(inode), inode->i_ino);
-	if (IS_ERR(page))
-		return PTR_ERR(page);
-
-	init_dent_inode(name, page);
-	f2fs_put_page(page, 1);
-
-	return 0;
-}
-
 void do_make_empty_dir(struct inode *inode, struct inode *parent,
 					struct f2fs_dentry_ptr *d)
 {
-	struct f2fs_dir_entry *de;
-
-	de = &d->dentry[0];
-	de->name_len = cpu_to_le16(1);
-	de->hash_code = 0;
-	de->ino = cpu_to_le32(inode->i_ino);
-	memcpy(d->filename[0], ".", 1);
-	set_de_type(de, inode->i_mode);
+	struct qstr dot = QSTR_INIT(".", 1);
+	struct qstr dotdot = QSTR_INIT("..", 2);
 
-	de = &d->dentry[1];
-	de->hash_code = 0;
-	de->name_len = cpu_to_le16(2);
-	de->ino = cpu_to_le32(parent->i_ino);
-	memcpy(d->filename[1], "..", 2);
-	set_de_type(de, parent->i_mode);
+	/* update dirent of "." */
+	f2fs_update_dentry(inode->i_ino, inode->i_mode, d, &dot, 0, 0);
 
-	test_and_set_bit_le(0, (void *)d->bitmap);
-	test_and_set_bit_le(1, (void *)d->bitmap);
+	/* update dirent of ".." */
+	f2fs_update_dentry(parent->i_ino, parent->i_mode, d, &dotdot, 0, 1);
 }
 
 static int make_empty_dir(struct inode *inode,
@@ -381,7 +350,7 @@ static int make_empty_dir(struct inode *inode,
 
 	dentry_blk = kmap_atomic(dentry_page);
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 1);
+	make_dentry_ptr_block(NULL, &d, dentry_blk);
 	do_make_empty_dir(inode, parent, &d);
 
 	kunmap_atomic(dentry_blk);
@@ -392,32 +361,38 @@ static int make_empty_dir(struct inode *inode,
 }
 
 struct page *init_inode_metadata(struct inode *inode, struct inode *dir,
-			const struct qstr *name, struct page *dpage)
+			const struct qstr *new_name, const struct qstr *orig_name,
+			struct page *dpage)
 {
 	struct page *page;
 	int err;
 
-	if (is_inode_flag_set(F2FS_I(inode), FI_NEW_INODE)) {
+	if (is_inode_flag_set(inode, FI_NEW_INODE)) {
 		page = new_inode_page(inode);
 		if (IS_ERR(page))
 			return page;
 
 		if (S_ISDIR(inode->i_mode)) {
+			/* in order to handle error case */
+			get_page(page);
 			err = make_empty_dir(inode, dir, page);
-			if (err)
-				goto error;
+			if (err) {
+				lock_page(page);
+				goto put_error;
+			}
+			put_page(page);
 		}
 
 		err = f2fs_init_acl(inode, dir, page, dpage);
 		if (err)
 			goto put_error;
 
-		err = f2fs_init_security(inode, dir, name, page);
+		err = f2fs_init_security(inode, dir, orig_name, page);
 		if (err)
 			goto put_error;
 
 		if (f2fs_encrypted_inode(dir) && f2fs_may_encrypt(inode)) {
-			err = f2fs_inherit_context(dir, inode, page);
+			err = fscrypt_inherit_context(dir, inode, page, false);
 			if (err)
 				goto put_error;
 		}
@@ -429,56 +404,52 @@ struct page *init_inode_metadata(struct inode *inode, struct inode *dir,
 		set_cold_node(inode, page);
 	}
 
-	if (name)
-		init_dent_inode(name, page);
+	if (new_name) {
+		init_dent_inode(new_name, page);
+		if (f2fs_encrypted_inode(dir))
+			file_set_enc_name(inode);
+	}
 
 	/*
 	 * This file should be checkpointed during fsync.
 	 * We lost i_pino from now on.
 	 */
-	if (is_inode_flag_set(F2FS_I(inode), FI_INC_LINK)) {
-		file_lost_pino(inode);
+	if (is_inode_flag_set(inode, FI_INC_LINK)) {
+		if (!S_ISDIR(inode->i_mode))
+			file_lost_pino(inode);
 		/*
 		 * If link the tmpfile to alias through linkat path,
 		 * we should remove this inode from orphan list.
 		 */
 		if (inode->i_nlink == 0)
 			remove_orphan_inode(F2FS_I_SB(dir), inode->i_ino);
-		inc_nlink(inode);
+		f2fs_i_links_write(inode, true);
 	}
 	return page;
 
 put_error:
+	clear_nlink(inode);
+	update_inode(inode, page);
 	f2fs_put_page(page, 1);
-error:
-	/* once the failed inode becomes a bad inode, i_mode is S_IFREG */
-	truncate_inode_pages(&inode->i_data, 0);
-	truncate_blocks(inode, 0, false);
-	remove_dirty_dir_inode(inode);
-	remove_inode_page(inode);
 	return ERR_PTR(err);
 }
 
 void update_parent_metadata(struct inode *dir, struct inode *inode,
 						unsigned int current_depth)
 {
-	if (inode && is_inode_flag_set(F2FS_I(inode), FI_NEW_INODE)) {
-		if (S_ISDIR(inode->i_mode)) {
-			inc_nlink(dir);
-			set_inode_flag(F2FS_I(dir), FI_UPDATE_DIR);
-		}
-		clear_inode_flag(F2FS_I(inode), FI_NEW_INODE);
+	if (inode && is_inode_flag_set(inode, FI_NEW_INODE)) {
+		if (S_ISDIR(inode->i_mode))
+			f2fs_i_links_write(dir, true);
+		clear_inode_flag(inode, FI_NEW_INODE);
 	}
-	dir->i_mtime = dir->i_ctime = CURRENT_TIME;
-	mark_inode_dirty(dir);
+	dir->i_mtime = dir->i_ctime = current_time(dir);
+	f2fs_mark_inode_dirty_sync(dir, false);
 
-	if (F2FS_I(dir)->i_current_depth != current_depth) {
-		F2FS_I(dir)->i_current_depth = current_depth;
-		set_inode_flag(F2FS_I(dir), FI_UPDATE_DIR);
-	}
+	if (F2FS_I(dir)->i_current_depth != current_depth)
+		f2fs_i_depth_write(dir, current_depth);
 
-	if (inode && is_inode_flag_set(F2FS_I(inode), FI_INC_LINK))
-		clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
+	if (inode && is_inode_flag_set(inode, FI_INC_LINK))
+		clear_inode_flag(inode, FI_INC_LINK);
 }
 
 int room_for_filename(const void *bitmap, int slots, int max_slots)
@@ -515,15 +486,16 @@ void f2fs_update_dentry(nid_t ino, umode_t mode, struct f2fs_dentry_ptr *d,
 	memcpy(d->filename[bit_pos], name->name, name->len);
 	de->ino = cpu_to_le32(ino);
 	set_de_type(de, mode);
-	for (i = 0; i < slots; i++)
-		test_and_set_bit_le(bit_pos + i, (void *)d->bitmap);
+	for (i = 0; i < slots; i++) {
+		__set_bit_le(bit_pos + i, (void *)d->bitmap);
+		/* avoid wrong garbage data for readdir */
+		if (i)
+			(de + i)->name_len = 0;
+	}
 }
 
-/*
- * Caller should grab and release a rwsem by calling f2fs_lock_op() and
- * f2fs_unlock_op().
- */
-int __f2fs_add_link(struct inode *dir, const struct qstr *name,
+int f2fs_add_regular_entry(struct inode *dir, const struct qstr *new_name,
+				const struct qstr *orig_name,
 				struct inode *inode, nid_t ino, umode_t mode)
 {
 	unsigned int bit_pos;
@@ -536,28 +508,11 @@ int __f2fs_add_link(struct inode *dir, const struct qstr *name,
 	struct f2fs_dentry_block *dentry_blk = NULL;
 	struct f2fs_dentry_ptr d;
 	struct page *page = NULL;
-	struct f2fs_filename fname;
-	struct qstr new_name;
-	int slots, err;
-
-	err = f2fs_fname_setup_filename(dir, name, 0, &fname);
-	if (err)
-		return err;
-
-	new_name.name = fname_name(&fname);
-	new_name.len = fname_len(&fname);
-
-	if (f2fs_has_inline_dentry(dir)) {
-		err = f2fs_add_inline_entry(dir, &new_name, inode, ino, mode);
-		if (!err || err != -EAGAIN)
-			goto out;
-		else
-			err = 0;
-	}
+	int slots, err = 0;
 
 	level = 0;
-	slots = GET_DENTRY_SLOTS(new_name.len);
-	dentry_hash = f2fs_dentry_hash(&new_name, NULL);
+	slots = GET_DENTRY_SLOTS(new_name->len);
+	dentry_hash = f2fs_dentry_hash(new_name, NULL);
 
 	current_depth = F2FS_I(dir)->i_current_depth;
 	if (F2FS_I(dir)->chash == dentry_hash) {
@@ -566,10 +521,14 @@ int __f2fs_add_link(struct inode *dir, const struct qstr *name,
 	}
 
 start:
-	if (unlikely(current_depth == MAX_DIR_HASH_DEPTH)) {
-		err = -ENOSPC;
-		goto out;
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(F2FS_I_SB(dir), FAULT_DIR_DEPTH)) {
+		f2fs_show_injection_info(FAULT_DIR_DEPTH);
+		return -ENOSPC;
 	}
+#endif
+	if (unlikely(current_depth == MAX_DIR_HASH_DEPTH))
+		return -ENOSPC;
 
 	/* Increase the depth, if required */
 	if (level == current_depth)
@@ -583,10 +542,8 @@ start:
 
 	for (block = bidx; block <= (bidx + nblock - 1); block++) {
 		dentry_page = get_new_data_page(dir, NULL, block, true);
-		if (IS_ERR(dentry_page)) {
-			err = PTR_ERR(dentry_page);
-			goto out;
-		}
+		if (IS_ERR(dentry_page))
+			return PTR_ERR(dentry_page);
 
 		dentry_blk = kmap(dentry_page);
 		bit_pos = room_for_filename(&dentry_blk->dentry_bitmap,
@@ -602,28 +559,25 @@ start:
 	++level;
 	goto start;
 add_dentry:
-	f2fs_wait_on_page_writeback(dentry_page, DATA);
+	f2fs_wait_on_page_writeback(dentry_page, DATA, true);
 
 	if (inode) {
 		down_write(&F2FS_I(inode)->i_sem);
-		page = init_inode_metadata(inode, dir, &new_name, NULL);
+		page = init_inode_metadata(inode, dir, new_name,
+						orig_name, NULL);
 		if (IS_ERR(page)) {
 			err = PTR_ERR(page);
 			goto fail;
 		}
-		if (f2fs_encrypted_inode(dir))
-			file_set_enc_name(inode);
 	}
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 1);
-	f2fs_update_dentry(ino, mode, &d, &new_name, dentry_hash, bit_pos);
+	make_dentry_ptr_block(NULL, &d, dentry_blk);
+	f2fs_update_dentry(ino, mode, &d, new_name, dentry_hash, bit_pos);
 
 	set_page_dirty(dentry_page);
 
 	if (inode) {
-		/* we don't need to mark_inode_dirty now */
-		F2FS_I(inode)->i_pino = dir->i_ino;
-		update_inode(inode, page);
+		f2fs_i_pino_write(inode, dir->i_ino);
 		f2fs_put_page(page, 1);
 	}
 
@@ -632,14 +586,69 @@ fail:
 	if (inode)
 		up_write(&F2FS_I(inode)->i_sem);
 
-	if (is_inode_flag_set(F2FS_I(dir), FI_UPDATE_DIR)) {
-		update_inode_page(dir);
-		clear_inode_flag(F2FS_I(dir), FI_UPDATE_DIR);
-	}
 	kunmap(dentry_page);
 	f2fs_put_page(dentry_page, 1);
-out:
-	f2fs_fname_free_filename(&fname);
+
+	return err;
+}
+
+int __f2fs_do_add_link(struct inode *dir, struct fscrypt_name *fname,
+				struct inode *inode, nid_t ino, umode_t mode)
+{
+	struct qstr new_name;
+	int err = -EAGAIN;
+
+	new_name.name = fname_name(fname);
+	new_name.len = fname_len(fname);
+
+	if (f2fs_has_inline_dentry(dir))
+		err = f2fs_add_inline_entry(dir, &new_name, fname->usr_fname,
+							inode, ino, mode);
+	if (err == -EAGAIN)
+		err = f2fs_add_regular_entry(dir, &new_name, fname->usr_fname,
+							inode, ino, mode);
+
+	f2fs_update_time(F2FS_I_SB(dir), REQ_TIME);
+	return err;
+}
+
+/*
+ * Caller should grab and release a rwsem by calling f2fs_lock_op() and
+ * f2fs_unlock_op().
+ */
+int __f2fs_add_link(struct inode *dir, const struct qstr *name,
+				struct inode *inode, nid_t ino, umode_t mode)
+{
+	struct fscrypt_name fname;
+	struct page *page = NULL;
+	struct f2fs_dir_entry *de = NULL;
+	int err;
+
+	err = fscrypt_setup_filename(dir, name, 0, &fname);
+	if (err)
+		return err;
+
+	/*
+	 * An immature stakable filesystem shows a race condition between lookup
+	 * and create. If we have same task when doing lookup and create, it's
+	 * definitely fine as expected by VFS normally. Otherwise, let's just
+	 * verify on-disk dentry one more time, which guarantees filesystem
+	 * consistency more.
+	 */
+	if (current != F2FS_I(dir)->task) {
+		de = __f2fs_find_entry(dir, &fname, &page);
+		F2FS_I(dir)->task = NULL;
+	}
+	if (de) {
+		f2fs_dentry_kunmap(dir, page);
+		f2fs_put_page(page, 0);
+		err = -EEXIST;
+	} else if (IS_ERR(page)) {
+		err = PTR_ERR(page);
+	} else {
+		err = __f2fs_do_add_link(dir, &fname, inode, ino, mode);
+	}
+	fscrypt_free_filename(&fname);
 	return err;
 }
 
@@ -649,46 +658,39 @@ int f2fs_do_tmpfile(struct inode *inode, struct inode *dir)
 	int err = 0;
 
 	down_write(&F2FS_I(inode)->i_sem);
-	page = init_inode_metadata(inode, dir, NULL, NULL);
+	page = init_inode_metadata(inode, dir, NULL, NULL, NULL);
 	if (IS_ERR(page)) {
 		err = PTR_ERR(page);
 		goto fail;
 	}
-	/* we don't need to mark_inode_dirty now */
-	update_inode(inode, page);
 	f2fs_put_page(page, 1);
 
-	clear_inode_flag(F2FS_I(inode), FI_NEW_INODE);
+	clear_inode_flag(inode, FI_NEW_INODE);
 fail:
 	up_write(&F2FS_I(inode)->i_sem);
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
 	return err;
 }
 
-void f2fs_drop_nlink(struct inode *dir, struct inode *inode, struct page *page)
+void f2fs_drop_nlink(struct inode *dir, struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
 
 	down_write(&F2FS_I(inode)->i_sem);
 
-	if (S_ISDIR(inode->i_mode)) {
-		drop_nlink(dir);
-		if (page)
-			update_inode(dir, page);
-		else
-			update_inode_page(dir);
-	}
-	inode->i_ctime = CURRENT_TIME;
+	if (S_ISDIR(inode->i_mode))
+		f2fs_i_links_write(dir, false);
+	inode->i_ctime = current_time(inode);
 
-	drop_nlink(inode);
+	f2fs_i_links_write(inode, false);
 	if (S_ISDIR(inode->i_mode)) {
-		drop_nlink(inode);
-		i_size_write(inode, 0);
+		f2fs_i_links_write(inode, false);
+		f2fs_i_size_write(inode, 0);
 	}
 	up_write(&F2FS_I(inode)->i_sem);
-	update_inode_page(inode);
 
 	if (inode->i_nlink == 0)
-		add_orphan_inode(sbi, inode->i_ino);
+		add_orphan_inode(inode);
 	else
 		release_orphan_inode(sbi);
 }
@@ -703,18 +705,22 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
 	struct	f2fs_dentry_block *dentry_blk;
 	unsigned int bit_pos;
 	int slots = GET_DENTRY_SLOTS(le16_to_cpu(dentry->name_len));
+	struct address_space *mapping = page_mapping(page);
+	unsigned long flags;
 	int i;
 
+	f2fs_update_time(F2FS_I_SB(dir), REQ_TIME);
+
 	if (f2fs_has_inline_dentry(dir))
 		return f2fs_delete_inline_entry(dentry, page, dir, inode);
 
 	lock_page(page);
-	f2fs_wait_on_page_writeback(page, DATA);
+	f2fs_wait_on_page_writeback(page, DATA, true);
 
 	dentry_blk = page_address(page);
 	bit_pos = dentry - dentry_blk->dentry;
 	for (i = 0; i < slots; i++)
-		clear_bit_le(bit_pos + i, &dentry_blk->dentry_bitmap);
+		__clear_bit_le(bit_pos + i, &dentry_blk->dentry_bitmap);
 
 	/* Let's check and deallocate this dentry page */
 	bit_pos = find_next_bit_le(&dentry_blk->dentry_bitmap,
@@ -723,17 +729,24 @@ void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
 	kunmap(page); /* kunmap - pair of f2fs_find_entry */
 	set_page_dirty(page);
 
-	dir->i_ctime = dir->i_mtime = CURRENT_TIME;
+	dir->i_ctime = dir->i_mtime = current_time(dir);
+	f2fs_mark_inode_dirty_sync(dir, false);
 
 	if (inode)
-		f2fs_drop_nlink(dir, inode, NULL);
+		f2fs_drop_nlink(dir, inode);
 
 	if (bit_pos == NR_DENTRY_IN_BLOCK &&
 			!truncate_hole(dir, page->index, page->index + 1)) {
+		spin_lock_irqsave(&mapping->tree_lock, flags);
+		radix_tree_tag_clear(&mapping->page_tree, page_index(page),
+				     PAGECACHE_TAG_DIRTY);
+		spin_unlock_irqrestore(&mapping->tree_lock, flags);
+
 		clear_page_dirty_for_io(page);
 		ClearPagePrivate(page);
 		ClearPageUptodate(page);
 		inode_dec_dirty_pages(dir);
+		remove_dirty_inode(dir);
 	}
 	f2fs_put_page(page, 1);
 }
@@ -776,13 +789,13 @@ bool f2fs_empty_dir(struct inode *dir)
 	return true;
 }
 
-bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
-				unsigned int start_pos, struct f2fs_str *fstr)
+int f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
+			unsigned int start_pos, struct fscrypt_str *fstr)
 {
 	unsigned char d_type = DT_UNKNOWN;
 	unsigned int bit_pos;
 	struct f2fs_dir_entry *de = NULL;
-	struct f2fs_str de_name = FSTR_INIT(NULL, 0);
+	struct fscrypt_str de_name = FSTR_INIT(NULL, 0);
 
 	bit_pos = ((unsigned long)ctx->pos % d->max);
 
@@ -792,29 +805,26 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 			break;
 
 		de = &d->dentry[bit_pos];
-		if (de->file_type < F2FS_FT_MAX)
-			d_type = f2fs_filetype_table[de->file_type];
-		else
-			d_type = DT_UNKNOWN;
+		if (de->name_len == 0) {
+			bit_pos++;
+			ctx->pos = start_pos + bit_pos;
+			continue;
+		}
+
+		d_type = get_de_type(de);
 
 		de_name.name = d->filename[bit_pos];
 		de_name.len = le16_to_cpu(de->name_len);
 
 		if (f2fs_encrypted_inode(d->inode)) {
 			int save_len = fstr->len;
-			int ret;
+			int err;
 
-			de_name.name = kmalloc(de_name.len, GFP_NOFS);
-			if (!de_name.name)
-				return false;
-
-			memcpy(de_name.name, d->filename[bit_pos], de_name.len);
-
-			ret = f2fs_fname_disk_to_usr(d->inode, &de->hash_code,
-							&de_name, fstr);
-			kfree(de_name.name);
-			if (ret < 0)
-				return true;
+			err = fscrypt_fname_disk_to_usr(d->inode,
+						(u32)de->hash_code, 0,
+						&de_name, fstr);
+			if (err)
+				return err;
 
 			de_name = *fstr;
 			fstr->len = save_len;
@@ -822,12 +832,12 @@ bool f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
 
 		if (!dir_emit(ctx, de_name.name, de_name.len,
 					le32_to_cpu(de->ino), d_type))
-			return true;
+			return 1;
 
 		bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len));
 		ctx->pos = start_pos + bit_pos;
 	}
-	return false;
+	return 0;
 }
 
 static int f2fs_readdir(struct file *file, struct dir_context *ctx)
@@ -839,16 +849,15 @@ static int f2fs_readdir(struct file *file, struct dir_context *ctx)
 	struct file_ra_state *ra = &file->f_ra;
 	unsigned int n = ((unsigned long)ctx->pos / NR_DENTRY_IN_BLOCK);
 	struct f2fs_dentry_ptr d;
-	struct f2fs_str fstr = FSTR_INIT(NULL, 0);
+	struct fscrypt_str fstr = FSTR_INIT(NULL, 0);
 	int err = 0;
 
 	if (f2fs_encrypted_inode(inode)) {
-		err = f2fs_get_encryption_info(inode);
-		if (err)
+		err = fscrypt_get_encryption_info(inode);
+		if (err && err != -ENOKEY)
 			return err;
 
-		err = f2fs_fname_crypto_alloc_buffer(inode, F2FS_NAME_LEN,
-								&fstr);
+		err = fscrypt_fname_alloc_buffer(inode, F2FS_NAME_LEN, &fstr);
 		if (err < 0)
 			return err;
 	}
@@ -865,29 +874,42 @@ static int f2fs_readdir(struct file *file, struct dir_context *ctx)
 
 	for (; n < npages; n++) {
 		dentry_page = get_lock_data_page(inode, n, false);
-		if (IS_ERR(dentry_page))
-			continue;
+		if (IS_ERR(dentry_page)) {
+			err = PTR_ERR(dentry_page);
+			if (err == -ENOENT) {
+				err = 0;
+				continue;
+			} else {
+				goto out;
+			}
+		}
 
 		dentry_blk = kmap(dentry_page);
 
-		make_dentry_ptr(inode, &d, (void *)dentry_blk, 1);
+		make_dentry_ptr_block(inode, &d, dentry_blk);
 
-		if (f2fs_fill_dentries(ctx, &d, n * NR_DENTRY_IN_BLOCK, &fstr))
-			goto stop;
+		err = f2fs_fill_dentries(ctx, &d,
+				n * NR_DENTRY_IN_BLOCK, &fstr);
+		if (err) {
+			kunmap(dentry_page);
+			f2fs_put_page(dentry_page, 1);
+			break;
+		}
 
 		ctx->pos = (n + 1) * NR_DENTRY_IN_BLOCK;
 		kunmap(dentry_page);
 		f2fs_put_page(dentry_page, 1);
-		dentry_page = NULL;
-	}
-stop:
-	if (dentry_page && !IS_ERR(dentry_page)) {
-		kunmap(dentry_page);
-		f2fs_put_page(dentry_page, 1);
 	}
 out:
-	f2fs_fname_crypto_free_buffer(&fstr);
-	return err;
+	fscrypt_fname_free_buffer(&fstr);
+	return err < 0 ? err : 0;
+}
+
+static int f2fs_dir_open(struct inode *inode, struct file *filp)
+{
+	if (f2fs_encrypted_inode(inode))
+		return fscrypt_get_encryption_info(inode) ? -EACCES : 0;
+	return 0;
 }
 
 const struct file_operations f2fs_dir_operations = {
@@ -895,6 +917,7 @@ const struct file_operations f2fs_dir_operations = {
 	.read		= generic_read_dir,
 	.iterate	= f2fs_readdir,
 	.fsync		= f2fs_sync_file,
+	.open		= f2fs_dir_open,
 	.unlocked_ioctl	= f2fs_ioctl,
 #ifdef CONFIG_COMPAT
 	.compat_ioctl   = f2fs_compat_ioctl,
diff --git a/fs/f2fs/extent_cache.c b/fs/f2fs/extent_cache.c
index 7ddba812e11b..ff2352a0ed15 100644
--- a/fs/f2fs/extent_cache.c
+++ b/fs/f2fs/extent_cache.c
@@ -18,6 +18,179 @@
 #include "node.h"
 #include <trace/events/f2fs.h>
 
+static struct rb_entry *__lookup_rb_tree_fast(struct rb_entry *cached_re,
+							unsigned int ofs)
+{
+	if (cached_re) {
+		if (cached_re->ofs <= ofs &&
+				cached_re->ofs + cached_re->len > ofs) {
+			return cached_re;
+		}
+	}
+	return NULL;
+}
+
+static struct rb_entry *__lookup_rb_tree_slow(struct rb_root *root,
+							unsigned int ofs)
+{
+	struct rb_node *node = root->rb_node;
+	struct rb_entry *re;
+
+	while (node) {
+		re = rb_entry(node, struct rb_entry, rb_node);
+
+		if (ofs < re->ofs)
+			node = node->rb_left;
+		else if (ofs >= re->ofs + re->len)
+			node = node->rb_right;
+		else
+			return re;
+	}
+	return NULL;
+}
+
+struct rb_entry *__lookup_rb_tree(struct rb_root *root,
+				struct rb_entry *cached_re, unsigned int ofs)
+{
+	struct rb_entry *re;
+
+	re = __lookup_rb_tree_fast(cached_re, ofs);
+	if (!re)
+		return __lookup_rb_tree_slow(root, ofs);
+
+	return re;
+}
+
+struct rb_node **__lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi,
+				struct rb_root *root, struct rb_node **parent,
+				unsigned int ofs)
+{
+	struct rb_node **p = &root->rb_node;
+	struct rb_entry *re;
+
+	while (*p) {
+		*parent = *p;
+		re = rb_entry(*parent, struct rb_entry, rb_node);
+
+		if (ofs < re->ofs)
+			p = &(*p)->rb_left;
+		else if (ofs >= re->ofs + re->len)
+			p = &(*p)->rb_right;
+		else
+			f2fs_bug_on(sbi, 1);
+	}
+
+	return p;
+}
+
+/*
+ * lookup rb entry in position of @ofs in rb-tree,
+ * if hit, return the entry, otherwise, return NULL
+ * @prev_ex: extent before ofs
+ * @next_ex: extent after ofs
+ * @insert_p: insert point for new extent at ofs
+ * in order to simpfy the insertion after.
+ * tree must stay unchanged between lookup and insertion.
+ */
+struct rb_entry *__lookup_rb_tree_ret(struct rb_root *root,
+				struct rb_entry *cached_re,
+				unsigned int ofs,
+				struct rb_entry **prev_entry,
+				struct rb_entry **next_entry,
+				struct rb_node ***insert_p,
+				struct rb_node **insert_parent,
+				bool force)
+{
+	struct rb_node **pnode = &root->rb_node;
+	struct rb_node *parent = NULL, *tmp_node;
+	struct rb_entry *re = cached_re;
+
+	*insert_p = NULL;
+	*insert_parent = NULL;
+	*prev_entry = NULL;
+	*next_entry = NULL;
+
+	if (RB_EMPTY_ROOT(root))
+		return NULL;
+
+	if (re) {
+		if (re->ofs <= ofs && re->ofs + re->len > ofs)
+			goto lookup_neighbors;
+	}
+
+	while (*pnode) {
+		parent = *pnode;
+		re = rb_entry(*pnode, struct rb_entry, rb_node);
+
+		if (ofs < re->ofs)
+			pnode = &(*pnode)->rb_left;
+		else if (ofs >= re->ofs + re->len)
+			pnode = &(*pnode)->rb_right;
+		else
+			goto lookup_neighbors;
+	}
+
+	*insert_p = pnode;
+	*insert_parent = parent;
+
+	re = rb_entry(parent, struct rb_entry, rb_node);
+	tmp_node = parent;
+	if (parent && ofs > re->ofs)
+		tmp_node = rb_next(parent);
+	*next_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+
+	tmp_node = parent;
+	if (parent && ofs < re->ofs)
+		tmp_node = rb_prev(parent);
+	*prev_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+	return NULL;
+
+lookup_neighbors:
+	if (ofs == re->ofs || force) {
+		/* lookup prev node for merging backward later */
+		tmp_node = rb_prev(&re->rb_node);
+		*prev_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+	}
+	if (ofs == re->ofs + re->len - 1 || force) {
+		/* lookup next node for merging frontward later */
+		tmp_node = rb_next(&re->rb_node);
+		*next_entry = rb_entry_safe(tmp_node, struct rb_entry, rb_node);
+	}
+	return re;
+}
+
+bool __check_rb_tree_consistence(struct f2fs_sb_info *sbi,
+						struct rb_root *root)
+{
+#ifdef CONFIG_F2FS_CHECK_FS
+	struct rb_node *cur = rb_first(root), *next;
+	struct rb_entry *cur_re, *next_re;
+
+	if (!cur)
+		return true;
+
+	while (cur) {
+		next = rb_next(cur);
+		if (!next)
+			return true;
+
+		cur_re = rb_entry(cur, struct rb_entry, rb_node);
+		next_re = rb_entry(next, struct rb_entry, rb_node);
+
+		if (cur_re->ofs + cur_re->len > next_re->ofs) {
+			f2fs_msg(sbi->sb, KERN_INFO, "inconsistent rbtree, "
+				"cur(%u, %u) next(%u, %u)",
+				cur_re->ofs, cur_re->len,
+				next_re->ofs, next_re->len);
+			return false;
+		}
+
+		cur = next;
+	}
+#endif
+	return true;
+}
+
 static struct kmem_cache *extent_tree_slab;
 static struct kmem_cache *extent_node_slab;
 
@@ -33,10 +206,11 @@ static struct extent_node *__attach_extent_node(struct f2fs_sb_info *sbi,
 
 	en->ei = *ei;
 	INIT_LIST_HEAD(&en->list);
+	en->et = et;
 
 	rb_link_node(&en->rb_node, parent, p);
 	rb_insert_color(&en->rb_node, &et->root);
-	et->count++;
+	atomic_inc(&et->node_cnt);
 	atomic_inc(&sbi->total_ext_node);
 	return en;
 }
@@ -45,11 +219,29 @@ static void __detach_extent_node(struct f2fs_sb_info *sbi,
 				struct extent_tree *et, struct extent_node *en)
 {
 	rb_erase(&en->rb_node, &et->root);
-	et->count--;
+	atomic_dec(&et->node_cnt);
 	atomic_dec(&sbi->total_ext_node);
 
 	if (et->cached_en == en)
 		et->cached_en = NULL;
+	kmem_cache_free(extent_node_slab, en);
+}
+
+/*
+ * Flow to release an extent_node:
+ * 1. list_del_init
+ * 2. __detach_extent_node
+ * 3. kmem_cache_free.
+ */
+static void __release_extent_node(struct f2fs_sb_info *sbi,
+			struct extent_tree *et, struct extent_node *en)
+{
+	spin_lock(&sbi->extent_lock);
+	f2fs_bug_on(sbi, list_empty(&en->list));
+	list_del_init(&en->list);
+	spin_unlock(&sbi->extent_lock);
+
+	__detach_extent_node(sbi, et, en);
 }
 
 static struct extent_tree *__grab_extent_tree(struct inode *inode)
@@ -58,7 +250,7 @@ static struct extent_tree *__grab_extent_tree(struct inode *inode)
 	struct extent_tree *et;
 	nid_t ino = inode->i_ino;
 
-	down_write(&sbi->extent_tree_lock);
+	mutex_lock(&sbi->extent_tree_lock);
 	et = radix_tree_lookup(&sbi->extent_tree_root, ino);
 	if (!et) {
 		et = f2fs_kmem_cache_alloc(extent_tree_slab, GFP_NOFS);
@@ -68,12 +260,14 @@ static struct extent_tree *__grab_extent_tree(struct inode *inode)
 		et->root = RB_ROOT;
 		et->cached_en = NULL;
 		rwlock_init(&et->lock);
-		atomic_set(&et->refcount, 0);
-		et->count = 0;
-		sbi->total_ext_tree++;
+		INIT_LIST_HEAD(&et->list);
+		atomic_set(&et->node_cnt, 0);
+		atomic_inc(&sbi->total_ext_tree);
+	} else {
+		atomic_dec(&sbi->total_zombie_tree);
+		list_del_init(&et->list);
 	}
-	atomic_inc(&et->refcount);
-	up_write(&sbi->extent_tree_lock);
+	mutex_unlock(&sbi->extent_tree_lock);
 
 	/* never died until evict_inode */
 	F2FS_I(inode)->extent_tree = et;
@@ -81,36 +275,6 @@ static struct extent_tree *__grab_extent_tree(struct inode *inode)
 	return et;
 }
 
-static struct extent_node *__lookup_extent_tree(struct f2fs_sb_info *sbi,
-				struct extent_tree *et, unsigned int fofs)
-{
-	struct rb_node *node = et->root.rb_node;
-	struct extent_node *en = et->cached_en;
-
-	if (en) {
-		struct extent_info *cei = &en->ei;
-
-		if (cei->fofs <= fofs && cei->fofs + cei->len > fofs) {
-			stat_inc_cached_node_hit(sbi);
-			return en;
-		}
-	}
-
-	while (node) {
-		en = rb_entry(node, struct extent_node, rb_node);
-
-		if (fofs < en->ei.fofs) {
-			node = node->rb_left;
-		} else if (fofs >= en->ei.fofs + en->ei.len) {
-			node = node->rb_right;
-		} else {
-			stat_inc_rbtree_node_hit(sbi);
-			return en;
-		}
-	}
-	return NULL;
-}
-
 static struct extent_node *__init_extent_tree(struct f2fs_sb_info *sbi,
 				struct extent_tree *et, struct extent_info *ei)
 {
@@ -127,32 +291,21 @@ static struct extent_node *__init_extent_tree(struct f2fs_sb_info *sbi,
 }
 
 static unsigned int __free_extent_tree(struct f2fs_sb_info *sbi,
-					struct extent_tree *et, bool free_all)
+					struct extent_tree *et)
 {
 	struct rb_node *node, *next;
 	struct extent_node *en;
-	unsigned int count = et->count;
+	unsigned int count = atomic_read(&et->node_cnt);
 
 	node = rb_first(&et->root);
 	while (node) {
 		next = rb_next(node);
 		en = rb_entry(node, struct extent_node, rb_node);
-
-		if (free_all) {
-			spin_lock(&sbi->extent_lock);
-			if (!list_empty(&en->list))
-				list_del_init(&en->list);
-			spin_unlock(&sbi->extent_lock);
-		}
-
-		if (free_all || list_empty(&en->list)) {
-			__detach_extent_node(sbi, et, en);
-			kmem_cache_free(extent_node_slab, en);
-		}
+		__release_extent_node(sbi, et, en);
 		node = next;
 	}
 
-	return count - et->count;
+	return count - atomic_read(&et->node_cnt);
 }
 
 static void __drop_largest_extent(struct inode *inode,
@@ -160,38 +313,38 @@ static void __drop_largest_extent(struct inode *inode,
 {
 	struct extent_info *largest = &F2FS_I(inode)->extent_tree->largest;
 
-	if (fofs < largest->fofs + largest->len && fofs + len > largest->fofs)
+	if (fofs < largest->fofs + largest->len && fofs + len > largest->fofs) {
 		largest->len = 0;
+		f2fs_mark_inode_dirty_sync(inode, true);
+	}
 }
 
-void f2fs_drop_largest_extent(struct inode *inode, pgoff_t fofs)
-{
-	if (!f2fs_may_extent_tree(inode))
-		return;
-
-	__drop_largest_extent(inode, fofs, 1);
-}
-
-void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
+/* return true, if inode page is changed */
+static bool __f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct extent_tree *et;
 	struct extent_node *en;
 	struct extent_info ei;
 
-	if (!f2fs_may_extent_tree(inode))
-		return;
+	if (!f2fs_may_extent_tree(inode)) {
+		/* drop largest extent */
+		if (i_ext && i_ext->len) {
+			i_ext->len = 0;
+			return true;
+		}
+		return false;
+	}
 
 	et = __grab_extent_tree(inode);
 
-	if (!i_ext || le32_to_cpu(i_ext->len) < F2FS_MIN_EXTENT_LEN)
-		return;
+	if (!i_ext || !i_ext->len)
+		return false;
 
-	set_extent_info(&ei, le32_to_cpu(i_ext->fofs),
-		le32_to_cpu(i_ext->blk), le32_to_cpu(i_ext->len));
+	get_extent_info(&ei, i_ext);
 
 	write_lock(&et->lock);
-	if (et->count)
+	if (atomic_read(&et->node_cnt))
 		goto out;
 
 	en = __init_extent_tree(sbi, et, &ei);
@@ -202,6 +355,17 @@ void f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
 	}
 out:
 	write_unlock(&et->lock);
+	return false;
+}
+
+bool f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext)
+{
+	bool ret =  __f2fs_init_extent_tree(inode, i_ext);
+
+	if (!F2FS_I(inode)->extent_tree)
+		set_inode_flag(inode, FI_NO_EXTENT);
+
+	return ret;
 }
 
 static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
@@ -226,16 +390,24 @@ static bool f2fs_lookup_extent_tree(struct inode *inode, pgoff_t pgofs,
 		goto out;
 	}
 
-	en = __lookup_extent_tree(sbi, et, pgofs);
-	if (en) {
-		*ei = en->ei;
-		spin_lock(&sbi->extent_lock);
-		if (!list_empty(&en->list))
-			list_move_tail(&en->list, &sbi->extent_list);
+	en = (struct extent_node *)__lookup_rb_tree(&et->root,
+				(struct rb_entry *)et->cached_en, pgofs);
+	if (!en)
+		goto out;
+
+	if (en == et->cached_en)
+		stat_inc_cached_node_hit(sbi);
+	else
+		stat_inc_rbtree_node_hit(sbi);
+
+	*ei = en->ei;
+	spin_lock(&sbi->extent_lock);
+	if (!list_empty(&en->list)) {
+		list_move_tail(&en->list, &sbi->extent_list);
 		et->cached_en = en;
-		spin_unlock(&sbi->extent_lock);
-		ret = true;
 	}
+	spin_unlock(&sbi->extent_lock);
+	ret = true;
 out:
 	stat_inc_total_hit(sbi);
 	read_unlock(&et->lock);
@@ -244,93 +416,12 @@ out:
 	return ret;
 }
 
-
-/*
- * lookup extent at @fofs, if hit, return the extent
- * if not, return NULL and
- * @prev_ex: extent before fofs
- * @next_ex: extent after fofs
- * @insert_p: insert point for new extent at fofs
- * in order to simpfy the insertion after.
- * tree must stay unchanged between lookup and insertion.
- */
-static struct extent_node *__lookup_extent_tree_ret(struct extent_tree *et,
-				unsigned int fofs,
-				struct extent_node **prev_ex,
-				struct extent_node **next_ex,
-				struct rb_node ***insert_p,
-				struct rb_node **insert_parent)
-{
-	struct rb_node **pnode = &et->root.rb_node;
-	struct rb_node *parent = NULL, *tmp_node;
-	struct extent_node *en = et->cached_en;
-
-	*insert_p = NULL;
-	*insert_parent = NULL;
-	*prev_ex = NULL;
-	*next_ex = NULL;
-
-	if (RB_EMPTY_ROOT(&et->root))
-		return NULL;
-
-	if (en) {
-		struct extent_info *cei = &en->ei;
-
-		if (cei->fofs <= fofs && cei->fofs + cei->len > fofs)
-			goto lookup_neighbors;
-	}
-
-	while (*pnode) {
-		parent = *pnode;
-		en = rb_entry(*pnode, struct extent_node, rb_node);
-
-		if (fofs < en->ei.fofs)
-			pnode = &(*pnode)->rb_left;
-		else if (fofs >= en->ei.fofs + en->ei.len)
-			pnode = &(*pnode)->rb_right;
-		else
-			goto lookup_neighbors;
-	}
-
-	*insert_p = pnode;
-	*insert_parent = parent;
-
-	en = rb_entry(parent, struct extent_node, rb_node);
-	tmp_node = parent;
-	if (parent && fofs > en->ei.fofs)
-		tmp_node = rb_next(parent);
-	*next_ex = tmp_node ?
-		rb_entry(tmp_node, struct extent_node, rb_node) : NULL;
-
-	tmp_node = parent;
-	if (parent && fofs < en->ei.fofs)
-		tmp_node = rb_prev(parent);
-	*prev_ex = tmp_node ?
-		rb_entry(tmp_node, struct extent_node, rb_node) : NULL;
-	return NULL;
-
-lookup_neighbors:
-	if (fofs == en->ei.fofs) {
-		/* lookup prev node for merging backward later */
-		tmp_node = rb_prev(&en->rb_node);
-		*prev_ex = tmp_node ?
-			rb_entry(tmp_node, struct extent_node, rb_node) : NULL;
-	}
-	if (fofs == en->ei.fofs + en->ei.len - 1) {
-		/* lookup next node for merging frontward later */
-		tmp_node = rb_next(&en->rb_node);
-		*next_ex = tmp_node ?
-			rb_entry(tmp_node, struct extent_node, rb_node) : NULL;
-	}
-	return en;
-}
-
-static struct extent_node *__try_merge_extent_node(struct f2fs_sb_info *sbi,
+static struct extent_node *__try_merge_extent_node(struct inode *inode,
 				struct extent_tree *et, struct extent_info *ei,
-				struct extent_node **den,
 				struct extent_node *prev_ex,
 				struct extent_node *next_ex)
 {
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct extent_node *en = NULL;
 
 	if (prev_ex && __is_back_mergeable(ei, &prev_ex->ei)) {
@@ -340,28 +431,35 @@ static struct extent_node *__try_merge_extent_node(struct f2fs_sb_info *sbi,
 	}
 
 	if (next_ex && __is_front_mergeable(ei, &next_ex->ei)) {
-		if (en) {
-			__detach_extent_node(sbi, et, prev_ex);
-			*den = prev_ex;
-		}
 		next_ex->ei.fofs = ei->fofs;
 		next_ex->ei.blk = ei->blk;
 		next_ex->ei.len += ei->len;
+		if (en)
+			__release_extent_node(sbi, et, prev_ex);
+
 		en = next_ex;
 	}
 
-	if (en) {
-		__try_update_largest_extent(et, en);
+	if (!en)
+		return NULL;
+
+	__try_update_largest_extent(inode, et, en);
+
+	spin_lock(&sbi->extent_lock);
+	if (!list_empty(&en->list)) {
+		list_move_tail(&en->list, &sbi->extent_list);
 		et->cached_en = en;
 	}
+	spin_unlock(&sbi->extent_lock);
 	return en;
 }
 
-static struct extent_node *__insert_extent_tree(struct f2fs_sb_info *sbi,
+static struct extent_node *__insert_extent_tree(struct inode *inode,
 				struct extent_tree *et, struct extent_info *ei,
 				struct rb_node **insert_p,
 				struct rb_node *insert_parent)
 {
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct rb_node **p = &et->root.rb_node;
 	struct rb_node *parent = NULL;
 	struct extent_node *en = NULL;
@@ -372,28 +470,23 @@ static struct extent_node *__insert_extent_tree(struct f2fs_sb_info *sbi,
 		goto do_insert;
 	}
 
-	while (*p) {
-		parent = *p;
-		en = rb_entry(parent, struct extent_node, rb_node);
-
-		if (ei->fofs < en->ei.fofs)
-			p = &(*p)->rb_left;
-		else if (ei->fofs >= en->ei.fofs + en->ei.len)
-			p = &(*p)->rb_right;
-		else
-			f2fs_bug_on(sbi, 1);
-	}
+	p = __lookup_rb_tree_for_insert(sbi, &et->root, &parent, ei->fofs);
 do_insert:
 	en = __attach_extent_node(sbi, et, ei, parent, p);
 	if (!en)
 		return NULL;
 
-	__try_update_largest_extent(et, en);
+	__try_update_largest_extent(inode, et, en);
+
+	/* update in global extent list */
+	spin_lock(&sbi->extent_lock);
+	list_add_tail(&en->list, &sbi->extent_list);
 	et->cached_en = en;
+	spin_unlock(&sbi->extent_lock);
 	return en;
 }
 
-static unsigned int f2fs_update_extent_tree_range(struct inode *inode,
+static void f2fs_update_extent_tree_range(struct inode *inode,
 				pgoff_t fofs, block_t blkaddr, unsigned int len)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
@@ -406,15 +499,15 @@ static unsigned int f2fs_update_extent_tree_range(struct inode *inode,
 	unsigned int pos = (unsigned int)fofs;
 
 	if (!et)
-		return false;
+		return;
 
 	trace_f2fs_update_extent_tree_range(inode, fofs, blkaddr, len);
 
 	write_lock(&et->lock);
 
-	if (is_inode_flag_set(F2FS_I(inode), FI_NO_EXTENT)) {
+	if (is_inode_flag_set(inode, FI_NO_EXTENT)) {
 		write_unlock(&et->lock);
-		return false;
+		return;
 	}
 
 	prev = et->largest;
@@ -427,8 +520,11 @@ static unsigned int f2fs_update_extent_tree_range(struct inode *inode,
 	__drop_largest_extent(inode, fofs, len);
 
 	/* 1. lookup first extent node in range [fofs, fofs + len - 1] */
-	en = __lookup_extent_tree_ret(et, fofs, &prev_en, &next_en,
-					&insert_p, &insert_parent);
+	en = (struct extent_node *)__lookup_rb_tree_ret(&et->root,
+					(struct rb_entry *)et->cached_en, fofs,
+					(struct rb_entry **)&prev_en,
+					(struct rb_entry **)&next_en,
+					&insert_p, &insert_parent, false);
 	if (!en)
 		en = next_en;
 
@@ -454,7 +550,7 @@ static unsigned int f2fs_update_extent_tree_range(struct inode *inode,
 				set_extent_info(&ei, end,
 						end - dei.fofs + dei.blk,
 						org_end - end);
-				en1 = __insert_extent_tree(sbi, et, &ei,
+				en1 = __insert_extent_tree(inode, et, &ei,
 							NULL, NULL);
 				next_en = en1;
 			} else {
@@ -469,15 +565,14 @@ static unsigned int f2fs_update_extent_tree_range(struct inode *inode,
 		if (!next_en) {
 			struct rb_node *node = rb_next(&en->rb_node);
 
-			next_en = node ?
-				rb_entry(node, struct extent_node, rb_node)
-				: NULL;
+			next_en = rb_entry_safe(node, struct extent_node,
+						rb_node);
 		}
 
 		if (parts)
-			__try_update_largest_extent(et, en);
+			__try_update_largest_extent(inode, et, en);
 		else
-			__detach_extent_node(sbi, et, en);
+			__release_extent_node(sbi, et, en);
 
 		/*
 		 * if original extent is split into zero or two parts, extent
@@ -488,143 +583,102 @@ static unsigned int f2fs_update_extent_tree_range(struct inode *inode,
 			insert_p = NULL;
 			insert_parent = NULL;
 		}
-
-		/* update in global extent list */
-		spin_lock(&sbi->extent_lock);
-		if (!parts && !list_empty(&en->list))
-			list_del(&en->list);
-		if (en1)
-			list_add_tail(&en1->list, &sbi->extent_list);
-		spin_unlock(&sbi->extent_lock);
-
-		/* release extent node */
-		if (!parts)
-			kmem_cache_free(extent_node_slab, en);
-
 		en = next_en;
 	}
 
 	/* 3. update extent in extent cache */
 	if (blkaddr) {
-		struct extent_node *den = NULL;
 
 		set_extent_info(&ei, fofs, blkaddr, len);
-		en1 = __try_merge_extent_node(sbi, et, &ei, &den,
-							prev_en, next_en);
-		if (!en1)
-			en1 = __insert_extent_tree(sbi, et, &ei,
+		if (!__try_merge_extent_node(inode, et, &ei, prev_en, next_en))
+			__insert_extent_tree(inode, et, &ei,
 						insert_p, insert_parent);
 
 		/* give up extent_cache, if split and small updates happen */
 		if (dei.len >= 1 &&
 				prev.len < F2FS_MIN_EXTENT_LEN &&
 				et->largest.len < F2FS_MIN_EXTENT_LEN) {
-			et->largest.len = 0;
-			set_inode_flag(F2FS_I(inode), FI_NO_EXTENT);
-		}
-
-		spin_lock(&sbi->extent_lock);
-		if (en1) {
-			if (list_empty(&en1->list))
-				list_add_tail(&en1->list, &sbi->extent_list);
-			else
-				list_move_tail(&en1->list, &sbi->extent_list);
+			__drop_largest_extent(inode, 0, UINT_MAX);
+			set_inode_flag(inode, FI_NO_EXTENT);
 		}
-		if (den && !list_empty(&den->list))
-			list_del(&den->list);
-		spin_unlock(&sbi->extent_lock);
-
-		if (den)
-			kmem_cache_free(extent_node_slab, den);
 	}
 
-	if (is_inode_flag_set(F2FS_I(inode), FI_NO_EXTENT))
-		__free_extent_tree(sbi, et, true);
+	if (is_inode_flag_set(inode, FI_NO_EXTENT))
+		__free_extent_tree(sbi, et);
 
 	write_unlock(&et->lock);
-
-	return !__is_extent_same(&prev, &et->largest);
 }
 
 unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *sbi, int nr_shrink)
 {
-	struct extent_tree *treevec[EXT_TREE_VEC_SIZE];
-	struct extent_node *en, *tmp;
-	unsigned long ino = F2FS_ROOT_INO(sbi);
-	struct radix_tree_root *root = &sbi->extent_tree_root;
-	unsigned int found;
+	struct extent_tree *et, *next;
+	struct extent_node *en;
 	unsigned int node_cnt = 0, tree_cnt = 0;
 	int remained;
 
 	if (!test_opt(sbi, EXTENT_CACHE))
 		return 0;
 
-	if (!down_write_trylock(&sbi->extent_tree_lock))
+	if (!atomic_read(&sbi->total_zombie_tree))
+		goto free_node;
+
+	if (!mutex_trylock(&sbi->extent_tree_lock))
 		goto out;
 
 	/* 1. remove unreferenced extent tree */
-	while ((found = radix_tree_gang_lookup(root,
-				(void **)treevec, ino, EXT_TREE_VEC_SIZE))) {
-		unsigned i;
-
-		ino = treevec[found - 1]->ino + 1;
-		for (i = 0; i < found; i++) {
-			struct extent_tree *et = treevec[i];
-
-			if (!atomic_read(&et->refcount)) {
-				write_lock(&et->lock);
-				node_cnt += __free_extent_tree(sbi, et, true);
-				write_unlock(&et->lock);
-
-				radix_tree_delete(root, et->ino);
-				kmem_cache_free(extent_tree_slab, et);
-				sbi->total_ext_tree--;
-				tree_cnt++;
-
-				if (node_cnt + tree_cnt >= nr_shrink)
-					goto unlock_out;
-			}
+	list_for_each_entry_safe(et, next, &sbi->zombie_list, list) {
+		if (atomic_read(&et->node_cnt)) {
+			write_lock(&et->lock);
+			node_cnt += __free_extent_tree(sbi, et);
+			write_unlock(&et->lock);
 		}
+		f2fs_bug_on(sbi, atomic_read(&et->node_cnt));
+		list_del_init(&et->list);
+		radix_tree_delete(&sbi->extent_tree_root, et->ino);
+		kmem_cache_free(extent_tree_slab, et);
+		atomic_dec(&sbi->total_ext_tree);
+		atomic_dec(&sbi->total_zombie_tree);
+		tree_cnt++;
+
+		if (node_cnt + tree_cnt >= nr_shrink)
+			goto unlock_out;
+		cond_resched();
 	}
-	up_write(&sbi->extent_tree_lock);
+	mutex_unlock(&sbi->extent_tree_lock);
 
+free_node:
 	/* 2. remove LRU extent entries */
-	if (!down_write_trylock(&sbi->extent_tree_lock))
+	if (!mutex_trylock(&sbi->extent_tree_lock))
 		goto out;
 
 	remained = nr_shrink - (node_cnt + tree_cnt);
 
 	spin_lock(&sbi->extent_lock);
-	list_for_each_entry_safe(en, tmp, &sbi->extent_list, list) {
-		if (!remained--)
+	for (; remained > 0; remained--) {
+		if (list_empty(&sbi->extent_list))
 			break;
-		list_del_init(&en->list);
-	}
-	spin_unlock(&sbi->extent_lock);
-
-	/*
-	 * reset ino for searching victims from beginning of global extent tree.
-	 */
-	ino = F2FS_ROOT_INO(sbi);
-
-	while ((found = radix_tree_gang_lookup(root,
-				(void **)treevec, ino, EXT_TREE_VEC_SIZE))) {
-		unsigned i;
+		en = list_first_entry(&sbi->extent_list,
+					struct extent_node, list);
+		et = en->et;
+		if (!write_trylock(&et->lock)) {
+			/* refresh this extent node's position in extent list */
+			list_move_tail(&en->list, &sbi->extent_list);
+			continue;
+		}
 
-		ino = treevec[found - 1]->ino + 1;
-		for (i = 0; i < found; i++) {
-			struct extent_tree *et = treevec[i];
+		list_del_init(&en->list);
+		spin_unlock(&sbi->extent_lock);
 
-			write_lock(&et->lock);
-			node_cnt += __free_extent_tree(sbi, et, false);
-			write_unlock(&et->lock);
+		__detach_extent_node(sbi, et, en);
 
-			if (node_cnt + tree_cnt >= nr_shrink)
-				goto unlock_out;
-		}
+		write_unlock(&et->lock);
+		node_cnt++;
+		spin_lock(&sbi->extent_lock);
 	}
+	spin_unlock(&sbi->extent_lock);
+
 unlock_out:
-	up_write(&sbi->extent_tree_lock);
+	mutex_unlock(&sbi->extent_tree_lock);
 out:
 	trace_f2fs_shrink_extent_tree(sbi, node_cnt, tree_cnt);
 
@@ -637,16 +691,29 @@ unsigned int f2fs_destroy_extent_node(struct inode *inode)
 	struct extent_tree *et = F2FS_I(inode)->extent_tree;
 	unsigned int node_cnt = 0;
 
-	if (!et)
+	if (!et || !atomic_read(&et->node_cnt))
 		return 0;
 
 	write_lock(&et->lock);
-	node_cnt = __free_extent_tree(sbi, et, true);
+	node_cnt = __free_extent_tree(sbi, et);
 	write_unlock(&et->lock);
 
 	return node_cnt;
 }
 
+void f2fs_drop_extent_tree(struct inode *inode)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct extent_tree *et = F2FS_I(inode)->extent_tree;
+
+	set_inode_flag(inode, FI_NO_EXTENT);
+
+	write_lock(&et->lock);
+	__free_extent_tree(sbi, et);
+	__drop_largest_extent(inode, 0, UINT_MAX);
+	write_unlock(&et->lock);
+}
+
 void f2fs_destroy_extent_tree(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
@@ -656,8 +723,12 @@ void f2fs_destroy_extent_tree(struct inode *inode)
 	if (!et)
 		return;
 
-	if (inode->i_nlink && !is_bad_inode(inode) && et->count) {
-		atomic_dec(&et->refcount);
+	if (inode->i_nlink && !is_bad_inode(inode) &&
+					atomic_read(&et->node_cnt)) {
+		mutex_lock(&sbi->extent_tree_lock);
+		list_add_tail(&et->list, &sbi->zombie_list);
+		atomic_inc(&sbi->total_zombie_tree);
+		mutex_unlock(&sbi->extent_tree_lock);
 		return;
 	}
 
@@ -665,13 +736,12 @@ void f2fs_destroy_extent_tree(struct inode *inode)
 	node_cnt = f2fs_destroy_extent_node(inode);
 
 	/* delete extent tree entry in radix tree */
-	down_write(&sbi->extent_tree_lock);
-	atomic_dec(&et->refcount);
-	f2fs_bug_on(sbi, atomic_read(&et->refcount) || et->count);
+	mutex_lock(&sbi->extent_tree_lock);
+	f2fs_bug_on(sbi, atomic_read(&et->node_cnt));
 	radix_tree_delete(&sbi->extent_tree_root, inode->i_ino);
 	kmem_cache_free(extent_tree_slab, et);
-	sbi->total_ext_tree--;
-	up_write(&sbi->extent_tree_lock);
+	atomic_dec(&sbi->total_ext_tree);
+	mutex_unlock(&sbi->extent_tree_lock);
 
 	F2FS_I(inode)->extent_tree = NULL;
 
@@ -689,20 +759,20 @@ bool f2fs_lookup_extent_cache(struct inode *inode, pgoff_t pgofs,
 
 void f2fs_update_extent_cache(struct dnode_of_data *dn)
 {
-	struct f2fs_inode_info *fi = F2FS_I(dn->inode);
 	pgoff_t fofs;
+	block_t blkaddr;
 
 	if (!f2fs_may_extent_tree(dn->inode))
 		return;
 
-	f2fs_bug_on(F2FS_I_SB(dn->inode), dn->data_blkaddr == NEW_ADDR);
-
-
-	fofs = start_bidx_of_node(ofs_of_node(dn->node_page), fi) +
-							dn->ofs_in_node;
+	if (dn->data_blkaddr == NEW_ADDR)
+		blkaddr = NULL_ADDR;
+	else
+		blkaddr = dn->data_blkaddr;
 
-	if (f2fs_update_extent_tree_range(dn->inode, fofs, dn->data_blkaddr, 1))
-		sync_inode_page(dn);
+	fofs = start_bidx_of_node(ofs_of_node(dn->node_page), dn->inode) +
+								dn->ofs_in_node;
+	f2fs_update_extent_tree_range(dn->inode, fofs, blkaddr, 1);
 }
 
 void f2fs_update_extent_cache_range(struct dnode_of_data *dn,
@@ -712,17 +782,18 @@ void f2fs_update_extent_cache_range(struct dnode_of_data *dn,
 	if (!f2fs_may_extent_tree(dn->inode))
 		return;
 
-	if (f2fs_update_extent_tree_range(dn->inode, fofs, blkaddr, len))
-		sync_inode_page(dn);
+	f2fs_update_extent_tree_range(dn->inode, fofs, blkaddr, len);
 }
 
 void init_extent_cache_info(struct f2fs_sb_info *sbi)
 {
 	INIT_RADIX_TREE(&sbi->extent_tree_root, GFP_NOIO);
-	init_rwsem(&sbi->extent_tree_lock);
+	mutex_init(&sbi->extent_tree_lock);
 	INIT_LIST_HEAD(&sbi->extent_list);
 	spin_lock_init(&sbi->extent_lock);
-	sbi->total_ext_tree = 0;
+	atomic_set(&sbi->total_ext_tree, 0);
+	INIT_LIST_HEAD(&sbi->zombie_list);
+	atomic_set(&sbi->total_zombie_tree, 0);
 	atomic_set(&sbi->total_ext_node, 0);
 }
 
diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h
index 2871576fbca4..c1a0aef8efc6 100644
--- a/fs/f2fs/f2fs.h
+++ b/fs/f2fs/f2fs.h
@@ -21,10 +21,18 @@
 #include <linux/sched.h>
 #include <linux/vmalloc.h>
 #include <linux/bio.h>
+#include <linux/blkdev.h>
+#include <linux/quotaops.h>
+#ifdef CONFIG_F2FS_FS_ENCRYPTION
+#include <linux/fscrypt_supp.h>
+#else
+#include <linux/fscrypt_notsupp.h>
+#endif
+#include <crypto/hash.h>
+#include <linux/writeback.h>
 
 #ifdef CONFIG_F2FS_CHECK_FS
 #define f2fs_bug_on(sbi, condition)	BUG_ON(condition)
-#define f2fs_down_write(x, y)	down_write_nest_lock(x, y)
 #else
 #define f2fs_bug_on(sbi, condition)					\
 	do {								\
@@ -33,7 +41,31 @@
 			set_sbi_flag(sbi, SBI_NEED_FSCK);		\
 		}							\
 	} while (0)
-#define f2fs_down_write(x, y)	down_write(x)
+#endif
+
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+enum {
+	FAULT_KMALLOC,
+	FAULT_PAGE_ALLOC,
+	FAULT_ALLOC_NID,
+	FAULT_ORPHAN,
+	FAULT_BLOCK,
+	FAULT_DIR_DEPTH,
+	FAULT_EVICT_INODE,
+	FAULT_TRUNCATE,
+	FAULT_IO,
+	FAULT_CHECKPOINT,
+	FAULT_MAX,
+};
+
+struct f2fs_fault_info {
+	atomic_t inject_ops;
+	unsigned int inject_rate;
+	unsigned int inject_type;
+};
+
+extern char *fault_name[FAULT_MAX];
+#define IS_FAULT_SET(fi, type) ((fi)->inject_type & (1 << (type)))
 #endif
 
 /*
@@ -54,10 +86,18 @@
 #define F2FS_MOUNT_FASTBOOT		0x00001000
 #define F2FS_MOUNT_EXTENT_CACHE		0x00002000
 #define F2FS_MOUNT_FORCE_FG_GC		0x00004000
-
-#define clear_opt(sbi, option)	(sbi->mount_opt.opt &= ~F2FS_MOUNT_##option)
-#define set_opt(sbi, option)	(sbi->mount_opt.opt |= F2FS_MOUNT_##option)
-#define test_opt(sbi, option)	(sbi->mount_opt.opt & F2FS_MOUNT_##option)
+#define F2FS_MOUNT_DATA_FLUSH		0x00008000
+#define F2FS_MOUNT_FAULT_INJECTION	0x00010000
+#define F2FS_MOUNT_ADAPTIVE		0x00020000
+#define F2FS_MOUNT_LFS			0x00040000
+#define F2FS_MOUNT_USRQUOTA		0x00080000
+#define F2FS_MOUNT_GRPQUOTA		0x00100000
+#define F2FS_MOUNT_PRJQUOTA		0x00200000
+#define F2FS_MOUNT_QUOTA		0x00400000
+
+#define clear_opt(sbi, option)	((sbi)->mount_opt.opt &= ~F2FS_MOUNT_##option)
+#define set_opt(sbi, option)	((sbi)->mount_opt.opt |= F2FS_MOUNT_##option)
+#define test_opt(sbi, option)	((sbi)->mount_opt.opt & F2FS_MOUNT_##option)
 
 #define ver_after(a, b)	(typecheck(unsigned long long, a) &&		\
 		typecheck(unsigned long long, b) &&			\
@@ -73,34 +113,84 @@ struct f2fs_mount_info {
 	unsigned int	opt;
 };
 
-#define F2FS_FEATURE_ENCRYPT	0x0001
+#define F2FS_FEATURE_ENCRYPT		0x0001
+#define F2FS_FEATURE_BLKZONED		0x0002
+#define F2FS_FEATURE_ATOMIC_WRITE	0x0004
+#define F2FS_FEATURE_EXTRA_ATTR		0x0008
+#define F2FS_FEATURE_PRJQUOTA		0x0010
+#define F2FS_FEATURE_INODE_CHKSUM	0x0020
 
 #define F2FS_HAS_FEATURE(sb, mask)					\
 	((F2FS_SB(sb)->raw_super->feature & cpu_to_le32(mask)) != 0)
 #define F2FS_SET_FEATURE(sb, mask)					\
-	F2FS_SB(sb)->raw_super->feature |= cpu_to_le32(mask)
+	(F2FS_SB(sb)->raw_super->feature |= cpu_to_le32(mask))
 #define F2FS_CLEAR_FEATURE(sb, mask)					\
-	F2FS_SB(sb)->raw_super->feature &= ~cpu_to_le32(mask)
+	(F2FS_SB(sb)->raw_super->feature &= ~cpu_to_le32(mask))
 
-#define CRCPOLY_LE 0xedb88320
+/* bio stuffs */
+#define REQ_OP_READ	READ
+#define REQ_OP_WRITE	WRITE
+#define bio_op(bio)	((bio)->bi_rw & 1)
 
-static inline __u32 f2fs_crc32(void *buf, size_t len)
+static inline void bio_set_op_attrs(struct bio *bio, unsigned op,
+		unsigned op_flags)
 {
-	unsigned char *p = (unsigned char *)buf;
-	__u32 crc = F2FS_SUPER_MAGIC;
-	int i;
+	bio->bi_rw = op | op_flags;
+}
 
-	while (len--) {
-		crc ^= *p++;
-		for (i = 0; i < 8; i++)
-			crc = (crc >> 1) ^ ((crc & 1) ? CRCPOLY_LE : 0);
-	}
-	return crc;
+static inline int wbc_to_write_flags(struct writeback_control *wbc)
+{
+	if (wbc->sync_mode == WB_SYNC_ALL)
+		return REQ_SYNC | REQ_NOIDLE;
+	return 0;
+}
+
+/**
+ * wq_has_sleeper - check if there are any waiting processes
+ * @wq: wait queue head
+ *
+ * Returns true if wq has waiting processes
+ *
+ * Please refer to the comment for waitqueue_active.
+ */
+static inline bool wq_has_sleeper(wait_queue_head_t *wq)
+{
+	/*
+	 * We need to be sure we are in sync with the
+	 * add_wait_queue modifications to the wait queue.
+	 *
+	 * This memory barrier should be paired with one on the
+	 * waiting side.
+	 */
+	smp_mb();
+	return waitqueue_active(wq);
+}
+
+static inline void inode_nohighmem(struct inode *inode)
+{
+	mapping_set_gfp_mask(inode->i_mapping, GFP_USER);
 }
 
-static inline bool f2fs_crc_valid(__u32 blk_crc, void *buf, size_t buf_size)
+/**
+ * current_time - Return FS time
+ * @inode: inode.
+ *
+ * Return the current time truncated to the time granularity supported by
+ * the fs.
+ *
+ * Note that inode and inode->sb cannot be NULL.
+ * Otherwise, the function warns and returns time without truncation.
+ */
+static inline struct timespec current_time(struct inode *inode)
 {
-	return f2fs_crc32(buf, buf_size) == blk_crc;
+	struct timespec now = current_kernel_time();
+
+	if (unlikely(!inode->i_sb)) {
+		WARN(1, "current_time() called with uninitialized super_block in the inode");
+		return now; 
+	}    
+
+	return timespec_trunc(now, inode->i_sb->s_time_gran);
 }
 
 /*
@@ -111,20 +201,24 @@ enum {
 	SIT_BITMAP
 };
 
-enum {
-	CP_UMOUNT,
-	CP_FASTBOOT,
-	CP_SYNC,
-	CP_RECOVERY,
-	CP_DISCARD,
-};
+#define	CP_UMOUNT	0x00000001
+#define	CP_FASTBOOT	0x00000002
+#define	CP_SYNC		0x00000004
+#define	CP_RECOVERY	0x00000008
+#define	CP_DISCARD	0x00000010
+#define CP_TRIMMED	0x00000020
 
-#define DEF_BATCHED_TRIM_SECTIONS	32
+#define DEF_BATCHED_TRIM_SECTIONS	2048
 #define BATCHED_TRIM_SEGMENTS(sbi)	\
-		(SM_I(sbi)->trim_sections * (sbi)->segs_per_sec)
+		(GET_SEG_FROM_SEC(sbi, SM_I(sbi)->trim_sections))
 #define BATCHED_TRIM_BLOCKS(sbi)	\
 		(BATCHED_TRIM_SEGMENTS(sbi) << (sbi)->log_blocks_per_seg)
+#define MAX_DISCARD_BLOCKS(sbi)		BLKS_PER_SEC(sbi)
+#define DISCARD_ISSUE_RATE		8
+#define DEF_MIN_DISCARD_ISSUE_TIME	50	/* 50 ms, if exists */
+#define DEF_MAX_DISCARD_ISSUE_TIME	60000	/* 60 s, if no candidates */
 #define DEF_CP_INTERVAL			60	/* 60 secs */
+#define DEF_IDLE_INTERVAL		5	/* 5 secs */
 
 struct cp_control {
 	int reason;
@@ -158,23 +252,79 @@ struct ino_entry {
 	nid_t ino;		/* inode number */
 };
 
-/*
- * for the list of directory inodes or gc inodes.
- * NOTE: there are two slab users for this structure, if we add/modify/delete
- * fields in structure for one of slab users, it may affect fields or size of
- * other one, in this condition, it's better to split both of slab and related
- * data structure.
- */
+/* for the list of inodes to be GCed */
 struct inode_entry {
 	struct list_head list;	/* list head */
 	struct inode *inode;	/* vfs inode pointer */
 };
 
-/* for the list of blockaddresses to be discarded */
+/* for the bitmap indicate blocks to be discarded */
 struct discard_entry {
 	struct list_head list;	/* list head */
-	block_t blkaddr;	/* block address to be discarded */
-	int len;		/* # of consecutive blocks of the discard */
+	block_t start_blkaddr;	/* start blockaddr of current segment */
+	unsigned char discard_map[SIT_VBLOCK_MAP_SIZE];	/* segment discard bitmap */
+};
+
+/* default discard granularity of inner discard thread, unit: block count */
+#define DEFAULT_DISCARD_GRANULARITY		16
+
+/* max discard pend list number */
+#define MAX_PLIST_NUM		512
+#define plist_idx(blk_num)	((blk_num) >= MAX_PLIST_NUM ?		\
+					(MAX_PLIST_NUM - 1) : (blk_num - 1))
+
+#define P_ACTIVE	0x01
+#define P_TRIM		0x02
+#define plist_issue(tag)	(((tag) & P_ACTIVE) || ((tag) & P_TRIM))
+
+enum {
+	D_PREP,
+	D_SUBMIT,
+	D_DONE,
+};
+
+struct discard_info {
+	block_t lstart;			/* logical start address */
+	block_t len;			/* length */
+	block_t start;			/* actual start address in dev */
+};
+
+struct discard_cmd {
+	struct rb_node rb_node;		/* rb node located in rb-tree */
+	union {
+		struct {
+			block_t lstart;	/* logical start address */
+			block_t len;	/* length */
+			block_t start;	/* actual start address in dev */
+		};
+		struct discard_info di;	/* discard info */
+
+	};
+	struct list_head list;		/* command list */
+	struct completion wait;		/* compleation */
+	struct block_device *bdev;	/* bdev */
+	unsigned short ref;		/* reference count */
+	unsigned char state;		/* state */
+	int error;			/* bio error */
+};
+
+struct discard_cmd_control {
+	struct task_struct *f2fs_issue_discard;	/* discard thread */
+	struct list_head entry_list;		/* 4KB discard entry list */
+	struct list_head pend_list[MAX_PLIST_NUM];/* store pending entries */
+	unsigned char pend_list_tag[MAX_PLIST_NUM];/* tag for pending entries */
+	struct list_head wait_list;		/* store on-flushing entries */
+	wait_queue_head_t discard_wait_queue;	/* waiting queue for wake-up */
+	unsigned int discard_wake;		/* to wake up discard thread */
+	struct mutex cmd_lock;
+	unsigned int nr_discards;		/* # of discards in the list */
+	unsigned int max_discards;		/* max. discards to be issued */
+	unsigned int discard_granularity;	/* discard granularity */
+	unsigned int undiscard_blks;		/* # of undiscard blocks */
+	atomic_t issued_discard;		/* # of issued discard */
+	atomic_t issing_discard;		/* # of issing discard */
+	atomic_t discard_cmd_cnt;		/* # of cached cmd count */
+	struct rb_root root;			/* root of discard rb-tree */
 };
 
 /* for the list of fsync inodes, used only during recovery */
@@ -183,40 +333,41 @@ struct fsync_inode_entry {
 	struct inode *inode;	/* vfs inode pointer */
 	block_t blkaddr;	/* block address locating the last fsync */
 	block_t last_dentry;	/* block address locating the last dentry */
-	block_t last_inode;	/* block address locating the last inode */
 };
 
-#define nats_in_cursum(sum)		(le16_to_cpu(sum->n_nats))
-#define sits_in_cursum(sum)		(le16_to_cpu(sum->n_sits))
+#define nats_in_cursum(jnl)		(le16_to_cpu((jnl)->n_nats))
+#define sits_in_cursum(jnl)		(le16_to_cpu((jnl)->n_sits))
 
-#define nat_in_journal(sum, i)		(sum->nat_j.entries[i].ne)
-#define nid_in_journal(sum, i)		(sum->nat_j.entries[i].nid)
-#define sit_in_journal(sum, i)		(sum->sit_j.entries[i].se)
-#define segno_in_journal(sum, i)	(sum->sit_j.entries[i].segno)
+#define nat_in_journal(jnl, i)		((jnl)->nat_j.entries[i].ne)
+#define nid_in_journal(jnl, i)		((jnl)->nat_j.entries[i].nid)
+#define sit_in_journal(jnl, i)		((jnl)->sit_j.entries[i].se)
+#define segno_in_journal(jnl, i)	((jnl)->sit_j.entries[i].segno)
 
-#define MAX_NAT_JENTRIES(sum)	(NAT_JOURNAL_ENTRIES - nats_in_cursum(sum))
-#define MAX_SIT_JENTRIES(sum)	(SIT_JOURNAL_ENTRIES - sits_in_cursum(sum))
+#define MAX_NAT_JENTRIES(jnl)	(NAT_JOURNAL_ENTRIES - nats_in_cursum(jnl))
+#define MAX_SIT_JENTRIES(jnl)	(SIT_JOURNAL_ENTRIES - sits_in_cursum(jnl))
 
-static inline int update_nats_in_cursum(struct f2fs_summary_block *rs, int i)
+static inline int update_nats_in_cursum(struct f2fs_journal *journal, int i)
 {
-	int before = nats_in_cursum(rs);
-	rs->n_nats = cpu_to_le16(before + i);
+	int before = nats_in_cursum(journal);
+
+	journal->n_nats = cpu_to_le16(before + i);
 	return before;
 }
 
-static inline int update_sits_in_cursum(struct f2fs_summary_block *rs, int i)
+static inline int update_sits_in_cursum(struct f2fs_journal *journal, int i)
 {
-	int before = sits_in_cursum(rs);
-	rs->n_sits = cpu_to_le16(before + i);
+	int before = sits_in_cursum(journal);
+
+	journal->n_sits = cpu_to_le16(before + i);
 	return before;
 }
 
-static inline bool __has_cursum_space(struct f2fs_summary_block *sum, int size,
-								int type)
+static inline bool __has_cursum_space(struct f2fs_journal *journal,
+							int size, int type)
 {
 	if (type == NAT_JOURNAL)
-		return size <= MAX_NAT_JENTRIES(sum);
-	return size <= MAX_SIT_JENTRIES(sum);
+		return size <= MAX_NAT_JENTRIES(journal);
+	return size <= MAX_SIT_JENTRIES(journal);
 }
 
 /*
@@ -232,15 +383,21 @@ static inline bool __has_cursum_space(struct f2fs_summary_block *sum, int size,
 #define F2FS_IOC_START_VOLATILE_WRITE	_IO(F2FS_IOCTL_MAGIC, 3)
 #define F2FS_IOC_RELEASE_VOLATILE_WRITE	_IO(F2FS_IOCTL_MAGIC, 4)
 #define F2FS_IOC_ABORT_VOLATILE_WRITE	_IO(F2FS_IOCTL_MAGIC, 5)
-#define F2FS_IOC_GARBAGE_COLLECT	_IO(F2FS_IOCTL_MAGIC, 6)
+#define F2FS_IOC_GARBAGE_COLLECT	_IOW(F2FS_IOCTL_MAGIC, 6, __u32)
 #define F2FS_IOC_WRITE_CHECKPOINT	_IO(F2FS_IOCTL_MAGIC, 7)
-
-#define F2FS_IOC_SET_ENCRYPTION_POLICY					\
-		_IOR('f', 19, struct f2fs_encryption_policy)
-#define F2FS_IOC_GET_ENCRYPTION_PWSALT					\
-		_IOW('f', 20, __u8[16])
-#define F2FS_IOC_GET_ENCRYPTION_POLICY					\
-		_IOW('f', 21, struct f2fs_encryption_policy)
+#define F2FS_IOC_DEFRAGMENT		_IOWR(F2FS_IOCTL_MAGIC, 8,	\
+						struct f2fs_defragment)
+#define F2FS_IOC_MOVE_RANGE		_IOWR(F2FS_IOCTL_MAGIC, 9,	\
+						struct f2fs_move_range)
+#define F2FS_IOC_FLUSH_DEVICE		_IOW(F2FS_IOCTL_MAGIC, 10,	\
+						struct f2fs_flush_device)
+#define F2FS_IOC_GARBAGE_COLLECT_RANGE	_IOW(F2FS_IOCTL_MAGIC, 11,	\
+						struct f2fs_gc_range)
+#define F2FS_IOC_GET_FEATURES		_IOR(F2FS_IOCTL_MAGIC, 12, __u32)
+
+#define F2FS_IOC_SET_ENCRYPTION_POLICY	FS_IOC_SET_ENCRYPTION_POLICY
+#define F2FS_IOC_GET_ENCRYPTION_POLICY	FS_IOC_GET_ENCRYPTION_POLICY
+#define F2FS_IOC_GET_ENCRYPTION_PWSALT	FS_IOC_GET_ENCRYPTION_PWSALT
 
 /*
  * should be same as XFS_IOC_GOINGDOWN.
@@ -256,59 +413,91 @@ static inline bool __has_cursum_space(struct f2fs_summary_block *sum, int size,
 /*
  * ioctl commands in 32 bit emulation
  */
-#define F2FS_IOC32_GETFLAGS             FS_IOC32_GETFLAGS
-#define F2FS_IOC32_SETFLAGS             FS_IOC32_SETFLAGS
+#define F2FS_IOC32_GETFLAGS		FS_IOC32_GETFLAGS
+#define F2FS_IOC32_SETFLAGS		FS_IOC32_SETFLAGS
+#define F2FS_IOC32_GETVERSION		FS_IOC32_GETVERSION
 #endif
 
-/*
- * For INODE and NODE manager
- */
-/* for directory operations */
-struct f2fs_str {
-	unsigned char *name;
-	u32 len;
+struct f2fs_gc_range {
+	u32 sync;
+	u64 start;
+	u64 len;
 };
 
-struct f2fs_filename {
-	const struct qstr *usr_fname;
-	struct f2fs_str disk_name;
-	f2fs_hash_t hash;
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	struct f2fs_str crypto_buf;
-#endif
+struct f2fs_defragment {
+	u64 start;
+	u64 len;
+};
+
+struct f2fs_move_range {
+	u32 dst_fd;		/* destination fd */
+	u64 pos_in;		/* start position in src_fd */
+	u64 pos_out;		/* start position in dst_fd */
+	u64 len;		/* size to move */
 };
 
-#define FSTR_INIT(n, l)		{ .name = n, .len = l }
-#define FSTR_TO_QSTR(f)		QSTR_INIT((f)->name, (f)->len)
-#define fname_name(p)		((p)->disk_name.name)
-#define fname_len(p)		((p)->disk_name.len)
+struct f2fs_flush_device {
+	u32 dev_num;		/* device number to flush */
+	u32 segments;		/* # of segments to flush */
+};
+
+/* for inline stuff */
+#define DEF_INLINE_RESERVED_SIZE	1
+static inline int get_extra_isize(struct inode *inode);
+#define MAX_INLINE_DATA(inode)	(sizeof(__le32) * \
+				(CUR_ADDRS_PER_INODE(inode) - \
+				DEF_INLINE_RESERVED_SIZE - \
+				F2FS_INLINE_XATTR_ADDRS))
+
+/* for inline dir */
+#define NR_INLINE_DENTRY(inode)	(MAX_INLINE_DATA(inode) * BITS_PER_BYTE / \
+				((SIZE_OF_DIR_ENTRY + F2FS_SLOT_LEN) * \
+				BITS_PER_BYTE + 1))
+#define INLINE_DENTRY_BITMAP_SIZE(inode)	((NR_INLINE_DENTRY(inode) + \
+					BITS_PER_BYTE - 1) / BITS_PER_BYTE)
+#define INLINE_RESERVED_SIZE(inode)	(MAX_INLINE_DATA(inode) - \
+				((SIZE_OF_DIR_ENTRY + F2FS_SLOT_LEN) * \
+				NR_INLINE_DENTRY(inode) + \
+				INLINE_DENTRY_BITMAP_SIZE(inode)))
 
+/*
+ * For INODE and NODE manager
+ */
+/* for directory operations */
 struct f2fs_dentry_ptr {
 	struct inode *inode;
-	const void *bitmap;
+	void *bitmap;
 	struct f2fs_dir_entry *dentry;
 	__u8 (*filename)[F2FS_SLOT_LEN];
 	int max;
+	int nr_bitmap;
 };
 
-static inline void make_dentry_ptr(struct inode *inode,
-		struct f2fs_dentry_ptr *d, void *src, int type)
+static inline void make_dentry_ptr_block(struct inode *inode,
+		struct f2fs_dentry_ptr *d, struct f2fs_dentry_block *t)
 {
 	d->inode = inode;
+	d->max = NR_DENTRY_IN_BLOCK;
+	d->nr_bitmap = SIZE_OF_DENTRY_BITMAP;
+	d->bitmap = &t->dentry_bitmap;
+	d->dentry = t->dentry;
+	d->filename = t->filename;
+}
 
-	if (type == 1) {
-		struct f2fs_dentry_block *t = (struct f2fs_dentry_block *)src;
-		d->max = NR_DENTRY_IN_BLOCK;
-		d->bitmap = &t->dentry_bitmap;
-		d->dentry = t->dentry;
-		d->filename = t->filename;
-	} else {
-		struct f2fs_inline_dentry *t = (struct f2fs_inline_dentry *)src;
-		d->max = NR_INLINE_DENTRY;
-		d->bitmap = &t->dentry_bitmap;
-		d->dentry = t->dentry;
-		d->filename = t->filename;
-	}
+static inline void make_dentry_ptr_inline(struct inode *inode,
+					struct f2fs_dentry_ptr *d, void *t)
+{
+	int entry_cnt = NR_INLINE_DENTRY(inode);
+	int bitmap_size = INLINE_DENTRY_BITMAP_SIZE(inode);
+	int reserved_size = INLINE_RESERVED_SIZE(inode);
+
+	d->inode = inode;
+	d->max = entry_cnt;
+	d->nr_bitmap = bitmap_size;
+	d->bitmap = t;
+	d->dentry = t + bitmap_size + reserved_size;
+	d->filename = t + bitmap_size + reserved_size +
+					SIZE_OF_DIR_ENTRY * entry_cnt;
 }
 
 /*
@@ -340,16 +529,31 @@ enum {
 /* number of extent info in extent cache we try to shrink */
 #define EXTENT_CACHE_SHRINK_NUMBER	128
 
+struct rb_entry {
+	struct rb_node rb_node;		/* rb node located in rb-tree */
+	unsigned int ofs;		/* start offset of the entry */
+	unsigned int len;		/* length of the entry */
+};
+
 struct extent_info {
 	unsigned int fofs;		/* start offset in a file */
-	u32 blk;			/* start block address of the extent */
 	unsigned int len;		/* length of the extent */
+	u32 blk;			/* start block address of the extent */
 };
 
 struct extent_node {
-	struct rb_node rb_node;		/* rb node located in rb-tree */
+	struct rb_node rb_node;
+	union {
+		struct {
+			unsigned int fofs;
+			unsigned int len;
+			u32 blk;
+		};
+		struct extent_info ei;	/* extent info */
+
+	};
 	struct list_head list;		/* node in global extent list of sbi */
-	struct extent_info ei;		/* extent info */
+	struct extent_tree *et;		/* extent tree pointer */
 };
 
 struct extent_tree {
@@ -357,9 +561,9 @@ struct extent_tree {
 	struct rb_root root;		/* root of extent info rb-tree */
 	struct extent_node *cached_en;	/* recently accessed extent node */
 	struct extent_info largest;	/* largested extent info */
+	struct list_head list;		/* to be used by sbi->zombie_list */
 	rwlock_t lock;			/* protect extent info rb-tree */
-	atomic_t refcount;		/* reference count of rb-tree */
-	unsigned int count;		/* # of extent node in rb-tree*/
+	atomic_t node_cnt;		/* # of extent node in rb-tree*/
 };
 
 /*
@@ -378,13 +582,17 @@ struct f2fs_map_blocks {
 	block_t m_lblk;
 	unsigned int m_len;
 	unsigned int m_flags;
+	pgoff_t *m_next_pgofs;		/* point next possible non-hole pgofs */
 };
 
 /* for flag in get_data_block */
-#define F2FS_GET_BLOCK_READ		0
-#define F2FS_GET_BLOCK_DIO		1
-#define F2FS_GET_BLOCK_FIEMAP		2
-#define F2FS_GET_BLOCK_BMAP		3
+enum {
+	F2FS_GET_BLOCK_DEFAULT,
+	F2FS_GET_BLOCK_FIEMAP,
+	F2FS_GET_BLOCK_BMAP,
+	F2FS_GET_BLOCK_PRE_DIO,
+	F2FS_GET_BLOCK_PRE_AIO,
+};
 
 /*
  * i_advise uses FADVISE_XXX_BIT. We can add additional hints later.
@@ -393,6 +601,7 @@ struct f2fs_map_blocks {
 #define FADVISE_LOST_PINO_BIT	0x02
 #define FADVISE_ENCRYPT_BIT	0x04
 #define FADVISE_ENC_NAME_BIT	0x08
+#define FADVISE_KEEP_SIZE_BIT	0x10
 
 #define file_is_cold(inode)	is_file(inode, FADVISE_COLD_BIT)
 #define file_wrong_pino(inode)	is_file(inode, FADVISE_LOST_PINO_BIT)
@@ -405,15 +614,8 @@ struct f2fs_map_blocks {
 #define file_clear_encrypt(inode) clear_file(inode, FADVISE_ENCRYPT_BIT)
 #define file_enc_name(inode)	is_file(inode, FADVISE_ENC_NAME_BIT)
 #define file_set_enc_name(inode) set_file(inode, FADVISE_ENC_NAME_BIT)
-
-/* Encryption algorithms */
-#define F2FS_ENCRYPTION_MODE_INVALID		0
-#define F2FS_ENCRYPTION_MODE_AES_256_XTS	1
-#define F2FS_ENCRYPTION_MODE_AES_256_GCM	2
-#define F2FS_ENCRYPTION_MODE_AES_256_CBC	3
-#define F2FS_ENCRYPTION_MODE_AES_256_CTS	4
-
-#include "f2fs_crypto.h"
+#define file_keep_isize(inode)	is_file(inode, FADVISE_KEEP_SIZE_BIT)
+#define file_set_keep_isize(inode) set_file(inode, FADVISE_KEEP_SIZE_BIT)
 
 #define DEF_DIR_LEVEL		0
 
@@ -432,27 +634,37 @@ struct f2fs_inode_info {
 	atomic_t dirty_pages;		/* # of dirty pages */
 	f2fs_hash_t chash;		/* hash value of given file name */
 	unsigned int clevel;		/* maximum level of given file name */
+	struct task_struct *task;	/* lookup and create consistency */
+	struct task_struct *cp_task;	/* separate cp/wb IO stats*/
 	nid_t i_xattr_nid;		/* node id that contains xattrs */
-	unsigned long long xattr_ver;	/* cp version of xattr modification */
-	struct inode_entry *dirty_dir;	/* the pointer of dirty dir */
+	loff_t	last_disk_size;		/* lastly written file size */
 
+#ifdef CONFIG_QUOTA
+	struct dquot *i_dquot[MAXQUOTAS];
+
+	/* quota space reservation, managed internally by quota code */
+	qsize_t i_reserved_quota;
+#endif
+	struct list_head dirty_list;	/* dirty list for dirs and files */
+	struct list_head gdirty_list;	/* linked in global dirty list */
 	struct list_head inmem_pages;	/* inmemory pages managed by f2fs */
+	struct task_struct *inmem_task;	/* store inmemory task */
 	struct mutex inmem_lock;	/* lock for inmemory pages */
-
 	struct extent_tree *extent_tree;	/* cached extent_tree entry */
+	struct rw_semaphore dio_rwsem[2];/* avoid racing between dio and gc */
+	struct rw_semaphore i_mmap_sem;
+	struct rw_semaphore i_xattr_sem; /* avoid racing between reading and changing EAs */
 
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	/* Encryption params */
-	struct f2fs_crypt_info *i_crypt_info;
-#endif
+	int i_extra_isize;		/* size of extra space located in i_addr */
+	kprojid_t i_projid;		/* id for project quota */
 };
 
 static inline void get_extent_info(struct extent_info *ext,
-					struct f2fs_extent i_ext)
+					struct f2fs_extent *i_ext)
 {
-	ext->fofs = le32_to_cpu(i_ext.fofs);
-	ext->blk = le32_to_cpu(i_ext.blk);
-	ext->len = le32_to_cpu(i_ext.len);
+	ext->fofs = le32_to_cpu(i_ext->fofs);
+	ext->blk = le32_to_cpu(i_ext->blk);
+	ext->len = le32_to_cpu(i_ext->len);
 }
 
 static inline void set_raw_extent(struct extent_info *ext,
@@ -471,11 +683,22 @@ static inline void set_extent_info(struct extent_info *ei, unsigned int fofs,
 	ei->len = len;
 }
 
-static inline bool __is_extent_same(struct extent_info *ei1,
-						struct extent_info *ei2)
+static inline bool __is_discard_mergeable(struct discard_info *back,
+						struct discard_info *front)
 {
-	return (ei1->fofs == ei2->fofs && ei1->blk == ei2->blk &&
-						ei1->len == ei2->len);
+	return back->lstart + back->len == front->lstart;
+}
+
+static inline bool __is_discard_back_mergeable(struct discard_info *cur,
+						struct discard_info *back)
+{
+	return __is_discard_mergeable(back, cur);
+}
+
+static inline bool __is_discard_front_mergeable(struct discard_info *cur,
+						struct discard_info *front)
+{
+	return __is_discard_mergeable(cur, front);
 }
 
 static inline bool __is_extent_mergeable(struct extent_info *back,
@@ -497,20 +720,30 @@ static inline bool __is_front_mergeable(struct extent_info *cur,
 	return __is_extent_mergeable(cur, front);
 }
 
-static inline void __try_update_largest_extent(struct extent_tree *et,
-						struct extent_node *en)
+extern void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync);
+static inline void __try_update_largest_extent(struct inode *inode,
+			struct extent_tree *et, struct extent_node *en)
 {
-	if (en->ei.len > et->largest.len)
+	if (en->ei.len > et->largest.len) {
 		et->largest = en->ei;
+		f2fs_mark_inode_dirty_sync(inode, true);
+	}
 }
 
+enum nid_list {
+	FREE_NID_LIST,
+	ALLOC_NID_LIST,
+	MAX_NID_LIST,
+};
+
 struct f2fs_nm_info {
 	block_t nat_blkaddr;		/* base disk address of NAT */
 	nid_t max_nid;			/* maximum possible node ids */
-	nid_t available_nids;		/* maximum available node ids */
+	nid_t available_nids;		/* # of available node ids */
 	nid_t next_scan_nid;		/* the next nid to be scanned */
 	unsigned int ram_thresh;	/* control the memory footprint */
 	unsigned int ra_nid_pages;	/* # of nid pages to be readaheaded */
+	unsigned int dirty_nats_ratio;	/* control dirty nats ratio threshold */
 
 	/* NAT cache management */
 	struct radix_tree_root nat_root;/* root of the nat entry cache */
@@ -519,16 +752,28 @@ struct f2fs_nm_info {
 	struct list_head nat_entries;	/* cached nat entry list (clean) */
 	unsigned int nat_cnt;		/* the # of cached nat entries */
 	unsigned int dirty_nat_cnt;	/* total num of nat entries in set */
+	unsigned int nat_blocks;	/* # of nat blocks */
 
 	/* free node ids management */
 	struct radix_tree_root free_nid_root;/* root of the free_nid cache */
-	struct list_head free_nid_list;	/* a list for free nids */
-	spinlock_t free_nid_list_lock;	/* protect free nid list */
-	unsigned int fcnt;		/* the number of free node id */
+	struct list_head nid_list[MAX_NID_LIST];/* lists for free nids */
+	unsigned int nid_cnt[MAX_NID_LIST];	/* the number of free node id */
+	spinlock_t nid_list_lock;	/* protect nid lists ops */
 	struct mutex build_lock;	/* lock for build free nids */
+	unsigned char (*free_nid_bitmap)[NAT_ENTRY_BITMAP_SIZE];
+	unsigned char *nat_block_bitmap;
+	unsigned short *free_nid_count;	/* free nid count of NAT block */
 
 	/* for checkpoint */
 	char *nat_bitmap;		/* NAT bitmap pointer */
+
+	unsigned int nat_bits_blocks;	/* # of nat bits blocks */
+	unsigned char *nat_bits;	/* NAT bits blocks */
+	unsigned char *full_nat_bits;	/* full NAT pages */
+	unsigned char *empty_nat_bits;	/* empty NAT pages */
+#ifdef CONFIG_F2FS_CHECK_FS
+	char *nat_bitmap_mir;		/* NAT bitmap mirror */
+#endif
 	int bitmap_size;		/* bitmap size */
 };
 
@@ -544,6 +789,9 @@ struct dnode_of_data {
 	nid_t nid;			/* node id of the direct node block */
 	unsigned int ofs_in_node;	/* data offset in the node page */
 	bool inode_page_locked;		/* inode page is locked or not */
+	bool node_changed;		/* is node block changed */
+	char cur_level;			/* level of hole node page */
+	char max_level;			/* level of current page located */
 	block_t	data_blkaddr;		/* block address of the node block */
 };
 
@@ -582,7 +830,6 @@ enum {
 	CURSEG_WARM_NODE,	/* direct node blocks of normal files */
 	CURSEG_COLD_NODE,	/* indirect node blocks */
 	NO_CHECK_TYPE,
-	CURSEG_DIRECT_IO,	/* to use for the direct IO path */
 };
 
 struct flush_cmd {
@@ -594,6 +841,8 @@ struct flush_cmd {
 struct flush_cmd_control {
 	struct task_struct *f2fs_issue_flush;	/* flush thread */
 	wait_queue_head_t flush_wait_queue;	/* waiting queue for wake-up */
+	atomic_t issued_flush;			/* # of issued flushes */
+	atomic_t issing_flush;			/* # of issing flushes */
 	struct llist_head issue_list;		/* list for command issue */
 	struct llist_node *dispatch_list;	/* list for command dispatch */
 };
@@ -616,11 +865,6 @@ struct f2fs_sm_info {
 	/* a threshold to reclaim prefree segments */
 	unsigned int rec_prefree_segments;
 
-	/* for small discard management */
-	struct list_head discard_list;		/* 4KB discard list */
-	int nr_discards;			/* # of discards in the list */
-	int max_discards;			/* max. discards to be issued */
-
 	/* for batched trimming */
 	unsigned int trim_sections;		/* # of sections to trim */
 
@@ -629,10 +873,13 @@ struct f2fs_sm_info {
 	unsigned int ipu_policy;	/* in-place-update policy */
 	unsigned int min_ipu_util;	/* in-place-update threshold */
 	unsigned int min_fsync_blocks;	/* threshold for fsync */
+	unsigned int min_hot_blocks;	/* threshold for hot block allocation */
 
 	/* for flush command control */
-	struct flush_cmd_control *cmd_control_info;
+	struct flush_cmd_control *fcc_info;
 
+	/* for discard command control */
+	struct discard_cmd_control *dcc_info;
 };
 
 /*
@@ -644,12 +891,16 @@ struct f2fs_sm_info {
  * f2fs monitors the number of several block types such as on-writeback,
  * dirty dentry blocks, dirty node blocks, and dirty meta blocks.
  */
+#define WB_DATA_TYPE(p)	(__is_cp_guaranteed(p) ? F2FS_WB_CP_DATA : F2FS_WB_DATA)
 enum count_type {
-	F2FS_WRITEBACK,
 	F2FS_DIRTY_DENTS,
+	F2FS_DIRTY_DATA,
 	F2FS_DIRTY_NODES,
 	F2FS_DIRTY_META,
 	F2FS_INMEM_PAGES,
+	F2FS_DIRTY_IMETA,
+	F2FS_WB_CP_DATA,
+	F2FS_WB_DATA,
 	NR_COUNT_TYPE,
 };
 
@@ -673,26 +924,89 @@ enum page_type {
 	META_FLUSH,
 	INMEM,		/* the below types are used by tracepoints only. */
 	INMEM_DROP,
+	INMEM_INVALIDATE,
+	INMEM_REVOKE,
 	IPU,
 	OPU,
 };
 
+enum temp_type {
+	HOT = 0,	/* must be zero for meta bio */
+	WARM,
+	COLD,
+	NR_TEMP_TYPE,
+};
+
+enum need_lock_type {
+	LOCK_REQ = 0,
+	LOCK_DONE,
+	LOCK_RETRY,
+};
+
+enum iostat_type {
+	APP_DIRECT_IO,			/* app direct IOs */
+	APP_BUFFERED_IO,		/* app buffered IOs */
+	APP_WRITE_IO,			/* app write IOs */
+	APP_MAPPED_IO,			/* app mapped IOs */
+	FS_DATA_IO,			/* data IOs from kworker/fsync/reclaimer */
+	FS_NODE_IO,			/* node IOs from kworker/fsync/reclaimer */
+	FS_META_IO,			/* meta IOs from kworker/reclaimer */
+	FS_GC_DATA_IO,			/* data IOs from forground gc */
+	FS_GC_NODE_IO,			/* node IOs from forground gc */
+	FS_CP_DATA_IO,			/* data IOs from checkpoint */
+	FS_CP_NODE_IO,			/* node IOs from checkpoint */
+	FS_CP_META_IO,			/* meta IOs from checkpoint */
+	FS_DISCARD,			/* discard */
+	NR_IO_TYPE,
+};
+
 struct f2fs_io_info {
 	struct f2fs_sb_info *sbi;	/* f2fs_sb_info pointer */
 	enum page_type type;	/* contains DATA/NODE/META/META_FLUSH */
-	int rw;			/* contains R/RS/W/WS with REQ_META/REQ_PRIO */
-	block_t blk_addr;	/* block address to be written */
+	enum temp_type temp;	/* contains HOT/WARM/COLD */
+	int op;			/* contains REQ_OP_ */
+	int op_flags;		/* req_flag_bits */
+	block_t new_blkaddr;	/* new block address to be written */
+	block_t old_blkaddr;	/* old block address before Cow */
 	struct page *page;	/* page to be written */
 	struct page *encrypted_page;	/* encrypted page */
+	struct list_head list;		/* serialize IOs */
+	bool submitted;		/* indicate IO submission */
+	int need_lock;		/* indicate we need to lock cp_rwsem */
+	bool in_list;		/* indicate fio is in io_list */
+	enum iostat_type io_type;	/* io type */
 };
 
-#define is_read_io(rw)	(((rw) & 1) == READ)
+#define is_read_io(rw) ((rw) == READ)
 struct f2fs_bio_info {
 	struct f2fs_sb_info *sbi;	/* f2fs superblock */
 	struct bio *bio;		/* bios to merge */
 	sector_t last_block_in_bio;	/* last block number */
 	struct f2fs_io_info fio;	/* store buffered io info. */
 	struct rw_semaphore io_rwsem;	/* blocking op for bio */
+	spinlock_t io_lock;		/* serialize DATA/NODE IOs */
+	struct list_head io_list;	/* track fios */
+};
+
+#define FDEV(i)				(sbi->devs[i])
+#define RDEV(i)				(raw_super->devs[i])
+struct f2fs_dev_info {
+	struct block_device *bdev;
+	char path[MAX_PATH_LEN];
+	unsigned int total_segments;
+	block_t start_blk;
+	block_t end_blk;
+#ifdef CONFIG_BLK_DEV_ZONED
+	unsigned int nr_blkz;			/* Total number of zones */
+	u8 *blkz_type;				/* Array of zones type */
+#endif
+};
+
+enum inode_type {
+	DIR_INODE,			/* for dirty dir inode */
+	FILE_INODE,			/* for dirty regular/symlink inode */
+	DIRTY_META,			/* for all dirtied inode metadata */
+	NR_INODE_TYPE,
 };
 
 /* for inner inode cache management */
@@ -709,14 +1023,27 @@ enum {
 	SBI_IS_CLOSE,				/* specify unmounting */
 	SBI_NEED_FSCK,				/* need fsck.f2fs to fix */
 	SBI_POR_DOING,				/* recovery is doing or not */
+	SBI_NEED_SB_WRITE,			/* need to recover superblock */
+	SBI_NEED_CP,				/* need to checkpoint */
+};
+
+enum {
+	CP_TIME,
+	REQ_TIME,
+	MAX_TIME,
 };
 
 struct f2fs_sb_info {
 	struct super_block *sb;			/* pointer to VFS super block */
 	struct proc_dir_entry *s_proc;		/* proc entry */
-	struct buffer_head *raw_super_buf;	/* buffer head of raw sb */
 	struct f2fs_super_block *raw_super;	/* raw super block pointer */
-	int s_flag;				/* flags for sbi */
+	int valid_super_block;			/* valid super block no */
+	unsigned long s_flag;				/* flags for sbi */
+
+#ifdef CONFIG_BLK_DEV_ZONED
+	unsigned int blocks_per_blkz;		/* F2FS blocks per zone */
+	unsigned int log_blocks_per_blkz;	/* log2 F2FS blocks per zone */
+#endif
 
 	/* for node-related operations */
 	struct f2fs_nm_info *nm_info;		/* node manager */
@@ -726,34 +1053,42 @@ struct f2fs_sb_info {
 	struct f2fs_sm_info *sm_info;		/* segment manager */
 
 	/* for bio operations */
-	struct f2fs_bio_info read_io;			/* for read bios */
-	struct f2fs_bio_info write_io[NR_PAGE_TYPE];	/* for write bios */
+	struct f2fs_bio_info *write_io[NR_PAGE_TYPE];	/* for write bios */
+	struct mutex wio_mutex[NR_PAGE_TYPE - 1][NR_TEMP_TYPE];
+						/* bio ordering for NODE/DATA */
+	int write_io_size_bits;			/* Write IO size bits */
+	mempool_t *write_io_dummy;		/* Dummy pages */
 
 	/* for checkpoint */
 	struct f2fs_checkpoint *ckpt;		/* raw checkpoint pointer */
+	int cur_cp_pack;			/* remain current cp pack */
+	spinlock_t cp_lock;			/* for flag in ckpt */
 	struct inode *meta_inode;		/* cache meta blocks */
 	struct mutex cp_mutex;			/* checkpoint procedure lock */
 	struct rw_semaphore cp_rwsem;		/* blocking FS operations */
 	struct rw_semaphore node_write;		/* locking node writes */
-	struct mutex writepages;		/* mutex for writepages() */
+	struct rw_semaphore node_change;	/* locking node change */
 	wait_queue_head_t cp_wait;
-	long cp_expires, cp_interval;		/* next expected periodic cp */
+	unsigned long last_time[MAX_TIME];	/* to store time in jiffies */
+	long interval_time[MAX_TIME];		/* to store thresholds */
 
 	struct inode_management im[MAX_INO_ENTRY];      /* manage inode cache */
 
 	/* for orphan inode, use 0'th array */
 	unsigned int max_orphans;		/* max orphan inodes */
 
-	/* for directory inode management */
-	struct list_head dir_inode_list;	/* dir inode list */
-	spinlock_t dir_inode_lock;		/* for dir inode list lock */
+	/* for inode management */
+	struct list_head inode_list[NR_INODE_TYPE];	/* dirty inode list */
+	spinlock_t inode_lock[NR_INODE_TYPE];	/* for dirty inode list lock */
 
 	/* for extent tree cache */
 	struct radix_tree_root extent_tree_root;/* cache extent cache entries */
-	struct rw_semaphore extent_tree_lock;	/* locking extent radix tree */
+	struct mutex extent_tree_lock;	/* locking extent radix tree */
 	struct list_head extent_list;		/* lru list for shrinker */
 	spinlock_t extent_lock;			/* locking extent lru list */
-	int total_ext_tree;			/* extent tree count */
+	atomic_t total_ext_tree;		/* extent tree count */
+	struct list_head zombie_list;		/* extent zombie tree list */
+	atomic_t total_zombie_tree;		/* extent zombie tree count */
 	atomic_t total_ext_node;		/* extent info count */
 
 	/* basic filesystem units */
@@ -770,17 +1105,28 @@ struct f2fs_sb_info {
 	unsigned int total_sections;		/* total section count */
 	unsigned int total_node_count;		/* total node block count */
 	unsigned int total_valid_node_count;	/* valid node block count */
-	unsigned int total_valid_inode_count;	/* valid inode count */
+	loff_t max_file_blocks;			/* max block index of file */
 	int active_logs;			/* # of active logs */
 	int dir_level;				/* directory level */
 
 	block_t user_block_count;		/* # of user blocks */
 	block_t total_valid_block_count;	/* # of valid blocks */
-	block_t alloc_valid_block_count;	/* # of allocated blocks */
 	block_t discard_blks;			/* discard command candidats */
 	block_t last_valid_block_count;		/* for recovery */
+	block_t reserved_blocks;		/* configurable reserved blocks */
+
 	u32 s_next_generation;			/* for NFS support */
-	atomic_t nr_pages[NR_COUNT_TYPE];	/* # of pages, see count_type */
+
+	/* # of pages, see count_type */
+	atomic_t nr_pages[NR_COUNT_TYPE];
+	/* # of allocated blocks */
+	struct percpu_counter alloc_valid_block_count;
+
+	/* writeback control */
+	atomic_t wb_sync_req;			/* count # of WB_SYNC threads */
+
+	/* valid inode count */
+	struct percpu_counter total_valid_inode_count;
 
 	struct f2fs_mount_info mount_opt;	/* mount options */
 
@@ -789,6 +1135,9 @@ struct f2fs_sb_info {
 	struct f2fs_gc_kthread	*gc_thread;	/* GC thread */
 	unsigned int cur_victim_sec;		/* current victim section num */
 
+	/* threshold for converting bg victims for fg */
+	u64 fggc_threshold;
+
 	/* maximum # of trials to find a victim segment for SSR and GC */
 	unsigned int max_victim_search;
 
@@ -808,25 +1157,159 @@ struct f2fs_sb_info {
 	atomic_t inline_xattr;			/* # of inline_xattr inodes */
 	atomic_t inline_inode;			/* # of inline_data inodes */
 	atomic_t inline_dir;			/* # of inline_dentry inodes */
+	atomic_t aw_cnt;			/* # of atomic writes */
+	atomic_t vw_cnt;			/* # of volatile writes */
+	atomic_t max_aw_cnt;			/* max # of atomic writes */
+	atomic_t max_vw_cnt;			/* max # of volatile writes */
 	int bg_gc;				/* background gc calls */
-	unsigned int n_dirty_dirs;		/* # of dir inodes */
+	unsigned int ndirty_inode[NR_INODE_TYPE];	/* # of dirty inodes */
 #endif
-	unsigned int last_victim[2];		/* last victim segment # */
 	spinlock_t stat_lock;			/* lock for stat operations */
 
+	/* For app/fs IO statistics */
+	spinlock_t iostat_lock;
+	unsigned long long write_iostat[NR_IO_TYPE];
+	bool iostat_enable;
+
 	/* For sysfs suppport */
 	struct kobject s_kobj;
 	struct completion s_kobj_unregister;
 
 	/* For shrinker support */
 	struct list_head s_list;
+	int s_ndevs;				/* number of devices */
+	struct f2fs_dev_info *devs;		/* for device list */
 	struct mutex umount_mutex;
 	unsigned int shrinker_run_no;
+
+	/* For write statistics */
+	u64 sectors_written_start;
+	u64 kbytes_written;
+
+	/* Reference to checksum algorithm driver via cryptoapi */
+	struct crypto_shash *s_chksum_driver;
+
+	/* Precomputed FS UUID checksum for seeding other checksums */
+	__u32 s_chksum_seed;
+
+	/* For fault injection */
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	struct f2fs_fault_info fault_info;
+#endif
+
+#ifdef CONFIG_QUOTA
+	/* Names of quota files with journalled quota */
+	char *s_qf_names[MAXQUOTAS];
+	int s_jquota_fmt;			/* Format of quota to use */
+#endif
 };
 
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+#define f2fs_show_injection_info(type)				\
+	printk("%sF2FS-fs : inject %s in %s of %pF\n",		\
+		KERN_INFO, fault_name[type],			\
+		__func__, __builtin_return_address(0))
+static inline bool time_to_inject(struct f2fs_sb_info *sbi, int type)
+{
+	struct f2fs_fault_info *ffi = &sbi->fault_info;
+
+	if (!ffi->inject_rate)
+		return false;
+
+	if (!IS_FAULT_SET(ffi, type))
+		return false;
+
+	atomic_inc(&ffi->inject_ops);
+	if (atomic_read(&ffi->inject_ops) >= ffi->inject_rate) {
+		atomic_set(&ffi->inject_ops, 0);
+		return true;
+	}
+	return false;
+}
+#endif
+
+/* For write statistics. Suppose sector size is 512 bytes,
+ * and the return value is in kbytes. s is of struct f2fs_sb_info.
+ */
+#define BD_PART_WRITTEN(s)						 \
+(((u64)part_stat_read((s)->sb->s_bdev->bd_part, sectors[1]) -		 \
+		(s)->sectors_written_start) >> 1)
+
+static inline void f2fs_update_time(struct f2fs_sb_info *sbi, int type)
+{
+	sbi->last_time[type] = jiffies;
+}
+
+static inline bool f2fs_time_over(struct f2fs_sb_info *sbi, int type)
+{
+	struct timespec ts = {sbi->interval_time[type], 0};
+	unsigned long interval = timespec_to_jiffies(&ts);
+
+	return time_after(jiffies, sbi->last_time[type] + interval);
+}
+
+static inline bool is_idle(struct f2fs_sb_info *sbi)
+{
+	struct block_device *bdev = sbi->sb->s_bdev;
+	struct request_queue *q = bdev_get_queue(bdev);
+	struct request_list *rl = &q->root_rl;
+
+	if (rl->count[BLK_RW_SYNC] || rl->count[BLK_RW_ASYNC])
+		return 0;
+
+	return f2fs_time_over(sbi, REQ_TIME);
+}
+
 /*
  * Inline functions
  */
+static inline u32 f2fs_crc32(struct f2fs_sb_info *sbi, const void *address,
+			   unsigned int length)
+{
+	SHASH_DESC_ON_STACK(shash, sbi->s_chksum_driver);
+	u32 *ctx = (u32 *)shash_desc_ctx(shash);
+	u32 retval;
+	int err;
+
+	shash->tfm = sbi->s_chksum_driver;
+	shash->flags = 0;
+	*ctx = F2FS_SUPER_MAGIC;
+
+	err = crypto_shash_update(shash, address, length);
+	BUG_ON(err);
+
+	retval = *ctx;
+	barrier_data(ctx);
+	return retval;
+}
+
+static inline bool f2fs_crc_valid(struct f2fs_sb_info *sbi, __u32 blk_crc,
+				  void *buf, size_t buf_size)
+{
+	return f2fs_crc32(sbi, buf, buf_size) == blk_crc;
+}
+
+static inline u32 f2fs_chksum(struct f2fs_sb_info *sbi, u32 crc,
+			      const void *address, unsigned int length)
+{
+	struct {
+		struct shash_desc shash;
+		char ctx[4];
+	} desc;
+	int err;
+
+	BUG_ON(crypto_shash_descsize(sbi->s_chksum_driver) != sizeof(desc.ctx));
+
+	desc.shash.tfm = sbi->s_chksum_driver;
+	desc.shash.flags = 0;
+	*(u32 *)desc.ctx = crc;
+
+	err = crypto_shash_update(&desc.shash, address, length);
+	BUG_ON(err);
+
+	return *(u32 *)desc.ctx;
+}
+
 static inline struct f2fs_inode_info *F2FS_I(struct inode *inode)
 {
 	return container_of(inode, struct f2fs_inode_info, vfs_inode);
@@ -909,17 +1392,17 @@ static inline struct address_space *NODE_MAPPING(struct f2fs_sb_info *sbi)
 
 static inline bool is_sbi_flag_set(struct f2fs_sb_info *sbi, unsigned int type)
 {
-	return sbi->s_flag & (0x01 << type);
+	return test_bit(type, &sbi->s_flag);
 }
 
 static inline void set_sbi_flag(struct f2fs_sb_info *sbi, unsigned int type)
 {
-	sbi->s_flag |= (0x01 << type);
+	set_bit(type, &sbi->s_flag);
 }
 
 static inline void clear_sbi_flag(struct f2fs_sb_info *sbi, unsigned int type)
 {
-	sbi->s_flag &= ~(0x01 << type);
+	clear_bit(type, &sbi->s_flag);
 }
 
 static inline unsigned long long cur_cp_version(struct f2fs_checkpoint *cp)
@@ -927,31 +1410,93 @@ static inline unsigned long long cur_cp_version(struct f2fs_checkpoint *cp)
 	return le64_to_cpu(cp->checkpoint_ver);
 }
 
-static inline bool is_set_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f)
+static inline __u64 cur_cp_crc(struct f2fs_checkpoint *cp)
+{
+	size_t crc_offset = le32_to_cpu(cp->checksum_offset);
+	return le32_to_cpu(*((__le32 *)((unsigned char *)cp + crc_offset)));
+}
+
+static inline bool __is_set_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f)
 {
 	unsigned int ckpt_flags = le32_to_cpu(cp->ckpt_flags);
+
 	return ckpt_flags & f;
 }
 
-static inline void set_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f)
+static inline bool is_set_ckpt_flags(struct f2fs_sb_info *sbi, unsigned int f)
 {
-	unsigned int ckpt_flags = le32_to_cpu(cp->ckpt_flags);
+	return __is_set_ckpt_flags(F2FS_CKPT(sbi), f);
+}
+
+static inline void __set_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f)
+{
+	unsigned int ckpt_flags;
+
+	ckpt_flags = le32_to_cpu(cp->ckpt_flags);
 	ckpt_flags |= f;
 	cp->ckpt_flags = cpu_to_le32(ckpt_flags);
 }
 
-static inline void clear_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f)
+static inline void set_ckpt_flags(struct f2fs_sb_info *sbi, unsigned int f)
 {
-	unsigned int ckpt_flags = le32_to_cpu(cp->ckpt_flags);
+	unsigned long flags;
+
+	spin_lock_irqsave(&sbi->cp_lock, flags);
+	__set_ckpt_flags(F2FS_CKPT(sbi), f);
+	spin_unlock_irqrestore(&sbi->cp_lock, flags);
+}
+
+static inline void __clear_ckpt_flags(struct f2fs_checkpoint *cp, unsigned int f)
+{
+	unsigned int ckpt_flags;
+
+	ckpt_flags = le32_to_cpu(cp->ckpt_flags);
 	ckpt_flags &= (~f);
 	cp->ckpt_flags = cpu_to_le32(ckpt_flags);
 }
 
+static inline void clear_ckpt_flags(struct f2fs_sb_info *sbi, unsigned int f)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&sbi->cp_lock, flags);
+	__clear_ckpt_flags(F2FS_CKPT(sbi), f);
+	spin_unlock_irqrestore(&sbi->cp_lock, flags);
+}
+
+static inline void disable_nat_bits(struct f2fs_sb_info *sbi, bool lock)
+{
+	unsigned long flags;
+
+	set_sbi_flag(sbi, SBI_NEED_FSCK);
+
+	if (lock)
+		spin_lock_irqsave(&sbi->cp_lock, flags);
+	__clear_ckpt_flags(F2FS_CKPT(sbi), CP_NAT_BITS_FLAG);
+	kfree(NM_I(sbi)->nat_bits);
+	NM_I(sbi)->nat_bits = NULL;
+	if (lock)
+		spin_unlock_irqrestore(&sbi->cp_lock, flags);
+}
+
+static inline bool enabled_nat_bits(struct f2fs_sb_info *sbi,
+					struct cp_control *cpc)
+{
+	bool set = is_set_ckpt_flags(sbi, CP_NAT_BITS_FLAG);
+
+	return (cpc) ? (cpc->reason & CP_UMOUNT) && set : set;
+}
+
 static inline void f2fs_lock_op(struct f2fs_sb_info *sbi)
 {
 	down_read(&sbi->cp_rwsem);
 }
 
+static inline int f2fs_trylock_op(struct f2fs_sb_info *sbi)
+{
+	return down_read_trylock(&sbi->cp_rwsem);
+}
+
 static inline void f2fs_unlock_op(struct f2fs_sb_info *sbi)
 {
 	up_read(&sbi->cp_rwsem);
@@ -959,7 +1504,7 @@ static inline void f2fs_unlock_op(struct f2fs_sb_info *sbi)
 
 static inline void f2fs_lock_all(struct f2fs_sb_info *sbi)
 {
-	f2fs_down_write(&sbi->cp_rwsem, &sbi->cp_mutex);
+	down_write(&sbi->cp_rwsem);
 }
 
 static inline void f2fs_unlock_all(struct f2fs_sb_info *sbi)
@@ -980,13 +1525,13 @@ static inline int __get_cp_reason(struct f2fs_sb_info *sbi)
 
 static inline bool __remain_node_summaries(int reason)
 {
-	return (reason == CP_UMOUNT || reason == CP_FASTBOOT);
+	return (reason & (CP_UMOUNT | CP_FASTBOOT));
 }
 
 static inline bool __exist_node_summaries(struct f2fs_sb_info *sbi)
 {
-	return (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG) ||
-			is_set_ckpt_flags(F2FS_CKPT(sbi), CP_FASTBOOT_FLAG));
+	return (is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG) ||
+			is_set_ckpt_flags(sbi, CP_FASTBOOT_FLAG));
 }
 
 /*
@@ -1001,17 +1546,14 @@ static inline int check_nid_range(struct f2fs_sb_info *sbi, nid_t nid)
 	return 0;
 }
 
-#define F2FS_DEFAULT_ALLOCATED_BLOCKS	1
-
 /*
  * Check whether the inode has blocks or not
  */
 static inline int F2FS_HAS_BLOCKS(struct inode *inode)
 {
-	if (F2FS_I(inode)->i_xattr_nid)
-		return inode->i_blocks > F2FS_DEFAULT_ALLOCATED_BLOCKS + 1;
-	else
-		return inode->i_blocks > F2FS_DEFAULT_ALLOCATED_BLOCKS;
+	block_t xattr_block = F2FS_I(inode)->i_xattr_nid ? 1 : 0;
+
+	return (inode->i_blocks >> F2FS_LOG_SECTORS_PER_BLOCK) > xattr_block;
 }
 
 static inline bool f2fs_has_xattr_block(unsigned int ofs)
@@ -1019,48 +1561,87 @@ static inline bool f2fs_has_xattr_block(unsigned int ofs)
 	return ofs == XATTR_NODE_OFFSET;
 }
 
-static inline bool inc_valid_block_count(struct f2fs_sb_info *sbi,
-				 struct inode *inode, blkcnt_t count)
+static inline void f2fs_i_blocks_write(struct inode *, block_t, bool, bool);
+static inline int inc_valid_block_count(struct f2fs_sb_info *sbi,
+				 struct inode *inode, blkcnt_t *count)
 {
-	block_t	valid_block_count;
+	blkcnt_t diff = 0, release = 0;
+	block_t avail_user_block_count;
+	int ret;
+
+	ret = dquot_reserve_block(inode, *count);
+	if (ret)
+		return ret;
+
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_BLOCK)) {
+		f2fs_show_injection_info(FAULT_BLOCK);
+		release = *count;
+		goto enospc;
+	}
+#endif
+	/*
+	 * let's increase this in prior to actual block count change in order
+	 * for f2fs_sync_file to avoid data races when deciding checkpoint.
+	 */
+	percpu_counter_add(&sbi->alloc_valid_block_count, (*count));
 
 	spin_lock(&sbi->stat_lock);
-	valid_block_count =
-		sbi->total_valid_block_count + (block_t)count;
-	if (unlikely(valid_block_count > sbi->user_block_count)) {
-		spin_unlock(&sbi->stat_lock);
-		return false;
+	sbi->total_valid_block_count += (block_t)(*count);
+	avail_user_block_count = sbi->user_block_count - sbi->reserved_blocks;
+	if (unlikely(sbi->total_valid_block_count > avail_user_block_count)) {
+		diff = sbi->total_valid_block_count - avail_user_block_count;
+		*count -= diff;
+		release = diff;
+		sbi->total_valid_block_count = avail_user_block_count;
+		if (!*count) {
+			spin_unlock(&sbi->stat_lock);
+			percpu_counter_sub(&sbi->alloc_valid_block_count, diff);
+			goto enospc;
+		}
 	}
-	inode->i_blocks += count;
-	sbi->total_valid_block_count = valid_block_count;
-	sbi->alloc_valid_block_count += (block_t)count;
 	spin_unlock(&sbi->stat_lock);
-	return true;
+
+	if (release)
+		dquot_release_reservation_block(inode, release);
+	f2fs_i_blocks_write(inode, *count, true, true);
+	return 0;
+
+enospc:
+	dquot_release_reservation_block(inode, release);
+	return -ENOSPC;
 }
 
 static inline void dec_valid_block_count(struct f2fs_sb_info *sbi,
 						struct inode *inode,
-						blkcnt_t count)
+						block_t count)
 {
+	blkcnt_t sectors = count << F2FS_LOG_SECTORS_PER_BLOCK;
+
 	spin_lock(&sbi->stat_lock);
 	f2fs_bug_on(sbi, sbi->total_valid_block_count < (block_t) count);
-	f2fs_bug_on(sbi, inode->i_blocks < count);
-	inode->i_blocks -= count;
+	f2fs_bug_on(sbi, inode->i_blocks < sectors);
 	sbi->total_valid_block_count -= (block_t)count;
 	spin_unlock(&sbi->stat_lock);
+	f2fs_i_blocks_write(inode, count, false, true);
 }
 
 static inline void inc_page_count(struct f2fs_sb_info *sbi, int count_type)
 {
 	atomic_inc(&sbi->nr_pages[count_type]);
+
+	if (count_type == F2FS_DIRTY_DATA || count_type == F2FS_INMEM_PAGES ||
+		count_type == F2FS_WB_CP_DATA || count_type == F2FS_WB_DATA)
+		return;
+
 	set_sbi_flag(sbi, SBI_IS_DIRTY);
 }
 
 static inline void inode_inc_dirty_pages(struct inode *inode)
 {
 	atomic_inc(&F2FS_I(inode)->dirty_pages);
-	if (S_ISDIR(inode->i_mode))
-		inc_page_count(F2FS_I_SB(inode), F2FS_DIRTY_DENTS);
+	inc_page_count(F2FS_I_SB(inode), S_ISDIR(inode->i_mode) ?
+				F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA);
 }
 
 static inline void dec_page_count(struct f2fs_sb_info *sbi, int count_type)
@@ -1075,12 +1656,11 @@ static inline void inode_dec_dirty_pages(struct inode *inode)
 		return;
 
 	atomic_dec(&F2FS_I(inode)->dirty_pages);
-
-	if (S_ISDIR(inode->i_mode))
-		dec_page_count(F2FS_I_SB(inode), F2FS_DIRTY_DENTS);
+	dec_page_count(F2FS_I_SB(inode), S_ISDIR(inode->i_mode) ?
+				F2FS_DIRTY_DENTS : F2FS_DIRTY_DATA);
 }
 
-static inline int get_pages(struct f2fs_sb_info *sbi, int count_type)
+static inline s64 get_pages(struct f2fs_sb_info *sbi, int count_type)
 {
 	return atomic_read(&sbi->nr_pages[count_type]);
 }
@@ -1092,10 +1672,11 @@ static inline int get_dirty_pages(struct inode *inode)
 
 static inline int get_blocktype_secs(struct f2fs_sb_info *sbi, int block_type)
 {
-	unsigned int pages_per_sec = sbi->segs_per_sec *
-					(1 << sbi->log_blocks_per_seg);
-	return ((get_pages(sbi, block_type) + pages_per_sec - 1)
-			>> sbi->log_blocks_per_seg) / sbi->segs_per_sec;
+	unsigned int pages_per_sec = sbi->segs_per_sec * sbi->blocks_per_seg;
+	unsigned int segs = (get_pages(sbi, block_type) + pages_per_sec - 1) >>
+						sbi->log_blocks_per_seg;
+
+	return segs / sbi->segs_per_sec;
 }
 
 static inline block_t valid_user_blocks(struct f2fs_sb_info *sbi)
@@ -1103,6 +1684,11 @@ static inline block_t valid_user_blocks(struct f2fs_sb_info *sbi)
 	return sbi->total_valid_block_count;
 }
 
+static inline block_t discard_blocks(struct f2fs_sb_info *sbi)
+{
+	return sbi->discard_blks;
+}
+
 static inline unsigned long __bitmap_size(struct f2fs_sb_info *sbi, int flag)
 {
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
@@ -1140,72 +1726,96 @@ static inline void *__bitmap_ptr(struct f2fs_sb_info *sbi, int flag)
 
 static inline block_t __start_cp_addr(struct f2fs_sb_info *sbi)
 {
-	block_t start_addr;
-	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
-	unsigned long long ckpt_version = cur_cp_version(ckpt);
-
-	start_addr = le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_blkaddr);
+	block_t start_addr = le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_blkaddr);
 
-	/*
-	 * odd numbered checkpoint should at cp segment 0
-	 * and even segment must be at cp segment 1
-	 */
-	if (!(ckpt_version & 1))
+	if (sbi->cur_cp_pack == 2)
 		start_addr += sbi->blocks_per_seg;
+	return start_addr;
+}
+
+static inline block_t __start_cp_next_addr(struct f2fs_sb_info *sbi)
+{
+	block_t start_addr = le32_to_cpu(F2FS_RAW_SUPER(sbi)->cp_blkaddr);
 
+	if (sbi->cur_cp_pack == 1)
+		start_addr += sbi->blocks_per_seg;
 	return start_addr;
 }
 
+static inline void __set_cp_next_pack(struct f2fs_sb_info *sbi)
+{
+	sbi->cur_cp_pack = (sbi->cur_cp_pack == 1) ? 2 : 1;
+}
+
 static inline block_t __start_sum_addr(struct f2fs_sb_info *sbi)
 {
 	return le32_to_cpu(F2FS_CKPT(sbi)->cp_pack_start_sum);
 }
 
-static inline bool inc_valid_node_count(struct f2fs_sb_info *sbi,
-						struct inode *inode)
+static inline int inc_valid_node_count(struct f2fs_sb_info *sbi,
+					struct inode *inode, bool is_inode)
 {
 	block_t	valid_block_count;
 	unsigned int valid_node_count;
+	bool quota = inode && !is_inode;
+
+	if (quota) {
+		int ret = dquot_reserve_block(inode, 1);
+		if (ret)
+			return ret;
+	}
 
 	spin_lock(&sbi->stat_lock);
 
 	valid_block_count = sbi->total_valid_block_count + 1;
-	if (unlikely(valid_block_count > sbi->user_block_count)) {
+	if (unlikely(valid_block_count + sbi->reserved_blocks >
+						sbi->user_block_count)) {
 		spin_unlock(&sbi->stat_lock);
-		return false;
+		goto enospc;
 	}
 
 	valid_node_count = sbi->total_valid_node_count + 1;
 	if (unlikely(valid_node_count > sbi->total_node_count)) {
 		spin_unlock(&sbi->stat_lock);
-		return false;
+		goto enospc;
 	}
 
-	if (inode)
-		inode->i_blocks++;
-
-	sbi->alloc_valid_block_count++;
 	sbi->total_valid_node_count++;
 	sbi->total_valid_block_count++;
 	spin_unlock(&sbi->stat_lock);
 
-	return true;
+	if (inode) {
+		if (is_inode)
+			f2fs_mark_inode_dirty_sync(inode, true);
+		else
+			f2fs_i_blocks_write(inode, 1, true, true);
+	}
+
+	percpu_counter_inc(&sbi->alloc_valid_block_count);
+	return 0;
+
+enospc:
+	if (quota)
+		dquot_release_reservation_block(inode, 1);
+	return -ENOSPC;
 }
 
 static inline void dec_valid_node_count(struct f2fs_sb_info *sbi,
-						struct inode *inode)
+					struct inode *inode, bool is_inode)
 {
 	spin_lock(&sbi->stat_lock);
 
 	f2fs_bug_on(sbi, !sbi->total_valid_block_count);
 	f2fs_bug_on(sbi, !sbi->total_valid_node_count);
-	f2fs_bug_on(sbi, !inode->i_blocks);
+	f2fs_bug_on(sbi, !is_inode && !inode->i_blocks);
 
-	inode->i_blocks--;
 	sbi->total_valid_node_count--;
 	sbi->total_valid_block_count--;
 
 	spin_unlock(&sbi->stat_lock);
+
+	if (!is_inode)
+		f2fs_i_blocks_write(inode, 1, false, true);
 }
 
 static inline unsigned int valid_node_count(struct f2fs_sb_info *sbi)
@@ -1215,28 +1825,33 @@ static inline unsigned int valid_node_count(struct f2fs_sb_info *sbi)
 
 static inline void inc_valid_inode_count(struct f2fs_sb_info *sbi)
 {
-	spin_lock(&sbi->stat_lock);
-	f2fs_bug_on(sbi, sbi->total_valid_inode_count == sbi->total_node_count);
-	sbi->total_valid_inode_count++;
-	spin_unlock(&sbi->stat_lock);
+	percpu_counter_inc(&sbi->total_valid_inode_count);
 }
 
 static inline void dec_valid_inode_count(struct f2fs_sb_info *sbi)
 {
-	spin_lock(&sbi->stat_lock);
-	f2fs_bug_on(sbi, !sbi->total_valid_inode_count);
-	sbi->total_valid_inode_count--;
-	spin_unlock(&sbi->stat_lock);
+	percpu_counter_dec(&sbi->total_valid_inode_count);
 }
 
-static inline unsigned int valid_inode_count(struct f2fs_sb_info *sbi)
+static inline s64 valid_inode_count(struct f2fs_sb_info *sbi)
 {
-	return sbi->total_valid_inode_count;
+	return percpu_counter_sum_positive(&sbi->total_valid_inode_count);
 }
 
 static inline struct page *f2fs_grab_cache_page(struct address_space *mapping,
 						pgoff_t index, bool for_write)
 {
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	struct page *page = find_lock_page(mapping, index);
+
+	if (page)
+		return page;
+
+	if (time_to_inject(F2FS_M_SB(mapping), FAULT_PAGE_ALLOC)) {
+		f2fs_show_injection_info(FAULT_PAGE_ALLOC);
+		return NULL;
+	}
+#endif
 	if (!for_write)
 		return grab_cache_page(mapping, index);
 	return grab_cache_page_write_begin(mapping, index, AOP_FLAG_NOFS);
@@ -1261,7 +1876,7 @@ static inline void f2fs_put_page(struct page *page, int unlock)
 		f2fs_bug_on(F2FS_P_SB(page), !PageLocked(page));
 		unlock_page(page);
 	}
-	page_cache_release(page);
+	put_page(page);
 }
 
 static inline void f2fs_put_dnode(struct dnode_of_data *dn)
@@ -1314,22 +1929,42 @@ static inline void f2fs_radix_tree_insert(struct radix_tree_root *root,
 static inline bool IS_INODE(struct page *page)
 {
 	struct f2fs_node *p = F2FS_NODE(page);
+
 	return RAW_IS_INODE(p);
 }
 
+static inline int offset_in_addr(struct f2fs_inode *i)
+{
+	return (i->i_inline & F2FS_EXTRA_ATTR) ?
+			(le16_to_cpu(i->i_extra_isize) / sizeof(__le32)) : 0;
+}
+
 static inline __le32 *blkaddr_in_node(struct f2fs_node *node)
 {
 	return RAW_IS_INODE(node) ? node->i.i_addr : node->dn.addr;
 }
 
-static inline block_t datablock_addr(struct page *node_page,
-		unsigned int offset)
+static inline int f2fs_has_extra_attr(struct inode *inode);
+static inline block_t datablock_addr(struct inode *inode,
+			struct page *node_page, unsigned int offset)
 {
 	struct f2fs_node *raw_node;
 	__le32 *addr_array;
+	int base = 0;
+	bool is_inode = IS_INODE(node_page);
+
 	raw_node = F2FS_NODE(node_page);
+
+	/* from GC path only */
+	if (!inode) {
+		if (is_inode)
+			base = offset_in_addr(&raw_node->i);
+	} else if (f2fs_has_extra_attr(inode) && is_inode) {
+		base = get_extra_isize(inode);
+	}
+
 	addr_array = blkaddr_in_node(raw_node);
-	return le32_to_cpu(addr_array[offset]);
+	return le32_to_cpu(addr_array[base + offset]);
 }
 
 static inline int f2fs_test_bit(unsigned int nr, char *addr)
@@ -1392,17 +2027,30 @@ static inline void f2fs_change_bit(unsigned int nr, char *addr)
 	*addr ^= mask;
 }
 
+#define F2FS_REG_FLMASK		(~(FS_DIRSYNC_FL | FS_TOPDIR_FL))
+#define F2FS_OTHER_FLMASK	(FS_NODUMP_FL | FS_NOATIME_FL)
+#define F2FS_FL_INHERITED	(FS_PROJINHERIT_FL)
+
+static inline __u32 f2fs_mask_flags(umode_t mode, __u32 flags)
+{
+	if (S_ISDIR(mode))
+		return flags;
+	else if (S_ISREG(mode))
+		return flags & F2FS_REG_FLMASK;
+	else
+		return flags & F2FS_OTHER_FLMASK;
+}
+
 /* used for f2fs_inode_info->flags */
 enum {
 	FI_NEW_INODE,		/* indicate newly allocated inode */
 	FI_DIRTY_INODE,		/* indicate inode is dirty or not */
+	FI_AUTO_RECOVER,	/* indicate inode is recoverable */
 	FI_DIRTY_DIR,		/* indicate directory has dirty pages */
 	FI_INC_LINK,		/* need to increment i_nlink */
 	FI_ACL_MODE,		/* indicate acl mode */
 	FI_NO_ALLOC,		/* should not allocate any blocks */
 	FI_FREE_NID,		/* free allocated nide */
-	FI_UPDATE_DIR,		/* should update inode block for consistency */
-	FI_DELAY_IPUT,		/* used for the recovery */
 	FI_NO_EXTENT,		/* not to use the extent cache */
 	FI_INLINE_XATTR,	/* used for inline xattr */
 	FI_INLINE_DATA,		/* used for inline data*/
@@ -1411,83 +2059,180 @@ enum {
 	FI_UPDATE_WRITE,	/* inode has in-place-update data */
 	FI_NEED_IPU,		/* used for ipu per file */
 	FI_ATOMIC_FILE,		/* indicate atomic file */
+	FI_ATOMIC_COMMIT,	/* indicate the state of atomical committing */
 	FI_VOLATILE_FILE,	/* indicate volatile file */
 	FI_FIRST_BLOCK_WRITTEN,	/* indicate #0 data block was written */
 	FI_DROP_CACHE,		/* drop dirty page cache */
 	FI_DATA_EXIST,		/* indicate data exists */
 	FI_INLINE_DOTS,		/* indicate inline dot dentries */
+	FI_DO_DEFRAG,		/* indicate defragment is running */
+	FI_DIRTY_FILE,		/* indicate regular/symlink has dirty pages */
+	FI_NO_PREALLOC,		/* indicate skipped preallocated blocks */
+	FI_HOT_DATA,		/* indicate file is hot */
+	FI_EXTRA_ATTR,		/* indicate file has extra attribute */
+	FI_PROJ_INHERIT,	/* indicate file inherits projectid */
 };
 
-static inline void set_inode_flag(struct f2fs_inode_info *fi, int flag)
+static inline void __mark_inode_dirty_flag(struct inode *inode,
+						int flag, bool set)
+{
+	switch (flag) {
+	case FI_INLINE_XATTR:
+	case FI_INLINE_DATA:
+	case FI_INLINE_DENTRY:
+		if (set)
+			return;
+	case FI_DATA_EXIST:
+	case FI_INLINE_DOTS:
+		f2fs_mark_inode_dirty_sync(inode, true);
+	}
+}
+
+static inline void set_inode_flag(struct inode *inode, int flag)
 {
-	if (!test_bit(flag, &fi->flags))
-		set_bit(flag, &fi->flags);
+	if (!test_bit(flag, &F2FS_I(inode)->flags))
+		set_bit(flag, &F2FS_I(inode)->flags);
+	__mark_inode_dirty_flag(inode, flag, true);
 }
 
-static inline int is_inode_flag_set(struct f2fs_inode_info *fi, int flag)
+static inline int is_inode_flag_set(struct inode *inode, int flag)
 {
-	return test_bit(flag, &fi->flags);
+	return test_bit(flag, &F2FS_I(inode)->flags);
 }
 
-static inline void clear_inode_flag(struct f2fs_inode_info *fi, int flag)
+static inline void clear_inode_flag(struct inode *inode, int flag)
 {
-	if (test_bit(flag, &fi->flags))
-		clear_bit(flag, &fi->flags);
+	if (test_bit(flag, &F2FS_I(inode)->flags))
+		clear_bit(flag, &F2FS_I(inode)->flags);
+	__mark_inode_dirty_flag(inode, flag, false);
 }
 
-static inline void set_acl_inode(struct f2fs_inode_info *fi, umode_t mode)
+static inline void set_acl_inode(struct inode *inode, umode_t mode)
 {
-	fi->i_acl_mode = mode;
-	set_inode_flag(fi, FI_ACL_MODE);
+	F2FS_I(inode)->i_acl_mode = mode;
+	set_inode_flag(inode, FI_ACL_MODE);
+	f2fs_mark_inode_dirty_sync(inode, false);
 }
 
-static inline void get_inline_info(struct f2fs_inode_info *fi,
-					struct f2fs_inode *ri)
+static inline void f2fs_i_links_write(struct inode *inode, bool inc)
 {
+	if (inc)
+		inc_nlink(inode);
+	else
+		drop_nlink(inode);
+	f2fs_mark_inode_dirty_sync(inode, true);
+}
+
+static inline void f2fs_i_blocks_write(struct inode *inode,
+					block_t diff, bool add, bool claim)
+{
+	bool clean = !is_inode_flag_set(inode, FI_DIRTY_INODE);
+	bool recover = is_inode_flag_set(inode, FI_AUTO_RECOVER);
+
+	/* add = 1, claim = 1 should be dquot_reserve_block in pair */
+	if (add) {
+		if (claim)
+			dquot_claim_block(inode, diff);
+		else
+			dquot_alloc_block_nofail(inode, diff);
+	} else {
+		dquot_free_block(inode, diff);
+	}
+
+	f2fs_mark_inode_dirty_sync(inode, true);
+	if (clean || recover)
+		set_inode_flag(inode, FI_AUTO_RECOVER);
+}
+
+static inline void f2fs_i_size_write(struct inode *inode, loff_t i_size)
+{
+	bool clean = !is_inode_flag_set(inode, FI_DIRTY_INODE);
+	bool recover = is_inode_flag_set(inode, FI_AUTO_RECOVER);
+
+	if (i_size_read(inode) == i_size)
+		return;
+
+	i_size_write(inode, i_size);
+	f2fs_mark_inode_dirty_sync(inode, true);
+	if (clean || recover)
+		set_inode_flag(inode, FI_AUTO_RECOVER);
+}
+
+static inline void f2fs_i_depth_write(struct inode *inode, unsigned int depth)
+{
+	F2FS_I(inode)->i_current_depth = depth;
+	f2fs_mark_inode_dirty_sync(inode, true);
+}
+
+static inline void f2fs_i_xnid_write(struct inode *inode, nid_t xnid)
+{
+	F2FS_I(inode)->i_xattr_nid = xnid;
+	f2fs_mark_inode_dirty_sync(inode, true);
+}
+
+static inline void f2fs_i_pino_write(struct inode *inode, nid_t pino)
+{
+	F2FS_I(inode)->i_pino = pino;
+	f2fs_mark_inode_dirty_sync(inode, true);
+}
+
+static inline void get_inline_info(struct inode *inode, struct f2fs_inode *ri)
+{
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+
 	if (ri->i_inline & F2FS_INLINE_XATTR)
-		set_inode_flag(fi, FI_INLINE_XATTR);
+		set_bit(FI_INLINE_XATTR, &fi->flags);
 	if (ri->i_inline & F2FS_INLINE_DATA)
-		set_inode_flag(fi, FI_INLINE_DATA);
+		set_bit(FI_INLINE_DATA, &fi->flags);
 	if (ri->i_inline & F2FS_INLINE_DENTRY)
-		set_inode_flag(fi, FI_INLINE_DENTRY);
+		set_bit(FI_INLINE_DENTRY, &fi->flags);
 	if (ri->i_inline & F2FS_DATA_EXIST)
-		set_inode_flag(fi, FI_DATA_EXIST);
+		set_bit(FI_DATA_EXIST, &fi->flags);
 	if (ri->i_inline & F2FS_INLINE_DOTS)
-		set_inode_flag(fi, FI_INLINE_DOTS);
+		set_bit(FI_INLINE_DOTS, &fi->flags);
+	if (ri->i_inline & F2FS_EXTRA_ATTR)
+		set_bit(FI_EXTRA_ATTR, &fi->flags);
 }
 
-static inline void set_raw_inline(struct f2fs_inode_info *fi,
-					struct f2fs_inode *ri)
+static inline void set_raw_inline(struct inode *inode, struct f2fs_inode *ri)
 {
 	ri->i_inline = 0;
 
-	if (is_inode_flag_set(fi, FI_INLINE_XATTR))
+	if (is_inode_flag_set(inode, FI_INLINE_XATTR))
 		ri->i_inline |= F2FS_INLINE_XATTR;
-	if (is_inode_flag_set(fi, FI_INLINE_DATA))
+	if (is_inode_flag_set(inode, FI_INLINE_DATA))
 		ri->i_inline |= F2FS_INLINE_DATA;
-	if (is_inode_flag_set(fi, FI_INLINE_DENTRY))
+	if (is_inode_flag_set(inode, FI_INLINE_DENTRY))
 		ri->i_inline |= F2FS_INLINE_DENTRY;
-	if (is_inode_flag_set(fi, FI_DATA_EXIST))
+	if (is_inode_flag_set(inode, FI_DATA_EXIST))
 		ri->i_inline |= F2FS_DATA_EXIST;
-	if (is_inode_flag_set(fi, FI_INLINE_DOTS))
+	if (is_inode_flag_set(inode, FI_INLINE_DOTS))
 		ri->i_inline |= F2FS_INLINE_DOTS;
+	if (is_inode_flag_set(inode, FI_EXTRA_ATTR))
+		ri->i_inline |= F2FS_EXTRA_ATTR;
+}
+
+static inline int f2fs_has_extra_attr(struct inode *inode)
+{
+	return is_inode_flag_set(inode, FI_EXTRA_ATTR);
 }
 
 static inline int f2fs_has_inline_xattr(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_INLINE_XATTR);
+	return is_inode_flag_set(inode, FI_INLINE_XATTR);
 }
 
-static inline unsigned int addrs_per_inode(struct f2fs_inode_info *fi)
+static inline unsigned int addrs_per_inode(struct inode *inode)
 {
-	if (f2fs_has_inline_xattr(&fi->vfs_inode))
-		return DEF_ADDRS_PER_INODE - F2FS_INLINE_XATTR_ADDRS;
-	return DEF_ADDRS_PER_INODE;
+	if (f2fs_has_inline_xattr(inode))
+		return CUR_ADDRS_PER_INODE(inode) - F2FS_INLINE_XATTR_ADDRS;
+	return CUR_ADDRS_PER_INODE(inode);
 }
 
 static inline void *inline_xattr_addr(struct page *page)
 {
 	struct f2fs_inode *ri = F2FS_INODE(page);
+
 	return (void *)&(ri->i_addr[DEF_ADDRS_PER_INODE -
 					F2FS_INLINE_XATTR_ADDRS]);
 }
@@ -1502,54 +2247,55 @@ static inline int inline_xattr_size(struct inode *inode)
 
 static inline int f2fs_has_inline_data(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_INLINE_DATA);
-}
-
-static inline void f2fs_clear_inline_inode(struct inode *inode)
-{
-	clear_inode_flag(F2FS_I(inode), FI_INLINE_DATA);
-	clear_inode_flag(F2FS_I(inode), FI_DATA_EXIST);
+	return is_inode_flag_set(inode, FI_INLINE_DATA);
 }
 
 static inline int f2fs_exist_data(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_DATA_EXIST);
+	return is_inode_flag_set(inode, FI_DATA_EXIST);
 }
 
 static inline int f2fs_has_inline_dots(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_INLINE_DOTS);
+	return is_inode_flag_set(inode, FI_INLINE_DOTS);
 }
 
 static inline bool f2fs_is_atomic_file(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_ATOMIC_FILE);
+	return is_inode_flag_set(inode, FI_ATOMIC_FILE);
+}
+
+static inline bool f2fs_is_commit_atomic_write(struct inode *inode)
+{
+	return is_inode_flag_set(inode, FI_ATOMIC_COMMIT);
 }
 
 static inline bool f2fs_is_volatile_file(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_VOLATILE_FILE);
+	return is_inode_flag_set(inode, FI_VOLATILE_FILE);
 }
 
 static inline bool f2fs_is_first_block_written(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_FIRST_BLOCK_WRITTEN);
+	return is_inode_flag_set(inode, FI_FIRST_BLOCK_WRITTEN);
 }
 
 static inline bool f2fs_is_drop_cache(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_DROP_CACHE);
+	return is_inode_flag_set(inode, FI_DROP_CACHE);
 }
 
-static inline void *inline_data_addr(struct page *page)
+static inline void *inline_data_addr(struct inode *inode, struct page *page)
 {
 	struct f2fs_inode *ri = F2FS_INODE(page);
-	return (void *)&(ri->i_addr[1]);
+	int extra_size = get_extra_isize(inode);
+
+	return (void *)&(ri->i_addr[extra_size + DEF_INLINE_RESERVED_SIZE]);
 }
 
 static inline int f2fs_has_inline_dentry(struct inode *inode)
 {
-	return is_inode_flag_set(F2FS_I(inode), FI_INLINE_DENTRY);
+	return is_inode_flag_set(inode, FI_INLINE_DENTRY);
 }
 
 static inline void f2fs_dentry_kunmap(struct inode *dir, struct page *page)
@@ -1566,27 +2312,41 @@ static inline int is_file(struct inode *inode, int type)
 static inline void set_file(struct inode *inode, int type)
 {
 	F2FS_I(inode)->i_advise |= type;
+	f2fs_mark_inode_dirty_sync(inode, true);
 }
 
 static inline void clear_file(struct inode *inode, int type)
 {
 	F2FS_I(inode)->i_advise &= ~type;
+	f2fs_mark_inode_dirty_sync(inode, true);
 }
 
-static inline int f2fs_readonly(struct super_block *sb)
+static inline bool f2fs_skip_inode_update(struct inode *inode, int dsync)
 {
-	return sb->s_flags & MS_RDONLY;
+	if (dsync) {
+		struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+		bool ret;
+
+		spin_lock(&sbi->inode_lock[DIRTY_META]);
+		ret = list_empty(&F2FS_I(inode)->gdirty_list);
+		spin_unlock(&sbi->inode_lock[DIRTY_META]);
+		return ret;
+	}
+	if (!is_inode_flag_set(inode, FI_AUTO_RECOVER) ||
+			file_keep_isize(inode) ||
+			i_size_read(inode) & PAGE_MASK)
+		return false;
+	return F2FS_I(inode)->last_disk_size == i_size_read(inode);
 }
 
-static inline bool f2fs_cp_error(struct f2fs_sb_info *sbi)
+static inline int f2fs_readonly(struct super_block *sb)
 {
-	return is_set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
+	return sb->s_flags & MS_RDONLY;
 }
 
-static inline void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi)
+static inline bool f2fs_cp_error(struct f2fs_sb_info *sbi)
 {
-	set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
-	sbi->sb->s_flags |= MS_RDONLY;
+	return is_set_ckpt_flags(sbi, CP_ERROR_FLAG);
 }
 
 static inline bool is_dot_dotdot(const struct qstr *str)
@@ -1602,16 +2362,26 @@ static inline bool is_dot_dotdot(const struct qstr *str)
 
 static inline bool f2fs_may_extent_tree(struct inode *inode)
 {
-	mode_t mode = inode->i_mode;
-
 	if (!test_opt(F2FS_I_SB(inode), EXTENT_CACHE) ||
-			is_inode_flag_set(F2FS_I(inode), FI_NO_EXTENT))
+			is_inode_flag_set(inode, FI_NO_EXTENT))
 		return false;
 
-	return S_ISREG(mode);
+	return S_ISREG(inode->i_mode);
+}
+
+static inline void *f2fs_kmalloc(struct f2fs_sb_info *sbi,
+					size_t size, gfp_t flags)
+{
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_KMALLOC)) {
+		f2fs_show_injection_info(FAULT_KMALLOC);
+		return NULL;
+	}
+#endif
+	return kmalloc(size, flags);
 }
 
-static inline void *f2fs_kvmalloc(size_t size, gfp_t flags)
+static inline void *kvmalloc(size_t size, gfp_t flags)
 {
 	void *ret;
 
@@ -1621,7 +2391,7 @@ static inline void *f2fs_kvmalloc(size_t size, gfp_t flags)
 	return ret;
 }
 
-static inline void *f2fs_kvzalloc(size_t size, gfp_t flags)
+static inline void *kvzalloc(size_t size, gfp_t flags)
 {
 	void *ret;
 
@@ -1631,41 +2401,79 @@ static inline void *f2fs_kvzalloc(size_t size, gfp_t flags)
 	return ret;
 }
 
+static inline int get_extra_isize(struct inode *inode)
+{
+	return F2FS_I(inode)->i_extra_isize / sizeof(__le32);
+}
+
 #define get_inode_mode(i) \
-	((is_inode_flag_set(F2FS_I(i), FI_ACL_MODE)) ? \
+	((is_inode_flag_set(i, FI_ACL_MODE)) ? \
 	 (F2FS_I(i)->i_acl_mode) : ((i)->i_mode))
 
-/* get offset of first page in next direct node */
-#define PGOFS_OF_NEXT_DNODE(pgofs, fi)				\
-	((pgofs < ADDRS_PER_INODE(fi)) ? ADDRS_PER_INODE(fi) :	\
-	(pgofs - ADDRS_PER_INODE(fi) + ADDRS_PER_BLOCK) /	\
-	ADDRS_PER_BLOCK * ADDRS_PER_BLOCK + ADDRS_PER_INODE(fi))
+#define F2FS_TOTAL_EXTRA_ATTR_SIZE			\
+	(offsetof(struct f2fs_inode, i_extra_end) -	\
+	offsetof(struct f2fs_inode, i_extra_isize))	\
+
+#define F2FS_OLD_ATTRIBUTE_SIZE	(offsetof(struct f2fs_inode, i_addr))
+#define F2FS_FITS_IN_INODE(f2fs_inode, extra_isize, field)		\
+		((offsetof(typeof(*f2fs_inode), field) +	\
+		sizeof((f2fs_inode)->field))			\
+		<= (F2FS_OLD_ATTRIBUTE_SIZE + extra_isize))	\
+
+static inline void f2fs_reset_iostat(struct f2fs_sb_info *sbi)
+{
+	int i;
+
+	spin_lock(&sbi->iostat_lock);
+	for (i = 0; i < NR_IO_TYPE; i++)
+		sbi->write_iostat[i] = 0;
+	spin_unlock(&sbi->iostat_lock);
+}
+
+static inline void f2fs_update_iostat(struct f2fs_sb_info *sbi,
+			enum iostat_type type, unsigned long long io_bytes)
+{
+	if (!sbi->iostat_enable)
+		return;
+	spin_lock(&sbi->iostat_lock);
+	sbi->write_iostat[type] += io_bytes;
+
+	if (type == APP_WRITE_IO || type == APP_DIRECT_IO)
+		sbi->write_iostat[APP_BUFFERED_IO] =
+			sbi->write_iostat[APP_WRITE_IO] -
+			sbi->write_iostat[APP_DIRECT_IO];
+	spin_unlock(&sbi->iostat_lock);
+}
 
 /*
  * file.c
  */
-int f2fs_sync_file(struct file *, loff_t, loff_t, int);
-void truncate_data_blocks(struct dnode_of_data *);
-int truncate_blocks(struct inode *, u64, bool);
-int f2fs_truncate(struct inode *, bool);
-int f2fs_getattr(struct vfsmount *, struct dentry *, struct kstat *);
-int f2fs_setattr(struct dentry *, struct iattr *);
-int truncate_hole(struct inode *, pgoff_t, pgoff_t);
-int truncate_data_blocks_range(struct dnode_of_data *, int);
-long f2fs_ioctl(struct file *, unsigned int, unsigned long);
-long f2fs_compat_ioctl(struct file *, unsigned int, unsigned long);
+int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync);
+void truncate_data_blocks(struct dnode_of_data *dn);
+int truncate_blocks(struct inode *inode, u64 from, bool lock);
+int f2fs_truncate(struct inode *inode);
+int f2fs_getattr(struct vfsmount *mnt, struct dentry *dentry,
+			struct kstat *stat);
+int f2fs_setattr(struct dentry *dentry, struct iattr *attr);
+int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end);
+int truncate_data_blocks_range(struct dnode_of_data *dn, int count);
+long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg);
+long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
 
 /*
  * inode.c
  */
-void f2fs_set_inode_flags(struct inode *);
-struct inode *f2fs_iget(struct super_block *, unsigned long);
-int try_to_free_nats(struct f2fs_sb_info *, int);
-void update_inode(struct inode *, struct page *);
-void update_inode_page(struct inode *);
-int f2fs_write_inode(struct inode *, struct writeback_control *);
-void f2fs_evict_inode(struct inode *);
-void handle_failed_inode(struct inode *);
+void f2fs_set_inode_flags(struct inode *inode);
+bool f2fs_inode_chksum_verify(struct f2fs_sb_info *sbi, struct page *page);
+void f2fs_inode_chksum_set(struct f2fs_sb_info *sbi, struct page *page);
+struct inode *f2fs_iget(struct super_block *sb, unsigned long ino);
+struct inode *f2fs_iget_retry(struct super_block *sb, unsigned long ino);
+int try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink);
+int update_inode(struct inode *inode, struct page *node_page);
+int update_inode_page(struct inode *inode);
+int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc);
+void f2fs_evict_inode(struct inode *inode);
+void handle_failed_inode(struct inode *inode);
 
 /*
  * namei.c
@@ -1675,35 +2483,45 @@ struct dentry *f2fs_get_parent(struct dentry *child);
 /*
  * dir.c
  */
-extern unsigned char f2fs_filetype_table[F2FS_FT_MAX];
-void set_de_type(struct f2fs_dir_entry *, umode_t);
-
-struct f2fs_dir_entry *find_target_dentry(struct f2fs_filename *,
-			f2fs_hash_t, int *, struct f2fs_dentry_ptr *);
-bool f2fs_fill_dentries(struct dir_context *, struct f2fs_dentry_ptr *,
-			unsigned int, struct f2fs_str *);
-void do_make_empty_dir(struct inode *, struct inode *,
-			struct f2fs_dentry_ptr *);
-struct page *init_inode_metadata(struct inode *, struct inode *,
-			const struct qstr *, struct page *);
-void update_parent_metadata(struct inode *, struct inode *, unsigned int);
-int room_for_filename(const void *, int, int);
-void f2fs_drop_nlink(struct inode *, struct inode *, struct page *);
-struct f2fs_dir_entry *f2fs_find_entry(struct inode *, struct qstr *,
-							struct page **);
-struct f2fs_dir_entry *f2fs_parent_dir(struct inode *, struct page **);
-ino_t f2fs_inode_by_name(struct inode *, struct qstr *);
-void f2fs_set_link(struct inode *, struct f2fs_dir_entry *,
-				struct page *, struct inode *);
-int update_dent_inode(struct inode *, struct inode *, const struct qstr *);
-void f2fs_update_dentry(nid_t ino, umode_t mode, struct f2fs_dentry_ptr *,
-			const struct qstr *, f2fs_hash_t , unsigned int);
-int __f2fs_add_link(struct inode *, const struct qstr *, struct inode *, nid_t,
-			umode_t);
-void f2fs_delete_entry(struct f2fs_dir_entry *, struct page *, struct inode *,
-							struct inode *);
-int f2fs_do_tmpfile(struct inode *, struct inode *);
-bool f2fs_empty_dir(struct inode *);
+void set_de_type(struct f2fs_dir_entry *de, umode_t mode);
+unsigned char get_de_type(struct f2fs_dir_entry *de);
+struct f2fs_dir_entry *find_target_dentry(struct fscrypt_name *fname,
+			f2fs_hash_t namehash, int *max_slots,
+			struct f2fs_dentry_ptr *d);
+int f2fs_fill_dentries(struct dir_context *ctx, struct f2fs_dentry_ptr *d,
+			unsigned int start_pos, struct fscrypt_str *fstr);
+void do_make_empty_dir(struct inode *inode, struct inode *parent,
+			struct f2fs_dentry_ptr *d);
+struct page *init_inode_metadata(struct inode *inode, struct inode *dir,
+			const struct qstr *new_name,
+			const struct qstr *orig_name, struct page *dpage);
+void update_parent_metadata(struct inode *dir, struct inode *inode,
+			unsigned int current_depth);
+int room_for_filename(const void *bitmap, int slots, int max_slots);
+void f2fs_drop_nlink(struct inode *dir, struct inode *inode);
+struct f2fs_dir_entry *__f2fs_find_entry(struct inode *dir,
+			struct fscrypt_name *fname, struct page **res_page);
+struct f2fs_dir_entry *f2fs_find_entry(struct inode *dir,
+			const struct qstr *child, struct page **res_page);
+struct f2fs_dir_entry *f2fs_parent_dir(struct inode *dir, struct page **p);
+ino_t f2fs_inode_by_name(struct inode *dir, const struct qstr *qstr,
+			struct page **page);
+void f2fs_set_link(struct inode *dir, struct f2fs_dir_entry *de,
+			struct page *page, struct inode *inode);
+void f2fs_update_dentry(nid_t ino, umode_t mode, struct f2fs_dentry_ptr *d,
+			const struct qstr *name, f2fs_hash_t name_hash,
+			unsigned int bit_pos);
+int f2fs_add_regular_entry(struct inode *dir, const struct qstr *new_name,
+			const struct qstr *orig_name,
+			struct inode *inode, nid_t ino, umode_t mode);
+int __f2fs_do_add_link(struct inode *dir, struct fscrypt_name *fname,
+			struct inode *inode, nid_t ino, umode_t mode);
+int __f2fs_add_link(struct inode *dir, const struct qstr *name,
+			struct inode *inode, nid_t ino, umode_t mode);
+void f2fs_delete_entry(struct f2fs_dir_entry *dentry, struct page *page,
+			struct inode *dir, struct inode *inode);
+int f2fs_do_tmpfile(struct inode *inode, struct inode *dir);
+bool f2fs_empty_dir(struct inode *dir);
 
 static inline int f2fs_add_link(struct dentry *dentry, struct inode *inode)
 {
@@ -1714,16 +2532,21 @@ static inline int f2fs_add_link(struct dentry *dentry, struct inode *inode)
 /*
  * super.c
  */
-int f2fs_commit_super(struct f2fs_sb_info *, bool);
-int f2fs_sync_fs(struct super_block *, int);
+int f2fs_inode_dirtied(struct inode *inode, bool sync);
+void f2fs_inode_synced(struct inode *inode);
+void f2fs_enable_quota_files(struct f2fs_sb_info *sbi);
+void f2fs_quota_off_umount(struct super_block *sb);
+int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover);
+int f2fs_sync_fs(struct super_block *sb, int sync);
 extern __printf(3, 4)
-void f2fs_msg(struct super_block *, const char *, const char *, ...);
+void f2fs_msg(struct super_block *sb, const char *level, const char *fmt, ...);
+int sanity_check_ckpt(struct f2fs_sb_info *sbi);
 
 /*
  * hash.c
  */
 f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
-				struct f2fs_filename *fname);
+				struct fscrypt_name *fname);
 
 /*
  * node.c
@@ -1731,141 +2554,190 @@ f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
 struct dnode_of_data;
 struct node_info;
 
-bool available_free_memory(struct f2fs_sb_info *, int);
-int need_dentry_mark(struct f2fs_sb_info *, nid_t);
-bool is_checkpointed_node(struct f2fs_sb_info *, nid_t);
-bool need_inode_block_update(struct f2fs_sb_info *, nid_t);
-void get_node_info(struct f2fs_sb_info *, nid_t, struct node_info *);
-int get_dnode_of_data(struct dnode_of_data *, pgoff_t, int);
-int truncate_inode_blocks(struct inode *, pgoff_t);
-int truncate_xattr_node(struct inode *, struct page *);
-int wait_on_node_pages_writeback(struct f2fs_sb_info *, nid_t);
-int remove_inode_page(struct inode *);
-struct page *new_inode_page(struct inode *);
-struct page *new_node_page(struct dnode_of_data *, unsigned int, struct page *);
-void ra_node_page(struct f2fs_sb_info *, nid_t);
-struct page *get_node_page(struct f2fs_sb_info *, pgoff_t);
-struct page *get_node_page_ra(struct page *, int);
-void sync_inode_page(struct dnode_of_data *);
-int sync_node_pages(struct f2fs_sb_info *, nid_t, struct writeback_control *);
-bool alloc_nid(struct f2fs_sb_info *, nid_t *);
-void alloc_nid_done(struct f2fs_sb_info *, nid_t);
-void alloc_nid_failed(struct f2fs_sb_info *, nid_t);
-int try_to_free_nids(struct f2fs_sb_info *, int);
-void recover_inline_xattr(struct inode *, struct page *);
-void recover_xattr_data(struct inode *, struct page *, block_t);
-int recover_inode_page(struct f2fs_sb_info *, struct page *);
-int restore_node_summary(struct f2fs_sb_info *, unsigned int,
-				struct f2fs_summary_block *);
-void flush_nat_entries(struct f2fs_sb_info *);
-int build_node_manager(struct f2fs_sb_info *);
-void destroy_node_manager(struct f2fs_sb_info *);
+bool available_free_memory(struct f2fs_sb_info *sbi, int type);
+int need_dentry_mark(struct f2fs_sb_info *sbi, nid_t nid);
+bool is_checkpointed_node(struct f2fs_sb_info *sbi, nid_t nid);
+bool need_inode_block_update(struct f2fs_sb_info *sbi, nid_t ino);
+void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni);
+pgoff_t get_next_page_offset(struct dnode_of_data *dn, pgoff_t pgofs);
+int get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode);
+int truncate_inode_blocks(struct inode *inode, pgoff_t from);
+int truncate_xattr_node(struct inode *inode, struct page *page);
+int wait_on_node_pages_writeback(struct f2fs_sb_info *sbi, nid_t ino);
+int remove_inode_page(struct inode *inode);
+struct page *new_inode_page(struct inode *inode);
+struct page *new_node_page(struct dnode_of_data *dn, unsigned int ofs);
+void ra_node_page(struct f2fs_sb_info *sbi, nid_t nid);
+struct page *get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid);
+struct page *get_node_page_ra(struct page *parent, int start);
+void move_node_page(struct page *node_page, int gc_type);
+int fsync_node_pages(struct f2fs_sb_info *sbi, struct inode *inode,
+			struct writeback_control *wbc, bool atomic);
+int sync_node_pages(struct f2fs_sb_info *sbi, struct writeback_control *wbc,
+			bool do_balance, enum iostat_type io_type);
+void build_free_nids(struct f2fs_sb_info *sbi, bool sync, bool mount);
+bool alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid);
+void alloc_nid_done(struct f2fs_sb_info *sbi, nid_t nid);
+void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid);
+int try_to_free_nids(struct f2fs_sb_info *sbi, int nr_shrink);
+void recover_inline_xattr(struct inode *inode, struct page *page);
+int recover_xattr_data(struct inode *inode, struct page *page,
+			block_t blkaddr);
+int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page);
+int restore_node_summary(struct f2fs_sb_info *sbi,
+			unsigned int segno, struct f2fs_summary_block *sum);
+void flush_nat_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc);
+int build_node_manager(struct f2fs_sb_info *sbi);
+void destroy_node_manager(struct f2fs_sb_info *sbi);
 int __init create_node_manager_caches(void);
 void destroy_node_manager_caches(void);
 
 /*
  * segment.c
  */
-void register_inmem_page(struct inode *, struct page *);
-int commit_inmem_pages(struct inode *, bool);
-void f2fs_balance_fs(struct f2fs_sb_info *);
-void f2fs_balance_fs_bg(struct f2fs_sb_info *);
-int f2fs_issue_flush(struct f2fs_sb_info *);
-int create_flush_cmd_control(struct f2fs_sb_info *);
-void destroy_flush_cmd_control(struct f2fs_sb_info *);
-void invalidate_blocks(struct f2fs_sb_info *, block_t);
-bool is_checkpointed_data(struct f2fs_sb_info *, block_t);
-void refresh_sit_entry(struct f2fs_sb_info *, block_t, block_t);
-void clear_prefree_segments(struct f2fs_sb_info *, struct cp_control *);
-void release_discard_addrs(struct f2fs_sb_info *);
-bool discard_next_dnode(struct f2fs_sb_info *, block_t);
-int npages_for_summary_flush(struct f2fs_sb_info *, bool);
-void allocate_new_segments(struct f2fs_sb_info *);
-int f2fs_trim_fs(struct f2fs_sb_info *, struct fstrim_range *);
-struct page *get_sum_page(struct f2fs_sb_info *, unsigned int);
-void update_meta_page(struct f2fs_sb_info *, void *, block_t);
-void write_meta_page(struct f2fs_sb_info *, struct page *);
-void write_node_page(unsigned int, struct f2fs_io_info *);
-void write_data_page(struct dnode_of_data *, struct f2fs_io_info *);
-void rewrite_data_page(struct f2fs_io_info *);
-void f2fs_replace_block(struct f2fs_sb_info *, struct dnode_of_data *,
-				block_t, block_t, unsigned char, bool);
-void allocate_data_block(struct f2fs_sb_info *, struct page *,
-		block_t, block_t *, struct f2fs_summary *, int);
-void f2fs_wait_on_page_writeback(struct page *, enum page_type);
-void f2fs_wait_on_encrypted_page_writeback(struct f2fs_sb_info *, block_t);
-void write_data_summaries(struct f2fs_sb_info *, block_t);
-void write_node_summaries(struct f2fs_sb_info *, block_t);
-int lookup_journal_in_cursum(struct f2fs_summary_block *,
-					int, unsigned int, int);
-void flush_sit_entries(struct f2fs_sb_info *, struct cp_control *);
-int build_segment_manager(struct f2fs_sb_info *);
-void destroy_segment_manager(struct f2fs_sb_info *);
+bool need_SSR(struct f2fs_sb_info *sbi);
+void register_inmem_page(struct inode *inode, struct page *page);
+void drop_inmem_pages(struct inode *inode);
+void drop_inmem_page(struct inode *inode, struct page *page);
+int commit_inmem_pages(struct inode *inode);
+void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need);
+void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi);
+int f2fs_issue_flush(struct f2fs_sb_info *sbi);
+int create_flush_cmd_control(struct f2fs_sb_info *sbi);
+void destroy_flush_cmd_control(struct f2fs_sb_info *sbi, bool free);
+void invalidate_blocks(struct f2fs_sb_info *sbi, block_t addr);
+bool is_checkpointed_data(struct f2fs_sb_info *sbi, block_t blkaddr);
+void refresh_sit_entry(struct f2fs_sb_info *sbi, block_t old, block_t new);
+void stop_discard_thread(struct f2fs_sb_info *sbi);
+void f2fs_wait_discard_bios(struct f2fs_sb_info *sbi, bool umount);
+void clear_prefree_segments(struct f2fs_sb_info *sbi, struct cp_control *cpc);
+void release_discard_addrs(struct f2fs_sb_info *sbi);
+int npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra);
+void allocate_new_segments(struct f2fs_sb_info *sbi);
+int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range);
+bool exist_trim_candidates(struct f2fs_sb_info *sbi, struct cp_control *cpc);
+struct page *get_sum_page(struct f2fs_sb_info *sbi, unsigned int segno);
+void update_meta_page(struct f2fs_sb_info *sbi, void *src, block_t blk_addr);
+void write_meta_page(struct f2fs_sb_info *sbi, struct page *page,
+						enum iostat_type io_type);
+void write_node_page(unsigned int nid, struct f2fs_io_info *fio);
+void write_data_page(struct dnode_of_data *dn, struct f2fs_io_info *fio);
+int rewrite_data_page(struct f2fs_io_info *fio);
+void __f2fs_replace_block(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
+			block_t old_blkaddr, block_t new_blkaddr,
+			bool recover_curseg, bool recover_newaddr);
+void f2fs_replace_block(struct f2fs_sb_info *sbi, struct dnode_of_data *dn,
+			block_t old_addr, block_t new_addr,
+			unsigned char version, bool recover_curseg,
+			bool recover_newaddr);
+void allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
+			block_t old_blkaddr, block_t *new_blkaddr,
+			struct f2fs_summary *sum, int type,
+			struct f2fs_io_info *fio, bool add_list);
+void f2fs_wait_on_page_writeback(struct page *page,
+			enum page_type type, bool ordered);
+void f2fs_wait_on_block_writeback(struct f2fs_sb_info *sbi, block_t blkaddr);
+void write_data_summaries(struct f2fs_sb_info *sbi, block_t start_blk);
+void write_node_summaries(struct f2fs_sb_info *sbi, block_t start_blk);
+int lookup_journal_in_cursum(struct f2fs_journal *journal, int type,
+			unsigned int val, int alloc);
+void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc);
+int build_segment_manager(struct f2fs_sb_info *sbi);
+void destroy_segment_manager(struct f2fs_sb_info *sbi);
 int __init create_segment_manager_caches(void);
 void destroy_segment_manager_caches(void);
 
 /*
  * checkpoint.c
  */
-struct page *grab_meta_page(struct f2fs_sb_info *, pgoff_t);
-struct page *get_meta_page(struct f2fs_sb_info *, pgoff_t);
-struct page *get_tmp_page(struct f2fs_sb_info *, pgoff_t);
-bool is_valid_blkaddr(struct f2fs_sb_info *, block_t, int);
-int ra_meta_pages(struct f2fs_sb_info *, block_t, int, int, bool);
-void ra_meta_pages_cond(struct f2fs_sb_info *, pgoff_t);
-long sync_meta_pages(struct f2fs_sb_info *, enum page_type, long);
-void add_dirty_inode(struct f2fs_sb_info *, nid_t, int type);
-void remove_dirty_inode(struct f2fs_sb_info *, nid_t, int type);
-void release_dirty_inode(struct f2fs_sb_info *);
-bool exist_written_data(struct f2fs_sb_info *, nid_t, int);
-int acquire_orphan_inode(struct f2fs_sb_info *);
-void release_orphan_inode(struct f2fs_sb_info *);
-void add_orphan_inode(struct f2fs_sb_info *, nid_t);
-void remove_orphan_inode(struct f2fs_sb_info *, nid_t);
-int recover_orphan_inodes(struct f2fs_sb_info *);
-int get_valid_checkpoint(struct f2fs_sb_info *);
-void update_dirty_page(struct inode *, struct page *);
-void add_dirty_dir_inode(struct inode *);
-void remove_dirty_dir_inode(struct inode *);
-void sync_dirty_dir_inodes(struct f2fs_sb_info *);
-void write_checkpoint(struct f2fs_sb_info *, struct cp_control *);
-void init_ino_entry_info(struct f2fs_sb_info *);
+void f2fs_stop_checkpoint(struct f2fs_sb_info *sbi, bool end_io);
+struct page *grab_meta_page(struct f2fs_sb_info *sbi, pgoff_t index);
+struct page *get_meta_page(struct f2fs_sb_info *sbi, pgoff_t index);
+struct page *get_tmp_page(struct f2fs_sb_info *sbi, pgoff_t index);
+bool is_valid_blkaddr(struct f2fs_sb_info *sbi, block_t blkaddr, int type);
+int ra_meta_pages(struct f2fs_sb_info *sbi, block_t start, int nrpages,
+			int type, bool sync);
+void ra_meta_pages_cond(struct f2fs_sb_info *sbi, pgoff_t index);
+long sync_meta_pages(struct f2fs_sb_info *sbi, enum page_type type,
+			long nr_to_write, enum iostat_type io_type);
+void add_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type);
+void remove_ino_entry(struct f2fs_sb_info *sbi, nid_t ino, int type);
+void release_ino_entry(struct f2fs_sb_info *sbi, bool all);
+bool exist_written_data(struct f2fs_sb_info *sbi, nid_t ino, int mode);
+int f2fs_sync_inode_meta(struct f2fs_sb_info *sbi);
+int acquire_orphan_inode(struct f2fs_sb_info *sbi);
+void release_orphan_inode(struct f2fs_sb_info *sbi);
+void add_orphan_inode(struct inode *inode);
+void remove_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino);
+int recover_orphan_inodes(struct f2fs_sb_info *sbi);
+int get_valid_checkpoint(struct f2fs_sb_info *sbi);
+void update_dirty_page(struct inode *inode, struct page *page);
+void remove_dirty_inode(struct inode *inode);
+int sync_dirty_inodes(struct f2fs_sb_info *sbi, enum inode_type type);
+int write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc);
+void init_ino_entry_info(struct f2fs_sb_info *sbi);
 int __init create_checkpoint_caches(void);
 void destroy_checkpoint_caches(void);
 
 /*
  * data.c
  */
-void f2fs_submit_merged_bio(struct f2fs_sb_info *, enum page_type, int);
-int f2fs_submit_page_bio(struct f2fs_io_info *);
-void f2fs_submit_page_mbio(struct f2fs_io_info *);
-void set_data_blkaddr(struct dnode_of_data *);
-int reserve_new_block(struct dnode_of_data *);
-int f2fs_get_block(struct dnode_of_data *, pgoff_t);
-int f2fs_reserve_block(struct dnode_of_data *, pgoff_t);
-struct page *get_read_data_page(struct inode *, pgoff_t, int, bool);
-struct page *find_data_page(struct inode *, pgoff_t);
-struct page *get_lock_data_page(struct inode *, pgoff_t, bool);
-struct page *get_new_data_page(struct inode *, struct page *, pgoff_t, bool);
-int do_write_data_page(struct f2fs_io_info *);
-int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *, u64, u64);
-void f2fs_invalidate_page(struct page *, unsigned int, unsigned int);
-int f2fs_release_page(struct page *, gfp_t);
+void f2fs_submit_merged_write(struct f2fs_sb_info *sbi, enum page_type type);
+void f2fs_submit_merged_write_cond(struct f2fs_sb_info *sbi,
+				struct inode *inode, nid_t ino, pgoff_t idx,
+				enum page_type type);
+void f2fs_flush_merged_writes(struct f2fs_sb_info *sbi);
+int f2fs_submit_page_bio(struct f2fs_io_info *fio);
+int f2fs_submit_page_write(struct f2fs_io_info *fio);
+struct block_device *f2fs_target_device(struct f2fs_sb_info *sbi,
+			block_t blk_addr, struct bio *bio);
+int f2fs_target_device_index(struct f2fs_sb_info *sbi, block_t blkaddr);
+void set_data_blkaddr(struct dnode_of_data *dn);
+void f2fs_update_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr);
+int reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count);
+int reserve_new_block(struct dnode_of_data *dn);
+int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index);
+int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from);
+int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index);
+struct page *get_read_data_page(struct inode *inode, pgoff_t index,
+			int op_flags, bool for_write);
+struct page *find_data_page(struct inode *inode, pgoff_t index);
+struct page *get_lock_data_page(struct inode *inode, pgoff_t index,
+			bool for_write);
+struct page *get_new_data_page(struct inode *inode,
+			struct page *ipage, pgoff_t index, bool new_i_size);
+int do_write_data_page(struct f2fs_io_info *fio);
+int f2fs_map_blocks(struct inode *inode, struct f2fs_map_blocks *map,
+			int create, int flag);
+int f2fs_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo,
+			u64 start, u64 len);
+void f2fs_set_page_dirty_nobuffers(struct page *page);
+int __f2fs_write_data_pages(struct address_space *mapping,
+						struct writeback_control *wbc,
+						enum iostat_type io_type);
+void f2fs_invalidate_page(struct page *page, unsigned int offset,
+			unsigned int length);
+int f2fs_release_page(struct page *page, gfp_t wait);
+#ifdef CONFIG_MIGRATION
+int f2fs_migrate_page(struct address_space *mapping, struct page *newpage,
+			struct page *page, enum migrate_mode mode);
+#endif
 
 /*
  * gc.c
  */
-int start_gc_thread(struct f2fs_sb_info *);
-void stop_gc_thread(struct f2fs_sb_info *);
-block_t start_bidx_of_node(unsigned int, struct f2fs_inode_info *);
-int f2fs_gc(struct f2fs_sb_info *, bool);
-void build_gc_manager(struct f2fs_sb_info *);
+int start_gc_thread(struct f2fs_sb_info *sbi);
+void stop_gc_thread(struct f2fs_sb_info *sbi);
+block_t start_bidx_of_node(unsigned int node_ofs, struct inode *inode);
+int f2fs_gc(struct f2fs_sb_info *sbi, bool sync, bool background,
+			unsigned int segno);
+void build_gc_manager(struct f2fs_sb_info *sbi);
 
 /*
  * recovery.c
  */
-int recover_fsync_data(struct f2fs_sb_info *);
-bool space_for_roll_forward(struct f2fs_sb_info *);
+int recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only);
+bool space_for_roll_forward(struct f2fs_sb_info *sbi);
 
 /*
  * debug.c
@@ -1878,18 +2750,25 @@ struct f2fs_stat_info {
 	int main_area_segs, main_area_sections, main_area_zones;
 	unsigned long long hit_largest, hit_cached, hit_rbtree;
 	unsigned long long hit_total, total_ext;
-	int ext_tree, ext_node;
-	int ndirty_node, ndirty_dent, ndirty_dirs, ndirty_meta;
-	int nats, dirty_nats, sits, dirty_sits, fnids;
+	int ext_tree, zombie_tree, ext_node;
+	int ndirty_node, ndirty_dent, ndirty_meta, ndirty_data, ndirty_imeta;
+	int inmem_pages;
+	unsigned int ndirty_dirs, ndirty_files, ndirty_all;
+	int nats, dirty_nats, sits, dirty_sits;
+	int free_nids, avail_nids, alloc_nids;
 	int total_count, utilization;
-	int bg_gc, inmem_pages, wb_pages;
-	int inline_xattr, inline_inode, inline_dir;
-	unsigned int valid_count, valid_node_count, valid_inode_count;
+	int bg_gc, nr_wb_cp_data, nr_wb_data;
+	int nr_flushing, nr_flushed, nr_discarding, nr_discarded;
+	int nr_discard_cmd;
+	unsigned int undiscard_blks;
+	int inline_xattr, inline_inode, inline_dir, append, update, orphans;
+	int aw_cnt, max_aw_cnt, vw_cnt, max_vw_cnt;
+	unsigned int valid_count, valid_node_count, valid_inode_count, discard_blks;
 	unsigned int bimodal, avg_vblocks;
 	int util_free, util_valid, util_invalid;
 	int rsvd_segs, overp_segs;
 	int dirty_count, node_pages, meta_pages;
-	int prefree_count, call_count, cp_count;
+	int prefree_count, call_count, cp_count, bg_cp_count;
 	int tot_segs, node_segs, data_segs, free_segs, free_secs;
 	int bg_node_segs, bg_data_segs;
 	int tot_blks, data_blks, node_blks;
@@ -1910,10 +2789,11 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 }
 
 #define stat_inc_cp_count(si)		((si)->cp_count++)
+#define stat_inc_bg_cp_count(si)	((si)->bg_cp_count++)
 #define stat_inc_call_count(si)		((si)->call_count++)
 #define stat_inc_bggc_count(sbi)	((sbi)->bg_gc++)
-#define stat_inc_dirty_dir(sbi)		((sbi)->n_dirty_dirs++)
-#define stat_dec_dirty_dir(sbi)		((sbi)->n_dirty_dirs--)
+#define stat_inc_dirty_inode(sbi, type)	((sbi)->ndirty_inode[type]++)
+#define stat_dec_dirty_inode(sbi, type)	((sbi)->ndirty_inode[type]--)
 #define stat_inc_total_hit(sbi)		(atomic64_inc(&(sbi)->total_hit_ext))
 #define stat_inc_rbtree_node_hit(sbi)	(atomic64_inc(&(sbi)->read_hit_rbtree))
 #define stat_inc_largest_node_hit(sbi)	(atomic64_inc(&(sbi)->read_hit_largest))
@@ -1954,11 +2834,33 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 		((sbi)->block_count[(curseg)->alloc_type]++)
 #define stat_inc_inplace_blocks(sbi)					\
 		(atomic_inc(&(sbi)->inplace_count))
+#define stat_inc_atomic_write(inode)					\
+		(atomic_inc(&F2FS_I_SB(inode)->aw_cnt))
+#define stat_dec_atomic_write(inode)					\
+		(atomic_dec(&F2FS_I_SB(inode)->aw_cnt))
+#define stat_update_max_atomic_write(inode)				\
+	do {								\
+		int cur = atomic_read(&F2FS_I_SB(inode)->aw_cnt);	\
+		int max = atomic_read(&F2FS_I_SB(inode)->max_aw_cnt);	\
+		if (cur > max)						\
+			atomic_set(&F2FS_I_SB(inode)->max_aw_cnt, cur);	\
+	} while (0)
+#define stat_inc_volatile_write(inode)					\
+		(atomic_inc(&F2FS_I_SB(inode)->vw_cnt))
+#define stat_dec_volatile_write(inode)					\
+		(atomic_dec(&F2FS_I_SB(inode)->vw_cnt))
+#define stat_update_max_volatile_write(inode)				\
+	do {								\
+		int cur = atomic_read(&F2FS_I_SB(inode)->vw_cnt);	\
+		int max = atomic_read(&F2FS_I_SB(inode)->max_vw_cnt);	\
+		if (cur > max)						\
+			atomic_set(&F2FS_I_SB(inode)->max_vw_cnt, cur);	\
+	} while (0)
 #define stat_inc_seg_count(sbi, type, gc_type)				\
 	do {								\
 		struct f2fs_stat_info *si = F2FS_STAT(sbi);		\
-		(si)->tot_segs++;					\
-		if (type == SUM_TYPE_DATA) {				\
+		si->tot_segs++;						\
+		if ((type) == SUM_TYPE_DATA) {				\
 			si->data_segs++;				\
 			si->bg_data_segs += (gc_type == BG_GC) ? 1 : 0;	\
 		} else {						\
@@ -1968,14 +2870,14 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 	} while (0)
 
 #define stat_inc_tot_blk_count(si, blks)				\
-	(si->tot_blks += (blks))
+	((si)->tot_blks += (blks))
 
 #define stat_inc_data_blk_count(sbi, blks, gc_type)			\
 	do {								\
 		struct f2fs_stat_info *si = F2FS_STAT(sbi);		\
 		stat_inc_tot_blk_count(si, blks);			\
 		si->data_blks += (blks);				\
-		si->bg_data_blks += (gc_type == BG_GC) ? (blks) : 0;	\
+		si->bg_data_blks += ((gc_type) == BG_GC) ? (blks) : 0;	\
 	} while (0)
 
 #define stat_inc_node_blk_count(sbi, blks, gc_type)			\
@@ -1983,40 +2885,47 @@ static inline struct f2fs_stat_info *F2FS_STAT(struct f2fs_sb_info *sbi)
 		struct f2fs_stat_info *si = F2FS_STAT(sbi);		\
 		stat_inc_tot_blk_count(si, blks);			\
 		si->node_blks += (blks);				\
-		si->bg_node_blks += (gc_type == BG_GC) ? (blks) : 0;	\
+		si->bg_node_blks += ((gc_type) == BG_GC) ? (blks) : 0;	\
 	} while (0)
 
-int f2fs_build_stats(struct f2fs_sb_info *);
-void f2fs_destroy_stats(struct f2fs_sb_info *);
-void __init f2fs_create_root_stats(void);
+int f2fs_build_stats(struct f2fs_sb_info *sbi);
+void f2fs_destroy_stats(struct f2fs_sb_info *sbi);
+int __init f2fs_create_root_stats(void);
 void f2fs_destroy_root_stats(void);
 #else
-#define stat_inc_cp_count(si)
-#define stat_inc_call_count(si)
-#define stat_inc_bggc_count(si)
-#define stat_inc_dirty_dir(sbi)
-#define stat_dec_dirty_dir(sbi)
-#define stat_inc_total_hit(sb)
-#define stat_inc_rbtree_node_hit(sb)
-#define stat_inc_largest_node_hit(sbi)
-#define stat_inc_cached_node_hit(sbi)
-#define stat_inc_inline_xattr(inode)
-#define stat_dec_inline_xattr(inode)
-#define stat_inc_inline_inode(inode)
-#define stat_dec_inline_inode(inode)
-#define stat_inc_inline_dir(inode)
-#define stat_dec_inline_dir(inode)
-#define stat_inc_seg_type(sbi, curseg)
-#define stat_inc_block_count(sbi, curseg)
-#define stat_inc_inplace_blocks(sbi)
-#define stat_inc_seg_count(sbi, type, gc_type)
-#define stat_inc_tot_blk_count(si, blks)
-#define stat_inc_data_blk_count(sbi, blks, gc_type)
-#define stat_inc_node_blk_count(sbi, blks, gc_type)
+#define stat_inc_cp_count(si)				do { } while (0)
+#define stat_inc_bg_cp_count(si)			do { } while (0)
+#define stat_inc_call_count(si)				do { } while (0)
+#define stat_inc_bggc_count(si)				do { } while (0)
+#define stat_inc_dirty_inode(sbi, type)			do { } while (0)
+#define stat_dec_dirty_inode(sbi, type)			do { } while (0)
+#define stat_inc_total_hit(sb)				do { } while (0)
+#define stat_inc_rbtree_node_hit(sb)			do { } while (0)
+#define stat_inc_largest_node_hit(sbi)			do { } while (0)
+#define stat_inc_cached_node_hit(sbi)			do { } while (0)
+#define stat_inc_inline_xattr(inode)			do { } while (0)
+#define stat_dec_inline_xattr(inode)			do { } while (0)
+#define stat_inc_inline_inode(inode)			do { } while (0)
+#define stat_dec_inline_inode(inode)			do { } while (0)
+#define stat_inc_inline_dir(inode)			do { } while (0)
+#define stat_dec_inline_dir(inode)			do { } while (0)
+#define stat_inc_atomic_write(inode)			do { } while (0)
+#define stat_dec_atomic_write(inode)			do { } while (0)
+#define stat_update_max_atomic_write(inode)		do { } while (0)
+#define stat_inc_volatile_write(inode)			do { } while (0)
+#define stat_dec_volatile_write(inode)			do { } while (0)
+#define stat_update_max_volatile_write(inode)		do { } while (0)
+#define stat_inc_seg_type(sbi, curseg)			do { } while (0)
+#define stat_inc_block_count(sbi, curseg)		do { } while (0)
+#define stat_inc_inplace_blocks(sbi)			do { } while (0)
+#define stat_inc_seg_count(sbi, type, gc_type)		do { } while (0)
+#define stat_inc_tot_blk_count(si, blks)		do { } while (0)
+#define stat_inc_data_blk_count(sbi, blks, gc_type)	do { } while (0)
+#define stat_inc_node_blk_count(sbi, blks, gc_type)	do { } while (0)
 
 static inline int f2fs_build_stats(struct f2fs_sb_info *sbi) { return 0; }
 static inline void f2fs_destroy_stats(struct f2fs_sb_info *sbi) { }
-static inline void __init f2fs_create_root_stats(void) { }
+static inline int __init f2fs_create_root_stats(void) { return 0; }
 static inline void f2fs_destroy_root_stats(void) { }
 #endif
 
@@ -2035,63 +2944,89 @@ extern struct kmem_cache *inode_entry_slab;
 /*
  * inline.c
  */
-bool f2fs_may_inline_data(struct inode *);
-bool f2fs_may_inline_dentry(struct inode *);
-void read_inline_data(struct page *, struct page *);
-bool truncate_inline_inode(struct page *, u64);
-int f2fs_read_inline_data(struct inode *, struct page *);
-int f2fs_convert_inline_page(struct dnode_of_data *, struct page *);
-int f2fs_convert_inline_inode(struct inode *);
-int f2fs_write_inline_data(struct inode *, struct page *);
-bool recover_inline_data(struct inode *, struct page *);
-struct f2fs_dir_entry *find_in_inline_dir(struct inode *,
-				struct f2fs_filename *, struct page **);
-struct f2fs_dir_entry *f2fs_parent_inline_dir(struct inode *, struct page **);
-int make_empty_inline_dir(struct inode *inode, struct inode *, struct page *);
-int f2fs_add_inline_entry(struct inode *, const struct qstr *, struct inode *,
-						nid_t, umode_t);
-void f2fs_delete_inline_entry(struct f2fs_dir_entry *, struct page *,
-						struct inode *, struct inode *);
-bool f2fs_empty_inline_dir(struct inode *);
-int f2fs_read_inline_dir(struct file *, struct dir_context *,
-						struct f2fs_str *);
-int f2fs_inline_data_fiemap(struct inode *,
-		struct fiemap_extent_info *, __u64, __u64);
+bool f2fs_may_inline_data(struct inode *inode);
+bool f2fs_may_inline_dentry(struct inode *inode);
+void read_inline_data(struct page *page, struct page *ipage);
+void truncate_inline_inode(struct inode *inode, struct page *ipage, u64 from);
+int f2fs_read_inline_data(struct inode *inode, struct page *page);
+int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page);
+int f2fs_convert_inline_inode(struct inode *inode);
+int f2fs_write_inline_data(struct inode *inode, struct page *page);
+bool recover_inline_data(struct inode *inode, struct page *npage);
+struct f2fs_dir_entry *find_in_inline_dir(struct inode *dir,
+			struct fscrypt_name *fname, struct page **res_page);
+int make_empty_inline_dir(struct inode *inode, struct inode *parent,
+			struct page *ipage);
+int f2fs_add_inline_entry(struct inode *dir, const struct qstr *new_name,
+			const struct qstr *orig_name,
+			struct inode *inode, nid_t ino, umode_t mode);
+void f2fs_delete_inline_entry(struct f2fs_dir_entry *dentry, struct page *page,
+			struct inode *dir, struct inode *inode);
+bool f2fs_empty_inline_dir(struct inode *dir);
+int f2fs_read_inline_dir(struct file *file, struct dir_context *ctx,
+			struct fscrypt_str *fstr);
+int f2fs_inline_data_fiemap(struct inode *inode,
+			struct fiemap_extent_info *fieinfo,
+			__u64 start, __u64 len);
 
 /*
  * shrinker.c
  */
-unsigned long f2fs_shrink_count(struct shrinker *, struct shrink_control *);
-unsigned long f2fs_shrink_scan(struct shrinker *, struct shrink_control *);
-void f2fs_join_shrinker(struct f2fs_sb_info *);
-void f2fs_leave_shrinker(struct f2fs_sb_info *);
+unsigned long f2fs_shrink_count(struct shrinker *shrink,
+			struct shrink_control *sc);
+unsigned long f2fs_shrink_scan(struct shrinker *shrink,
+			struct shrink_control *sc);
+void f2fs_join_shrinker(struct f2fs_sb_info *sbi);
+void f2fs_leave_shrinker(struct f2fs_sb_info *sbi);
 
 /*
  * extent_cache.c
  */
-unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *, int);
-void f2fs_drop_largest_extent(struct inode *, pgoff_t);
-void f2fs_init_extent_tree(struct inode *, struct f2fs_extent *);
-unsigned int f2fs_destroy_extent_node(struct inode *);
-void f2fs_destroy_extent_tree(struct inode *);
-bool f2fs_lookup_extent_cache(struct inode *, pgoff_t, struct extent_info *);
-void f2fs_update_extent_cache(struct dnode_of_data *);
+struct rb_entry *__lookup_rb_tree(struct rb_root *root,
+				struct rb_entry *cached_re, unsigned int ofs);
+struct rb_node **__lookup_rb_tree_for_insert(struct f2fs_sb_info *sbi,
+				struct rb_root *root, struct rb_node **parent,
+				unsigned int ofs);
+struct rb_entry *__lookup_rb_tree_ret(struct rb_root *root,
+		struct rb_entry *cached_re, unsigned int ofs,
+		struct rb_entry **prev_entry, struct rb_entry **next_entry,
+		struct rb_node ***insert_p, struct rb_node **insert_parent,
+		bool force);
+bool __check_rb_tree_consistence(struct f2fs_sb_info *sbi,
+						struct rb_root *root);
+unsigned int f2fs_shrink_extent_tree(struct f2fs_sb_info *sbi, int nr_shrink);
+bool f2fs_init_extent_tree(struct inode *inode, struct f2fs_extent *i_ext);
+void f2fs_drop_extent_tree(struct inode *inode);
+unsigned int f2fs_destroy_extent_node(struct inode *inode);
+void f2fs_destroy_extent_tree(struct inode *inode);
+bool f2fs_lookup_extent_cache(struct inode *inode, pgoff_t pgofs,
+			struct extent_info *ei);
+void f2fs_update_extent_cache(struct dnode_of_data *dn);
 void f2fs_update_extent_cache_range(struct dnode_of_data *dn,
-						pgoff_t, block_t, unsigned int);
-void init_extent_cache_info(struct f2fs_sb_info *);
+			pgoff_t fofs, block_t blkaddr, unsigned int len);
+void init_extent_cache_info(struct f2fs_sb_info *sbi);
 int __init create_extent_cache(void);
 void destroy_extent_cache(void);
 
 /*
+ * sysfs.c
+ */
+int __init f2fs_init_sysfs(void);
+void f2fs_exit_sysfs(void);
+int f2fs_register_sysfs(struct f2fs_sb_info *sbi);
+void f2fs_unregister_sysfs(struct f2fs_sb_info *sbi);
+
+/*
  * crypto support
  */
-static inline int f2fs_encrypted_inode(struct inode *inode)
+static inline bool f2fs_encrypted_inode(struct inode *inode)
 {
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
 	return file_is_encrypt(inode);
-#else
-	return 0;
-#endif
+}
+
+static inline bool f2fs_encrypted_file(struct inode *inode)
+{
+	return f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode);
 }
 
 static inline void f2fs_set_encrypted_inode(struct inode *inode)
@@ -2103,101 +3038,79 @@ static inline void f2fs_set_encrypted_inode(struct inode *inode)
 
 static inline bool f2fs_bio_encrypted(struct bio *bio)
 {
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	return unlikely(bio->bi_private != NULL);
-#else
-	return false;
-#endif
+	return bio->bi_private != NULL;
 }
 
 static inline int f2fs_sb_has_crypto(struct super_block *sb)
 {
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
 	return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_ENCRYPT);
-#else
-	return 0;
-#endif
 }
 
-static inline bool f2fs_may_encrypt(struct inode *inode)
+static inline int f2fs_sb_mounted_blkzoned(struct super_block *sb)
 {
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	mode_t mode = inode->i_mode;
+	return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_BLKZONED);
+}
 
-	return (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode));
-#else
-	return 0;
-#endif
+static inline int f2fs_sb_has_extra_attr(struct super_block *sb)
+{
+	return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_EXTRA_ATTR);
 }
 
-/* crypto_policy.c */
-int f2fs_is_child_context_consistent_with_parent(struct inode *,
-							struct inode *);
-int f2fs_inherit_context(struct inode *, struct inode *, struct page *);
-int f2fs_process_policy(const struct f2fs_encryption_policy *, struct inode *);
-int f2fs_get_policy(struct inode *, struct f2fs_encryption_policy *);
-
-/* crypt.c */
-extern struct kmem_cache *f2fs_crypt_info_cachep;
-bool f2fs_valid_contents_enc_mode(uint32_t);
-uint32_t f2fs_validate_encryption_key_size(uint32_t, uint32_t);
-struct f2fs_crypto_ctx *f2fs_get_crypto_ctx(struct inode *);
-void f2fs_release_crypto_ctx(struct f2fs_crypto_ctx *);
-struct page *f2fs_encrypt(struct inode *, struct page *);
-int f2fs_decrypt(struct f2fs_crypto_ctx *, struct page *);
-int f2fs_decrypt_one(struct inode *, struct page *);
-void f2fs_end_io_crypto_work(struct f2fs_crypto_ctx *, struct bio *);
-
-/* crypto_key.c */
-void f2fs_free_encryption_info(struct inode *, struct f2fs_crypt_info *);
-
-/* crypto_fname.c */
-bool f2fs_valid_filenames_enc_mode(uint32_t);
-u32 f2fs_fname_crypto_round_up(u32, u32);
-int f2fs_fname_crypto_alloc_buffer(struct inode *, u32, struct f2fs_str *);
-int f2fs_fname_disk_to_usr(struct inode *, f2fs_hash_t *,
-			const struct f2fs_str *, struct f2fs_str *);
-int f2fs_fname_usr_to_disk(struct inode *, const struct qstr *,
-			struct f2fs_str *);
+static inline int f2fs_sb_has_project_quota(struct super_block *sb)
+{
+	return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_PRJQUOTA);
+}
 
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-void f2fs_restore_and_release_control_page(struct page **);
-void f2fs_restore_control_page(struct page *);
+static inline int f2fs_sb_has_inode_chksum(struct super_block *sb)
+{
+	return F2FS_HAS_FEATURE(sb, F2FS_FEATURE_INODE_CHKSUM);
+}
 
-int __init f2fs_init_crypto(void);
-int f2fs_crypto_initialize(void);
-void f2fs_exit_crypto(void);
+#ifdef CONFIG_BLK_DEV_ZONED
+static inline int get_blkz_type(struct f2fs_sb_info *sbi,
+			struct block_device *bdev, block_t blkaddr)
+{
+	unsigned int zno = blkaddr >> sbi->log_blocks_per_blkz;
+	int i;
 
-int f2fs_has_encryption_key(struct inode *);
+	for (i = 0; i < sbi->s_ndevs; i++)
+		if (FDEV(i).bdev == bdev)
+			return FDEV(i).blkz_type[zno];
+	return -EINVAL;
+}
+#endif
 
-int f2fs_get_encryption_info(struct inode *inode);
+static inline bool f2fs_discard_en(struct f2fs_sb_info *sbi)
+{
+	struct request_queue *q = bdev_get_queue(sbi->sb->s_bdev);
 
-void f2fs_fname_crypto_free_buffer(struct f2fs_str *);
-int f2fs_fname_setup_filename(struct inode *, const struct qstr *,
-				int lookup, struct f2fs_filename *);
-void f2fs_fname_free_filename(struct f2fs_filename *);
-#else
-static inline void f2fs_restore_and_release_control_page(struct page **p) { }
-static inline void f2fs_restore_control_page(struct page *p) { }
+	return blk_queue_discard(q) || f2fs_sb_mounted_blkzoned(sbi->sb);
+}
 
-static inline int __init f2fs_init_crypto(void) { return 0; }
-static inline void f2fs_exit_crypto(void) { }
+static inline void set_opt_mode(struct f2fs_sb_info *sbi, unsigned int mt)
+{
+	clear_opt(sbi, ADAPTIVE);
+	clear_opt(sbi, LFS);
 
-static inline int f2fs_has_encryption_key(struct inode *i) { return 0; }
-static inline int f2fs_get_encryption_info(struct inode *i) { return 0; }
-static inline void f2fs_fname_crypto_free_buffer(struct f2fs_str *p) { }
+	switch (mt) {
+	case F2FS_MOUNT_ADAPTIVE:
+		set_opt(sbi, ADAPTIVE);
+		break;
+	case F2FS_MOUNT_LFS:
+		set_opt(sbi, LFS);
+		break;
+	}
+}
 
-static inline int f2fs_fname_setup_filename(struct inode *dir,
-					const struct qstr *iname,
-					int lookup, struct f2fs_filename *fname)
+static inline bool f2fs_may_encrypt(struct inode *inode)
 {
-	memset(fname, 0, sizeof(struct f2fs_filename));
-	fname->usr_fname = iname;
-	fname->disk_name.name = (unsigned char *)iname->name;
-	fname->disk_name.len = iname->len;
+#ifdef CONFIG_F2FS_FS_ENCRYPTION
+	umode_t mode = inode->i_mode;
+
+	return (S_ISREG(mode) || S_ISDIR(mode) || S_ISLNK(mode));
+#else
 	return 0;
+#endif
 }
 
-static inline void f2fs_fname_free_filename(struct f2fs_filename *fname) { }
-#endif
 #endif
diff --git a/fs/f2fs/f2fs_crypto.h b/fs/f2fs/f2fs_crypto.h
deleted file mode 100644
index f113f1a1e8c1..000000000000
--- a/fs/f2fs/f2fs_crypto.h
+++ /dev/null
@@ -1,150 +0,0 @@
-/*
- * linux/fs/f2fs/f2fs_crypto.h
- *
- * Copied from linux/fs/ext4/ext4_crypto.h
- *
- * Copyright (C) 2015, Google, Inc.
- *
- * This contains encryption header content for f2fs
- *
- * Written by Michael Halcrow, 2015.
- * Modified by Jaegeuk Kim, 2015.
- */
-#ifndef _F2FS_CRYPTO_H
-#define _F2FS_CRYPTO_H
-
-#include <linux/fs.h>
-
-#define F2FS_KEY_DESCRIPTOR_SIZE	8
-
-/* Policy provided via an ioctl on the topmost directory */
-struct f2fs_encryption_policy {
-	char version;
-	char contents_encryption_mode;
-	char filenames_encryption_mode;
-	char flags;
-	char master_key_descriptor[F2FS_KEY_DESCRIPTOR_SIZE];
-} __attribute__((__packed__));
-
-#define F2FS_ENCRYPTION_CONTEXT_FORMAT_V1	1
-#define F2FS_KEY_DERIVATION_NONCE_SIZE		16
-
-#define F2FS_POLICY_FLAGS_PAD_4		0x00
-#define F2FS_POLICY_FLAGS_PAD_8		0x01
-#define F2FS_POLICY_FLAGS_PAD_16	0x02
-#define F2FS_POLICY_FLAGS_PAD_32	0x03
-#define F2FS_POLICY_FLAGS_PAD_MASK	0x03
-#define F2FS_POLICY_FLAGS_VALID		0x03
-
-/**
- * Encryption context for inode
- *
- * Protector format:
- *  1 byte: Protector format (1 = this version)
- *  1 byte: File contents encryption mode
- *  1 byte: File names encryption mode
- *  1 byte: Flags
- *  8 bytes: Master Key descriptor
- *  16 bytes: Encryption Key derivation nonce
- */
-struct f2fs_encryption_context {
-	char format;
-	char contents_encryption_mode;
-	char filenames_encryption_mode;
-	char flags;
-	char master_key_descriptor[F2FS_KEY_DESCRIPTOR_SIZE];
-	char nonce[F2FS_KEY_DERIVATION_NONCE_SIZE];
-} __attribute__((__packed__));
-
-/* Encryption parameters */
-#define F2FS_XTS_TWEAK_SIZE 16
-#define F2FS_AES_128_ECB_KEY_SIZE 16
-#define F2FS_AES_256_GCM_KEY_SIZE 32
-#define F2FS_AES_256_CBC_KEY_SIZE 32
-#define F2FS_AES_256_CTS_KEY_SIZE 32
-#define F2FS_AES_256_XTS_KEY_SIZE 64
-#define F2FS_MAX_KEY_SIZE 64
-
-#define F2FS_KEY_DESC_PREFIX "f2fs:"
-#define F2FS_KEY_DESC_PREFIX_SIZE 5
-
-struct f2fs_encryption_key {
-	__u32 mode;
-	char raw[F2FS_MAX_KEY_SIZE];
-	__u32 size;
-} __attribute__((__packed__));
-
-struct f2fs_crypt_info {
-	char		ci_data_mode;
-	char		ci_filename_mode;
-	char		ci_flags;
-	struct crypto_ablkcipher *ci_ctfm;
-	char		ci_master_key[F2FS_KEY_DESCRIPTOR_SIZE];
-};
-
-#define F2FS_CTX_REQUIRES_FREE_ENCRYPT_FL             0x00000001
-#define F2FS_WRITE_PATH_FL			      0x00000002
-
-struct f2fs_crypto_ctx {
-	union {
-		struct {
-			struct page *bounce_page;       /* Ciphertext page */
-			struct page *control_page;      /* Original page  */
-		} w;
-		struct {
-			struct bio *bio;
-			struct work_struct work;
-		} r;
-		struct list_head free_list;     /* Free list */
-	};
-	char flags;                      /* Flags */
-};
-
-struct f2fs_completion_result {
-	struct completion completion;
-	int res;
-};
-
-#define DECLARE_F2FS_COMPLETION_RESULT(ecr) \
-	struct f2fs_completion_result ecr = { \
-		COMPLETION_INITIALIZER((ecr).completion), 0 }
-
-static inline int f2fs_encryption_key_size(int mode)
-{
-	switch (mode) {
-	case F2FS_ENCRYPTION_MODE_AES_256_XTS:
-		return F2FS_AES_256_XTS_KEY_SIZE;
-	case F2FS_ENCRYPTION_MODE_AES_256_GCM:
-		return F2FS_AES_256_GCM_KEY_SIZE;
-	case F2FS_ENCRYPTION_MODE_AES_256_CBC:
-		return F2FS_AES_256_CBC_KEY_SIZE;
-	case F2FS_ENCRYPTION_MODE_AES_256_CTS:
-		return F2FS_AES_256_CTS_KEY_SIZE;
-	default:
-		BUG();
-	}
-	return 0;
-}
-
-#define F2FS_FNAME_NUM_SCATTER_ENTRIES	4
-#define F2FS_CRYPTO_BLOCK_SIZE		16
-#define F2FS_FNAME_CRYPTO_DIGEST_SIZE	32
-
-/**
- * For encrypted symlinks, the ciphertext length is stored at the beginning
- * of the string in little-endian format.
- */
-struct f2fs_encrypted_symlink_data {
-	__le16 len;
-	char encrypted_path[1];
-} __attribute__((__packed__));
-
-/**
- * This function is used to calculate the disk space required to
- * store a filename of length l in encrypted symlink format.
- */
-static inline u32 encrypted_symlink_data_len(u32 l)
-{
-	return (l + sizeof(struct f2fs_encrypted_symlink_data) - 1);
-}
-#endif	/* _F2FS_CRYPTO_H */
diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c
index 01eed94b01ea..a9e1655a6bf8 100644
--- a/fs/f2fs/file.c
+++ b/fs/f2fs/file.c
@@ -21,6 +21,9 @@
 #include <linux/mount.h>
 #include <linux/pagevec.h>
 #include <linux/random.h>
+#include <linux/uio.h>
+#include <linux/uuid.h>
+#include <linux/file.h>
 
 #include "f2fs.h"
 #include "node.h"
@@ -31,6 +34,19 @@
 #include "trace.h"
 #include <trace/events/f2fs.h>
 
+static int f2fs_filemap_fault(struct vm_area_struct *vma,
+					struct vm_fault *vmf)
+{
+	struct inode *inode = file_inode(vma->vm_file);
+	int err;
+
+	down_read(&F2FS_I(inode)->i_mmap_sem);
+	err = filemap_fault(vma, vmf);
+	up_read(&F2FS_I(inode)->i_mmap_sem);
+
+	return err;
+}
+
 static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 						struct vm_fault *vmf)
 {
@@ -40,8 +56,6 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 	struct dnode_of_data dn;
 	int err;
 
-	f2fs_balance_fs(sbi);
-
 	sb_start_pagefault(inode->i_sb);
 
 	f2fs_bug_on(sbi, f2fs_has_inline_data(inode));
@@ -57,14 +71,17 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 	f2fs_put_dnode(&dn);
 	f2fs_unlock_op(sbi);
 
+	f2fs_balance_fs(sbi, dn.node_changed);
+
 	file_update_time(vma->vm_file);
+	down_read(&F2FS_I(inode)->i_mmap_sem);
 	lock_page(page);
 	if (unlikely(page->mapping != inode->i_mapping ||
 			page_offset(page) > i_size_read(inode) ||
 			!PageUptodate(page))) {
 		unlock_page(page);
 		err = -EFAULT;
-		goto out;
+		goto out_sem;
 	}
 
 	/*
@@ -74,33 +91,37 @@ static int f2fs_vm_page_mkwrite(struct vm_area_struct *vma,
 		goto mapped;
 
 	/* page is wholly or partially inside EOF */
-	if (((loff_t)(page->index + 1) << PAGE_CACHE_SHIFT) >
+	if (((loff_t)(page->index + 1) << PAGE_SHIFT) >
 						i_size_read(inode)) {
 		unsigned offset;
-		offset = i_size_read(inode) & ~PAGE_CACHE_MASK;
-		zero_user_segment(page, offset, PAGE_CACHE_SIZE);
+		offset = i_size_read(inode) & ~PAGE_MASK;
+		zero_user_segment(page, offset, PAGE_SIZE);
 	}
 	set_page_dirty(page);
-	SetPageUptodate(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
+
+	f2fs_update_iostat(sbi, APP_MAPPED_IO, F2FS_BLKSIZE);
 
 	trace_f2fs_vm_page_mkwrite(page, DATA);
 mapped:
 	/* fill the page */
-	f2fs_wait_on_page_writeback(page, DATA);
+	f2fs_wait_on_page_writeback(page, DATA, false);
 
 	/* wait for GCed encrypted page writeback */
-	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
-		f2fs_wait_on_encrypted_page_writeback(sbi, dn.data_blkaddr);
+	if (f2fs_encrypted_file(inode))
+		f2fs_wait_on_block_writeback(sbi, dn.data_blkaddr);
 
-	/* if gced page is attached, don't write to cold segment */
-	clear_cold_data(page);
+out_sem:
+	up_read(&F2FS_I(inode)->i_mmap_sem);
 out:
 	sb_end_pagefault(inode->i_sb);
+	f2fs_update_time(sbi, REQ_TIME);
 	return block_page_mkwrite_return(err);
 }
 
 static const struct vm_operations_struct f2fs_file_vm_ops = {
-	.fault		= filemap_fault,
+	.fault		= f2fs_filemap_fault,
 	.map_pages	= filemap_map_pages,
 	.page_mkwrite	= f2fs_vm_page_mkwrite,
 };
@@ -115,11 +136,6 @@ static int get_parent_ino(struct inode *inode, nid_t *pino)
 	if (!dentry)
 		return 0;
 
-	if (update_dent_inode(inode, inode, &dentry->d_name)) {
-		dput(dentry);
-		return 0;
-	}
-
 	*pino = parent_ino(dentry);
 	dput(dentry);
 	return 1;
@@ -132,7 +148,7 @@ static inline bool need_do_checkpoint(struct inode *inode)
 
 	if (!S_ISREG(inode->i_mode) || inode->i_nlink != 1)
 		need_cp = true;
-	else if (file_enc_name(inode) && need_dentry_mark(sbi, inode->i_ino))
+	else if (is_sbi_flag_set(sbi, SBI_NEED_CP))
 		need_cp = true;
 	else if (file_wrong_pino(inode))
 		need_cp = true;
@@ -140,8 +156,6 @@ static inline bool need_do_checkpoint(struct inode *inode)
 		need_cp = true;
 	else if (!is_checkpointed_node(sbi, F2FS_I(inode)->i_pino))
 		need_cp = true;
-	else if (F2FS_I(inode)->xattr_ver == cur_cp_version(F2FS_CKPT(sbi)))
-		need_cp = true;
 	else if (test_opt(sbi, FASTBOOT))
 		need_cp = true;
 	else if (sbi->active_logs == 2)
@@ -167,24 +181,18 @@ static void try_to_fix_pino(struct inode *inode)
 	nid_t pino;
 
 	down_write(&fi->i_sem);
-	fi->xattr_ver = 0;
 	if (file_wrong_pino(inode) && inode->i_nlink == 1 &&
 			get_parent_ino(inode, &pino)) {
-		fi->i_pino = pino;
+		f2fs_i_pino_write(inode, pino);
 		file_got_pino(inode);
-		up_write(&fi->i_sem);
-
-		mark_inode_dirty_sync(inode);
-		f2fs_write_inode(inode, NULL);
-	} else {
-		up_write(&fi->i_sem);
 	}
+	up_write(&fi->i_sem);
 }
 
-int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
+static int f2fs_do_sync_file(struct file *file, loff_t start, loff_t end,
+						int datasync, bool atomic)
 {
 	struct inode *inode = file->f_mapping->host;
-	struct f2fs_inode_info *fi = F2FS_I(inode);
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	nid_t ino = inode->i_ino;
 	int ret = 0;
@@ -201,10 +209,10 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	trace_f2fs_sync_file_enter(inode);
 
 	/* if fdatasync is triggered, let's do in-place-update */
-	if (get_dirty_pages(inode) <= SM_I(sbi)->min_fsync_blocks)
-		set_inode_flag(fi, FI_NEED_IPU);
+	if (datasync || get_dirty_pages(inode) <= SM_I(sbi)->min_fsync_blocks)
+		set_inode_flag(inode, FI_NEED_IPU);
 	ret = filemap_write_and_wait_range(inode->i_mapping, start, end);
-	clear_inode_flag(fi, FI_NEED_IPU);
+	clear_inode_flag(inode, FI_NEED_IPU);
 
 	if (ret) {
 		trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
@@ -212,7 +220,7 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	}
 
 	/* if the inode is dirty, let's recover all the time */
-	if (!datasync) {
+	if (!f2fs_skip_inode_update(inode, datasync)) {
 		f2fs_write_inode(inode, NULL);
 		goto go_write;
 	}
@@ -220,29 +228,26 @@ int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
 	/*
 	 * if there is no written data, don't waste time to write recovery info.
 	 */
-	if (!is_inode_flag_set(fi, FI_APPEND_WRITE) &&
+	if (!is_inode_flag_set(inode, FI_APPEND_WRITE) &&
 			!exist_written_data(sbi, ino, APPEND_INO)) {
 
 		/* it may call write_inode just prior to fsync */
 		if (need_inode_page_update(sbi, ino))
 			goto go_write;
 
-		if (is_inode_flag_set(fi, FI_UPDATE_WRITE) ||
+		if (is_inode_flag_set(inode, FI_UPDATE_WRITE) ||
 				exist_written_data(sbi, ino, UPDATE_INO))
 			goto flush_out;
 		goto out;
 	}
 go_write:
-	/* guarantee free sections for fsync */
-	f2fs_balance_fs(sbi);
-
 	/*
 	 * Both of fdatasync() and fsync() are able to be recovered from
 	 * sudden-power-off.
 	 */
-	down_read(&fi->i_sem);
+	down_read(&F2FS_I(inode)->i_sem);
 	need_cp = need_do_checkpoint(inode);
-	up_read(&fi->i_sem);
+	up_read(&F2FS_I(inode)->i_sem);
 
 	if (need_cp) {
 		/* all the dirty node pages should be flushed for POR */
@@ -253,40 +258,61 @@ go_write:
 		 * will be used only for fsynced inodes after checkpoint.
 		 */
 		try_to_fix_pino(inode);
-		clear_inode_flag(fi, FI_APPEND_WRITE);
-		clear_inode_flag(fi, FI_UPDATE_WRITE);
+		clear_inode_flag(inode, FI_APPEND_WRITE);
+		clear_inode_flag(inode, FI_UPDATE_WRITE);
 		goto out;
 	}
 sync_nodes:
-	sync_node_pages(sbi, ino, &wbc);
+	ret = fsync_node_pages(sbi, inode, &wbc, atomic);
+	if (ret)
+		goto out;
 
 	/* if cp_error was enabled, we should avoid infinite loop */
-	if (unlikely(f2fs_cp_error(sbi)))
+	if (unlikely(f2fs_cp_error(sbi))) {
+		ret = -EIO;
 		goto out;
+	}
 
 	if (need_inode_block_update(sbi, ino)) {
-		mark_inode_dirty_sync(inode);
+		f2fs_mark_inode_dirty_sync(inode, true);
 		f2fs_write_inode(inode, NULL);
 		goto sync_nodes;
 	}
 
-	ret = wait_on_node_pages_writeback(sbi, ino);
-	if (ret)
-		goto out;
+	/*
+	 * If it's atomic_write, it's just fine to keep write ordering. So
+	 * here we don't need to wait for node write completion, since we use
+	 * node chain which serializes node blocks. If one of node writes are
+	 * reordered, we can see simply broken chain, resulting in stopping
+	 * roll-forward recovery. It means we'll recover all or none node blocks
+	 * given fsync mark.
+	 */
+	if (!atomic) {
+		ret = wait_on_node_pages_writeback(sbi, ino);
+		if (ret)
+			goto out;
+	}
 
 	/* once recovery info is written, don't need to tack this */
-	remove_dirty_inode(sbi, ino, APPEND_INO);
-	clear_inode_flag(fi, FI_APPEND_WRITE);
+	remove_ino_entry(sbi, ino, APPEND_INO);
+	clear_inode_flag(inode, FI_APPEND_WRITE);
 flush_out:
-	remove_dirty_inode(sbi, ino, UPDATE_INO);
-	clear_inode_flag(fi, FI_UPDATE_WRITE);
-	ret = f2fs_issue_flush(sbi);
+	remove_ino_entry(sbi, ino, UPDATE_INO);
+	clear_inode_flag(inode, FI_UPDATE_WRITE);
+	if (!atomic)
+		ret = f2fs_issue_flush(sbi);
+	f2fs_update_time(sbi, REQ_TIME);
 out:
 	trace_f2fs_sync_file_exit(inode, need_cp, datasync, ret);
 	f2fs_trace_ios(NULL, 1);
 	return ret;
 }
 
+int f2fs_sync_file(struct file *file, loff_t start, loff_t end, int datasync)
+{
+	return f2fs_do_sync_file(file, start, end, datasync, false);
+}
+
 static pgoff_t __get_first_dirty_index(struct address_space *mapping,
 						pgoff_t pgofs, int whence)
 {
@@ -300,7 +326,7 @@ static pgoff_t __get_first_dirty_index(struct address_space *mapping,
 	pagevec_init(&pvec, 0);
 	nr_pages = pagevec_lookup_tag(&pvec, mapping, &pgofs,
 					PAGECACHE_TAG_DIRTY, 1);
-	pgofs = nr_pages ? pvec.pages[0]->index : LONG_MAX;
+	pgofs = nr_pages ? pvec.pages[0]->index : ULONG_MAX;
 	pagevec_release(&pvec);
 	return pgofs;
 }
@@ -332,7 +358,7 @@ static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
 	loff_t isize;
 	int err = 0;
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 
 	isize = i_size_read(inode);
 	if (offset >= isize)
@@ -345,34 +371,34 @@ static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
 		goto found;
 	}
 
-	pgofs = (pgoff_t)(offset >> PAGE_CACHE_SHIFT);
+	pgofs = (pgoff_t)(offset >> PAGE_SHIFT);
 
 	dirty = __get_first_dirty_index(inode->i_mapping, pgofs, whence);
 
-	for (; data_ofs < isize; data_ofs = (loff_t)pgofs << PAGE_CACHE_SHIFT) {
+	for (; data_ofs < isize; data_ofs = (loff_t)pgofs << PAGE_SHIFT) {
 		set_new_dnode(&dn, inode, NULL, NULL, 0);
-		err = get_dnode_of_data(&dn, pgofs, LOOKUP_NODE_RA);
+		err = get_dnode_of_data(&dn, pgofs, LOOKUP_NODE);
 		if (err && err != -ENOENT) {
 			goto fail;
 		} else if (err == -ENOENT) {
 			/* direct node does not exists */
 			if (whence == SEEK_DATA) {
-				pgofs = PGOFS_OF_NEXT_DNODE(pgofs,
-							F2FS_I(inode));
+				pgofs = get_next_page_offset(&dn, pgofs);
 				continue;
 			} else {
 				goto found;
 			}
 		}
 
-		end_offset = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
+		end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
 
 		/* find data/hole in dnode block */
 		for (; dn.ofs_in_node < end_offset;
 				dn.ofs_in_node++, pgofs++,
-				data_ofs = (loff_t)pgofs << PAGE_CACHE_SHIFT) {
+				data_ofs = (loff_t)pgofs << PAGE_SHIFT) {
 			block_t blkaddr;
-			blkaddr = datablock_addr(dn.node_page, dn.ofs_in_node);
+			blkaddr = datablock_addr(dn.inode,
+					dn.node_page, dn.ofs_in_node);
 
 			if (__found_offset(blkaddr, dirty, pgofs, whence)) {
 				f2fs_put_dnode(&dn);
@@ -387,10 +413,10 @@ static loff_t f2fs_seek_block(struct file *file, loff_t offset, int whence)
 found:
 	if (whence == SEEK_HOLE && data_ofs > isize)
 		data_ofs = isize;
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return vfs_setpos(file, data_ofs, maxbytes);
 fail:
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 	return -ENXIO;
 }
 
@@ -418,19 +444,12 @@ static loff_t f2fs_llseek(struct file *file, loff_t offset, int whence)
 static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct inode *inode = file_inode(file);
-
-	if (f2fs_encrypted_inode(inode)) {
-		int err = f2fs_get_encryption_info(inode);
-		if (err)
-			return 0;
-	}
+	int err;
 
 	/* we don't need to use inline_data strictly */
-	if (f2fs_has_inline_data(inode)) {
-		int err = f2fs_convert_inline_inode(inode);
-		if (err)
-			return err;
-	}
+	err = f2fs_convert_inline_inode(inode);
+	if (err)
+		return err;
 
 	file_accessed(file);
 	vma->vm_ops = &f2fs_file_vm_ops;
@@ -439,14 +458,23 @@ static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma)
 
 static int f2fs_file_open(struct inode *inode, struct file *filp)
 {
-	int ret = generic_file_open(inode, filp);
+	struct dentry *dir;
 
-	if (!ret && f2fs_encrypted_inode(inode)) {
-		ret = f2fs_get_encryption_info(inode);
+	if (f2fs_encrypted_inode(inode)) {
+		int ret = fscrypt_get_encryption_info(inode);
 		if (ret)
-			ret = -EACCES;
+			return -EACCES;
+		if (!fscrypt_has_encryption_key(inode))
+			return -ENOKEY;
 	}
-	return ret;
+	dir = dget_parent(file_dentry(filp));
+	if (f2fs_encrypted_inode(d_inode(dir)) &&
+			!fscrypt_has_permitted_context(d_inode(dir), inode)) {
+		dput(dir);
+		return -EPERM;
+	}
+	dput(dir);
+	return dquot_file_open(inode, filp);
 }
 
 int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
@@ -455,9 +483,13 @@ int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 	struct f2fs_node *raw_node;
 	int nr_free = 0, ofs = dn->ofs_in_node, len = count;
 	__le32 *addr;
+	int base = 0;
+
+	if (IS_INODE(dn->node_page) && f2fs_has_extra_attr(dn->inode))
+		base = get_extra_isize(dn->inode);
 
 	raw_node = F2FS_NODE(dn->node_page);
-	addr = blkaddr_in_node(raw_node) + ofs;
+	addr = blkaddr_in_node(raw_node) + base + ofs;
 
 	for (; count > 0; count--, addr++, dn->ofs_in_node++) {
 		block_t blkaddr = le32_to_cpu(*addr);
@@ -468,8 +500,7 @@ int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 		set_data_blkaddr(dn);
 		invalidate_blocks(sbi, blkaddr);
 		if (dn->ofs_in_node == 0 && IS_INODE(dn->node_page))
-			clear_inode_flag(F2FS_I(dn->inode),
-						FI_FIRST_BLOCK_WRITTEN);
+			clear_inode_flag(dn->inode, FI_FIRST_BLOCK_WRITTEN);
 		nr_free++;
 	}
 
@@ -480,14 +511,13 @@ int truncate_data_blocks_range(struct dnode_of_data *dn, int count)
 		 * we will invalidate all blkaddr in the whole range.
 		 */
 		fofs = start_bidx_of_node(ofs_of_node(dn->node_page),
-						F2FS_I(dn->inode)) + ofs;
+							dn->inode) + ofs;
 		f2fs_update_extent_cache_range(dn, fofs, 0, len);
 		dec_valid_block_count(sbi, dn->inode, nr_free);
-		set_page_dirty(dn->node_page);
-		sync_inode_page(dn);
 	}
 	dn->ofs_in_node = ofs;
 
+	f2fs_update_time(sbi, REQ_TIME);
 	trace_f2fs_truncate_data_blocks_range(dn->inode, dn->nid,
 					 dn->ofs_in_node, nr_free);
 	return nr_free;
@@ -501,8 +531,8 @@ void truncate_data_blocks(struct dnode_of_data *dn)
 static int truncate_partial_data_page(struct inode *inode, u64 from,
 								bool cache_only)
 {
-	unsigned offset = from & (PAGE_CACHE_SIZE - 1);
-	pgoff_t index = from >> PAGE_CACHE_SHIFT;
+	unsigned offset = from & (PAGE_SIZE - 1);
+	pgoff_t index = from >> PAGE_SHIFT;
 	struct address_space *mapping = inode->i_mapping;
 	struct page *page;
 
@@ -510,7 +540,7 @@ static int truncate_partial_data_page(struct inode *inode, u64 from,
 		return 0;
 
 	if (cache_only) {
-		page = f2fs_grab_cache_page(mapping, index, false);
+		page = find_lock_page(mapping, index);
 		if (page && PageUptodate(page))
 			goto truncate_out;
 		f2fs_put_page(page, 1);
@@ -519,11 +549,14 @@ static int truncate_partial_data_page(struct inode *inode, u64 from,
 
 	page = get_lock_data_page(inode, index, true);
 	if (IS_ERR(page))
-		return 0;
+		return PTR_ERR(page) == -ENOENT ? 0 : PTR_ERR(page);
 truncate_out:
-	f2fs_wait_on_page_writeback(page, DATA);
-	zero_user(page, offset, PAGE_CACHE_SIZE - offset);
-	if (!cache_only || !f2fs_encrypted_inode(inode) || !S_ISREG(inode->i_mode))
+	f2fs_wait_on_page_writeback(page, DATA, true);
+	zero_user(page, offset, PAGE_SIZE - offset);
+
+	/* An encrypted inode should have a key and truncate the last page. */
+	f2fs_bug_on(F2FS_I_SB(inode), cache_only && f2fs_encrypted_inode(inode));
+	if (!cache_only)
 		set_page_dirty(page);
 	f2fs_put_page(page, 1);
 	return 0;
@@ -543,6 +576,9 @@ int truncate_blocks(struct inode *inode, u64 from, bool lock)
 
 	free_from = (pgoff_t)F2FS_BYTES_TO_BLK(from + blocksize - 1);
 
+	if (free_from >= sbi->max_file_blocks)
+		goto free_partial;
+
 	if (lock)
 		f2fs_lock_op(sbi);
 
@@ -553,22 +589,21 @@ int truncate_blocks(struct inode *inode, u64 from, bool lock)
 	}
 
 	if (f2fs_has_inline_data(inode)) {
-		if (truncate_inline_inode(ipage, from))
-			set_page_dirty(ipage);
+		truncate_inline_inode(inode, ipage, from);
 		f2fs_put_page(ipage, 1);
 		truncate_page = true;
 		goto out;
 	}
 
 	set_new_dnode(&dn, inode, ipage, NULL, 0);
-	err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE);
+	err = get_dnode_of_data(&dn, free_from, LOOKUP_NODE_RA);
 	if (err) {
 		if (err == -ENOENT)
 			goto free_next;
 		goto out;
 	}
 
-	count = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
+	count = ADDRS_PER_PAGE(dn.node_page, inode);
 
 	count -= dn.ofs_in_node;
 	f2fs_bug_on(sbi, count < 0);
@@ -584,7 +619,7 @@ free_next:
 out:
 	if (lock)
 		f2fs_unlock_op(sbi);
-
+free_partial:
 	/* lastly zero out the first data page */
 	if (!err)
 		err = truncate_partial_data_page(inode, from, truncate_page);
@@ -593,7 +628,7 @@ out:
 	return err;
 }
 
-int f2fs_truncate(struct inode *inode, bool lock)
+int f2fs_truncate(struct inode *inode)
 {
 	int err;
 
@@ -603,19 +638,25 @@ int f2fs_truncate(struct inode *inode, bool lock)
 
 	trace_f2fs_truncate(inode);
 
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(F2FS_I_SB(inode), FAULT_TRUNCATE)) {
+		f2fs_show_injection_info(FAULT_TRUNCATE);
+		return -EIO;
+	}
+#endif
 	/* we should check inline_data size */
-	if (f2fs_has_inline_data(inode) && !f2fs_may_inline_data(inode)) {
+	if (!f2fs_may_inline_data(inode)) {
 		err = f2fs_convert_inline_inode(inode);
 		if (err)
 			return err;
 	}
 
-	err = truncate_blocks(inode, i_size_read(inode), lock);
+	err = truncate_blocks(inode, i_size_read(inode), true);
 	if (err)
 		return err;
 
-	inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-	mark_inode_dirty(inode);
+	inode->i_mtime = inode->i_ctime = current_time(inode);
+	f2fs_mark_inode_dirty_sync(inode, false);
 	return 0;
 }
 
@@ -624,14 +665,17 @@ int f2fs_getattr(struct vfsmount *mnt,
 {
 	struct inode *inode = d_inode(dentry);
 	generic_fillattr(inode, stat);
-	stat->blocks <<= 3;
+
+	/* we need to show initial sectors used for inline_data/dentries */
+	if (f2fs_has_inline_data(inode) || f2fs_has_inline_dentry(inode))
+		stat->blocks += (stat->size + 511) >> 9;
+
 	return 0;
 }
 
 #ifdef CONFIG_F2FS_FS_POSIX_ACL
 static void __setattr_copy(struct inode *inode, const struct iattr *attr)
 {
-	struct f2fs_inode_info *fi = F2FS_I(inode);
 	unsigned int ia_valid = attr->ia_valid;
 
 	if (ia_valid & ATTR_UID)
@@ -652,7 +696,7 @@ static void __setattr_copy(struct inode *inode, const struct iattr *attr)
 
 		if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
 			mode &= ~S_ISGID;
-		set_acl_inode(fi, mode);
+		set_acl_inode(inode, mode);
 	}
 }
 #else
@@ -662,45 +706,80 @@ static void __setattr_copy(struct inode *inode, const struct iattr *attr)
 int f2fs_setattr(struct dentry *dentry, struct iattr *attr)
 {
 	struct inode *inode = d_inode(dentry);
-	struct f2fs_inode_info *fi = F2FS_I(inode);
 	int err;
+	bool size_changed = false;
 
 	err = inode_change_ok(inode, attr);
 	if (err)
 		return err;
 
+	if (is_quota_modification(inode, attr)) {
+		err = dquot_initialize(inode);
+		if (err)
+			return err;
+	}
+	if ((attr->ia_valid & ATTR_UID &&
+		!uid_eq(attr->ia_uid, inode->i_uid)) ||
+		(attr->ia_valid & ATTR_GID &&
+		!gid_eq(attr->ia_gid, inode->i_gid))) {
+		err = dquot_transfer(inode, attr);
+		if (err)
+			return err;
+	}
+
 	if (attr->ia_valid & ATTR_SIZE) {
-		if (f2fs_encrypted_inode(inode) &&
-				f2fs_get_encryption_info(inode))
-			return -EACCES;
+		if (f2fs_encrypted_inode(inode)) {
+			err = fscrypt_get_encryption_info(inode);
+			if (err)
+				return err;
+			if (!fscrypt_has_encryption_key(inode))
+				return -ENOKEY;
+		}
 
 		if (attr->ia_size <= i_size_read(inode)) {
+			down_write(&F2FS_I(inode)->i_mmap_sem);
 			truncate_setsize(inode, attr->ia_size);
-			err = f2fs_truncate(inode, true);
+			err = f2fs_truncate(inode);
+			up_write(&F2FS_I(inode)->i_mmap_sem);
 			if (err)
 				return err;
-			f2fs_balance_fs(F2FS_I_SB(inode));
 		} else {
 			/*
 			 * do not trim all blocks after i_size if target size is
 			 * larger than i_size.
 			 */
+			down_write(&F2FS_I(inode)->i_mmap_sem);
 			truncate_setsize(inode, attr->ia_size);
-			inode->i_mtime = inode->i_ctime = CURRENT_TIME;
+			up_write(&F2FS_I(inode)->i_mmap_sem);
+
+			/* should convert inline inode here */
+			if (!f2fs_may_inline_data(inode)) {
+				err = f2fs_convert_inline_inode(inode);
+				if (err)
+					return err;
+			}
+			inode->i_mtime = inode->i_ctime = current_time(inode);
 		}
+
+		size_changed = true;
 	}
 
 	__setattr_copy(inode, attr);
 
 	if (attr->ia_valid & ATTR_MODE) {
 		err = posix_acl_chmod(inode, get_inode_mode(inode));
-		if (err || is_inode_flag_set(fi, FI_ACL_MODE)) {
-			inode->i_mode = fi->i_acl_mode;
-			clear_inode_flag(fi, FI_ACL_MODE);
+		if (err || is_inode_flag_set(inode, FI_ACL_MODE)) {
+			inode->i_mode = F2FS_I(inode)->i_acl_mode;
+			clear_inode_flag(inode, FI_ACL_MODE);
 		}
 	}
 
-	mark_inode_dirty(inode);
+	/* file size may changed here */
+	f2fs_mark_inode_dirty_sync(inode, size_changed);
+
+	/* inode change will produce dirty node pages flushed by checkpoint */
+	f2fs_balance_fs(F2FS_I_SB(inode), true);
+
 	return err;
 }
 
@@ -727,7 +806,7 @@ static int fill_zero(struct inode *inode, pgoff_t index,
 	if (!len)
 		return 0;
 
-	f2fs_balance_fs(sbi);
+	f2fs_balance_fs(sbi, true);
 
 	f2fs_lock_op(sbi);
 	page = get_new_data_page(inode, NULL, index, false);
@@ -736,7 +815,7 @@ static int fill_zero(struct inode *inode, pgoff_t index,
 	if (IS_ERR(page))
 		return PTR_ERR(page);
 
-	f2fs_wait_on_page_writeback(page, DATA);
+	f2fs_wait_on_page_writeback(page, DATA, true);
 	zero_user(page, start, len);
 	set_page_dirty(page);
 	f2fs_put_page(page, 1);
@@ -761,7 +840,7 @@ int truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end)
 			return err;
 		}
 
-		end_offset = ADDRS_PER_PAGE(dn.node_page, F2FS_I(inode));
+		end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
 		count = min(end_offset - dn.ofs_in_node, pg_end - pg_start);
 
 		f2fs_bug_on(F2FS_I_SB(inode), count == 0 || count > end_offset);
@@ -778,19 +857,17 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
 {
 	pgoff_t pg_start, pg_end;
 	loff_t off_start, off_end;
-	int ret = 0;
+	int ret;
 
-	if (f2fs_has_inline_data(inode)) {
-		ret = f2fs_convert_inline_inode(inode);
-		if (ret)
-			return ret;
-	}
+	ret = f2fs_convert_inline_inode(inode);
+	if (ret)
+		return ret;
 
-	pg_start = ((unsigned long long) offset) >> PAGE_CACHE_SHIFT;
-	pg_end = ((unsigned long long) offset + len) >> PAGE_CACHE_SHIFT;
+	pg_start = ((unsigned long long) offset) >> PAGE_SHIFT;
+	pg_end = ((unsigned long long) offset + len) >> PAGE_SHIFT;
 
-	off_start = offset & (PAGE_CACHE_SIZE - 1);
-	off_end = (offset + len) & (PAGE_CACHE_SIZE - 1);
+	off_start = offset & (PAGE_SIZE - 1);
+	off_end = (offset + len) & (PAGE_SIZE - 1);
 
 	if (pg_start == pg_end) {
 		ret = fill_zero(inode, pg_start, off_start,
@@ -800,7 +877,7 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
 	} else {
 		if (off_start) {
 			ret = fill_zero(inode, pg_start++, off_start,
-						PAGE_CACHE_SIZE - off_start);
+						PAGE_SIZE - off_start);
 			if (ret)
 				return ret;
 		}
@@ -815,99 +892,218 @@ static int punch_hole(struct inode *inode, loff_t offset, loff_t len)
 			loff_t blk_start, blk_end;
 			struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 
-			f2fs_balance_fs(sbi);
+			f2fs_balance_fs(sbi, true);
 
-			blk_start = (loff_t)pg_start << PAGE_CACHE_SHIFT;
-			blk_end = (loff_t)pg_end << PAGE_CACHE_SHIFT;
+			blk_start = (loff_t)pg_start << PAGE_SHIFT;
+			blk_end = (loff_t)pg_end << PAGE_SHIFT;
+			down_write(&F2FS_I(inode)->i_mmap_sem);
 			truncate_inode_pages_range(mapping, blk_start,
 					blk_end - 1);
 
 			f2fs_lock_op(sbi);
 			ret = truncate_hole(inode, pg_start, pg_end);
 			f2fs_unlock_op(sbi);
+			up_write(&F2FS_I(inode)->i_mmap_sem);
 		}
 	}
 
 	return ret;
 }
 
-static int __exchange_data_block(struct inode *inode, pgoff_t src,
-					pgoff_t dst, bool full)
+static int __read_out_blkaddrs(struct inode *inode, block_t *blkaddr,
+				int *do_replace, pgoff_t off, pgoff_t len)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct dnode_of_data dn;
-	block_t new_addr;
-	bool do_replace = false;
-	int ret;
+	int ret, done, i;
 
+next_dnode:
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
-	ret = get_dnode_of_data(&dn, src, LOOKUP_NODE_RA);
+	ret = get_dnode_of_data(&dn, off, LOOKUP_NODE_RA);
 	if (ret && ret != -ENOENT) {
 		return ret;
 	} else if (ret == -ENOENT) {
-		new_addr = NULL_ADDR;
-	} else {
-		new_addr = dn.data_blkaddr;
-		if (!is_checkpointed_data(sbi, new_addr)) {
-			dn.data_blkaddr = NULL_ADDR;
+		if (dn.max_level == 0)
+			return -ENOENT;
+		done = min((pgoff_t)ADDRS_PER_BLOCK - dn.ofs_in_node, len);
+		blkaddr += done;
+		do_replace += done;
+		goto next;
+	}
+
+	done = min((pgoff_t)ADDRS_PER_PAGE(dn.node_page, inode) -
+							dn.ofs_in_node, len);
+	for (i = 0; i < done; i++, blkaddr++, do_replace++, dn.ofs_in_node++) {
+		*blkaddr = datablock_addr(dn.inode,
+					dn.node_page, dn.ofs_in_node);
+		if (!is_checkpointed_data(sbi, *blkaddr)) {
+
+			if (test_opt(sbi, LFS)) {
+				f2fs_put_dnode(&dn);
+				return -ENOTSUPP;
+			}
+
 			/* do not invalidate this block address */
-			set_data_blkaddr(&dn);
-			f2fs_update_extent_cache(&dn);
-			do_replace = true;
+			f2fs_update_data_blkaddr(&dn, NULL_ADDR);
+			*do_replace = 1;
 		}
-		f2fs_put_dnode(&dn);
 	}
+	f2fs_put_dnode(&dn);
+next:
+	len -= done;
+	off += done;
+	if (len)
+		goto next_dnode;
+	return 0;
+}
 
-	if (new_addr == NULL_ADDR)
-		return full ? truncate_hole(inode, dst, dst + 1) : 0;
+static int __roll_back_blkaddrs(struct inode *inode, block_t *blkaddr,
+				int *do_replace, pgoff_t off, int len)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct dnode_of_data dn;
+	int ret, i;
 
-	if (do_replace) {
-		struct page *ipage = get_node_page(sbi, inode->i_ino);
-		struct node_info ni;
+	for (i = 0; i < len; i++, do_replace++, blkaddr++) {
+		if (*do_replace == 0)
+			continue;
 
-		if (IS_ERR(ipage)) {
-			ret = PTR_ERR(ipage);
-			goto err_out;
+		set_new_dnode(&dn, inode, NULL, NULL, 0);
+		ret = get_dnode_of_data(&dn, off + i, LOOKUP_NODE_RA);
+		if (ret) {
+			dec_valid_block_count(sbi, inode, 1);
+			invalidate_blocks(sbi, *blkaddr);
+		} else {
+			f2fs_update_data_blkaddr(&dn, *blkaddr);
 		}
+		f2fs_put_dnode(&dn);
+	}
+	return 0;
+}
 
-		set_new_dnode(&dn, inode, ipage, NULL, 0);
-		ret = f2fs_reserve_block(&dn, dst);
-		if (ret)
-			goto err_out;
+static int __clone_blkaddrs(struct inode *src_inode, struct inode *dst_inode,
+			block_t *blkaddr, int *do_replace,
+			pgoff_t src, pgoff_t dst, pgoff_t len, bool full)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(src_inode);
+	pgoff_t i = 0;
+	int ret;
+
+	while (i < len) {
+		if (blkaddr[i] == NULL_ADDR && !full) {
+			i++;
+			continue;
+		}
 
-		truncate_data_blocks_range(&dn, 1);
+		if (do_replace[i] || blkaddr[i] == NULL_ADDR) {
+			struct dnode_of_data dn;
+			struct node_info ni;
+			size_t new_size;
+			pgoff_t ilen;
 
-		get_node_info(sbi, dn.nid, &ni);
-		f2fs_replace_block(sbi, &dn, dn.data_blkaddr, new_addr,
-				ni.version, true);
-		f2fs_put_dnode(&dn);
-	} else {
-		struct page *psrc, *pdst;
+			set_new_dnode(&dn, dst_inode, NULL, NULL, 0);
+			ret = get_dnode_of_data(&dn, dst + i, ALLOC_NODE);
+			if (ret)
+				return ret;
+
+			get_node_info(sbi, dn.nid, &ni);
+			ilen = min((pgoff_t)
+				ADDRS_PER_PAGE(dn.node_page, dst_inode) -
+						dn.ofs_in_node, len - i);
+			do {
+				dn.data_blkaddr = datablock_addr(dn.inode,
+						dn.node_page, dn.ofs_in_node);
+				truncate_data_blocks_range(&dn, 1);
+
+				if (do_replace[i]) {
+					f2fs_i_blocks_write(src_inode,
+							1, false, false);
+					f2fs_i_blocks_write(dst_inode,
+							1, true, false);
+					f2fs_replace_block(sbi, &dn, dn.data_blkaddr,
+					blkaddr[i], ni.version, true, false);
+
+					do_replace[i] = 0;
+				}
+				dn.ofs_in_node++;
+				i++;
+				new_size = (dst + i) << PAGE_SHIFT;
+				if (dst_inode->i_size < new_size)
+					f2fs_i_size_write(dst_inode, new_size);
+			} while (--ilen && (do_replace[i] || blkaddr[i] == NULL_ADDR));
 
-		psrc = get_lock_data_page(inode, src, true);
-		if (IS_ERR(psrc))
-			return PTR_ERR(psrc);
-		pdst = get_new_data_page(inode, NULL, dst, false);
-		if (IS_ERR(pdst)) {
+			f2fs_put_dnode(&dn);
+		} else {
+			struct page *psrc, *pdst;
+
+			psrc = get_lock_data_page(src_inode, src + i, true);
+			if (IS_ERR(psrc))
+				return PTR_ERR(psrc);
+			pdst = get_new_data_page(dst_inode, NULL, dst + i,
+								true);
+			if (IS_ERR(pdst)) {
+				f2fs_put_page(psrc, 1);
+				return PTR_ERR(pdst);
+			}
+			f2fs_copy_page(psrc, pdst);
+			set_page_dirty(pdst);
+			f2fs_put_page(pdst, 1);
 			f2fs_put_page(psrc, 1);
-			return PTR_ERR(pdst);
-		}
-		f2fs_copy_page(psrc, pdst);
-		set_page_dirty(pdst);
-		f2fs_put_page(pdst, 1);
-		f2fs_put_page(psrc, 1);
 
-		return truncate_hole(inode, src, src + 1);
+			ret = truncate_hole(src_inode, src + i, src + i + 1);
+			if (ret)
+				return ret;
+			i++;
+		}
 	}
 	return 0;
+}
 
-err_out:
-	if (!get_dnode_of_data(&dn, src, LOOKUP_NODE)) {
-		dn.data_blkaddr = new_addr;
-		set_data_blkaddr(&dn);
-		f2fs_update_extent_cache(&dn);
-		f2fs_put_dnode(&dn);
+static int __exchange_data_block(struct inode *src_inode,
+			struct inode *dst_inode, pgoff_t src, pgoff_t dst,
+			pgoff_t len, bool full)
+{
+	block_t *src_blkaddr;
+	int *do_replace;
+	pgoff_t olen;
+	int ret;
+
+	while (len) {
+		olen = min((pgoff_t)4 * ADDRS_PER_BLOCK, len);
+
+		src_blkaddr = kvzalloc(sizeof(block_t) * olen, GFP_KERNEL);
+		if (!src_blkaddr)
+			return -ENOMEM;
+
+		do_replace = kvzalloc(sizeof(int) * olen, GFP_KERNEL);
+		if (!do_replace) {
+			kvfree(src_blkaddr);
+			return -ENOMEM;
+		}
+
+		ret = __read_out_blkaddrs(src_inode, src_blkaddr,
+					do_replace, src, olen);
+		if (ret)
+			goto roll_back;
+
+		ret = __clone_blkaddrs(src_inode, dst_inode, src_blkaddr,
+					do_replace, src, dst, olen, full);
+		if (ret)
+			goto roll_back;
+
+		src += olen;
+		dst += olen;
+		len -= olen;
+
+		kvfree(src_blkaddr);
+		kvfree(do_replace);
 	}
+	return 0;
+
+roll_back:
+	__roll_back_blkaddrs(src_inode, src_blkaddr, do_replace, src, len);
+	kvfree(src_blkaddr);
+	kvfree(do_replace);
 	return ret;
 }
 
@@ -915,16 +1111,15 @@ static int f2fs_do_collapse(struct inode *inode, pgoff_t start, pgoff_t end)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	pgoff_t nrpages = (i_size_read(inode) + PAGE_SIZE - 1) / PAGE_SIZE;
-	int ret = 0;
+	int ret;
 
-	for (; end < nrpages; start++, end++) {
-		f2fs_balance_fs(sbi);
-		f2fs_lock_op(sbi);
-		ret = __exchange_data_block(inode, end, start, true);
-		f2fs_unlock_op(sbi);
-		if (ret)
-			break;
-	}
+	f2fs_balance_fs(sbi, true);
+	f2fs_lock_op(sbi);
+
+	f2fs_drop_extent_tree(inode);
+
+	ret = __exchange_data_block(inode, inode, end, start, nrpages - end, true);
+	f2fs_unlock_op(sbi);
 	return ret;
 }
 
@@ -941,27 +1136,24 @@ static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 	if (offset & (F2FS_BLKSIZE - 1) || len & (F2FS_BLKSIZE - 1))
 		return -EINVAL;
 
-	f2fs_balance_fs(F2FS_I_SB(inode));
-
-	if (f2fs_has_inline_data(inode)) {
-		ret = f2fs_convert_inline_inode(inode);
-		if (ret)
-			return ret;
-	}
+	ret = f2fs_convert_inline_inode(inode);
+	if (ret)
+		return ret;
 
-	pg_start = offset >> PAGE_CACHE_SHIFT;
-	pg_end = (offset + len) >> PAGE_CACHE_SHIFT;
+	pg_start = offset >> PAGE_SHIFT;
+	pg_end = (offset + len) >> PAGE_SHIFT;
 
+	down_write(&F2FS_I(inode)->i_mmap_sem);
 	/* write out all dirty pages from offset */
 	ret = filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
 	if (ret)
-		return ret;
+		goto out;
 
 	truncate_pagecache(inode, offset);
 
 	ret = f2fs_do_collapse(inode, pg_start, pg_end);
 	if (ret)
-		return ret;
+		goto out;
 
 	/* write out all moved pages, if possible */
 	filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
@@ -972,7 +1164,53 @@ static int f2fs_collapse_range(struct inode *inode, loff_t offset, loff_t len)
 
 	ret = truncate_blocks(inode, new_size, true);
 	if (!ret)
-		i_size_write(inode, new_size);
+		f2fs_i_size_write(inode, new_size);
+
+out:
+	up_write(&F2FS_I(inode)->i_mmap_sem);
+	return ret;
+}
+
+static int f2fs_do_zero_range(struct dnode_of_data *dn, pgoff_t start,
+								pgoff_t end)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
+	pgoff_t index = start;
+	unsigned int ofs_in_node = dn->ofs_in_node;
+	blkcnt_t count = 0;
+	int ret;
+
+	for (; index < end; index++, dn->ofs_in_node++) {
+		if (datablock_addr(dn->inode, dn->node_page,
+					dn->ofs_in_node) == NULL_ADDR)
+			count++;
+	}
+
+	dn->ofs_in_node = ofs_in_node;
+	ret = reserve_new_blocks(dn, count);
+	if (ret)
+		return ret;
+
+	dn->ofs_in_node = ofs_in_node;
+	for (index = start; index < end; index++, dn->ofs_in_node++) {
+		dn->data_blkaddr = datablock_addr(dn->inode,
+					dn->node_page, dn->ofs_in_node);
+		/*
+		 * reserve_new_blocks will not guarantee entire block
+		 * allocation.
+		 */
+		if (dn->data_blkaddr == NULL_ADDR) {
+			ret = -ENOSPC;
+			break;
+		}
+		if (dn->data_blkaddr != NEW_ADDR) {
+			invalidate_blocks(sbi, dn->data_blkaddr);
+			dn->data_blkaddr = NEW_ADDR;
+			set_data_blkaddr(dn);
+		}
+	}
+
+	f2fs_update_extent_cache_range(dn, start, 0, index - start);
 
 	return ret;
 }
@@ -991,80 +1229,70 @@ static int f2fs_zero_range(struct inode *inode, loff_t offset, loff_t len,
 	if (ret)
 		return ret;
 
-	f2fs_balance_fs(sbi);
-
-	if (f2fs_has_inline_data(inode)) {
-		ret = f2fs_convert_inline_inode(inode);
-		if (ret)
-			return ret;
-	}
+	ret = f2fs_convert_inline_inode(inode);
+	if (ret)
+		return ret;
 
+	down_write(&F2FS_I(inode)->i_mmap_sem);
 	ret = filemap_write_and_wait_range(mapping, offset, offset + len - 1);
 	if (ret)
-		return ret;
+		goto out_sem;
 
 	truncate_pagecache_range(inode, offset, offset + len - 1);
 
-	pg_start = ((unsigned long long) offset) >> PAGE_CACHE_SHIFT;
-	pg_end = ((unsigned long long) offset + len) >> PAGE_CACHE_SHIFT;
+	pg_start = ((unsigned long long) offset) >> PAGE_SHIFT;
+	pg_end = ((unsigned long long) offset + len) >> PAGE_SHIFT;
 
-	off_start = offset & (PAGE_CACHE_SIZE - 1);
-	off_end = (offset + len) & (PAGE_CACHE_SIZE - 1);
+	off_start = offset & (PAGE_SIZE - 1);
+	off_end = (offset + len) & (PAGE_SIZE - 1);
 
 	if (pg_start == pg_end) {
 		ret = fill_zero(inode, pg_start, off_start,
 						off_end - off_start);
 		if (ret)
-			return ret;
+			goto out_sem;
 
-		if (offset + len > new_size)
-			new_size = offset + len;
 		new_size = max_t(loff_t, new_size, offset + len);
 	} else {
 		if (off_start) {
 			ret = fill_zero(inode, pg_start++, off_start,
-						PAGE_CACHE_SIZE - off_start);
+						PAGE_SIZE - off_start);
 			if (ret)
-				return ret;
+				goto out_sem;
 
 			new_size = max_t(loff_t, new_size,
-					(loff_t)pg_start << PAGE_CACHE_SHIFT);
+					(loff_t)pg_start << PAGE_SHIFT);
 		}
 
-		for (index = pg_start; index < pg_end; index++) {
+		for (index = pg_start; index < pg_end;) {
 			struct dnode_of_data dn;
-			struct page *ipage;
+			unsigned int end_offset;
+			pgoff_t end;
 
 			f2fs_lock_op(sbi);
 
-			ipage = get_node_page(sbi, inode->i_ino);
-			if (IS_ERR(ipage)) {
-				ret = PTR_ERR(ipage);
-				f2fs_unlock_op(sbi);
-				goto out;
-			}
-
-			set_new_dnode(&dn, inode, ipage, NULL, 0);
-			ret = f2fs_reserve_block(&dn, index);
+			set_new_dnode(&dn, inode, NULL, NULL, 0);
+			ret = get_dnode_of_data(&dn, index, ALLOC_NODE);
 			if (ret) {
 				f2fs_unlock_op(sbi);
 				goto out;
 			}
 
-			if (dn.data_blkaddr != NEW_ADDR) {
-				invalidate_blocks(sbi, dn.data_blkaddr);
+			end_offset = ADDRS_PER_PAGE(dn.node_page, inode);
+			end = min(pg_end, end_offset - dn.ofs_in_node + index);
 
-				dn.data_blkaddr = NEW_ADDR;
-				set_data_blkaddr(&dn);
-
-				dn.data_blkaddr = NULL_ADDR;
-				f2fs_update_extent_cache(&dn);
-			}
+			ret = f2fs_do_zero_range(&dn, index, end);
 			f2fs_put_dnode(&dn);
 			f2fs_unlock_op(sbi);
 
+			f2fs_balance_fs(sbi, dn.node_changed);
+
+			if (ret)
+				goto out;
+
+			index = end;
 			new_size = max_t(loff_t, new_size,
-				(loff_t)(index + 1) << PAGE_CACHE_SHIFT);
+					(loff_t)index << PAGE_SHIFT);
 		}
 
 		if (off_end) {
@@ -1077,11 +1305,10 @@ static int f2fs_zero_range(struct inode *inode, loff_t offset, loff_t len,
 	}
 
 out:
-	if (!(mode & FALLOC_FL_KEEP_SIZE) && i_size_read(inode) < new_size) {
-		i_size_write(inode, new_size);
-		mark_inode_dirty(inode);
-		update_inode_page(inode);
-	}
+	if (!(mode & FALLOC_FL_KEEP_SIZE) && i_size_read(inode) < new_size)
+		f2fs_i_size_write(inode, new_size);
+out_sem:
+	up_write(&F2FS_I(inode)->i_mmap_sem);
 
 	return ret;
 }
@@ -1089,13 +1316,14 @@ out:
 static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	pgoff_t pg_start, pg_end, delta, nrpages, idx;
+	pgoff_t nr, pg_start, pg_end, delta, idx;
 	loff_t new_size;
 	int ret = 0;
 
 	new_size = i_size_read(inode) + len;
-	if (new_size > inode->i_sb->s_maxbytes)
-		return -EFBIG;
+	ret = inode_newsize_ok(inode, new_size);
+	if (ret)
+		return ret;
 
 	if (offset >= i_size_read(inode))
 		return -EINVAL;
@@ -1104,36 +1332,41 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len)
 	if (offset & (F2FS_BLKSIZE - 1) || len & (F2FS_BLKSIZE - 1))
 		return -EINVAL;
 
-	f2fs_balance_fs(sbi);
+	ret = f2fs_convert_inline_inode(inode);
+	if (ret)
+		return ret;
 
-	if (f2fs_has_inline_data(inode)) {
-		ret = f2fs_convert_inline_inode(inode);
-		if (ret)
-			return ret;
-	}
+	f2fs_balance_fs(sbi, true);
 
+	down_write(&F2FS_I(inode)->i_mmap_sem);
 	ret = truncate_blocks(inode, i_size_read(inode), true);
 	if (ret)
-		return ret;
+		goto out;
 
 	/* write out all dirty pages from offset */
 	ret = filemap_write_and_wait_range(inode->i_mapping, offset, LLONG_MAX);
 	if (ret)
-		return ret;
+		goto out;
 
 	truncate_pagecache(inode, offset);
 
-	pg_start = offset >> PAGE_CACHE_SHIFT;
-	pg_end = (offset + len) >> PAGE_CACHE_SHIFT;
+	pg_start = offset >> PAGE_SHIFT;
+	pg_end = (offset + len) >> PAGE_SHIFT;
 	delta = pg_end - pg_start;
-	nrpages = (i_size_read(inode) + PAGE_SIZE - 1) / PAGE_SIZE;
+	idx = (i_size_read(inode) + PAGE_SIZE - 1) / PAGE_SIZE;
+
+	while (!ret && idx > pg_start) {
+		nr = idx - pg_start;
+		if (nr > delta)
+			nr = delta;
+		idx -= nr;
 
-	for (idx = nrpages - 1; idx >= pg_start && idx != -1; idx--) {
 		f2fs_lock_op(sbi);
-		ret = __exchange_data_block(inode, idx, idx + delta, false);
+		f2fs_drop_extent_tree(inode);
+
+		ret = __exchange_data_block(inode, inode, idx,
+					idx + delta, nr, false);
 		f2fs_unlock_op(sbi);
-		if (ret)
-			break;
 	}
 
 	/* write out all moved pages, if possible */
@@ -1141,7 +1374,9 @@ static int f2fs_insert_range(struct inode *inode, loff_t offset, loff_t len)
 	truncate_pagecache(inode, offset);
 
 	if (!ret)
-		i_size_write(inode, new_size);
+		f2fs_i_size_write(inode, new_size);
+out:
+	up_write(&F2FS_I(inode)->i_mmap_sem);
 	return ret;
 }
 
@@ -1149,62 +1384,50 @@ static int expand_inode_data(struct inode *inode, loff_t offset,
 					loff_t len, int mode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	pgoff_t index, pg_start, pg_end;
+	struct f2fs_map_blocks map = { .m_next_pgofs = NULL };
+	pgoff_t pg_end;
 	loff_t new_size = i_size_read(inode);
-	loff_t off_start, off_end;
-	int ret = 0;
+	loff_t off_end;
+	int err;
 
-	f2fs_balance_fs(sbi);
+	err = inode_newsize_ok(inode, (len + offset));
+	if (err)
+		return err;
 
-	ret = inode_newsize_ok(inode, (len + offset));
-	if (ret)
-		return ret;
+	err = f2fs_convert_inline_inode(inode);
+	if (err)
+		return err;
 
-	if (f2fs_has_inline_data(inode)) {
-		ret = f2fs_convert_inline_inode(inode);
-		if (ret)
-			return ret;
-	}
+	f2fs_balance_fs(sbi, true);
 
-	pg_start = ((unsigned long long) offset) >> PAGE_CACHE_SHIFT;
-	pg_end = ((unsigned long long) offset + len) >> PAGE_CACHE_SHIFT;
+	pg_end = ((unsigned long long)offset + len) >> PAGE_SHIFT;
+	off_end = (offset + len) & (PAGE_SIZE - 1);
 
-	off_start = offset & (PAGE_CACHE_SIZE - 1);
-	off_end = (offset + len) & (PAGE_CACHE_SIZE - 1);
+	map.m_lblk = ((unsigned long long)offset) >> PAGE_SHIFT;
+	map.m_len = pg_end - map.m_lblk;
+	if (off_end)
+		map.m_len++;
 
-	f2fs_lock_op(sbi);
+	err = f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_AIO);
+	if (err) {
+		pgoff_t last_off;
 
-	for (index = pg_start; index <= pg_end; index++) {
-		struct dnode_of_data dn;
+		if (!map.m_len)
+			return err;
 
-		if (index == pg_end && !off_end)
-			goto noalloc;
+		last_off = map.m_lblk + map.m_len - 1;
 
-		set_new_dnode(&dn, inode, NULL, NULL, 0);
-		ret = f2fs_reserve_block(&dn, index);
-		if (ret)
-			break;
-noalloc:
-		if (pg_start == pg_end)
-			new_size = offset + len;
-		else if (index == pg_start && off_start)
-			new_size = (loff_t)(index + 1) << PAGE_CACHE_SHIFT;
-		else if (index == pg_end)
-			new_size = ((loff_t)index << PAGE_CACHE_SHIFT) +
-								off_end;
-		else
-			new_size += PAGE_CACHE_SIZE;
-	}
-
-	if (!(mode & FALLOC_FL_KEEP_SIZE) &&
-		i_size_read(inode) < new_size) {
-		i_size_write(inode, new_size);
-		mark_inode_dirty(inode);
-		update_inode_page(inode);
+		/* update new size to the failed position */
+		new_size = (last_off == pg_end) ? offset + len:
+					(loff_t)(last_off + 1) << PAGE_SHIFT;
+	} else {
+		new_size = ((loff_t)pg_end << PAGE_SHIFT) + off_end;
 	}
-	f2fs_unlock_op(sbi);
 
-	return ret;
+	if (!(mode & FALLOC_FL_KEEP_SIZE) && i_size_read(inode) < new_size)
+		f2fs_i_size_write(inode, new_size);
+
+	return err;
 }
 
 static long f2fs_fallocate(struct file *file, int mode,
@@ -1226,7 +1449,7 @@ static long f2fs_fallocate(struct file *file, int mode,
 			FALLOC_FL_INSERT_RANGE))
 		return -EOPNOTSUPP;
 
-	mutex_lock(&inode->i_mutex);
+	inode_lock(inode);
 
 	if (mode & FALLOC_FL_PUNCH_HOLE) {
 		if (offset >= inode->i_size)
@@ -1244,12 +1467,15 @@ static long f2fs_fallocate(struct file *file, int mode,
 	}
 
 	if (!ret) {
-		inode->i_mtime = inode->i_ctime = CURRENT_TIME;
-		mark_inode_dirty(inode);
+		inode->i_mtime = inode->i_ctime = current_time(inode);
+		f2fs_mark_inode_dirty_sync(inode, false);
+		if (mode & FALLOC_FL_KEEP_SIZE)
+			file_set_keep_isize(inode);
+		f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
 	}
 
 out:
-	mutex_unlock(&inode->i_mutex);
+	inode_unlock(inode);
 
 	trace_f2fs_fallocate(inode, mode, offset, len, ret);
 	return ret;
@@ -1257,28 +1483,41 @@ out:
 
 static int f2fs_release_file(struct inode *inode, struct file *filp)
 {
+	/*
+	 * f2fs_relase_file is called at every close calls. So we should
+	 * not drop any inmemory pages by close called by other process.
+	 */
+	if (!(filp->f_mode & FMODE_WRITE) ||
+			atomic_read(&inode->i_writecount) != 1)
+		return 0;
+
 	/* some remained atomic pages should discarded */
 	if (f2fs_is_atomic_file(inode))
-		commit_inmem_pages(inode, true);
+		drop_inmem_pages(inode);
 	if (f2fs_is_volatile_file(inode)) {
-		set_inode_flag(F2FS_I(inode), FI_DROP_CACHE);
+		clear_inode_flag(inode, FI_VOLATILE_FILE);
+		stat_dec_volatile_write(inode);
+		set_inode_flag(inode, FI_DROP_CACHE);
 		filemap_fdatawrite(inode->i_mapping);
-		clear_inode_flag(F2FS_I(inode), FI_DROP_CACHE);
+		clear_inode_flag(inode, FI_DROP_CACHE);
 	}
 	return 0;
 }
 
-#define F2FS_REG_FLMASK		(~(FS_DIRSYNC_FL | FS_TOPDIR_FL))
-#define F2FS_OTHER_FLMASK	(FS_NODUMP_FL | FS_NOATIME_FL)
-
-static inline __u32 f2fs_mask_flags(umode_t mode, __u32 flags)
+static int f2fs_file_flush(struct file *file, fl_owner_t id)
 {
-	if (S_ISDIR(mode))
-		return flags;
-	else if (S_ISREG(mode))
-		return flags & F2FS_REG_FLMASK;
-	else
-		return flags & F2FS_OTHER_FLMASK;
+	struct inode *inode = file_inode(file);
+
+	/*
+	 * If the process doing a transaction is crashed, we should do
+	 * roll-back. Otherwise, other reader/write can see corrupted database
+	 * until all the writers close its file. Since this should be done
+	 * before dropping file lock, it needs to do in ->flush.
+	 */
+	if (f2fs_is_atomic_file(inode) &&
+			F2FS_I(inode)->inmem_task == current)
+		drop_inmem_pages(inode);
+	return 0;
 }
 
 static int f2fs_ioc_getflags(struct file *filp, unsigned long arg)
@@ -1293,47 +1532,48 @@ static int f2fs_ioc_setflags(struct file *filp, unsigned long arg)
 {
 	struct inode *inode = file_inode(filp);
 	struct f2fs_inode_info *fi = F2FS_I(inode);
-	unsigned int flags = fi->i_flags & FS_FL_USER_VISIBLE;
+	unsigned int flags;
 	unsigned int oldflags;
 	int ret;
 
+	if (!inode_owner_or_capable(inode))
+		return -EACCES;
+
+	if (get_user(flags, (int __user *)arg))
+		return -EFAULT;
+
 	ret = mnt_want_write_file(filp);
 	if (ret)
 		return ret;
 
-	if (!inode_owner_or_capable(inode)) {
-		ret = -EACCES;
-		goto out;
-	}
+	inode_lock(inode);
 
-	if (get_user(flags, (int __user *)arg)) {
-		ret = -EFAULT;
-		goto out;
+	/* Is it quota file? Do not allow user to mess with it */
+	if (IS_NOQUOTA(inode)) {
+		ret = -EPERM;
+		goto unlock_out;
 	}
 
 	flags = f2fs_mask_flags(inode->i_mode, flags);
 
-	mutex_lock(&inode->i_mutex);
-
 	oldflags = fi->i_flags;
 
 	if ((flags ^ oldflags) & (FS_APPEND_FL | FS_IMMUTABLE_FL)) {
 		if (!capable(CAP_LINUX_IMMUTABLE)) {
-			mutex_unlock(&inode->i_mutex);
 			ret = -EPERM;
-			goto out;
+			goto unlock_out;
 		}
 	}
 
 	flags = flags & FS_FL_USER_MODIFIABLE;
 	flags |= oldflags & ~FS_FL_USER_MODIFIABLE;
 	fi->i_flags = flags;
-	mutex_unlock(&inode->i_mutex);
 
+	inode->i_ctime = current_time(inode);
 	f2fs_set_inode_flags(inode);
-	inode->i_ctime = CURRENT_TIME;
-	mark_inode_dirty(inode);
-out:
+	f2fs_mark_inode_dirty_sync(inode, false);
+unlock_out:
+	inode_unlock(inode);
 	mnt_drop_write_file(filp);
 	return ret;
 }
@@ -1353,17 +1593,47 @@ static int f2fs_ioc_start_atomic_write(struct file *filp)
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
-	f2fs_balance_fs(F2FS_I_SB(inode));
+	if (!S_ISREG(inode->i_mode))
+		return -EINVAL;
+
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	inode_lock(inode);
 
 	if (f2fs_is_atomic_file(inode))
-		return 0;
+		goto out;
 
 	ret = f2fs_convert_inline_inode(inode);
 	if (ret)
-		return ret;
+		goto out;
 
-	set_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
-	return 0;
+	set_inode_flag(inode, FI_ATOMIC_FILE);
+	set_inode_flag(inode, FI_HOT_DATA);
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
+
+	if (!get_dirty_pages(inode))
+		goto inc_stat;
+
+	f2fs_msg(F2FS_I_SB(inode)->sb, KERN_WARNING,
+		"Unexpected flush for atomic writes: ino=%lu, npages=%u",
+					inode->i_ino, get_dirty_pages(inode));
+	ret = filemap_write_and_wait_range(inode->i_mapping, 0, LLONG_MAX);
+	if (ret) {
+		clear_inode_flag(inode, FI_ATOMIC_FILE);
+		clear_inode_flag(inode, FI_HOT_DATA);
+		goto out;
+	}
+
+inc_stat:
+	F2FS_I(inode)->inmem_task = current;
+	stat_inc_atomic_write(inode);
+	stat_update_max_atomic_write(inode);
+out:
+	inode_unlock(inode);
+	mnt_drop_write_file(filp);
+	return ret;
 }
 
 static int f2fs_ioc_commit_atomic_write(struct file *filp)
@@ -1374,22 +1644,31 @@ static int f2fs_ioc_commit_atomic_write(struct file *filp)
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
-	if (f2fs_is_volatile_file(inode))
-		return 0;
-
 	ret = mnt_want_write_file(filp);
 	if (ret)
 		return ret;
 
+	inode_lock(inode);
+
+	if (f2fs_is_volatile_file(inode))
+		goto err_out;
+
 	if (f2fs_is_atomic_file(inode)) {
-		clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
-		ret = commit_inmem_pages(inode, false);
+		ret = commit_inmem_pages(inode);
 		if (ret)
 			goto err_out;
-	}
 
-	ret = f2fs_sync_file(filp, 0, LLONG_MAX, 0);
+		ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 0, true);
+		if (!ret) {
+			clear_inode_flag(inode, FI_ATOMIC_FILE);
+			clear_inode_flag(inode, FI_HOT_DATA);
+			stat_dec_atomic_write(inode);
+		}
+	} else {
+		ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 1, false);
+	}
 err_out:
+	inode_unlock(inode);
 	mnt_drop_write_file(filp);
 	return ret;
 }
@@ -1402,31 +1681,60 @@ static int f2fs_ioc_start_volatile_write(struct file *filp)
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
+	if (!S_ISREG(inode->i_mode))
+		return -EINVAL;
+
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	inode_lock(inode);
+
 	if (f2fs_is_volatile_file(inode))
-		return 0;
+		goto out;
 
 	ret = f2fs_convert_inline_inode(inode);
 	if (ret)
-		return ret;
+		goto out;
 
-	set_inode_flag(F2FS_I(inode), FI_VOLATILE_FILE);
-	return 0;
+	stat_inc_volatile_write(inode);
+	stat_update_max_volatile_write(inode);
+
+	set_inode_flag(inode, FI_VOLATILE_FILE);
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
+out:
+	inode_unlock(inode);
+	mnt_drop_write_file(filp);
+	return ret;
 }
 
 static int f2fs_ioc_release_volatile_write(struct file *filp)
 {
 	struct inode *inode = file_inode(filp);
+	int ret;
 
 	if (!inode_owner_or_capable(inode))
 		return -EACCES;
 
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	inode_lock(inode);
+
 	if (!f2fs_is_volatile_file(inode))
-		return 0;
+		goto out;
 
-	if (!f2fs_is_first_block_written(inode))
-		return truncate_partial_data_page(inode, 0, true);
+	if (!f2fs_is_first_block_written(inode)) {
+		ret = truncate_partial_data_page(inode, 0, true);
+		goto out;
+	}
 
-	return punch_hole(inode, 0, F2FS_BLKSIZE);
+	ret = punch_hole(inode, 0, F2FS_BLKSIZE);
+out:
+	inode_unlock(inode);
+	mnt_drop_write_file(filp);
+	return ret;
 }
 
 static int f2fs_ioc_abort_volatile_write(struct file *filp)
@@ -1441,13 +1749,20 @@ static int f2fs_ioc_abort_volatile_write(struct file *filp)
 	if (ret)
 		return ret;
 
-	f2fs_balance_fs(F2FS_I_SB(inode));
+	inode_lock(inode);
 
-	clear_inode_flag(F2FS_I(inode), FI_ATOMIC_FILE);
-	clear_inode_flag(F2FS_I(inode), FI_VOLATILE_FILE);
-	commit_inmem_pages(inode, true);
+	if (f2fs_is_atomic_file(inode))
+		drop_inmem_pages(inode);
+	if (f2fs_is_volatile_file(inode)) {
+		clear_inode_flag(inode, FI_VOLATILE_FILE);
+		stat_dec_volatile_write(inode);
+		ret = f2fs_do_sync_file(filp, 0, LLONG_MAX, 0, true);
+	}
+
+	inode_unlock(inode);
 
 	mnt_drop_write_file(filp);
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
 	return ret;
 }
 
@@ -1457,6 +1772,7 @@ static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg)
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct super_block *sb = sbi->sb;
 	__u32 in;
+	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -1464,30 +1780,38 @@ static int f2fs_ioc_shutdown(struct file *filp, unsigned long arg)
 	if (get_user(in, (__u32 __user *)arg))
 		return -EFAULT;
 
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
 	switch (in) {
 	case F2FS_GOING_DOWN_FULLSYNC:
 		sb = freeze_bdev(sb->s_bdev);
 		if (sb && !IS_ERR(sb)) {
-			f2fs_stop_checkpoint(sbi);
+			f2fs_stop_checkpoint(sbi, false);
 			thaw_bdev(sb->s_bdev, sb);
 		}
 		break;
 	case F2FS_GOING_DOWN_METASYNC:
 		/* do checkpoint only */
 		f2fs_sync_fs(sb, 1);
-		f2fs_stop_checkpoint(sbi);
+		f2fs_stop_checkpoint(sbi, false);
 		break;
 	case F2FS_GOING_DOWN_NOSYNC:
-		f2fs_stop_checkpoint(sbi);
+		f2fs_stop_checkpoint(sbi, false);
 		break;
 	case F2FS_GOING_DOWN_METAFLUSH:
-		sync_meta_pages(sbi, META, LONG_MAX);
-		f2fs_stop_checkpoint(sbi);
+		sync_meta_pages(sbi, META, LONG_MAX, FS_META_IO);
+		f2fs_stop_checkpoint(sbi, false);
 		break;
 	default:
-		return -EINVAL;
+		ret = -EINVAL;
+		goto out;
 	}
-	return 0;
+	f2fs_update_time(sbi, REQ_TIME);
+out:
+	mnt_drop_write_file(filp);
+	return ret;
 }
 
 static int f2fs_ioc_fitrim(struct file *filp, unsigned long arg)
@@ -1508,15 +1832,21 @@ static int f2fs_ioc_fitrim(struct file *filp, unsigned long arg)
 				sizeof(range)))
 		return -EFAULT;
 
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
 	range.minlen = max((unsigned int)range.minlen,
 				q->limits.discard_granularity);
 	ret = f2fs_trim_fs(F2FS_SB(sb), &range);
+	mnt_drop_write_file(filp);
 	if (ret < 0)
 		return ret;
 
 	if (copy_to_user((struct fstrim_range __user *)arg, &range,
 				sizeof(range)))
 		return -EFAULT;
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
 	return 0;
 }
 
@@ -1532,51 +1862,16 @@ static bool uuid_is_nonzero(__u8 u[16])
 
 static int f2fs_ioc_set_encryption_policy(struct file *filp, unsigned long arg)
 {
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	struct f2fs_encryption_policy policy;
 	struct inode *inode = file_inode(filp);
-	int err;
-
-	if (copy_from_user(&policy, (struct f2fs_encryption_policy __user *)arg,
-				sizeof(policy)))
-		return -EFAULT;
-
-	err = mnt_want_write_file(filp);
-	if (err)
-		return err;
-
-	mutex_lock(&inode->i_mutex);
 
-	err = f2fs_process_policy(&policy, inode);
-
-	mutex_unlock(&inode->i_mutex);
-
-	mnt_drop_write_file(filp);
+	f2fs_update_time(F2FS_I_SB(inode), REQ_TIME);
 
-	return err;
-#else
-	return -EOPNOTSUPP;
-#endif
+	return fscrypt_ioctl_set_policy(filp, (const void __user *)arg);
 }
 
 static int f2fs_ioc_get_encryption_policy(struct file *filp, unsigned long arg)
 {
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	struct f2fs_encryption_policy policy;
-	struct inode *inode = file_inode(filp);
-	int err;
-
-	err = f2fs_get_policy(inode, &policy);
-	if (err)
-		return err;
-
-	if (copy_to_user((struct f2fs_encryption_policy __user *)arg, &policy,
-							sizeof(policy)))
-		return -EFAULT;
-	return 0;
-#else
-	return -EOPNOTSUPP;
-#endif
+	return fscrypt_ioctl_get_policy(filp, (void __user *)arg);
 }
 
 static int f2fs_ioc_get_encryption_pwsalt(struct file *filp, unsigned long arg)
@@ -1599,13 +1894,13 @@ static int f2fs_ioc_get_encryption_pwsalt(struct file *filp, unsigned long arg)
 	generate_random_uuid(sbi->raw_super->encrypt_pw_salt);
 
 	err = f2fs_commit_super(sbi, false);
-
-	mnt_drop_write_file(filp);
 	if (err) {
 		/* undo new data */
 		memset(sbi->raw_super->encrypt_pw_salt, 0, 16);
+		mnt_drop_write_file(filp);
 		return err;
 	}
+	mnt_drop_write_file(filp);
 got_it:
 	if (copy_to_user((__u8 __user *)arg, sbi->raw_super->encrypt_pw_salt,
 									16))
@@ -1618,6 +1913,7 @@ static int f2fs_ioc_gc(struct file *filp, unsigned long arg)
 	struct inode *inode = file_inode(filp);
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	__u32 sync;
+	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -1628,21 +1924,74 @@ static int f2fs_ioc_gc(struct file *filp, unsigned long arg)
 	if (f2fs_readonly(sbi->sb))
 		return -EROFS;
 
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
 	if (!sync) {
-		if (!mutex_trylock(&sbi->gc_mutex))
-			return -EBUSY;
+		if (!mutex_trylock(&sbi->gc_mutex)) {
+			ret = -EBUSY;
+			goto out;
+		}
+	} else {
+		mutex_lock(&sbi->gc_mutex);
+	}
+
+	ret = f2fs_gc(sbi, sync, true, NULL_SEGNO);
+out:
+	mnt_drop_write_file(filp);
+	return ret;
+}
+
+static int f2fs_ioc_gc_range(struct file *filp, unsigned long arg)
+{
+	struct inode *inode = file_inode(filp);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct f2fs_gc_range range;
+	u64 end;
+	int ret;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (copy_from_user(&range, (struct f2fs_gc_range __user *)arg,
+							sizeof(range)))
+		return -EFAULT;
+
+	if (f2fs_readonly(sbi->sb))
+		return -EROFS;
+
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	end = range.start + range.len;
+	if (range.start < MAIN_BLKADDR(sbi) || end >= MAX_BLKADDR(sbi))
+		return -EINVAL;
+do_more:
+	if (!range.sync) {
+		if (!mutex_trylock(&sbi->gc_mutex)) {
+			ret = -EBUSY;
+			goto out;
+		}
 	} else {
 		mutex_lock(&sbi->gc_mutex);
 	}
 
-	return f2fs_gc(sbi, sync);
+	ret = f2fs_gc(sbi, range.sync, true, GET_SEGNO(sbi, range.start));
+	range.start += sbi->blocks_per_seg;
+	if (range.start <= end)
+		goto do_more;
+out:
+	mnt_drop_write_file(filp);
+	return ret;
 }
 
 static int f2fs_ioc_write_checkpoint(struct file *filp, unsigned long arg)
 {
 	struct inode *inode = file_inode(filp);
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct cp_control cpc;
+	int ret;
 
 	if (!capable(CAP_SYS_ADMIN))
 		return -EPERM;
@@ -1650,15 +1999,417 @@ static int f2fs_ioc_write_checkpoint(struct file *filp, unsigned long arg)
 	if (f2fs_readonly(sbi->sb))
 		return -EROFS;
 
-	cpc.reason = __get_cp_reason(sbi);
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	ret = f2fs_sync_fs(sbi->sb, 1);
+
+	mnt_drop_write_file(filp);
+	return ret;
+}
+
+static int f2fs_defragment_range(struct f2fs_sb_info *sbi,
+					struct file *filp,
+					struct f2fs_defragment *range)
+{
+	struct inode *inode = file_inode(filp);
+	struct f2fs_map_blocks map = { .m_next_pgofs = NULL };
+	struct extent_info ei = {0,0,0};
+	pgoff_t pg_start, pg_end;
+	unsigned int blk_per_seg = sbi->blocks_per_seg;
+	unsigned int total = 0, sec_num;
+	block_t blk_end = 0;
+	bool fragmented = false;
+	int err;
+
+	/* if in-place-update policy is enabled, don't waste time here */
+	if (need_inplace_update_policy(inode, NULL))
+		return -EINVAL;
+
+	pg_start = range->start >> PAGE_SHIFT;
+	pg_end = (range->start + range->len) >> PAGE_SHIFT;
+
+	f2fs_balance_fs(sbi, true);
+
+	inode_lock(inode);
+
+	/* writeback all dirty pages in the range */
+	err = filemap_write_and_wait_range(inode->i_mapping, range->start,
+						range->start + range->len - 1);
+	if (err)
+		goto out;
+
+	/*
+	 * lookup mapping info in extent cache, skip defragmenting if physical
+	 * block addresses are continuous.
+	 */
+	if (f2fs_lookup_extent_cache(inode, pg_start, &ei)) {
+		if (ei.fofs + ei.len >= pg_end)
+			goto out;
+	}
 
-	mutex_lock(&sbi->gc_mutex);
-	write_checkpoint(sbi, &cpc);
-	mutex_unlock(&sbi->gc_mutex);
+	map.m_lblk = pg_start;
+
+	/*
+	 * lookup mapping info in dnode page cache, skip defragmenting if all
+	 * physical block addresses are continuous even if there are hole(s)
+	 * in logical blocks.
+	 */
+	while (map.m_lblk < pg_end) {
+		map.m_len = pg_end - map.m_lblk;
+		err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_DEFAULT);
+		if (err)
+			goto out;
+
+		if (!(map.m_flags & F2FS_MAP_FLAGS)) {
+			map.m_lblk++;
+			continue;
+		}
+
+		if (blk_end && blk_end != map.m_pblk) {
+			fragmented = true;
+			break;
+		}
+		blk_end = map.m_pblk + map.m_len;
+
+		map.m_lblk += map.m_len;
+	}
+
+	if (!fragmented)
+		goto out;
+
+	map.m_lblk = pg_start;
+	map.m_len = pg_end - pg_start;
+
+	sec_num = (map.m_len + BLKS_PER_SEC(sbi) - 1) / BLKS_PER_SEC(sbi);
+
+	/*
+	 * make sure there are enough free section for LFS allocation, this can
+	 * avoid defragment running in SSR mode when free section are allocated
+	 * intensively
+	 */
+	if (has_not_enough_free_secs(sbi, 0, sec_num)) {
+		err = -EAGAIN;
+		goto out;
+	}
+
+	while (map.m_lblk < pg_end) {
+		pgoff_t idx;
+		int cnt = 0;
+
+do_map:
+		map.m_len = pg_end - map.m_lblk;
+		err = f2fs_map_blocks(inode, &map, 0, F2FS_GET_BLOCK_DEFAULT);
+		if (err)
+			goto clear_out;
+
+		if (!(map.m_flags & F2FS_MAP_FLAGS)) {
+			map.m_lblk++;
+			continue;
+		}
+
+		set_inode_flag(inode, FI_DO_DEFRAG);
+
+		idx = map.m_lblk;
+		while (idx < map.m_lblk + map.m_len && cnt < blk_per_seg) {
+			struct page *page;
+
+			page = get_lock_data_page(inode, idx, true);
+			if (IS_ERR(page)) {
+				err = PTR_ERR(page);
+				goto clear_out;
+			}
+
+			set_page_dirty(page);
+			f2fs_put_page(page, 1);
+
+			idx++;
+			cnt++;
+			total++;
+		}
+
+		map.m_lblk = idx;
+
+		if (idx < pg_end && cnt < blk_per_seg)
+			goto do_map;
+
+		clear_inode_flag(inode, FI_DO_DEFRAG);
+
+		err = filemap_fdatawrite(inode->i_mapping);
+		if (err)
+			goto out;
+	}
+clear_out:
+	clear_inode_flag(inode, FI_DO_DEFRAG);
+out:
+	inode_unlock(inode);
+	if (!err)
+		range->len = (u64)total << PAGE_SHIFT;
+	return err;
+}
+
+static int f2fs_ioc_defragment(struct file *filp, unsigned long arg)
+{
+	struct inode *inode = file_inode(filp);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct f2fs_defragment range;
+	int err;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (!S_ISREG(inode->i_mode) || f2fs_is_atomic_file(inode))
+		return -EINVAL;
+
+	if (f2fs_readonly(sbi->sb))
+		return -EROFS;
+
+	if (copy_from_user(&range, (struct f2fs_defragment __user *)arg,
+							sizeof(range)))
+		return -EFAULT;
+
+	/* verify alignment of offset & size */
+	if (range.start & (F2FS_BLKSIZE - 1) || range.len & (F2FS_BLKSIZE - 1))
+		return -EINVAL;
+
+	if (unlikely((range.start + range.len) >> PAGE_SHIFT >
+					sbi->max_file_blocks))
+		return -EINVAL;
+
+	err = mnt_want_write_file(filp);
+	if (err)
+		return err;
+
+	err = f2fs_defragment_range(sbi, filp, &range);
+	mnt_drop_write_file(filp);
+
+	f2fs_update_time(sbi, REQ_TIME);
+	if (err < 0)
+		return err;
+
+	if (copy_to_user((struct f2fs_defragment __user *)arg, &range,
+							sizeof(range)))
+		return -EFAULT;
 
 	return 0;
 }
 
+static int f2fs_move_file_range(struct file *file_in, loff_t pos_in,
+			struct file *file_out, loff_t pos_out, size_t len)
+{
+	struct inode *src = file_inode(file_in);
+	struct inode *dst = file_inode(file_out);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(src);
+	size_t olen = len, dst_max_i_size = 0;
+	size_t dst_osize;
+	int ret;
+
+	if (file_in->f_path.mnt != file_out->f_path.mnt ||
+				src->i_sb != dst->i_sb)
+		return -EXDEV;
+
+	if (unlikely(f2fs_readonly(src->i_sb)))
+		return -EROFS;
+
+	if (!S_ISREG(src->i_mode) || !S_ISREG(dst->i_mode))
+		return -EINVAL;
+
+	if (f2fs_encrypted_inode(src) || f2fs_encrypted_inode(dst))
+		return -EOPNOTSUPP;
+
+	if (src == dst) {
+		if (pos_in == pos_out)
+			return 0;
+		if (pos_out > pos_in && pos_out < pos_in + len)
+			return -EINVAL;
+	}
+
+	inode_lock(src);
+	if (src != dst) {
+		if (!inode_trylock(dst)) {
+			ret = -EBUSY;
+			goto out;
+		}
+	}
+
+	ret = -EINVAL;
+	if (pos_in + len > src->i_size || pos_in + len < pos_in)
+		goto out_unlock;
+	if (len == 0)
+		olen = len = src->i_size - pos_in;
+	if (pos_in + len == src->i_size)
+		len = ALIGN(src->i_size, F2FS_BLKSIZE) - pos_in;
+	if (len == 0) {
+		ret = 0;
+		goto out_unlock;
+	}
+
+	dst_osize = dst->i_size;
+	if (pos_out + olen > dst->i_size)
+		dst_max_i_size = pos_out + olen;
+
+	/* verify the end result is block aligned */
+	if (!IS_ALIGNED(pos_in, F2FS_BLKSIZE) ||
+			!IS_ALIGNED(pos_in + len, F2FS_BLKSIZE) ||
+			!IS_ALIGNED(pos_out, F2FS_BLKSIZE))
+		goto out_unlock;
+
+	ret = f2fs_convert_inline_inode(src);
+	if (ret)
+		goto out_unlock;
+
+	ret = f2fs_convert_inline_inode(dst);
+	if (ret)
+		goto out_unlock;
+
+	/* write out all dirty pages from offset */
+	ret = filemap_write_and_wait_range(src->i_mapping,
+					pos_in, pos_in + len);
+	if (ret)
+		goto out_unlock;
+
+	ret = filemap_write_and_wait_range(dst->i_mapping,
+					pos_out, pos_out + len);
+	if (ret)
+		goto out_unlock;
+
+	f2fs_balance_fs(sbi, true);
+	f2fs_lock_op(sbi);
+	ret = __exchange_data_block(src, dst, pos_in >> F2FS_BLKSIZE_BITS,
+				pos_out >> F2FS_BLKSIZE_BITS,
+				len >> F2FS_BLKSIZE_BITS, false);
+
+	if (!ret) {
+		if (dst_max_i_size)
+			f2fs_i_size_write(dst, dst_max_i_size);
+		else if (dst_osize != dst->i_size)
+			f2fs_i_size_write(dst, dst_osize);
+	}
+	f2fs_unlock_op(sbi);
+out_unlock:
+	if (src != dst)
+		inode_unlock(dst);
+out:
+	inode_unlock(src);
+	return ret;
+}
+
+static int f2fs_ioc_move_range(struct file *filp, unsigned long arg)
+{
+	struct f2fs_move_range range;
+	struct fd dst;
+	int err;
+
+	if (!(filp->f_mode & FMODE_READ) ||
+			!(filp->f_mode & FMODE_WRITE))
+		return -EBADF;
+
+	if (copy_from_user(&range, (struct f2fs_move_range __user *)arg,
+							sizeof(range)))
+		return -EFAULT;
+
+	dst = fdget(range.dst_fd);
+	if (!dst.file)
+		return -EBADF;
+
+	if (!(dst.file->f_mode & FMODE_WRITE)) {
+		err = -EBADF;
+		goto err_out;
+	}
+
+	err = mnt_want_write_file(filp);
+	if (err)
+		goto err_out;
+
+	err = f2fs_move_file_range(filp, range.pos_in, dst.file,
+					range.pos_out, range.len);
+
+	mnt_drop_write_file(filp);
+	if (err)
+		goto err_out;
+
+	if (copy_to_user((struct f2fs_move_range __user *)arg,
+						&range, sizeof(range)))
+		err = -EFAULT;
+err_out:
+	fdput(dst);
+	return err;
+}
+
+static int f2fs_ioc_flush_device(struct file *filp, unsigned long arg)
+{
+	struct inode *inode = file_inode(filp);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct sit_info *sm = SIT_I(sbi);
+	unsigned int start_segno = 0, end_segno = 0;
+	unsigned int dev_start_segno = 0, dev_end_segno = 0;
+	struct f2fs_flush_device range;
+	int ret;
+
+	if (!capable(CAP_SYS_ADMIN))
+		return -EPERM;
+
+	if (f2fs_readonly(sbi->sb))
+		return -EROFS;
+
+	if (copy_from_user(&range, (struct f2fs_flush_device __user *)arg,
+							sizeof(range)))
+		return -EFAULT;
+
+	if (sbi->s_ndevs <= 1 || sbi->s_ndevs - 1 <= range.dev_num ||
+			sbi->segs_per_sec != 1) {
+		f2fs_msg(sbi->sb, KERN_WARNING,
+			"Can't flush %u in %d for segs_per_sec %u != 1\n",
+				range.dev_num, sbi->s_ndevs,
+				sbi->segs_per_sec);
+		return -EINVAL;
+	}
+
+	ret = mnt_want_write_file(filp);
+	if (ret)
+		return ret;
+
+	if (range.dev_num != 0)
+		dev_start_segno = GET_SEGNO(sbi, FDEV(range.dev_num).start_blk);
+	dev_end_segno = GET_SEGNO(sbi, FDEV(range.dev_num).end_blk);
+
+	start_segno = sm->last_victim[FLUSH_DEVICE];
+	if (start_segno < dev_start_segno || start_segno >= dev_end_segno)
+		start_segno = dev_start_segno;
+	end_segno = min(start_segno + range.segments, dev_end_segno);
+
+	while (start_segno < end_segno) {
+		if (!mutex_trylock(&sbi->gc_mutex)) {
+			ret = -EBUSY;
+			goto out;
+		}
+		sm->last_victim[GC_CB] = end_segno + 1;
+		sm->last_victim[GC_GREEDY] = end_segno + 1;
+		sm->last_victim[ALLOC_NEXT] = end_segno + 1;
+		ret = f2fs_gc(sbi, true, true, start_segno);
+		if (ret == -EAGAIN)
+			ret = 0;
+		else if (ret < 0)
+			break;
+		start_segno++;
+	}
+out:
+	mnt_drop_write_file(filp);
+	return ret;
+}
+
+static int f2fs_ioc_get_features(struct file *filp, unsigned long arg)
+{
+	struct inode *inode = file_inode(filp);
+	u32 sb_feature = le32_to_cpu(F2FS_I_SB(inode)->raw_super->feature);
+
+	/* Must validate to set it with SQLite behavior in Android. */
+	sb_feature |= F2FS_FEATURE_ATOMIC_WRITE;
+
+	return put_user(sb_feature, (u32 __user *)arg);
+}
+
 long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 {
 	switch (cmd) {
@@ -1690,8 +2441,18 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 		return f2fs_ioc_get_encryption_pwsalt(filp, arg);
 	case F2FS_IOC_GARBAGE_COLLECT:
 		return f2fs_ioc_gc(filp, arg);
+	case F2FS_IOC_GARBAGE_COLLECT_RANGE:
+		return f2fs_ioc_gc_range(filp, arg);
 	case F2FS_IOC_WRITE_CHECKPOINT:
 		return f2fs_ioc_write_checkpoint(filp, arg);
+	case F2FS_IOC_DEFRAGMENT:
+		return f2fs_ioc_defragment(filp, arg);
+	case F2FS_IOC_MOVE_RANGE:
+		return f2fs_ioc_move_range(filp, arg);
+	case F2FS_IOC_FLUSH_DEVICE:
+		return f2fs_ioc_flush_device(filp, arg);
+	case F2FS_IOC_GET_FEATURES:
+		return f2fs_ioc_get_features(filp, arg);
 	default:
 		return -ENOTTY;
 	}
@@ -1699,14 +2460,42 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg)
 
 static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
-	struct inode *inode = file_inode(iocb->ki_filp);
+	struct file *file = iocb->ki_filp;
+	struct inode *inode = file_inode(file);
+	struct blk_plug plug;
+	ssize_t ret;
 
-	if (f2fs_encrypted_inode(inode) &&
-				!f2fs_has_encryption_key(inode) &&
-				f2fs_get_encryption_info(inode))
-		return -EACCES;
+	inode_lock(inode);
+	ret = generic_write_checks(iocb, from);
+	if (ret > 0) {
+		int err;
+
+		if (iov_iter_fault_in_readable(from, iov_iter_count(from)))
+			set_inode_flag(inode, FI_NO_PREALLOC);
+
+		err = f2fs_preallocate_blocks(iocb, from);
+		if (err) {
+			inode_unlock(inode);
+			return err;
+		}
+		blk_start_plug(&plug);
+		ret = __generic_file_write_iter(iocb, from);
+		blk_finish_plug(&plug);
+		clear_inode_flag(inode, FI_NO_PREALLOC);
+
+		if (ret > 0)
+			f2fs_update_iostat(F2FS_I_SB(inode), APP_WRITE_IO, ret);
+	}
+	inode_unlock(inode);
+
+	if (ret > 0) {
+		ssize_t err;
 
-	return generic_file_write_iter(iocb, from);
+		err = generic_write_sync(file, iocb->ki_pos - ret, ret);
+		if (err < 0)
+			ret = err;
+	}
+	return ret;
 }
 
 #ifdef CONFIG_COMPAT
@@ -1719,6 +2508,26 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 	case F2FS_IOC32_SETFLAGS:
 		cmd = F2FS_IOC_SETFLAGS;
 		break;
+	case F2FS_IOC32_GETVERSION:
+		cmd = F2FS_IOC_GETVERSION;
+		break;
+	case F2FS_IOC_START_ATOMIC_WRITE:
+	case F2FS_IOC_COMMIT_ATOMIC_WRITE:
+	case F2FS_IOC_START_VOLATILE_WRITE:
+	case F2FS_IOC_RELEASE_VOLATILE_WRITE:
+	case F2FS_IOC_ABORT_VOLATILE_WRITE:
+	case F2FS_IOC_SHUTDOWN:
+	case F2FS_IOC_SET_ENCRYPTION_POLICY:
+	case F2FS_IOC_GET_ENCRYPTION_PWSALT:
+	case F2FS_IOC_GET_ENCRYPTION_POLICY:
+	case F2FS_IOC_GARBAGE_COLLECT:
+	case F2FS_IOC_GARBAGE_COLLECT_RANGE:
+	case F2FS_IOC_WRITE_CHECKPOINT:
+	case F2FS_IOC_DEFRAGMENT:
+	case F2FS_IOC_MOVE_RANGE:
+	case F2FS_IOC_FLUSH_DEVICE:
+	case F2FS_IOC_GET_FEATURES:
+		break;
 	default:
 		return -ENOIOCTLCMD;
 	}
@@ -1733,6 +2542,7 @@ const struct file_operations f2fs_file_operations = {
 	.open		= f2fs_file_open,
 	.release	= f2fs_release_file,
 	.mmap		= f2fs_file_mmap,
+	.flush		= f2fs_file_flush,
 	.fsync		= f2fs_sync_file,
 	.fallocate	= f2fs_fallocate,
 	.unlocked_ioctl	= f2fs_ioctl,
diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c
index fedbf67a0842..bd16e6631cf3 100644
--- a/fs/f2fs/gc.c
+++ b/fs/f2fs/gc.c
@@ -16,7 +16,6 @@
 #include <linux/kthread.h>
 #include <linux/delay.h>
 #include <linux/freezer.h>
-#include <linux/blkdev.h>
 
 #include "f2fs.h"
 #include "node.h"
@@ -29,17 +28,23 @@ static int gc_thread_func(void *data)
 	struct f2fs_sb_info *sbi = data;
 	struct f2fs_gc_kthread *gc_th = sbi->gc_thread;
 	wait_queue_head_t *wq = &sbi->gc_thread->gc_wait_queue_head;
-	long wait_ms;
+	unsigned int wait_ms;
 
 	wait_ms = gc_th->min_sleep_time;
 
+	set_freezable();
 	do {
+		wait_event_interruptible_timeout(*wq,
+				kthread_should_stop() || freezing(current) ||
+				gc_th->gc_wake,
+				msecs_to_jiffies(wait_ms));
+
+		/* give it a try one time */
+		if (gc_th->gc_wake)
+			gc_th->gc_wake = 0;
+
 		if (try_to_freeze())
 			continue;
-		else
-			wait_event_interruptible_timeout(*wq,
-						kthread_should_stop(),
-						msecs_to_jiffies(wait_ms));
 		if (kthread_should_stop())
 			break;
 
@@ -48,6 +53,16 @@ static int gc_thread_func(void *data)
 			continue;
 		}
 
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+		if (time_to_inject(sbi, FAULT_CHECKPOINT)) {
+			f2fs_show_injection_info(FAULT_CHECKPOINT);
+			f2fs_stop_checkpoint(sbi, false);
+		}
+#endif
+
+		if (!sb_start_write_trylock(sbi->sb))
+			continue;
+
 		/*
 		 * [GC triggering condition]
 		 * 0. GC is not conducted currently.
@@ -62,23 +77,28 @@ static int gc_thread_func(void *data)
 		 * So, I'd like to wait some time to collect dirty segments.
 		 */
 		if (!mutex_trylock(&sbi->gc_mutex))
-			continue;
+			goto next;
+
+		if (gc_th->gc_urgent) {
+			wait_ms = gc_th->urgent_sleep_time;
+			goto do_gc;
+		}
 
 		if (!is_idle(sbi)) {
 			increase_sleep_time(gc_th, &wait_ms);
 			mutex_unlock(&sbi->gc_mutex);
-			continue;
+			goto next;
 		}
 
 		if (has_enough_invalid_blocks(sbi))
 			decrease_sleep_time(gc_th, &wait_ms);
 		else
 			increase_sleep_time(gc_th, &wait_ms);
-
+do_gc:
 		stat_inc_bggc_count(sbi);
 
 		/* if return value is not zero, no victim was selected */
-		if (f2fs_gc(sbi, test_opt(sbi, FORCE_FG_GC)))
+		if (f2fs_gc(sbi, test_opt(sbi, FORCE_FG_GC), true, NULL_SEGNO))
 			wait_ms = gc_th->no_gc_sleep_time;
 
 		trace_f2fs_background_gc(sbi->sb, wait_ms,
@@ -86,6 +106,8 @@ static int gc_thread_func(void *data)
 
 		/* balancing f2fs's metadata periodically */
 		f2fs_balance_fs_bg(sbi);
+next:
+		sb_end_write(sbi->sb);
 
 	} while (!kthread_should_stop());
 	return 0;
@@ -97,17 +119,20 @@ int start_gc_thread(struct f2fs_sb_info *sbi)
 	dev_t dev = sbi->sb->s_bdev->bd_dev;
 	int err = 0;
 
-	gc_th = kmalloc(sizeof(struct f2fs_gc_kthread), GFP_KERNEL);
+	gc_th = f2fs_kmalloc(sbi, sizeof(struct f2fs_gc_kthread), GFP_KERNEL);
 	if (!gc_th) {
 		err = -ENOMEM;
 		goto out;
 	}
 
+	gc_th->urgent_sleep_time = DEF_GC_THREAD_URGENT_SLEEP_TIME;
 	gc_th->min_sleep_time = DEF_GC_THREAD_MIN_SLEEP_TIME;
 	gc_th->max_sleep_time = DEF_GC_THREAD_MAX_SLEEP_TIME;
 	gc_th->no_gc_sleep_time = DEF_GC_THREAD_NOGC_SLEEP_TIME;
 
 	gc_th->gc_idle = 0;
+	gc_th->gc_urgent = 0;
+	gc_th->gc_wake= 0;
 
 	sbi->gc_thread = gc_th;
 	init_waitqueue_head(&sbi->gc_thread->gc_wait_queue_head);
@@ -162,10 +187,15 @@ static void select_policy(struct f2fs_sb_info *sbi, int gc_type,
 		p->ofs_unit = sbi->segs_per_sec;
 	}
 
-	if (p->max_search > sbi->max_victim_search)
+	/* we need to check every dirty segments in the FG_GC case */
+	if (gc_type != FG_GC && p->max_search > sbi->max_victim_search)
 		p->max_search = sbi->max_victim_search;
 
-	p->offset = sbi->last_victim[p->gc_mode];
+	/* let's select beginning hot/small space first */
+	if (type == CURSEG_HOT_DATA || IS_NODESEG(type))
+		p->offset = 0;
+	else
+		p->offset = SIT_I(sbi)->last_victim[p->gc_mode];
 }
 
 static unsigned int get_max_cost(struct f2fs_sb_info *sbi,
@@ -173,9 +203,9 @@ static unsigned int get_max_cost(struct f2fs_sb_info *sbi,
 {
 	/* SSR allocates in a segment unit */
 	if (p->alloc_mode == SSR)
-		return 1 << sbi->log_blocks_per_seg;
+		return sbi->blocks_per_seg;
 	if (p->gc_mode == GC_GREEDY)
-		return (1 << sbi->log_blocks_per_seg) * p->ofs_unit;
+		return 2 * sbi->blocks_per_seg * p->ofs_unit;
 	else if (p->gc_mode == GC_CB)
 		return UINT_MAX;
 	else /* No other gc_mode */
@@ -195,8 +225,12 @@ static unsigned int check_bg_victims(struct f2fs_sb_info *sbi)
 	for_each_set_bit(secno, dirty_i->victim_secmap, MAIN_SECS(sbi)) {
 		if (sec_usage_check(sbi, secno))
 			continue;
+
+		if (no_fggc_candidate(sbi, secno))
+			continue;
+
 		clear_bit(secno, dirty_i->victim_secmap);
-		return secno * sbi->segs_per_sec;
+		return GET_SEG_FROM_SEC(sbi, secno);
 	}
 	return NULL_SEGNO;
 }
@@ -204,8 +238,8 @@ static unsigned int check_bg_victims(struct f2fs_sb_info *sbi)
 static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	unsigned int secno = GET_SECNO(sbi, segno);
-	unsigned int start = secno * sbi->segs_per_sec;
+	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
+	unsigned int start = GET_SEG_FROM_SEC(sbi, secno);
 	unsigned long long mtime = 0;
 	unsigned int vblocks;
 	unsigned char age = 0;
@@ -214,7 +248,7 @@ static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 
 	for (i = 0; i < sbi->segs_per_sec; i++)
 		mtime += get_seg_entry(sbi, start + i)->mtime;
-	vblocks = get_valid_blocks(sbi, segno, sbi->segs_per_sec);
+	vblocks = get_valid_blocks(sbi, segno, true);
 
 	mtime = div_u64(mtime, sbi->segs_per_sec);
 	vblocks = div_u64(vblocks, sbi->segs_per_sec);
@@ -233,6 +267,16 @@ static unsigned int get_cb_cost(struct f2fs_sb_info *sbi, unsigned int segno)
 	return UINT_MAX - ((100 * (100 - u) * age) / (100 + u));
 }
 
+static unsigned int get_greedy_cost(struct f2fs_sb_info *sbi,
+						unsigned int segno)
+{
+	unsigned int valid_blocks =
+			get_valid_blocks(sbi, segno, true);
+
+	return IS_DATASEG(get_seg_entry(sbi, segno)->type) ?
+				valid_blocks * 2 : valid_blocks;
+}
+
 static inline unsigned int get_gc_cost(struct f2fs_sb_info *sbi,
 			unsigned int segno, struct victim_sel_policy *p)
 {
@@ -241,11 +285,23 @@ static inline unsigned int get_gc_cost(struct f2fs_sb_info *sbi,
 
 	/* alloc_mode == LFS */
 	if (p->gc_mode == GC_GREEDY)
-		return get_valid_blocks(sbi, segno, sbi->segs_per_sec);
+		return get_greedy_cost(sbi, segno);
 	else
 		return get_cb_cost(sbi, segno);
 }
 
+static unsigned int count_bits(const unsigned long *addr,
+				unsigned int offset, unsigned int len)
+{
+	unsigned int end = offset + len, sum = 0;
+
+	while (offset < end) {
+		if (test_bit(offset++, addr))
+			++sum;
+	}
+	return sum;
+}
+
 /*
  * This function is called from two paths.
  * One is garbage collection and the other is SSR segment selection.
@@ -258,10 +314,11 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 		unsigned int *result, int gc_type, int type, char alloc_mode)
 {
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
+	struct sit_info *sm = SIT_I(sbi);
 	struct victim_sel_policy p;
-	unsigned int secno, max_cost;
+	unsigned int secno, last_victim;
 	unsigned int last_segment = MAIN_SEGS(sbi);
-	int nsearched = 0;
+	unsigned int nsearched = 0;
 
 	mutex_lock(&dirty_i->seglist_lock);
 
@@ -269,11 +326,20 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 	select_policy(sbi, gc_type, type, &p);
 
 	p.min_segno = NULL_SEGNO;
-	p.min_cost = max_cost = get_max_cost(sbi, &p);
+	p.min_cost = get_max_cost(sbi, &p);
+
+	if (*result != NULL_SEGNO) {
+		if (IS_DATASEG(get_seg_entry(sbi, *result)->type) &&
+			get_valid_blocks(sbi, *result, false) &&
+			!sec_usage_check(sbi, GET_SEC_FROM_SEG(sbi, *result)))
+			p.min_segno = *result;
+		goto out;
+	}
 
 	if (p.max_search == 0)
 		goto out;
 
+	last_victim = sm->last_victim[p.gc_mode];
 	if (p.alloc_mode == LFS && gc_type == FG_GC) {
 		p.min_segno = check_bg_victims(sbi);
 		if (p.min_segno != NULL_SEGNO)
@@ -286,9 +352,10 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 
 		segno = find_next_bit(p.dirty_segmap, last_segment, p.offset);
 		if (segno >= last_segment) {
-			if (sbi->last_victim[p.gc_mode]) {
-				last_segment = sbi->last_victim[p.gc_mode];
-				sbi->last_victim[p.gc_mode] = 0;
+			if (sm->last_victim[p.gc_mode]) {
+				last_segment =
+					sm->last_victim[p.gc_mode];
+				sm->last_victim[p.gc_mode] = 0;
 				p.offset = 0;
 				continue;
 			}
@@ -296,34 +363,45 @@ static int get_victim_by_default(struct f2fs_sb_info *sbi,
 		}
 
 		p.offset = segno + p.ofs_unit;
-		if (p.ofs_unit > 1)
+		if (p.ofs_unit > 1) {
 			p.offset -= segno % p.ofs_unit;
+			nsearched += count_bits(p.dirty_segmap,
+						p.offset - p.ofs_unit,
+						p.ofs_unit);
+		} else {
+			nsearched++;
+		}
 
-		secno = GET_SECNO(sbi, segno);
+		secno = GET_SEC_FROM_SEG(sbi, segno);
 
 		if (sec_usage_check(sbi, secno))
-			continue;
+			goto next;
 		if (gc_type == BG_GC && test_bit(secno, dirty_i->victim_secmap))
-			continue;
+			goto next;
+		if (gc_type == FG_GC && p.alloc_mode == LFS &&
+					no_fggc_candidate(sbi, secno))
+			goto next;
 
 		cost = get_gc_cost(sbi, segno, &p);
 
 		if (p.min_cost > cost) {
 			p.min_segno = segno;
 			p.min_cost = cost;
-		} else if (unlikely(cost == max_cost)) {
-			continue;
 		}
-
-		if (nsearched++ >= p.max_search) {
-			sbi->last_victim[p.gc_mode] = segno;
+next:
+		if (nsearched >= p.max_search) {
+			if (!sm->last_victim[p.gc_mode] && segno <= last_victim)
+				sm->last_victim[p.gc_mode] = last_victim + 1;
+			else
+				sm->last_victim[p.gc_mode] = segno + 1;
+			sm->last_victim[p.gc_mode] %= MAIN_SEGS(sbi);
 			break;
 		}
 	}
 	if (p.min_segno != NULL_SEGNO) {
 got_it:
 		if (p.alloc_mode == LFS) {
-			secno = GET_SECNO(sbi, p.min_segno);
+			secno = GET_SEC_FROM_SEG(sbi, p.min_segno);
 			if (gc_type == FG_GC)
 				sbi->cur_victim_sec = secno;
 			else
@@ -400,13 +478,13 @@ static int check_valid_map(struct f2fs_sb_info *sbi,
  * On validity, copy that node with cold status, otherwise (invalid node)
  * ignore that.
  */
-static int gc_node_segment(struct f2fs_sb_info *sbi,
+static void gc_node_segment(struct f2fs_sb_info *sbi,
 		struct f2fs_summary *sum, unsigned int segno, int gc_type)
 {
-	bool initial = true;
 	struct f2fs_summary *entry;
 	block_t start_addr;
 	int off;
+	int phase = 0;
 
 	start_addr = START_BLOCK(sbi, segno);
 
@@ -419,16 +497,24 @@ next_step:
 		struct node_info ni;
 
 		/* stop BG_GC if there is not enough free sections. */
-		if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0))
-			return 0;
+		if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0, 0))
+			return;
 
 		if (check_valid_map(sbi, segno, off) == 0)
 			continue;
 
-		if (initial) {
+		if (phase == 0) {
+			ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), 1,
+							META_NAT, true);
+			continue;
+		}
+
+		if (phase == 1) {
 			ra_node_page(sbi, nid);
 			continue;
 		}
+
+		/* phase == 2 */
 		node_page = get_node_page(sbi, nid);
 		if (IS_ERR(node_page))
 			continue;
@@ -445,36 +531,12 @@ next_step:
 			continue;
 		}
 
-		/* set page dirty and write it */
-		if (gc_type == FG_GC) {
-			f2fs_wait_on_page_writeback(node_page, NODE);
-			set_page_dirty(node_page);
-		} else {
-			if (!PageWriteback(node_page))
-				set_page_dirty(node_page);
-		}
-		f2fs_put_page(node_page, 1);
+		move_node_page(node_page, gc_type);
 		stat_inc_node_blk_count(sbi, 1, gc_type);
 	}
 
-	if (initial) {
-		initial = false;
+	if (++phase < 3)
 		goto next_step;
-	}
-
-	if (gc_type == FG_GC) {
-		struct writeback_control wbc = {
-			.sync_mode = WB_SYNC_ALL,
-			.nr_to_write = LONG_MAX,
-			.for_reclaim = 0,
-		};
-		sync_node_pages(sbi, 0, &wbc);
-
-		/* return 1 only if FG_GC succefully reclaimed one */
-		if (get_valid_blocks(sbi, segno, 1) == 0)
-			return 1;
-	}
-	return 0;
 }
 
 /*
@@ -484,7 +546,7 @@ next_step:
  * as indirect or double indirect node blocks, are given, it must be a caller's
  * bug.
  */
-block_t start_bidx_of_node(unsigned int node_ofs, struct f2fs_inode_info *fi)
+block_t start_bidx_of_node(unsigned int node_ofs, struct inode *inode)
 {
 	unsigned int indirect_blks = 2 * NIDS_PER_BLOCK + 4;
 	unsigned int bidx;
@@ -501,7 +563,7 @@ block_t start_bidx_of_node(unsigned int node_ofs, struct f2fs_inode_info *fi)
 		int dec = (node_ofs - indirect_blks - 3) / (NIDS_PER_BLOCK + 1);
 		bidx = node_ofs - 5 - dec;
 	}
-	return bidx * ADDRS_PER_BLOCK + ADDRS_PER_INODE(fi);
+	return bidx * ADDRS_PER_BLOCK + ADDRS_PER_INODE(inode);
 }
 
 static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
@@ -522,12 +584,14 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
 	get_node_info(sbi, nid, dni);
 
 	if (sum->version != dni->version) {
-		f2fs_put_page(node_page, 1);
-		return false;
+		f2fs_msg(sbi->sb, KERN_WARNING,
+				"%s: valid data with mismatched node version.",
+				__func__);
+		set_sbi_flag(sbi, SBI_NEED_FSCK);
 	}
 
 	*nofs = ofs_of_node(node_page);
-	source_blkaddr = datablock_addr(node_page, ofs_in_node);
+	source_blkaddr = datablock_addr(NULL, node_page, ofs_in_node);
 	f2fs_put_page(node_page, 1);
 
 	if (source_blkaddr != blkaddr)
@@ -535,18 +599,27 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
 	return true;
 }
 
-static void move_encrypted_block(struct inode *inode, block_t bidx)
+/*
+ * Move data block via META_MAPPING while keeping locked data page.
+ * This can be used to move blocks, aka LBAs, directly on disk.
+ */
+static void move_data_block(struct inode *inode, block_t bidx,
+					unsigned int segno, int off)
 {
 	struct f2fs_io_info fio = {
 		.sbi = F2FS_I_SB(inode),
 		.type = DATA,
-		.rw = READ_SYNC,
+		.temp = COLD,
+		.op = REQ_OP_READ,
+		.op_flags = REQ_SYNC,
 		.encrypted_page = NULL,
+		.in_list = false,
 	};
 	struct dnode_of_data dn;
 	struct f2fs_summary sum;
 	struct node_info ni;
 	struct page *page;
+	block_t newaddr;
 	int err;
 
 	/* do not read out */
@@ -554,6 +627,12 @@ static void move_encrypted_block(struct inode *inode, block_t bidx)
 	if (!page)
 		return;
 
+	if (!check_valid_map(F2FS_I_SB(inode), segno, off))
+		goto out;
+
+	if (f2fs_is_atomic_file(inode))
+		goto out;
+
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
 	err = get_dnode_of_data(&dn, bidx, LOOKUP_NODE);
 	if (err)
@@ -568,21 +647,24 @@ static void move_encrypted_block(struct inode *inode, block_t bidx)
 	 * don't cache encrypted data into meta inode until previous dirty
 	 * data were writebacked to avoid racing between GC and flush.
 	 */
-	f2fs_wait_on_page_writeback(page, DATA);
+	f2fs_wait_on_page_writeback(page, DATA, true);
 
 	get_node_info(fio.sbi, dn.nid, &ni);
 	set_summary(&sum, dn.nid, dn.ofs_in_node, ni.version);
 
 	/* read page */
 	fio.page = page;
-	fio.blk_addr = dn.data_blkaddr;
+	fio.new_blkaddr = fio.old_blkaddr = dn.data_blkaddr;
 
-	fio.encrypted_page = pagecache_get_page(META_MAPPING(fio.sbi),
-					fio.blk_addr,
-					FGP_LOCK|FGP_CREAT,
-					GFP_NOFS);
-	if (!fio.encrypted_page)
-		goto put_out;
+	allocate_data_block(fio.sbi, NULL, fio.old_blkaddr, &newaddr,
+					&sum, CURSEG_COLD_DATA, NULL, false);
+
+	fio.encrypted_page = pagecache_get_page(META_MAPPING(fio.sbi), newaddr,
+					FGP_LOCK | FGP_CREAT, GFP_NOFS);
+	if (!fio.encrypted_page) {
+		err = -ENOMEM;
+		goto recover_block;
+	}
 
 	err = f2fs_submit_page_bio(&fio);
 	if (err)
@@ -591,40 +673,50 @@ static void move_encrypted_block(struct inode *inode, block_t bidx)
 	/* write page */
 	lock_page(fio.encrypted_page);
 
-	if (unlikely(!PageUptodate(fio.encrypted_page)))
+	if (unlikely(fio.encrypted_page->mapping != META_MAPPING(fio.sbi))) {
+		err = -EIO;
 		goto put_page_out;
-	if (unlikely(fio.encrypted_page->mapping != META_MAPPING(fio.sbi)))
+	}
+	if (unlikely(!PageUptodate(fio.encrypted_page))) {
+		err = -EIO;
 		goto put_page_out;
+	}
 
 	set_page_dirty(fio.encrypted_page);
-	f2fs_wait_on_page_writeback(fio.encrypted_page, DATA);
+	f2fs_wait_on_page_writeback(fio.encrypted_page, DATA, true);
 	if (clear_page_dirty_for_io(fio.encrypted_page))
 		dec_page_count(fio.sbi, F2FS_DIRTY_META);
 
 	set_page_writeback(fio.encrypted_page);
 
 	/* allocate block address */
-	f2fs_wait_on_page_writeback(dn.node_page, NODE);
-	allocate_data_block(fio.sbi, NULL, fio.blk_addr,
-					&fio.blk_addr, &sum, CURSEG_COLD_DATA);
-	fio.rw = WRITE_SYNC;
-	f2fs_submit_page_mbio(&fio);
-
-	dn.data_blkaddr = fio.blk_addr;
-	set_data_blkaddr(&dn);
-	f2fs_update_extent_cache(&dn);
-	set_inode_flag(F2FS_I(inode), FI_APPEND_WRITE);
+	f2fs_wait_on_page_writeback(dn.node_page, NODE, true);
+
+	fio.op = REQ_OP_WRITE;
+	fio.op_flags = REQ_SYNC | REQ_NOIDLE;
+	fio.new_blkaddr = newaddr;
+	f2fs_submit_page_write(&fio);
+
+	f2fs_update_iostat(fio.sbi, FS_GC_DATA_IO, F2FS_BLKSIZE);
+
+	f2fs_update_data_blkaddr(&dn, newaddr);
+	set_inode_flag(inode, FI_APPEND_WRITE);
 	if (page->index == 0)
-		set_inode_flag(F2FS_I(inode), FI_FIRST_BLOCK_WRITTEN);
+		set_inode_flag(inode, FI_FIRST_BLOCK_WRITTEN);
 put_page_out:
 	f2fs_put_page(fio.encrypted_page, 1);
+recover_block:
+	if (err)
+		__f2fs_replace_block(fio.sbi, &sum, newaddr, fio.old_blkaddr,
+								true, true);
 put_out:
 	f2fs_put_dnode(&dn);
 out:
 	f2fs_put_page(page, 1);
 }
 
-static void move_data_page(struct inode *inode, block_t bidx, int gc_type)
+static void move_data_page(struct inode *inode, block_t bidx, int gc_type,
+							unsigned int segno, int off)
 {
 	struct page *page;
 
@@ -632,6 +724,12 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type)
 	if (IS_ERR(page))
 		return;
 
+	if (!check_valid_map(F2FS_I_SB(inode), segno, off))
+		goto out;
+
+	if (f2fs_is_atomic_file(inode))
+		goto out;
+
 	if (gc_type == BG_GC) {
 		if (PageWriteback(page))
 			goto out;
@@ -641,17 +739,33 @@ static void move_data_page(struct inode *inode, block_t bidx, int gc_type)
 		struct f2fs_io_info fio = {
 			.sbi = F2FS_I_SB(inode),
 			.type = DATA,
-			.rw = WRITE_SYNC,
+			.temp = COLD,
+			.op = REQ_OP_WRITE,
+			.op_flags = REQ_SYNC,
+			.old_blkaddr = NULL_ADDR,
 			.page = page,
 			.encrypted_page = NULL,
+			.need_lock = LOCK_REQ,
+			.io_type = FS_GC_DATA_IO,
 		};
+		bool is_dirty = PageDirty(page);
+		int err;
+
+retry:
 		set_page_dirty(page);
-		f2fs_wait_on_page_writeback(page, DATA);
-		if (clear_page_dirty_for_io(page))
+		f2fs_wait_on_page_writeback(page, DATA, true);
+		if (clear_page_dirty_for_io(page)) {
 			inode_dec_dirty_pages(inode);
+			remove_dirty_inode(inode);
+		}
+
 		set_cold_data(page);
-		do_write_data_page(&fio);
-		clear_cold_data(page);
+
+		err = do_write_data_page(&fio);
+		if (err == -ENOMEM && is_dirty) {
+			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			goto retry;
+		}
 	}
 out:
 	f2fs_put_page(page, 1);
@@ -664,7 +778,7 @@ out:
  * If the parent node is not valid or the data block address is different,
  * the victim data block is ignored.
  */
-static int gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
+static void gc_data_segment(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
 		struct gc_inode_list *gc_list, unsigned int segno, int gc_type)
 {
 	struct super_block *sb = sbi->sb;
@@ -684,16 +798,23 @@ next_step:
 		struct node_info dni; /* dnode info for the data */
 		unsigned int ofs_in_node, nofs;
 		block_t start_bidx;
+		nid_t nid = le32_to_cpu(entry->nid);
 
 		/* stop BG_GC if there is not enough free sections. */
-		if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0))
-			return 0;
+		if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0, 0))
+			return;
 
 		if (check_valid_map(sbi, segno, off) == 0)
 			continue;
 
 		if (phase == 0) {
-			ra_node_page(sbi, le32_to_cpu(entry->nid));
+			ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), 1,
+							META_NAT, true);
+			continue;
+		}
+
+		if (phase == 1) {
+			ra_node_page(sbi, nid);
 			continue;
 		}
 
@@ -701,28 +822,28 @@ next_step:
 		if (!is_alive(sbi, entry, &dni, start_addr + off, &nofs))
 			continue;
 
-		if (phase == 1) {
+		if (phase == 2) {
 			ra_node_page(sbi, dni.ino);
 			continue;
 		}
 
 		ofs_in_node = le16_to_cpu(entry->ofs_in_node);
 
-		if (phase == 2) {
+		if (phase == 3) {
 			inode = f2fs_iget(sb, dni.ino);
 			if (IS_ERR(inode) || is_bad_inode(inode))
 				continue;
 
 			/* if encrypted inode, let's go phase 3 */
-			if (f2fs_encrypted_inode(inode) &&
-						S_ISREG(inode->i_mode)) {
+			if (f2fs_encrypted_file(inode)) {
 				add_gc_inode(gc_list, inode);
 				continue;
 			}
 
-			start_bidx = start_bidx_of_node(nofs, F2FS_I(inode));
+			start_bidx = start_bidx_of_node(nofs, inode);
 			data_page = get_read_data_page(inode,
-					start_bidx + ofs_in_node, READA, true);
+					start_bidx + ofs_in_node, REQ_RAHEAD,
+					true);
 			if (IS_ERR(data_page)) {
 				iput(inode);
 				continue;
@@ -733,30 +854,45 @@ next_step:
 			continue;
 		}
 
-		/* phase 3 */
+		/* phase 4 */
 		inode = find_gc_inode(gc_list, dni.ino);
 		if (inode) {
-			start_bidx = start_bidx_of_node(nofs, F2FS_I(inode))
+			struct f2fs_inode_info *fi = F2FS_I(inode);
+			bool locked = false;
+
+			if (S_ISREG(inode->i_mode)) {
+				if (!down_write_trylock(&fi->dio_rwsem[READ]))
+					continue;
+				if (!down_write_trylock(
+						&fi->dio_rwsem[WRITE])) {
+					up_write(&fi->dio_rwsem[READ]);
+					continue;
+				}
+				locked = true;
+
+				/* wait for all inflight aio data */
+				inode_dio_wait(inode);
+			}
+
+			start_bidx = start_bidx_of_node(nofs, inode)
 								+ ofs_in_node;
-			if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
-				move_encrypted_block(inode, start_bidx);
+			if (f2fs_encrypted_file(inode))
+				move_data_block(inode, start_bidx, segno, off);
 			else
-				move_data_page(inode, start_bidx, gc_type);
+				move_data_page(inode, start_bidx, gc_type,
+								segno, off);
+
+			if (locked) {
+				up_write(&fi->dio_rwsem[WRITE]);
+				up_write(&fi->dio_rwsem[READ]);
+			}
+
 			stat_inc_data_blk_count(sbi, 1, gc_type);
 		}
 	}
 
-	if (++phase < 4)
+	if (++phase < 5)
 		goto next_step;
-
-	if (gc_type == FG_GC) {
-		f2fs_submit_merged_bio(sbi, DATA, WRITE);
-
-		/* return 1 only if FG_GC succefully reclaimed one */
-		if (get_valid_blocks(sbi, segno, 1) == 0)
-			return 1;
-	}
-	return 0;
 }
 
 static int __get_victim(struct f2fs_sb_info *sbi, unsigned int *victim,
@@ -772,108 +908,168 @@ static int __get_victim(struct f2fs_sb_info *sbi, unsigned int *victim,
 	return ret;
 }
 
-static int do_garbage_collect(struct f2fs_sb_info *sbi, unsigned int segno,
+static int do_garbage_collect(struct f2fs_sb_info *sbi,
+				unsigned int start_segno,
 				struct gc_inode_list *gc_list, int gc_type)
 {
 	struct page *sum_page;
 	struct f2fs_summary_block *sum;
 	struct blk_plug plug;
-	int nfree = 0;
+	unsigned int segno = start_segno;
+	unsigned int end_segno = start_segno + sbi->segs_per_sec;
+	int seg_freed = 0;
+	unsigned char type = IS_DATASEG(get_seg_entry(sbi, segno)->type) ?
+						SUM_TYPE_DATA : SUM_TYPE_NODE;
+
+	/* readahead multi ssa blocks those have contiguous address */
+	if (sbi->segs_per_sec > 1)
+		ra_meta_pages(sbi, GET_SUM_BLOCK(sbi, segno),
+					sbi->segs_per_sec, META_SSA, true);
 
-	/* read segment summary of victim */
-	sum_page = get_sum_page(sbi, segno);
+	/* reference all summary page */
+	while (segno < end_segno) {
+		sum_page = get_sum_page(sbi, segno++);
+		unlock_page(sum_page);
+	}
 
 	blk_start_plug(&plug);
 
-	sum = page_address(sum_page);
+	for (segno = start_segno; segno < end_segno; segno++) {
 
-	/*
-	 * this is to avoid deadlock:
-	 * - lock_page(sum_page)         - f2fs_replace_block
-	 *  - check_valid_map()            - mutex_lock(sentry_lock)
-	 *   - mutex_lock(sentry_lock)     - change_curseg()
-	 *                                  - lock_page(sum_page)
-	 */
-	unlock_page(sum_page);
-
-	switch (GET_SUM_TYPE((&sum->footer))) {
-	case SUM_TYPE_NODE:
-		nfree = gc_node_segment(sbi, sum->entries, segno, gc_type);
-		break;
-	case SUM_TYPE_DATA:
-		nfree = gc_data_segment(sbi, sum->entries, gc_list,
-							segno, gc_type);
-		break;
+		/* find segment summary of victim */
+		sum_page = find_get_page(META_MAPPING(sbi),
+					GET_SUM_BLOCK(sbi, segno));
+		f2fs_put_page(sum_page, 0);
+
+		if (get_valid_blocks(sbi, segno, false) == 0 ||
+				!PageUptodate(sum_page) ||
+				unlikely(f2fs_cp_error(sbi)))
+			goto next;
+
+		sum = page_address(sum_page);
+		f2fs_bug_on(sbi, type != GET_SUM_TYPE((&sum->footer)));
+
+		/*
+		 * this is to avoid deadlock:
+		 * - lock_page(sum_page)         - f2fs_replace_block
+		 *  - check_valid_map()            - mutex_lock(sentry_lock)
+		 *   - mutex_lock(sentry_lock)     - change_curseg()
+		 *                                  - lock_page(sum_page)
+		 */
+		if (type == SUM_TYPE_NODE)
+			gc_node_segment(sbi, sum->entries, segno, gc_type);
+		else
+			gc_data_segment(sbi, sum->entries, gc_list, segno,
+								gc_type);
+
+		stat_inc_seg_count(sbi, type, gc_type);
+
+		if (gc_type == FG_GC &&
+				get_valid_blocks(sbi, segno, false) == 0)
+			seg_freed++;
+next:
+		f2fs_put_page(sum_page, 0);
 	}
+
+	if (gc_type == FG_GC)
+		f2fs_submit_merged_write(sbi,
+				(type == SUM_TYPE_NODE) ? NODE : DATA);
+
 	blk_finish_plug(&plug);
 
-	stat_inc_seg_count(sbi, GET_SUM_TYPE((&sum->footer)), gc_type);
 	stat_inc_call_count(sbi->stat_info);
 
-	f2fs_put_page(sum_page, 0);
-	return nfree;
+	return seg_freed;
 }
 
-int f2fs_gc(struct f2fs_sb_info *sbi, bool sync)
+int f2fs_gc(struct f2fs_sb_info *sbi, bool sync,
+			bool background, unsigned int segno)
 {
-	unsigned int segno, i;
 	int gc_type = sync ? FG_GC : BG_GC;
-	int sec_freed = 0;
-	int ret = -EINVAL;
+	int sec_freed = 0, seg_freed = 0, total_freed = 0;
+	int ret = 0;
 	struct cp_control cpc;
+	unsigned int init_segno = segno;
 	struct gc_inode_list gc_list = {
 		.ilist = LIST_HEAD_INIT(gc_list.ilist),
 		.iroot = RADIX_TREE_INIT(GFP_NOFS),
 	};
 
+	trace_f2fs_gc_begin(sbi->sb, sync, background,
+				get_pages(sbi, F2FS_DIRTY_NODES),
+				get_pages(sbi, F2FS_DIRTY_DENTS),
+				get_pages(sbi, F2FS_DIRTY_IMETA),
+				free_sections(sbi),
+				free_segments(sbi),
+				reserved_segments(sbi),
+				prefree_segments(sbi));
+
 	cpc.reason = __get_cp_reason(sbi);
 gc_more:
-	segno = NULL_SEGNO;
-
-	if (unlikely(!(sbi->sb->s_flags & MS_ACTIVE)))
+	if (unlikely(!(sbi->sb->s_flags & MS_ACTIVE))) {
+		ret = -EINVAL;
 		goto stop;
-	if (unlikely(f2fs_cp_error(sbi)))
-		goto stop;
-
-	if (gc_type == BG_GC && has_not_enough_free_secs(sbi, sec_freed)) {
-		gc_type = FG_GC;
-		if (__get_victim(sbi, &segno, gc_type) || prefree_segments(sbi))
-			write_checkpoint(sbi, &cpc);
 	}
-
-	if (segno == NULL_SEGNO && !__get_victim(sbi, &segno, gc_type))
+	if (unlikely(f2fs_cp_error(sbi))) {
+		ret = -EIO;
 		goto stop;
-	ret = 0;
-
-	/* readahead multi ssa blocks those have contiguous address */
-	if (sbi->segs_per_sec > 1)
-		ra_meta_pages(sbi, GET_SUM_BLOCK(sbi, segno), sbi->segs_per_sec,
-							META_SSA, true);
+	}
 
-	for (i = 0; i < sbi->segs_per_sec; i++) {
+	if (gc_type == BG_GC && has_not_enough_free_secs(sbi, 0, 0)) {
 		/*
-		 * for FG_GC case, halt gcing left segments once failed one
-		 * of segments in selected section to avoid long latency.
+		 * For example, if there are many prefree_segments below given
+		 * threshold, we can make them free by checkpoint. Then, we
+		 * secure free segments which doesn't need fggc any more.
 		 */
-		if (!do_garbage_collect(sbi, segno + i, &gc_list, gc_type) &&
-				gc_type == FG_GC)
-			break;
+		if (prefree_segments(sbi)) {
+			ret = write_checkpoint(sbi, &cpc);
+			if (ret)
+				goto stop;
+		}
+		if (has_not_enough_free_secs(sbi, 0, 0))
+			gc_type = FG_GC;
+	}
+
+	/* f2fs_balance_fs doesn't need to do BG_GC in critical path. */
+	if (gc_type == BG_GC && !background) {
+		ret = -EINVAL;
+		goto stop;
+	}
+	if (!__get_victim(sbi, &segno, gc_type)) {
+		ret = -ENODATA;
+		goto stop;
 	}
 
-	if (i == sbi->segs_per_sec && gc_type == FG_GC)
+	seg_freed = do_garbage_collect(sbi, segno, &gc_list, gc_type);
+	if (gc_type == FG_GC && seg_freed == sbi->segs_per_sec)
 		sec_freed++;
+	total_freed += seg_freed;
 
 	if (gc_type == FG_GC)
 		sbi->cur_victim_sec = NULL_SEGNO;
 
 	if (!sync) {
-		if (has_not_enough_free_secs(sbi, sec_freed))
+		if (has_not_enough_free_secs(sbi, sec_freed, 0)) {
+			segno = NULL_SEGNO;
 			goto gc_more;
+		}
 
 		if (gc_type == FG_GC)
-			write_checkpoint(sbi, &cpc);
+			ret = write_checkpoint(sbi, &cpc);
 	}
 stop:
+	SIT_I(sbi)->last_victim[ALLOC_NEXT] = 0;
+	SIT_I(sbi)->last_victim[FLUSH_DEVICE] = init_segno;
+
+	trace_f2fs_gc_end(sbi->sb, ret, total_freed, sec_freed,
+				get_pages(sbi, F2FS_DIRTY_NODES),
+				get_pages(sbi, F2FS_DIRTY_DENTS),
+				get_pages(sbi, F2FS_DIRTY_IMETA),
+				free_sections(sbi),
+				free_segments(sbi),
+				reserved_segments(sbi),
+				prefree_segments(sbi));
+
 	mutex_unlock(&sbi->gc_mutex);
 
 	put_gc_inode(&gc_list);
@@ -885,5 +1081,20 @@ stop:
 
 void build_gc_manager(struct f2fs_sb_info *sbi)
 {
+	u64 main_count, resv_count, ovp_count;
+
 	DIRTY_I(sbi)->v_ops = &default_v_ops;
+
+	/* threshold of # of valid blocks in a section for victims of FG_GC */
+	main_count = SM_I(sbi)->main_segments << sbi->log_blocks_per_seg;
+	resv_count = SM_I(sbi)->reserved_segments << sbi->log_blocks_per_seg;
+	ovp_count = SM_I(sbi)->ovp_segments << sbi->log_blocks_per_seg;
+
+	sbi->fggc_threshold = div64_u64((main_count - ovp_count) *
+				BLKS_PER_SEC(sbi), (main_count - resv_count));
+
+	/* give warm/cold data area from slower device */
+	if (sbi->s_ndevs && sbi->segs_per_sec == 1)
+		SIT_I(sbi)->last_victim[ALLOC_NEXT] =
+				GET_SEGNO(sbi, FDEV(0).end_blk) + 1;
 }
diff --git a/fs/f2fs/gc.h b/fs/f2fs/gc.h
index b4a65be9f7d3..9325191fab2d 100644
--- a/fs/f2fs/gc.h
+++ b/fs/f2fs/gc.h
@@ -13,6 +13,7 @@
 						 * whether IO subsystem is idle
 						 * or not
 						 */
+#define DEF_GC_THREAD_URGENT_SLEEP_TIME	500	/* 500 ms */
 #define DEF_GC_THREAD_MIN_SLEEP_TIME	30000	/* milliseconds */
 #define DEF_GC_THREAD_MAX_SLEEP_TIME	60000
 #define DEF_GC_THREAD_NOGC_SLEEP_TIME	300000	/* wait 5 min */
@@ -27,12 +28,15 @@ struct f2fs_gc_kthread {
 	wait_queue_head_t gc_wait_queue_head;
 
 	/* for gc sleep time */
+	unsigned int urgent_sleep_time;
 	unsigned int min_sleep_time;
 	unsigned int max_sleep_time;
 	unsigned int no_gc_sleep_time;
 
 	/* for changing gc mode */
 	unsigned int gc_idle;
+	unsigned int gc_urgent;
+	unsigned int gc_wake;
 };
 
 struct gc_inode_list {
@@ -65,25 +69,32 @@ static inline block_t limit_free_user_blocks(struct f2fs_sb_info *sbi)
 }
 
 static inline void increase_sleep_time(struct f2fs_gc_kthread *gc_th,
-								long *wait)
+							unsigned int *wait)
 {
+	unsigned int min_time = gc_th->min_sleep_time;
+	unsigned int max_time = gc_th->max_sleep_time;
+
 	if (*wait == gc_th->no_gc_sleep_time)
 		return;
 
-	*wait += gc_th->min_sleep_time;
-	if (*wait > gc_th->max_sleep_time)
-		*wait = gc_th->max_sleep_time;
+	if ((long long)*wait + (long long)min_time > (long long)max_time)
+		*wait = max_time;
+	else
+		*wait += min_time;
 }
 
 static inline void decrease_sleep_time(struct f2fs_gc_kthread *gc_th,
-								long *wait)
+							unsigned int *wait)
 {
+	unsigned int min_time = gc_th->min_sleep_time;
+
 	if (*wait == gc_th->no_gc_sleep_time)
 		*wait = gc_th->max_sleep_time;
 
-	*wait -= gc_th->min_sleep_time;
-	if (*wait <= gc_th->min_sleep_time)
-		*wait = gc_th->min_sleep_time;
+	if ((long long)*wait - (long long)min_time < (long long)min_time)
+		*wait = min_time;
+	else
+		*wait -= min_time;
 }
 
 static inline bool has_enough_invalid_blocks(struct f2fs_sb_info *sbi)
@@ -100,11 +111,3 @@ static inline bool has_enough_invalid_blocks(struct f2fs_sb_info *sbi)
 		return true;
 	return false;
 }
-
-static inline int is_idle(struct f2fs_sb_info *sbi)
-{
-	struct block_device *bdev = sbi->sb->s_bdev;
-	struct request_queue *q = bdev_get_queue(bdev);
-	struct request_list *rl = &q->root_rl;
-	return !(rl->count[BLK_RW_SYNC]) && !(rl->count[BLK_RW_ASYNC]);
-}
diff --git a/fs/f2fs/hash.c b/fs/f2fs/hash.c
index b238d2fec3e5..eb2e031ea887 100644
--- a/fs/f2fs/hash.c
+++ b/fs/f2fs/hash.c
@@ -71,7 +71,7 @@ static void str2hashbuf(const unsigned char *msg, size_t len,
 }
 
 f2fs_hash_t f2fs_dentry_hash(const struct qstr *name_info,
-				struct f2fs_filename *fname)
+				struct fscrypt_name *fname)
 {
 	__u32 hash;
 	f2fs_hash_t f2fs_hash;
diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c
index ad80f916b64d..fbf22b0f667f 100644
--- a/fs/f2fs/inline.c
+++ b/fs/f2fs/inline.c
@@ -13,22 +13,20 @@
 
 #include "f2fs.h"
 #include "node.h"
+#include <trace/events/android_fs.h>
 
 bool f2fs_may_inline_data(struct inode *inode)
 {
-	if (!test_opt(F2FS_I_SB(inode), INLINE_DATA))
-		return false;
-
 	if (f2fs_is_atomic_file(inode))
 		return false;
 
 	if (!S_ISREG(inode->i_mode) && !S_ISLNK(inode->i_mode))
 		return false;
 
-	if (i_size_read(inode) > MAX_INLINE_DATA)
+	if (i_size_read(inode) > MAX_INLINE_DATA(inode))
 		return false;
 
-	if (f2fs_encrypted_inode(inode) && S_ISREG(inode->i_mode))
+	if (f2fs_encrypted_file(inode))
 		return false;
 
 	return true;
@@ -47,6 +45,7 @@ bool f2fs_may_inline_dentry(struct inode *inode)
 
 void read_inline_data(struct page *page, struct page *ipage)
 {
+	struct inode *inode = page->mapping->host;
 	void *src_addr, *dst_addr;
 
 	if (PageUptodate(page))
@@ -54,72 +53,92 @@ void read_inline_data(struct page *page, struct page *ipage)
 
 	f2fs_bug_on(F2FS_P_SB(page), page->index);
 
-	zero_user_segment(page, MAX_INLINE_DATA, PAGE_CACHE_SIZE);
+	zero_user_segment(page, MAX_INLINE_DATA(inode), PAGE_SIZE);
 
 	/* Copy the whole inline data block */
-	src_addr = inline_data_addr(ipage);
+	src_addr = inline_data_addr(inode, ipage);
 	dst_addr = kmap_atomic(page);
-	memcpy(dst_addr, src_addr, MAX_INLINE_DATA);
+	memcpy(dst_addr, src_addr, MAX_INLINE_DATA(inode));
 	flush_dcache_page(page);
 	kunmap_atomic(dst_addr);
-	SetPageUptodate(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
 }
 
-bool truncate_inline_inode(struct page *ipage, u64 from)
+void truncate_inline_inode(struct inode *inode, struct page *ipage, u64 from)
 {
 	void *addr;
 
-	if (from >= MAX_INLINE_DATA)
-		return false;
+	if (from >= MAX_INLINE_DATA(inode))
+		return;
 
-	addr = inline_data_addr(ipage);
+	addr = inline_data_addr(inode, ipage);
 
-	f2fs_wait_on_page_writeback(ipage, NODE);
-	memset(addr + from, 0, MAX_INLINE_DATA - from);
+	f2fs_wait_on_page_writeback(ipage, NODE, true);
+	memset(addr + from, 0, MAX_INLINE_DATA(inode) - from);
+	set_page_dirty(ipage);
 
-	return true;
+	if (from == 0)
+		clear_inode_flag(inode, FI_DATA_EXIST);
 }
 
 int f2fs_read_inline_data(struct inode *inode, struct page *page)
 {
 	struct page *ipage;
 
+	if (trace_android_fs_dataread_start_enabled()) {
+		char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+		path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    inode);
+		trace_android_fs_dataread_start(inode, page_offset(page),
+						PAGE_SIZE, current->pid,
+						path, current->comm);
+	}
+
 	ipage = get_node_page(F2FS_I_SB(inode), inode->i_ino);
 	if (IS_ERR(ipage)) {
+		trace_android_fs_dataread_end(inode, page_offset(page),
+					      PAGE_SIZE);
 		unlock_page(page);
 		return PTR_ERR(ipage);
 	}
 
 	if (!f2fs_has_inline_data(inode)) {
 		f2fs_put_page(ipage, 1);
+		trace_android_fs_dataread_end(inode, page_offset(page),
+					      PAGE_SIZE);
 		return -EAGAIN;
 	}
 
 	if (page->index)
-		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
+		zero_user_segment(page, 0, PAGE_SIZE);
 	else
 		read_inline_data(page, ipage);
 
-	SetPageUptodate(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
 	f2fs_put_page(ipage, 1);
+	trace_android_fs_dataread_end(inode, page_offset(page),
+				      PAGE_SIZE);
 	unlock_page(page);
 	return 0;
 }
 
 int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page)
 {
-	void *src_addr, *dst_addr;
 	struct f2fs_io_info fio = {
 		.sbi = F2FS_I_SB(dn->inode),
 		.type = DATA,
-		.rw = WRITE_SYNC | REQ_PRIO,
+		.op = REQ_OP_WRITE,
+		.op_flags = REQ_SYNC | REQ_NOIDLE | REQ_PRIO,
 		.page = page,
 		.encrypted_page = NULL,
+		.io_type = FS_DATA_IO,
 	};
 	int dirty, err;
 
-	f2fs_bug_on(F2FS_I_SB(dn->inode), page->index);
-
 	if (!f2fs_exist_data(dn->inode))
 		goto clear_out;
 
@@ -127,21 +146,9 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page)
 	if (err)
 		return err;
 
-	f2fs_wait_on_page_writeback(page, DATA);
-
-	if (PageUptodate(page))
-		goto no_update;
-
-	zero_user_segment(page, MAX_INLINE_DATA, PAGE_CACHE_SIZE);
+	f2fs_bug_on(F2FS_P_SB(page), PageWriteback(page));
 
-	/* Copy the whole inline data block */
-	src_addr = inline_data_addr(dn->inode_page);
-	dst_addr = kmap_atomic(page);
-	memcpy(dst_addr, src_addr, MAX_INLINE_DATA);
-	flush_dcache_page(page);
-	kunmap_atomic(dst_addr);
-	SetPageUptodate(page);
-no_update:
+	read_inline_data(page, dn->inode_page);
 	set_page_dirty(page);
 
 	/* clear dirty state */
@@ -149,23 +156,24 @@ no_update:
 
 	/* write data page to try to make data consistent */
 	set_page_writeback(page);
-	fio.blk_addr = dn->data_blkaddr;
+	fio.old_blkaddr = dn->data_blkaddr;
+	set_inode_flag(dn->inode, FI_HOT_DATA);
 	write_data_page(dn, &fio);
-	set_data_blkaddr(dn);
-	f2fs_update_extent_cache(dn);
-	f2fs_wait_on_page_writeback(page, DATA);
-	if (dirty)
+	f2fs_wait_on_page_writeback(page, DATA, true);
+	if (dirty) {
 		inode_dec_dirty_pages(dn->inode);
+		remove_dirty_inode(dn->inode);
+	}
 
 	/* this converted inline_data should be recovered. */
-	set_inode_flag(F2FS_I(dn->inode), FI_APPEND_WRITE);
+	set_inode_flag(dn->inode, FI_APPEND_WRITE);
 
 	/* clear inline data and flag after data writeback */
-	truncate_inline_inode(dn->inode_page, 0);
+	truncate_inline_inode(dn->inode, dn->inode_page, 0);
+	clear_inline_node(dn->inode_page);
 clear_out:
 	stat_dec_inline_inode(dn->inode);
-	f2fs_clear_inline_inode(dn->inode);
-	sync_inode_page(dn);
+	clear_inode_flag(dn->inode, FI_INLINE_DATA);
 	f2fs_put_dnode(dn);
 	return 0;
 }
@@ -177,7 +185,10 @@ int f2fs_convert_inline_inode(struct inode *inode)
 	struct page *ipage, *page;
 	int err = 0;
 
-	page = grab_cache_page(inode->i_mapping, 0);
+	if (!f2fs_has_inline_data(inode))
+		return 0;
+
+	page = f2fs_grab_cache_page(inode->i_mapping, 0, false);
 	if (!page)
 		return -ENOMEM;
 
@@ -199,6 +210,9 @@ out:
 	f2fs_unlock_op(sbi);
 
 	f2fs_put_page(page, 1);
+
+	f2fs_balance_fs(sbi, dn.node_changed);
+
 	return err;
 }
 
@@ -206,6 +220,8 @@ int f2fs_write_inline_data(struct inode *inode, struct page *page)
 {
 	void *src_addr, *dst_addr;
 	struct dnode_of_data dn;
+	struct address_space *mapping = page_mapping(page);
+	unsigned long flags;
 	int err;
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
@@ -220,16 +236,22 @@ int f2fs_write_inline_data(struct inode *inode, struct page *page)
 
 	f2fs_bug_on(F2FS_I_SB(inode), page->index);
 
-	f2fs_wait_on_page_writeback(dn.inode_page, NODE);
+	f2fs_wait_on_page_writeback(dn.inode_page, NODE, true);
 	src_addr = kmap_atomic(page);
-	dst_addr = inline_data_addr(dn.inode_page);
-	memcpy(dst_addr, src_addr, MAX_INLINE_DATA);
+	dst_addr = inline_data_addr(inode, dn.inode_page);
+	memcpy(dst_addr, src_addr, MAX_INLINE_DATA(inode));
 	kunmap_atomic(src_addr);
+	set_page_dirty(dn.inode_page);
+
+	spin_lock_irqsave(&mapping->tree_lock, flags);
+	radix_tree_tag_clear(&mapping->page_tree, page_index(page),
+			     PAGECACHE_TAG_DIRTY);
+	spin_unlock_irqrestore(&mapping->tree_lock, flags);
 
-	set_inode_flag(F2FS_I(inode), FI_APPEND_WRITE);
-	set_inode_flag(F2FS_I(inode), FI_DATA_EXIST);
+	set_inode_flag(inode, FI_APPEND_WRITE);
+	set_inode_flag(inode, FI_DATA_EXIST);
 
-	sync_inode_page(&dn);
+	clear_inline_node(dn.inode_page);
 	f2fs_put_dnode(&dn);
 	return 0;
 }
@@ -258,16 +280,16 @@ process_inline:
 		ipage = get_node_page(sbi, inode->i_ino);
 		f2fs_bug_on(sbi, IS_ERR(ipage));
 
-		f2fs_wait_on_page_writeback(ipage, NODE);
+		f2fs_wait_on_page_writeback(ipage, NODE, true);
 
-		src_addr = inline_data_addr(npage);
-		dst_addr = inline_data_addr(ipage);
-		memcpy(dst_addr, src_addr, MAX_INLINE_DATA);
+		src_addr = inline_data_addr(inode, npage);
+		dst_addr = inline_data_addr(inode, ipage);
+		memcpy(dst_addr, src_addr, MAX_INLINE_DATA(inode));
 
-		set_inode_flag(F2FS_I(inode), FI_INLINE_DATA);
-		set_inode_flag(F2FS_I(inode), FI_DATA_EXIST);
+		set_inode_flag(inode, FI_INLINE_DATA);
+		set_inode_flag(inode, FI_DATA_EXIST);
 
-		update_inode(inode, ipage);
+		set_page_dirty(ipage);
 		f2fs_put_page(ipage, 1);
 		return true;
 	}
@@ -275,10 +297,8 @@ process_inline:
 	if (f2fs_has_inline_data(inode)) {
 		ipage = get_node_page(sbi, inode->i_ino);
 		f2fs_bug_on(sbi, IS_ERR(ipage));
-		if (!truncate_inline_inode(ipage, 0))
-			return false;
-		f2fs_clear_inline_inode(inode);
-		update_inode(inode, ipage);
+		truncate_inline_inode(inode, ipage, 0);
+		clear_inode_flag(inode, FI_INLINE_DATA);
 		f2fs_put_page(ipage, 1);
 	} else if (ri && (ri->i_inline & F2FS_INLINE_DATA)) {
 		if (truncate_blocks(inode, 0, false))
@@ -289,25 +309,27 @@ process_inline:
 }
 
 struct f2fs_dir_entry *find_in_inline_dir(struct inode *dir,
-			struct f2fs_filename *fname, struct page **res_page)
+			struct fscrypt_name *fname, struct page **res_page)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(dir->i_sb);
-	struct f2fs_inline_dentry *inline_dentry;
 	struct qstr name = FSTR_TO_QSTR(&fname->disk_name);
 	struct f2fs_dir_entry *de;
 	struct f2fs_dentry_ptr d;
 	struct page *ipage;
+	void *inline_dentry;
 	f2fs_hash_t namehash;
 
 	ipage = get_node_page(sbi, dir->i_ino);
-	if (IS_ERR(ipage))
+	if (IS_ERR(ipage)) {
+		*res_page = ipage;
 		return NULL;
+	}
 
 	namehash = f2fs_dentry_hash(&name, fname);
 
-	inline_dentry = inline_data_addr(ipage);
+	inline_dentry = inline_data_addr(dir, ipage);
 
-	make_dentry_ptr(NULL, &d, (void *)inline_dentry, 2);
+	make_dentry_ptr_inline(dir, &d, inline_dentry);
 	de = find_target_dentry(fname, namehash, NULL, &d);
 	unlock_page(ipage);
 	if (de)
@@ -315,51 +337,25 @@ struct f2fs_dir_entry *find_in_inline_dir(struct inode *dir,
 	else
 		f2fs_put_page(ipage, 0);
 
-	/*
-	 * For the most part, it should be a bug when name_len is zero.
-	 * We stop here for figuring out where the bugs has occurred.
-	 */
-	f2fs_bug_on(sbi, d.max < 0);
-	return de;
-}
-
-struct f2fs_dir_entry *f2fs_parent_inline_dir(struct inode *dir,
-							struct page **p)
-{
-	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
-	struct page *ipage;
-	struct f2fs_dir_entry *de;
-	struct f2fs_inline_dentry *dentry_blk;
-
-	ipage = get_node_page(sbi, dir->i_ino);
-	if (IS_ERR(ipage))
-		return NULL;
-
-	dentry_blk = inline_data_addr(ipage);
-	de = &dentry_blk->dentry[1];
-	*p = ipage;
-	unlock_page(ipage);
 	return de;
 }
 
 int make_empty_inline_dir(struct inode *inode, struct inode *parent,
 							struct page *ipage)
 {
-	struct f2fs_inline_dentry *dentry_blk;
 	struct f2fs_dentry_ptr d;
+	void *inline_dentry;
 
-	dentry_blk = inline_data_addr(ipage);
+	inline_dentry = inline_data_addr(inode, ipage);
 
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 2);
+	make_dentry_ptr_inline(inode, &d, inline_dentry);
 	do_make_empty_dir(inode, parent, &d);
 
 	set_page_dirty(ipage);
 
 	/* update i_size to MAX_INLINE_DATA */
-	if (i_size_read(inode) < MAX_INLINE_DATA) {
-		i_size_write(inode, MAX_INLINE_DATA);
-		set_inode_flag(F2FS_I(inode), FI_UPDATE_DIR);
-	}
+	if (i_size_read(inode) < MAX_INLINE_DATA(inode))
+		f2fs_i_size_write(inode, MAX_INLINE_DATA(inode));
 	return 0;
 }
 
@@ -367,15 +363,16 @@ int make_empty_inline_dir(struct inode *inode, struct inode *parent,
  * NOTE: ipage is grabbed by caller, but if any error occurs, we should
  * release ipage in this function.
  */
-static int f2fs_convert_inline_dir(struct inode *dir, struct page *ipage,
-				struct f2fs_inline_dentry *inline_dentry)
+static int f2fs_move_inline_dirents(struct inode *dir, struct page *ipage,
+							void *inline_dentry)
 {
 	struct page *page;
 	struct dnode_of_data dn;
 	struct f2fs_dentry_block *dentry_blk;
+	struct f2fs_dentry_ptr src, dst;
 	int err;
 
-	page = grab_cache_page(dir->i_mapping, 0);
+	page = f2fs_grab_cache_page(dir->i_mapping, 0, false);
 	if (!page) {
 		f2fs_put_page(ipage, 1);
 		return -ENOMEM;
@@ -386,59 +383,152 @@ static int f2fs_convert_inline_dir(struct inode *dir, struct page *ipage,
 	if (err)
 		goto out;
 
-	f2fs_wait_on_page_writeback(page, DATA);
-	zero_user_segment(page, MAX_INLINE_DATA, PAGE_CACHE_SIZE);
+	f2fs_wait_on_page_writeback(page, DATA, true);
+	zero_user_segment(page, MAX_INLINE_DATA(dir), PAGE_SIZE);
 
 	dentry_blk = kmap_atomic(page);
 
+	make_dentry_ptr_inline(dir, &src, inline_dentry);
+	make_dentry_ptr_block(dir, &dst, dentry_blk);
+
 	/* copy data from inline dentry block to new dentry block */
-	memcpy(dentry_blk->dentry_bitmap, inline_dentry->dentry_bitmap,
-					INLINE_DENTRY_BITMAP_SIZE);
-	memset(dentry_blk->dentry_bitmap + INLINE_DENTRY_BITMAP_SIZE, 0,
-			SIZE_OF_DENTRY_BITMAP - INLINE_DENTRY_BITMAP_SIZE);
+	memcpy(dst.bitmap, src.bitmap, src.nr_bitmap);
+	memset(dst.bitmap + src.nr_bitmap, 0, dst.nr_bitmap - src.nr_bitmap);
 	/*
 	 * we do not need to zero out remainder part of dentry and filename
 	 * field, since we have used bitmap for marking the usage status of
 	 * them, besides, we can also ignore copying/zeroing reserved space
 	 * of dentry block, because them haven't been used so far.
 	 */
-	memcpy(dentry_blk->dentry, inline_dentry->dentry,
-			sizeof(struct f2fs_dir_entry) * NR_INLINE_DENTRY);
-	memcpy(dentry_blk->filename, inline_dentry->filename,
-					NR_INLINE_DENTRY * F2FS_SLOT_LEN);
+	memcpy(dst.dentry, src.dentry, SIZE_OF_DIR_ENTRY * src.max);
+	memcpy(dst.filename, src.filename, src.max * F2FS_SLOT_LEN);
 
 	kunmap_atomic(dentry_blk);
-	SetPageUptodate(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
 	set_page_dirty(page);
 
 	/* clear inline dir and flag after data writeback */
-	truncate_inline_inode(ipage, 0);
+	truncate_inline_inode(dir, ipage, 0);
 
 	stat_dec_inline_dir(dir);
-	clear_inode_flag(F2FS_I(dir), FI_INLINE_DENTRY);
+	clear_inode_flag(dir, FI_INLINE_DENTRY);
 
-	if (i_size_read(dir) < PAGE_CACHE_SIZE) {
-		i_size_write(dir, PAGE_CACHE_SIZE);
-		set_inode_flag(F2FS_I(dir), FI_UPDATE_DIR);
-	}
-
-	sync_inode_page(&dn);
+	f2fs_i_depth_write(dir, 1);
+	if (i_size_read(dir) < PAGE_SIZE)
+		f2fs_i_size_write(dir, PAGE_SIZE);
 out:
 	f2fs_put_page(page, 1);
 	return err;
 }
 
-int f2fs_add_inline_entry(struct inode *dir, const struct qstr *name,
-			struct inode *inode, nid_t ino, umode_t mode)
+static int f2fs_add_inline_entries(struct inode *dir, void *inline_dentry)
+{
+	struct f2fs_dentry_ptr d;
+	unsigned long bit_pos = 0;
+	int err = 0;
+
+	make_dentry_ptr_inline(dir, &d, inline_dentry);
+
+	while (bit_pos < d.max) {
+		struct f2fs_dir_entry *de;
+		struct qstr new_name;
+		nid_t ino;
+		umode_t fake_mode;
+
+		if (!test_bit_le(bit_pos, d.bitmap)) {
+			bit_pos++;
+			continue;
+		}
+
+		de = &d.dentry[bit_pos];
+
+		if (unlikely(!de->name_len)) {
+			bit_pos++;
+			continue;
+		}
+
+		new_name.name = d.filename[bit_pos];
+		new_name.len = le16_to_cpu(de->name_len);
+
+		ino = le32_to_cpu(de->ino);
+		fake_mode = get_de_type(de) << S_SHIFT;
+
+		err = f2fs_add_regular_entry(dir, &new_name, NULL, NULL,
+							ino, fake_mode);
+		if (err)
+			goto punch_dentry_pages;
+
+		bit_pos += GET_DENTRY_SLOTS(le16_to_cpu(de->name_len));
+	}
+	return 0;
+punch_dentry_pages:
+	truncate_inode_pages(&dir->i_data, 0);
+	truncate_blocks(dir, 0, false);
+	remove_dirty_inode(dir);
+	return err;
+}
+
+static int f2fs_move_rehashed_dirents(struct inode *dir, struct page *ipage,
+							void *inline_dentry)
+{
+	void *backup_dentry;
+	int err;
+
+	backup_dentry = f2fs_kmalloc(F2FS_I_SB(dir),
+				MAX_INLINE_DATA(dir), GFP_F2FS_ZERO);
+	if (!backup_dentry) {
+		f2fs_put_page(ipage, 1);
+		return -ENOMEM;
+	}
+
+	memcpy(backup_dentry, inline_dentry, MAX_INLINE_DATA(dir));
+	truncate_inline_inode(dir, ipage, 0);
+
+	unlock_page(ipage);
+
+	err = f2fs_add_inline_entries(dir, backup_dentry);
+	if (err)
+		goto recover;
+
+	lock_page(ipage);
+
+	stat_dec_inline_dir(dir);
+	clear_inode_flag(dir, FI_INLINE_DENTRY);
+	kfree(backup_dentry);
+	return 0;
+recover:
+	lock_page(ipage);
+	memcpy(inline_dentry, backup_dentry, MAX_INLINE_DATA(dir));
+	f2fs_i_depth_write(dir, 0);
+	f2fs_i_size_write(dir, MAX_INLINE_DATA(dir));
+	set_page_dirty(ipage);
+	f2fs_put_page(ipage, 1);
+
+	kfree(backup_dentry);
+	return err;
+}
+
+static int f2fs_convert_inline_dir(struct inode *dir, struct page *ipage,
+							void *inline_dentry)
+{
+	if (!F2FS_I(dir)->i_dir_level)
+		return f2fs_move_inline_dirents(dir, ipage, inline_dentry);
+	else
+		return f2fs_move_rehashed_dirents(dir, ipage, inline_dentry);
+}
+
+int f2fs_add_inline_entry(struct inode *dir, const struct qstr *new_name,
+				const struct qstr *orig_name,
+				struct inode *inode, nid_t ino, umode_t mode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
 	struct page *ipage;
 	unsigned int bit_pos;
 	f2fs_hash_t name_hash;
-	size_t namelen = name->len;
-	struct f2fs_inline_dentry *dentry_blk = NULL;
+	void *inline_dentry = NULL;
 	struct f2fs_dentry_ptr d;
-	int slots = GET_DENTRY_SLOTS(namelen);
+	int slots = GET_DENTRY_SLOTS(new_name->len);
 	struct page *page = NULL;
 	int err = 0;
 
@@ -446,11 +536,12 @@ int f2fs_add_inline_entry(struct inode *dir, const struct qstr *name,
 	if (IS_ERR(ipage))
 		return PTR_ERR(ipage);
 
-	dentry_blk = inline_data_addr(ipage);
-	bit_pos = room_for_filename(&dentry_blk->dentry_bitmap,
-						slots, NR_INLINE_DENTRY);
-	if (bit_pos >= NR_INLINE_DENTRY) {
-		err = f2fs_convert_inline_dir(dir, ipage, dentry_blk);
+	inline_dentry = inline_data_addr(dir, ipage);
+	make_dentry_ptr_inline(dir, &d, inline_dentry);
+
+	bit_pos = room_for_filename(d.bitmap, slots, d.max);
+	if (bit_pos >= d.max) {
+		err = f2fs_convert_inline_dir(dir, ipage, inline_dentry);
 		if (err)
 			return err;
 		err = -EAGAIN;
@@ -459,25 +550,24 @@ int f2fs_add_inline_entry(struct inode *dir, const struct qstr *name,
 
 	if (inode) {
 		down_write(&F2FS_I(inode)->i_sem);
-		page = init_inode_metadata(inode, dir, name, ipage);
+		page = init_inode_metadata(inode, dir, new_name,
+						orig_name, ipage);
 		if (IS_ERR(page)) {
 			err = PTR_ERR(page);
 			goto fail;
 		}
 	}
 
-	f2fs_wait_on_page_writeback(ipage, NODE);
+	f2fs_wait_on_page_writeback(ipage, NODE, true);
 
-	name_hash = f2fs_dentry_hash(name, NULL);
-	make_dentry_ptr(NULL, &d, (void *)dentry_blk, 2);
-	f2fs_update_dentry(ino, mode, &d, name, name_hash, bit_pos);
+	name_hash = f2fs_dentry_hash(new_name, NULL);
+	f2fs_update_dentry(ino, mode, &d, new_name, name_hash, bit_pos);
 
 	set_page_dirty(ipage);
 
 	/* we don't need to mark_inode_dirty now */
 	if (inode) {
-		F2FS_I(inode)->i_pino = dir->i_ino;
-		update_inode(inode, page);
+		f2fs_i_pino_write(inode, dir->i_ino);
 		f2fs_put_page(page, 1);
 	}
 
@@ -485,11 +575,6 @@ int f2fs_add_inline_entry(struct inode *dir, const struct qstr *name,
 fail:
 	if (inode)
 		up_write(&F2FS_I(inode)->i_sem);
-
-	if (is_inode_flag_set(F2FS_I(dir), FI_UPDATE_DIR)) {
-		update_inode(dir, ipage);
-		clear_inode_flag(F2FS_I(dir), FI_UPDATE_DIR);
-	}
 out:
 	f2fs_put_page(ipage, 1);
 	return err;
@@ -498,28 +583,30 @@ out:
 void f2fs_delete_inline_entry(struct f2fs_dir_entry *dentry, struct page *page,
 					struct inode *dir, struct inode *inode)
 {
-	struct f2fs_inline_dentry *inline_dentry;
+	struct f2fs_dentry_ptr d;
+	void *inline_dentry;
 	int slots = GET_DENTRY_SLOTS(le16_to_cpu(dentry->name_len));
 	unsigned int bit_pos;
 	int i;
 
 	lock_page(page);
-	f2fs_wait_on_page_writeback(page, NODE);
+	f2fs_wait_on_page_writeback(page, NODE, true);
 
-	inline_dentry = inline_data_addr(page);
-	bit_pos = dentry - inline_dentry->dentry;
+	inline_dentry = inline_data_addr(dir, page);
+	make_dentry_ptr_inline(dir, &d, inline_dentry);
+
+	bit_pos = dentry - d.dentry;
 	for (i = 0; i < slots; i++)
-		test_and_clear_bit_le(bit_pos + i,
-				&inline_dentry->dentry_bitmap);
+		__clear_bit_le(bit_pos + i, d.bitmap);
 
 	set_page_dirty(page);
+	f2fs_put_page(page, 1);
 
-	dir->i_ctime = dir->i_mtime = CURRENT_TIME;
+	dir->i_ctime = dir->i_mtime = current_time(dir);
+	f2fs_mark_inode_dirty_sync(dir, false);
 
 	if (inode)
-		f2fs_drop_nlink(dir, inode, page);
-
-	f2fs_put_page(page, 1);
+		f2fs_drop_nlink(dir, inode);
 }
 
 bool f2fs_empty_inline_dir(struct inode *dir)
@@ -527,49 +614,54 @@ bool f2fs_empty_inline_dir(struct inode *dir)
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
 	struct page *ipage;
 	unsigned int bit_pos = 2;
-	struct f2fs_inline_dentry *dentry_blk;
+	void *inline_dentry;
+	struct f2fs_dentry_ptr d;
 
 	ipage = get_node_page(sbi, dir->i_ino);
 	if (IS_ERR(ipage))
 		return false;
 
-	dentry_blk = inline_data_addr(ipage);
-	bit_pos = find_next_bit_le(&dentry_blk->dentry_bitmap,
-					NR_INLINE_DENTRY,
-					bit_pos);
+	inline_dentry = inline_data_addr(dir, ipage);
+	make_dentry_ptr_inline(dir, &d, inline_dentry);
+
+	bit_pos = find_next_bit_le(d.bitmap, d.max, bit_pos);
 
 	f2fs_put_page(ipage, 1);
 
-	if (bit_pos < NR_INLINE_DENTRY)
+	if (bit_pos < d.max)
 		return false;
 
 	return true;
 }
 
 int f2fs_read_inline_dir(struct file *file, struct dir_context *ctx,
-				struct f2fs_str *fstr)
+				struct fscrypt_str *fstr)
 {
 	struct inode *inode = file_inode(file);
-	struct f2fs_inline_dentry *inline_dentry = NULL;
 	struct page *ipage = NULL;
 	struct f2fs_dentry_ptr d;
+	void *inline_dentry = NULL;
+	int err;
+
+	make_dentry_ptr_inline(inode, &d, inline_dentry);
 
-	if (ctx->pos == NR_INLINE_DENTRY)
+	if (ctx->pos == d.max)
 		return 0;
 
 	ipage = get_node_page(F2FS_I_SB(inode), inode->i_ino);
 	if (IS_ERR(ipage))
 		return PTR_ERR(ipage);
 
-	inline_dentry = inline_data_addr(ipage);
+	inline_dentry = inline_data_addr(inode, ipage);
 
-	make_dentry_ptr(inode, &d, (void *)inline_dentry, 2);
+	make_dentry_ptr_inline(inode, &d, inline_dentry);
 
-	if (!f2fs_fill_dentries(ctx, &d, 0, fstr))
-		ctx->pos = NR_INLINE_DENTRY;
+	err = f2fs_fill_dentries(ctx, &d, 0, fstr);
+	if (!err)
+		ctx->pos = d.max;
 
 	f2fs_put_page(ipage, 1);
-	return 0;
+	return err < 0 ? err : 0;
 }
 
 int f2fs_inline_data_fiemap(struct inode *inode,
@@ -591,7 +683,7 @@ int f2fs_inline_data_fiemap(struct inode *inode,
 		goto out;
 	}
 
-	ilen = min_t(size_t, MAX_INLINE_DATA, i_size_read(inode));
+	ilen = min_t(size_t, MAX_INLINE_DATA(inode), i_size_read(inode));
 	if (start >= ilen)
 		goto out;
 	if (start + len < ilen)
@@ -600,7 +692,8 @@ int f2fs_inline_data_fiemap(struct inode *inode,
 
 	get_node_info(F2FS_I_SB(inode), inode->i_ino, &ni);
 	byteaddr = (__u64)ni.blk_addr << inode->i_sb->s_blocksize_bits;
-	byteaddr += (char *)inline_data_addr(ipage) - (char *)F2FS_INODE(ipage);
+	byteaddr += (char *)inline_data_addr(inode, ipage) -
+					(char *)F2FS_INODE(ipage);
 	err = fiemap_fill_next_extent(fieinfo, start, byteaddr, ilen, flags);
 out:
 	f2fs_put_page(ipage, 1);
diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c
index 97e20decacb4..50c88e37ed66 100644
--- a/fs/f2fs/inode.c
+++ b/fs/f2fs/inode.c
@@ -11,13 +11,23 @@
 #include <linux/fs.h>
 #include <linux/f2fs_fs.h>
 #include <linux/buffer_head.h>
+#include <linux/backing-dev.h>
 #include <linux/writeback.h>
 
 #include "f2fs.h"
 #include "node.h"
+#include "segment.h"
 
 #include <trace/events/f2fs.h>
 
+void f2fs_mark_inode_dirty_sync(struct inode *inode, bool sync)
+{
+	if (f2fs_inode_dirtied(inode, sync))
+		return;
+
+	mark_inode_dirty_sync(inode);
+}
+
 void f2fs_set_inode_flags(struct inode *inode)
 {
 	unsigned int flags = F2FS_I(inode)->i_flags;
@@ -39,20 +49,22 @@ void f2fs_set_inode_flags(struct inode *inode)
 
 static void __get_inode_rdev(struct inode *inode, struct f2fs_inode *ri)
 {
+	int extra_size = get_extra_isize(inode);
+
 	if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode) ||
 			S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) {
-		if (ri->i_addr[0])
-			inode->i_rdev =
-				old_decode_dev(le32_to_cpu(ri->i_addr[0]));
+		if (ri->i_addr[extra_size])
+			inode->i_rdev = old_decode_dev(
+				le32_to_cpu(ri->i_addr[extra_size]));
 		else
-			inode->i_rdev =
-				new_decode_dev(le32_to_cpu(ri->i_addr[1]));
+			inode->i_rdev = new_decode_dev(
+				le32_to_cpu(ri->i_addr[extra_size + 1]));
 	}
 }
 
 static bool __written_first_block(struct f2fs_inode *ri)
 {
-	block_t addr = le32_to_cpu(ri->i_addr[0]);
+	block_t addr = le32_to_cpu(ri->i_addr[offset_in_addr(ri)]);
 
 	if (addr != NEW_ADDR && addr != NULL_ADDR)
 		return true;
@@ -61,32 +73,34 @@ static bool __written_first_block(struct f2fs_inode *ri)
 
 static void __set_inode_rdev(struct inode *inode, struct f2fs_inode *ri)
 {
+	int extra_size = get_extra_isize(inode);
+
 	if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode)) {
 		if (old_valid_dev(inode->i_rdev)) {
-			ri->i_addr[0] =
+			ri->i_addr[extra_size] =
 				cpu_to_le32(old_encode_dev(inode->i_rdev));
-			ri->i_addr[1] = 0;
+			ri->i_addr[extra_size + 1] = 0;
 		} else {
-			ri->i_addr[0] = 0;
-			ri->i_addr[1] =
+			ri->i_addr[extra_size] = 0;
+			ri->i_addr[extra_size + 1] =
 				cpu_to_le32(new_encode_dev(inode->i_rdev));
-			ri->i_addr[2] = 0;
+			ri->i_addr[extra_size + 2] = 0;
 		}
 	}
 }
 
 static void __recover_inline_status(struct inode *inode, struct page *ipage)
 {
-	void *inline_data = inline_data_addr(ipage);
+	void *inline_data = inline_data_addr(inode, ipage);
 	__le32 *start = inline_data;
-	__le32 *end = start + MAX_INLINE_DATA / sizeof(__le32);
+	__le32 *end = start + MAX_INLINE_DATA(inode) / sizeof(__le32);
 
 	while (start < end) {
 		if (*start++) {
-			f2fs_wait_on_page_writeback(ipage, NODE);
+			f2fs_wait_on_page_writeback(ipage, NODE, true);
 
-			set_inode_flag(F2FS_I(inode), FI_DATA_EXIST);
-			set_raw_inline(F2FS_I(inode), F2FS_INODE(ipage));
+			set_inode_flag(inode, FI_DATA_EXIST);
+			set_raw_inline(inode, F2FS_INODE(ipage));
 			set_page_dirty(ipage);
 			return;
 		}
@@ -94,12 +108,84 @@ static void __recover_inline_status(struct inode *inode, struct page *ipage)
 	return;
 }
 
+static bool f2fs_enable_inode_chksum(struct f2fs_sb_info *sbi, struct page *page)
+{
+	struct f2fs_inode *ri = &F2FS_NODE(page)->i;
+	int extra_isize = le32_to_cpu(ri->i_extra_isize);
+
+	if (!f2fs_sb_has_inode_chksum(sbi->sb))
+		return false;
+
+	if (!RAW_IS_INODE(F2FS_NODE(page)) || !(ri->i_inline & F2FS_EXTRA_ATTR))
+		return false;
+
+	if (!F2FS_FITS_IN_INODE(ri, extra_isize, i_inode_checksum))
+		return false;
+
+	return true;
+}
+
+static __u32 f2fs_inode_chksum(struct f2fs_sb_info *sbi, struct page *page)
+{
+	struct f2fs_node *node = F2FS_NODE(page);
+	struct f2fs_inode *ri = &node->i;
+	__le32 ino = node->footer.ino;
+	__le32 gen = ri->i_generation;
+	__u32 chksum, chksum_seed;
+	__u32 dummy_cs = 0;
+	unsigned int offset = offsetof(struct f2fs_inode, i_inode_checksum);
+	unsigned int cs_size = sizeof(dummy_cs);
+
+	chksum = f2fs_chksum(sbi, sbi->s_chksum_seed, (__u8 *)&ino,
+							sizeof(ino));
+	chksum_seed = f2fs_chksum(sbi, chksum, (__u8 *)&gen, sizeof(gen));
+
+	chksum = f2fs_chksum(sbi, chksum_seed, (__u8 *)ri, offset);
+	chksum = f2fs_chksum(sbi, chksum, (__u8 *)&dummy_cs, cs_size);
+	offset += cs_size;
+	chksum = f2fs_chksum(sbi, chksum, (__u8 *)ri + offset,
+						F2FS_BLKSIZE - offset);
+	return chksum;
+}
+
+bool f2fs_inode_chksum_verify(struct f2fs_sb_info *sbi, struct page *page)
+{
+	struct f2fs_inode *ri;
+	__u32 provided, calculated;
+
+	if (!f2fs_enable_inode_chksum(sbi, page) ||
+			PageDirty(page) || PageWriteback(page))
+		return true;
+
+	ri = &F2FS_NODE(page)->i;
+	provided = le32_to_cpu(ri->i_inode_checksum);
+	calculated = f2fs_inode_chksum(sbi, page);
+
+	if (provided != calculated)
+		f2fs_msg(sbi->sb, KERN_WARNING,
+			"checksum invalid, ino = %x, %x vs. %x",
+			ino_of_node(page), provided, calculated);
+
+	return provided == calculated;
+}
+
+void f2fs_inode_chksum_set(struct f2fs_sb_info *sbi, struct page *page)
+{
+	struct f2fs_inode *ri = &F2FS_NODE(page)->i;
+
+	if (!f2fs_enable_inode_chksum(sbi, page))
+		return;
+
+	ri->i_inode_checksum = cpu_to_le32(f2fs_inode_chksum(sbi, page));
+}
+
 static int do_read_inode(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct f2fs_inode_info *fi = F2FS_I(inode);
 	struct page *node_page;
 	struct f2fs_inode *ri;
+	projid_t i_projid;
 
 	/* Check if ino is within scope */
 	if (check_nid_range(sbi, inode->i_ino)) {
@@ -120,7 +206,7 @@ static int do_read_inode(struct inode *inode)
 	i_gid_write(inode, le32_to_cpu(ri->i_gid));
 	set_nlink(inode, le32_to_cpu(ri->i_links));
 	inode->i_size = le64_to_cpu(ri->i_size);
-	inode->i_blocks = le64_to_cpu(ri->i_blocks);
+	inode->i_blocks = SECTOR_FROM_BLOCK(le64_to_cpu(ri->i_blocks) - 1);
 
 	inode->i_atime.tv_sec = le64_to_cpu(ri->i_atime);
 	inode->i_ctime.tv_sec = le64_to_cpu(ri->i_ctime);
@@ -138,9 +224,13 @@ static int do_read_inode(struct inode *inode)
 	fi->i_pino = le32_to_cpu(ri->i_pino);
 	fi->i_dir_level = ri->i_dir_level;
 
-	f2fs_init_extent_tree(inode, &ri->i_ext);
+	if (f2fs_init_extent_tree(inode, &ri->i_ext))
+		set_page_dirty(node_page);
+
+	get_inline_info(inode, ri);
 
-	get_inline_info(fi, ri);
+	fi->i_extra_isize = f2fs_has_extra_attr(inode) ?
+					le16_to_cpu(ri->i_extra_isize) : 0;
 
 	/* check data exist */
 	if (f2fs_has_inline_data(inode) && !f2fs_exist_data(inode))
@@ -150,7 +240,20 @@ static int do_read_inode(struct inode *inode)
 	__get_inode_rdev(inode, ri);
 
 	if (__written_first_block(ri))
-		set_inode_flag(F2FS_I(inode), FI_FIRST_BLOCK_WRITTEN);
+		set_inode_flag(inode, FI_FIRST_BLOCK_WRITTEN);
+
+	if (!need_inode_block_update(sbi, inode->i_ino))
+		fi->last_disk_size = inode->i_size;
+
+	if (fi->i_flags & FS_PROJINHERIT_FL)
+		set_inode_flag(inode, FI_PROJ_INHERIT);
+
+	if (f2fs_has_extra_attr(inode) && f2fs_sb_has_project_quota(sbi->sb) &&
+			F2FS_FITS_IN_INODE(ri, fi->i_extra_isize, i_projid))
+		i_projid = (projid_t)le32_to_cpu(ri->i_projid);
+	else
+		i_projid = F2FS_DEF_PROJID;
+	fi->i_projid = make_kprojid(&init_user_ns, i_projid);
 
 	f2fs_put_page(node_page, 1);
 
@@ -202,6 +305,7 @@ make_now:
 			inode->i_op = &f2fs_encrypted_symlink_inode_operations;
 		else
 			inode->i_op = &f2fs_symlink_inode_operations;
+		inode_nohighmem(inode);
 		inode->i_mapping->a_ops = &f2fs_dblock_aops;
 	} else if (S_ISCHR(inode->i_mode) || S_ISBLK(inode->i_mode) ||
 			S_ISFIFO(inode->i_mode) || S_ISSOCK(inode->i_mode)) {
@@ -211,6 +315,7 @@ make_now:
 		ret = -EIO;
 		goto bad_inode;
 	}
+	f2fs_set_inode_flags(inode);
 	unlock_new_inode(inode);
 	trace_f2fs_iget(inode);
 	return inode;
@@ -221,11 +326,28 @@ bad_inode:
 	return ERR_PTR(ret);
 }
 
-void update_inode(struct inode *inode, struct page *node_page)
+struct inode *f2fs_iget_retry(struct super_block *sb, unsigned long ino)
+{
+	struct inode *inode;
+retry:
+	inode = f2fs_iget(sb, ino);
+	if (IS_ERR(inode)) {
+		if (PTR_ERR(inode) == -ENOMEM) {
+			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			goto retry;
+		}
+	}
+	return inode;
+}
+
+int update_inode(struct inode *inode, struct page *node_page)
 {
 	struct f2fs_inode *ri;
+	struct extent_tree *et = F2FS_I(inode)->extent_tree;
+
+	f2fs_inode_synced(inode);
 
-	f2fs_wait_on_page_writeback(node_page, NODE);
+	f2fs_wait_on_page_writeback(node_page, NODE, true);
 
 	ri = F2FS_INODE(node_page);
 
@@ -235,14 +357,16 @@ void update_inode(struct inode *inode, struct page *node_page)
 	ri->i_gid = cpu_to_le32(i_gid_read(inode));
 	ri->i_links = cpu_to_le32(inode->i_nlink);
 	ri->i_size = cpu_to_le64(i_size_read(inode));
-	ri->i_blocks = cpu_to_le64(inode->i_blocks);
+	ri->i_blocks = cpu_to_le64(SECTOR_TO_BLOCK(inode->i_blocks) + 1);
 
-	if (F2FS_I(inode)->extent_tree)
-		set_raw_extent(&F2FS_I(inode)->extent_tree->largest,
-							&ri->i_ext);
-	else
+	if (et) {
+		read_lock(&et->lock);
+		set_raw_extent(&et->largest, &ri->i_ext);
+		read_unlock(&et->lock);
+	} else {
 		memset(&ri->i_ext, 0, sizeof(ri->i_ext));
-	set_raw_inline(F2FS_I(inode), ri);
+	}
+	set_raw_inline(inode, ri);
 
 	ri->i_atime = cpu_to_le64(inode->i_atime.tv_sec);
 	ri->i_ctime = cpu_to_le64(inode->i_ctime.tv_sec);
@@ -257,17 +381,35 @@ void update_inode(struct inode *inode, struct page *node_page)
 	ri->i_generation = cpu_to_le32(inode->i_generation);
 	ri->i_dir_level = F2FS_I(inode)->i_dir_level;
 
+	if (f2fs_has_extra_attr(inode)) {
+		ri->i_extra_isize = cpu_to_le16(F2FS_I(inode)->i_extra_isize);
+
+		if (f2fs_sb_has_project_quota(F2FS_I_SB(inode)->sb) &&
+			F2FS_FITS_IN_INODE(ri, F2FS_I(inode)->i_extra_isize,
+								i_projid)) {
+			projid_t i_projid;
+
+			i_projid = from_kprojid(&init_user_ns,
+						F2FS_I(inode)->i_projid);
+			ri->i_projid = cpu_to_le32(i_projid);
+		}
+	}
+
 	__set_inode_rdev(inode, ri);
 	set_cold_node(inode, node_page);
-	set_page_dirty(node_page);
 
-	clear_inode_flag(F2FS_I(inode), FI_DIRTY_INODE);
+	/* deleted inode */
+	if (inode->i_nlink == 0)
+		clear_inline_node(node_page);
+
+	return set_page_dirty(node_page);
 }
 
-void update_inode_page(struct inode *inode)
+int update_inode_page(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct page *node_page;
+	int ret = 0;
 retry:
 	node_page = get_node_page(sbi, inode->i_ino);
 	if (IS_ERR(node_page)) {
@@ -276,12 +418,13 @@ retry:
 			cond_resched();
 			goto retry;
 		} else if (err != -ENOENT) {
-			f2fs_stop_checkpoint(sbi);
+			f2fs_stop_checkpoint(sbi, false);
 		}
-		return;
+		return 0;
 	}
-	update_inode(inode, node_page);
+	ret = update_inode(inode, node_page);
 	f2fs_put_page(node_page, 1);
+	return ret;
 }
 
 int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
@@ -292,7 +435,7 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 			inode->i_ino == F2FS_META_INO(sbi))
 		return 0;
 
-	if (!is_inode_flag_set(F2FS_I(inode), FI_DIRTY_INODE))
+	if (!is_inode_flag_set(inode, FI_DIRTY_INODE))
 		return 0;
 
 	/*
@@ -300,8 +443,8 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 	 * during the urgent cleaning time when runing out of free sections.
 	 */
 	update_inode_page(inode);
-
-	f2fs_balance_fs(sbi);
+	if (wbc && wbc->nr_to_write)
+		f2fs_balance_fs(sbi, true);
 	return 0;
 }
 
@@ -311,13 +454,12 @@ int f2fs_write_inode(struct inode *inode, struct writeback_control *wbc)
 void f2fs_evict_inode(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	struct f2fs_inode_info *fi = F2FS_I(inode);
-	nid_t xnid = fi->i_xattr_nid;
+	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
 	int err = 0;
 
 	/* some remained atomic pages should discarded */
 	if (f2fs_is_atomic_file(inode))
-		commit_inmem_pages(inode, true);
+		drop_inmem_pages(inode);
 
 	trace_f2fs_evict_inode(inode);
 	truncate_inode_pages_final(&inode->i_data);
@@ -327,65 +469,80 @@ void f2fs_evict_inode(struct inode *inode)
 		goto out_clear;
 
 	f2fs_bug_on(sbi, get_dirty_pages(inode));
-	remove_dirty_dir_inode(inode);
+	remove_dirty_inode(inode);
 
 	f2fs_destroy_extent_tree(inode);
 
 	if (inode->i_nlink || is_bad_inode(inode))
 		goto no_delete;
 
+	dquot_initialize(inode);
+
+	remove_ino_entry(sbi, inode->i_ino, APPEND_INO);
+	remove_ino_entry(sbi, inode->i_ino, UPDATE_INO);
+
 	sb_start_intwrite(inode->i_sb);
-	set_inode_flag(fi, FI_NO_ALLOC);
+	set_inode_flag(inode, FI_NO_ALLOC);
 	i_size_write(inode, 0);
-
+retry:
 	if (F2FS_HAS_BLOCKS(inode))
-		err = f2fs_truncate(inode, true);
+		err = f2fs_truncate(inode);
 
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_EVICT_INODE)) {
+		f2fs_show_injection_info(FAULT_EVICT_INODE);
+		err = -EIO;
+	}
+#endif
 	if (!err) {
 		f2fs_lock_op(sbi);
 		err = remove_inode_page(inode);
 		f2fs_unlock_op(sbi);
+		if (err == -ENOENT)
+			err = 0;
+	}
+
+	/* give more chances, if ENOMEM case */
+	if (err == -ENOMEM) {
+		err = 0;
+		goto retry;
 	}
 
+	if (err)
+		update_inode_page(inode);
+	dquot_free_inode(inode);
 	sb_end_intwrite(inode->i_sb);
 no_delete:
+	dquot_drop(inode);
+
 	stat_dec_inline_xattr(inode);
 	stat_dec_inline_dir(inode);
 	stat_dec_inline_inode(inode);
 
-	invalidate_mapping_pages(NODE_MAPPING(sbi), inode->i_ino, inode->i_ino);
+	if (!is_set_ckpt_flags(sbi, CP_ERROR_FLAG))
+		f2fs_bug_on(sbi, is_inode_flag_set(inode, FI_DIRTY_INODE));
+
+	/* ino == 0, if f2fs_new_inode() was failed t*/
+	if (inode->i_ino)
+		invalidate_mapping_pages(NODE_MAPPING(sbi), inode->i_ino,
+							inode->i_ino);
 	if (xnid)
 		invalidate_mapping_pages(NODE_MAPPING(sbi), xnid, xnid);
-	if (is_inode_flag_set(fi, FI_APPEND_WRITE))
-		add_dirty_inode(sbi, inode->i_ino, APPEND_INO);
-	if (is_inode_flag_set(fi, FI_UPDATE_WRITE))
-		add_dirty_inode(sbi, inode->i_ino, UPDATE_INO);
-	if (is_inode_flag_set(fi, FI_FREE_NID)) {
-		if (err && err != -ENOENT)
-			alloc_nid_done(sbi, inode->i_ino);
-		else
-			alloc_nid_failed(sbi, inode->i_ino);
-		clear_inode_flag(fi, FI_FREE_NID);
+	if (inode->i_nlink) {
+		if (is_inode_flag_set(inode, FI_APPEND_WRITE))
+			add_ino_entry(sbi, inode->i_ino, APPEND_INO);
+		if (is_inode_flag_set(inode, FI_UPDATE_WRITE))
+			add_ino_entry(sbi, inode->i_ino, UPDATE_INO);
 	}
-
-	if (err && err != -ENOENT) {
-		if (!exist_written_data(sbi, inode->i_ino, ORPHAN_INO)) {
-			/*
-			 * get here because we failed to release resource
-			 * of inode previously, reminder our user to run fsck
-			 * for fixing.
-			 */
-			set_sbi_flag(sbi, SBI_NEED_FSCK);
-			f2fs_msg(sbi->sb, KERN_WARNING,
-				"inode (ino:%lu) resource leak, run fsck "
-				"to fix this issue!", inode->i_ino);
-		}
+	if (is_inode_flag_set(inode, FI_FREE_NID)) {
+		alloc_nid_failed(sbi, inode->i_ino);
+		clear_inode_flag(inode, FI_FREE_NID);
+	} else {
+		f2fs_bug_on(sbi, err &&
+			!exist_written_data(sbi, inode->i_ino, ORPHAN_INO));
 	}
 out_clear:
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	if (fi->i_crypt_info)
-		f2fs_free_encryption_info(inode, fi->i_crypt_info);
-#endif
+	fscrypt_put_encryption_info(inode, NULL);
 	clear_inode(inode);
 }
 
@@ -393,37 +550,45 @@ out_clear:
 void handle_failed_inode(struct inode *inode)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	int err = 0;
+	struct node_info ni;
 
+	/*
+	 * clear nlink of inode in order to release resource of inode
+	 * immediately.
+	 */
 	clear_nlink(inode);
-	make_bad_inode(inode);
-	unlock_new_inode(inode);
 
-	i_size_write(inode, 0);
-	if (F2FS_HAS_BLOCKS(inode))
-		err = f2fs_truncate(inode, false);
+	/*
+	 * we must call this to avoid inode being remained as dirty, resulting
+	 * in a panic when flushing dirty inodes in gdirty_list.
+	 */
+	update_inode_page(inode);
+	f2fs_inode_synced(inode);
 
-	if (!err)
-		err = remove_inode_page(inode);
+	/* don't make bad inode, since it becomes a regular file. */
+	unlock_new_inode(inode);
 
 	/*
-	 * if we skip truncate_node in remove_inode_page bacause we failed
-	 * before, it's better to find another way to release resource of
-	 * this inode (e.g. valid block count, node block or nid). Here we
-	 * choose to add this inode to orphan list, so that we can call iput
-	 * for releasing in orphan recovery flow.
-	 *
 	 * Note: we should add inode to orphan list before f2fs_unlock_op()
 	 * so we can prevent losing this orphan when encoutering checkpoint
 	 * and following suddenly power-off.
 	 */
-	if (err && err != -ENOENT) {
-		err = acquire_orphan_inode(sbi);
-		if (!err)
-			add_orphan_inode(sbi, inode->i_ino);
+	get_node_info(sbi, inode->i_ino, &ni);
+
+	if (ni.blk_addr != NULL_ADDR) {
+		int err = acquire_orphan_inode(sbi);
+		if (err) {
+			set_sbi_flag(sbi, SBI_NEED_FSCK);
+			f2fs_msg(sbi->sb, KERN_WARNING,
+				"Too many orphan inodes, run fsck to fix.");
+		} else {
+			add_orphan_inode(inode);
+		}
+		alloc_nid_done(sbi, inode->i_ino);
+	} else {
+		set_inode_flag(inode, FI_FREE_NID);
 	}
 
-	set_inode_flag(F2FS_I(inode), FI_FREE_NID);
 	f2fs_unlock_op(sbi);
 
 	/* iput will drop the inode object */
diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c
index 2c32110f9fc0..d92b8e9064cb 100644
--- a/fs/f2fs/namei.c
+++ b/fs/f2fs/namei.c
@@ -15,6 +15,7 @@
 #include <linux/ctype.h>
 #include <linux/dcache.h>
 #include <linux/namei.h>
+#include <linux/quotaops.h>
 
 #include "f2fs.h"
 #include "node.h"
@@ -42,28 +43,53 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode)
 	}
 	f2fs_unlock_op(sbi);
 
+	nid_free = true;
+
 	inode_init_owner(inode, dir, mode);
 
 	inode->i_ino = ino;
 	inode->i_blocks = 0;
-	inode->i_mtime = inode->i_atime = inode->i_ctime = CURRENT_TIME;
+	inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode);
 	inode->i_generation = sbi->s_next_generation++;
 
 	err = insert_inode_locked(inode);
 	if (err) {
 		err = -EINVAL;
-		nid_free = true;
 		goto fail;
 	}
 
+	if (f2fs_sb_has_project_quota(sbi->sb) &&
+		(F2FS_I(dir)->i_flags & FS_PROJINHERIT_FL))
+		F2FS_I(inode)->i_projid = F2FS_I(dir)->i_projid;
+	else
+		F2FS_I(inode)->i_projid = make_kprojid(&init_user_ns,
+							F2FS_DEF_PROJID);
+
+	err = dquot_initialize(inode);
+	if (err)
+		goto fail_drop;
+
+	err = dquot_alloc_inode(inode);
+	if (err)
+		goto fail_drop;
+
 	/* If the directory encrypted, then we should encrypt the inode. */
 	if (f2fs_encrypted_inode(dir) && f2fs_may_encrypt(inode))
 		f2fs_set_encrypted_inode(inode);
 
-	if (f2fs_may_inline_data(inode))
-		set_inode_flag(F2FS_I(inode), FI_INLINE_DATA);
+	set_inode_flag(inode, FI_NEW_INODE);
+
+	if (f2fs_sb_has_extra_attr(sbi->sb)) {
+		set_inode_flag(inode, FI_EXTRA_ATTR);
+		F2FS_I(inode)->i_extra_isize = F2FS_TOTAL_EXTRA_ATTR_SIZE;
+	}
+
+	if (test_opt(sbi, INLINE_XATTR))
+		set_inode_flag(inode, FI_INLINE_XATTR);
+	if (test_opt(sbi, INLINE_DATA) && f2fs_may_inline_data(inode))
+		set_inode_flag(inode, FI_INLINE_DATA);
 	if (f2fs_may_inline_dentry(inode))
-		set_inode_flag(F2FS_I(inode), FI_INLINE_DENTRY);
+		set_inode_flag(inode, FI_INLINE_DENTRY);
 
 	f2fs_init_extent_tree(inode, NULL);
 
@@ -71,15 +97,33 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode)
 	stat_inc_inline_inode(inode);
 	stat_inc_inline_dir(inode);
 
+	F2FS_I(inode)->i_flags =
+		f2fs_mask_flags(mode, F2FS_I(dir)->i_flags & F2FS_FL_INHERITED);
+
+	if (S_ISDIR(inode->i_mode))
+		F2FS_I(inode)->i_flags |= FS_INDEX_FL;
+
+	if (F2FS_I(inode)->i_flags & FS_PROJINHERIT_FL)
+		set_inode_flag(inode, FI_PROJ_INHERIT);
+
 	trace_f2fs_new_inode(inode, 0);
-	mark_inode_dirty(inode);
 	return inode;
 
 fail:
 	trace_f2fs_new_inode(inode, err);
 	make_bad_inode(inode);
 	if (nid_free)
-		set_inode_flag(F2FS_I(inode), FI_FREE_NID);
+		set_inode_flag(inode, FI_FREE_NID);
+	iput(inode);
+	return ERR_PTR(err);
+fail_drop:
+	trace_f2fs_new_inode(inode, err);
+	dquot_drop(inode);
+	inode->i_flags |= S_NOQUOTA;
+	if (nid_free)
+		set_inode_flag(inode, FI_FREE_NID);
+	clear_nlink(inode);
+	unlock_new_inode(inode);
 	iput(inode);
 	return ERR_PTR(err);
 }
@@ -88,18 +132,23 @@ static int is_multimedia_file(const unsigned char *s, const char *sub)
 {
 	size_t slen = strlen(s);
 	size_t sublen = strlen(sub);
+	int i;
 
 	/*
 	 * filename format of multimedia file should be defined as:
-	 * "filename + '.' + extension".
+	 * "filename + '.' + extension + (optional: '.' + temp extension)".
 	 */
 	if (slen < sublen + 2)
 		return 0;
 
-	if (s[slen - sublen - 1] != '.')
-		return 0;
+	for (i = 1; i < slen - sublen; i++) {
+		if (s[i] != '.')
+			continue;
+		if (!strncasecmp(s + i + 1, sub, sublen))
+			return 1;
+	}
 
-	return !strncasecmp(s + slen - sublen, sub, sublen);
+	return 0;
 }
 
 /*
@@ -128,7 +177,9 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 	nid_t ino = 0;
 	int err;
 
-	f2fs_balance_fs(sbi);
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
 
 	inode = f2fs_new_inode(dir, mode);
 	if (IS_ERR(inode))
@@ -155,6 +206,8 @@ static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 
 	if (IS_DIRSYNC(dir))
 		f2fs_sync_fs(sbi->sb, 1);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 out:
 	handle_failed_inode(inode);
@@ -169,15 +222,24 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
 	int err;
 
 	if (f2fs_encrypted_inode(dir) &&
-		!f2fs_is_child_context_consistent_with_parent(dir, inode))
+			!fscrypt_has_permitted_context(dir, inode))
 		return -EPERM;
 
-	f2fs_balance_fs(sbi);
+	if (is_inode_flag_set(dir, FI_PROJ_INHERIT) &&
+			(!projid_eq(F2FS_I(dir)->i_projid,
+			F2FS_I(old_dentry->d_inode)->i_projid)))
+		return -EXDEV;
+
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
+
+	f2fs_balance_fs(sbi, true);
 
-	inode->i_ctime = CURRENT_TIME;
+	inode->i_ctime = current_time(inode);
 	ihold(inode);
 
-	set_inode_flag(F2FS_I(inode), FI_INC_LINK);
+	set_inode_flag(inode, FI_INC_LINK);
 	f2fs_lock_op(sbi);
 	err = f2fs_add_link(dentry, inode);
 	if (err)
@@ -190,7 +252,7 @@ static int f2fs_link(struct dentry *old_dentry, struct inode *dir,
 		f2fs_sync_fs(sbi->sb, 1);
 	return 0;
 out:
-	clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
+	clear_inode_flag(inode, FI_INC_LINK);
 	iput(inode);
 	f2fs_unlock_op(sbi);
 	return err;
@@ -199,10 +261,14 @@ out:
 struct dentry *f2fs_get_parent(struct dentry *child)
 {
 	struct qstr dotdot = QSTR_INIT("..", 2);
-	unsigned long ino = f2fs_inode_by_name(d_inode(child), &dotdot);
-	if (!ino)
+	struct page *page;
+	unsigned long ino = f2fs_inode_by_name(d_inode(child), &dotdot, &page);
+	if (!ino) {
+		if (IS_ERR(page))
+			return ERR_CAST(page);
 		return ERR_PTR(-ENOENT);
-	return d_obtain_alias(f2fs_iget(d_inode(child)->i_sb, ino));
+	}
+	return d_obtain_alias(f2fs_iget(child->d_sb, ino));
 }
 
 static int __recover_dot_dentries(struct inode *dir, nid_t pino)
@@ -214,12 +280,28 @@ static int __recover_dot_dentries(struct inode *dir, nid_t pino)
 	struct page *page;
 	int err = 0;
 
+	if (f2fs_readonly(sbi->sb)) {
+		f2fs_msg(sbi->sb, KERN_INFO,
+			"skip recovering inline_dots inode (ino:%lu, pino:%u) "
+			"in readonly mountpoint", dir->i_ino, pino);
+		return 0;
+	}
+
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
+
+	f2fs_balance_fs(sbi, true);
+
 	f2fs_lock_op(sbi);
 
 	de = f2fs_find_entry(dir, &dot, &page);
 	if (de) {
 		f2fs_dentry_kunmap(dir, page);
 		f2fs_put_page(page, 0);
+	} else if (IS_ERR(page)) {
+		err = PTR_ERR(page);
+		goto out;
 	} else {
 		err = __f2fs_add_link(dir, &dot, NULL, dir->i_ino, S_IFDIR);
 		if (err)
@@ -230,14 +312,14 @@ static int __recover_dot_dentries(struct inode *dir, nid_t pino)
 	if (de) {
 		f2fs_dentry_kunmap(dir, page);
 		f2fs_put_page(page, 0);
+	} else if (IS_ERR(page)) {
+		err = PTR_ERR(page);
 	} else {
 		err = __f2fs_add_link(dir, &dotdot, NULL, pino, S_IFDIR);
 	}
 out:
-	if (!err) {
-		clear_inode_flag(F2FS_I(dir), FI_INLINE_DOTS);
-		mark_inode_dirty(dir);
-	}
+	if (!err)
+		clear_inode_flag(dir, FI_INLINE_DOTS);
 
 	f2fs_unlock_op(sbi);
 	return err;
@@ -251,13 +333,32 @@ static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry,
 	struct page *page;
 	nid_t ino;
 	int err = 0;
+	unsigned int root_ino = F2FS_ROOT_INO(F2FS_I_SB(dir));
+
+	if (f2fs_encrypted_inode(dir)) {
+		int res = fscrypt_get_encryption_info(dir);
+
+		/*
+		 * DCACHE_ENCRYPTED_WITH_KEY is set if the dentry is
+		 * created while the directory was encrypted and we
+		 * don't have access to the key.
+		 */
+		if (fscrypt_has_encryption_key(dir))
+			fscrypt_set_encrypted_dentry(dentry);
+		fscrypt_set_d_op(dentry);
+		if (res && res != -ENOKEY)
+			return ERR_PTR(res);
+	}
 
 	if (dentry->d_name.len > F2FS_NAME_LEN)
 		return ERR_PTR(-ENAMETOOLONG);
 
 	de = f2fs_find_entry(dir, &dentry->d_name, &page);
-	if (!de)
+	if (!de) {
+		if (IS_ERR(page))
+			return (struct dentry *)page;
 		return d_splice_alias(inode, dentry);
+	}
 
 	ino = le32_to_cpu(de->ino);
 	f2fs_dentry_kunmap(dir, page);
@@ -267,15 +368,30 @@ static struct dentry *f2fs_lookup(struct inode *dir, struct dentry *dentry,
 	if (IS_ERR(inode))
 		return ERR_CAST(inode);
 
+	if ((dir->i_ino == root_ino) && f2fs_has_inline_dots(dir)) {
+		err = __recover_dot_dentries(dir, root_ino);
+		if (err)
+			goto err_out;
+	}
+
 	if (f2fs_has_inline_dots(inode)) {
 		err = __recover_dot_dentries(inode, dir->i_ino);
 		if (err)
 			goto err_out;
 	}
+	if (f2fs_encrypted_inode(dir) &&
+	    (S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) &&
+	    !fscrypt_has_permitted_context(dir, inode)) {
+		f2fs_msg(inode->i_sb, KERN_WARNING,
+			 "Inconsistent encryption contexts: %lu/%lu",
+			 dir->i_ino, inode->i_ino);
+		err = -EPERM;
+		goto err_out;
+	}
 	return d_splice_alias(inode, dentry);
 
 err_out:
-	iget_failed(inode);
+	iput(inode);
 	return ERR_PTR(err);
 }
 
@@ -288,11 +404,19 @@ static int f2fs_unlink(struct inode *dir, struct dentry *dentry)
 	int err = -ENOENT;
 
 	trace_f2fs_unlink_enter(dir, dentry);
-	f2fs_balance_fs(sbi);
+
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
 
 	de = f2fs_find_entry(dir, &dentry->d_name, &page);
-	if (!de)
+	if (!de) {
+		if (IS_ERR(page))
+			err = PTR_ERR(page);
 		goto fail;
+	}
+
+	f2fs_balance_fs(sbi, true);
 
 	f2fs_lock_op(sbi);
 	err = acquire_orphan_inode(sbi);
@@ -305,9 +429,6 @@ static int f2fs_unlink(struct inode *dir, struct dentry *dentry)
 	f2fs_delete_entry(de, page, dir, inode);
 	f2fs_unlock_op(sbi);
 
-	/* In order to evict this inode, we set it dirty */
-	mark_inode_dirty(inode);
-
 	if (IS_DIRSYNC(dir))
 		f2fs_sync_fs(sbi->sb, 1);
 fail:
@@ -332,16 +453,28 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dir);
 	struct inode *inode;
 	size_t len = strlen(symname);
-	size_t p_len;
-	char *p_str;
-	struct f2fs_str disk_link = FSTR_INIT(NULL, 0);
-	struct f2fs_encrypted_symlink_data *sd = NULL;
+	struct fscrypt_str disk_link = FSTR_INIT((char *)symname, len + 1);
+	struct fscrypt_symlink_data *sd = NULL;
 	int err;
 
-	if (len > dir->i_sb->s_blocksize)
+	if (f2fs_encrypted_inode(dir)) {
+		err = fscrypt_get_encryption_info(dir);
+		if (err)
+			return err;
+
+		if (!fscrypt_has_encryption_key(dir))
+			return -ENOKEY;
+
+		disk_link.len = (fscrypt_fname_encrypted_size(dir, len) +
+				sizeof(struct fscrypt_symlink_data));
+	}
+
+	if (disk_link.len > dir->i_sb->s_blocksize)
 		return -ENAMETOOLONG;
 
-	f2fs_balance_fs(sbi);
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
 
 	inode = f2fs_new_inode(dir, S_IFLNK | S_IRWXUGO);
 	if (IS_ERR(inode))
@@ -351,6 +484,7 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 		inode->i_op = &f2fs_encrypted_symlink_inode_operations;
 	else
 		inode->i_op = &f2fs_symlink_inode_operations;
+	inode_nohighmem(inode);
 	inode->i_mapping->a_ops = &f2fs_dblock_aops;
 
 	f2fs_lock_op(sbi);
@@ -360,42 +494,36 @@ static int f2fs_symlink(struct inode *dir, struct dentry *dentry,
 	f2fs_unlock_op(sbi);
 	alloc_nid_done(sbi, inode->i_ino);
 
-	if (f2fs_encrypted_inode(dir)) {
+	if (f2fs_encrypted_inode(inode)) {
 		struct qstr istr = QSTR_INIT(symname, len);
+		struct fscrypt_str ostr;
 
-		err = f2fs_get_encryption_info(inode);
-		if (err)
+		sd = kzalloc(disk_link.len, GFP_NOFS);
+		if (!sd) {
+			err = -ENOMEM;
 			goto err_out;
+		}
 
-		err = f2fs_fname_crypto_alloc_buffer(inode, len, &disk_link);
+		err = fscrypt_get_encryption_info(inode);
 		if (err)
 			goto err_out;
 
-		err = f2fs_fname_usr_to_disk(inode, &istr, &disk_link);
-		if (err < 0)
-			goto err_out;
-
-		p_len = encrypted_symlink_data_len(disk_link.len) + 1;
-
-		if (p_len > dir->i_sb->s_blocksize) {
-			err = -ENAMETOOLONG;
+		if (!fscrypt_has_encryption_key(inode)) {
+			err = -ENOKEY;
 			goto err_out;
 		}
 
-		sd = kzalloc(p_len, GFP_NOFS);
-		if (!sd) {
-			err = -ENOMEM;
+		ostr.name = sd->encrypted_path;
+		ostr.len = disk_link.len;
+		err = fscrypt_fname_usr_to_disk(inode, &istr, &ostr);
+		if (err)
 			goto err_out;
-		}
-		memcpy(sd->encrypted_path, disk_link.name, disk_link.len);
-		sd->len = cpu_to_le16(disk_link.len);
-		p_str = (char *)sd;
-	} else {
-		p_len = len + 1;
-		p_str = (char *)symname;
+
+		sd->len = cpu_to_le16(ostr.len);
+		disk_link.name = (char *)sd;
 	}
 
-	err = page_symlink(inode, p_str, p_len);
+	err = page_symlink(inode, disk_link.name, disk_link.len);
 
 err_out:
 	d_instantiate(dentry, inode);
@@ -411,7 +539,8 @@ err_out:
 	 * performance regression.
 	 */
 	if (!err) {
-		filemap_write_and_wait_range(inode->i_mapping, 0, p_len - 1);
+		filemap_write_and_wait_range(inode->i_mapping, 0,
+							disk_link.len - 1);
 
 		if (IS_DIRSYNC(dir))
 			f2fs_sync_fs(sbi->sb, 1);
@@ -420,7 +549,8 @@ err_out:
 	}
 
 	kfree(sd);
-	f2fs_fname_crypto_free_buffer(&disk_link);
+
+	f2fs_balance_fs(sbi, true);
 	return err;
 out:
 	handle_failed_inode(inode);
@@ -433,7 +563,9 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	struct inode *inode;
 	int err;
 
-	f2fs_balance_fs(sbi);
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
 
 	inode = f2fs_new_inode(dir, S_IFDIR | mode);
 	if (IS_ERR(inode))
@@ -444,7 +576,7 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 	inode->i_mapping->a_ops = &f2fs_dblock_aops;
 	mapping_set_gfp_mask(inode->i_mapping, GFP_F2FS_HIGH_ZERO);
 
-	set_inode_flag(F2FS_I(inode), FI_INC_LINK);
+	set_inode_flag(inode, FI_INC_LINK);
 	f2fs_lock_op(sbi);
 	err = f2fs_add_link(dentry, inode);
 	if (err)
@@ -458,10 +590,12 @@ static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 
 	if (IS_DIRSYNC(dir))
 		f2fs_sync_fs(sbi->sb, 1);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 
 out_fail:
-	clear_inode_flag(F2FS_I(inode), FI_INC_LINK);
+	clear_inode_flag(inode, FI_INC_LINK);
 	handle_failed_inode(inode);
 	return err;
 }
@@ -481,7 +615,9 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 	struct inode *inode;
 	int err = 0;
 
-	f2fs_balance_fs(sbi);
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
 
 	inode = f2fs_new_inode(dir, mode);
 	if (IS_ERR(inode))
@@ -503,6 +639,8 @@ static int f2fs_mknod(struct inode *dir, struct dentry *dentry,
 
 	if (IS_DIRSYNC(dir))
 		f2fs_sync_fs(sbi->sb, 1);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 out:
 	handle_failed_inode(inode);
@@ -516,8 +654,9 @@ static int __f2fs_tmpfile(struct inode *dir, struct dentry *dentry,
 	struct inode *inode;
 	int err;
 
-	if (!whiteout)
-		f2fs_balance_fs(sbi);
+	err = dquot_initialize(dir);
+	if (err)
+		return err;
 
 	inode = f2fs_new_inode(dir, mode);
 	if (IS_ERR(inode))
@@ -545,18 +684,20 @@ static int __f2fs_tmpfile(struct inode *dir, struct dentry *dentry,
 	 * add this non-linked tmpfile to orphan list, in this way we could
 	 * remove all unused data of tmpfile after abnormal power-off.
 	 */
-	add_orphan_inode(sbi, inode->i_ino);
-	f2fs_unlock_op(sbi);
-
+	add_orphan_inode(inode);
 	alloc_nid_done(sbi, inode->i_ino);
 
 	if (whiteout) {
-		inode_dec_link_count(inode);
+		f2fs_i_links_write(inode, false);
 		*whiteout = inode;
 	} else {
 		d_tmpfile(dentry, inode);
 	}
+	/* link_count was changed by d_tmpfile as well. */
+	f2fs_unlock_op(sbi);
 	unlock_new_inode(inode);
+
+	f2fs_balance_fs(sbi, true);
 	return 0;
 
 release_out:
@@ -569,7 +710,7 @@ out:
 static int f2fs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode)
 {
 	if (f2fs_encrypted_inode(dir)) {
-		int err = f2fs_get_encryption_info(dir);
+		int err = fscrypt_get_encryption_info(dir);
 		if (err)
 			return err;
 	}
@@ -595,26 +736,48 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	struct f2fs_dir_entry *old_dir_entry = NULL;
 	struct f2fs_dir_entry *old_entry;
 	struct f2fs_dir_entry *new_entry;
+	bool is_old_inline = f2fs_has_inline_dentry(old_dir);
 	int err = -ENOENT;
 
+	if ((f2fs_encrypted_inode(old_dir) &&
+			!fscrypt_has_encryption_key(old_dir)) ||
+			(f2fs_encrypted_inode(new_dir) &&
+			!fscrypt_has_encryption_key(new_dir)))
+		return -ENOKEY;
+
 	if ((old_dir != new_dir) && f2fs_encrypted_inode(new_dir) &&
-		!f2fs_is_child_context_consistent_with_parent(new_dir,
-							old_inode)) {
+			!fscrypt_has_permitted_context(new_dir, old_inode)) {
 		err = -EPERM;
 		goto out;
 	}
 
-	f2fs_balance_fs(sbi);
+	if (is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
+			(!projid_eq(F2FS_I(new_dir)->i_projid,
+			F2FS_I(old_dentry->d_inode)->i_projid)))
+		return -EXDEV;
+
+	err = dquot_initialize(old_dir);
+	if (err)
+		goto out;
+
+	err = dquot_initialize(new_dir);
+	if (err)
+		goto out;
 
 	old_entry = f2fs_find_entry(old_dir, &old_dentry->d_name, &old_page);
-	if (!old_entry)
+	if (!old_entry) {
+		if (IS_ERR(old_page))
+			err = PTR_ERR(old_page);
 		goto out;
+	}
 
 	if (S_ISDIR(old_inode->i_mode)) {
-		err = -EIO;
 		old_dir_entry = f2fs_parent_dir(old_inode, &old_dir_page);
-		if (!old_dir_entry)
+		if (!old_dir_entry) {
+			if (IS_ERR(old_dir_page))
+				err = PTR_ERR(old_dir_page);
 			goto out_old;
+		}
 	}
 
 	if (flags & RENAME_WHITEOUT) {
@@ -632,8 +795,13 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		err = -ENOENT;
 		new_entry = f2fs_find_entry(new_dir, &new_dentry->d_name,
 						&new_page);
-		if (!new_entry)
+		if (!new_entry) {
+			if (IS_ERR(new_page))
+				err = PTR_ERR(new_page);
 			goto out_whiteout;
+		}
+
+		f2fs_balance_fs(sbi, true);
 
 		f2fs_lock_op(sbi);
 
@@ -641,31 +809,22 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (err)
 			goto put_out_dir;
 
-		if (update_dent_inode(old_inode, new_inode,
-						&new_dentry->d_name)) {
-			release_orphan_inode(sbi);
-			goto put_out_dir;
-		}
-
 		f2fs_set_link(new_dir, new_entry, new_page, old_inode);
 
-		new_inode->i_ctime = CURRENT_TIME;
+		new_inode->i_ctime = current_time(new_inode);
 		down_write(&F2FS_I(new_inode)->i_sem);
 		if (old_dir_entry)
-			drop_nlink(new_inode);
-		drop_nlink(new_inode);
+			f2fs_i_links_write(new_inode, false);
+		f2fs_i_links_write(new_inode, false);
 		up_write(&F2FS_I(new_inode)->i_sem);
 
-		mark_inode_dirty(new_inode);
-
 		if (!new_inode->i_nlink)
-			add_orphan_inode(sbi, new_inode->i_ino);
+			add_orphan_inode(new_inode);
 		else
 			release_orphan_inode(sbi);
-
-		update_inode_page(old_inode);
-		update_inode_page(new_inode);
 	} else {
+		f2fs_balance_fs(sbi, true);
+
 		f2fs_lock_op(sbi);
 
 		err = f2fs_add_link(new_dentry, old_inode);
@@ -674,26 +833,47 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 			goto out_whiteout;
 		}
 
-		if (old_dir_entry) {
-			inc_nlink(new_dir);
-			update_inode_page(new_dir);
+		if (old_dir_entry)
+			f2fs_i_links_write(new_dir, true);
+
+		/*
+		 * old entry and new entry can locate in the same inline
+		 * dentry in inode, when attaching new entry in inline dentry,
+		 * it could force inline dentry conversion, after that,
+		 * old_entry and old_page will point to wrong address, in
+		 * order to avoid this, let's do the check and update here.
+		 */
+		if (is_old_inline && !f2fs_has_inline_dentry(old_dir)) {
+			f2fs_put_page(old_page, 0);
+			old_page = NULL;
+
+			old_entry = f2fs_find_entry(old_dir,
+						&old_dentry->d_name, &old_page);
+			if (!old_entry) {
+				err = -ENOENT;
+				if (IS_ERR(old_page))
+					err = PTR_ERR(old_page);
+				f2fs_unlock_op(sbi);
+				goto out_whiteout;
+			}
 		}
 	}
 
 	down_write(&F2FS_I(old_inode)->i_sem);
-	file_lost_pino(old_inode);
-	if (new_inode && file_enc_name(new_inode))
-		file_set_enc_name(old_inode);
+	if (!old_dir_entry || whiteout)
+		file_lost_pino(old_inode);
+	else
+		F2FS_I(old_inode)->i_pino = new_dir->i_ino;
 	up_write(&F2FS_I(old_inode)->i_sem);
 
-	old_inode->i_ctime = CURRENT_TIME;
-	mark_inode_dirty(old_inode);
+	old_inode->i_ctime = current_time(old_inode);
+	f2fs_mark_inode_dirty_sync(old_inode, false);
 
 	f2fs_delete_entry(old_entry, old_page, old_dir, NULL);
 
 	if (whiteout) {
 		whiteout->i_state |= I_LINKABLE;
-		set_inode_flag(F2FS_I(whiteout), FI_INC_LINK);
+		set_inode_flag(whiteout, FI_INC_LINK);
 		err = f2fs_add_link(old_dentry, whiteout);
 		if (err)
 			goto put_out_dir;
@@ -705,14 +885,11 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry,
 		if (old_dir != new_dir && !whiteout) {
 			f2fs_set_link(old_inode, old_dir_entry,
 						old_dir_page, new_dir);
-			update_inode_page(old_inode);
 		} else {
 			f2fs_dentry_kunmap(old_inode, old_dir_page);
 			f2fs_put_page(old_dir_page, 0);
 		}
-		drop_nlink(old_dir);
-		mark_inode_dirty(old_dir);
-		update_inode_page(old_dir);
+		f2fs_i_links_write(old_dir, false);
 	}
 
 	f2fs_unlock_op(sbi);
@@ -755,40 +932,68 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	int old_nlink = 0, new_nlink = 0;
 	int err = -ENOENT;
 
+	if ((f2fs_encrypted_inode(old_dir) &&
+			!fscrypt_has_encryption_key(old_dir)) ||
+			(f2fs_encrypted_inode(new_dir) &&
+			!fscrypt_has_encryption_key(new_dir)))
+		return -ENOKEY;
+
 	if ((f2fs_encrypted_inode(old_dir) || f2fs_encrypted_inode(new_dir)) &&
-		(old_dir != new_dir) &&
-		(!f2fs_is_child_context_consistent_with_parent(new_dir,
-								old_inode) ||
-		!f2fs_is_child_context_consistent_with_parent(old_dir,
-								new_inode)))
+			(old_dir != new_dir) &&
+			(!fscrypt_has_permitted_context(new_dir, old_inode) ||
+			 !fscrypt_has_permitted_context(old_dir, new_inode)))
 		return -EPERM;
 
-	f2fs_balance_fs(sbi);
+	if ((is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
+			!projid_eq(F2FS_I(new_dir)->i_projid,
+			F2FS_I(old_dentry->d_inode)->i_projid)) ||
+	    (is_inode_flag_set(new_dir, FI_PROJ_INHERIT) &&
+			!projid_eq(F2FS_I(old_dir)->i_projid,
+			F2FS_I(new_dentry->d_inode)->i_projid)))
+		return -EXDEV;
+
+	err = dquot_initialize(old_dir);
+	if (err)
+		goto out;
+
+	err = dquot_initialize(new_dir);
+	if (err)
+		goto out;
 
 	old_entry = f2fs_find_entry(old_dir, &old_dentry->d_name, &old_page);
-	if (!old_entry)
+	if (!old_entry) {
+		if (IS_ERR(old_page))
+			err = PTR_ERR(old_page);
 		goto out;
+	}
 
 	new_entry = f2fs_find_entry(new_dir, &new_dentry->d_name, &new_page);
-	if (!new_entry)
+	if (!new_entry) {
+		if (IS_ERR(new_page))
+			err = PTR_ERR(new_page);
 		goto out_old;
+	}
 
 	/* prepare for updating ".." directory entry info later */
 	if (old_dir != new_dir) {
 		if (S_ISDIR(old_inode->i_mode)) {
-			err = -EIO;
 			old_dir_entry = f2fs_parent_dir(old_inode,
 							&old_dir_page);
-			if (!old_dir_entry)
+			if (!old_dir_entry) {
+				if (IS_ERR(old_dir_page))
+					err = PTR_ERR(old_dir_page);
 				goto out_new;
+			}
 		}
 
 		if (S_ISDIR(new_inode->i_mode)) {
-			err = -EIO;
 			new_dir_entry = f2fs_parent_dir(new_inode,
 							&new_dir_page);
-			if (!new_dir_entry)
+			if (!new_dir_entry) {
+				if (IS_ERR(new_dir_page))
+					err = PTR_ERR(new_dir_page);
 				goto out_old_dir;
+			}
 		}
 	}
 
@@ -802,24 +1007,14 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 		old_nlink = old_dir_entry ? -1 : 1;
 		new_nlink = -old_nlink;
 		err = -EMLINK;
-		if ((old_nlink > 0 && old_inode->i_nlink >= F2FS_LINK_MAX) ||
-			(new_nlink > 0 && new_inode->i_nlink >= F2FS_LINK_MAX))
+		if ((old_nlink > 0 && old_dir->i_nlink >= F2FS_LINK_MAX) ||
+			(new_nlink > 0 && new_dir->i_nlink >= F2FS_LINK_MAX))
 			goto out_new_dir;
 	}
 
-	f2fs_lock_op(sbi);
-
-	err = update_dent_inode(old_inode, new_inode, &new_dentry->d_name);
-	if (err)
-		goto out_unlock;
-	if (file_enc_name(new_inode))
-		file_set_enc_name(old_inode);
+	f2fs_balance_fs(sbi, true);
 
-	err = update_dent_inode(new_inode, old_inode, &old_dentry->d_name);
-	if (err)
-		goto out_undo;
-	if (file_enc_name(old_inode))
-		file_set_enc_name(new_inode);
+	f2fs_lock_op(sbi);
 
 	/* update ".." directory entry info of old dentry */
 	if (old_dir_entry)
@@ -836,19 +1031,13 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	file_lost_pino(old_inode);
 	up_write(&F2FS_I(old_inode)->i_sem);
 
-	update_inode_page(old_inode);
-
-	old_dir->i_ctime = CURRENT_TIME;
+	old_dir->i_ctime = current_time(old_dir);
 	if (old_nlink) {
 		down_write(&F2FS_I(old_dir)->i_sem);
-		if (old_nlink < 0)
-			drop_nlink(old_dir);
-		else
-			inc_nlink(old_dir);
+		f2fs_i_links_write(old_dir, old_nlink > 0);
 		up_write(&F2FS_I(old_dir)->i_sem);
 	}
-	mark_inode_dirty(old_dir);
-	update_inode_page(old_dir);
+	f2fs_mark_inode_dirty_sync(old_dir, false);
 
 	/* update directory entry info of new dir inode */
 	f2fs_set_link(new_dir, new_entry, new_page, old_inode);
@@ -857,33 +1046,19 @@ static int f2fs_cross_rename(struct inode *old_dir, struct dentry *old_dentry,
 	file_lost_pino(new_inode);
 	up_write(&F2FS_I(new_inode)->i_sem);
 
-	update_inode_page(new_inode);
-
-	new_dir->i_ctime = CURRENT_TIME;
+	new_dir->i_ctime = current_time(new_dir);
 	if (new_nlink) {
 		down_write(&F2FS_I(new_dir)->i_sem);
-		if (new_nlink < 0)
-			drop_nlink(new_dir);
-		else
-			inc_nlink(new_dir);
+		f2fs_i_links_write(new_dir, new_nlink > 0);
 		up_write(&F2FS_I(new_dir)->i_sem);
 	}
-	mark_inode_dirty(new_dir);
-	update_inode_page(new_dir);
+	f2fs_mark_inode_dirty_sync(new_dir, false);
 
 	f2fs_unlock_op(sbi);
 
 	if (IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir))
 		f2fs_sync_fs(sbi->sb, 1);
 	return 0;
-out_undo:
-	/*
-	 * Still we may fail to recover name info of f2fs_inode here
-	 * Drop it, once its name is set as encrypted
-	 */
-	update_dent_inode(old_inode, old_inode, &old_dentry->d_name);
-out_unlock:
-	f2fs_unlock_op(sbi);
 out_new_dir:
 	if (new_dir_entry) {
 		f2fs_dentry_kunmap(new_inode, new_dir_page);
@@ -922,89 +1097,82 @@ static int f2fs_rename2(struct inode *old_dir, struct dentry *old_dentry,
 	return f2fs_rename(old_dir, old_dentry, new_dir, new_dentry, flags);
 }
 
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
 static const char *f2fs_encrypted_follow_link(struct dentry *dentry, void **cookie)
 {
 	struct page *cpage = NULL;
 	char *caddr, *paddr = NULL;
-	struct f2fs_str cstr;
-	struct f2fs_str pstr = FSTR_INIT(NULL, 0);
+	struct fscrypt_str cstr = FSTR_INIT(NULL, 0);
+	struct fscrypt_str pstr = FSTR_INIT(NULL, 0);
+	struct fscrypt_symlink_data *sd;
 	struct inode *inode = d_inode(dentry);
-	struct f2fs_encrypted_symlink_data *sd;
-	loff_t size = min_t(loff_t, i_size_read(inode), PAGE_SIZE - 1);
 	u32 max_size = inode->i_sb->s_blocksize;
 	int res;
 
-	res = f2fs_get_encryption_info(inode);
+	res = fscrypt_get_encryption_info(inode);
 	if (res)
 		return ERR_PTR(res);
 
 	cpage = read_mapping_page(inode->i_mapping, 0, NULL);
 	if (IS_ERR(cpage))
 		return ERR_CAST(cpage);
-	caddr = kmap(cpage);
-	caddr[size] = 0;
+	caddr = page_address(cpage);
 
 	/* Symlink is encrypted */
-	sd = (struct f2fs_encrypted_symlink_data *)caddr;
+	sd = (struct fscrypt_symlink_data *)caddr;
+	cstr.name = sd->encrypted_path;
 	cstr.len = le16_to_cpu(sd->len);
-	cstr.name = kmalloc(cstr.len, GFP_NOFS);
-	if (!cstr.name) {
-		res = -ENOMEM;
-		goto errout;
-	}
-	memcpy(cstr.name, sd->encrypted_path, cstr.len);
 
 	/* this is broken symlink case */
-	if (cstr.name[0] == 0 && cstr.len == 0) {
+	if (unlikely(cstr.len == 0)) {
 		res = -ENOENT;
 		goto errout;
 	}
 
-	if ((cstr.len + sizeof(struct f2fs_encrypted_symlink_data) - 1) >
-								max_size) {
+	if ((cstr.len + sizeof(struct fscrypt_symlink_data) - 1) > max_size) {
 		/* Symlink data on the disk is corrupted */
 		res = -EIO;
 		goto errout;
 	}
-	res = f2fs_fname_crypto_alloc_buffer(inode, cstr.len, &pstr);
+	res = fscrypt_fname_alloc_buffer(inode, cstr.len, &pstr);
 	if (res)
 		goto errout;
 
-	res = f2fs_fname_disk_to_usr(inode, NULL, &cstr, &pstr);
-	if (res < 0)
+	res = fscrypt_fname_disk_to_usr(inode, 0, 0, &cstr, &pstr);
+	if (res)
 		goto errout;
 
-	kfree(cstr.name);
+	/* this is broken symlink case */
+	if (unlikely(pstr.name[0] == 0)) {
+		res = -ENOENT;
+		goto errout;
+	}
 
 	paddr = pstr.name;
 
 	/* Null-terminate the name */
-	paddr[res] = '\0';
+	paddr[pstr.len] = '\0';
 
-	kunmap(cpage);
-	page_cache_release(cpage);
+	put_page(cpage);
 	return *cookie = paddr;
 errout:
-	kfree(cstr.name);
-	f2fs_fname_crypto_free_buffer(&pstr);
-	kunmap(cpage);
-	page_cache_release(cpage);
+	fscrypt_fname_free_buffer(&pstr);
+	put_page(cpage);
 	return ERR_PTR(res);
 }
 
 const struct inode_operations f2fs_encrypted_symlink_inode_operations = {
 	.readlink       = generic_readlink,
-	.follow_link    = f2fs_encrypted_follow_link,
-	.put_link       = kfree_put_link,
+	.follow_link	= f2fs_encrypted_follow_link,
+	.put_link	= kfree_put_link,
 	.getattr	= f2fs_getattr,
 	.setattr	= f2fs_setattr,
+#ifdef CONFIG_F2FS_FS_XATTR
 	.setxattr	= generic_setxattr,
 	.getxattr	= generic_getxattr,
 	.listxattr	= f2fs_listxattr,
 	.removexattr	= generic_removexattr,
-};
 #endif
+};
 
 const struct inode_operations f2fs_dir_inode_operations = {
 	.create		= f2fs_create,
diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c
index 7bcbc6e9c40d..32474db18ad9 100644
--- a/fs/f2fs/node.c
+++ b/fs/f2fs/node.c
@@ -19,10 +19,11 @@
 #include "f2fs.h"
 #include "node.h"
 #include "segment.h"
+#include "xattr.h"
 #include "trace.h"
 #include <trace/events/f2fs.h>
 
-#define on_build_free_nids(nmi) mutex_is_locked(&nm_i->build_lock)
+#define on_build_free_nids(nmi) mutex_is_locked(&(nm_i)->build_lock)
 
 static struct kmem_cache *nat_entry_slab;
 static struct kmem_cache *free_nid_slab;
@@ -45,13 +46,15 @@ bool available_free_memory(struct f2fs_sb_info *sbi, int type)
 	 * give 25%, 25%, 50%, 50%, 50% memory for each components respectively
 	 */
 	if (type == FREE_NIDS) {
-		mem_size = (nm_i->fcnt * sizeof(struct free_nid)) >>
-							PAGE_CACHE_SHIFT;
+		mem_size = (nm_i->nid_cnt[FREE_NID_LIST] *
+				sizeof(struct free_nid)) >> PAGE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 2);
 	} else if (type == NAT_ENTRIES) {
 		mem_size = (nm_i->nat_cnt * sizeof(struct nat_entry)) >>
-							PAGE_CACHE_SHIFT;
+							PAGE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 2);
+		if (excess_cached_nats(sbi))
+			res = false;
 	} else if (type == DIRTY_DENTS) {
 		if (sbi->sb->s_bdi->wb.dirty_exceeded)
 			return false;
@@ -61,17 +64,19 @@ bool available_free_memory(struct f2fs_sb_info *sbi, int type)
 		int i;
 
 		for (i = 0; i <= UPDATE_INO; i++)
-			mem_size += (sbi->im[i].ino_num *
-				sizeof(struct ino_entry)) >> PAGE_CACHE_SHIFT;
+			mem_size += sbi->im[i].ino_num *
+						sizeof(struct ino_entry);
+		mem_size >>= PAGE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1);
 	} else if (type == EXTENT_CACHE) {
-		mem_size = (sbi->total_ext_tree * sizeof(struct extent_tree) +
+		mem_size = (atomic_read(&sbi->total_ext_tree) *
+				sizeof(struct extent_tree) +
 				atomic_read(&sbi->total_ext_node) *
-				sizeof(struct extent_node)) >> PAGE_CACHE_SHIFT;
+				sizeof(struct extent_node)) >> PAGE_SHIFT;
 		res = mem_size < ((avail_ram * nm_i->ram_thresh / 100) >> 1);
 	} else {
-		if (sbi->sb->s_bdi->wb.dirty_exceeded)
-			return false;
+		if (!sbi->sb->s_bdi->wb.dirty_exceeded)
+			return true;
 	}
 	return res;
 }
@@ -120,7 +125,7 @@ static struct page *get_next_nat_page(struct f2fs_sb_info *sbi, nid_t nid)
 
 	src_addr = page_address(src_page);
 	dst_addr = page_address(dst_page);
-	memcpy(dst_addr, src_addr, PAGE_CACHE_SIZE);
+	memcpy(dst_addr, src_addr, PAGE_SIZE);
 	set_page_dirty(dst_page);
 	f2fs_put_page(src_page, 1);
 
@@ -154,9 +159,6 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
 	nid_t set = NAT_BLOCK_OFFSET(ne->ni.nid);
 	struct nat_entry_set *head;
 
-	if (get_nat_flag(ne, IS_DIRTY))
-		return;
-
 	head = radix_tree_lookup(&nm_i->nat_set_root, set);
 	if (!head) {
 		head = f2fs_kmem_cache_alloc(nat_entry_set_slab, GFP_NOFS);
@@ -167,25 +169,27 @@ static void __set_nat_cache_dirty(struct f2fs_nm_info *nm_i,
 		head->entry_cnt = 0;
 		f2fs_radix_tree_insert(&nm_i->nat_set_root, set, head);
 	}
-	list_move_tail(&ne->list, &head->entry_list);
+
+	if (get_nat_flag(ne, IS_DIRTY))
+		goto refresh_list;
+
 	nm_i->dirty_nat_cnt++;
 	head->entry_cnt++;
 	set_nat_flag(ne, IS_DIRTY, true);
+refresh_list:
+	if (nat_get_blkaddr(ne) == NEW_ADDR)
+		list_del_init(&ne->list);
+	else
+		list_move_tail(&ne->list, &head->entry_list);
 }
 
 static void __clear_nat_cache_dirty(struct f2fs_nm_info *nm_i,
-						struct nat_entry *ne)
+		struct nat_entry_set *set, struct nat_entry *ne)
 {
-	nid_t set = NAT_BLOCK_OFFSET(ne->ni.nid);
-	struct nat_entry_set *head;
-
-	head = radix_tree_lookup(&nm_i->nat_set_root, set);
-	if (head) {
-		list_move_tail(&ne->list, &nm_i->nat_entries);
-		set_nat_flag(ne, IS_DIRTY, false);
-		head->entry_cnt--;
-		nm_i->dirty_nat_cnt--;
-	}
+	list_move_tail(&ne->list, &nm_i->nat_entries);
+	set_nat_flag(ne, IS_DIRTY, false);
+	set->entry_cnt--;
+	nm_i->dirty_nat_cnt--;
 }
 
 static unsigned int __gang_lookup_nat_set(struct f2fs_nm_info *nm_i,
@@ -242,12 +246,24 @@ bool need_inode_block_update(struct f2fs_sb_info *sbi, nid_t ino)
 	return need_update;
 }
 
-static struct nat_entry *grab_nat_entry(struct f2fs_nm_info *nm_i, nid_t nid)
+static struct nat_entry *grab_nat_entry(struct f2fs_nm_info *nm_i, nid_t nid,
+								bool no_fail)
 {
 	struct nat_entry *new;
 
-	new = f2fs_kmem_cache_alloc(nat_entry_slab, GFP_NOFS);
-	f2fs_radix_tree_insert(&nm_i->nat_root, nid, new);
+	if (no_fail) {
+		new = f2fs_kmem_cache_alloc(nat_entry_slab, GFP_NOFS);
+		f2fs_radix_tree_insert(&nm_i->nat_root, nid, new);
+	} else {
+		new = kmem_cache_alloc(nat_entry_slab, GFP_NOFS);
+		if (!new)
+			return NULL;
+		if (radix_tree_insert(&nm_i->nat_root, nid, new)) {
+			kmem_cache_free(nat_entry_slab, new);
+			return NULL;
+		}
+	}
+
 	memset(new, 0, sizeof(struct nat_entry));
 	nat_set_nid(new, nid);
 	nat_reset_flag(new);
@@ -256,18 +272,23 @@ static struct nat_entry *grab_nat_entry(struct f2fs_nm_info *nm_i, nid_t nid)
 	return new;
 }
 
-static void cache_nat_entry(struct f2fs_nm_info *nm_i, nid_t nid,
+static void cache_nat_entry(struct f2fs_sb_info *sbi, nid_t nid,
 						struct f2fs_nat_entry *ne)
 {
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct nat_entry *e;
 
-	down_write(&nm_i->nat_tree_lock);
 	e = __lookup_nat_cache(nm_i, nid);
 	if (!e) {
-		e = grab_nat_entry(nm_i, nid);
-		node_info_from_raw_nat(&e->ni, ne);
+		e = grab_nat_entry(nm_i, nid, false);
+		if (e)
+			node_info_from_raw_nat(&e->ni, ne);
+	} else {
+		f2fs_bug_on(sbi, nat_get_ino(e) != le32_to_cpu(ne->ino) ||
+				nat_get_blkaddr(e) !=
+					le32_to_cpu(ne->block_addr) ||
+				nat_get_version(e) != ne->version);
 	}
-	up_write(&nm_i->nat_tree_lock);
 }
 
 static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
@@ -279,7 +300,7 @@ static void set_node_addr(struct f2fs_sb_info *sbi, struct node_info *ni,
 	down_write(&nm_i->nat_tree_lock);
 	e = __lookup_nat_cache(nm_i, ni->nid);
 	if (!e) {
-		e = grab_nat_entry(nm_i, ni->nid);
+		e = grab_nat_entry(nm_i, ni->nid, true);
 		copy_node_info(&e->ni, ni);
 		f2fs_bug_on(sbi, ni->blk_addr == NEW_ADDR);
 	} else if (new_blkaddr == NEW_ADDR) {
@@ -355,12 +376,13 @@ void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
 	nid_t start_nid = START_NID(nid);
 	struct f2fs_nat_block *nat_blk;
 	struct page *page = NULL;
 	struct f2fs_nat_entry ne;
 	struct nat_entry *e;
+	pgoff_t index;
 	int i;
 
 	ni->nid = nid;
@@ -372,43 +394,103 @@ void get_node_info(struct f2fs_sb_info *sbi, nid_t nid, struct node_info *ni)
 		ni->ino = nat_get_ino(e);
 		ni->blk_addr = nat_get_blkaddr(e);
 		ni->version = nat_get_version(e);
-	}
-	up_read(&nm_i->nat_tree_lock);
-	if (e)
+		up_read(&nm_i->nat_tree_lock);
 		return;
+	}
 
 	memset(&ne, 0, sizeof(struct f2fs_nat_entry));
 
 	/* Check current segment summary */
-	mutex_lock(&curseg->curseg_mutex);
-	i = lookup_journal_in_cursum(sum, NAT_JOURNAL, nid, 0);
+	down_read(&curseg->journal_rwsem);
+	i = lookup_journal_in_cursum(journal, NAT_JOURNAL, nid, 0);
 	if (i >= 0) {
-		ne = nat_in_journal(sum, i);
+		ne = nat_in_journal(journal, i);
 		node_info_from_raw_nat(ni, &ne);
 	}
-	mutex_unlock(&curseg->curseg_mutex);
-	if (i >= 0)
+	up_read(&curseg->journal_rwsem);
+	if (i >= 0) {
+		up_read(&nm_i->nat_tree_lock);
 		goto cache;
+	}
 
 	/* Fill node_info from nat page */
-	page = get_current_nat_page(sbi, start_nid);
+	index = current_nat_addr(sbi, nid);
+	up_read(&nm_i->nat_tree_lock);
+
+	page = get_meta_page(sbi, index);
 	nat_blk = (struct f2fs_nat_block *)page_address(page);
 	ne = nat_blk->entries[nid - start_nid];
 	node_info_from_raw_nat(ni, &ne);
 	f2fs_put_page(page, 1);
 cache:
 	/* cache nat entry */
-	cache_nat_entry(NM_I(sbi), nid, &ne);
+	down_write(&nm_i->nat_tree_lock);
+	cache_nat_entry(sbi, nid, &ne);
+	up_write(&nm_i->nat_tree_lock);
+}
+
+/*
+ * readahead MAX_RA_NODE number of node pages.
+ */
+static void ra_node_pages(struct page *parent, int start, int n)
+{
+	struct f2fs_sb_info *sbi = F2FS_P_SB(parent);
+	struct blk_plug plug;
+	int i, end;
+	nid_t nid;
+
+	blk_start_plug(&plug);
+
+	/* Then, try readahead for siblings of the desired node */
+	end = start + n;
+	end = min(end, NIDS_PER_BLOCK);
+	for (i = start; i < end; i++) {
+		nid = get_nid(parent, i, false);
+		ra_node_page(sbi, nid);
+	}
+
+	blk_finish_plug(&plug);
+}
+
+pgoff_t get_next_page_offset(struct dnode_of_data *dn, pgoff_t pgofs)
+{
+	const long direct_index = ADDRS_PER_INODE(dn->inode);
+	const long direct_blks = ADDRS_PER_BLOCK;
+	const long indirect_blks = ADDRS_PER_BLOCK * NIDS_PER_BLOCK;
+	unsigned int skipped_unit = ADDRS_PER_BLOCK;
+	int cur_level = dn->cur_level;
+	int max_level = dn->max_level;
+	pgoff_t base = 0;
+
+	if (!dn->max_level)
+		return pgofs + 1;
+
+	while (max_level-- > cur_level)
+		skipped_unit *= NIDS_PER_BLOCK;
+
+	switch (dn->max_level) {
+	case 3:
+		base += 2 * indirect_blks;
+	case 2:
+		base += 2 * direct_blks;
+	case 1:
+		base += direct_index;
+		break;
+	default:
+		f2fs_bug_on(F2FS_I_SB(dn->inode), 1);
+	}
+
+	return ((pgofs - base) / skipped_unit + 1) * skipped_unit + base;
 }
 
 /*
  * The maximum depth is four.
  * Offset[0] will have raw inode offset.
  */
-static int get_node_path(struct f2fs_inode_info *fi, long block,
+static int get_node_path(struct inode *inode, long block,
 				int offset[4], unsigned int noffset[4])
 {
-	const long direct_index = ADDRS_PER_INODE(fi);
+	const long direct_index = ADDRS_PER_INODE(inode);
 	const long direct_blks = ADDRS_PER_BLOCK;
 	const long dptrs_per_blk = NIDS_PER_BLOCK;
 	const long indirect_blks = ADDRS_PER_BLOCK * NIDS_PER_BLOCK;
@@ -473,7 +555,7 @@ static int get_node_path(struct f2fs_inode_info *fi, long block,
 		level = 3;
 		goto got;
 	} else {
-		BUG();
+		return -E2BIG;
 	}
 got:
 	return level;
@@ -493,10 +575,12 @@ int get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
 	int offset[4];
 	unsigned int noffset[4];
 	nid_t nids[4];
-	int level, i;
+	int level, i = 0;
 	int err = 0;
 
-	level = get_node_path(F2FS_I(dn->inode), index, offset, noffset);
+	level = get_node_path(dn->inode, index, offset, noffset);
+	if (level < 0)
+		return level;
 
 	nids[0] = dn->inode->i_ino;
 	npage[0] = dn->inode_page;
@@ -532,7 +616,7 @@ int get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
 			}
 
 			dn->nid = nids[i];
-			npage[i] = new_node_page(dn, noffset[i], NULL);
+			npage[i] = new_node_page(dn, noffset[i]);
 			if (IS_ERR(npage[i])) {
 				alloc_nid_failed(sbi, nids[i]);
 				err = PTR_ERR(npage[i]);
@@ -573,7 +657,8 @@ int get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode)
 	dn->nid = nids[level];
 	dn->ofs_in_node = offset[level];
 	dn->node_page = npage[level];
-	dn->data_blkaddr = datablock_addr(dn->node_page, dn->ofs_in_node);
+	dn->data_blkaddr = datablock_addr(dn->inode,
+				dn->node_page, dn->ofs_in_node);
 	return 0;
 
 release_pages:
@@ -583,6 +668,11 @@ release_pages:
 release_out:
 	dn->inode_page = NULL;
 	dn->node_page = NULL;
+	if (err == -ENOENT) {
+		dn->cur_level = i;
+		dn->max_level = level;
+		dn->ofs_in_node = offset[level];
+	}
 	return err;
 }
 
@@ -592,24 +682,19 @@ static void truncate_node(struct dnode_of_data *dn)
 	struct node_info ni;
 
 	get_node_info(sbi, dn->nid, &ni);
-	if (dn->inode->i_blocks == 0) {
-		f2fs_bug_on(sbi, ni.blk_addr != NULL_ADDR);
-		goto invalidate;
-	}
 	f2fs_bug_on(sbi, ni.blk_addr == NULL_ADDR);
 
 	/* Deallocate node address */
 	invalidate_blocks(sbi, ni.blk_addr);
-	dec_valid_node_count(sbi, dn->inode);
+	dec_valid_node_count(sbi, dn->inode, dn->nid == dn->inode->i_ino);
 	set_node_addr(sbi, &ni, NULL_ADDR, false);
 
 	if (dn->nid == dn->inode->i_ino) {
 		remove_orphan_inode(sbi, dn->nid);
 		dec_valid_inode_count(sbi);
-	} else {
-		sync_inode_page(dn);
+		f2fs_inode_synced(dn->inode);
 	}
-invalidate:
+
 	clear_node_page_dirty(dn->node_page);
 	set_sbi_flag(sbi, SBI_IS_DIRTY);
 
@@ -666,6 +751,8 @@ static int truncate_nodes(struct dnode_of_data *dn, unsigned int nofs,
 		return PTR_ERR(page);
 	}
 
+	ra_node_pages(page, ofs, NIDS_PER_BLOCK);
+
 	rn = F2FS_NODE(page);
 	if (depth < 3) {
 		for (i = ofs; i < NIDS_PER_BLOCK; i++, freed++) {
@@ -676,7 +763,8 @@ static int truncate_nodes(struct dnode_of_data *dn, unsigned int nofs,
 			ret = truncate_dnode(&rdn);
 			if (ret < 0)
 				goto out_err;
-			set_nid(page, i, 0, false);
+			if (set_nid(page, i, 0, false))
+				dn->node_changed = true;
 		}
 	} else {
 		child_nofs = nofs + ofs * (NIDS_PER_BLOCK + 1) + 1;
@@ -689,7 +777,8 @@ static int truncate_nodes(struct dnode_of_data *dn, unsigned int nofs,
 			rdn.nid = child_nid;
 			ret = truncate_nodes(&rdn, child_nofs, 0, depth - 1);
 			if (ret == (NIDS_PER_BLOCK + 1)) {
-				set_nid(page, i, 0, false);
+				if (set_nid(page, i, 0, false))
+					dn->node_changed = true;
 				child_nofs += ret;
 			} else if (ret < 0 && ret != -ENOENT) {
 				goto out_err;
@@ -741,6 +830,8 @@ static int truncate_partial_nodes(struct dnode_of_data *dn,
 		nid[i + 1] = get_nid(pages[i], offset[i + 1], false);
 	}
 
+	ra_node_pages(pages[idx], offset[idx + 1], NIDS_PER_BLOCK);
+
 	/* free direct nodes linked to a partial indirect node */
 	for (i = offset[idx + 1]; i < NIDS_PER_BLOCK; i++) {
 		child_nid = get_nid(pages[idx], i, false);
@@ -750,7 +841,8 @@ static int truncate_partial_nodes(struct dnode_of_data *dn,
 		err = truncate_dnode(dn);
 		if (err < 0)
 			goto fail;
-		set_nid(pages[idx], i, 0, false);
+		if (set_nid(pages[idx], i, 0, false))
+			dn->node_changed = true;
 	}
 
 	if (offset[idx + 1] == 0) {
@@ -787,8 +879,10 @@ int truncate_inode_blocks(struct inode *inode, pgoff_t from)
 
 	trace_f2fs_truncate_inode_blocks_enter(inode, from);
 
-	level = get_node_path(F2FS_I(inode), from, offset, noffset);
-restart:
+	level = get_node_path(inode, from, offset, noffset);
+	if (level < 0)
+		return level;
+
 	page = get_node_page(sbi, inode->i_ino);
 	if (IS_ERR(page)) {
 		trace_f2fs_truncate_inode_blocks_exit(inode, PTR_ERR(page));
@@ -852,11 +946,8 @@ skip_partial:
 		if (offset[1] == 0 &&
 				ri->i_nid[offset[0] - NODE_DIR1_BLOCK]) {
 			lock_page(page);
-			if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
-				f2fs_put_page(page, 1);
-				goto restart;
-			}
-			f2fs_wait_on_page_writeback(page, NODE);
+			BUG_ON(page->mapping != NODE_MAPPING(sbi));
+			f2fs_wait_on_page_writeback(page, NODE, true);
 			ri->i_nid[offset[0] - NODE_DIR1_BLOCK] = 0;
 			set_page_dirty(page);
 			unlock_page(page);
@@ -885,10 +976,7 @@ int truncate_xattr_node(struct inode *inode, struct page *page)
 	if (IS_ERR(npage))
 		return PTR_ERR(npage);
 
-	F2FS_I(inode)->i_xattr_nid = 0;
-
-	/* need to do checkpoint during fsync */
-	F2FS_I(inode)->xattr_ver = cur_cp_version(F2FS_CKPT(sbi));
+	f2fs_i_xnid_write(inode, 0);
 
 	set_new_dnode(&dn, inode, page, npage, nid);
 
@@ -925,7 +1013,7 @@ int remove_inode_page(struct inode *inode)
 
 	/* 0 is possible, after f2fs_new_inode() has failed */
 	f2fs_bug_on(F2FS_I_SB(inode),
-			inode->i_blocks != 0 && inode->i_blocks != 1);
+			inode->i_blocks != 0 && inode->i_blocks != 8);
 
 	/* will put inode & node pages */
 	truncate_node(&dn);
@@ -940,54 +1028,50 @@ struct page *new_inode_page(struct inode *inode)
 	set_new_dnode(&dn, inode, NULL, NULL, inode->i_ino);
 
 	/* caller should f2fs_put_page(page, 1); */
-	return new_node_page(&dn, 0, NULL);
+	return new_node_page(&dn, 0);
 }
 
-struct page *new_node_page(struct dnode_of_data *dn,
-				unsigned int ofs, struct page *ipage)
+struct page *new_node_page(struct dnode_of_data *dn, unsigned int ofs)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(dn->inode);
-	struct node_info old_ni, new_ni;
+	struct node_info new_ni;
 	struct page *page;
 	int err;
 
-	if (unlikely(is_inode_flag_set(F2FS_I(dn->inode), FI_NO_ALLOC)))
+	if (unlikely(is_inode_flag_set(dn->inode, FI_NO_ALLOC)))
 		return ERR_PTR(-EPERM);
 
-	page = grab_cache_page(NODE_MAPPING(sbi), dn->nid);
+	page = f2fs_grab_cache_page(NODE_MAPPING(sbi), dn->nid, false);
 	if (!page)
 		return ERR_PTR(-ENOMEM);
 
-	if (unlikely(!inc_valid_node_count(sbi, dn->inode))) {
-		err = -ENOSPC;
+	if (unlikely((err = inc_valid_node_count(sbi, dn->inode, !ofs))))
 		goto fail;
-	}
-
-	get_node_info(sbi, dn->nid, &old_ni);
 
-	/* Reinitialize old_ni with new node page */
-	f2fs_bug_on(sbi, old_ni.blk_addr != NULL_ADDR);
-	new_ni = old_ni;
+#ifdef CONFIG_F2FS_CHECK_FS
+	get_node_info(sbi, dn->nid, &new_ni);
+	f2fs_bug_on(sbi, new_ni.blk_addr != NULL_ADDR);
+#endif
+	new_ni.nid = dn->nid;
 	new_ni.ino = dn->inode->i_ino;
+	new_ni.blk_addr = NULL_ADDR;
+	new_ni.flag = 0;
+	new_ni.version = 0;
 	set_node_addr(sbi, &new_ni, NEW_ADDR, false);
 
-	f2fs_wait_on_page_writeback(page, NODE);
+	f2fs_wait_on_page_writeback(page, NODE, true);
 	fill_node_footer(page, dn->nid, dn->inode->i_ino, ofs, true);
 	set_cold_node(dn->inode, page);
-	SetPageUptodate(page);
-	set_page_dirty(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
+	if (set_page_dirty(page))
+		dn->node_changed = true;
 
 	if (f2fs_has_xattr_block(ofs))
-		F2FS_I(dn->inode)->i_xattr_nid = dn->nid;
+		f2fs_i_xnid_write(dn->inode, dn->nid);
 
-	dn->node_page = page;
-	if (ipage)
-		update_inode(dn->inode, ipage);
-	else
-		sync_inode_page(dn);
 	if (ofs == 0)
 		inc_valid_inode_count(sbi);
-
 	return page;
 
 fail:
@@ -1001,18 +1085,22 @@ fail:
  * 0: f2fs_put_page(page, 0)
  * LOCKED_PAGE or error: f2fs_put_page(page, 1)
  */
-static int read_node_page(struct page *page, int rw)
+static int read_node_page(struct page *page, int op_flags)
 {
 	struct f2fs_sb_info *sbi = F2FS_P_SB(page);
 	struct node_info ni;
 	struct f2fs_io_info fio = {
 		.sbi = sbi,
 		.type = NODE,
-		.rw = rw,
+		.op = REQ_OP_READ,
+		.op_flags = op_flags,
 		.page = page,
 		.encrypted_page = NULL,
 	};
 
+	if (PageUptodate(page))
+		return LOCKED_PAGE;
+
 	get_node_info(sbi, page->index, &ni);
 
 	if (unlikely(ni.blk_addr == NULL_ADDR)) {
@@ -1020,10 +1108,7 @@ static int read_node_page(struct page *page, int rw)
 		return -ENOENT;
 	}
 
-	if (PageUptodate(page))
-		return LOCKED_PAGE;
-
-	fio.blk_addr = ni.blk_addr;
+	fio.new_blkaddr = fio.old_blkaddr = ni.blk_addr;
 	return f2fs_submit_page_bio(&fio);
 }
 
@@ -1035,134 +1120,445 @@ void ra_node_page(struct f2fs_sb_info *sbi, nid_t nid)
 	struct page *apage;
 	int err;
 
-	apage = find_get_page(NODE_MAPPING(sbi), nid);
-	if (apage && PageUptodate(apage)) {
-		f2fs_put_page(apage, 0);
+	if (!nid)
+		return;
+	f2fs_bug_on(sbi, check_nid_range(sbi, nid));
+
+	rcu_read_lock();
+	apage = radix_tree_lookup(&NODE_MAPPING(sbi)->page_tree, nid);
+	rcu_read_unlock();
+	if (apage)
 		return;
-	}
-	f2fs_put_page(apage, 0);
 
-	apage = grab_cache_page(NODE_MAPPING(sbi), nid);
+	apage = f2fs_grab_cache_page(NODE_MAPPING(sbi), nid, false);
 	if (!apage)
 		return;
 
-	err = read_node_page(apage, READA);
+	err = read_node_page(apage, REQ_RAHEAD);
 	f2fs_put_page(apage, err ? 1 : 0);
 }
 
-struct page *get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid)
+static struct page *__get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid,
+					struct page *parent, int start)
 {
 	struct page *page;
 	int err;
+
+	if (!nid)
+		return ERR_PTR(-ENOENT);
+	f2fs_bug_on(sbi, check_nid_range(sbi, nid));
 repeat:
-	page = grab_cache_page(NODE_MAPPING(sbi), nid);
+	page = f2fs_grab_cache_page(NODE_MAPPING(sbi), nid, false);
 	if (!page)
 		return ERR_PTR(-ENOMEM);
 
-	err = read_node_page(page, READ_SYNC);
+	err = read_node_page(page, REQ_SYNC);
 	if (err < 0) {
 		f2fs_put_page(page, 1);
 		return ERR_PTR(err);
-	} else if (err != LOCKED_PAGE) {
-		lock_page(page);
+	} else if (err == LOCKED_PAGE) {
+		err = 0;
+		goto page_hit;
 	}
 
-	if (unlikely(!PageUptodate(page) || nid != nid_of_node(page))) {
-		ClearPageUptodate(page);
-		f2fs_put_page(page, 1);
-		return ERR_PTR(-EIO);
-	}
+	if (parent)
+		ra_node_pages(parent, start + 1, MAX_RA_NODE);
+
+	lock_page(page);
+
 	if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
 		f2fs_put_page(page, 1);
 		goto repeat;
 	}
+
+	if (unlikely(!PageUptodate(page))) {
+		err = -EIO;
+		goto out_err;
+	}
+
+	if (!f2fs_inode_chksum_verify(sbi, page)) {
+		err = -EBADMSG;
+		goto out_err;
+	}
+page_hit:
+	if(unlikely(nid != nid_of_node(page))) {
+		f2fs_msg(sbi->sb, KERN_WARNING, "inconsistent node block, "
+			"nid:%lu, node_footer[nid:%u,ino:%u,ofs:%u,cpver:%llu,blkaddr:%u]",
+			nid, nid_of_node(page), ino_of_node(page),
+			ofs_of_node(page), cpver_of_node(page),
+			next_blkaddr_of_node(page));
+		err = -EINVAL;
+out_err:
+		ClearPageUptodate(page);
+		f2fs_put_page(page, 1);
+		return ERR_PTR(err);
+	}
 	return page;
 }
 
-/*
- * Return a locked page for the desired node page.
- * And, readahead MAX_RA_NODE number of node pages.
- */
+struct page *get_node_page(struct f2fs_sb_info *sbi, pgoff_t nid)
+{
+	return __get_node_page(sbi, nid, NULL, 0);
+}
+
 struct page *get_node_page_ra(struct page *parent, int start)
 {
 	struct f2fs_sb_info *sbi = F2FS_P_SB(parent);
-	struct blk_plug plug;
+	nid_t nid = get_nid(parent, start, false);
+
+	return __get_node_page(sbi, nid, parent, start);
+}
+
+static void flush_inline_data(struct f2fs_sb_info *sbi, nid_t ino)
+{
+	struct inode *inode;
 	struct page *page;
-	int err, i, end;
-	nid_t nid;
+	int ret;
 
-	/* First, try getting the desired direct node. */
-	nid = get_nid(parent, start, false);
-	if (!nid)
-		return ERR_PTR(-ENOENT);
-repeat:
-	page = grab_cache_page(NODE_MAPPING(sbi), nid);
+	/* should flush inline_data before evict_inode */
+	inode = ilookup(sbi->sb, ino);
+	if (!inode)
+		return;
+
+	page = pagecache_get_page(inode->i_mapping, 0, FGP_LOCK|FGP_NOWAIT, 0);
 	if (!page)
-		return ERR_PTR(-ENOMEM);
+		goto iput_out;
 
-	err = read_node_page(page, READ_SYNC);
-	if (err < 0) {
-		f2fs_put_page(page, 1);
-		return ERR_PTR(err);
-	} else if (err == LOCKED_PAGE) {
-		goto page_hit;
+	if (!PageUptodate(page))
+		goto page_out;
+
+	if (!PageDirty(page))
+		goto page_out;
+
+	if (!clear_page_dirty_for_io(page))
+		goto page_out;
+
+	ret = f2fs_write_inline_data(inode, page);
+	inode_dec_dirty_pages(inode);
+	remove_dirty_inode(inode);
+	if (ret)
+		set_page_dirty(page);
+page_out:
+	f2fs_put_page(page, 1);
+iput_out:
+	iput(inode);
+}
+
+void move_node_page(struct page *node_page, int gc_type)
+{
+	if (gc_type == FG_GC) {
+		struct f2fs_sb_info *sbi = F2FS_P_SB(node_page);
+		struct writeback_control wbc = {
+			.sync_mode = WB_SYNC_ALL,
+			.nr_to_write = 1,
+			.for_reclaim = 0,
+		};
+
+		set_page_dirty(node_page);
+		f2fs_wait_on_page_writeback(node_page, NODE, true);
+
+		f2fs_bug_on(sbi, PageWriteback(node_page));
+		if (!clear_page_dirty_for_io(node_page))
+			goto out_page;
+
+		if (NODE_MAPPING(sbi)->a_ops->writepage(node_page, &wbc))
+			unlock_page(node_page);
+		goto release_page;
+	} else {
+		/* set page dirty and write it */
+		if (!PageWriteback(node_page))
+			set_page_dirty(node_page);
+	}
+out_page:
+	unlock_page(node_page);
+release_page:
+	f2fs_put_page(node_page, 0);
+}
+
+static struct page *last_fsync_dnode(struct f2fs_sb_info *sbi, nid_t ino)
+{
+	pgoff_t index, end;
+	struct pagevec pvec;
+	struct page *last_page = NULL;
+
+	pagevec_init(&pvec, 0);
+	index = 0;
+	end = ULONG_MAX;
+
+	while (index <= end) {
+		int i, nr_pages;
+		nr_pages = pagevec_lookup_tag(&pvec, NODE_MAPPING(sbi), &index,
+				PAGECACHE_TAG_DIRTY,
+				min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1);
+		if (nr_pages == 0)
+			break;
+
+		for (i = 0; i < nr_pages; i++) {
+			struct page *page = pvec.pages[i];
+
+			if (unlikely(f2fs_cp_error(sbi))) {
+				f2fs_put_page(last_page, 0);
+				pagevec_release(&pvec);
+				return ERR_PTR(-EIO);
+			}
+
+			if (!IS_DNODE(page) || !is_cold_node(page))
+				continue;
+			if (ino_of_node(page) != ino)
+				continue;
+
+			lock_page(page);
+
+			if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
+continue_unlock:
+				unlock_page(page);
+				continue;
+			}
+			if (ino_of_node(page) != ino)
+				goto continue_unlock;
+
+			if (!PageDirty(page)) {
+				/* someone wrote it for us */
+				goto continue_unlock;
+			}
+
+			if (last_page)
+				f2fs_put_page(last_page, 0);
+
+			get_page(page);
+			last_page = page;
+			unlock_page(page);
+		}
+		pagevec_release(&pvec);
+		cond_resched();
 	}
+	return last_page;
+}
 
-	blk_start_plug(&plug);
+static int __write_node_page(struct page *page, bool atomic, bool *submitted,
+				struct writeback_control *wbc, bool do_balance,
+				enum iostat_type io_type)
+{
+	struct f2fs_sb_info *sbi = F2FS_P_SB(page);
+	nid_t nid;
+	struct node_info ni;
+	struct f2fs_io_info fio = {
+		.sbi = sbi,
+		.type = NODE,
+		.op = REQ_OP_WRITE,
+		.op_flags = wbc_to_write_flags(wbc),
+		.page = page,
+		.encrypted_page = NULL,
+		.submitted = false,
+		.io_type = io_type,
+	};
 
-	/* Then, try readahead for siblings of the desired node */
-	end = start + MAX_RA_NODE;
-	end = min(end, NIDS_PER_BLOCK);
-	for (i = start + 1; i < end; i++) {
-		nid = get_nid(parent, i, false);
-		if (!nid)
-			continue;
-		ra_node_page(sbi, nid);
+	trace_f2fs_writepage(page, NODE);
+
+	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+		goto redirty_out;
+	if (unlikely(f2fs_cp_error(sbi)))
+		goto redirty_out;
+
+	/* get old block addr of this node page */
+	nid = nid_of_node(page);
+	f2fs_bug_on(sbi, page->index != nid);
+
+	if (wbc->for_reclaim) {
+		if (!down_read_trylock(&sbi->node_write))
+			goto redirty_out;
+	} else {
+		down_read(&sbi->node_write);
 	}
 
-	blk_finish_plug(&plug);
+	get_node_info(sbi, nid, &ni);
 
-	lock_page(page);
-	if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
-		f2fs_put_page(page, 1);
-		goto repeat;
+	/* This page is already truncated */
+	if (unlikely(ni.blk_addr == NULL_ADDR)) {
+		ClearPageUptodate(page);
+		dec_page_count(sbi, F2FS_DIRTY_NODES);
+		up_read(&sbi->node_write);
+		unlock_page(page);
+		return 0;
 	}
-page_hit:
-	if (unlikely(!PageUptodate(page))) {
-		f2fs_put_page(page, 1);
-		return ERR_PTR(-EIO);
+
+	if (atomic && !test_opt(sbi, NOBARRIER))
+		fio.op_flags |= WRITE_FLUSH_FUA;
+
+	set_page_writeback(page);
+	fio.old_blkaddr = ni.blk_addr;
+	write_node_page(nid, &fio);
+	set_node_addr(sbi, &ni, fio.new_blkaddr, is_fsync_dnode(page));
+	dec_page_count(sbi, F2FS_DIRTY_NODES);
+	up_read(&sbi->node_write);
+
+	if (wbc->for_reclaim) {
+		f2fs_submit_merged_write_cond(sbi, page->mapping->host, 0,
+						page->index, NODE);
+		submitted = NULL;
 	}
-	return page;
+
+	unlock_page(page);
+
+	if (unlikely(f2fs_cp_error(sbi))) {
+		f2fs_submit_merged_write(sbi, NODE);
+		submitted = NULL;
+	}
+	if (submitted)
+		*submitted = fio.submitted;
+
+	if (do_balance)
+		f2fs_balance_fs(sbi, false);
+	return 0;
+
+redirty_out:
+	redirty_page_for_writepage(wbc, page);
+	return AOP_WRITEPAGE_ACTIVATE;
 }
 
-void sync_inode_page(struct dnode_of_data *dn)
+static int f2fs_write_node_page(struct page *page,
+				struct writeback_control *wbc)
 {
-	if (IS_INODE(dn->node_page) || dn->inode_page == dn->node_page) {
-		update_inode(dn->inode, dn->node_page);
-	} else if (dn->inode_page) {
-		if (!dn->inode_page_locked)
-			lock_page(dn->inode_page);
-		update_inode(dn->inode, dn->inode_page);
-		if (!dn->inode_page_locked)
-			unlock_page(dn->inode_page);
-	} else {
-		update_inode_page(dn->inode);
+	return __write_node_page(page, false, NULL, wbc, false, FS_NODE_IO);
+}
+
+int fsync_node_pages(struct f2fs_sb_info *sbi, struct inode *inode,
+			struct writeback_control *wbc, bool atomic)
+{
+	pgoff_t index, end;
+	pgoff_t last_idx = ULONG_MAX;
+	struct pagevec pvec;
+	int ret = 0;
+	struct page *last_page = NULL;
+	bool marked = false;
+	nid_t ino = inode->i_ino;
+
+	if (atomic) {
+		last_page = last_fsync_dnode(sbi, ino);
+		if (IS_ERR_OR_NULL(last_page))
+			return PTR_ERR_OR_ZERO(last_page);
 	}
+retry:
+	pagevec_init(&pvec, 0);
+	index = 0;
+	end = ULONG_MAX;
+
+	while (index <= end) {
+		int i, nr_pages;
+		nr_pages = pagevec_lookup_tag(&pvec, NODE_MAPPING(sbi), &index,
+				PAGECACHE_TAG_DIRTY,
+				min(end - index, (pgoff_t)PAGEVEC_SIZE-1) + 1);
+		if (nr_pages == 0)
+			break;
+
+		for (i = 0; i < nr_pages; i++) {
+			struct page *page = pvec.pages[i];
+			bool submitted = false;
+
+			if (unlikely(f2fs_cp_error(sbi))) {
+				f2fs_put_page(last_page, 0);
+				pagevec_release(&pvec);
+				ret = -EIO;
+				goto out;
+			}
+
+			if (!IS_DNODE(page) || !is_cold_node(page))
+				continue;
+			if (ino_of_node(page) != ino)
+				continue;
+
+			lock_page(page);
+
+			if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
+continue_unlock:
+				unlock_page(page);
+				continue;
+			}
+			if (ino_of_node(page) != ino)
+				goto continue_unlock;
+
+			if (!PageDirty(page) && page != last_page) {
+				/* someone wrote it for us */
+				goto continue_unlock;
+			}
+
+			f2fs_wait_on_page_writeback(page, NODE, true);
+			BUG_ON(PageWriteback(page));
+
+			set_fsync_mark(page, 0);
+			set_dentry_mark(page, 0);
+
+			if (!atomic || page == last_page) {
+				set_fsync_mark(page, 1);
+				if (IS_INODE(page)) {
+					if (is_inode_flag_set(inode,
+								FI_DIRTY_INODE))
+						update_inode(inode, page);
+					set_dentry_mark(page,
+						need_dentry_mark(sbi, ino));
+				}
+				/*  may be written by other thread */
+				if (!PageDirty(page))
+					set_page_dirty(page);
+			}
+
+			if (!clear_page_dirty_for_io(page))
+				goto continue_unlock;
+
+			ret = __write_node_page(page, atomic &&
+						page == last_page,
+						&submitted, wbc, true,
+						FS_NODE_IO);
+			if (ret) {
+				unlock_page(page);
+				f2fs_put_page(last_page, 0);
+				break;
+			} else if (submitted) {
+				last_idx = page->index;
+			}
+
+			if (page == last_page) {
+				f2fs_put_page(page, 0);
+				marked = true;
+				break;
+			}
+		}
+		pagevec_release(&pvec);
+		cond_resched();
+
+		if (ret || marked)
+			break;
+	}
+	if (!ret && atomic && !marked) {
+		f2fs_msg(sbi->sb, KERN_DEBUG,
+			"Retry to write fsync mark: ino=%u, idx=%lx",
+					ino, last_page->index);
+		lock_page(last_page);
+		f2fs_wait_on_page_writeback(last_page, NODE, true);
+		set_page_dirty(last_page);
+		unlock_page(last_page);
+		goto retry;
+	}
+out:
+	if (last_idx != ULONG_MAX)
+		f2fs_submit_merged_write_cond(sbi, NULL, ino, last_idx, NODE);
+	return ret ? -EIO: 0;
 }
 
-int sync_node_pages(struct f2fs_sb_info *sbi, nid_t ino,
-					struct writeback_control *wbc)
+int sync_node_pages(struct f2fs_sb_info *sbi, struct writeback_control *wbc,
+				bool do_balance, enum iostat_type io_type)
 {
 	pgoff_t index, end;
 	struct pagevec pvec;
-	int step = ino ? 2 : 0;
-	int nwritten = 0, wrote = 0;
+	int step = 0;
+	int nwritten = 0;
+	int ret = 0;
 
 	pagevec_init(&pvec, 0);
 
 next_step:
 	index = 0;
-	end = LONG_MAX;
+	end = ULONG_MAX;
 
 	while (index <= end) {
 		int i, nr_pages;
@@ -1174,6 +1570,13 @@ next_step:
 
 		for (i = 0; i < nr_pages; i++) {
 			struct page *page = pvec.pages[i];
+			bool submitted = false;
+
+			if (unlikely(f2fs_cp_error(sbi))) {
+				pagevec_release(&pvec);
+				ret = -EIO;
+				goto out;
+			}
 
 			/*
 			 * flushing sequence with step:
@@ -1189,14 +1592,8 @@ next_step:
 			if (step == 2 && (!IS_DNODE(page) ||
 						!is_cold_node(page)))
 				continue;
-
-			/*
-			 * If an fsync mode,
-			 * we should not skip writing node pages.
-			 */
-			if (ino && ino_of_node(page) == ino)
-				lock_page(page);
-			else if (!trylock_page(page))
+lock_node:
+			if (!trylock_page(page))
 				continue;
 
 			if (unlikely(page->mapping != NODE_MAPPING(sbi))) {
@@ -1204,33 +1601,35 @@ continue_unlock:
 				unlock_page(page);
 				continue;
 			}
-			if (ino && ino_of_node(page) != ino)
-				goto continue_unlock;
 
 			if (!PageDirty(page)) {
 				/* someone wrote it for us */
 				goto continue_unlock;
 			}
 
+			/* flush inline_data */
+			if (is_inline_node(page)) {
+				clear_inline_node(page);
+				unlock_page(page);
+				flush_inline_data(sbi, ino_of_node(page));
+				goto lock_node;
+			}
+
+			f2fs_wait_on_page_writeback(page, NODE, true);
+
+			BUG_ON(PageWriteback(page));
 			if (!clear_page_dirty_for_io(page))
 				goto continue_unlock;
 
-			/* called by fsync() */
-			if (ino && IS_DNODE(page)) {
-				set_fsync_mark(page, 1);
-				if (IS_INODE(page))
-					set_dentry_mark(page,
-						need_dentry_mark(sbi, ino));
-				nwritten++;
-			} else {
-				set_fsync_mark(page, 0);
-				set_dentry_mark(page, 0);
-			}
+			set_fsync_mark(page, 0);
+			set_dentry_mark(page, 0);
 
-			if (NODE_MAPPING(sbi)->a_ops->writepage(page, wbc))
+			ret = __write_node_page(page, false, &submitted,
+						wbc, do_balance, io_type);
+			if (ret)
 				unlock_page(page);
-			else
-				wrote++;
+			else if (submitted)
+				nwritten++;
 
 			if (--wbc->nr_to_write == 0)
 				break;
@@ -1248,15 +1647,15 @@ continue_unlock:
 		step++;
 		goto next_step;
 	}
-
-	if (wrote)
-		f2fs_submit_merged_bio(sbi, NODE, WRITE);
-	return nwritten;
+out:
+	if (nwritten)
+		f2fs_submit_merged_write(sbi, NODE);
+	return ret;
 }
 
 int wait_on_node_pages_writeback(struct f2fs_sb_info *sbi, nid_t ino)
 {
-	pgoff_t index = 0, end = LONG_MAX;
+	pgoff_t index = 0, end = ULONG_MAX;
 	struct pagevec pvec;
 	int ret2 = 0, ret = 0;
 
@@ -1278,7 +1677,7 @@ int wait_on_node_pages_writeback(struct f2fs_sb_info *sbi, nid_t ino)
 				continue;
 
 			if (ino && ino_of_node(page) == ino) {
-				f2fs_wait_on_page_writeback(page, NODE);
+				f2fs_wait_on_page_writeback(page, NODE, true);
 				if (TestClearPageError(page))
 					ret = -EIO;
 			}
@@ -1296,76 +1695,15 @@ int wait_on_node_pages_writeback(struct f2fs_sb_info *sbi, nid_t ino)
 	return ret;
 }
 
-static int f2fs_write_node_page(struct page *page,
-				struct writeback_control *wbc)
-{
-	struct f2fs_sb_info *sbi = F2FS_P_SB(page);
-	nid_t nid;
-	struct node_info ni;
-	struct f2fs_io_info fio = {
-		.sbi = sbi,
-		.type = NODE,
-		.rw = (wbc->sync_mode == WB_SYNC_ALL) ? WRITE_SYNC : WRITE,
-		.page = page,
-		.encrypted_page = NULL,
-	};
-
-	trace_f2fs_writepage(page, NODE);
-
-	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
-		goto redirty_out;
-	if (unlikely(f2fs_cp_error(sbi)))
-		goto redirty_out;
-
-	f2fs_wait_on_page_writeback(page, NODE);
-
-	/* get old block addr of this node page */
-	nid = nid_of_node(page);
-	f2fs_bug_on(sbi, page->index != nid);
-
-	if (wbc->for_reclaim) {
-		if (!down_read_trylock(&sbi->node_write))
-			goto redirty_out;
-	} else {
-		down_read(&sbi->node_write);
-	}
-
-	get_node_info(sbi, nid, &ni);
-
-	/* This page is already truncated */
-	if (unlikely(ni.blk_addr == NULL_ADDR)) {
-		ClearPageUptodate(page);
-		dec_page_count(sbi, F2FS_DIRTY_NODES);
-		up_read(&sbi->node_write);
-		unlock_page(page);
-		return 0;
-	}
-
-	set_page_writeback(page);
-	fio.blk_addr = ni.blk_addr;
-	write_node_page(nid, &fio);
-	set_node_addr(sbi, &ni, fio.blk_addr, is_fsync_dnode(page));
-	dec_page_count(sbi, F2FS_DIRTY_NODES);
-	up_read(&sbi->node_write);
-	unlock_page(page);
-
-	if (wbc->for_reclaim)
-		f2fs_submit_merged_bio(sbi, NODE, WRITE);
-
-	return 0;
-
-redirty_out:
-	redirty_page_for_writepage(wbc, page);
-	return AOP_WRITEPAGE_ACTIVATE;
-}
-
 static int f2fs_write_node_pages(struct address_space *mapping,
 			    struct writeback_control *wbc)
 {
 	struct f2fs_sb_info *sbi = F2FS_M_SB(mapping);
+	struct blk_plug plug;
 	long diff;
 
-	trace_f2fs_writepages(mapping->host, wbc, NODE);
+	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+		goto skip_write;
 
 	/* balancing f2fs's metadata in background */
 	f2fs_balance_fs_bg(sbi);
@@ -1374,14 +1712,19 @@ static int f2fs_write_node_pages(struct address_space *mapping,
 	if (get_pages(sbi, F2FS_DIRTY_NODES) < nr_pages_to_skip(sbi, NODE))
 		goto skip_write;
 
+	trace_f2fs_writepages(mapping->host, wbc, NODE);
+
 	diff = nr_pages_to_write(sbi, NODE, wbc);
 	wbc->sync_mode = WB_SYNC_NONE;
-	sync_node_pages(sbi, 0, wbc);
+	blk_start_plug(&plug);
+	sync_node_pages(sbi, wbc, true, FS_NODE_IO);
+	blk_finish_plug(&plug);
 	wbc->nr_to_write = max((long)0, wbc->nr_to_write - diff);
 	return 0;
 
 skip_write:
 	wbc->pages_skipped += get_pages(sbi, F2FS_DIRTY_NODES);
+	trace_f2fs_writepages(mapping->host, wbc, NODE);
 	return 0;
 }
 
@@ -1389,9 +1732,10 @@ static int f2fs_set_node_page_dirty(struct page *page)
 {
 	trace_f2fs_set_page_dirty(page, NODE);
 
-	SetPageUptodate(page);
+	if (!PageUptodate(page))
+		SetPageUptodate(page);
 	if (!PageDirty(page)) {
-		__set_page_dirty_nobuffers(page);
+		f2fs_set_page_dirty_nobuffers(page);
 		inc_page_count(F2FS_P_SB(page), F2FS_DIRTY_NODES);
 		SetPagePrivate(page);
 		f2fs_trace_pid(page);
@@ -1409,6 +1753,9 @@ const struct address_space_operations f2fs_node_aops = {
 	.set_page_dirty	= f2fs_set_node_page_dirty,
 	.invalidatepage	= f2fs_invalidate_page,
 	.releasepage	= f2fs_release_page,
+#ifdef CONFIG_MIGRATION
+	.migratepage    = f2fs_migrate_page,
+#endif
 };
 
 static struct free_nid *__lookup_free_nid_list(struct f2fs_nm_info *nm_i,
@@ -1417,121 +1764,251 @@ static struct free_nid *__lookup_free_nid_list(struct f2fs_nm_info *nm_i,
 	return radix_tree_lookup(&nm_i->free_nid_root, n);
 }
 
-static void __del_from_free_nid_list(struct f2fs_nm_info *nm_i,
-						struct free_nid *i)
+static int __insert_nid_to_list(struct f2fs_sb_info *sbi,
+			struct free_nid *i, enum nid_list list, bool new)
 {
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+
+	if (new) {
+		int err = radix_tree_insert(&nm_i->free_nid_root, i->nid, i);
+		if (err)
+			return err;
+	}
+
+	f2fs_bug_on(sbi, list == FREE_NID_LIST ? i->state != NID_NEW :
+						i->state != NID_ALLOC);
+	nm_i->nid_cnt[list]++;
+	list_add_tail(&i->list, &nm_i->nid_list[list]);
+	return 0;
+}
+
+static void __remove_nid_from_list(struct f2fs_sb_info *sbi,
+			struct free_nid *i, enum nid_list list, bool reuse)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+
+	f2fs_bug_on(sbi, list == FREE_NID_LIST ? i->state != NID_NEW :
+						i->state != NID_ALLOC);
+	nm_i->nid_cnt[list]--;
 	list_del(&i->list);
-	radix_tree_delete(&nm_i->free_nid_root, i->nid);
+	if (!reuse)
+		radix_tree_delete(&nm_i->free_nid_root, i->nid);
 }
 
-static int add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
+/* return if the nid is recognized as free */
+static bool add_free_nid(struct f2fs_sb_info *sbi, nid_t nid, bool build)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
-	struct free_nid *i;
+	struct free_nid *i, *e;
 	struct nat_entry *ne;
-	bool allocated = false;
-
-	if (!available_free_memory(sbi, FREE_NIDS))
-		return -1;
+	int err = -EINVAL;
+	bool ret = false;
 
 	/* 0 nid should not be used */
 	if (unlikely(nid == 0))
-		return 0;
-
-	if (build) {
-		/* do not add allocated nids */
-		down_read(&nm_i->nat_tree_lock);
-		ne = __lookup_nat_cache(nm_i, nid);
-		if (ne &&
-			(!get_nat_flag(ne, IS_CHECKPOINTED) ||
-				nat_get_blkaddr(ne) != NULL_ADDR))
-			allocated = true;
-		up_read(&nm_i->nat_tree_lock);
-		if (allocated)
-			return 0;
-	}
+		return false;
 
 	i = f2fs_kmem_cache_alloc(free_nid_slab, GFP_NOFS);
 	i->nid = nid;
 	i->state = NID_NEW;
 
-	if (radix_tree_preload(GFP_NOFS)) {
-		kmem_cache_free(free_nid_slab, i);
-		return 0;
-	}
+	if (radix_tree_preload(GFP_NOFS))
+		goto err;
 
-	spin_lock(&nm_i->free_nid_list_lock);
-	if (radix_tree_insert(&nm_i->free_nid_root, i->nid, i)) {
-		spin_unlock(&nm_i->free_nid_list_lock);
-		radix_tree_preload_end();
-		kmem_cache_free(free_nid_slab, i);
-		return 0;
+	spin_lock(&nm_i->nid_list_lock);
+
+	if (build) {
+		/*
+		 *   Thread A             Thread B
+		 *  - f2fs_create
+		 *   - f2fs_new_inode
+		 *    - alloc_nid
+		 *     - __insert_nid_to_list(ALLOC_NID_LIST)
+		 *                     - f2fs_balance_fs_bg
+		 *                      - build_free_nids
+		 *                       - __build_free_nids
+		 *                        - scan_nat_page
+		 *                         - add_free_nid
+		 *                          - __lookup_nat_cache
+		 *  - f2fs_add_link
+		 *   - init_inode_metadata
+		 *    - new_inode_page
+		 *     - new_node_page
+		 *      - set_node_addr
+		 *  - alloc_nid_done
+		 *   - __remove_nid_from_list(ALLOC_NID_LIST)
+		 *                         - __insert_nid_to_list(FREE_NID_LIST)
+		 */
+		ne = __lookup_nat_cache(nm_i, nid);
+		if (ne && (!get_nat_flag(ne, IS_CHECKPOINTED) ||
+				nat_get_blkaddr(ne) != NULL_ADDR))
+			goto err_out;
+
+		e = __lookup_free_nid_list(nm_i, nid);
+		if (e) {
+			if (e->state == NID_NEW)
+				ret = true;
+			goto err_out;
+		}
 	}
-	list_add_tail(&i->list, &nm_i->free_nid_list);
-	nm_i->fcnt++;
-	spin_unlock(&nm_i->free_nid_list_lock);
+	ret = true;
+	err = __insert_nid_to_list(sbi, i, FREE_NID_LIST, true);
+err_out:
+	spin_unlock(&nm_i->nid_list_lock);
 	radix_tree_preload_end();
-	return 1;
+err:
+	if (err)
+		kmem_cache_free(free_nid_slab, i);
+	return ret;
 }
 
-static void remove_free_nid(struct f2fs_nm_info *nm_i, nid_t nid)
+static void remove_free_nid(struct f2fs_sb_info *sbi, nid_t nid)
 {
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct free_nid *i;
 	bool need_free = false;
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
 	i = __lookup_free_nid_list(nm_i, nid);
 	if (i && i->state == NID_NEW) {
-		__del_from_free_nid_list(nm_i, i);
-		nm_i->fcnt--;
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, false);
 		need_free = true;
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+	spin_unlock(&nm_i->nid_list_lock);
 
 	if (need_free)
 		kmem_cache_free(free_nid_slab, i);
 }
 
+static void update_free_nid_bitmap(struct f2fs_sb_info *sbi, nid_t nid,
+							bool set, bool build)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	unsigned int nat_ofs = NAT_BLOCK_OFFSET(nid);
+	unsigned int nid_ofs = nid - START_NID(nid);
+
+	if (!test_bit_le(nat_ofs, nm_i->nat_block_bitmap))
+		return;
+
+	if (set)
+		__set_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]);
+	else
+		__clear_bit_le(nid_ofs, nm_i->free_nid_bitmap[nat_ofs]);
+
+	if (set)
+		nm_i->free_nid_count[nat_ofs]++;
+	else if (!build)
+		nm_i->free_nid_count[nat_ofs]--;
+}
+
 static void scan_nat_page(struct f2fs_sb_info *sbi,
 			struct page *nat_page, nid_t start_nid)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct f2fs_nat_block *nat_blk = page_address(nat_page);
 	block_t blk_addr;
+	unsigned int nat_ofs = NAT_BLOCK_OFFSET(start_nid);
 	int i;
 
+	if (test_bit_le(nat_ofs, nm_i->nat_block_bitmap))
+		return;
+
+	__set_bit_le(nat_ofs, nm_i->nat_block_bitmap);
+
 	i = start_nid % NAT_ENTRY_PER_BLOCK;
 
 	for (; i < NAT_ENTRY_PER_BLOCK; i++, start_nid++) {
+		bool freed = false;
 
 		if (unlikely(start_nid >= nm_i->max_nid))
 			break;
 
 		blk_addr = le32_to_cpu(nat_blk->entries[i].block_addr);
 		f2fs_bug_on(sbi, blk_addr == NEW_ADDR);
-		if (blk_addr == NULL_ADDR) {
-			if (add_free_nid(sbi, start_nid, true) < 0)
-				break;
+		if (blk_addr == NULL_ADDR)
+			freed = add_free_nid(sbi, start_nid, true);
+		spin_lock(&NM_I(sbi)->nid_list_lock);
+		update_free_nid_bitmap(sbi, start_nid, freed, true);
+		spin_unlock(&NM_I(sbi)->nid_list_lock);
+	}
+}
+
+static void scan_free_nid_bits(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
+	struct f2fs_journal *journal = curseg->journal;
+	unsigned int i, idx;
+
+	down_read(&nm_i->nat_tree_lock);
+
+	for (i = 0; i < nm_i->nat_blocks; i++) {
+		if (!test_bit_le(i, nm_i->nat_block_bitmap))
+			continue;
+		if (!nm_i->free_nid_count[i])
+			continue;
+		for (idx = 0; idx < NAT_ENTRY_PER_BLOCK; idx++) {
+			nid_t nid;
+
+			if (!test_bit_le(idx, nm_i->free_nid_bitmap[i]))
+				continue;
+
+			nid = i * NAT_ENTRY_PER_BLOCK + idx;
+			add_free_nid(sbi, nid, true);
+
+			if (nm_i->nid_cnt[FREE_NID_LIST] >= MAX_FREE_NIDS)
+				goto out;
 		}
 	}
+out:
+	down_read(&curseg->journal_rwsem);
+	for (i = 0; i < nats_in_cursum(journal); i++) {
+		block_t addr;
+		nid_t nid;
+
+		addr = le32_to_cpu(nat_in_journal(journal, i).block_addr);
+		nid = le32_to_cpu(nid_in_journal(journal, i));
+		if (addr == NULL_ADDR)
+			add_free_nid(sbi, nid, true);
+		else
+			remove_free_nid(sbi, nid);
+	}
+	up_read(&curseg->journal_rwsem);
+	up_read(&nm_i->nat_tree_lock);
 }
 
-static void build_free_nids(struct f2fs_sb_info *sbi)
+static void __build_free_nids(struct f2fs_sb_info *sbi, bool sync, bool mount)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
 	int i = 0;
 	nid_t nid = nm_i->next_scan_nid;
 
+	if (unlikely(nid >= nm_i->max_nid))
+		nid = 0;
+
 	/* Enough entries */
-	if (nm_i->fcnt > NAT_ENTRY_PER_BLOCK)
+	if (nm_i->nid_cnt[FREE_NID_LIST] >= NAT_ENTRY_PER_BLOCK)
 		return;
 
+	if (!sync && !available_free_memory(sbi, FREE_NIDS))
+		return;
+
+	if (!mount) {
+		/* try to find free nids in free_nid_bitmap */
+		scan_free_nid_bits(sbi);
+
+		if (nm_i->nid_cnt[FREE_NID_LIST])
+			return;
+	}
+
 	/* readahead nat pages to be scanned */
 	ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nid), FREE_NID_PAGES,
 							META_NAT, true);
 
+	down_read(&nm_i->nat_tree_lock);
+
 	while (1) {
 		struct page *page = get_current_nat_page(sbi, nid);
 
@@ -1550,21 +2027,31 @@ static void build_free_nids(struct f2fs_sb_info *sbi)
 	nm_i->next_scan_nid = nid;
 
 	/* find free nids from current sum_pages */
-	mutex_lock(&curseg->curseg_mutex);
-	for (i = 0; i < nats_in_cursum(sum); i++) {
-		block_t addr = le32_to_cpu(nat_in_journal(sum, i).block_addr);
-		nid = le32_to_cpu(nid_in_journal(sum, i));
+	down_read(&curseg->journal_rwsem);
+	for (i = 0; i < nats_in_cursum(journal); i++) {
+		block_t addr;
+
+		addr = le32_to_cpu(nat_in_journal(journal, i).block_addr);
+		nid = le32_to_cpu(nid_in_journal(journal, i));
 		if (addr == NULL_ADDR)
 			add_free_nid(sbi, nid, true);
 		else
-			remove_free_nid(nm_i, nid);
+			remove_free_nid(sbi, nid);
 	}
-	mutex_unlock(&curseg->curseg_mutex);
+	up_read(&curseg->journal_rwsem);
+	up_read(&nm_i->nat_tree_lock);
 
 	ra_meta_pages(sbi, NAT_BLOCK_OFFSET(nm_i->next_scan_nid),
 					nm_i->ra_nid_pages, META_NAT, false);
 }
 
+void build_free_nids(struct f2fs_sb_info *sbi, bool sync, bool mount)
+{
+	mutex_lock(&NM_I(sbi)->build_lock);
+	__build_free_nids(sbi, sync, mount);
+	mutex_unlock(&NM_I(sbi)->build_lock);
+}
+
 /*
  * If this function returns success, caller can obtain a new nid
  * from second parameter of this function.
@@ -1575,40 +2062,40 @@ bool alloc_nid(struct f2fs_sb_info *sbi, nid_t *nid)
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct free_nid *i = NULL;
 retry:
-	if (unlikely(sbi->total_valid_node_count + 1 > nm_i->available_nids))
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_ALLOC_NID)) {
+		f2fs_show_injection_info(FAULT_ALLOC_NID);
 		return false;
+	}
+#endif
+	spin_lock(&nm_i->nid_list_lock);
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	if (unlikely(nm_i->available_nids == 0)) {
+		spin_unlock(&nm_i->nid_list_lock);
+		return false;
+	}
 
 	/* We should not use stale free nids created by build_free_nids */
-	if (nm_i->fcnt && !on_build_free_nids(nm_i)) {
-		struct node_info ni;
-
-		f2fs_bug_on(sbi, list_empty(&nm_i->free_nid_list));
-		list_for_each_entry(i, &nm_i->free_nid_list, list)
-			if (i->state == NID_NEW)
-				break;
-
-		f2fs_bug_on(sbi, i->state != NID_NEW);
+	if (nm_i->nid_cnt[FREE_NID_LIST] && !on_build_free_nids(nm_i)) {
+		f2fs_bug_on(sbi, list_empty(&nm_i->nid_list[FREE_NID_LIST]));
+		i = list_first_entry(&nm_i->nid_list[FREE_NID_LIST],
+					struct free_nid, list);
 		*nid = i->nid;
+
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, true);
 		i->state = NID_ALLOC;
-		nm_i->fcnt--;
-		spin_unlock(&nm_i->free_nid_list_lock);
-
-		/* check nid is allocated already */
-		get_node_info(sbi, *nid, &ni);
-		if (ni.blk_addr != NULL_ADDR) {
-			alloc_nid_done(sbi, *nid);
-			goto retry;
-		}
+		__insert_nid_to_list(sbi, i, ALLOC_NID_LIST, false);
+		nm_i->available_nids--;
+
+		update_free_nid_bitmap(sbi, *nid, false, false);
+
+		spin_unlock(&nm_i->nid_list_lock);
 		return true;
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+	spin_unlock(&nm_i->nid_list_lock);
 
 	/* Let's scan nat pages and its caches to get free nids */
-	mutex_lock(&nm_i->build_lock);
-	build_free_nids(sbi);
-	mutex_unlock(&nm_i->build_lock);
+	build_free_nids(sbi, true, false);
 	goto retry;
 }
 
@@ -1620,11 +2107,11 @@ void alloc_nid_done(struct f2fs_sb_info *sbi, nid_t nid)
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct free_nid *i;
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
 	i = __lookup_free_nid_list(nm_i, nid);
-	f2fs_bug_on(sbi, !i || i->state != NID_ALLOC);
-	__del_from_free_nid_list(nm_i, i);
-	spin_unlock(&nm_i->free_nid_list_lock);
+	f2fs_bug_on(sbi, !i);
+	__remove_nid_from_list(sbi, i, ALLOC_NID_LIST, false);
+	spin_unlock(&nm_i->nid_list_lock);
 
 	kmem_cache_free(free_nid_slab, i);
 }
@@ -1641,17 +2128,24 @@ void alloc_nid_failed(struct f2fs_sb_info *sbi, nid_t nid)
 	if (!nid)
 		return;
 
-	spin_lock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
 	i = __lookup_free_nid_list(nm_i, nid);
-	f2fs_bug_on(sbi, !i || i->state != NID_ALLOC);
+	f2fs_bug_on(sbi, !i);
+
 	if (!available_free_memory(sbi, FREE_NIDS)) {
-		__del_from_free_nid_list(nm_i, i);
+		__remove_nid_from_list(sbi, i, ALLOC_NID_LIST, false);
 		need_free = true;
 	} else {
+		__remove_nid_from_list(sbi, i, ALLOC_NID_LIST, true);
 		i->state = NID_NEW;
-		nm_i->fcnt++;
+		__insert_nid_to_list(sbi, i, FREE_NID_LIST, false);
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+
+	nm_i->available_nids++;
+
+	update_free_nid_bitmap(sbi, nid, true, false);
+
+	spin_unlock(&nm_i->nid_list_lock);
 
 	if (need_free)
 		kmem_cache_free(free_nid_slab, i);
@@ -1663,21 +2157,24 @@ int try_to_free_nids(struct f2fs_sb_info *sbi, int nr_shrink)
 	struct free_nid *i, *next;
 	int nr = nr_shrink;
 
+	if (nm_i->nid_cnt[FREE_NID_LIST] <= MAX_FREE_NIDS)
+		return 0;
+
 	if (!mutex_trylock(&nm_i->build_lock))
 		return 0;
 
-	spin_lock(&nm_i->free_nid_list_lock);
-	list_for_each_entry_safe(i, next, &nm_i->free_nid_list, list) {
-		if (nr_shrink <= 0 || nm_i->fcnt <= NAT_ENTRY_PER_BLOCK)
+	spin_lock(&nm_i->nid_list_lock);
+	list_for_each_entry_safe(i, next, &nm_i->nid_list[FREE_NID_LIST],
+									list) {
+		if (nr_shrink <= 0 ||
+				nm_i->nid_cnt[FREE_NID_LIST] <= MAX_FREE_NIDS)
 			break;
-		if (i->state == NID_ALLOC)
-			continue;
-		__del_from_free_nid_list(nm_i, i);
+
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, false);
 		kmem_cache_free(free_nid_slab, i);
-		nm_i->fcnt--;
 		nr_shrink--;
 	}
-	spin_unlock(&nm_i->free_nid_list_lock);
+	spin_unlock(&nm_i->nid_list_lock);
 	mutex_unlock(&nm_i->build_lock);
 
 	return nr - nr_shrink;
@@ -1695,7 +2192,7 @@ void recover_inline_xattr(struct inode *inode, struct page *page)
 
 	ri = F2FS_INODE(page);
 	if (!(ri->i_inline & F2FS_INLINE_XATTR)) {
-		clear_inode_flag(F2FS_I(inode), FI_INLINE_XATTR);
+		clear_inode_flag(inode, FI_INLINE_XATTR);
 		goto update_inode;
 	}
 
@@ -1703,47 +2200,54 @@ void recover_inline_xattr(struct inode *inode, struct page *page)
 	src_addr = inline_xattr_addr(page);
 	inline_size = inline_xattr_size(inode);
 
-	f2fs_wait_on_page_writeback(ipage, NODE);
+	f2fs_wait_on_page_writeback(ipage, NODE, true);
 	memcpy(dst_addr, src_addr, inline_size);
 update_inode:
 	update_inode(inode, ipage);
 	f2fs_put_page(ipage, 1);
 }
 
-void recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
+int recover_xattr_data(struct inode *inode, struct page *page, block_t blkaddr)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	nid_t prev_xnid = F2FS_I(inode)->i_xattr_nid;
-	nid_t new_xnid = nid_of_node(page);
+	nid_t new_xnid;
+	struct dnode_of_data dn;
 	struct node_info ni;
+	struct page *xpage;
 
-	/* 1: invalidate the previous xattr nid */
 	if (!prev_xnid)
 		goto recover_xnid;
 
-	/* Deallocate node address */
+	/* 1: invalidate the previous xattr nid */
 	get_node_info(sbi, prev_xnid, &ni);
 	f2fs_bug_on(sbi, ni.blk_addr == NULL_ADDR);
 	invalidate_blocks(sbi, ni.blk_addr);
-	dec_valid_node_count(sbi, inode);
+	dec_valid_node_count(sbi, inode, false);
 	set_node_addr(sbi, &ni, NULL_ADDR, false);
 
 recover_xnid:
-	/* 2: allocate new xattr nid */
-	if (unlikely(!inc_valid_node_count(sbi, inode)))
-		f2fs_bug_on(sbi, 1);
-
-	remove_free_nid(NM_I(sbi), new_xnid);
-	get_node_info(sbi, new_xnid, &ni);
-	ni.ino = inode->i_ino;
-	set_node_addr(sbi, &ni, NEW_ADDR, false);
-	F2FS_I(inode)->i_xattr_nid = new_xnid;
+	/* 2: update xattr nid in inode */
+	if (!alloc_nid(sbi, &new_xnid))
+		return -ENOSPC;
 
-	/* 3: update xattr blkaddr */
-	refresh_sit_entry(sbi, NEW_ADDR, blkaddr);
-	set_node_addr(sbi, &ni, blkaddr, false);
+	set_new_dnode(&dn, inode, NULL, NULL, new_xnid);
+	xpage = new_node_page(&dn, XATTR_NODE_OFFSET);
+	if (IS_ERR(xpage)) {
+		alloc_nid_failed(sbi, new_xnid);
+		return PTR_ERR(xpage);
+	}
 
+	alloc_nid_done(sbi, new_xnid);
 	update_inode_page(inode);
+
+	/* 3: update and set xattr node page dirty */
+	memcpy(F2FS_NODE(xpage), F2FS_NODE(page), VALID_XATTR_BLOCK_SIZE);
+
+	set_page_dirty(xpage);
+	f2fs_put_page(xpage, 1);
+
+	return 0;
 }
 
 int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
@@ -1757,15 +2261,18 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 
 	if (unlikely(old_ni.blk_addr != NULL_ADDR))
 		return -EINVAL;
-
-	ipage = grab_cache_page(NODE_MAPPING(sbi), ino);
-	if (!ipage)
-		return -ENOMEM;
+retry:
+	ipage = f2fs_grab_cache_page(NODE_MAPPING(sbi), ino, false);
+	if (!ipage) {
+		congestion_wait(BLK_RW_ASYNC, HZ/50);
+		goto retry;
+	}
 
 	/* Should not use this inode from free nid list */
-	remove_free_nid(NM_I(sbi), ino);
+	remove_free_nid(sbi, ino);
 
-	SetPageUptodate(ipage);
+	if (!PageUptodate(ipage))
+		SetPageUptodate(ipage);
 	fill_node_footer(ipage, ino, ino, 0, true);
 
 	src = F2FS_INODE(page);
@@ -1776,12 +2283,19 @@ int recover_inode_page(struct f2fs_sb_info *sbi, struct page *page)
 	dst->i_blocks = cpu_to_le64(1);
 	dst->i_links = cpu_to_le32(1);
 	dst->i_xattr_nid = 0;
-	dst->i_inline = src->i_inline & F2FS_INLINE_XATTR;
+	dst->i_inline = src->i_inline & (F2FS_INLINE_XATTR | F2FS_EXTRA_ATTR);
+	if (dst->i_inline & F2FS_EXTRA_ATTR) {
+		dst->i_extra_isize = src->i_extra_isize;
+		if (f2fs_sb_has_project_quota(sbi->sb) &&
+			F2FS_FITS_IN_INODE(src, le16_to_cpu(src->i_extra_isize),
+								i_projid))
+			dst->i_projid = src->i_projid;
+	}
 
 	new_ni = old_ni;
 	new_ni.ino = ino;
 
-	if (unlikely(!inc_valid_node_count(sbi, NULL)))
+	if (unlikely(inc_valid_node_count(sbi, NULL, true)))
 		WARN_ON(1);
 	set_node_addr(sbi, &new_ni, NEW_ADDR, false);
 	inc_valid_inode_count(sbi);
@@ -1796,7 +2310,6 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
 	struct f2fs_node *rn;
 	struct f2fs_summary *sum_entry;
 	block_t addr;
-	int bio_blocks = MAX_BIO_BLOCKS(sbi);
 	int i, idx, last_offset, nrpages;
 
 	/* scan the node segment */
@@ -1805,7 +2318,7 @@ int restore_node_summary(struct f2fs_sb_info *sbi,
 	sum_entry = &sum->entries[0];
 
 	for (i = 0; i < last_offset; i += nrpages, addr += nrpages) {
-		nrpages = min(last_offset - i, bio_blocks);
+		nrpages = min(last_offset - i, BIO_MAX_PAGES);
 
 		/* readahead node pages */
 		ra_meta_pages(sbi, addr, nrpages, META_POR, true);
@@ -1831,28 +2344,39 @@ static void remove_nats_in_journal(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
 	int i;
 
-	mutex_lock(&curseg->curseg_mutex);
-	for (i = 0; i < nats_in_cursum(sum); i++) {
+	down_write(&curseg->journal_rwsem);
+	for (i = 0; i < nats_in_cursum(journal); i++) {
 		struct nat_entry *ne;
 		struct f2fs_nat_entry raw_ne;
-		nid_t nid = le32_to_cpu(nid_in_journal(sum, i));
+		nid_t nid = le32_to_cpu(nid_in_journal(journal, i));
 
-		raw_ne = nat_in_journal(sum, i);
+		raw_ne = nat_in_journal(journal, i);
 
-		down_write(&nm_i->nat_tree_lock);
 		ne = __lookup_nat_cache(nm_i, nid);
 		if (!ne) {
-			ne = grab_nat_entry(nm_i, nid);
+			ne = grab_nat_entry(nm_i, nid, true);
 			node_info_from_raw_nat(&ne->ni, &raw_ne);
 		}
+
+		/*
+		 * if a free nat in journal has not been used after last
+		 * checkpoint, we should remove it from available nids,
+		 * since later we will add it again.
+		 */
+		if (!get_nat_flag(ne, IS_DIRTY) &&
+				le32_to_cpu(raw_ne.block_addr) == NULL_ADDR) {
+			spin_lock(&nm_i->nid_list_lock);
+			nm_i->available_nids--;
+			spin_unlock(&nm_i->nid_list_lock);
+		}
+
 		__set_nat_cache_dirty(nm_i, ne);
-		up_write(&nm_i->nat_tree_lock);
 	}
-	update_nats_in_cursum(sum, -i);
-	mutex_unlock(&curseg->curseg_mutex);
+	update_nats_in_cursum(journal, -i);
+	up_write(&curseg->journal_rwsem);
 }
 
 static void __adjust_nat_entry_set(struct nat_entry_set *nes,
@@ -1873,28 +2397,59 @@ add_out:
 	list_add_tail(&nes->set_list, head);
 }
 
+static void __update_nat_bits(struct f2fs_sb_info *sbi, nid_t start_nid,
+						struct page *page)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	unsigned int nat_index = start_nid / NAT_ENTRY_PER_BLOCK;
+	struct f2fs_nat_block *nat_blk = page_address(page);
+	int valid = 0;
+	int i;
+
+	if (!enabled_nat_bits(sbi, NULL))
+		return;
+
+	for (i = 0; i < NAT_ENTRY_PER_BLOCK; i++) {
+		if (start_nid == 0 && i == 0)
+			valid++;
+		if (nat_blk->entries[i].block_addr)
+			valid++;
+	}
+	if (valid == 0) {
+		__set_bit_le(nat_index, nm_i->empty_nat_bits);
+		__clear_bit_le(nat_index, nm_i->full_nat_bits);
+		return;
+	}
+
+	__clear_bit_le(nat_index, nm_i->empty_nat_bits);
+	if (valid == NAT_ENTRY_PER_BLOCK)
+		__set_bit_le(nat_index, nm_i->full_nat_bits);
+	else
+		__clear_bit_le(nat_index, nm_i->full_nat_bits);
+}
+
 static void __flush_nat_entry_set(struct f2fs_sb_info *sbi,
-					struct nat_entry_set *set)
+		struct nat_entry_set *set, struct cp_control *cpc)
 {
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
 	nid_t start_nid = set->set * NAT_ENTRY_PER_BLOCK;
 	bool to_journal = true;
 	struct f2fs_nat_block *nat_blk;
 	struct nat_entry *ne, *cur;
 	struct page *page = NULL;
-	struct f2fs_nm_info *nm_i = NM_I(sbi);
 
 	/*
 	 * there are two steps to flush nat entries:
 	 * #1, flush nat entries to journal in current hot data summary block.
 	 * #2, flush nat entries to nat page.
 	 */
-	if (!__has_cursum_space(sum, set->entry_cnt, NAT_JOURNAL))
+	if (enabled_nat_bits(sbi, cpc) ||
+		!__has_cursum_space(journal, set->entry_cnt, NAT_JOURNAL))
 		to_journal = false;
 
 	if (to_journal) {
-		mutex_lock(&curseg->curseg_mutex);
+		down_write(&curseg->journal_rwsem);
 	} else {
 		page = get_next_nat_page(sbi, start_nid);
 		nat_blk = page_address(page);
@@ -1907,50 +2462,55 @@ static void __flush_nat_entry_set(struct f2fs_sb_info *sbi,
 		nid_t nid = nat_get_nid(ne);
 		int offset;
 
-		if (nat_get_blkaddr(ne) == NEW_ADDR)
-			continue;
+		f2fs_bug_on(sbi, nat_get_blkaddr(ne) == NEW_ADDR);
 
 		if (to_journal) {
-			offset = lookup_journal_in_cursum(sum,
+			offset = lookup_journal_in_cursum(journal,
 							NAT_JOURNAL, nid, 1);
 			f2fs_bug_on(sbi, offset < 0);
-			raw_ne = &nat_in_journal(sum, offset);
-			nid_in_journal(sum, offset) = cpu_to_le32(nid);
+			raw_ne = &nat_in_journal(journal, offset);
+			nid_in_journal(journal, offset) = cpu_to_le32(nid);
 		} else {
 			raw_ne = &nat_blk->entries[nid - start_nid];
 		}
 		raw_nat_from_node_info(raw_ne, &ne->ni);
-
-		down_write(&NM_I(sbi)->nat_tree_lock);
 		nat_reset_flag(ne);
-		__clear_nat_cache_dirty(NM_I(sbi), ne);
-		up_write(&NM_I(sbi)->nat_tree_lock);
-
-		if (nat_get_blkaddr(ne) == NULL_ADDR)
+		__clear_nat_cache_dirty(NM_I(sbi), set, ne);
+		if (nat_get_blkaddr(ne) == NULL_ADDR) {
 			add_free_nid(sbi, nid, false);
+			spin_lock(&NM_I(sbi)->nid_list_lock);
+			NM_I(sbi)->available_nids++;
+			update_free_nid_bitmap(sbi, nid, true, false);
+			spin_unlock(&NM_I(sbi)->nid_list_lock);
+		} else {
+			spin_lock(&NM_I(sbi)->nid_list_lock);
+			update_free_nid_bitmap(sbi, nid, false, false);
+			spin_unlock(&NM_I(sbi)->nid_list_lock);
+		}
 	}
 
-	if (to_journal)
-		mutex_unlock(&curseg->curseg_mutex);
-	else
+	if (to_journal) {
+		up_write(&curseg->journal_rwsem);
+	} else {
+		__update_nat_bits(sbi, start_nid, page);
 		f2fs_put_page(page, 1);
+	}
 
-	f2fs_bug_on(sbi, set->entry_cnt);
-
-	down_write(&nm_i->nat_tree_lock);
-	radix_tree_delete(&NM_I(sbi)->nat_set_root, set->set);
-	up_write(&nm_i->nat_tree_lock);
-	kmem_cache_free(nat_entry_set_slab, set);
+	/* Allow dirty nats by node block allocation in write_begin */
+	if (!set->entry_cnt) {
+		radix_tree_delete(&NM_I(sbi)->nat_set_root, set->set);
+		kmem_cache_free(nat_entry_set_slab, set);
+	}
 }
 
 /*
  * This function is called during the checkpointing process.
  */
-void flush_nat_entries(struct f2fs_sb_info *sbi)
+void flush_nat_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
 	struct nat_entry_set *setvec[SETVEC_SIZE];
 	struct nat_entry_set *set, *tmp;
 	unsigned int found;
@@ -1959,30 +2519,109 @@ void flush_nat_entries(struct f2fs_sb_info *sbi)
 
 	if (!nm_i->dirty_nat_cnt)
 		return;
+
+	down_write(&nm_i->nat_tree_lock);
+
 	/*
 	 * if there are no enough space in journal to store dirty nat
 	 * entries, remove all entries from journal and merge them
 	 * into nat entry set.
 	 */
-	if (!__has_cursum_space(sum, nm_i->dirty_nat_cnt, NAT_JOURNAL))
+	if (enabled_nat_bits(sbi, cpc) ||
+		!__has_cursum_space(journal, nm_i->dirty_nat_cnt, NAT_JOURNAL))
 		remove_nats_in_journal(sbi);
 
-	down_write(&nm_i->nat_tree_lock);
 	while ((found = __gang_lookup_nat_set(nm_i,
 					set_idx, SETVEC_SIZE, setvec))) {
 		unsigned idx;
 		set_idx = setvec[found - 1]->set + 1;
 		for (idx = 0; idx < found; idx++)
 			__adjust_nat_entry_set(setvec[idx], &sets,
-							MAX_NAT_JENTRIES(sum));
+						MAX_NAT_JENTRIES(journal));
 	}
-	up_write(&nm_i->nat_tree_lock);
 
 	/* flush dirty nats in nat entry set */
 	list_for_each_entry_safe(set, tmp, &sets, set_list)
-		__flush_nat_entry_set(sbi, set);
+		__flush_nat_entry_set(sbi, set, cpc);
+
+	up_write(&nm_i->nat_tree_lock);
+	/* Allow dirty nats by node block allocation in write_begin */
+}
+
+static int __get_nat_bitmaps(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	unsigned int nat_bits_bytes = nm_i->nat_blocks / BITS_PER_BYTE;
+	unsigned int i;
+	__u64 cp_ver = cur_cp_version(ckpt);
+	block_t nat_bits_addr;
+
+	if (!enabled_nat_bits(sbi, NULL))
+		return 0;
+
+	nm_i->nat_bits_blocks = F2FS_BYTES_TO_BLK((nat_bits_bytes << 1) + 8 +
+						F2FS_BLKSIZE - 1);
+	nm_i->nat_bits = kzalloc(nm_i->nat_bits_blocks << F2FS_BLKSIZE_BITS,
+						GFP_KERNEL);
+	if (!nm_i->nat_bits)
+		return -ENOMEM;
+
+	nat_bits_addr = __start_cp_addr(sbi) + sbi->blocks_per_seg -
+						nm_i->nat_bits_blocks;
+	for (i = 0; i < nm_i->nat_bits_blocks; i++) {
+		struct page *page = get_meta_page(sbi, nat_bits_addr++);
+
+		memcpy(nm_i->nat_bits + (i << F2FS_BLKSIZE_BITS),
+					page_address(page), F2FS_BLKSIZE);
+		f2fs_put_page(page, 1);
+	}
+
+	cp_ver |= (cur_cp_crc(ckpt) << 32);
+	if (cpu_to_le64(cp_ver) != *(__le64 *)nm_i->nat_bits) {
+		disable_nat_bits(sbi, true);
+		return 0;
+	}
+
+	nm_i->full_nat_bits = nm_i->nat_bits + 8;
+	nm_i->empty_nat_bits = nm_i->full_nat_bits + nat_bits_bytes;
+
+	f2fs_msg(sbi->sb, KERN_NOTICE, "Found nat_bits in checkpoint");
+	return 0;
+}
 
-	f2fs_bug_on(sbi, nm_i->dirty_nat_cnt);
+static inline void load_free_nid_bitmap(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+	unsigned int i = 0;
+	nid_t nid, last_nid;
+
+	if (!enabled_nat_bits(sbi, NULL))
+		return;
+
+	for (i = 0; i < nm_i->nat_blocks; i++) {
+		i = find_next_bit_le(nm_i->empty_nat_bits, nm_i->nat_blocks, i);
+		if (i >= nm_i->nat_blocks)
+			break;
+
+		__set_bit_le(i, nm_i->nat_block_bitmap);
+
+		nid = i * NAT_ENTRY_PER_BLOCK;
+		last_nid = (i + 1) * NAT_ENTRY_PER_BLOCK;
+
+		spin_lock(&NM_I(sbi)->nid_list_lock);
+		for (; nid < last_nid; nid++)
+			update_free_nid_bitmap(sbi, nid, true, true);
+		spin_unlock(&NM_I(sbi)->nid_list_lock);
+	}
+
+	for (i = 0; i < nm_i->nat_blocks; i++) {
+		i = find_next_bit_le(nm_i->full_nat_bits, nm_i->nat_blocks, i);
+		if (i >= nm_i->nat_blocks)
+			break;
+
+		__set_bit_le(i, nm_i->nat_block_bitmap);
+	}
 }
 
 static int init_node_manager(struct f2fs_sb_info *sbi)
@@ -1990,31 +2629,35 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
 	struct f2fs_super_block *sb_raw = F2FS_RAW_SUPER(sbi);
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	unsigned char *version_bitmap;
-	unsigned int nat_segs, nat_blocks;
+	unsigned int nat_segs;
+	int err;
 
 	nm_i->nat_blkaddr = le32_to_cpu(sb_raw->nat_blkaddr);
 
 	/* segment_count_nat includes pair segment so divide to 2. */
 	nat_segs = le32_to_cpu(sb_raw->segment_count_nat) >> 1;
-	nat_blocks = nat_segs << le32_to_cpu(sb_raw->log_blocks_per_seg);
-
-	nm_i->max_nid = NAT_ENTRY_PER_BLOCK * nat_blocks;
+	nm_i->nat_blocks = nat_segs << le32_to_cpu(sb_raw->log_blocks_per_seg);
+	nm_i->max_nid = NAT_ENTRY_PER_BLOCK * nm_i->nat_blocks;
 
 	/* not used nids: 0, node, meta, (and root counted as valid node) */
-	nm_i->available_nids = nm_i->max_nid - F2FS_RESERVED_NODE_NUM;
-	nm_i->fcnt = 0;
+	nm_i->available_nids = nm_i->max_nid - sbi->total_valid_node_count -
+							F2FS_RESERVED_NODE_NUM;
+	nm_i->nid_cnt[FREE_NID_LIST] = 0;
+	nm_i->nid_cnt[ALLOC_NID_LIST] = 0;
 	nm_i->nat_cnt = 0;
 	nm_i->ram_thresh = DEF_RAM_THRESHOLD;
 	nm_i->ra_nid_pages = DEF_RA_NID_PAGES;
+	nm_i->dirty_nats_ratio = DEF_DIRTY_NAT_RATIO_THRESHOLD;
 
 	INIT_RADIX_TREE(&nm_i->free_nid_root, GFP_ATOMIC);
-	INIT_LIST_HEAD(&nm_i->free_nid_list);
+	INIT_LIST_HEAD(&nm_i->nid_list[FREE_NID_LIST]);
+	INIT_LIST_HEAD(&nm_i->nid_list[ALLOC_NID_LIST]);
 	INIT_RADIX_TREE(&nm_i->nat_root, GFP_NOIO);
 	INIT_RADIX_TREE(&nm_i->nat_set_root, GFP_NOIO);
 	INIT_LIST_HEAD(&nm_i->nat_entries);
 
 	mutex_init(&nm_i->build_lock);
-	spin_lock_init(&nm_i->free_nid_list_lock);
+	spin_lock_init(&nm_i->nid_list_lock);
 	init_rwsem(&nm_i->nat_tree_lock);
 
 	nm_i->next_scan_nid = le32_to_cpu(sbi->ckpt->next_free_nid);
@@ -2027,6 +2670,39 @@ static int init_node_manager(struct f2fs_sb_info *sbi)
 					GFP_KERNEL);
 	if (!nm_i->nat_bitmap)
 		return -ENOMEM;
+
+	err = __get_nat_bitmaps(sbi);
+	if (err)
+		return err;
+
+#ifdef CONFIG_F2FS_CHECK_FS
+	nm_i->nat_bitmap_mir = kmemdup(version_bitmap, nm_i->bitmap_size,
+					GFP_KERNEL);
+	if (!nm_i->nat_bitmap_mir)
+		return -ENOMEM;
+#endif
+
+	return 0;
+}
+
+static int init_free_nid_cache(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_nm_info *nm_i = NM_I(sbi);
+
+	nm_i->free_nid_bitmap = kvzalloc(nm_i->nat_blocks *
+					NAT_ENTRY_BITMAP_SIZE, GFP_KERNEL);
+	if (!nm_i->free_nid_bitmap)
+		return -ENOMEM;
+
+	nm_i->nat_block_bitmap = kvzalloc(nm_i->nat_blocks / 8,
+								GFP_KERNEL);
+	if (!nm_i->nat_block_bitmap)
+		return -ENOMEM;
+
+	nm_i->free_nid_count = kvzalloc(nm_i->nat_blocks *
+					sizeof(unsigned short), GFP_KERNEL);
+	if (!nm_i->free_nid_count)
+		return -ENOMEM;
 	return 0;
 }
 
@@ -2042,7 +2718,14 @@ int build_node_manager(struct f2fs_sb_info *sbi)
 	if (err)
 		return err;
 
-	build_free_nids(sbi);
+	err = init_free_nid_cache(sbi);
+	if (err)
+		return err;
+
+	/* load free nid status from nat_bits table */
+	load_free_nid_bitmap(sbi);
+
+	build_free_nids(sbi, true, true);
 	return 0;
 }
 
@@ -2059,17 +2742,18 @@ void destroy_node_manager(struct f2fs_sb_info *sbi)
 		return;
 
 	/* destroy free nid list */
-	spin_lock(&nm_i->free_nid_list_lock);
-	list_for_each_entry_safe(i, next_i, &nm_i->free_nid_list, list) {
-		f2fs_bug_on(sbi, i->state == NID_ALLOC);
-		__del_from_free_nid_list(nm_i, i);
-		nm_i->fcnt--;
-		spin_unlock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
+	list_for_each_entry_safe(i, next_i, &nm_i->nid_list[FREE_NID_LIST],
+									list) {
+		__remove_nid_from_list(sbi, i, FREE_NID_LIST, false);
+		spin_unlock(&nm_i->nid_list_lock);
 		kmem_cache_free(free_nid_slab, i);
-		spin_lock(&nm_i->free_nid_list_lock);
+		spin_lock(&nm_i->nid_list_lock);
 	}
-	f2fs_bug_on(sbi, nm_i->fcnt);
-	spin_unlock(&nm_i->free_nid_list_lock);
+	f2fs_bug_on(sbi, nm_i->nid_cnt[FREE_NID_LIST]);
+	f2fs_bug_on(sbi, nm_i->nid_cnt[ALLOC_NID_LIST]);
+	f2fs_bug_on(sbi, !list_empty(&nm_i->nid_list[ALLOC_NID_LIST]));
+	spin_unlock(&nm_i->nid_list_lock);
 
 	/* destroy nat cache */
 	down_write(&nm_i->nat_tree_lock);
@@ -2099,7 +2783,15 @@ void destroy_node_manager(struct f2fs_sb_info *sbi)
 	}
 	up_write(&nm_i->nat_tree_lock);
 
+	kvfree(nm_i->nat_block_bitmap);
+	kvfree(nm_i->free_nid_bitmap);
+	kvfree(nm_i->free_nid_count);
+
 	kfree(nm_i->nat_bitmap);
+	kfree(nm_i->nat_bits);
+#ifdef CONFIG_F2FS_CHECK_FS
+	kfree(nm_i->nat_bitmap_mir);
+#endif
 	sbi->nm_info = NULL;
 	kfree(nm_i);
 }
diff --git a/fs/f2fs/node.h b/fs/f2fs/node.h
index e4fffd2d98c4..bb53e9955ff2 100644
--- a/fs/f2fs/node.h
+++ b/fs/f2fs/node.h
@@ -9,21 +9,27 @@
  * published by the Free Software Foundation.
  */
 /* start node id of a node block dedicated to the given node id */
-#define	START_NID(nid) ((nid / NAT_ENTRY_PER_BLOCK) * NAT_ENTRY_PER_BLOCK)
+#define	START_NID(nid) (((nid) / NAT_ENTRY_PER_BLOCK) * NAT_ENTRY_PER_BLOCK)
 
 /* node block offset on the NAT area dedicated to the given start node id */
-#define	NAT_BLOCK_OFFSET(start_nid) (start_nid / NAT_ENTRY_PER_BLOCK)
+#define	NAT_BLOCK_OFFSET(start_nid) ((start_nid) / NAT_ENTRY_PER_BLOCK)
 
 /* # of pages to perform synchronous readahead before building free nids */
-#define FREE_NID_PAGES 4
+#define FREE_NID_PAGES	8
+#define MAX_FREE_NIDS	(NAT_ENTRY_PER_BLOCK * FREE_NID_PAGES)
 
-#define DEF_RA_NID_PAGES	4	/* # of nid pages to be readaheaded */
+#define DEF_RA_NID_PAGES	0	/* # of nid pages to be readaheaded */
 
 /* maximum readahead size for node during getting data blocks */
 #define MAX_RA_NODE		128
 
 /* control the memory footprint threshold (10MB per 1GB ram) */
-#define DEF_RAM_THRESHOLD	10
+#define DEF_RAM_THRESHOLD	1
+
+/* control dirty nats ratio threshold (default: 10% over max nid count) */
+#define DEF_DIRTY_NAT_RATIO_THRESHOLD		10
+/* control total # of nats */
+#define DEF_NAT_CACHE_THRESHOLD			100000
 
 /* vector size for gang look-up from nat cache that consists of radix tree */
 #define NATVEC_SIZE	64
@@ -56,16 +62,16 @@ struct nat_entry {
 	struct node_info ni;	/* in-memory node information */
 };
 
-#define nat_get_nid(nat)		(nat->ni.nid)
-#define nat_set_nid(nat, n)		(nat->ni.nid = n)
-#define nat_get_blkaddr(nat)		(nat->ni.blk_addr)
-#define nat_set_blkaddr(nat, b)		(nat->ni.blk_addr = b)
-#define nat_get_ino(nat)		(nat->ni.ino)
-#define nat_set_ino(nat, i)		(nat->ni.ino = i)
-#define nat_get_version(nat)		(nat->ni.version)
-#define nat_set_version(nat, v)		(nat->ni.version = v)
+#define nat_get_nid(nat)		((nat)->ni.nid)
+#define nat_set_nid(nat, n)		((nat)->ni.nid = (n))
+#define nat_get_blkaddr(nat)		((nat)->ni.blk_addr)
+#define nat_set_blkaddr(nat, b)		((nat)->ni.blk_addr = (b))
+#define nat_get_ino(nat)		((nat)->ni.ino)
+#define nat_set_ino(nat, i)		((nat)->ni.ino = (i))
+#define nat_get_version(nat)		((nat)->ni.version)
+#define nat_set_version(nat, v)		((nat)->ni.version = (v))
 
-#define inc_node_version(version)	(++version)
+#define inc_node_version(version)	(++(version))
 
 static inline void copy_node_info(struct node_info *dst,
 						struct node_info *src)
@@ -117,6 +123,17 @@ static inline void raw_nat_from_node_info(struct f2fs_nat_entry *raw_ne,
 	raw_ne->version = ni->version;
 }
 
+static inline bool excess_dirty_nats(struct f2fs_sb_info *sbi)
+{
+	return NM_I(sbi)->dirty_nat_cnt >= NM_I(sbi)->max_nid *
+					NM_I(sbi)->dirty_nats_ratio / 100;
+}
+
+static inline bool excess_cached_nats(struct f2fs_sb_info *sbi)
+{
+	return NM_I(sbi)->nat_cnt >= DEF_NAT_CACHE_THRESHOLD;
+}
+
 enum mem_type {
 	FREE_NIDS,	/* indicates the free nid list */
 	NAT_ENTRIES,	/* indicates the cached nat entry */
@@ -152,14 +169,15 @@ static inline void next_free_nid(struct f2fs_sb_info *sbi, nid_t *nid)
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	struct free_nid *fnid;
 
-	spin_lock(&nm_i->free_nid_list_lock);
-	if (nm_i->fcnt <= 0) {
-		spin_unlock(&nm_i->free_nid_list_lock);
+	spin_lock(&nm_i->nid_list_lock);
+	if (nm_i->nid_cnt[FREE_NID_LIST] <= 0) {
+		spin_unlock(&nm_i->nid_list_lock);
 		return;
 	}
-	fnid = list_entry(nm_i->free_nid_list.next, struct free_nid, list);
+	fnid = list_first_entry(&nm_i->nid_list[FREE_NID_LIST],
+						struct free_nid, list);
 	*nid = fnid->nid;
-	spin_unlock(&nm_i->free_nid_list_lock);
+	spin_unlock(&nm_i->nid_list_lock);
 }
 
 /*
@@ -168,6 +186,12 @@ static inline void next_free_nid(struct f2fs_sb_info *sbi, nid_t *nid)
 static inline void get_nat_bitmap(struct f2fs_sb_info *sbi, void *addr)
 {
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
+
+#ifdef CONFIG_F2FS_CHECK_FS
+	if (memcmp(nm_i->nat_bitmap, nm_i->nat_bitmap_mir,
+						nm_i->bitmap_size))
+		f2fs_bug_on(sbi, 1);
+#endif
 	memcpy(addr, nm_i->nat_bitmap, nm_i->bitmap_size);
 }
 
@@ -176,14 +200,17 @@ static inline pgoff_t current_nat_addr(struct f2fs_sb_info *sbi, nid_t start)
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 	pgoff_t block_off;
 	pgoff_t block_addr;
-	int seg_off;
 
+	/*
+	 * block_off = segment_off * 512 + off_in_segment
+	 * OLD = (segment_off * 512) * 2 + off_in_segment
+	 * NEW = 2 * (segment_off * 512 + off_in_segment) - off_in_segment
+	 */
 	block_off = NAT_BLOCK_OFFSET(start);
-	seg_off = block_off >> sbi->log_blocks_per_seg;
 
 	block_addr = (pgoff_t)(nm_i->nat_blkaddr +
-		(seg_off << sbi->log_blocks_per_seg << 1) +
-		(block_off & ((1 << sbi->log_blocks_per_seg) - 1)));
+		(block_off << 1) -
+		(block_off & (sbi->blocks_per_seg - 1)));
 
 	if (f2fs_test_bit(block_off, nm_i->nat_bitmap))
 		block_addr += sbi->blocks_per_seg;
@@ -197,11 +224,7 @@ static inline pgoff_t next_nat_addr(struct f2fs_sb_info *sbi,
 	struct f2fs_nm_info *nm_i = NM_I(sbi);
 
 	block_addr -= nm_i->nat_blkaddr;
-	if ((block_addr >> sbi->log_blocks_per_seg) % 2)
-		block_addr -= sbi->blocks_per_seg;
-	else
-		block_addr += sbi->blocks_per_seg;
-
+	block_addr ^= 1 << sbi->log_blocks_per_seg;
 	return block_addr + nm_i->nat_blkaddr;
 }
 
@@ -210,6 +233,40 @@ static inline void set_to_next_nat(struct f2fs_nm_info *nm_i, nid_t start_nid)
 	unsigned int block_off = NAT_BLOCK_OFFSET(start_nid);
 
 	f2fs_change_bit(block_off, nm_i->nat_bitmap);
+#ifdef CONFIG_F2FS_CHECK_FS
+	f2fs_change_bit(block_off, nm_i->nat_bitmap_mir);
+#endif
+}
+
+static inline nid_t ino_of_node(struct page *node_page)
+{
+	struct f2fs_node *rn = F2FS_NODE(node_page);
+	return le32_to_cpu(rn->footer.ino);
+}
+
+static inline nid_t nid_of_node(struct page *node_page)
+{
+	struct f2fs_node *rn = F2FS_NODE(node_page);
+	return le32_to_cpu(rn->footer.nid);
+}
+
+static inline unsigned int ofs_of_node(struct page *node_page)
+{
+	struct f2fs_node *rn = F2FS_NODE(node_page);
+	unsigned flag = le32_to_cpu(rn->footer.flag);
+	return flag >> OFFSET_BIT_SHIFT;
+}
+
+static inline __u64 cpver_of_node(struct page *node_page)
+{
+	struct f2fs_node *rn = F2FS_NODE(node_page);
+	return le64_to_cpu(rn->footer.cp_ver);
+}
+
+static inline block_t next_blkaddr_of_node(struct page *node_page)
+{
+	struct f2fs_node *rn = F2FS_NODE(node_page);
+	return le32_to_cpu(rn->footer.next_blkaddr);
 }
 
 static inline void fill_node_footer(struct page *page, nid_t nid,
@@ -242,40 +299,24 @@ static inline void fill_node_footer_blkaddr(struct page *page, block_t blkaddr)
 {
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(F2FS_P_SB(page));
 	struct f2fs_node *rn = F2FS_NODE(page);
+	__u64 cp_ver = cur_cp_version(ckpt);
 
-	rn->footer.cp_ver = ckpt->checkpoint_ver;
-	rn->footer.next_blkaddr = cpu_to_le32(blkaddr);
-}
-
-static inline nid_t ino_of_node(struct page *node_page)
-{
-	struct f2fs_node *rn = F2FS_NODE(node_page);
-	return le32_to_cpu(rn->footer.ino);
-}
+	if (__is_set_ckpt_flags(ckpt, CP_CRC_RECOVERY_FLAG))
+		cp_ver |= (cur_cp_crc(ckpt) << 32);
 
-static inline nid_t nid_of_node(struct page *node_page)
-{
-	struct f2fs_node *rn = F2FS_NODE(node_page);
-	return le32_to_cpu(rn->footer.nid);
+	rn->footer.cp_ver = cpu_to_le64(cp_ver);
+	rn->footer.next_blkaddr = cpu_to_le32(blkaddr);
 }
 
-static inline unsigned int ofs_of_node(struct page *node_page)
+static inline bool is_recoverable_dnode(struct page *page)
 {
-	struct f2fs_node *rn = F2FS_NODE(node_page);
-	unsigned flag = le32_to_cpu(rn->footer.flag);
-	return flag >> OFFSET_BIT_SHIFT;
-}
+	struct f2fs_checkpoint *ckpt = F2FS_CKPT(F2FS_P_SB(page));
+	__u64 cp_ver = cur_cp_version(ckpt);
 
-static inline unsigned long long cpver_of_node(struct page *node_page)
-{
-	struct f2fs_node *rn = F2FS_NODE(node_page);
-	return le64_to_cpu(rn->footer.cp_ver);
-}
+	if (__is_set_ckpt_flags(ckpt, CP_CRC_RECOVERY_FLAG))
+		cp_ver |= (cur_cp_crc(ckpt) << 32);
 
-static inline block_t next_blkaddr_of_node(struct page *node_page)
-{
-	struct f2fs_node *rn = F2FS_NODE(node_page);
-	return le32_to_cpu(rn->footer.next_blkaddr);
+	return cp_ver == cpver_of_node(page);
 }
 
 /*
@@ -304,7 +345,7 @@ static inline bool IS_DNODE(struct page *node_page)
 	unsigned int ofs = ofs_of_node(node_page);
 
 	if (f2fs_has_xattr_block(ofs))
-		return false;
+		return true;
 
 	if (ofs == 3 || ofs == 4 + NIDS_PER_BLOCK ||
 			ofs == 5 + 2 * NIDS_PER_BLOCK)
@@ -317,17 +358,17 @@ static inline bool IS_DNODE(struct page *node_page)
 	return true;
 }
 
-static inline void set_nid(struct page *p, int off, nid_t nid, bool i)
+static inline int set_nid(struct page *p, int off, nid_t nid, bool i)
 {
 	struct f2fs_node *rn = F2FS_NODE(p);
 
-	f2fs_wait_on_page_writeback(p, NODE);
+	f2fs_wait_on_page_writeback(p, NODE, true);
 
 	if (i)
 		rn->i.i_nid[off - NODE_DIR1_BLOCK] = cpu_to_le32(nid);
 	else
 		rn->in.nid[off] = cpu_to_le32(nid);
-	set_page_dirty(p);
+	return set_page_dirty(p);
 }
 
 static inline nid_t get_nid(struct page *p, int off, bool i)
@@ -370,6 +411,21 @@ static inline int is_node(struct page *page, int type)
 #define is_fsync_dnode(page)	is_node(page, FSYNC_BIT_SHIFT)
 #define is_dent_dnode(page)	is_node(page, DENT_BIT_SHIFT)
 
+static inline int is_inline_node(struct page *page)
+{
+	return PageChecked(page);
+}
+
+static inline void set_inline_node(struct page *page)
+{
+	SetPageChecked(page);
+}
+
+static inline void clear_inline_node(struct page *page)
+{
+	ClearPageChecked(page);
+}
+
 static inline void set_cold_node(struct inode *inode, struct page *page)
 {
 	struct f2fs_node *rn = F2FS_NODE(page);
diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c
index e32f349f341b..9626758bc762 100644
--- a/fs/f2fs/recovery.c
+++ b/fs/f2fs/recovery.c
@@ -49,8 +49,9 @@ static struct kmem_cache *fsync_entry_slab;
 
 bool space_for_roll_forward(struct f2fs_sb_info *sbi)
 {
-	if (sbi->last_valid_block_count + sbi->alloc_valid_block_count
-			> sbi->user_block_count)
+	s64 nalloc = percpu_counter_sum_positive(&sbi->alloc_valid_block_count);
+
+	if (sbi->last_valid_block_count + nalloc > sbi->user_block_count)
 		return false;
 	return true;
 }
@@ -67,42 +68,86 @@ static struct fsync_inode_entry *get_fsync_inode(struct list_head *head,
 	return NULL;
 }
 
-static int recover_dentry(struct inode *inode, struct page *ipage)
+static struct fsync_inode_entry *add_fsync_inode(struct f2fs_sb_info *sbi,
+			struct list_head *head, nid_t ino, bool quota_inode)
+{
+	struct inode *inode;
+	struct fsync_inode_entry *entry;
+	int err;
+
+	inode = f2fs_iget_retry(sbi->sb, ino);
+	if (IS_ERR(inode))
+		return ERR_CAST(inode);
+
+	err = dquot_initialize(inode);
+	if (err)
+		goto err_out;
+
+	if (quota_inode) {
+		err = dquot_alloc_inode(inode);
+		if (err)
+			goto err_out;
+	}
+
+	entry = f2fs_kmem_cache_alloc(fsync_entry_slab, GFP_F2FS_ZERO);
+	entry->inode = inode;
+	list_add_tail(&entry->list, head);
+
+	return entry;
+err_out:
+	iput(inode);
+	return ERR_PTR(err);
+}
+
+static void del_fsync_inode(struct fsync_inode_entry *entry)
+{
+	iput(entry->inode);
+	list_del(&entry->list);
+	kmem_cache_free(fsync_entry_slab, entry);
+}
+
+static int recover_dentry(struct inode *inode, struct page *ipage,
+						struct list_head *dir_list)
 {
 	struct f2fs_inode *raw_inode = F2FS_INODE(ipage);
 	nid_t pino = le32_to_cpu(raw_inode->i_pino);
 	struct f2fs_dir_entry *de;
-	struct qstr name;
+	struct fscrypt_name fname;
 	struct page *page;
 	struct inode *dir, *einode;
+	struct fsync_inode_entry *entry;
 	int err = 0;
+	char *name;
 
-	dir = f2fs_iget(inode->i_sb, pino);
-	if (IS_ERR(dir)) {
-		err = PTR_ERR(dir);
-		goto out;
+	entry = get_fsync_inode(dir_list, pino);
+	if (!entry) {
+		entry = add_fsync_inode(F2FS_I_SB(inode), dir_list,
+							pino, false);
+		if (IS_ERR(entry)) {
+			dir = ERR_CAST(entry);
+			err = PTR_ERR(entry);
+			goto out;
+		}
 	}
 
-	if (file_enc_name(inode)) {
-		iput(dir);
-		return 0;
-	}
+	dir = entry->inode;
 
-	name.len = le32_to_cpu(raw_inode->i_namelen);
-	name.name = raw_inode->i_name;
+	memset(&fname, 0, sizeof(struct fscrypt_name));
+	fname.disk_name.len = le32_to_cpu(raw_inode->i_namelen);
+	fname.disk_name.name = raw_inode->i_name;
 
-	if (unlikely(name.len > F2FS_NAME_LEN)) {
+	if (unlikely(fname.disk_name.len > F2FS_NAME_LEN)) {
 		WARN_ON(1);
 		err = -ENAMETOOLONG;
-		goto out_err;
+		goto out;
 	}
 retry:
-	de = f2fs_find_entry(dir, &name, &page);
+	de = __f2fs_find_entry(dir, &fname, &page);
 	if (de && inode->i_ino == le32_to_cpu(de->ino))
 		goto out_unmap_put;
 
 	if (de) {
-		einode = f2fs_iget(inode->i_sb, le32_to_cpu(de->ino));
+		einode = f2fs_iget_retry(inode->i_sb, le32_to_cpu(de->ino));
 		if (IS_ERR(einode)) {
 			WARN_ON(1);
 			err = PTR_ERR(einode);
@@ -110,6 +155,13 @@ retry:
 				err = -EEXIST;
 			goto out_unmap_put;
 		}
+
+		err = dquot_initialize(einode);
+		if (err) {
+			iput(einode);
+			goto out_unmap_put;
+		}
+
 		err = acquire_orphan_inode(F2FS_I_SB(inode));
 		if (err) {
 			iput(einode);
@@ -118,29 +170,27 @@ retry:
 		f2fs_delete_entry(de, page, dir, einode);
 		iput(einode);
 		goto retry;
-	}
-	err = __f2fs_add_link(dir, &name, inode, inode->i_ino, inode->i_mode);
-	if (err)
-		goto out_err;
-
-	if (is_inode_flag_set(F2FS_I(dir), FI_DELAY_IPUT)) {
-		iput(dir);
+	} else if (IS_ERR(page)) {
+		err = PTR_ERR(page);
 	} else {
-		add_dirty_dir_inode(dir);
-		set_inode_flag(F2FS_I(dir), FI_DELAY_IPUT);
+		err = __f2fs_do_add_link(dir, &fname, inode,
+					inode->i_ino, inode->i_mode);
 	}
-
+	if (err == -ENOMEM)
+		goto retry;
 	goto out;
 
 out_unmap_put:
 	f2fs_dentry_kunmap(dir, page);
 	f2fs_put_page(page, 0);
-out_err:
-	iput(dir);
 out:
+	if (file_enc_name(inode))
+		name = "<encrypted>";
+	else
+		name = raw_inode->i_name;
 	f2fs_msg(inode->i_sb, KERN_NOTICE,
 			"%s: ino = %x, name = %s, dir = %lx, err = %d",
-			__func__, ino_of_node(ipage), raw_inode->i_name,
+			__func__, ino_of_node(ipage), name,
 			IS_ERR(dir) ? 0 : dir->i_ino, err);
 	return err;
 }
@@ -151,14 +201,16 @@ static void recover_inode(struct inode *inode, struct page *page)
 	char *name;
 
 	inode->i_mode = le16_to_cpu(raw->i_mode);
-	i_size_write(inode, le64_to_cpu(raw->i_size));
-	inode->i_atime.tv_sec = le64_to_cpu(raw->i_mtime);
+	f2fs_i_size_write(inode, le64_to_cpu(raw->i_size));
+	inode->i_atime.tv_sec = le64_to_cpu(raw->i_atime);
 	inode->i_ctime.tv_sec = le64_to_cpu(raw->i_ctime);
 	inode->i_mtime.tv_sec = le64_to_cpu(raw->i_mtime);
-	inode->i_atime.tv_nsec = le32_to_cpu(raw->i_mtime_nsec);
+	inode->i_atime.tv_nsec = le32_to_cpu(raw->i_atime_nsec);
 	inode->i_ctime.tv_nsec = le32_to_cpu(raw->i_ctime_nsec);
 	inode->i_mtime.tv_nsec = le32_to_cpu(raw->i_mtime_nsec);
 
+	F2FS_I(inode)->i_advise = raw->i_advise;
+
 	if (file_enc_name(inode))
 		name = "<encrypted>";
 	else
@@ -168,9 +220,9 @@ static void recover_inode(struct inode *inode, struct page *page)
 			ino_of_node(page), name);
 }
 
-static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head)
+static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head,
+				bool check_only)
 {
-	unsigned long long cp_ver = cur_cp_version(F2FS_CKPT(sbi));
 	struct curseg_info *curseg;
 	struct page *page = NULL;
 	block_t blkaddr;
@@ -180,8 +232,6 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head)
 	curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
 	blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
 
-	ra_meta_pages(sbi, blkaddr, 1, META_POR, true);
-
 	while (1) {
 		struct fsync_inode_entry *entry;
 
@@ -190,7 +240,7 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head)
 
 		page = get_tmp_page(sbi, blkaddr);
 
-		if (cp_ver != cpver_of_node(page))
+		if (!is_recoverable_dnode(page))
 			break;
 
 		if (!is_fsync_dnode(page))
@@ -198,41 +248,35 @@ static int find_fsync_dnodes(struct f2fs_sb_info *sbi, struct list_head *head)
 
 		entry = get_fsync_inode(head, ino_of_node(page));
 		if (!entry) {
-			if (IS_INODE(page) && is_dent_dnode(page)) {
+			bool quota_inode = false;
+
+			if (!check_only &&
+					IS_INODE(page) && is_dent_dnode(page)) {
 				err = recover_inode_page(sbi, page);
 				if (err)
 					break;
+				quota_inode = true;
 			}
 
-			/* add this fsync inode to the list */
-			entry = kmem_cache_alloc(fsync_entry_slab, GFP_F2FS_ZERO);
-			if (!entry) {
-				err = -ENOMEM;
-				break;
-			}
 			/*
 			 * CP | dnode(F) | inode(DF)
 			 * For this case, we should not give up now.
 			 */
-			entry->inode = f2fs_iget(sbi->sb, ino_of_node(page));
-			if (IS_ERR(entry->inode)) {
-				err = PTR_ERR(entry->inode);
-				kmem_cache_free(fsync_entry_slab, entry);
+			entry = add_fsync_inode(sbi, head, ino_of_node(page),
+								quota_inode);
+			if (IS_ERR(entry)) {
+				err = PTR_ERR(entry);
 				if (err == -ENOENT) {
 					err = 0;
 					goto next;
 				}
 				break;
 			}
-			list_add_tail(&entry->list, head);
 		}
 		entry->blkaddr = blkaddr;
 
-		if (IS_INODE(page)) {
-			entry->last_inode = blkaddr;
-			if (is_dent_dnode(page))
-				entry->last_dentry = blkaddr;
-		}
+		if (IS_INODE(page) && is_dent_dnode(page))
+			entry->last_dentry = blkaddr;
 next:
 		/* check next segment */
 		blkaddr = next_blkaddr_of_node(page);
@@ -248,11 +292,8 @@ static void destroy_fsync_dnodes(struct list_head *head)
 {
 	struct fsync_inode_entry *entry, *tmp;
 
-	list_for_each_entry_safe(entry, tmp, head, list) {
-		iput(entry->inode);
-		list_del(&entry->list);
-		kmem_cache_free(fsync_entry_slab, entry);
-	}
+	list_for_each_entry_safe(entry, tmp, head, list)
+		del_fsync_inode(entry);
 }
 
 static int check_index_in_prev_nodes(struct f2fs_sb_info *sbi,
@@ -313,16 +354,23 @@ got_it:
 	f2fs_put_page(node_page, 1);
 
 	if (ino != dn->inode->i_ino) {
+		int ret;
+
 		/* Deallocate previous index in the node page */
-		inode = f2fs_iget(sbi->sb, ino);
+		inode = f2fs_iget_retry(sbi->sb, ino);
 		if (IS_ERR(inode))
 			return PTR_ERR(inode);
+
+		ret = dquot_initialize(inode);
+		if (ret) {
+			iput(inode);
+			return ret;
+		}
 	} else {
 		inode = dn->inode;
 	}
 
-	bidx = start_bidx_of_node(offset, F2FS_I(inode)) +
-			le16_to_cpu(sum.ofs_in_node);
+	bidx = start_bidx_of_node(offset, inode) + le16_to_cpu(sum.ofs_in_node);
 
 	/*
 	 * if inode page is locked, unlock temporarily, but its reference
@@ -347,7 +395,8 @@ out:
 	return 0;
 
 truncate_out:
-	if (datablock_addr(tdn.node_page, tdn.ofs_in_node) == blkaddr)
+	if (datablock_addr(tdn.inode, tdn.node_page,
+					tdn.ofs_in_node) == blkaddr)
 		truncate_data_blocks_range(&tdn, 1);
 	if (dn->inode->i_ino == nid && !dn->inode_page_locked)
 		unlock_page(dn->inode_page);
@@ -357,21 +406,18 @@ truncate_out:
 static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 					struct page *page, block_t blkaddr)
 {
-	struct f2fs_inode_info *fi = F2FS_I(inode);
-	unsigned int start, end;
 	struct dnode_of_data dn;
 	struct node_info ni;
+	unsigned int start, end;
 	int err = 0, recovered = 0;
 
 	/* step 1: recover xattr */
 	if (IS_INODE(page)) {
 		recover_inline_xattr(inode, page);
 	} else if (f2fs_has_xattr_block(ofs_of_node(page))) {
-		/*
-		 * Deprecated; xattr blocks should be found from cold log.
-		 * But, we should remain this for backward compatibility.
-		 */
-		recover_xattr_data(inode, page, blkaddr);
+		err = recover_xattr_data(inode, page, blkaddr);
+		if (!err)
+			recovered++;
 		goto out;
 	}
 
@@ -380,16 +426,21 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 		goto out;
 
 	/* step 3: recover data indices */
-	start = start_bidx_of_node(ofs_of_node(page), fi);
-	end = start + ADDRS_PER_PAGE(page, fi);
+	start = start_bidx_of_node(ofs_of_node(page), inode);
+	end = start + ADDRS_PER_PAGE(page, inode);
 
 	set_new_dnode(&dn, inode, NULL, NULL, 0);
-
+retry_dn:
 	err = get_dnode_of_data(&dn, start, ALLOC_NODE);
-	if (err)
+	if (err) {
+		if (err == -ENOMEM) {
+			congestion_wait(BLK_RW_ASYNC, HZ/50);
+			goto retry_dn;
+		}
 		goto out;
+	}
 
-	f2fs_wait_on_page_writeback(dn.node_page, NODE);
+	f2fs_wait_on_page_writeback(dn.node_page, NODE, true);
 
 	get_node_info(sbi, dn.nid, &ni);
 	f2fs_bug_on(sbi, ni.ino != ino_of_node(page));
@@ -398,8 +449,8 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 	for (; start < end; start++, dn.ofs_in_node++) {
 		block_t src, dest;
 
-		src = datablock_addr(dn.node_page, dn.ofs_in_node);
-		dest = datablock_addr(page, dn.ofs_in_node);
+		src = datablock_addr(dn.inode, dn.node_page, dn.ofs_in_node);
+		dest = datablock_addr(dn.inode, page, dn.ofs_in_node);
 
 		/* skip recovering if dest is the same as src */
 		if (src == dest)
@@ -411,14 +462,18 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 			continue;
 		}
 
+		if (!file_keep_isize(inode) &&
+			(i_size_read(inode) <= ((loff_t)start << PAGE_SHIFT)))
+			f2fs_i_size_write(inode,
+				(loff_t)(start + 1) << PAGE_SHIFT);
+
 		/*
 		 * dest is reserved block, invalidate src block
 		 * and then reserve one new block in dnode page.
 		 */
 		if (dest == NEW_ADDR) {
 			truncate_data_blocks_range(&dn, 1);
-			err = reserve_new_block(&dn);
-			f2fs_bug_on(sbi, err);
+			reserve_new_block(&dn);
 			continue;
 		}
 
@@ -427,25 +482,33 @@ static int do_recover_data(struct f2fs_sb_info *sbi, struct inode *inode,
 
 			if (src == NULL_ADDR) {
 				err = reserve_new_block(&dn);
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+				while (err)
+					err = reserve_new_block(&dn);
+#endif
 				/* We should not get -ENOSPC */
 				f2fs_bug_on(sbi, err);
+				if (err)
+					goto err;
 			}
-
+retry_prev:
 			/* Check the previous node page having this index */
 			err = check_index_in_prev_nodes(sbi, dest, &dn);
-			if (err)
+			if (err) {
+				if (err == -ENOMEM) {
+					congestion_wait(BLK_RW_ASYNC, HZ/50);
+					goto retry_prev;
+				}
 				goto err;
+			}
 
 			/* write dummy data page */
 			f2fs_replace_block(sbi, &dn, src, dest,
-							ni.version, false);
+						ni.version, false, false);
 			recovered++;
 		}
 	}
 
-	if (IS_INODE(dn.node_page))
-		sync_inode_page(&dn);
-
 	copy_node_footer(dn.node_page, page);
 	fill_node_footer(dn.node_page, dn.nid, ni.ino,
 					ofs_of_node(page), false);
@@ -454,22 +517,23 @@ err:
 	f2fs_put_dnode(&dn);
 out:
 	f2fs_msg(sbi->sb, KERN_NOTICE,
-		"recover_data: ino = %lx, recovered = %d blocks, err = %d",
-		inode->i_ino, recovered, err);
+		"recover_data: ino = %lx (i_size: %s) recovered = %d, err = %d",
+		inode->i_ino,
+		file_keep_isize(inode) ? "keep" : "recover",
+		recovered, err);
 	return err;
 }
 
-static int recover_data(struct f2fs_sb_info *sbi,
-				struct list_head *head, int type)
+static int recover_data(struct f2fs_sb_info *sbi, struct list_head *inode_list,
+						struct list_head *dir_list)
 {
-	unsigned long long cp_ver = cur_cp_version(F2FS_CKPT(sbi));
 	struct curseg_info *curseg;
 	struct page *page = NULL;
 	int err = 0;
 	block_t blkaddr;
 
 	/* get node pages in the current segment */
-	curseg = CURSEG_I(sbi, type);
+	curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
 	blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
 
 	while (1) {
@@ -482,12 +546,12 @@ static int recover_data(struct f2fs_sb_info *sbi,
 
 		page = get_tmp_page(sbi, blkaddr);
 
-		if (cp_ver != cpver_of_node(page)) {
+		if (!is_recoverable_dnode(page)) {
 			f2fs_put_page(page, 1);
 			break;
 		}
 
-		entry = get_fsync_inode(head, ino_of_node(page));
+		entry = get_fsync_inode(inode_list, ino_of_node(page));
 		if (!entry)
 			goto next;
 		/*
@@ -495,10 +559,10 @@ static int recover_data(struct f2fs_sb_info *sbi,
 		 * In this case, we can lose the latest inode(x).
 		 * So, call recover_inode for the inode update.
 		 */
-		if (entry->last_inode == blkaddr)
+		if (IS_INODE(page))
 			recover_inode(entry->inode, page);
 		if (entry->last_dentry == blkaddr) {
-			err = recover_dentry(entry->inode, page);
+			err = recover_dentry(entry->inode, page, dir_list);
 			if (err) {
 				f2fs_put_page(page, 1);
 				break;
@@ -510,11 +574,8 @@ static int recover_data(struct f2fs_sb_info *sbi,
 			break;
 		}
 
-		if (entry->blkaddr == blkaddr) {
-			iput(entry->inode);
-			list_del(&entry->list);
-			kmem_cache_free(fsync_entry_slab, entry);
-		}
+		if (entry->blkaddr == blkaddr)
+			del_fsync_inode(entry);
 next:
 		/* check next segment */
 		blkaddr = next_blkaddr_of_node(page);
@@ -525,47 +586,62 @@ next:
 	return err;
 }
 
-int recover_fsync_data(struct f2fs_sb_info *sbi)
+int recover_fsync_data(struct f2fs_sb_info *sbi, bool check_only)
 {
-	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_WARM_NODE);
 	struct list_head inode_list;
-	block_t blkaddr;
+	struct list_head dir_list;
 	int err;
+	int ret = 0;
+	unsigned long s_flags = sbi->sb->s_flags;
 	bool need_writecp = false;
 
+	if (s_flags & MS_RDONLY) {
+		f2fs_msg(sbi->sb, KERN_INFO, "orphan cleanup on readonly fs");
+		sbi->sb->s_flags &= ~MS_RDONLY;
+	}
+
+#ifdef CONFIG_QUOTA
+	/* Needed for iput() to work correctly and not trash data */
+	sbi->sb->s_flags |= MS_ACTIVE;
+	/* Turn on quotas so that they are updated correctly */
+	f2fs_enable_quota_files(sbi);
+#endif
+
 	fsync_entry_slab = f2fs_kmem_cache_create("f2fs_fsync_inode_entry",
 			sizeof(struct fsync_inode_entry));
-	if (!fsync_entry_slab)
-		return -ENOMEM;
+	if (!fsync_entry_slab) {
+		err = -ENOMEM;
+		goto out;
+	}
 
 	INIT_LIST_HEAD(&inode_list);
+	INIT_LIST_HEAD(&dir_list);
 
 	/* prevent checkpoint */
 	mutex_lock(&sbi->cp_mutex);
 
-	blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
-
 	/* step #1: find fsynced inode numbers */
-	err = find_fsync_dnodes(sbi, &inode_list);
-	if (err)
-		goto out;
+	err = find_fsync_dnodes(sbi, &inode_list, check_only);
+	if (err || list_empty(&inode_list))
+		goto skip;
 
-	if (list_empty(&inode_list))
-		goto out;
+	if (check_only) {
+		ret = 1;
+		goto skip;
+	}
 
 	need_writecp = true;
 
 	/* step #2: recover data */
-	err = recover_data(sbi, &inode_list, CURSEG_WARM_NODE);
+	err = recover_data(sbi, &inode_list, &dir_list);
 	if (!err)
 		f2fs_bug_on(sbi, !list_empty(&inode_list));
-out:
+skip:
 	destroy_fsync_dnodes(&inode_list);
-	kmem_cache_destroy(fsync_entry_slab);
 
 	/* truncate meta pages to be used by the recovery */
 	truncate_inode_pages_range(META_MAPPING(sbi),
-			(loff_t)MAIN_BLKADDR(sbi) << PAGE_CACHE_SHIFT, -1);
+			(loff_t)MAIN_BLKADDR(sbi) << PAGE_SHIFT, -1);
 
 	if (err) {
 		truncate_inode_pages_final(NODE_MAPPING(sbi));
@@ -573,31 +649,25 @@ out:
 	}
 
 	clear_sbi_flag(sbi, SBI_POR_DOING);
-	if (err) {
-		bool invalidate = false;
+	mutex_unlock(&sbi->cp_mutex);
 
-		if (discard_next_dnode(sbi, blkaddr))
-			invalidate = true;
+	/* let's drop all the directory inodes for clean checkpoint */
+	destroy_fsync_dnodes(&dir_list);
 
-		/* Flush all the NAT/SIT pages */
-		while (get_pages(sbi, F2FS_DIRTY_META))
-			sync_meta_pages(sbi, META, LONG_MAX);
-
-		/* invalidate temporary meta page */
-		if (invalidate)
-			invalidate_mapping_pages(META_MAPPING(sbi),
-							blkaddr, blkaddr);
-
-		set_ckpt_flags(sbi->ckpt, CP_ERROR_FLAG);
-		mutex_unlock(&sbi->cp_mutex);
-	} else if (need_writecp) {
+	if (!err && need_writecp) {
 		struct cp_control cpc = {
 			.reason = CP_RECOVERY,
 		};
-		mutex_unlock(&sbi->cp_mutex);
-		write_checkpoint(sbi, &cpc);
-	} else {
-		mutex_unlock(&sbi->cp_mutex);
+		err = write_checkpoint(sbi, &cpc);
 	}
-	return err;
+
+	kmem_cache_destroy(fsync_entry_slab);
+out:
+#ifdef CONFIG_QUOTA
+	/* Turn quotas off */
+	f2fs_quota_off_umount(sbi->sb);
+#endif
+	sbi->sb->s_flags = s_flags; /* Restore MS_RDONLY status */
+
+	return ret ? ret: err;
 }
diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c
index f77b3258454a..f5c494389483 100644
--- a/fs/f2fs/segment.c
+++ b/fs/f2fs/segment.c
@@ -16,16 +16,20 @@
 #include <linux/kthread.h>
 #include <linux/swap.h>
 #include <linux/timer.h>
+#include <linux/freezer.h>
+#include <linux/sched.h>
 
 #include "f2fs.h"
 #include "segment.h"
 #include "node.h"
+#include "gc.h"
 #include "trace.h"
 #include <trace/events/f2fs.h>
 
 #define __reverse_ffz(x) __reverse_ffs(~(x))
 
 static struct kmem_cache *discard_entry_slab;
+static struct kmem_cache *discard_cmd_slab;
 static struct kmem_cache *sit_entry_set_slab;
 static struct kmem_cache *inmem_entry_slab;
 
@@ -86,6 +90,7 @@ static inline unsigned long __reverse_ffs(unsigned long word)
 /*
  * __find_rev_next(_zero)_bit is copied from lib/find_next_bit.c because
  * f2fs_set_bit makes MSB and LSB reversed in a byte.
+ * @size must be integral times of unsigned long.
  * Example:
  *                             MSB <--> LSB
  *   f2fs_set_bit(0, bitmap) => 1000 0000
@@ -95,94 +100,88 @@ static unsigned long __find_rev_next_bit(const unsigned long *addr,
 			unsigned long size, unsigned long offset)
 {
 	const unsigned long *p = addr + BIT_WORD(offset);
-	unsigned long result = offset & ~(BITS_PER_LONG - 1);
+	unsigned long result = size;
 	unsigned long tmp;
 
 	if (offset >= size)
 		return size;
 
-	size -= result;
+	size -= (offset & ~(BITS_PER_LONG - 1));
 	offset %= BITS_PER_LONG;
-	if (!offset)
-		goto aligned;
-
-	tmp = __reverse_ulong((unsigned char *)p);
-	tmp &= ~0UL >> offset;
-
-	if (size < BITS_PER_LONG)
-		goto found_first;
-	if (tmp)
-		goto found_middle;
-
-	size -= BITS_PER_LONG;
-	result += BITS_PER_LONG;
-	p++;
-aligned:
-	while (size & ~(BITS_PER_LONG-1)) {
+
+	while (1) {
+		if (*p == 0)
+			goto pass;
+
 		tmp = __reverse_ulong((unsigned char *)p);
+
+		tmp &= ~0UL >> offset;
+		if (size < BITS_PER_LONG)
+			tmp &= (~0UL << (BITS_PER_LONG - size));
 		if (tmp)
-			goto found_middle;
-		result += BITS_PER_LONG;
+			goto found;
+pass:
+		if (size <= BITS_PER_LONG)
+			break;
 		size -= BITS_PER_LONG;
+		offset = 0;
 		p++;
 	}
-	if (!size)
-		return result;
-
-	tmp = __reverse_ulong((unsigned char *)p);
-found_first:
-	tmp &= (~0UL << (BITS_PER_LONG - size));
-	if (!tmp)		/* Are any bits set? */
-		return result + size;   /* Nope. */
-found_middle:
-	return result + __reverse_ffs(tmp);
+	return result;
+found:
+	return result - size + __reverse_ffs(tmp);
 }
 
 static unsigned long __find_rev_next_zero_bit(const unsigned long *addr,
 			unsigned long size, unsigned long offset)
 {
 	const unsigned long *p = addr + BIT_WORD(offset);
-	unsigned long result = offset & ~(BITS_PER_LONG - 1);
+	unsigned long result = size;
 	unsigned long tmp;
 
 	if (offset >= size)
 		return size;
 
-	size -= result;
+	size -= (offset & ~(BITS_PER_LONG - 1));
 	offset %= BITS_PER_LONG;
-	if (!offset)
-		goto aligned;
-
-	tmp = __reverse_ulong((unsigned char *)p);
-	tmp |= ~((~0UL << offset) >> offset);
-
-	if (size < BITS_PER_LONG)
-		goto found_first;
-	if (tmp != ~0UL)
-		goto found_middle;
-
-	size -= BITS_PER_LONG;
-	result += BITS_PER_LONG;
-	p++;
-aligned:
-	while (size & ~(BITS_PER_LONG - 1)) {
+
+	while (1) {
+		if (*p == ~0UL)
+			goto pass;
+
 		tmp = __reverse_ulong((unsigned char *)p);
+
+		if (offset)
+			tmp |= ~0UL << (BITS_PER_LONG - offset);
+		if (size < BITS_PER_LONG)
+			tmp |= ~0UL >> size;
 		if (tmp != ~0UL)
-			goto found_middle;
-		result += BITS_PER_LONG;
+			goto found;
+pass:
+		if (size <= BITS_PER_LONG)
+			break;
 		size -= BITS_PER_LONG;
+		offset = 0;
 		p++;
 	}
-	if (!size)
-		return result;
+	return result;
+found:
+	return result - size + __reverse_ffz(tmp);
+}
 
-	tmp = __reverse_ulong((unsigned char *)p);
-found_first:
-	tmp |= ~(~0UL << (BITS_PER_LONG - size));
-	if (tmp == ~0UL)	/* Are any bits zero? */
-		return result + size;   /* Nope. */
-found_middle:
-	return result + __reverse_ffz(tmp);
+bool need_SSR(struct f2fs_sb_info *sbi)
+{
+	int node_secs = get_blocktype_secs(sbi, F2FS_DIRTY_NODES);
+	int dent_secs = get_blocktype_secs(sbi, F2FS_DIRTY_DENTS);
+	int imeta_secs = get_blocktype_secs(sbi, F2FS_DIRTY_IMETA);
+
+	if (test_opt(sbi, LFS))
+		return false;
+	if (sbi->gc_thread && sbi->gc_thread->gc_urgent)
+		return true;
+
+	return free_sections(sbi) <= (node_secs + 2 * dent_secs + imeta_secs +
+						2 * reserved_sections(sbi));
 }
 
 void register_inmem_page(struct inode *inode, struct page *page)
@@ -211,69 +210,199 @@ void register_inmem_page(struct inode *inode, struct page *page)
 	trace_f2fs_register_inmem_page(page, INMEM);
 }
 
-int commit_inmem_pages(struct inode *inode, bool abort)
+static int __revoke_inmem_pages(struct inode *inode,
+				struct list_head *head, bool drop, bool recover)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct inmem_pages *cur, *tmp;
+	int err = 0;
+
+	list_for_each_entry_safe(cur, tmp, head, list) {
+		struct page *page = cur->page;
+
+		if (drop)
+			trace_f2fs_commit_inmem_page(page, INMEM_DROP);
+
+		lock_page(page);
+
+		if (recover) {
+			struct dnode_of_data dn;
+			struct node_info ni;
+
+			trace_f2fs_commit_inmem_page(page, INMEM_REVOKE);
+retry:
+			set_new_dnode(&dn, inode, NULL, NULL, 0);
+			err = get_dnode_of_data(&dn, page->index, LOOKUP_NODE);
+			if (err) {
+				if (err == -ENOMEM) {
+					congestion_wait(BLK_RW_ASYNC, HZ/50);
+					cond_resched();
+					goto retry;
+				}
+				err = -EAGAIN;
+				goto next;
+			}
+			get_node_info(sbi, dn.nid, &ni);
+			f2fs_replace_block(sbi, &dn, dn.data_blkaddr,
+					cur->old_addr, ni.version, true, true);
+			f2fs_put_dnode(&dn);
+		}
+next:
+		/* we don't need to invalidate this in the sccessful status */
+		if (drop || recover)
+			ClearPageUptodate(page);
+		set_page_private(page, 0);
+		ClearPagePrivate(page);
+		f2fs_put_page(page, 1);
+
+		list_del(&cur->list);
+		kmem_cache_free(inmem_entry_slab, cur);
+		dec_page_count(F2FS_I_SB(inode), F2FS_INMEM_PAGES);
+	}
+	return err;
+}
+
+void drop_inmem_pages(struct inode *inode)
+{
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+
+	mutex_lock(&fi->inmem_lock);
+	__revoke_inmem_pages(inode, &fi->inmem_pages, true, false);
+	mutex_unlock(&fi->inmem_lock);
+
+	clear_inode_flag(inode, FI_ATOMIC_FILE);
+	clear_inode_flag(inode, FI_HOT_DATA);
+	stat_dec_atomic_write(inode);
+}
+
+void drop_inmem_page(struct inode *inode, struct page *page)
+{
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct list_head *head = &fi->inmem_pages;
+	struct inmem_pages *cur = NULL;
+
+	f2fs_bug_on(sbi, !IS_ATOMIC_WRITTEN_PAGE(page));
+
+	mutex_lock(&fi->inmem_lock);
+	list_for_each_entry(cur, head, list) {
+		if (cur->page == page)
+			break;
+	}
+
+	f2fs_bug_on(sbi, !cur || cur->page != page);
+	list_del(&cur->list);
+	mutex_unlock(&fi->inmem_lock);
+
+	dec_page_count(sbi, F2FS_INMEM_PAGES);
+	kmem_cache_free(inmem_entry_slab, cur);
+
+	ClearPageUptodate(page);
+	set_page_private(page, 0);
+	ClearPagePrivate(page);
+	f2fs_put_page(page, 0);
+
+	trace_f2fs_commit_inmem_page(page, INMEM_INVALIDATE);
+}
+
+static int __commit_inmem_pages(struct inode *inode,
+					struct list_head *revoke_list)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct f2fs_inode_info *fi = F2FS_I(inode);
 	struct inmem_pages *cur, *tmp;
-	bool submit_bio = false;
 	struct f2fs_io_info fio = {
 		.sbi = sbi,
 		.type = DATA,
-		.rw = WRITE_SYNC | REQ_PRIO,
-		.encrypted_page = NULL,
+		.op = REQ_OP_WRITE,
+		.op_flags = REQ_SYNC | REQ_PRIO,
+		.io_type = FS_DATA_IO,
 	};
+	pgoff_t last_idx = ULONG_MAX;
 	int err = 0;
 
-	/*
-	 * The abort is true only when f2fs_evict_inode is called.
-	 * Basically, the f2fs_evict_inode doesn't produce any data writes, so
-	 * that we don't need to call f2fs_balance_fs.
-	 * Otherwise, f2fs_gc in f2fs_balance_fs can wait forever until this
-	 * inode becomes free by iget_locked in f2fs_iget.
-	 */
-	if (!abort) {
-		f2fs_balance_fs(sbi);
-		f2fs_lock_op(sbi);
-	}
-
-	mutex_lock(&fi->inmem_lock);
 	list_for_each_entry_safe(cur, tmp, &fi->inmem_pages, list) {
-		lock_page(cur->page);
-		if (!abort) {
-			if (cur->page->mapping == inode->i_mapping) {
-				set_page_dirty(cur->page);
-				f2fs_wait_on_page_writeback(cur->page, DATA);
-				if (clear_page_dirty_for_io(cur->page))
-					inode_dec_dirty_pages(inode);
-				trace_f2fs_commit_inmem_page(cur->page, INMEM);
-				fio.page = cur->page;
-				err = do_write_data_page(&fio);
-				if (err) {
-					unlock_page(cur->page);
-					break;
+		struct page *page = cur->page;
+
+		lock_page(page);
+		if (page->mapping == inode->i_mapping) {
+			trace_f2fs_commit_inmem_page(page, INMEM);
+
+			set_page_dirty(page);
+			f2fs_wait_on_page_writeback(page, DATA, true);
+			if (clear_page_dirty_for_io(page)) {
+				inode_dec_dirty_pages(inode);
+				remove_dirty_inode(inode);
+			}
+retry:
+			fio.page = page;
+			fio.old_blkaddr = NULL_ADDR;
+			fio.encrypted_page = NULL;
+			fio.need_lock = LOCK_DONE;
+			err = do_write_data_page(&fio);
+			if (err) {
+				if (err == -ENOMEM) {
+					congestion_wait(BLK_RW_ASYNC, HZ/50);
+					cond_resched();
+					goto retry;
 				}
-				clear_cold_data(cur->page);
-				submit_bio = true;
+				unlock_page(page);
+				break;
 			}
-		} else {
-			trace_f2fs_commit_inmem_page(cur->page, INMEM_DROP);
+			/* record old blkaddr for revoking */
+			cur->old_addr = fio.old_blkaddr;
+			last_idx = page->index;
 		}
-		set_page_private(cur->page, 0);
-		ClearPagePrivate(cur->page);
-		f2fs_put_page(cur->page, 1);
+		unlock_page(page);
+		list_move_tail(&cur->list, revoke_list);
+	}
 
-		list_del(&cur->list);
-		kmem_cache_free(inmem_entry_slab, cur);
-		dec_page_count(F2FS_I_SB(inode), F2FS_INMEM_PAGES);
+	if (last_idx != ULONG_MAX)
+		f2fs_submit_merged_write_cond(sbi, inode, 0, last_idx, DATA);
+
+	if (!err)
+		__revoke_inmem_pages(inode, revoke_list, false, false);
+
+	return err;
+}
+
+int commit_inmem_pages(struct inode *inode)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	struct f2fs_inode_info *fi = F2FS_I(inode);
+	struct list_head revoke_list;
+	int err;
+
+	INIT_LIST_HEAD(&revoke_list);
+	f2fs_balance_fs(sbi, true);
+	f2fs_lock_op(sbi);
+
+	set_inode_flag(inode, FI_ATOMIC_COMMIT);
+
+	mutex_lock(&fi->inmem_lock);
+	err = __commit_inmem_pages(inode, &revoke_list);
+	if (err) {
+		int ret;
+		/*
+		 * try to revoke all committed pages, but still we could fail
+		 * due to no memory or other reason, if that happened, EAGAIN
+		 * will be returned, which means in such case, transaction is
+		 * already not integrity, caller should use journal to do the
+		 * recovery or rewrite & commit last transaction. For other
+		 * error number, revoking was done by filesystem itself.
+		 */
+		ret = __revoke_inmem_pages(inode, &revoke_list, false, true);
+		if (ret)
+			err = ret;
+
+		/* drop all uncommitted pages */
+		__revoke_inmem_pages(inode, &fi->inmem_pages, true, false);
 	}
 	mutex_unlock(&fi->inmem_lock);
 
-	if (!abort) {
-		f2fs_unlock_op(sbi);
-		if (submit_bio)
-			f2fs_submit_merged_bio(sbi, DATA, WRITE);
-	}
+	clear_inode_flag(inode, FI_ATOMIC_COMMIT);
+
+	f2fs_unlock_op(sbi);
 	return err;
 }
 
@@ -281,15 +410,26 @@ int commit_inmem_pages(struct inode *inode, bool abort)
  * This function balances dirty node and dentry pages.
  * In addition, it controls garbage collection.
  */
-void f2fs_balance_fs(struct f2fs_sb_info *sbi)
+void f2fs_balance_fs(struct f2fs_sb_info *sbi, bool need)
 {
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (time_to_inject(sbi, FAULT_CHECKPOINT)) {
+		f2fs_show_injection_info(FAULT_CHECKPOINT);
+		f2fs_stop_checkpoint(sbi, false);
+	}
+#endif
+
+	/* balance_fs_bg is able to be pending */
+	if (need && excess_cached_nats(sbi))
+		f2fs_balance_fs_bg(sbi);
+
 	/*
 	 * We should do GC or end up with checkpoint, if there are so many dirty
 	 * dir/node pages without enough free segments.
 	 */
-	if (has_not_enough_free_secs(sbi, 0)) {
+	if (has_not_enough_free_secs(sbi, 0, 0)) {
 		mutex_lock(&sbi->gc_mutex);
-		f2fs_gc(sbi, false);
+		f2fs_gc(sbi, false, false, NULL_SEGNO);
 	}
 }
 
@@ -304,47 +444,94 @@ void f2fs_balance_fs_bg(struct f2fs_sb_info *sbi)
 		try_to_free_nats(sbi, NAT_ENTRY_PER_BLOCK);
 
 	if (!available_free_memory(sbi, FREE_NIDS))
-		try_to_free_nids(sbi, NAT_ENTRY_PER_BLOCK * FREE_NID_PAGES);
+		try_to_free_nids(sbi, MAX_FREE_NIDS);
+	else
+		build_free_nids(sbi, false, false);
+
+	if (!is_idle(sbi) && !excess_dirty_nats(sbi))
+		return;
 
 	/* checkpoint is the only way to shrink partial cached entries */
 	if (!available_free_memory(sbi, NAT_ENTRIES) ||
-			excess_prefree_segs(sbi) ||
 			!available_free_memory(sbi, INO_ENTRIES) ||
-			jiffies > sbi->cp_expires)
+			excess_prefree_segs(sbi) ||
+			excess_dirty_nats(sbi) ||
+			f2fs_time_over(sbi, CP_TIME)) {
+		if (test_opt(sbi, DATA_FLUSH)) {
+			struct blk_plug plug;
+
+			blk_start_plug(&plug);
+			sync_dirty_inodes(sbi, FILE_INODE);
+			blk_finish_plug(&plug);
+		}
 		f2fs_sync_fs(sbi->sb, true);
+		stat_inc_bg_cp_count(sbi->stat_info);
+	}
+}
+
+static int __submit_flush_wait(struct f2fs_sb_info *sbi,
+				struct block_device *bdev)
+{
+	struct bio *bio = f2fs_bio_alloc(0);
+	int ret;
+
+	bio->bi_rw = REQ_OP_WRITE;
+	bio->bi_bdev = bdev;
+	ret = submit_bio_wait(WRITE_FLUSH, bio);
+	bio_put(bio);
+
+	trace_f2fs_issue_flush(bdev, test_opt(sbi, NOBARRIER),
+				test_opt(sbi, FLUSH_MERGE), ret);
+	return ret;
+}
+
+static int submit_flush_wait(struct f2fs_sb_info *sbi)
+{
+	int ret = __submit_flush_wait(sbi, sbi->sb->s_bdev);
+	int i;
+
+	if (!sbi->s_ndevs || ret)
+		return ret;
+
+	for (i = 1; i < sbi->s_ndevs; i++) {
+		ret = __submit_flush_wait(sbi, FDEV(i).bdev);
+		if (ret)
+			break;
+	}
+	return ret;
 }
 
 static int issue_flush_thread(void *data)
 {
 	struct f2fs_sb_info *sbi = data;
-	struct flush_cmd_control *fcc = SM_I(sbi)->cmd_control_info;
+	struct flush_cmd_control *fcc = SM_I(sbi)->fcc_info;
 	wait_queue_head_t *q = &fcc->flush_wait_queue;
 repeat:
 	if (kthread_should_stop())
 		return 0;
 
+	sb_start_intwrite(sbi->sb);
+
 	if (!llist_empty(&fcc->issue_list)) {
-		struct bio *bio;
 		struct flush_cmd *cmd, *next;
 		int ret;
 
-		bio = f2fs_bio_alloc(0);
-
 		fcc->dispatch_list = llist_del_all(&fcc->issue_list);
 		fcc->dispatch_list = llist_reverse_order(fcc->dispatch_list);
 
-		bio->bi_bdev = sbi->sb->s_bdev;
-		ret = submit_bio_wait(WRITE_FLUSH, bio);
+		ret = submit_flush_wait(sbi);
+		atomic_inc(&fcc->issued_flush);
 
 		llist_for_each_entry_safe(cmd, next,
 					  fcc->dispatch_list, llnode) {
 			cmd->ret = ret;
 			complete(&cmd->wait);
 		}
-		bio_put(bio);
 		fcc->dispatch_list = NULL;
 	}
 
+	sb_end_intwrite(sbi->sb);
+
 	wait_event_interruptible(*q,
 		kthread_should_stop() || !llist_empty(&fcc->issue_list));
 	goto repeat;
@@ -352,22 +539,24 @@ repeat:
 
 int f2fs_issue_flush(struct f2fs_sb_info *sbi)
 {
-	struct flush_cmd_control *fcc = SM_I(sbi)->cmd_control_info;
+	struct flush_cmd_control *fcc = SM_I(sbi)->fcc_info;
 	struct flush_cmd cmd;
-
-	trace_f2fs_issue_flush(sbi->sb, test_opt(sbi, NOBARRIER),
-					test_opt(sbi, FLUSH_MERGE));
+	int ret;
 
 	if (test_opt(sbi, NOBARRIER))
 		return 0;
 
 	if (!test_opt(sbi, FLUSH_MERGE)) {
-		struct bio *bio = f2fs_bio_alloc(0);
-		int ret;
+		ret = submit_flush_wait(sbi);
+		atomic_inc(&fcc->issued_flush);
+		return ret;
+	}
+
+	if (atomic_inc_return(&fcc->issing_flush) == 1) {
+		ret = submit_flush_wait(sbi);
+		atomic_dec(&fcc->issing_flush);
 
-		bio->bi_bdev = sbi->sb->s_bdev;
-		ret = submit_bio_wait(WRITE_FLUSH, bio);
-		bio_put(bio);
+		atomic_inc(&fcc->issued_flush);
 		return ret;
 	}
 
@@ -375,10 +564,38 @@ int f2fs_issue_flush(struct f2fs_sb_info *sbi)
 
 	llist_add(&cmd.llnode, &fcc->issue_list);
 
-	if (!fcc->dispatch_list)
+	/* update issue_list before we wake up issue_flush thread */
+	smp_mb();
+
+	if (waitqueue_active(&fcc->flush_wait_queue))
 		wake_up(&fcc->flush_wait_queue);
 
-	wait_for_completion(&cmd.wait);
+	if (fcc->f2fs_issue_flush) {
+		wait_for_completion(&cmd.wait);
+		atomic_dec(&fcc->issing_flush);
+	} else {
+		struct llist_node *list;
+
+		list = llist_del_all(&fcc->issue_list);
+		if (!list) {
+			wait_for_completion(&cmd.wait);
+			atomic_dec(&fcc->issing_flush);
+		} else {
+			struct flush_cmd *tmp, *next;
+
+			ret = submit_flush_wait(sbi);
+
+			llist_for_each_entry_safe(tmp, next, list, llnode) {
+				if (tmp == &cmd) {
+					cmd.ret = ret;
+					atomic_dec(&fcc->issing_flush);
+					continue;
+				}
+				tmp->ret = ret;
+				complete(&tmp->wait);
+			}
+		}
+	}
 
 	return cmd.ret;
 }
@@ -389,32 +606,51 @@ int create_flush_cmd_control(struct f2fs_sb_info *sbi)
 	struct flush_cmd_control *fcc;
 	int err = 0;
 
+	if (SM_I(sbi)->fcc_info) {
+		fcc = SM_I(sbi)->fcc_info;
+		if (fcc->f2fs_issue_flush)
+			return err;
+		goto init_thread;
+	}
+
 	fcc = kzalloc(sizeof(struct flush_cmd_control), GFP_KERNEL);
 	if (!fcc)
 		return -ENOMEM;
+	atomic_set(&fcc->issued_flush, 0);
+	atomic_set(&fcc->issing_flush, 0);
 	init_waitqueue_head(&fcc->flush_wait_queue);
 	init_llist_head(&fcc->issue_list);
-	SM_I(sbi)->cmd_control_info = fcc;
+	SM_I(sbi)->fcc_info = fcc;
+	if (!test_opt(sbi, FLUSH_MERGE))
+		return err;
+
+init_thread:
 	fcc->f2fs_issue_flush = kthread_run(issue_flush_thread, sbi,
 				"f2fs_flush-%u:%u", MAJOR(dev), MINOR(dev));
 	if (IS_ERR(fcc->f2fs_issue_flush)) {
 		err = PTR_ERR(fcc->f2fs_issue_flush);
 		kfree(fcc);
-		SM_I(sbi)->cmd_control_info = NULL;
+		SM_I(sbi)->fcc_info = NULL;
 		return err;
 	}
 
 	return err;
 }
 
-void destroy_flush_cmd_control(struct f2fs_sb_info *sbi)
+void destroy_flush_cmd_control(struct f2fs_sb_info *sbi, bool free)
 {
-	struct flush_cmd_control *fcc = SM_I(sbi)->cmd_control_info;
+	struct flush_cmd_control *fcc = SM_I(sbi)->fcc_info;
+
+	if (fcc && fcc->f2fs_issue_flush) {
+		struct task_struct *flush_thread = fcc->f2fs_issue_flush;
 
-	if (fcc && fcc->f2fs_issue_flush)
-		kthread_stop(fcc->f2fs_issue_flush);
-	kfree(fcc);
-	SM_I(sbi)->cmd_control_info = NULL;
+		fcc->f2fs_issue_flush = NULL;
+		kthread_stop(flush_thread);
+	}
+	if (free) {
+		kfree(fcc);
+		SM_I(sbi)->fcc_info = NULL;
+	}
 }
 
 static void __locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno,
@@ -457,8 +693,8 @@ static void __remove_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno,
 		if (test_and_clear_bit(segno, dirty_i->dirty_segmap[t]))
 			dirty_i->nr_dirty[t]--;
 
-		if (get_valid_blocks(sbi, segno, sbi->segs_per_sec) == 0)
-			clear_bit(GET_SECNO(sbi, segno),
+		if (get_valid_blocks(sbi, segno, true) == 0)
+			clear_bit(GET_SEC_FROM_SEG(sbi, segno),
 						dirty_i->victim_secmap);
 	}
 }
@@ -478,7 +714,7 @@ static void locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno)
 
 	mutex_lock(&dirty_i->seglist_lock);
 
-	valid_blocks = get_valid_blocks(sbi, segno, 0);
+	valid_blocks = get_valid_blocks(sbi, segno, false);
 
 	if (valid_blocks == 0) {
 		__locate_dirty_segment(sbi, segno, PRE);
@@ -493,74 +729,727 @@ static void locate_dirty_segment(struct f2fs_sb_info *sbi, unsigned int segno)
 	mutex_unlock(&dirty_i->seglist_lock);
 }
 
-static int f2fs_issue_discard(struct f2fs_sb_info *sbi,
-				block_t blkstart, block_t blklen)
+static struct discard_cmd *__create_discard_cmd(struct f2fs_sb_info *sbi,
+		struct block_device *bdev, block_t lstart,
+		block_t start, block_t len)
 {
-	sector_t start = SECTOR_FROM_BLOCK(blkstart);
-	sector_t len = SECTOR_FROM_BLOCK(blklen);
-	struct seg_entry *se;
-	unsigned int offset;
-	block_t i;
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct list_head *pend_list;
+	struct discard_cmd *dc;
 
-	for (i = blkstart; i < blkstart + blklen; i++) {
-		se = get_seg_entry(sbi, GET_SEGNO(sbi, i));
-		offset = GET_BLKOFF_FROM_SEG0(sbi, i);
+	f2fs_bug_on(sbi, !len);
 
-		if (!f2fs_test_and_set_bit(offset, se->discard_map))
-			sbi->discard_blks--;
+	pend_list = &dcc->pend_list[plist_idx(len)];
+
+	dc = f2fs_kmem_cache_alloc(discard_cmd_slab, GFP_NOFS);
+	INIT_LIST_HEAD(&dc->list);
+	dc->bdev = bdev;
+	dc->lstart = lstart;
+	dc->start = start;
+	dc->len = len;
+	dc->ref = 0;
+	dc->state = D_PREP;
+	dc->error = 0;
+	init_completion(&dc->wait);
+	list_add_tail(&dc->list, pend_list);
+	atomic_inc(&dcc->discard_cmd_cnt);
+	dcc->undiscard_blks += len;
+
+	return dc;
+}
+
+static struct discard_cmd *__attach_discard_cmd(struct f2fs_sb_info *sbi,
+				struct block_device *bdev, block_t lstart,
+				block_t start, block_t len,
+				struct rb_node *parent, struct rb_node **p)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct discard_cmd *dc;
+
+	dc = __create_discard_cmd(sbi, bdev, lstart, start, len);
+
+	rb_link_node(&dc->rb_node, parent, p);
+	rb_insert_color(&dc->rb_node, &dcc->root);
+
+	return dc;
+}
+
+static void __detach_discard_cmd(struct discard_cmd_control *dcc,
+							struct discard_cmd *dc)
+{
+	if (dc->state == D_DONE)
+		atomic_dec(&dcc->issing_discard);
+
+	list_del(&dc->list);
+	rb_erase(&dc->rb_node, &dcc->root);
+	dcc->undiscard_blks -= dc->len;
+
+	kmem_cache_free(discard_cmd_slab, dc);
+
+	atomic_dec(&dcc->discard_cmd_cnt);
+}
+
+static void __remove_discard_cmd(struct f2fs_sb_info *sbi,
+							struct discard_cmd *dc)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+
+	f2fs_bug_on(sbi, dc->ref);
+
+	if (dc->error == -EOPNOTSUPP)
+		dc->error = 0;
+
+	if (dc->error)
+		f2fs_msg(sbi->sb, KERN_INFO,
+			"Issue discard(%u, %u, %u) failed, ret: %d",
+			dc->lstart, dc->start, dc->len, dc->error);
+	__detach_discard_cmd(dcc, dc);
+}
+
+static void f2fs_submit_discard_endio(struct bio *bio)
+{
+	struct discard_cmd *dc = (struct discard_cmd *)bio->bi_private;
+
+	dc->error = bio->bi_error;
+	dc->state = D_DONE;
+	complete_all(&dc->wait);
+	bio_put(bio);
+}
+
+/* copied from block/blk-lib.c in 4.10-rc1 */
+static int __blkdev_issue_discard(struct block_device *bdev, sector_t sector,
+		sector_t nr_sects, gfp_t gfp_mask, int flags,
+		struct bio **biop)
+{
+	struct request_queue *q = bdev_get_queue(bdev);
+	struct bio *bio = *biop;
+	unsigned int granularity;
+	int op = REQ_WRITE | REQ_DISCARD;
+	int alignment;
+	sector_t bs_mask;
+
+	if (!q)
+		return -ENXIO;
+
+	if (!blk_queue_discard(q))
+		return -EOPNOTSUPP;
+
+	if (flags & BLKDEV_DISCARD_SECURE) {
+		if (!blk_queue_secdiscard(q))
+			return -EOPNOTSUPP;
+		op |= REQ_SECURE;
+	}
+
+	bs_mask = (bdev_logical_block_size(bdev) >> 9) - 1;
+	if ((sector | nr_sects) & bs_mask)
+		return -EINVAL;
+
+	/* Zero-sector (unknown) and one-sector granularities are the same.  */
+	granularity = max(q->limits.discard_granularity >> 9, 1U);
+	alignment = (bdev_discard_alignment(bdev) >> 9) % granularity;
+
+	while (nr_sects) {
+		unsigned int req_sects;
+		sector_t end_sect, tmp;
+
+		/* Make sure bi_size doesn't overflow */
+		req_sects = min_t(sector_t, nr_sects, UINT_MAX >> 9);
+
+		/**
+		 * If splitting a request, and the next starting sector would be
+		 * misaligned, stop the discard at the previous aligned sector.
+		 */
+		end_sect = sector + req_sects;
+		tmp = end_sect;
+		if (req_sects < nr_sects &&
+		    sector_div(tmp, granularity) != alignment) {
+			end_sect = end_sect - alignment;
+			sector_div(end_sect, granularity);
+			end_sect = end_sect * granularity + alignment;
+			req_sects = end_sect - sector;
+		}
+
+		if (bio) {
+			int ret = submit_bio_wait(op, bio);
+			bio_put(bio);
+			if (ret)
+				return ret;
+		}
+		bio = f2fs_bio_alloc(1);
+		bio->bi_iter.bi_sector = sector;
+		bio->bi_bdev = bdev;
+		bio_set_op_attrs(bio, op, 0);
+
+		bio->bi_iter.bi_size = req_sects << 9;
+		nr_sects -= req_sects;
+		sector = end_sect;
+
+		/*
+		 * We can loop for a long time in here, if someone does
+		 * full device discards (like mkfs). Be nice and allow
+		 * us to schedule out to avoid softlocking if preempt
+		 * is disabled.
+		 */
+		cond_resched();
 	}
-	trace_f2fs_issue_discard(sbi->sb, blkstart, blklen);
-	return blkdev_issue_discard(sbi->sb->s_bdev, start, len, GFP_NOFS, 0);
+
+	*biop = bio;
+	return 0;
 }
 
-bool discard_next_dnode(struct f2fs_sb_info *sbi, block_t blkaddr)
+void __check_sit_bitmap(struct f2fs_sb_info *sbi,
+				block_t start, block_t end)
 {
-	int err = -ENOTSUPP;
+#ifdef CONFIG_F2FS_CHECK_FS
+	struct seg_entry *sentry;
+	unsigned int segno;
+	block_t blk = start;
+	unsigned long offset, size, max_blocks = sbi->blocks_per_seg;
+	unsigned long *map;
 
-	if (test_opt(sbi, DISCARD)) {
-		struct seg_entry *se = get_seg_entry(sbi,
-				GET_SEGNO(sbi, blkaddr));
-		unsigned int offset = GET_BLKOFF_FROM_SEG0(sbi, blkaddr);
+	while (blk < end) {
+		segno = GET_SEGNO(sbi, blk);
+		sentry = get_seg_entry(sbi, segno);
+		offset = GET_BLKOFF_FROM_SEG0(sbi, blk);
 
-		if (f2fs_test_bit(offset, se->discard_map))
-			return false;
+		if (end < START_BLOCK(sbi, segno + 1))
+			size = GET_BLKOFF_FROM_SEG0(sbi, end);
+		else
+			size = max_blocks;
+		map = (unsigned long *)(sentry->cur_valid_map);
+		offset = __find_rev_next_bit(map, size, offset);
+		f2fs_bug_on(sbi, offset != size);
+		blk = START_BLOCK(sbi, segno + 1);
+	}
+#endif
+}
+
+/* this function is copied from blkdev_issue_discard from block/blk-lib.c */
+static void __submit_discard_cmd(struct f2fs_sb_info *sbi,
+				struct discard_cmd *dc)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct bio *bio = NULL;
+
+	if (dc->state != D_PREP)
+		return;
 
-		err = f2fs_issue_discard(sbi, blkaddr, 1);
+	trace_f2fs_issue_discard(dc->bdev, dc->start, dc->len);
+
+	dc->error = __blkdev_issue_discard(dc->bdev,
+				SECTOR_FROM_BLOCK(dc->start),
+				SECTOR_FROM_BLOCK(dc->len),
+				GFP_NOFS, 0, &bio);
+	if (!dc->error) {
+		/* should keep before submission to avoid D_DONE right away */
+		dc->state = D_SUBMIT;
+		atomic_inc(&dcc->issued_discard);
+		atomic_inc(&dcc->issing_discard);
+		if (bio) {
+			bio->bi_private = dc;
+			bio->bi_end_io = f2fs_submit_discard_endio;
+			submit_bio(REQ_SYNC, bio);
+			list_move_tail(&dc->list, &dcc->wait_list);
+			__check_sit_bitmap(sbi, dc->start, dc->start + dc->len);
+
+			f2fs_update_iostat(sbi, FS_DISCARD, 1);
+		}
+	} else {
+		__remove_discard_cmd(sbi, dc);
 	}
+}
 
-	if (err) {
-		update_meta_page(sbi, NULL, blkaddr);
-		return true;
+static struct discard_cmd *__insert_discard_tree(struct f2fs_sb_info *sbi,
+				struct block_device *bdev, block_t lstart,
+				block_t start, block_t len,
+				struct rb_node **insert_p,
+				struct rb_node *insert_parent)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct rb_node **p = &dcc->root.rb_node;
+	struct rb_node *parent = NULL;
+	struct discard_cmd *dc = NULL;
+
+	if (insert_p && insert_parent) {
+		parent = insert_parent;
+		p = insert_p;
+		goto do_insert;
 	}
-	return false;
+
+	p = __lookup_rb_tree_for_insert(sbi, &dcc->root, &parent, lstart);
+do_insert:
+	dc = __attach_discard_cmd(sbi, bdev, lstart, start, len, parent, p);
+	if (!dc)
+		return NULL;
+
+	return dc;
 }
 
-static void __add_discard_entry(struct f2fs_sb_info *sbi,
-		struct cp_control *cpc, struct seg_entry *se,
-		unsigned int start, unsigned int end)
+static void __relocate_discard_cmd(struct discard_cmd_control *dcc,
+						struct discard_cmd *dc)
 {
-	struct list_head *head = &SM_I(sbi)->discard_list;
-	struct discard_entry *new, *last;
+	list_move_tail(&dc->list, &dcc->pend_list[plist_idx(dc->len)]);
+}
+
+static void __punch_discard_cmd(struct f2fs_sb_info *sbi,
+				struct discard_cmd *dc, block_t blkaddr)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct discard_info di = dc->di;
+	bool modified = false;
+
+	if (dc->state == D_DONE || dc->len == 1) {
+		__remove_discard_cmd(sbi, dc);
+		return;
+	}
+
+	dcc->undiscard_blks -= di.len;
+
+	if (blkaddr > di.lstart) {
+		dc->len = blkaddr - dc->lstart;
+		dcc->undiscard_blks += dc->len;
+		__relocate_discard_cmd(dcc, dc);
+		modified = true;
+	}
 
-	if (!list_empty(head)) {
-		last = list_last_entry(head, struct discard_entry, list);
-		if (START_BLOCK(sbi, cpc->trim_start) + start ==
-						last->blkaddr + last->len) {
-			last->len += end - start;
-			goto done;
+	if (blkaddr < di.lstart + di.len - 1) {
+		if (modified) {
+			__insert_discard_tree(sbi, dc->bdev, blkaddr + 1,
+					di.start + blkaddr + 1 - di.lstart,
+					di.lstart + di.len - 1 - blkaddr,
+					NULL, NULL);
+		} else {
+			dc->lstart++;
+			dc->len--;
+			dc->start++;
+			dcc->undiscard_blks += dc->len;
+			__relocate_discard_cmd(dcc, dc);
 		}
 	}
+}
 
-	new = f2fs_kmem_cache_alloc(discard_entry_slab, GFP_NOFS);
-	INIT_LIST_HEAD(&new->list);
-	new->blkaddr = START_BLOCK(sbi, cpc->trim_start) + start;
-	new->len = end - start;
-	list_add_tail(&new->list, head);
-done:
-	SM_I(sbi)->nr_discards += end - start;
+static void __update_discard_tree_range(struct f2fs_sb_info *sbi,
+				struct block_device *bdev, block_t lstart,
+				block_t start, block_t len)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct discard_cmd *prev_dc = NULL, *next_dc = NULL;
+	struct discard_cmd *dc;
+	struct discard_info di = {0};
+	struct rb_node **insert_p = NULL, *insert_parent = NULL;
+	block_t end = lstart + len;
+
+	mutex_lock(&dcc->cmd_lock);
+
+	dc = (struct discard_cmd *)__lookup_rb_tree_ret(&dcc->root,
+					NULL, lstart,
+					(struct rb_entry **)&prev_dc,
+					(struct rb_entry **)&next_dc,
+					&insert_p, &insert_parent, true);
+	if (dc)
+		prev_dc = dc;
+
+	if (!prev_dc) {
+		di.lstart = lstart;
+		di.len = next_dc ? next_dc->lstart - lstart : len;
+		di.len = min(di.len, len);
+		di.start = start;
+	}
+
+	while (1) {
+		struct rb_node *node;
+		bool merged = false;
+		struct discard_cmd *tdc = NULL;
+
+		if (prev_dc) {
+			di.lstart = prev_dc->lstart + prev_dc->len;
+			if (di.lstart < lstart)
+				di.lstart = lstart;
+			if (di.lstart >= end)
+				break;
+
+			if (!next_dc || next_dc->lstart > end)
+				di.len = end - di.lstart;
+			else
+				di.len = next_dc->lstart - di.lstart;
+			di.start = start + di.lstart - lstart;
+		}
+
+		if (!di.len)
+			goto next;
+
+		if (prev_dc && prev_dc->state == D_PREP &&
+			prev_dc->bdev == bdev &&
+			__is_discard_back_mergeable(&di, &prev_dc->di)) {
+			prev_dc->di.len += di.len;
+			dcc->undiscard_blks += di.len;
+			__relocate_discard_cmd(dcc, prev_dc);
+			di = prev_dc->di;
+			tdc = prev_dc;
+			merged = true;
+		}
+
+		if (next_dc && next_dc->state == D_PREP &&
+			next_dc->bdev == bdev &&
+			__is_discard_front_mergeable(&di, &next_dc->di)) {
+			next_dc->di.lstart = di.lstart;
+			next_dc->di.len += di.len;
+			next_dc->di.start = di.start;
+			dcc->undiscard_blks += di.len;
+			__relocate_discard_cmd(dcc, next_dc);
+			if (tdc)
+				__remove_discard_cmd(sbi, tdc);
+			merged = true;
+		}
+
+		if (!merged) {
+			__insert_discard_tree(sbi, bdev, di.lstart, di.start,
+							di.len, NULL, NULL);
+		}
+ next:
+		prev_dc = next_dc;
+		if (!prev_dc)
+			break;
+
+		node = rb_next(&prev_dc->rb_node);
+		next_dc = rb_entry_safe(node, struct discard_cmd, rb_node);
+	}
+
+	mutex_unlock(&dcc->cmd_lock);
 }
 
-static void add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc)
+static int __queue_discard_cmd(struct f2fs_sb_info *sbi,
+		struct block_device *bdev, block_t blkstart, block_t blklen)
+{
+	block_t lblkstart = blkstart;
+
+	trace_f2fs_queue_discard(bdev, blkstart, blklen);
+
+	if (sbi->s_ndevs) {
+		int devi = f2fs_target_device_index(sbi, blkstart);
+
+		blkstart -= FDEV(devi).start_blk;
+	}
+	__update_discard_tree_range(sbi, bdev, lblkstart, blkstart, blklen);
+	return 0;
+}
+
+static int __issue_discard_cmd(struct f2fs_sb_info *sbi, bool issue_cond)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct list_head *pend_list;
+	struct discard_cmd *dc, *tmp;
+	struct blk_plug plug;
+	int iter = 0, issued = 0;
+	int i;
+	bool io_interrupted = false;
+
+	mutex_lock(&dcc->cmd_lock);
+	f2fs_bug_on(sbi,
+		!__check_rb_tree_consistence(sbi, &dcc->root));
+	blk_start_plug(&plug);
+	for (i = MAX_PLIST_NUM - 1;
+			i >= 0 && plist_issue(dcc->pend_list_tag[i]); i--) {
+		pend_list = &dcc->pend_list[i];
+		list_for_each_entry_safe(dc, tmp, pend_list, list) {
+			f2fs_bug_on(sbi, dc->state != D_PREP);
+
+			/* Hurry up to finish fstrim */
+			if (dcc->pend_list_tag[i] & P_TRIM) {
+				__submit_discard_cmd(sbi, dc);
+				issued++;
+
+				if (fatal_signal_pending(current))
+					break;
+				continue;
+			}
+
+			if (!issue_cond) {
+				__submit_discard_cmd(sbi, dc);
+				issued++;
+				continue;
+			}
+
+			if (is_idle(sbi)) {
+				__submit_discard_cmd(sbi, dc);
+				issued++;
+			} else {
+				io_interrupted = true;
+			}
+
+			if (++iter >= DISCARD_ISSUE_RATE)
+				goto out;
+		}
+		if (list_empty(pend_list) && dcc->pend_list_tag[i] & P_TRIM)
+			dcc->pend_list_tag[i] &= (~P_TRIM);
+	}
+out:
+	blk_finish_plug(&plug);
+	mutex_unlock(&dcc->cmd_lock);
+
+	if (!issued && io_interrupted)
+		issued = -1;
+
+	return issued;
+}
+
+static void __drop_discard_cmd(struct f2fs_sb_info *sbi)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct list_head *pend_list;
+	struct discard_cmd *dc, *tmp;
+	int i;
+
+	mutex_lock(&dcc->cmd_lock);
+	for (i = MAX_PLIST_NUM - 1; i >= 0; i--) {
+		pend_list = &dcc->pend_list[i];
+		list_for_each_entry_safe(dc, tmp, pend_list, list) {
+			f2fs_bug_on(sbi, dc->state != D_PREP);
+			__remove_discard_cmd(sbi, dc);
+		}
+	}
+	mutex_unlock(&dcc->cmd_lock);
+}
+
+static void __wait_one_discard_bio(struct f2fs_sb_info *sbi,
+							struct discard_cmd *dc)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+
+	wait_for_completion_io(&dc->wait);
+	mutex_lock(&dcc->cmd_lock);
+	f2fs_bug_on(sbi, dc->state != D_DONE);
+	dc->ref--;
+	if (!dc->ref)
+		__remove_discard_cmd(sbi, dc);
+	mutex_unlock(&dcc->cmd_lock);
+}
+
+static void __wait_discard_cmd(struct f2fs_sb_info *sbi, bool wait_cond)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct list_head *wait_list = &(dcc->wait_list);
+	struct discard_cmd *dc, *tmp;
+	bool need_wait;
+
+next:
+	need_wait = false;
+
+	mutex_lock(&dcc->cmd_lock);
+	list_for_each_entry_safe(dc, tmp, wait_list, list) {
+		if (!wait_cond || (dc->state == D_DONE && !dc->ref)) {
+			wait_for_completion_io(&dc->wait);
+			__remove_discard_cmd(sbi, dc);
+		} else {
+			dc->ref++;
+			need_wait = true;
+			break;
+		}
+	}
+	mutex_unlock(&dcc->cmd_lock);
+
+	if (need_wait) {
+		__wait_one_discard_bio(sbi, dc);
+		goto next;
+	}
+}
+
+/* This should be covered by global mutex, &sit_i->sentry_lock */
+void f2fs_wait_discard_bio(struct f2fs_sb_info *sbi, block_t blkaddr)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct discard_cmd *dc;
+	bool need_wait = false;
+
+	mutex_lock(&dcc->cmd_lock);
+	dc = (struct discard_cmd *)__lookup_rb_tree(&dcc->root, NULL, blkaddr);
+	if (dc) {
+		if (dc->state == D_PREP) {
+			__punch_discard_cmd(sbi, dc, blkaddr);
+		} else {
+			dc->ref++;
+			need_wait = true;
+		}
+	}
+	mutex_unlock(&dcc->cmd_lock);
+
+	if (need_wait)
+		__wait_one_discard_bio(sbi, dc);
+}
+
+void stop_discard_thread(struct f2fs_sb_info *sbi)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+
+	if (dcc && dcc->f2fs_issue_discard) {
+		struct task_struct *discard_thread = dcc->f2fs_issue_discard;
+
+		dcc->f2fs_issue_discard = NULL;
+		kthread_stop(discard_thread);
+	}
+}
+
+/* This comes from f2fs_put_super and f2fs_trim_fs */
+void f2fs_wait_discard_bios(struct f2fs_sb_info *sbi, bool umount)
+{
+	__issue_discard_cmd(sbi, false);
+	__drop_discard_cmd(sbi);
+	__wait_discard_cmd(sbi, !umount);
+}
+
+static void mark_discard_range_all(struct f2fs_sb_info *sbi)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	int i;
+
+	mutex_lock(&dcc->cmd_lock);
+	for (i = 0; i < MAX_PLIST_NUM; i++)
+		dcc->pend_list_tag[i] |= P_TRIM;
+	mutex_unlock(&dcc->cmd_lock);
+}
+
+static int issue_discard_thread(void *data)
+{
+	struct f2fs_sb_info *sbi = data;
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	wait_queue_head_t *q = &dcc->discard_wait_queue;
+	unsigned int wait_ms = DEF_MIN_DISCARD_ISSUE_TIME;
+	int issued;
+
+	set_freezable();
+
+	do {
+		wait_event_interruptible_timeout(*q,
+				kthread_should_stop() || freezing(current) ||
+				dcc->discard_wake,
+				msecs_to_jiffies(wait_ms));
+		if (try_to_freeze())
+			continue;
+		if (kthread_should_stop())
+			return 0;
+
+		if (dcc->discard_wake) {
+			dcc->discard_wake = 0;
+			if (sbi->gc_thread && sbi->gc_thread->gc_urgent)
+				mark_discard_range_all(sbi);
+		}
+
+		sb_start_intwrite(sbi->sb);
+
+		issued = __issue_discard_cmd(sbi, true);
+		if (issued) {
+			__wait_discard_cmd(sbi, true);
+			wait_ms = DEF_MIN_DISCARD_ISSUE_TIME;
+		} else {
+			wait_ms = DEF_MAX_DISCARD_ISSUE_TIME;
+		}
+
+		sb_end_intwrite(sbi->sb);
+
+	} while (!kthread_should_stop());
+	return 0;
+}
+
+#ifdef CONFIG_BLK_DEV_ZONED
+static int __f2fs_issue_discard_zone(struct f2fs_sb_info *sbi,
+		struct block_device *bdev, block_t blkstart, block_t blklen)
+{
+	sector_t sector, nr_sects;
+	block_t lblkstart = blkstart;
+	int devi = 0;
+
+	if (sbi->s_ndevs) {
+		devi = f2fs_target_device_index(sbi, blkstart);
+		blkstart -= FDEV(devi).start_blk;
+	}
+
+	/*
+	 * We need to know the type of the zone: for conventional zones,
+	 * use regular discard if the drive supports it. For sequential
+	 * zones, reset the zone write pointer.
+	 */
+	switch (get_blkz_type(sbi, bdev, blkstart)) {
+
+	case BLK_ZONE_TYPE_CONVENTIONAL:
+		if (!blk_queue_discard(bdev_get_queue(bdev)))
+			return 0;
+		return __queue_discard_cmd(sbi, bdev, lblkstart, blklen);
+	case BLK_ZONE_TYPE_SEQWRITE_REQ:
+	case BLK_ZONE_TYPE_SEQWRITE_PREF:
+		sector = SECTOR_FROM_BLOCK(blkstart);
+		nr_sects = SECTOR_FROM_BLOCK(blklen);
+
+		if (sector & (bdev_zone_sectors(bdev) - 1) ||
+				nr_sects != bdev_zone_sectors(bdev)) {
+			f2fs_msg(sbi->sb, KERN_INFO,
+				"(%d) %s: Unaligned discard attempted (block %x + %x)",
+				devi, sbi->s_ndevs ? FDEV(devi).path: "",
+				blkstart, blklen);
+			return -EIO;
+		}
+		trace_f2fs_issue_reset_zone(bdev, blkstart);
+		return blkdev_reset_zones(bdev, sector,
+					  nr_sects, GFP_NOFS);
+	default:
+		/* Unknown zone type: broken device ? */
+		return -EIO;
+	}
+}
+#endif
+
+static int __issue_discard_async(struct f2fs_sb_info *sbi,
+		struct block_device *bdev, block_t blkstart, block_t blklen)
+{
+#ifdef CONFIG_BLK_DEV_ZONED
+	if (f2fs_sb_mounted_blkzoned(sbi->sb) &&
+				bdev_zoned_model(bdev) != BLK_ZONED_NONE)
+		return __f2fs_issue_discard_zone(sbi, bdev, blkstart, blklen);
+#endif
+	return __queue_discard_cmd(sbi, bdev, blkstart, blklen);
+}
+
+static int f2fs_issue_discard(struct f2fs_sb_info *sbi,
+				block_t blkstart, block_t blklen)
+{
+	sector_t start = blkstart, len = 0;
+	struct block_device *bdev;
+	struct seg_entry *se;
+	unsigned int offset;
+	block_t i;
+	int err = 0;
+
+	bdev = f2fs_target_device(sbi, blkstart, NULL);
+
+	for (i = blkstart; i < blkstart + blklen; i++, len++) {
+		if (i != start) {
+			struct block_device *bdev2 =
+				f2fs_target_device(sbi, i, NULL);
+
+			if (bdev2 != bdev) {
+				err = __issue_discard_async(sbi, bdev,
+						start, len);
+				if (err)
+					return err;
+				bdev = bdev2;
+				start = i;
+				len = 0;
+			}
+		}
+
+		se = get_seg_entry(sbi, GET_SEGNO(sbi, i));
+		offset = GET_BLKOFF_FROM_SEG0(sbi, i);
+
+		if (!f2fs_test_and_set_bit(offset, se->discard_map))
+			sbi->discard_blks--;
+	}
+
+	if (len)
+		err = __issue_discard_async(sbi, bdev, start, len);
+	return err;
+}
+
+static bool add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc,
+							bool check_only)
 {
 	int entries = SIT_VBLOCK_MAP_SIZE / sizeof(unsigned long);
 	int max_blocks = sbi->blocks_per_seg;
@@ -570,16 +1459,19 @@ static void add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	unsigned long *discard_map = (unsigned long *)se->discard_map;
 	unsigned long *dmap = SIT_I(sbi)->tmp_map;
 	unsigned int start = 0, end = -1;
-	bool force = (cpc->reason == CP_DISCARD);
+	bool force = (cpc->reason & CP_DISCARD);
+	struct discard_entry *de = NULL;
+	struct list_head *head = &SM_I(sbi)->dcc_info->entry_list;
 	int i;
 
-	if (se->valid_blocks == max_blocks)
-		return;
+	if (se->valid_blocks == max_blocks || !f2fs_discard_en(sbi))
+		return false;
 
 	if (!force) {
 		if (!test_opt(sbi, DISCARD) || !se->valid_blocks ||
-		    SM_I(sbi)->nr_discards >= SM_I(sbi)->max_discards)
-			return;
+			SM_I(sbi)->dcc_info->nr_discards >=
+				SM_I(sbi)->dcc_info->max_discards)
+			return false;
 	}
 
 	/* SIT_VBLOCK_MAP_SIZE should be multiple of sizeof(unsigned long) */
@@ -587,19 +1479,38 @@ static void add_discard_addrs(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 		dmap[i] = force ? ~ckpt_map[i] & ~discard_map[i] :
 				(cur_map[i] ^ ckpt_map[i]) & ckpt_map[i];
 
-	while (force || SM_I(sbi)->nr_discards <= SM_I(sbi)->max_discards) {
+	while (force || SM_I(sbi)->dcc_info->nr_discards <=
+				SM_I(sbi)->dcc_info->max_discards) {
 		start = __find_rev_next_bit(dmap, max_blocks, end + 1);
 		if (start >= max_blocks)
 			break;
 
 		end = __find_rev_next_zero_bit(dmap, max_blocks, start + 1);
-		__add_discard_entry(sbi, cpc, se, start, end);
+		if (force && start && end != max_blocks
+					&& (end - start) < cpc->trim_minlen)
+			continue;
+
+		if (check_only)
+			return true;
+
+		if (!de) {
+			de = f2fs_kmem_cache_alloc(discard_entry_slab,
+								GFP_F2FS_ZERO);
+			de->start_blkaddr = START_BLOCK(sbi, cpc->trim_start);
+			list_add_tail(&de->list, head);
+		}
+
+		for (i = start; i < end; i++)
+			__set_bit_le(i, (void *)de->discard_map);
+
+		SM_I(sbi)->dcc_info->nr_discards += end - start;
 	}
+	return false;
 }
 
 void release_discard_addrs(struct f2fs_sb_info *sbi)
 {
-	struct list_head *head = &(SM_I(sbi)->discard_list);
+	struct list_head *head = &(SM_I(sbi)->dcc_info->entry_list);
 	struct discard_entry *entry, *this;
 
 	/* drop caches */
@@ -625,11 +1536,14 @@ static void set_prefree_as_free_segments(struct f2fs_sb_info *sbi)
 
 void clear_prefree_segments(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 {
-	struct list_head *head = &(SM_I(sbi)->discard_list);
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	struct list_head *head = &dcc->entry_list;
 	struct discard_entry *entry, *this;
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
 	unsigned long *prefree_map = dirty_i->dirty_segmap[PRE];
 	unsigned int start = 0, end = -1;
+	unsigned int secno, start_segno;
+	bool force = (cpc->reason & CP_DISCARD);
 
 	mutex_lock(&dirty_i->seglist_lock);
 
@@ -649,22 +1563,127 @@ void clear_prefree_segments(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 		if (!test_opt(sbi, DISCARD))
 			continue;
 
-		f2fs_issue_discard(sbi, START_BLOCK(sbi, start),
+		if (force && start >= cpc->trim_start &&
+					(end - 1) <= cpc->trim_end)
+				continue;
+
+		if (!test_opt(sbi, LFS) || sbi->segs_per_sec == 1) {
+			f2fs_issue_discard(sbi, START_BLOCK(sbi, start),
 				(end - start) << sbi->log_blocks_per_seg);
+			continue;
+		}
+next:
+		secno = GET_SEC_FROM_SEG(sbi, start);
+		start_segno = GET_SEG_FROM_SEC(sbi, secno);
+		if (!IS_CURSEC(sbi, secno) &&
+			!get_valid_blocks(sbi, start, true))
+			f2fs_issue_discard(sbi, START_BLOCK(sbi, start_segno),
+				sbi->segs_per_sec << sbi->log_blocks_per_seg);
+
+		start = start_segno + sbi->segs_per_sec;
+		if (start < end)
+			goto next;
+		else
+			end = start - 1;
 	}
 	mutex_unlock(&dirty_i->seglist_lock);
 
 	/* send small discards */
 	list_for_each_entry_safe(entry, this, head, list) {
-		if (cpc->reason == CP_DISCARD && entry->len < cpc->trim_minlen)
-			goto skip;
-		f2fs_issue_discard(sbi, entry->blkaddr, entry->len);
-		cpc->trimmed += entry->len;
+		unsigned int cur_pos = 0, next_pos, len, total_len = 0;
+		bool is_valid = test_bit_le(0, entry->discard_map);
+
+find_next:
+		if (is_valid) {
+			next_pos = find_next_zero_bit_le(entry->discard_map,
+					sbi->blocks_per_seg, cur_pos);
+			len = next_pos - cur_pos;
+
+			if (f2fs_sb_mounted_blkzoned(sbi->sb) ||
+			    (force && len < cpc->trim_minlen))
+				goto skip;
+
+			f2fs_issue_discard(sbi, entry->start_blkaddr + cur_pos,
+									len);
+			cpc->trimmed += len;
+			total_len += len;
+		} else {
+			next_pos = find_next_bit_le(entry->discard_map,
+					sbi->blocks_per_seg, cur_pos);
+		}
 skip:
+		cur_pos = next_pos;
+		is_valid = !is_valid;
+
+		if (cur_pos < sbi->blocks_per_seg)
+			goto find_next;
+
 		list_del(&entry->list);
-		SM_I(sbi)->nr_discards -= entry->len;
+		dcc->nr_discards -= total_len;
 		kmem_cache_free(discard_entry_slab, entry);
 	}
+
+	wake_up_discard_thread(sbi, false);
+}
+
+static int create_discard_cmd_control(struct f2fs_sb_info *sbi)
+{
+	dev_t dev = sbi->sb->s_bdev->bd_dev;
+	struct discard_cmd_control *dcc;
+	int err = 0, i;
+
+	if (SM_I(sbi)->dcc_info) {
+		dcc = SM_I(sbi)->dcc_info;
+		goto init_thread;
+	}
+
+	dcc = kzalloc(sizeof(struct discard_cmd_control), GFP_KERNEL);
+	if (!dcc)
+		return -ENOMEM;
+
+	dcc->discard_granularity = DEFAULT_DISCARD_GRANULARITY;
+	INIT_LIST_HEAD(&dcc->entry_list);
+	for (i = 0; i < MAX_PLIST_NUM; i++) {
+		INIT_LIST_HEAD(&dcc->pend_list[i]);
+		if (i >= dcc->discard_granularity - 1)
+			dcc->pend_list_tag[i] |= P_ACTIVE;
+	}
+	INIT_LIST_HEAD(&dcc->wait_list);
+	mutex_init(&dcc->cmd_lock);
+	atomic_set(&dcc->issued_discard, 0);
+	atomic_set(&dcc->issing_discard, 0);
+	atomic_set(&dcc->discard_cmd_cnt, 0);
+	dcc->nr_discards = 0;
+	dcc->max_discards = MAIN_SEGS(sbi) << sbi->log_blocks_per_seg;
+	dcc->undiscard_blks = 0;
+	dcc->root = RB_ROOT;
+
+	init_waitqueue_head(&dcc->discard_wait_queue);
+	SM_I(sbi)->dcc_info = dcc;
+init_thread:
+	dcc->f2fs_issue_discard = kthread_run(issue_discard_thread, sbi,
+				"f2fs_discard-%u:%u", MAJOR(dev), MINOR(dev));
+	if (IS_ERR(dcc->f2fs_issue_discard)) {
+		err = PTR_ERR(dcc->f2fs_issue_discard);
+		kfree(dcc);
+		SM_I(sbi)->dcc_info = NULL;
+		return err;
+	}
+
+	return err;
+}
+
+static void destroy_discard_cmd_control(struct f2fs_sb_info *sbi)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+
+	if (!dcc)
+		return;
+
+	stop_discard_thread(sbi);
+
+	kfree(dcc);
+	SM_I(sbi)->dcc_info = NULL;
 }
 
 static bool __mark_sit_entry_dirty(struct f2fs_sb_info *sbi, unsigned int segno)
@@ -693,6 +1712,10 @@ static void update_sit_entry(struct f2fs_sb_info *sbi, block_t blkaddr, int del)
 	struct seg_entry *se;
 	unsigned int segno, offset;
 	long int new_vblocks;
+	bool exist;
+#ifdef CONFIG_F2FS_CHECK_FS
+	bool mir_exist;
+#endif
 
 	segno = GET_SEGNO(sbi, blkaddr);
 
@@ -709,14 +1732,56 @@ static void update_sit_entry(struct f2fs_sb_info *sbi, block_t blkaddr, int del)
 
 	/* Update valid block bitmap */
 	if (del > 0) {
-		if (f2fs_test_and_set_bit(offset, se->cur_valid_map))
+		exist = f2fs_test_and_set_bit(offset, se->cur_valid_map);
+#ifdef CONFIG_F2FS_CHECK_FS
+		mir_exist = f2fs_test_and_set_bit(offset,
+						se->cur_valid_map_mir);
+		if (unlikely(exist != mir_exist)) {
+			f2fs_msg(sbi->sb, KERN_ERR, "Inconsistent error "
+				"when setting bitmap, blk:%u, old bit:%d",
+				blkaddr, exist);
 			f2fs_bug_on(sbi, 1);
-		if (!f2fs_test_and_set_bit(offset, se->discard_map))
+		}
+#endif
+		if (unlikely(exist)) {
+			f2fs_msg(sbi->sb, KERN_ERR,
+				"Bitmap was wrongly set, blk:%u", blkaddr);
+			f2fs_bug_on(sbi, 1);
+			se->valid_blocks--;
+			del = 0;
+		}
+
+		if (f2fs_discard_en(sbi) &&
+			!f2fs_test_and_set_bit(offset, se->discard_map))
 			sbi->discard_blks--;
+
+		/* don't overwrite by SSR to keep node chain */
+		if (se->type == CURSEG_WARM_NODE) {
+			if (!f2fs_test_and_set_bit(offset, se->ckpt_valid_map))
+				se->ckpt_valid_blocks++;
+		}
 	} else {
-		if (!f2fs_test_and_clear_bit(offset, se->cur_valid_map))
+		exist = f2fs_test_and_clear_bit(offset, se->cur_valid_map);
+#ifdef CONFIG_F2FS_CHECK_FS
+		mir_exist = f2fs_test_and_clear_bit(offset,
+						se->cur_valid_map_mir);
+		if (unlikely(exist != mir_exist)) {
+			f2fs_msg(sbi->sb, KERN_ERR, "Inconsistent error "
+				"when clearing bitmap, blk:%u, old bit:%d",
+				blkaddr, exist);
 			f2fs_bug_on(sbi, 1);
-		if (f2fs_test_and_clear_bit(offset, se->discard_map))
+		}
+#endif
+		if (unlikely(!exist)) {
+			f2fs_msg(sbi->sb, KERN_ERR,
+				"Bitmap was wrongly cleared, blk:%u", blkaddr);
+			f2fs_bug_on(sbi, 1);
+			se->valid_blocks++;
+			del = 0;
+		}
+
+		if (f2fs_discard_en(sbi) &&
+			f2fs_test_and_clear_bit(offset, se->discard_map))
 			sbi->discard_blks++;
 	}
 	if (!f2fs_test_bit(offset, se->ckpt_valid_map))
@@ -817,12 +1882,12 @@ int npages_for_summary_flush(struct f2fs_sb_info *sbi, bool for_ra)
 		}
 	}
 
-	sum_in_page = (PAGE_CACHE_SIZE - 2 * SUM_JOURNAL_SIZE -
+	sum_in_page = (PAGE_SIZE - 2 * SUM_JOURNAL_SIZE -
 			SUM_FOOTER_SIZE) / SUMMARY_SIZE;
 	if (valid_sum_count <= sum_in_page)
 		return 1;
 	else if ((valid_sum_count - sum_in_page) <=
-		(PAGE_CACHE_SIZE - SUM_FOOTER_SIZE) / SUMMARY_SIZE)
+		(PAGE_SIZE - SUM_FOOTER_SIZE) / SUMMARY_SIZE)
 		return 2;
 	return 3;
 }
@@ -841,9 +1906,9 @@ void update_meta_page(struct f2fs_sb_info *sbi, void *src, block_t blk_addr)
 	void *dst = page_address(page);
 
 	if (src)
-		memcpy(dst, src, PAGE_CACHE_SIZE);
+		memcpy(dst, src, PAGE_SIZE);
 	else
-		memset(dst, 0, PAGE_CACHE_SIZE);
+		memset(dst, 0, PAGE_SIZE);
 	set_page_dirty(page);
 	f2fs_put_page(page, 1);
 }
@@ -854,6 +1919,31 @@ static void write_sum_page(struct f2fs_sb_info *sbi,
 	update_meta_page(sbi, (void *)sum_blk, blk_addr);
 }
 
+static void write_current_sum_page(struct f2fs_sb_info *sbi,
+						int type, block_t blk_addr)
+{
+	struct curseg_info *curseg = CURSEG_I(sbi, type);
+	struct page *page = grab_meta_page(sbi, blk_addr);
+	struct f2fs_summary_block *src = curseg->sum_blk;
+	struct f2fs_summary_block *dst;
+
+	dst = (struct f2fs_summary_block *)page_address(page);
+
+	mutex_lock(&curseg->curseg_mutex);
+
+	down_read(&curseg->journal_rwsem);
+	memcpy(&dst->journal, curseg->journal, SUM_JOURNAL_SIZE);
+	up_read(&curseg->journal_rwsem);
+
+	memcpy(dst->entries, src->entries, SUM_ENTRY_SIZE);
+	memcpy(&dst->footer, &src->footer, SUM_FOOTER_SIZE);
+
+	mutex_unlock(&curseg->curseg_mutex);
+
+	set_page_dirty(page);
+	f2fs_put_page(page, 1);
+}
+
 static int is_next_segment_free(struct f2fs_sb_info *sbi, int type)
 {
 	struct curseg_info *curseg = CURSEG_I(sbi, type);
@@ -875,8 +1965,8 @@ static void get_new_segment(struct f2fs_sb_info *sbi,
 	struct free_segmap_info *free_i = FREE_I(sbi);
 	unsigned int segno, secno, zoneno;
 	unsigned int total_zones = MAIN_SECS(sbi) / sbi->secs_per_zone;
-	unsigned int hint = *newseg / sbi->segs_per_sec;
-	unsigned int old_zoneno = GET_ZONENO_FROM_SEGNO(sbi, *newseg);
+	unsigned int hint = GET_SEC_FROM_SEG(sbi, *newseg);
+	unsigned int old_zoneno = GET_ZONE_FROM_SEG(sbi, *newseg);
 	unsigned int left_start = hint;
 	bool init = true;
 	int go_left = 0;
@@ -886,9 +1976,8 @@ static void get_new_segment(struct f2fs_sb_info *sbi,
 
 	if (!new_sec && ((*newseg + 1) % sbi->segs_per_sec)) {
 		segno = find_next_zero_bit(free_i->free_segmap,
-					MAIN_SEGS(sbi), *newseg + 1);
-		if (segno - *newseg < sbi->segs_per_sec -
-					(*newseg % sbi->segs_per_sec))
+			GET_SEG_FROM_SEC(sbi, hint + 1), *newseg + 1);
+		if (segno < GET_SEG_FROM_SEC(sbi, hint + 1))
 			goto got_it;
 	}
 find_other_zone:
@@ -919,8 +2008,8 @@ find_other_zone:
 	secno = left_start;
 skip_left:
 	hint = secno;
-	segno = secno * sbi->segs_per_sec;
-	zoneno = secno / sbi->secs_per_zone;
+	segno = GET_SEG_FROM_SEC(sbi, secno);
+	zoneno = GET_ZONE_FROM_SEC(sbi, secno);
 
 	/* give up on finding another zone */
 	if (!init)
@@ -964,7 +2053,7 @@ static void reset_curseg(struct f2fs_sb_info *sbi, int type, int modified)
 	struct summary_footer *sum_footer;
 
 	curseg->segno = curseg->next_segno;
-	curseg->zone = GET_ZONENO_FROM_SEGNO(sbi, curseg->segno);
+	curseg->zone = GET_ZONE_FROM_SEG(sbi, curseg->segno);
 	curseg->next_blkoff = 0;
 	curseg->next_segno = NULL_SEGNO;
 
@@ -977,6 +2066,20 @@ static void reset_curseg(struct f2fs_sb_info *sbi, int type, int modified)
 	__set_sit_entry_type(sbi, type, curseg->segno, modified);
 }
 
+static unsigned int __get_next_segno(struct f2fs_sb_info *sbi, int type)
+{
+	/* if segs_per_sec is large than 1, we need to keep original policy. */
+	if (sbi->segs_per_sec != 1)
+		return CURSEG_I(sbi, type)->segno;
+
+	if (type == CURSEG_HOT_DATA || IS_NODESEG(type))
+		return 0;
+
+	if (SIT_I(sbi)->last_victim[ALLOC_NEXT])
+		return SIT_I(sbi)->last_victim[ALLOC_NEXT];
+	return CURSEG_I(sbi, type)->segno;
+}
+
 /*
  * Allocate a current working segment.
  * This function always allocates a free segment in LFS manner.
@@ -995,6 +2098,7 @@ static void new_curseg(struct f2fs_sb_info *sbi, int type, bool new_sec)
 	if (test_opt(sbi, NOHEAP))
 		dir = ALLOC_RIGHT;
 
+	segno = __get_next_segno(sbi, type);
 	get_new_segment(sbi, &segno, new_sec, dir);
 	curseg->next_segno = segno;
 	reset_curseg(sbi, type, 1);
@@ -1037,7 +2141,7 @@ static void __refresh_next_blkoff(struct f2fs_sb_info *sbi,
  * This function always allocates a used segment(from dirty seglist) by SSR
  * manner, so it should recover the existing segment information of valid blocks
  */
-static void change_curseg(struct f2fs_sb_info *sbi, int type, bool reuse)
+static void change_curseg(struct f2fs_sb_info *sbi, int type)
 {
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, type);
@@ -1058,28 +2162,53 @@ static void change_curseg(struct f2fs_sb_info *sbi, int type, bool reuse)
 	curseg->alloc_type = SSR;
 	__next_free_blkoff(sbi, curseg, 0);
 
-	if (reuse) {
-		sum_page = get_sum_page(sbi, new_segno);
-		sum_node = (struct f2fs_summary_block *)page_address(sum_page);
-		memcpy(curseg->sum_blk, sum_node, SUM_ENTRY_SIZE);
-		f2fs_put_page(sum_page, 1);
-	}
+	sum_page = get_sum_page(sbi, new_segno);
+	sum_node = (struct f2fs_summary_block *)page_address(sum_page);
+	memcpy(curseg->sum_blk, sum_node, SUM_ENTRY_SIZE);
+	f2fs_put_page(sum_page, 1);
 }
 
 static int get_ssr_segment(struct f2fs_sb_info *sbi, int type)
 {
 	struct curseg_info *curseg = CURSEG_I(sbi, type);
 	const struct victim_selection *v_ops = DIRTY_I(sbi)->v_ops;
+	unsigned segno = NULL_SEGNO;
+	int i, cnt;
+	bool reversed = false;
 
-	if (IS_NODESEG(type) || !has_not_enough_free_secs(sbi, 0))
-		return v_ops->get_victim(sbi,
-				&(curseg)->next_segno, BG_GC, type, SSR);
+	/* need_SSR() already forces to do this */
+	if (v_ops->get_victim(sbi, &segno, BG_GC, type, SSR)) {
+		curseg->next_segno = segno;
+		return 1;
+	}
+
+	/* For node segments, let's do SSR more intensively */
+	if (IS_NODESEG(type)) {
+		if (type >= CURSEG_WARM_NODE) {
+			reversed = true;
+			i = CURSEG_COLD_NODE;
+		} else {
+			i = CURSEG_HOT_NODE;
+		}
+		cnt = NR_CURSEG_NODE_TYPE;
+	} else {
+		if (type >= CURSEG_WARM_DATA) {
+			reversed = true;
+			i = CURSEG_COLD_DATA;
+		} else {
+			i = CURSEG_HOT_DATA;
+		}
+		cnt = NR_CURSEG_DATA_TYPE;
+	}
 
-	/* For data segments, let's do SSR more intensively */
-	for (; type >= CURSEG_HOT_DATA; type--)
-		if (v_ops->get_victim(sbi, &(curseg)->next_segno,
-						BG_GC, type, SSR))
+	for (; cnt-- > 0; reversed ? i-- : i++) {
+		if (i == type)
+			continue;
+		if (v_ops->get_victim(sbi, &segno, BG_GC, i, SSR)) {
+			curseg->next_segno = segno;
 			return 1;
+		}
+	}
 	return 0;
 }
 
@@ -1094,46 +2223,62 @@ static void allocate_segment_by_default(struct f2fs_sb_info *sbi,
 
 	if (force)
 		new_curseg(sbi, type, true);
-	else if (type == CURSEG_WARM_NODE)
+	else if (!is_set_ckpt_flags(sbi, CP_CRC_RECOVERY_FLAG) &&
+					type == CURSEG_WARM_NODE)
 		new_curseg(sbi, type, false);
 	else if (curseg->alloc_type == LFS && is_next_segment_free(sbi, type))
 		new_curseg(sbi, type, false);
 	else if (need_SSR(sbi) && get_ssr_segment(sbi, type))
-		change_curseg(sbi, type, true);
+		change_curseg(sbi, type);
 	else
 		new_curseg(sbi, type, false);
 
 	stat_inc_seg_type(sbi, curseg);
 }
 
-static void __allocate_new_segments(struct f2fs_sb_info *sbi, int type)
-{
-	struct curseg_info *curseg = CURSEG_I(sbi, type);
-	unsigned int old_segno;
-
-	old_segno = curseg->segno;
-	SIT_I(sbi)->s_ops->allocate_segment(sbi, type, true);
-	locate_dirty_segment(sbi, old_segno);
-}
-
 void allocate_new_segments(struct f2fs_sb_info *sbi)
 {
+	struct curseg_info *curseg;
+	unsigned int old_segno;
 	int i;
 
-	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++)
-		__allocate_new_segments(sbi, i);
+	for (i = CURSEG_HOT_DATA; i <= CURSEG_COLD_DATA; i++) {
+		curseg = CURSEG_I(sbi, i);
+		old_segno = curseg->segno;
+		SIT_I(sbi)->s_ops->allocate_segment(sbi, i, true);
+		locate_dirty_segment(sbi, old_segno);
+	}
 }
 
 static const struct segment_allocation default_salloc_ops = {
 	.allocate_segment = allocate_segment_by_default,
 };
 
+bool exist_trim_candidates(struct f2fs_sb_info *sbi, struct cp_control *cpc)
+{
+	__u64 trim_start = cpc->trim_start;
+	bool has_candidate = false;
+
+	mutex_lock(&SIT_I(sbi)->sentry_lock);
+	for (; cpc->trim_start <= cpc->trim_end; cpc->trim_start++) {
+		if (add_discard_addrs(sbi, cpc, true)) {
+			has_candidate = true;
+			break;
+		}
+	}
+	mutex_unlock(&SIT_I(sbi)->sentry_lock);
+
+	cpc->trim_start = trim_start;
+	return has_candidate;
+}
+
 int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 {
 	__u64 start = F2FS_BYTES_TO_BLK(range->start);
 	__u64 end = start + F2FS_BYTES_TO_BLK(range->len) - 1;
 	unsigned int start_segno, end_segno;
 	struct cp_control cpc;
+	int err = 0;
 
 	if (start >= MAX_BLKADDR(sbi) || range->len < sbi->blocksize)
 		return -EINVAL;
@@ -1142,6 +2287,12 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 	if (end <= MAIN_BLKADDR(sbi))
 		goto out;
 
+	if (is_sbi_flag_set(sbi, SBI_NEED_FSCK)) {
+		f2fs_msg(sbi->sb, KERN_WARNING,
+			"Found FS corruption, run fsck to fix.");
+		goto out;
+	}
+
 	/* start/end segment number in main_area */
 	start_segno = (start <= MAIN_BLKADDR(sbi)) ? 0 : GET_SEGNO(sbi, start);
 	end_segno = (end >= MAX_BLKADDR(sbi)) ? MAIN_SEGS(sbi) - 1 :
@@ -1164,12 +2315,19 @@ int f2fs_trim_fs(struct f2fs_sb_info *sbi, struct fstrim_range *range)
 				sbi->segs_per_sec) - 1, end_segno);
 
 		mutex_lock(&sbi->gc_mutex);
-		write_checkpoint(sbi, &cpc);
+		err = write_checkpoint(sbi, &cpc);
 		mutex_unlock(&sbi->gc_mutex);
+		if (err)
+			break;
+
+		schedule();
 	}
+	/* It's time to issue all the filed discards */
+	mark_discard_range_all(sbi);
+	f2fs_wait_discard_bios(sbi, false);
 out:
 	range->len = F2FS_BLK_TO_BYTES(cpc.trimmed);
-	return 0;
+	return err;
 }
 
 static bool __has_curseg_space(struct f2fs_sb_info *sbi, int type)
@@ -1180,87 +2338,91 @@ static bool __has_curseg_space(struct f2fs_sb_info *sbi, int type)
 	return false;
 }
 
-static int __get_segment_type_2(struct page *page, enum page_type p_type)
+static int __get_segment_type_2(struct f2fs_io_info *fio)
 {
-	if (p_type == DATA)
+	if (fio->type == DATA)
 		return CURSEG_HOT_DATA;
 	else
 		return CURSEG_HOT_NODE;
 }
 
-static int __get_segment_type_4(struct page *page, enum page_type p_type)
+static int __get_segment_type_4(struct f2fs_io_info *fio)
 {
-	if (p_type == DATA) {
-		struct inode *inode = page->mapping->host;
+	if (fio->type == DATA) {
+		struct inode *inode = fio->page->mapping->host;
 
 		if (S_ISDIR(inode->i_mode))
 			return CURSEG_HOT_DATA;
 		else
 			return CURSEG_COLD_DATA;
 	} else {
-		if (IS_DNODE(page) && is_cold_node(page))
+		if (IS_DNODE(fio->page) && is_cold_node(fio->page))
 			return CURSEG_WARM_NODE;
 		else
 			return CURSEG_COLD_NODE;
 	}
 }
 
-static int __get_segment_type_6(struct page *page, enum page_type p_type)
+static int __get_segment_type_6(struct f2fs_io_info *fio)
 {
-	if (p_type == DATA) {
-		struct inode *inode = page->mapping->host;
+	if (fio->type == DATA) {
+		struct inode *inode = fio->page->mapping->host;
 
-		if (S_ISDIR(inode->i_mode))
-			return CURSEG_HOT_DATA;
-		else if (is_cold_data(page) || file_is_cold(inode))
+		if (is_cold_data(fio->page) || file_is_cold(inode))
 			return CURSEG_COLD_DATA;
-		else
-			return CURSEG_WARM_DATA;
+		if (is_inode_flag_set(inode, FI_HOT_DATA))
+			return CURSEG_HOT_DATA;
+		return CURSEG_WARM_DATA;
 	} else {
-		if (IS_DNODE(page))
-			return is_cold_node(page) ? CURSEG_WARM_NODE :
+		if (IS_DNODE(fio->page))
+			return is_cold_node(fio->page) ? CURSEG_WARM_NODE :
 						CURSEG_HOT_NODE;
-		else
-			return CURSEG_COLD_NODE;
+		return CURSEG_COLD_NODE;
 	}
 }
 
-static int __get_segment_type(struct page *page, enum page_type p_type)
+static int __get_segment_type(struct f2fs_io_info *fio)
 {
-	switch (F2FS_P_SB(page)->active_logs) {
+	int type = 0;
+
+	switch (fio->sbi->active_logs) {
 	case 2:
-		return __get_segment_type_2(page, p_type);
+		type = __get_segment_type_2(fio);
+		break;
 	case 4:
-		return __get_segment_type_4(page, p_type);
+		type = __get_segment_type_4(fio);
+		break;
+	case 6:
+		type = __get_segment_type_6(fio);
+		break;
+	default:
+		f2fs_bug_on(fio->sbi, true);
 	}
-	/* NR_CURSEG_TYPE(6) logs by default */
-	f2fs_bug_on(F2FS_P_SB(page),
-		F2FS_P_SB(page)->active_logs != NR_CURSEG_TYPE);
-	return __get_segment_type_6(page, p_type);
+
+	if (IS_HOT(type))
+		fio->temp = HOT;
+	else if (IS_WARM(type))
+		fio->temp = WARM;
+	else
+		fio->temp = COLD;
+	return type;
 }
 
 void allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 		block_t old_blkaddr, block_t *new_blkaddr,
-		struct f2fs_summary *sum, int type)
+		struct f2fs_summary *sum, int type,
+		struct f2fs_io_info *fio, bool add_list)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	struct curseg_info *curseg;
-	bool direct_io = (type == CURSEG_DIRECT_IO);
-
-	type = direct_io ? CURSEG_WARM_DATA : type;
-
-	curseg = CURSEG_I(sbi, type);
+	struct curseg_info *curseg = CURSEG_I(sbi, type);
 
 	mutex_lock(&curseg->curseg_mutex);
 	mutex_lock(&sit_i->sentry_lock);
 
-	/* direct_io'ed data is aligned to the segment for better performance */
-	if (direct_io && curseg->next_blkoff &&
-				!has_not_enough_free_secs(sbi, 0))
-		__allocate_new_segments(sbi, type);
-
 	*new_blkaddr = NEXT_FREE_BLKADDR(sbi, curseg);
 
+	f2fs_wait_discard_bio(sbi, *new_blkaddr);
+
 	/*
 	 * __add_sum_entry should be resided under the curseg_mutex
 	 * because, this function updates a summary entry in the
@@ -1275,46 +2437,72 @@ void allocate_data_block(struct f2fs_sb_info *sbi, struct page *page,
 	if (!__has_curseg_space(sbi, type))
 		sit_i->s_ops->allocate_segment(sbi, type, false);
 	/*
-	 * SIT information should be updated before segment allocation,
-	 * since SSR needs latest valid block information.
+	 * SIT information should be updated after segment allocation,
+	 * since we need to keep dirty segments precisely under SSR.
 	 */
 	refresh_sit_entry(sbi, old_blkaddr, *new_blkaddr);
 
 	mutex_unlock(&sit_i->sentry_lock);
 
-	if (page && IS_NODESEG(type))
+	if (page && IS_NODESEG(type)) {
 		fill_node_footer_blkaddr(page, NEXT_FREE_BLKADDR(sbi, curseg));
 
+		f2fs_inode_chksum_set(sbi, page);
+	}
+
+	if (add_list) {
+		struct f2fs_bio_info *io;
+
+		INIT_LIST_HEAD(&fio->list);
+		fio->in_list = true;
+		io = sbi->write_io[fio->type] + fio->temp;
+		spin_lock(&io->io_lock);
+		list_add_tail(&fio->list, &io->io_list);
+		spin_unlock(&io->io_lock);
+	}
+
 	mutex_unlock(&curseg->curseg_mutex);
 }
 
 static void do_write_page(struct f2fs_summary *sum, struct f2fs_io_info *fio)
 {
-	int type = __get_segment_type(fio->page, fio->type);
+	int type = __get_segment_type(fio);
+	int err;
 
-	allocate_data_block(fio->sbi, fio->page, fio->blk_addr,
-					&fio->blk_addr, sum, type);
+reallocate:
+	allocate_data_block(fio->sbi, fio->page, fio->old_blkaddr,
+			&fio->new_blkaddr, sum, type, fio, true);
 
 	/* writeout dirty page into bdev */
-	f2fs_submit_page_mbio(fio);
+	err = f2fs_submit_page_write(fio);
+	if (err == -EAGAIN) {
+		fio->old_blkaddr = fio->new_blkaddr;
+		goto reallocate;
+	}
 }
 
-void write_meta_page(struct f2fs_sb_info *sbi, struct page *page)
+void write_meta_page(struct f2fs_sb_info *sbi, struct page *page,
+					enum iostat_type io_type)
 {
 	struct f2fs_io_info fio = {
 		.sbi = sbi,
 		.type = META,
-		.rw = WRITE_SYNC | REQ_META | REQ_PRIO,
-		.blk_addr = page->index,
+		.op = REQ_OP_WRITE,
+		.op_flags = REQ_SYNC | REQ_NOIDLE | REQ_META | REQ_PRIO,
+		.old_blkaddr = page->index,
+		.new_blkaddr = page->index,
 		.page = page,
 		.encrypted_page = NULL,
+		.in_list = false,
 	};
 
 	if (unlikely(page->index >= MAIN_BLKADDR(sbi)))
-		fio.rw &= ~REQ_META;
+		fio.op_flags &= ~REQ_META;
 
 	set_page_writeback(page);
-	f2fs_submit_page_mbio(&fio);
+	f2fs_submit_page_write(&fio);
+
+	f2fs_update_iostat(sbi, io_type, F2FS_BLKSIZE);
 }
 
 void write_node_page(unsigned int nid, struct f2fs_io_info *fio)
@@ -1323,6 +2511,8 @@ void write_node_page(unsigned int nid, struct f2fs_io_info *fio)
 
 	set_summary(&sum, nid, 0, 0);
 	do_write_page(&sum, fio);
+
+	f2fs_update_iostat(fio->sbi, fio->io_type, F2FS_BLKSIZE);
 }
 
 void write_data_page(struct dnode_of_data *dn, struct f2fs_io_info *fio)
@@ -1335,19 +2525,28 @@ void write_data_page(struct dnode_of_data *dn, struct f2fs_io_info *fio)
 	get_node_info(sbi, dn->nid, &ni);
 	set_summary(&sum, dn->nid, dn->ofs_in_node, ni.version);
 	do_write_page(&sum, fio);
-	dn->data_blkaddr = fio->blk_addr;
+	f2fs_update_data_blkaddr(dn, fio->new_blkaddr);
+
+	f2fs_update_iostat(sbi, fio->io_type, F2FS_BLKSIZE);
 }
 
-void rewrite_data_page(struct f2fs_io_info *fio)
+int rewrite_data_page(struct f2fs_io_info *fio)
 {
+	int err;
+
+	fio->new_blkaddr = fio->old_blkaddr;
 	stat_inc_inplace_blocks(fio->sbi);
-	f2fs_submit_page_mbio(fio);
+
+	err = f2fs_submit_page_bio(fio);
+
+	f2fs_update_iostat(fio->sbi, fio->io_type, F2FS_BLKSIZE);
+
+	return err;
 }
 
-static void __f2fs_replace_block(struct f2fs_sb_info *sbi,
-				struct f2fs_summary *sum,
+void __f2fs_replace_block(struct f2fs_sb_info *sbi, struct f2fs_summary *sum,
 				block_t old_blkaddr, block_t new_blkaddr,
-				bool recover_curseg)
+				bool recover_curseg, bool recover_newaddr)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
 	struct curseg_info *curseg;
@@ -1384,13 +2583,13 @@ static void __f2fs_replace_block(struct f2fs_sb_info *sbi,
 	/* change the current segment */
 	if (segno != curseg->segno) {
 		curseg->next_segno = segno;
-		change_curseg(sbi, type, true);
+		change_curseg(sbi, type);
 	}
 
 	curseg->next_blkoff = GET_BLKOFF_FROM_SEG0(sbi, new_blkaddr);
 	__add_sum_entry(sbi, type, sum);
 
-	if (!recover_curseg)
+	if (!recover_curseg || recover_newaddr)
 		update_sit_entry(sbi, new_blkaddr, 1);
 	if (GET_SEGNO(sbi, old_blkaddr) != NULL_SEGNO)
 		update_sit_entry(sbi, old_blkaddr, -1);
@@ -1403,7 +2602,7 @@ static void __f2fs_replace_block(struct f2fs_sb_info *sbi,
 	if (recover_curseg) {
 		if (old_cursegno != curseg->segno) {
 			curseg->next_segno = old_cursegno;
-			change_curseg(sbi, type, true);
+			change_curseg(sbi, type);
 		}
 		curseg->next_blkoff = old_blkoff;
 	}
@@ -1414,82 +2613,44 @@ static void __f2fs_replace_block(struct f2fs_sb_info *sbi,
 
 void f2fs_replace_block(struct f2fs_sb_info *sbi, struct dnode_of_data *dn,
 				block_t old_addr, block_t new_addr,
-				unsigned char version, bool recover_curseg)
+				unsigned char version, bool recover_curseg,
+				bool recover_newaddr)
 {
 	struct f2fs_summary sum;
 
 	set_summary(&sum, dn->nid, dn->ofs_in_node, version);
 
-	__f2fs_replace_block(sbi, &sum, old_addr, new_addr, recover_curseg);
+	__f2fs_replace_block(sbi, &sum, old_addr, new_addr,
+					recover_curseg, recover_newaddr);
 
-	dn->data_blkaddr = new_addr;
-	set_data_blkaddr(dn);
-	f2fs_update_extent_cache(dn);
-}
-
-static inline bool is_merged_page(struct f2fs_sb_info *sbi,
-					struct page *page, enum page_type type)
-{
-	enum page_type btype = PAGE_TYPE_OF_BIO(type);
-	struct f2fs_bio_info *io = &sbi->write_io[btype];
-	struct bio_vec *bvec;
-	struct page *target;
-	int i;
-
-	down_read(&io->io_rwsem);
-	if (!io->bio) {
-		up_read(&io->io_rwsem);
-		return false;
-	}
-
-	bio_for_each_segment_all(bvec, io->bio, i) {
-
-		if (bvec->bv_page->mapping) {
-			target = bvec->bv_page;
-		} else {
-			struct f2fs_crypto_ctx *ctx;
-
-			/* encrypted page */
-			ctx = (struct f2fs_crypto_ctx *)page_private(
-								bvec->bv_page);
-			target = ctx->w.control_page;
-		}
-
-		if (page == target) {
-			up_read(&io->io_rwsem);
-			return true;
-		}
-	}
-
-	up_read(&io->io_rwsem);
-	return false;
+	f2fs_update_data_blkaddr(dn, new_addr);
 }
 
 void f2fs_wait_on_page_writeback(struct page *page,
-				enum page_type type)
+				enum page_type type, bool ordered)
 {
 	if (PageWriteback(page)) {
 		struct f2fs_sb_info *sbi = F2FS_P_SB(page);
 
-		if (is_merged_page(sbi, page, type))
-			f2fs_submit_merged_bio(sbi, type, WRITE);
-		wait_on_page_writeback(page);
+		f2fs_submit_merged_write_cond(sbi, page->mapping->host,
+						0, page->index, type);
+		if (ordered)
+			wait_on_page_writeback(page);
+		else
+			wait_for_stable_page(page);
 	}
 }
 
-void f2fs_wait_on_encrypted_page_writeback(struct f2fs_sb_info *sbi,
-							block_t blkaddr)
+void f2fs_wait_on_block_writeback(struct f2fs_sb_info *sbi, block_t blkaddr)
 {
 	struct page *cpage;
 
-	if (blkaddr == NEW_ADDR)
+	if (blkaddr == NEW_ADDR || blkaddr == NULL_ADDR)
 		return;
 
-	f2fs_bug_on(sbi, blkaddr == NULL_ADDR);
-
 	cpage = find_lock_page(META_MAPPING(sbi), blkaddr);
 	if (cpage) {
-		f2fs_wait_on_page_writeback(cpage, DATA);
+		f2fs_wait_on_page_writeback(cpage, DATA, true);
 		f2fs_put_page(cpage, 1);
 	}
 }
@@ -1510,12 +2671,11 @@ static int read_compacted_summaries(struct f2fs_sb_info *sbi)
 
 	/* Step 1: restore nat cache */
 	seg_i = CURSEG_I(sbi, CURSEG_HOT_DATA);
-	memcpy(&seg_i->sum_blk->n_nats, kaddr, SUM_JOURNAL_SIZE);
+	memcpy(seg_i->journal, kaddr, SUM_JOURNAL_SIZE);
 
 	/* Step 2: restore sit cache */
 	seg_i = CURSEG_I(sbi, CURSEG_COLD_DATA);
-	memcpy(&seg_i->sum_blk->n_sits, kaddr + SUM_JOURNAL_SIZE,
-						SUM_JOURNAL_SIZE);
+	memcpy(seg_i->journal, kaddr + SUM_JOURNAL_SIZE, SUM_JOURNAL_SIZE);
 	offset = 2 * SUM_JOURNAL_SIZE;
 
 	/* Step 3: restore summary entries */
@@ -1539,7 +2699,7 @@ static int read_compacted_summaries(struct f2fs_sb_info *sbi)
 			s = (struct f2fs_summary *)(kaddr + offset);
 			seg_i->sum_blk->entries[j] = *s;
 			offset += SUMMARY_SIZE;
-			if (offset + SUMMARY_SIZE <= PAGE_CACHE_SIZE -
+			if (offset + SUMMARY_SIZE <= PAGE_SIZE -
 						SUM_FOOTER_SIZE)
 				continue;
 
@@ -1611,7 +2771,14 @@ static int read_normal_summaries(struct f2fs_sb_info *sbi, int type)
 	/* set uncompleted segment to curseg */
 	curseg = CURSEG_I(sbi, type);
 	mutex_lock(&curseg->curseg_mutex);
-	memcpy(curseg->sum_blk, sum, PAGE_CACHE_SIZE);
+
+	/* update journal info */
+	down_write(&curseg->journal_rwsem);
+	memcpy(curseg->journal, &sum->journal, SUM_JOURNAL_SIZE);
+	up_write(&curseg->journal_rwsem);
+
+	memcpy(curseg->sum_blk->entries, sum->entries, SUM_ENTRY_SIZE);
+	memcpy(&curseg->sum_blk->footer, &sum->footer, SUM_FOOTER_SIZE);
 	curseg->next_segno = segno;
 	reset_curseg(sbi, type, 0);
 	curseg->alloc_type = ckpt->alloc_type[type];
@@ -1623,10 +2790,12 @@ static int read_normal_summaries(struct f2fs_sb_info *sbi, int type)
 
 static int restore_curseg_summaries(struct f2fs_sb_info *sbi)
 {
+	struct f2fs_journal *sit_j = CURSEG_I(sbi, CURSEG_COLD_DATA)->journal;
+	struct f2fs_journal *nat_j = CURSEG_I(sbi, CURSEG_HOT_DATA)->journal;
 	int type = CURSEG_HOT_DATA;
 	int err;
 
-	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_COMPACT_SUM_FLAG)) {
+	if (is_set_ckpt_flags(sbi, CP_COMPACT_SUM_FLAG)) {
 		int npages = npages_for_summary_flush(sbi, true);
 
 		if (npages >= 2)
@@ -1649,6 +2818,11 @@ static int restore_curseg_summaries(struct f2fs_sb_info *sbi)
 			return err;
 	}
 
+	/* sanity check for summary blocks */
+	if (nats_in_cursum(nat_j) > NAT_JOURNAL_ENTRIES ||
+			sits_in_cursum(sit_j) > SIT_JOURNAL_ENTRIES)
+		return -EINVAL;
+
 	return 0;
 }
 
@@ -1666,13 +2840,12 @@ static void write_compacted_summaries(struct f2fs_sb_info *sbi, block_t blkaddr)
 
 	/* Step 1: write nat cache */
 	seg_i = CURSEG_I(sbi, CURSEG_HOT_DATA);
-	memcpy(kaddr, &seg_i->sum_blk->n_nats, SUM_JOURNAL_SIZE);
+	memcpy(kaddr, seg_i->journal, SUM_JOURNAL_SIZE);
 	written_size += SUM_JOURNAL_SIZE;
 
 	/* Step 2: write sit cache */
 	seg_i = CURSEG_I(sbi, CURSEG_COLD_DATA);
-	memcpy(kaddr + written_size, &seg_i->sum_blk->n_sits,
-						SUM_JOURNAL_SIZE);
+	memcpy(kaddr + written_size, seg_i->journal, SUM_JOURNAL_SIZE);
 	written_size += SUM_JOURNAL_SIZE;
 
 	/* Step 3: write summary entries */
@@ -1694,7 +2867,7 @@ static void write_compacted_summaries(struct f2fs_sb_info *sbi, block_t blkaddr)
 			*summary = seg_i->sum_blk->entries[j];
 			written_size += SUMMARY_SIZE;
 
-			if (written_size + SUMMARY_SIZE <= PAGE_CACHE_SIZE -
+			if (written_size + SUMMARY_SIZE <= PAGE_SIZE -
 							SUM_FOOTER_SIZE)
 				continue;
 
@@ -1718,17 +2891,13 @@ static void write_normal_summaries(struct f2fs_sb_info *sbi,
 	else
 		end = type + NR_CURSEG_NODE_TYPE;
 
-	for (i = type; i < end; i++) {
-		struct curseg_info *sum = CURSEG_I(sbi, i);
-		mutex_lock(&sum->curseg_mutex);
-		write_sum_page(sbi, sum->sum_blk, blkaddr + (i - type));
-		mutex_unlock(&sum->curseg_mutex);
-	}
+	for (i = type; i < end; i++)
+		write_current_sum_page(sbi, i, blkaddr + (i - type));
 }
 
 void write_data_summaries(struct f2fs_sb_info *sbi, block_t start_blk)
 {
-	if (is_set_ckpt_flags(F2FS_CKPT(sbi), CP_COMPACT_SUM_FLAG))
+	if (is_set_ckpt_flags(sbi, CP_COMPACT_SUM_FLAG))
 		write_compacted_summaries(sbi, start_blk);
 	else
 		write_normal_summaries(sbi, start_blk, CURSEG_HOT_DATA);
@@ -1739,24 +2908,24 @@ void write_node_summaries(struct f2fs_sb_info *sbi, block_t start_blk)
 	write_normal_summaries(sbi, start_blk, CURSEG_HOT_NODE);
 }
 
-int lookup_journal_in_cursum(struct f2fs_summary_block *sum, int type,
+int lookup_journal_in_cursum(struct f2fs_journal *journal, int type,
 					unsigned int val, int alloc)
 {
 	int i;
 
 	if (type == NAT_JOURNAL) {
-		for (i = 0; i < nats_in_cursum(sum); i++) {
-			if (le32_to_cpu(nid_in_journal(sum, i)) == val)
+		for (i = 0; i < nats_in_cursum(journal); i++) {
+			if (le32_to_cpu(nid_in_journal(journal, i)) == val)
 				return i;
 		}
-		if (alloc && nats_in_cursum(sum) < NAT_JOURNAL_ENTRIES)
-			return update_nats_in_cursum(sum, 1);
+		if (alloc && __has_cursum_space(journal, 1, NAT_JOURNAL))
+			return update_nats_in_cursum(journal, 1);
 	} else if (type == SIT_JOURNAL) {
-		for (i = 0; i < sits_in_cursum(sum); i++)
-			if (le32_to_cpu(segno_in_journal(sum, i)) == val)
+		for (i = 0; i < sits_in_cursum(journal); i++)
+			if (le32_to_cpu(segno_in_journal(journal, i)) == val)
 				return i;
-		if (alloc && sits_in_cursum(sum) < SIT_JOURNAL_ENTRIES)
-			return update_sits_in_cursum(sum, 1);
+		if (alloc && __has_cursum_space(journal, 1, SIT_JOURNAL))
+			return update_sits_in_cursum(journal, 1);
 	}
 	return -1;
 }
@@ -1785,7 +2954,7 @@ static struct page *get_next_sit_page(struct f2fs_sb_info *sbi,
 
 	src_addr = page_address(src_page);
 	dst_addr = page_address(dst_page);
-	memcpy(dst_addr, src_addr, PAGE_CACHE_SIZE);
+	memcpy(dst_addr, src_addr, PAGE_SIZE);
 
 	set_page_dirty(dst_page);
 	f2fs_put_page(src_page, 1);
@@ -1860,20 +3029,22 @@ static void add_sits_in_set(struct f2fs_sb_info *sbi)
 static void remove_sits_in_journal(struct f2fs_sb_info *sbi)
 {
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
 	int i;
 
-	for (i = sits_in_cursum(sum) - 1; i >= 0; i--) {
+	down_write(&curseg->journal_rwsem);
+	for (i = 0; i < sits_in_cursum(journal); i++) {
 		unsigned int segno;
 		bool dirtied;
 
-		segno = le32_to_cpu(segno_in_journal(sum, i));
+		segno = le32_to_cpu(segno_in_journal(journal, i));
 		dirtied = __mark_sit_entry_dirty(sbi, segno);
 
 		if (!dirtied)
 			add_sit_entry(segno, &SM_I(sbi)->sit_entry_set);
 	}
-	update_sits_in_cursum(sum, -sits_in_cursum(sum));
+	update_sits_in_cursum(journal, -i);
+	up_write(&curseg->journal_rwsem);
 }
 
 /*
@@ -1885,13 +3056,12 @@ void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	struct sit_info *sit_i = SIT_I(sbi);
 	unsigned long *bitmap = sit_i->dirty_sentries_bitmap;
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
 	struct sit_entry_set *ses, *tmp;
 	struct list_head *head = &SM_I(sbi)->sit_entry_set;
 	bool to_journal = true;
 	struct seg_entry *se;
 
-	mutex_lock(&curseg->curseg_mutex);
 	mutex_lock(&sit_i->sentry_lock);
 
 	if (!sit_i->dirty_sentries)
@@ -1908,7 +3078,7 @@ void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	 * entries, remove all entries from journal and add and account
 	 * them in sit entry set.
 	 */
-	if (!__has_cursum_space(sum, sit_i->dirty_sentries, SIT_JOURNAL))
+	if (!__has_cursum_space(journal, sit_i->dirty_sentries, SIT_JOURNAL))
 		remove_sits_in_journal(sbi);
 
 	/*
@@ -1925,10 +3095,12 @@ void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 		unsigned int segno = start_segno;
 
 		if (to_journal &&
-			!__has_cursum_space(sum, ses->entry_cnt, SIT_JOURNAL))
+			!__has_cursum_space(journal, ses->entry_cnt, SIT_JOURNAL))
 			to_journal = false;
 
-		if (!to_journal) {
+		if (to_journal) {
+			down_write(&curseg->journal_rwsem);
+		} else {
 			page = get_next_sit_page(sbi, start_segno);
 			raw_sit = page_address(page);
 		}
@@ -1940,19 +3112,19 @@ void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 			se = get_seg_entry(sbi, segno);
 
 			/* add discard candidates */
-			if (cpc->reason != CP_DISCARD) {
+			if (!(cpc->reason & CP_DISCARD)) {
 				cpc->trim_start = segno;
-				add_discard_addrs(sbi, cpc);
+				add_discard_addrs(sbi, cpc, false);
 			}
 
 			if (to_journal) {
-				offset = lookup_journal_in_cursum(sum,
+				offset = lookup_journal_in_cursum(journal,
 							SIT_JOURNAL, segno, 1);
 				f2fs_bug_on(sbi, offset < 0);
-				segno_in_journal(sum, offset) =
+				segno_in_journal(journal, offset) =
 							cpu_to_le32(segno);
 				seg_info_to_raw_sit(se,
-						&sit_in_journal(sum, offset));
+					&sit_in_journal(journal, offset));
 			} else {
 				sit_offset = SIT_ENTRY_OFFSET(sit_i, segno);
 				seg_info_to_raw_sit(se,
@@ -1964,7 +3136,9 @@ void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 			ses->entry_cnt--;
 		}
 
-		if (!to_journal)
+		if (to_journal)
+			up_write(&curseg->journal_rwsem);
+		else
 			f2fs_put_page(page, 1);
 
 		f2fs_bug_on(sbi, ses->entry_cnt);
@@ -1974,12 +3148,15 @@ void flush_sit_entries(struct f2fs_sb_info *sbi, struct cp_control *cpc)
 	f2fs_bug_on(sbi, !list_empty(head));
 	f2fs_bug_on(sbi, sit_i->dirty_sentries);
 out:
-	if (cpc->reason == CP_DISCARD) {
+	if (cpc->reason & CP_DISCARD) {
+		__u64 trim_start = cpc->trim_start;
+
 		for (; cpc->trim_start <= cpc->trim_end; cpc->trim_start++)
-			add_discard_addrs(sbi, cpc);
+			add_discard_addrs(sbi, cpc, false);
+
+		cpc->trim_start = trim_start;
 	}
 	mutex_unlock(&sit_i->sentry_lock);
-	mutex_unlock(&curseg->curseg_mutex);
 
 	set_prefree_as_free_segments(sbi);
 }
@@ -1987,10 +3164,9 @@ out:
 static int build_sit_info(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
-	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
 	struct sit_info *sit_i;
 	unsigned int sit_segs, start;
-	char *src_bitmap, *dst_bitmap;
+	char *src_bitmap;
 	unsigned int bitmap_size;
 
 	/* allocate memory for SIT information */
@@ -2000,13 +3176,13 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 
 	SM_I(sbi)->sit_info = sit_i;
 
-	sit_i->sentries = f2fs_kvzalloc(MAIN_SEGS(sbi) *
+	sit_i->sentries = kvzalloc(MAIN_SEGS(sbi) *
 					sizeof(struct seg_entry), GFP_KERNEL);
 	if (!sit_i->sentries)
 		return -ENOMEM;
 
 	bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
-	sit_i->dirty_sentries_bitmap = f2fs_kvzalloc(bitmap_size, GFP_KERNEL);
+	sit_i->dirty_sentries_bitmap = kvzalloc(bitmap_size, GFP_KERNEL);
 	if (!sit_i->dirty_sentries_bitmap)
 		return -ENOMEM;
 
@@ -2015,12 +3191,23 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 		sit_i->sentries[start].ckpt_valid_map
 			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
-		sit_i->sentries[start].discard_map
-			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
 		if (!sit_i->sentries[start].cur_valid_map ||
-				!sit_i->sentries[start].ckpt_valid_map ||
-				!sit_i->sentries[start].discard_map)
+				!sit_i->sentries[start].ckpt_valid_map)
 			return -ENOMEM;
+
+#ifdef CONFIG_F2FS_CHECK_FS
+		sit_i->sentries[start].cur_valid_map_mir
+			= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+		if (!sit_i->sentries[start].cur_valid_map_mir)
+			return -ENOMEM;
+#endif
+
+		if (f2fs_discard_en(sbi)) {
+			sit_i->sentries[start].discard_map
+				= kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
+			if (!sit_i->sentries[start].discard_map)
+				return -ENOMEM;
+		}
 	}
 
 	sit_i->tmp_map = kzalloc(SIT_VBLOCK_MAP_SIZE, GFP_KERNEL);
@@ -2028,7 +3215,7 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 		return -ENOMEM;
 
 	if (sbi->segs_per_sec > 1) {
-		sit_i->sec_entries = f2fs_kvzalloc(MAIN_SECS(sbi) *
+		sit_i->sec_entries = kvzalloc(MAIN_SECS(sbi) *
 					sizeof(struct sec_entry), GFP_KERNEL);
 		if (!sit_i->sec_entries)
 			return -ENOMEM;
@@ -2041,17 +3228,22 @@ static int build_sit_info(struct f2fs_sb_info *sbi)
 	bitmap_size = __bitmap_size(sbi, SIT_BITMAP);
 	src_bitmap = __bitmap_ptr(sbi, SIT_BITMAP);
 
-	dst_bitmap = kmemdup(src_bitmap, bitmap_size, GFP_KERNEL);
-	if (!dst_bitmap)
+	sit_i->sit_bitmap = kmemdup(src_bitmap, bitmap_size, GFP_KERNEL);
+	if (!sit_i->sit_bitmap)
 		return -ENOMEM;
 
+#ifdef CONFIG_F2FS_CHECK_FS
+	sit_i->sit_bitmap_mir = kmemdup(src_bitmap, bitmap_size, GFP_KERNEL);
+	if (!sit_i->sit_bitmap_mir)
+		return -ENOMEM;
+#endif
+
 	/* init SIT information */
 	sit_i->s_ops = &default_salloc_ops;
 
 	sit_i->sit_base_addr = le32_to_cpu(raw_super->sit_blkaddr);
 	sit_i->sit_blocks = sit_segs << sbi->log_blocks_per_seg;
-	sit_i->written_valid_blocks = le64_to_cpu(ckpt->valid_block_count);
-	sit_i->sit_bitmap = dst_bitmap;
+	sit_i->written_valid_blocks = 0;
 	sit_i->bitmap_size = bitmap_size;
 	sit_i->dirty_sentries = 0;
 	sit_i->sents_per_block = SIT_ENTRY_PER_BLOCK;
@@ -2074,12 +3266,12 @@ static int build_free_segmap(struct f2fs_sb_info *sbi)
 	SM_I(sbi)->free_info = free_i;
 
 	bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
-	free_i->free_segmap = f2fs_kvmalloc(bitmap_size, GFP_KERNEL);
+	free_i->free_segmap = kvmalloc(bitmap_size, GFP_KERNEL);
 	if (!free_i->free_segmap)
 		return -ENOMEM;
 
 	sec_bitmap_size = f2fs_bitmap_size(MAIN_SECS(sbi));
-	free_i->free_secmap = f2fs_kvmalloc(sec_bitmap_size, GFP_KERNEL);
+	free_i->free_secmap = kvmalloc(sec_bitmap_size, GFP_KERNEL);
 	if (!free_i->free_secmap)
 		return -ENOMEM;
 
@@ -2108,9 +3300,14 @@ static int build_curseg(struct f2fs_sb_info *sbi)
 
 	for (i = 0; i < NR_CURSEG_TYPE; i++) {
 		mutex_init(&array[i].curseg_mutex);
-		array[i].sum_blk = kzalloc(PAGE_CACHE_SIZE, GFP_KERNEL);
+		array[i].sum_blk = kzalloc(PAGE_SIZE, GFP_KERNEL);
 		if (!array[i].sum_blk)
 			return -ENOMEM;
+		init_rwsem(&array[i].journal_rwsem);
+		array[i].journal = kzalloc(sizeof(struct f2fs_journal),
+							GFP_KERNEL);
+		if (!array[i].journal)
+			return -ENOMEM;
 		array[i].segno = NULL_SEGNO;
 		array[i].next_blkoff = 0;
 	}
@@ -2121,54 +3318,85 @@ static void build_sit_entries(struct f2fs_sb_info *sbi)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
 	struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_COLD_DATA);
-	struct f2fs_summary_block *sum = curseg->sum_blk;
+	struct f2fs_journal *journal = curseg->journal;
+	struct seg_entry *se;
+	struct f2fs_sit_entry sit;
 	int sit_blk_cnt = SIT_BLK_CNT(sbi);
 	unsigned int i, start, end;
 	unsigned int readed, start_blk = 0;
-	int nrpages = MAX_BIO_BLOCKS(sbi);
 
 	do {
-		readed = ra_meta_pages(sbi, start_blk, nrpages, META_SIT, true);
+		readed = ra_meta_pages(sbi, start_blk, BIO_MAX_PAGES,
+							META_SIT, true);
 
 		start = start_blk * sit_i->sents_per_block;
 		end = (start_blk + readed) * sit_i->sents_per_block;
 
 		for (; start < end && start < MAIN_SEGS(sbi); start++) {
-			struct seg_entry *se = &sit_i->sentries[start];
 			struct f2fs_sit_block *sit_blk;
-			struct f2fs_sit_entry sit;
 			struct page *page;
 
-			mutex_lock(&curseg->curseg_mutex);
-			for (i = 0; i < sits_in_cursum(sum); i++) {
-				if (le32_to_cpu(segno_in_journal(sum, i))
-								== start) {
-					sit = sit_in_journal(sum, i);
-					mutex_unlock(&curseg->curseg_mutex);
-					goto got_it;
-				}
-			}
-			mutex_unlock(&curseg->curseg_mutex);
-
+			se = &sit_i->sentries[start];
 			page = get_current_sit_page(sbi, start);
 			sit_blk = (struct f2fs_sit_block *)page_address(page);
 			sit = sit_blk->entries[SIT_ENTRY_OFFSET(sit_i, start)];
 			f2fs_put_page(page, 1);
-got_it:
+
 			check_block_count(sbi, start, &sit);
 			seg_info_from_raw_sit(se, &sit);
 
 			/* build discard map only one time */
-			memcpy(se->discard_map, se->cur_valid_map, SIT_VBLOCK_MAP_SIZE);
-			sbi->discard_blks += sbi->blocks_per_seg - se->valid_blocks;
-
-			if (sbi->segs_per_sec > 1) {
-				struct sec_entry *e = get_sec_entry(sbi, start);
-				e->valid_blocks += se->valid_blocks;
+			if (f2fs_discard_en(sbi)) {
+				if (is_set_ckpt_flags(sbi, CP_TRIMMED_FLAG)) {
+					memset(se->discard_map, 0xff,
+						SIT_VBLOCK_MAP_SIZE);
+				} else {
+					memcpy(se->discard_map,
+						se->cur_valid_map,
+						SIT_VBLOCK_MAP_SIZE);
+					sbi->discard_blks +=
+						sbi->blocks_per_seg -
+						se->valid_blocks;
+				}
 			}
+
+			if (sbi->segs_per_sec > 1)
+				get_sec_entry(sbi, start)->valid_blocks +=
+							se->valid_blocks;
 		}
 		start_blk += readed;
 	} while (start_blk < sit_blk_cnt);
+
+	down_read(&curseg->journal_rwsem);
+	for (i = 0; i < sits_in_cursum(journal); i++) {
+		unsigned int old_valid_blocks;
+
+		start = le32_to_cpu(segno_in_journal(journal, i));
+		se = &sit_i->sentries[start];
+		sit = sit_in_journal(journal, i);
+
+		old_valid_blocks = se->valid_blocks;
+
+		check_block_count(sbi, start, &sit);
+		seg_info_from_raw_sit(se, &sit);
+
+		if (f2fs_discard_en(sbi)) {
+			if (is_set_ckpt_flags(sbi, CP_TRIMMED_FLAG)) {
+				memset(se->discard_map, 0xff,
+							SIT_VBLOCK_MAP_SIZE);
+			} else {
+				memcpy(se->discard_map, se->cur_valid_map,
+							SIT_VBLOCK_MAP_SIZE);
+				sbi->discard_blks += old_valid_blocks -
+							se->valid_blocks;
+			}
+		}
+
+		if (sbi->segs_per_sec > 1)
+			get_sec_entry(sbi, start)->valid_blocks +=
+				se->valid_blocks - old_valid_blocks;
+	}
+	up_read(&curseg->journal_rwsem);
 }
 
 static void init_free_segmap(struct f2fs_sb_info *sbi)
@@ -2180,6 +3408,9 @@ static void init_free_segmap(struct f2fs_sb_info *sbi)
 		struct seg_entry *sentry = get_seg_entry(sbi, start);
 		if (!sentry->valid_blocks)
 			__set_free(sbi, start);
+		else
+			SIT_I(sbi)->written_valid_blocks +=
+						sentry->valid_blocks;
 	}
 
 	/* set use the current segments */
@@ -2202,7 +3433,7 @@ static void init_dirty_segmap(struct f2fs_sb_info *sbi)
 		if (segno >= MAIN_SEGS(sbi))
 			break;
 		offset = segno + 1;
-		valid_blocks = get_valid_blocks(sbi, segno, 0);
+		valid_blocks = get_valid_blocks(sbi, segno, false);
 		if (valid_blocks == sbi->blocks_per_seg || !valid_blocks)
 			continue;
 		if (valid_blocks > sbi->blocks_per_seg) {
@@ -2220,7 +3451,7 @@ static int init_victim_secmap(struct f2fs_sb_info *sbi)
 	struct dirty_seglist_info *dirty_i = DIRTY_I(sbi);
 	unsigned int bitmap_size = f2fs_bitmap_size(MAIN_SECS(sbi));
 
-	dirty_i->victim_secmap = f2fs_kvzalloc(bitmap_size, GFP_KERNEL);
+	dirty_i->victim_secmap = kvzalloc(bitmap_size, GFP_KERNEL);
 	if (!dirty_i->victim_secmap)
 		return -ENOMEM;
 	return 0;
@@ -2242,7 +3473,7 @@ static int build_dirty_segmap(struct f2fs_sb_info *sbi)
 	bitmap_size = f2fs_bitmap_size(MAIN_SEGS(sbi));
 
 	for (i = 0; i < NR_DIRTY_TYPE; i++) {
-		dirty_i->dirty_segmap[i] = f2fs_kvzalloc(bitmap_size, GFP_KERNEL);
+		dirty_i->dirty_segmap[i] = kvzalloc(bitmap_size, GFP_KERNEL);
 		if (!dirty_i->dirty_segmap[i])
 			return -ENOMEM;
 	}
@@ -2301,24 +3532,29 @@ int build_segment_manager(struct f2fs_sb_info *sbi)
 	sm_info->ssa_blkaddr = le32_to_cpu(raw_super->ssa_blkaddr);
 	sm_info->rec_prefree_segments = sm_info->main_segments *
 					DEF_RECLAIM_PREFREE_SEGMENTS / 100;
-	sm_info->ipu_policy = 1 << F2FS_IPU_FSYNC;
+	if (sm_info->rec_prefree_segments > DEF_MAX_RECLAIM_PREFREE_SEGMENTS)
+		sm_info->rec_prefree_segments = DEF_MAX_RECLAIM_PREFREE_SEGMENTS;
+
+	if (!test_opt(sbi, LFS))
+		sm_info->ipu_policy = 1 << F2FS_IPU_FSYNC;
 	sm_info->min_ipu_util = DEF_MIN_IPU_UTIL;
 	sm_info->min_fsync_blocks = DEF_MIN_FSYNC_BLOCKS;
-
-	INIT_LIST_HEAD(&sm_info->discard_list);
-	sm_info->nr_discards = 0;
-	sm_info->max_discards = 0;
+	sm_info->min_hot_blocks = DEF_MIN_HOT_BLOCKS;
 
 	sm_info->trim_sections = DEF_BATCHED_TRIM_SECTIONS;
 
 	INIT_LIST_HEAD(&sm_info->sit_entry_set);
 
-	if (test_opt(sbi, FLUSH_MERGE) && !f2fs_readonly(sbi->sb)) {
+	if (!f2fs_readonly(sbi->sb)) {
 		err = create_flush_cmd_control(sbi);
 		if (err)
 			return err;
 	}
 
+	err = create_discard_cmd_control(sbi);
+	if (err)
+		return err;
+
 	err = build_sit_info(sbi);
 	if (err)
 		return err;
@@ -2383,8 +3619,10 @@ static void destroy_curseg(struct f2fs_sb_info *sbi)
 	if (!array)
 		return;
 	SM_I(sbi)->curseg_array = NULL;
-	for (i = 0; i < NR_CURSEG_TYPE; i++)
+	for (i = 0; i < NR_CURSEG_TYPE; i++) {
 		kfree(array[i].sum_blk);
+		kfree(array[i].journal);
+	}
 	kfree(array);
 }
 
@@ -2410,6 +3648,9 @@ static void destroy_sit_info(struct f2fs_sb_info *sbi)
 	if (sit_i->sentries) {
 		for (start = 0; start < MAIN_SEGS(sbi); start++) {
 			kfree(sit_i->sentries[start].cur_valid_map);
+#ifdef CONFIG_F2FS_CHECK_FS
+			kfree(sit_i->sentries[start].cur_valid_map_mir);
+#endif
 			kfree(sit_i->sentries[start].ckpt_valid_map);
 			kfree(sit_i->sentries[start].discard_map);
 		}
@@ -2422,6 +3663,9 @@ static void destroy_sit_info(struct f2fs_sb_info *sbi)
 
 	SM_I(sbi)->sit_info = NULL;
 	kfree(sit_i->sit_bitmap);
+#ifdef CONFIG_F2FS_CHECK_FS
+	kfree(sit_i->sit_bitmap_mir);
+#endif
 	kfree(sit_i);
 }
 
@@ -2431,7 +3675,8 @@ void destroy_segment_manager(struct f2fs_sb_info *sbi)
 
 	if (!sm_info)
 		return;
-	destroy_flush_cmd_control(sbi);
+	destroy_flush_cmd_control(sbi, true);
+	destroy_discard_cmd_control(sbi);
 	destroy_dirty_segmap(sbi);
 	destroy_curseg(sbi);
 	destroy_free_segmap(sbi);
@@ -2447,10 +3692,15 @@ int __init create_segment_manager_caches(void)
 	if (!discard_entry_slab)
 		goto fail;
 
+	discard_cmd_slab = f2fs_kmem_cache_create("discard_cmd",
+			sizeof(struct discard_cmd));
+	if (!discard_cmd_slab)
+		goto destroy_discard_entry;
+
 	sit_entry_set_slab = f2fs_kmem_cache_create("sit_entry_set",
 			sizeof(struct sit_entry_set));
 	if (!sit_entry_set_slab)
-		goto destory_discard_entry;
+		goto destroy_discard_cmd;
 
 	inmem_entry_slab = f2fs_kmem_cache_create("inmem_page_entry",
 			sizeof(struct inmem_pages));
@@ -2460,7 +3710,9 @@ int __init create_segment_manager_caches(void)
 
 destroy_sit_entry_set:
 	kmem_cache_destroy(sit_entry_set_slab);
-destory_discard_entry:
+destroy_discard_cmd:
+	kmem_cache_destroy(discard_cmd_slab);
+destroy_discard_entry:
 	kmem_cache_destroy(discard_entry_slab);
 fail:
 	return -ENOMEM;
@@ -2469,6 +3721,7 @@ fail:
 void destroy_segment_manager_caches(void)
 {
 	kmem_cache_destroy(sit_entry_set_slab);
+	kmem_cache_destroy(discard_cmd_slab);
 	kmem_cache_destroy(discard_entry_slab);
 	kmem_cache_destroy(inmem_entry_slab);
 }
diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h
index ee44d346ea44..ffa11274b0ce 100644
--- a/fs/f2fs/segment.h
+++ b/fs/f2fs/segment.h
@@ -16,80 +16,93 @@
 #define NULL_SECNO			((unsigned int)(~0))
 
 #define DEF_RECLAIM_PREFREE_SEGMENTS	5	/* 5% over total segments */
+#define DEF_MAX_RECLAIM_PREFREE_SEGMENTS	4096	/* 8GB in maximum */
+
+#define F2FS_MIN_SEGMENTS	9 /* SB + 2 (CP + SIT + NAT) + SSA + MAIN */
 
 /* L: Logical segment # in volume, R: Relative segment # in main area */
-#define GET_L2R_SEGNO(free_i, segno)	(segno - free_i->start_segno)
-#define GET_R2L_SEGNO(free_i, segno)	(segno + free_i->start_segno)
+#define GET_L2R_SEGNO(free_i, segno)	((segno) - (free_i)->start_segno)
+#define GET_R2L_SEGNO(free_i, segno)	((segno) + (free_i)->start_segno)
+
+#define IS_DATASEG(t)	((t) <= CURSEG_COLD_DATA)
+#define IS_NODESEG(t)	((t) >= CURSEG_HOT_NODE)
 
-#define IS_DATASEG(t)	(t <= CURSEG_COLD_DATA)
-#define IS_NODESEG(t)	(t >= CURSEG_HOT_NODE)
+#define IS_HOT(t)	((t) == CURSEG_HOT_NODE || (t) == CURSEG_HOT_DATA)
+#define IS_WARM(t)	((t) == CURSEG_WARM_NODE || (t) == CURSEG_WARM_DATA)
+#define IS_COLD(t)	((t) == CURSEG_COLD_NODE || (t) == CURSEG_COLD_DATA)
 
 #define IS_CURSEG(sbi, seg)						\
-	((seg == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno) ||	\
-	 (seg == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno) ||	\
-	 (seg == CURSEG_I(sbi, CURSEG_COLD_DATA)->segno) ||	\
-	 (seg == CURSEG_I(sbi, CURSEG_HOT_NODE)->segno) ||	\
-	 (seg == CURSEG_I(sbi, CURSEG_WARM_NODE)->segno) ||	\
-	 (seg == CURSEG_I(sbi, CURSEG_COLD_NODE)->segno))
+	(((seg) == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno) ||	\
+	 ((seg) == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno) ||	\
+	 ((seg) == CURSEG_I(sbi, CURSEG_COLD_DATA)->segno) ||	\
+	 ((seg) == CURSEG_I(sbi, CURSEG_HOT_NODE)->segno) ||	\
+	 ((seg) == CURSEG_I(sbi, CURSEG_WARM_NODE)->segno) ||	\
+	 ((seg) == CURSEG_I(sbi, CURSEG_COLD_NODE)->segno))
 
 #define IS_CURSEC(sbi, secno)						\
-	((secno == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno /		\
-	  sbi->segs_per_sec) ||	\
-	 (secno == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno /		\
-	  sbi->segs_per_sec) ||	\
-	 (secno == CURSEG_I(sbi, CURSEG_COLD_DATA)->segno /		\
-	  sbi->segs_per_sec) ||	\
-	 (secno == CURSEG_I(sbi, CURSEG_HOT_NODE)->segno /		\
-	  sbi->segs_per_sec) ||	\
-	 (secno == CURSEG_I(sbi, CURSEG_WARM_NODE)->segno /		\
-	  sbi->segs_per_sec) ||	\
-	 (secno == CURSEG_I(sbi, CURSEG_COLD_NODE)->segno /		\
-	  sbi->segs_per_sec))	\
+	(((secno) == CURSEG_I(sbi, CURSEG_HOT_DATA)->segno /		\
+	  (sbi)->segs_per_sec) ||	\
+	 ((secno) == CURSEG_I(sbi, CURSEG_WARM_DATA)->segno /		\
+	  (sbi)->segs_per_sec) ||	\
+	 ((secno) == CURSEG_I(sbi, CURSEG_COLD_DATA)->segno /		\
+	  (sbi)->segs_per_sec) ||	\
+	 ((secno) == CURSEG_I(sbi, CURSEG_HOT_NODE)->segno /		\
+	  (sbi)->segs_per_sec) ||	\
+	 ((secno) == CURSEG_I(sbi, CURSEG_WARM_NODE)->segno /		\
+	  (sbi)->segs_per_sec) ||	\
+	 ((secno) == CURSEG_I(sbi, CURSEG_COLD_NODE)->segno /		\
+	  (sbi)->segs_per_sec))	\
 
 #define MAIN_BLKADDR(sbi)	(SM_I(sbi)->main_blkaddr)
 #define SEG0_BLKADDR(sbi)	(SM_I(sbi)->seg0_blkaddr)
 
 #define MAIN_SEGS(sbi)	(SM_I(sbi)->main_segments)
-#define MAIN_SECS(sbi)	(sbi->total_sections)
+#define MAIN_SECS(sbi)	((sbi)->total_sections)
 
 #define TOTAL_SEGS(sbi)	(SM_I(sbi)->segment_count)
-#define TOTAL_BLKS(sbi)	(TOTAL_SEGS(sbi) << sbi->log_blocks_per_seg)
+#define TOTAL_BLKS(sbi)	(TOTAL_SEGS(sbi) << (sbi)->log_blocks_per_seg)
 
 #define MAX_BLKADDR(sbi)	(SEG0_BLKADDR(sbi) + TOTAL_BLKS(sbi))
-#define SEGMENT_SIZE(sbi)	(1ULL << (sbi->log_blocksize +		\
-					sbi->log_blocks_per_seg))
+#define SEGMENT_SIZE(sbi)	(1ULL << ((sbi)->log_blocksize +	\
+					(sbi)->log_blocks_per_seg))
 
 #define START_BLOCK(sbi, segno)	(SEG0_BLKADDR(sbi) +			\
-	 (GET_R2L_SEGNO(FREE_I(sbi), segno) << sbi->log_blocks_per_seg))
+	 (GET_R2L_SEGNO(FREE_I(sbi), segno) << (sbi)->log_blocks_per_seg))
 
 #define NEXT_FREE_BLKADDR(sbi, curseg)					\
-	(START_BLOCK(sbi, curseg->segno) + curseg->next_blkoff)
+	(START_BLOCK(sbi, (curseg)->segno) + (curseg)->next_blkoff)
 
 #define GET_SEGOFF_FROM_SEG0(sbi, blk_addr)	((blk_addr) - SEG0_BLKADDR(sbi))
 #define GET_SEGNO_FROM_SEG0(sbi, blk_addr)				\
-	(GET_SEGOFF_FROM_SEG0(sbi, blk_addr) >> sbi->log_blocks_per_seg)
+	(GET_SEGOFF_FROM_SEG0(sbi, blk_addr) >> (sbi)->log_blocks_per_seg)
 #define GET_BLKOFF_FROM_SEG0(sbi, blk_addr)				\
-	(GET_SEGOFF_FROM_SEG0(sbi, blk_addr) & (sbi->blocks_per_seg - 1))
+	(GET_SEGOFF_FROM_SEG0(sbi, blk_addr) & ((sbi)->blocks_per_seg - 1))
 
 #define GET_SEGNO(sbi, blk_addr)					\
-	(((blk_addr == NULL_ADDR) || (blk_addr == NEW_ADDR)) ?		\
+	((((blk_addr) == NULL_ADDR) || ((blk_addr) == NEW_ADDR)) ?	\
 	NULL_SEGNO : GET_L2R_SEGNO(FREE_I(sbi),			\
 		GET_SEGNO_FROM_SEG0(sbi, blk_addr)))
-#define GET_SECNO(sbi, segno)					\
-	((segno) / sbi->segs_per_sec)
-#define GET_ZONENO_FROM_SEGNO(sbi, segno)				\
-	((segno / sbi->segs_per_sec) / sbi->secs_per_zone)
+#define BLKS_PER_SEC(sbi)					\
+	((sbi)->segs_per_sec * (sbi)->blocks_per_seg)
+#define GET_SEC_FROM_SEG(sbi, segno)				\
+	((segno) / (sbi)->segs_per_sec)
+#define GET_SEG_FROM_SEC(sbi, secno)				\
+	((secno) * (sbi)->segs_per_sec)
+#define GET_ZONE_FROM_SEC(sbi, secno)				\
+	((secno) / (sbi)->secs_per_zone)
+#define GET_ZONE_FROM_SEG(sbi, segno)				\
+	GET_ZONE_FROM_SEC(sbi, GET_SEC_FROM_SEG(sbi, segno))
 
 #define GET_SUM_BLOCK(sbi, segno)				\
-	((sbi->sm_info->ssa_blkaddr) + segno)
+	((sbi)->sm_info->ssa_blkaddr + (segno))
 
 #define GET_SUM_TYPE(footer) ((footer)->entry_type)
-#define SET_SUM_TYPE(footer, type) ((footer)->entry_type = type)
+#define SET_SUM_TYPE(footer, type) ((footer)->entry_type = (type))
 
 #define SIT_ENTRY_OFFSET(sit_i, segno)					\
-	(segno % sit_i->sents_per_block)
+	((segno) % (sit_i)->sents_per_block)
 #define SIT_BLOCK_OFFSET(segno)					\
-	(segno / SIT_ENTRY_PER_BLOCK)
+	((segno) / SIT_ENTRY_PER_BLOCK)
 #define	START_SEGNO(segno)		\
 	(SIT_BLOCK_OFFSET(segno) * SIT_ENTRY_PER_BLOCK)
 #define SIT_BLK_CNT(sbi)			\
@@ -100,9 +113,7 @@
 #define SECTOR_FROM_BLOCK(blk_addr)					\
 	(((sector_t)blk_addr) << F2FS_LOG_SECTORS_PER_BLOCK)
 #define SECTOR_TO_BLOCK(sectors)					\
-	(sectors >> F2FS_LOG_SECTORS_PER_BLOCK)
-#define MAX_BIO_BLOCKS(sbi)						\
-	((int)min((int)max_hw_blocks(sbi), BIO_MAX_PAGES))
+	((sectors) >> F2FS_LOG_SECTORS_PER_BLOCK)
 
 /*
  * indicate a block allocation direction: RIGHT and LEFT.
@@ -131,7 +142,10 @@ enum {
  */
 enum {
 	GC_CB = 0,
-	GC_GREEDY
+	GC_GREEDY,
+	ALLOC_NEXT,
+	FLUSH_DEVICE,
+	MAX_GC_POLICY,
 };
 
 /*
@@ -158,16 +172,20 @@ struct victim_sel_policy {
 };
 
 struct seg_entry {
-	unsigned short valid_blocks;	/* # of valid blocks */
+	unsigned int type:6;		/* segment type like CURSEG_XXX_TYPE */
+	unsigned int valid_blocks:10;	/* # of valid blocks */
+	unsigned int ckpt_valid_blocks:10;	/* # of valid blocks last cp */
+	unsigned int padding:6;		/* padding */
 	unsigned char *cur_valid_map;	/* validity bitmap of blocks */
+#ifdef CONFIG_F2FS_CHECK_FS
+	unsigned char *cur_valid_map_mir;	/* mirror of current valid bitmap */
+#endif
 	/*
 	 * # of valid blocks and the validity bitmap stored in the the last
 	 * checkpoint pack. This information is used by the SSR mode.
 	 */
-	unsigned short ckpt_valid_blocks;
-	unsigned char *ckpt_valid_map;
+	unsigned char *ckpt_valid_map;	/* validity bitmap of blocks last cp */
 	unsigned char *discard_map;
-	unsigned char type;		/* segment type like CURSEG_XXX_TYPE */
 	unsigned long long mtime;	/* modification time of the segment */
 };
 
@@ -183,14 +201,18 @@ struct segment_allocation {
  * this value is set in page as a private data which indicate that
  * the page is atomically written, and it is in inmem_pages list.
  */
-#define ATOMIC_WRITTEN_PAGE		0x0000ffff
+#define ATOMIC_WRITTEN_PAGE		((unsigned long)-1)
+#define DUMMY_WRITTEN_PAGE		((unsigned long)-2)
 
 #define IS_ATOMIC_WRITTEN_PAGE(page)			\
 		(page_private(page) == (unsigned long)ATOMIC_WRITTEN_PAGE)
+#define IS_DUMMY_WRITTEN_PAGE(page)			\
+		(page_private(page) == (unsigned long)DUMMY_WRITTEN_PAGE)
 
 struct inmem_pages {
 	struct list_head list;
 	struct page *page;
+	block_t old_addr;		/* for revoking when fail to commit */
 };
 
 struct sit_info {
@@ -200,6 +222,9 @@ struct sit_info {
 	block_t sit_blocks;		/* # of blocks used by SIT area */
 	block_t written_valid_blocks;	/* # of valid blocks in main area */
 	char *sit_bitmap;		/* SIT bitmap pointer */
+#ifdef CONFIG_F2FS_CHECK_FS
+	char *sit_bitmap_mir;		/* SIT bitmap mirror */
+#endif
 	unsigned int bitmap_size;	/* SIT bitmap size */
 
 	unsigned long *tmp_map;			/* bitmap for temporal use */
@@ -215,6 +240,8 @@ struct sit_info {
 	unsigned long long mounted_time;	/* mount time */
 	unsigned long long min_mtime;		/* min. modification time */
 	unsigned long long max_mtime;		/* max. modification time */
+
+	unsigned int last_victim[MAX_GC_POLICY]; /* last victim segment # */
 };
 
 struct free_segmap_info {
@@ -257,6 +284,8 @@ struct victim_selection {
 struct curseg_info {
 	struct mutex curseg_mutex;		/* lock for consistency */
 	struct f2fs_summary_block *sum_blk;	/* cached summary block */
+	struct rw_semaphore journal_rwsem;	/* protect journal area */
+	struct f2fs_journal *journal;		/* cached journal info */
 	unsigned char alloc_type;		/* current allocation type */
 	unsigned int segno;			/* current segment number */
 	unsigned short next_blkoff;		/* next block offset to write */
@@ -289,17 +318,17 @@ static inline struct sec_entry *get_sec_entry(struct f2fs_sb_info *sbi,
 						unsigned int segno)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
-	return &sit_i->sec_entries[GET_SECNO(sbi, segno)];
+	return &sit_i->sec_entries[GET_SEC_FROM_SEG(sbi, segno)];
 }
 
 static inline unsigned int get_valid_blocks(struct f2fs_sb_info *sbi,
-				unsigned int segno, int section)
+				unsigned int segno, bool use_section)
 {
 	/*
 	 * In order to get # of valid blocks in a section instantly from many
 	 * segments, f2fs manages two counting structures separately.
 	 */
-	if (section > 1)
+	if (use_section && sbi->segs_per_sec > 1)
 		return get_sec_entry(sbi, segno)->valid_blocks;
 	else
 		return get_seg_entry(sbi, segno)->valid_blocks;
@@ -312,6 +341,9 @@ static inline void seg_info_from_raw_sit(struct seg_entry *se,
 	se->ckpt_valid_blocks = GET_SIT_VBLOCKS(rs);
 	memcpy(se->cur_valid_map, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
 	memcpy(se->ckpt_valid_map, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
+#ifdef CONFIG_F2FS_CHECK_FS
+	memcpy(se->cur_valid_map_mir, rs->valid_map, SIT_VBLOCK_MAP_SIZE);
+#endif
 	se->type = GET_SIT_TYPE(rs);
 	se->mtime = le64_to_cpu(rs->mtime);
 }
@@ -341,8 +373,8 @@ static inline unsigned int find_next_inuse(struct free_segmap_info *free_i,
 static inline void __set_free(struct f2fs_sb_info *sbi, unsigned int segno)
 {
 	struct free_segmap_info *free_i = FREE_I(sbi);
-	unsigned int secno = segno / sbi->segs_per_sec;
-	unsigned int start_segno = secno * sbi->segs_per_sec;
+	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
+	unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno);
 	unsigned int next;
 
 	spin_lock(&free_i->segmap_lock);
@@ -362,7 +394,8 @@ static inline void __set_inuse(struct f2fs_sb_info *sbi,
 		unsigned int segno)
 {
 	struct free_segmap_info *free_i = FREE_I(sbi);
-	unsigned int secno = segno / sbi->segs_per_sec;
+	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
+
 	set_bit(segno, free_i->free_segmap);
 	free_i->free_segments--;
 	if (!test_and_set_bit(secno, free_i->free_secmap))
@@ -373,8 +406,8 @@ static inline void __set_test_and_free(struct f2fs_sb_info *sbi,
 		unsigned int segno)
 {
 	struct free_segmap_info *free_i = FREE_I(sbi);
-	unsigned int secno = segno / sbi->segs_per_sec;
-	unsigned int start_segno = secno * sbi->segs_per_sec;
+	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
+	unsigned int start_segno = GET_SEG_FROM_SEC(sbi, secno);
 	unsigned int next;
 
 	spin_lock(&free_i->segmap_lock);
@@ -395,7 +428,8 @@ static inline void __set_test_and_inuse(struct f2fs_sb_info *sbi,
 		unsigned int segno)
 {
 	struct free_segmap_info *free_i = FREE_I(sbi);
-	unsigned int secno = segno / sbi->segs_per_sec;
+	unsigned int secno = GET_SEC_FROM_SEG(sbi, segno);
+
 	spin_lock(&free_i->segmap_lock);
 	if (!test_and_set_bit(segno, free_i->free_segmap)) {
 		free_i->free_segments--;
@@ -409,6 +443,12 @@ static inline void get_sit_bitmap(struct f2fs_sb_info *sbi,
 		void *dst_addr)
 {
 	struct sit_info *sit_i = SIT_I(sbi);
+
+#ifdef CONFIG_F2FS_CHECK_FS
+	if (memcmp(sit_i->sit_bitmap, sit_i->sit_bitmap_mir,
+						sit_i->bitmap_size))
+		f2fs_bug_on(sbi, 1);
+#endif
 	memcpy(dst_addr, sit_i->sit_bitmap, sit_i->bitmap_size);
 }
 
@@ -452,34 +492,24 @@ static inline int overprovision_segments(struct f2fs_sb_info *sbi)
 	return SM_I(sbi)->ovp_segments;
 }
 
-static inline int overprovision_sections(struct f2fs_sb_info *sbi)
-{
-	return ((unsigned int) overprovision_segments(sbi)) / sbi->segs_per_sec;
-}
-
 static inline int reserved_sections(struct f2fs_sb_info *sbi)
 {
-	return ((unsigned int) reserved_segments(sbi)) / sbi->segs_per_sec;
-}
-
-static inline bool need_SSR(struct f2fs_sb_info *sbi)
-{
-	int node_secs = get_blocktype_secs(sbi, F2FS_DIRTY_NODES);
-	int dent_secs = get_blocktype_secs(sbi, F2FS_DIRTY_DENTS);
-	return free_sections(sbi) <= (node_secs + 2 * dent_secs +
-						reserved_sections(sbi) + 1);
+	return GET_SEC_FROM_SEG(sbi, (unsigned int)reserved_segments(sbi));
 }
 
-static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi, int freed)
+static inline bool has_not_enough_free_secs(struct f2fs_sb_info *sbi,
+					int freed, int needed)
 {
 	int node_secs = get_blocktype_secs(sbi, F2FS_DIRTY_NODES);
 	int dent_secs = get_blocktype_secs(sbi, F2FS_DIRTY_DENTS);
+	int imeta_secs = get_blocktype_secs(sbi, F2FS_DIRTY_IMETA);
 
 	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
 		return false;
 
-	return (free_sections(sbi) + freed) <= (node_secs + 2 * dent_secs +
-						reserved_sections(sbi));
+	return (free_sections(sbi) + freed) <=
+		(node_secs + 2 * dent_secs + imeta_secs +
+		reserved_sections(sbi) + needed);
 }
 
 static inline bool excess_prefree_segs(struct f2fs_sb_info *sbi)
@@ -509,6 +539,7 @@ static inline int utilization(struct f2fs_sb_info *sbi)
  */
 #define DEF_MIN_IPU_UTIL	70
 #define DEF_MIN_FSYNC_BLOCKS	8
+#define DEF_MIN_HOT_BLOCKS	16
 
 enum {
 	F2FS_IPU_FORCE,
@@ -516,17 +547,22 @@ enum {
 	F2FS_IPU_UTIL,
 	F2FS_IPU_SSR_UTIL,
 	F2FS_IPU_FSYNC,
+	F2FS_IPU_ASYNC,
 };
 
-static inline bool need_inplace_update(struct inode *inode)
+static inline bool need_inplace_update_policy(struct inode *inode,
+				struct f2fs_io_info *fio)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	unsigned int policy = SM_I(sbi)->ipu_policy;
 
-	/* IPU can be done only for the user data */
-	if (S_ISDIR(inode->i_mode) || f2fs_is_atomic_file(inode))
+	if (test_opt(sbi, LFS))
 		return false;
 
+	/* if this is cold file, we should overwrite to avoid fragmentation */
+	if (file_is_cold(inode))
+		return true;
+
 	if (policy & (0x1 << F2FS_IPU_FORCE))
 		return true;
 	if (policy & (0x1 << F2FS_IPU_SSR) && need_SSR(sbi))
@@ -538,9 +574,18 @@ static inline bool need_inplace_update(struct inode *inode)
 			utilization(sbi) > SM_I(sbi)->min_ipu_util)
 		return true;
 
+	/*
+	 * IPU for rewrite async pages
+	 */
+	if (policy & (0x1 << F2FS_IPU_ASYNC) &&
+			fio && fio->op == REQ_OP_WRITE &&
+			!(fio->op_flags & REQ_SYNC) &&
+			!f2fs_encrypted_inode(inode))
+		return true;
+
 	/* this is only set during fdatasync */
 	if (policy & (0x1 << F2FS_IPU_FSYNC) &&
-			is_inode_flag_set(F2FS_I(inode), FI_NEED_IPU))
+			is_inode_flag_set(inode, FI_NEED_IPU))
 		return true;
 
 	return false;
@@ -573,8 +618,8 @@ static inline void check_seg_range(struct f2fs_sb_info *sbi, unsigned int segno)
 
 static inline void verify_block_addr(struct f2fs_sb_info *sbi, block_t blk_addr)
 {
-	f2fs_bug_on(sbi, blk_addr < SEG0_BLKADDR(sbi)
-					|| blk_addr >= MAX_BLKADDR(sbi));
+	BUG_ON(blk_addr < SEG0_BLKADDR(sbi)
+			|| blk_addr >= MAX_BLKADDR(sbi));
 }
 
 /*
@@ -618,6 +663,12 @@ static inline pgoff_t current_sit_addr(struct f2fs_sb_info *sbi,
 
 	check_seg_range(sbi, start);
 
+#ifdef CONFIG_F2FS_CHECK_FS
+	if (f2fs_test_bit(offset, sit_i->sit_bitmap) !=
+			f2fs_test_bit(offset, sit_i->sit_bitmap_mir))
+		f2fs_bug_on(sbi, 1);
+#endif
+
 	/* calculate sit block address */
 	if (f2fs_test_bit(offset, sit_i->sit_bitmap))
 		blk_addr += sit_i->sit_blocks;
@@ -643,6 +694,9 @@ static inline void set_to_next_sit(struct sit_info *sit_i, unsigned int start)
 	unsigned int block_off = SIT_BLOCK_OFFSET(start);
 
 	f2fs_change_bit(block_off, sit_i->sit_bitmap);
+#ifdef CONFIG_F2FS_CHECK_FS
+	f2fs_change_bit(block_off, sit_i->sit_bitmap_mir);
+#endif
 }
 
 static inline unsigned long long get_mtime(struct f2fs_sb_info *sbi)
@@ -673,26 +727,28 @@ static inline block_t sum_blk_addr(struct f2fs_sb_info *sbi, int base, int type)
 				- (base + 1) + type;
 }
 
-static inline bool sec_usage_check(struct f2fs_sb_info *sbi, unsigned int secno)
+static inline bool no_fggc_candidate(struct f2fs_sb_info *sbi,
+						unsigned int secno)
 {
-	if (IS_CURSEC(sbi, secno) || (sbi->cur_victim_sec == secno))
+	if (get_valid_blocks(sbi, GET_SEG_FROM_SEC(sbi, secno), true) >=
+						sbi->fggc_threshold)
 		return true;
 	return false;
 }
 
-static inline unsigned int max_hw_blocks(struct f2fs_sb_info *sbi)
+static inline bool sec_usage_check(struct f2fs_sb_info *sbi, unsigned int secno)
 {
-	struct block_device *bdev = sbi->sb->s_bdev;
-	struct request_queue *q = bdev_get_queue(bdev);
-	return SECTOR_TO_BLOCK(queue_max_sectors(q));
+	if (IS_CURSEC(sbi, secno) || (sbi->cur_victim_sec == secno))
+		return true;
+	return false;
 }
 
 /*
  * It is very important to gather dirty pages and write at once, so that we can
  * submit a big bio without interfering other data writes.
  * By default, 512 pages for directory data,
- * 512 pages (2MB) * 3 for three types of nodes, and
- * max_bio_blocks for meta are set.
+ * 512 pages (2MB) * 8 for nodes, and
+ * 256 pages * 8 for meta are set.
  */
 static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type)
 {
@@ -702,9 +758,9 @@ static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type)
 	if (type == DATA)
 		return sbi->blocks_per_seg;
 	else if (type == NODE)
-		return 3 * sbi->blocks_per_seg;
+		return 8 * sbi->blocks_per_seg;
 	else if (type == META)
-		return MAX_BIO_BLOCKS(sbi);
+		return 8 * BIO_MAX_PAGES;
 	else
 		return 0;
 }
@@ -721,14 +777,35 @@ static inline long nr_pages_to_write(struct f2fs_sb_info *sbi, int type,
 		return 0;
 
 	nr_to_write = wbc->nr_to_write;
-
-	if (type == DATA)
-		desired = 4096;
-	else if (type == NODE)
-		desired = 3 * max_hw_blocks(sbi);
-	else
-		desired = MAX_BIO_BLOCKS(sbi);
+	desired = BIO_MAX_PAGES;
+	if (type == NODE)
+		desired <<= 1;
 
 	wbc->nr_to_write = desired;
 	return desired - nr_to_write;
 }
+
+static inline void wake_up_discard_thread(struct f2fs_sb_info *sbi, bool force)
+{
+	struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+	bool wakeup = false;
+	int i;
+
+	if (force)
+		goto wake_up;
+
+	mutex_lock(&dcc->cmd_lock);
+	for (i = MAX_PLIST_NUM - 1;
+			i >= 0 && plist_issue(dcc->pend_list_tag[i]); i--) {
+		if (!list_empty(&dcc->pend_list[i])) {
+			wakeup = true;
+			break;
+		}
+	}
+	mutex_unlock(&dcc->cmd_lock);
+	if (!wakeup)
+		return;
+wake_up:
+	dcc->discard_wake = 1;
+	wake_up_interruptible_all(&dcc->discard_wait_queue);
+}
diff --git a/fs/f2fs/shrinker.c b/fs/f2fs/shrinker.c
index da0d8e0b55a5..5c60fc28ec75 100644
--- a/fs/f2fs/shrinker.c
+++ b/fs/f2fs/shrinker.c
@@ -13,6 +13,7 @@
 #include <linux/f2fs_fs.h>
 
 #include "f2fs.h"
+#include "node.h"
 
 static LIST_HEAD(f2fs_list);
 static DEFINE_SPINLOCK(f2fs_list_lock);
@@ -20,19 +21,22 @@ static unsigned int shrinker_run_no;
 
 static unsigned long __count_nat_entries(struct f2fs_sb_info *sbi)
 {
-	return NM_I(sbi)->nat_cnt - NM_I(sbi)->dirty_nat_cnt;
+	long count = NM_I(sbi)->nat_cnt - NM_I(sbi)->dirty_nat_cnt;
+
+	return count > 0 ? count : 0;
 }
 
 static unsigned long __count_free_nids(struct f2fs_sb_info *sbi)
 {
-	if (NM_I(sbi)->fcnt > NAT_ENTRY_PER_BLOCK)
-		return NM_I(sbi)->fcnt - NAT_ENTRY_PER_BLOCK;
-	return 0;
+	long count = NM_I(sbi)->nid_cnt[FREE_NID_LIST] - MAX_FREE_NIDS;
+
+	return count > 0 ? count : 0;
 }
 
 static unsigned long __count_extent_cache(struct f2fs_sb_info *sbi)
 {
-	return sbi->total_ext_tree + atomic_read(&sbi->total_ext_node);
+	return atomic_read(&sbi->total_zombie_tree) +
+				atomic_read(&sbi->total_ext_node);
 }
 
 unsigned long f2fs_shrink_count(struct shrinker *shrink,
diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c
index 4f666368aa85..482bb0333806 100644
--- a/fs/f2fs/super.c
+++ b/fs/f2fs/super.c
@@ -22,8 +22,10 @@
 #include <linux/random.h>
 #include <linux/exportfs.h>
 #include <linux/blkdev.h>
+#include <linux/quotaops.h>
 #include <linux/f2fs_fs.h>
 #include <linux/sysfs.h>
+#include <linux/quota.h>
 
 #include "f2fs.h"
 #include "node.h"
@@ -35,9 +37,37 @@
 #define CREATE_TRACE_POINTS
 #include <trace/events/f2fs.h>
 
-static struct proc_dir_entry *f2fs_proc_root;
 static struct kmem_cache *f2fs_inode_cachep;
-static struct kset *f2fs_kset;
+
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+
+char *fault_name[FAULT_MAX] = {
+	[FAULT_KMALLOC]		= "kmalloc",
+	[FAULT_PAGE_ALLOC]	= "page alloc",
+	[FAULT_ALLOC_NID]	= "alloc nid",
+	[FAULT_ORPHAN]		= "orphan",
+	[FAULT_BLOCK]		= "no more block",
+	[FAULT_DIR_DEPTH]	= "too big dir depth",
+	[FAULT_EVICT_INODE]	= "evict_inode fail",
+	[FAULT_TRUNCATE]	= "truncate fail",
+	[FAULT_IO]		= "IO error",
+	[FAULT_CHECKPOINT]	= "checkpoint error",
+};
+
+static void f2fs_build_fault_attr(struct f2fs_sb_info *sbi,
+						unsigned int rate)
+{
+	struct f2fs_fault_info *ffi = &sbi->fault_info;
+
+	if (rate) {
+		atomic_set(&ffi->inject_ops, 0);
+		ffi->inject_rate = rate;
+		ffi->inject_type = (1 << FAULT_MAX) - 1;
+	} else {
+		memset(ffi, 0, sizeof(struct f2fs_fault_info));
+	}
+}
+#endif
 
 /* f2fs-wide shrinker description */
 static struct shrinker f2fs_shrinker_info = {
@@ -51,7 +81,9 @@ enum {
 	Opt_disable_roll_forward,
 	Opt_norecovery,
 	Opt_discard,
+	Opt_nodiscard,
 	Opt_noheap,
+	Opt_heap,
 	Opt_user_xattr,
 	Opt_nouser_xattr,
 	Opt_acl,
@@ -59,14 +91,37 @@ enum {
 	Opt_active_logs,
 	Opt_disable_ext_identify,
 	Opt_inline_xattr,
+	Opt_noinline_xattr,
 	Opt_inline_data,
 	Opt_inline_dentry,
+	Opt_noinline_dentry,
 	Opt_flush_merge,
+	Opt_noflush_merge,
 	Opt_nobarrier,
 	Opt_fastboot,
 	Opt_extent_cache,
 	Opt_noextent_cache,
 	Opt_noinline_data,
+	Opt_data_flush,
+	Opt_mode,
+	Opt_io_size_bits,
+	Opt_fault_injection,
+	Opt_lazytime,
+	Opt_nolazytime,
+	Opt_quota,
+	Opt_noquota,
+	Opt_usrquota,
+	Opt_grpquota,
+	Opt_prjquota,
+	Opt_usrjquota,
+	Opt_grpjquota,
+	Opt_prjjquota,
+	Opt_offusrjquota,
+	Opt_offgrpjquota,
+	Opt_offprjjquota,
+	Opt_jqfmt_vfsold,
+	Opt_jqfmt_vfsv0,
+	Opt_jqfmt_vfsv1,
 	Opt_err,
 };
 
@@ -75,7 +130,9 @@ static match_table_t f2fs_tokens = {
 	{Opt_disable_roll_forward, "disable_roll_forward"},
 	{Opt_norecovery, "norecovery"},
 	{Opt_discard, "discard"},
+	{Opt_nodiscard, "nodiscard"},
 	{Opt_noheap, "no_heap"},
+	{Opt_heap, "heap"},
 	{Opt_user_xattr, "user_xattr"},
 	{Opt_nouser_xattr, "nouser_xattr"},
 	{Opt_acl, "acl"},
@@ -83,190 +140,156 @@ static match_table_t f2fs_tokens = {
 	{Opt_active_logs, "active_logs=%u"},
 	{Opt_disable_ext_identify, "disable_ext_identify"},
 	{Opt_inline_xattr, "inline_xattr"},
+	{Opt_noinline_xattr, "noinline_xattr"},
 	{Opt_inline_data, "inline_data"},
 	{Opt_inline_dentry, "inline_dentry"},
+	{Opt_noinline_dentry, "noinline_dentry"},
 	{Opt_flush_merge, "flush_merge"},
+	{Opt_noflush_merge, "noflush_merge"},
 	{Opt_nobarrier, "nobarrier"},
 	{Opt_fastboot, "fastboot"},
 	{Opt_extent_cache, "extent_cache"},
 	{Opt_noextent_cache, "noextent_cache"},
 	{Opt_noinline_data, "noinline_data"},
+	{Opt_data_flush, "data_flush"},
+	{Opt_mode, "mode=%s"},
+	{Opt_io_size_bits, "io_bits=%u"},
+	{Opt_fault_injection, "fault_injection=%u"},
+	{Opt_lazytime, "lazytime"},
+	{Opt_nolazytime, "nolazytime"},
+	{Opt_quota, "quota"},
+	{Opt_noquota, "noquota"},
+	{Opt_usrquota, "usrquota"},
+	{Opt_grpquota, "grpquota"},
+	{Opt_prjquota, "prjquota"},
+	{Opt_usrjquota, "usrjquota=%s"},
+	{Opt_grpjquota, "grpjquota=%s"},
+	{Opt_prjjquota, "prjjquota=%s"},
+	{Opt_offusrjquota, "usrjquota="},
+	{Opt_offgrpjquota, "grpjquota="},
+	{Opt_offprjjquota, "prjjquota="},
+	{Opt_jqfmt_vfsold, "jqfmt=vfsold"},
+	{Opt_jqfmt_vfsv0, "jqfmt=vfsv0"},
+	{Opt_jqfmt_vfsv1, "jqfmt=vfsv1"},
 	{Opt_err, NULL},
 };
 
-/* Sysfs support for f2fs */
-enum {
-	GC_THREAD,	/* struct f2fs_gc_thread */
-	SM_INFO,	/* struct f2fs_sm_info */
-	NM_INFO,	/* struct f2fs_nm_info */
-	F2FS_SBI,	/* struct f2fs_sb_info */
-};
-
-struct f2fs_attr {
-	struct attribute attr;
-	ssize_t (*show)(struct f2fs_attr *, struct f2fs_sb_info *, char *);
-	ssize_t (*store)(struct f2fs_attr *, struct f2fs_sb_info *,
-			 const char *, size_t);
-	int struct_type;
-	int offset;
-};
-
-static unsigned char *__struct_ptr(struct f2fs_sb_info *sbi, int struct_type)
-{
-	if (struct_type == GC_THREAD)
-		return (unsigned char *)sbi->gc_thread;
-	else if (struct_type == SM_INFO)
-		return (unsigned char *)SM_I(sbi);
-	else if (struct_type == NM_INFO)
-		return (unsigned char *)NM_I(sbi);
-	else if (struct_type == F2FS_SBI)
-		return (unsigned char *)sbi;
-	return NULL;
-}
-
-static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
-			struct f2fs_sb_info *sbi, char *buf)
-{
-	unsigned char *ptr = NULL;
-	unsigned int *ui;
-
-	ptr = __struct_ptr(sbi, a->struct_type);
-	if (!ptr)
-		return -EINVAL;
-
-	ui = (unsigned int *)(ptr + a->offset);
-
-	return snprintf(buf, PAGE_SIZE, "%u\n", *ui);
-}
-
-static ssize_t f2fs_sbi_store(struct f2fs_attr *a,
-			struct f2fs_sb_info *sbi,
-			const char *buf, size_t count)
+void f2fs_msg(struct super_block *sb, const char *level, const char *fmt, ...)
 {
-	unsigned char *ptr;
-	unsigned long t;
-	unsigned int *ui;
-	ssize_t ret;
-
-	ptr = __struct_ptr(sbi, a->struct_type);
-	if (!ptr)
-		return -EINVAL;
-
-	ui = (unsigned int *)(ptr + a->offset);
+	struct va_format vaf;
+	va_list args;
 
-	ret = kstrtoul(skip_spaces(buf), 0, &t);
-	if (ret < 0)
-		return ret;
-	*ui = t;
-	return count;
+	va_start(args, fmt);
+	vaf.fmt = fmt;
+	vaf.va = &args;
+	printk_ratelimited("%sF2FS-fs (%s): %pV\n", level, sb->s_id, &vaf);
+	va_end(args);
 }
 
-static ssize_t f2fs_attr_show(struct kobject *kobj,
-				struct attribute *attr, char *buf)
+static void init_once(void *foo)
 {
-	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
-								s_kobj);
-	struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr);
+	struct f2fs_inode_info *fi = (struct f2fs_inode_info *) foo;
 
-	return a->show ? a->show(a, sbi, buf) : 0;
+	inode_init_once(&fi->vfs_inode);
 }
 
-static ssize_t f2fs_attr_store(struct kobject *kobj, struct attribute *attr,
-						const char *buf, size_t len)
+#ifdef CONFIG_QUOTA
+static const char * const quotatypes[] = INITQFNAMES;
+#define QTYPE2NAME(t) (quotatypes[t])
+static int f2fs_set_qf_name(struct super_block *sb, int qtype,
+							substring_t *args)
 {
-	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
-									s_kobj);
-	struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr);
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	char *qname;
+	int ret = -EINVAL;
 
-	return a->store ? a->store(a, sbi, buf, len) : 0;
+	if (sb_any_quota_loaded(sb) && !sbi->s_qf_names[qtype]) {
+		f2fs_msg(sb, KERN_ERR,
+			"Cannot change journaled "
+			"quota options when quota turned on");
+		return -EINVAL;
+	}
+	qname = match_strdup(args);
+	if (!qname) {
+		f2fs_msg(sb, KERN_ERR,
+			"Not enough memory for storing quotafile name");
+		return -EINVAL;
+	}
+	if (sbi->s_qf_names[qtype]) {
+		if (strcmp(sbi->s_qf_names[qtype], qname) == 0)
+			ret = 0;
+		else
+			f2fs_msg(sb, KERN_ERR,
+				 "%s quota file already specified",
+				 QTYPE2NAME(qtype));
+		goto errout;
+	}
+	if (strchr(qname, '/')) {
+		f2fs_msg(sb, KERN_ERR,
+			"quotafile must be on filesystem root");
+		goto errout;
+	}
+	sbi->s_qf_names[qtype] = qname;
+	set_opt(sbi, QUOTA);
+	return 0;
+errout:
+	kfree(qname);
+	return ret;
 }
 
-static void f2fs_sb_release(struct kobject *kobj)
+static int f2fs_clear_qf_name(struct super_block *sb, int qtype)
 {
-	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
-								s_kobj);
-	complete(&sbi->s_kobj_unregister);
-}
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
 
-#define F2FS_ATTR_OFFSET(_struct_type, _name, _mode, _show, _store, _offset) \
-static struct f2fs_attr f2fs_attr_##_name = {			\
-	.attr = {.name = __stringify(_name), .mode = _mode },	\
-	.show	= _show,					\
-	.store	= _store,					\
-	.struct_type = _struct_type,				\
-	.offset = _offset					\
+	if (sb_any_quota_loaded(sb) && sbi->s_qf_names[qtype]) {
+		f2fs_msg(sb, KERN_ERR, "Cannot change journaled quota options"
+			" when quota turned on");
+		return -EINVAL;
+	}
+	kfree(sbi->s_qf_names[qtype]);
+	sbi->s_qf_names[qtype] = NULL;
+	return 0;
 }
 
-#define F2FS_RW_ATTR(struct_type, struct_name, name, elname)	\
-	F2FS_ATTR_OFFSET(struct_type, name, 0644,		\
-		f2fs_sbi_show, f2fs_sbi_store,			\
-		offsetof(struct struct_name, elname))
-
-F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_min_sleep_time, min_sleep_time);
-F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_max_sleep_time, max_sleep_time);
-F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_no_gc_sleep_time, no_gc_sleep_time);
-F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_idle, gc_idle);
-F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, reclaim_segments, rec_prefree_segments);
-F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, max_small_discards, max_discards);
-F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
-F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, ipu_policy, ipu_policy);
-F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_ipu_util, min_ipu_util);
-F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
-F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ram_thresh, ram_thresh);
-F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ra_nid_pages, ra_nid_pages);
-F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, max_victim_search, max_victim_search);
-F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, dir_level, dir_level);
-F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, cp_interval, cp_interval);
-
-#define ATTR_LIST(name) (&f2fs_attr_##name.attr)
-static struct attribute *f2fs_attrs[] = {
-	ATTR_LIST(gc_min_sleep_time),
-	ATTR_LIST(gc_max_sleep_time),
-	ATTR_LIST(gc_no_gc_sleep_time),
-	ATTR_LIST(gc_idle),
-	ATTR_LIST(reclaim_segments),
-	ATTR_LIST(max_small_discards),
-	ATTR_LIST(batched_trim_sections),
-	ATTR_LIST(ipu_policy),
-	ATTR_LIST(min_ipu_util),
-	ATTR_LIST(min_fsync_blocks),
-	ATTR_LIST(max_victim_search),
-	ATTR_LIST(dir_level),
-	ATTR_LIST(ram_thresh),
-	ATTR_LIST(ra_nid_pages),
-	ATTR_LIST(cp_interval),
-	NULL,
-};
-
-static const struct sysfs_ops f2fs_attr_ops = {
-	.show	= f2fs_attr_show,
-	.store	= f2fs_attr_store,
-};
-
-static struct kobj_type f2fs_ktype = {
-	.default_attrs	= f2fs_attrs,
-	.sysfs_ops	= &f2fs_attr_ops,
-	.release	= f2fs_sb_release,
-};
-
-void f2fs_msg(struct super_block *sb, const char *level, const char *fmt, ...)
+static int f2fs_check_quota_options(struct f2fs_sb_info *sbi)
 {
-	struct va_format vaf;
-	va_list args;
-
-	va_start(args, fmt);
-	vaf.fmt = fmt;
-	vaf.va = &args;
-	printk("%sF2FS-fs (%s): %pV\n", level, sb->s_id, &vaf);
-	va_end(args);
-}
-
-static void init_once(void *foo)
-{
-	struct f2fs_inode_info *fi = (struct f2fs_inode_info *) foo;
+	/*
+	 * We do the test below only for project quotas. 'usrquota' and
+	 * 'grpquota' mount options are allowed even without quota feature
+	 * to support legacy quotas in quota files.
+	 */
+	if (test_opt(sbi, PRJQUOTA) && !f2fs_sb_has_project_quota(sbi->sb)) {
+		f2fs_msg(sbi->sb, KERN_ERR, "Project quota feature not enabled. "
+			 "Cannot enable project quota enforcement.");
+		return -1;
+	}
+	if (sbi->s_qf_names[USRQUOTA] || sbi->s_qf_names[GRPQUOTA] ||
+			sbi->s_qf_names[PRJQUOTA]) {
+		if (test_opt(sbi, USRQUOTA) && sbi->s_qf_names[USRQUOTA])
+			clear_opt(sbi, USRQUOTA);
+
+		if (test_opt(sbi, GRPQUOTA) && sbi->s_qf_names[GRPQUOTA])
+			clear_opt(sbi, GRPQUOTA);
+
+		if (test_opt(sbi, PRJQUOTA) && sbi->s_qf_names[PRJQUOTA])
+			clear_opt(sbi, PRJQUOTA);
+
+		if (test_opt(sbi, GRPQUOTA) || test_opt(sbi, USRQUOTA) ||
+				test_opt(sbi, PRJQUOTA)) {
+			f2fs_msg(sbi->sb, KERN_ERR, "old and new quota "
+					"format mixing");
+			return -1;
+		}
 
-	inode_init_once(&fi->vfs_inode);
+		if (!sbi->s_jquota_fmt) {
+			f2fs_msg(sbi->sb, KERN_ERR, "journaled quota format "
+					"not specified");
+			return -1;
+		}
+	}
+	return 0;
 }
+#endif
 
 static int parse_options(struct super_block *sb, char *options)
 {
@@ -275,6 +298,9 @@ static int parse_options(struct super_block *sb, char *options)
 	substring_t args[MAX_OPT_ARGS];
 	char *p, *name;
 	int arg = 0;
+#ifdef CONFIG_QUOTA
+	int ret;
+#endif
 
 	if (!options)
 		return 0;
@@ -324,15 +350,26 @@ static int parse_options(struct super_block *sb, char *options)
 			q = bdev_get_queue(sb->s_bdev);
 			if (blk_queue_discard(q)) {
 				set_opt(sbi, DISCARD);
-			} else {
+			} else if (!f2fs_sb_mounted_blkzoned(sb)) {
 				f2fs_msg(sb, KERN_WARNING,
 					"mounting with \"discard\" option, but "
 					"the device does not support discard");
 			}
 			break;
+		case Opt_nodiscard:
+			if (f2fs_sb_mounted_blkzoned(sb)) {
+				f2fs_msg(sb, KERN_WARNING,
+					"discard is required for zoned block devices");
+				return -EINVAL;
+			}
+			clear_opt(sbi, DISCARD);
+			break;
 		case Opt_noheap:
 			set_opt(sbi, NOHEAP);
 			break;
+		case Opt_heap:
+			clear_opt(sbi, NOHEAP);
+			break;
 #ifdef CONFIG_F2FS_FS_XATTR
 		case Opt_user_xattr:
 			set_opt(sbi, XATTR_USER);
@@ -343,6 +380,9 @@ static int parse_options(struct super_block *sb, char *options)
 		case Opt_inline_xattr:
 			set_opt(sbi, INLINE_XATTR);
 			break;
+		case Opt_noinline_xattr:
+			clear_opt(sbi, INLINE_XATTR);
+			break;
 #else
 		case Opt_user_xattr:
 			f2fs_msg(sb, KERN_INFO,
@@ -356,6 +396,10 @@ static int parse_options(struct super_block *sb, char *options)
 			f2fs_msg(sb, KERN_INFO,
 				"inline_xattr options not supported");
 			break;
+		case Opt_noinline_xattr:
+			f2fs_msg(sb, KERN_INFO,
+				"noinline_xattr options not supported");
+			break;
 #endif
 #ifdef CONFIG_F2FS_FS_POSIX_ACL
 		case Opt_acl:
@@ -388,9 +432,15 @@ static int parse_options(struct super_block *sb, char *options)
 		case Opt_inline_dentry:
 			set_opt(sbi, INLINE_DENTRY);
 			break;
+		case Opt_noinline_dentry:
+			clear_opt(sbi, INLINE_DENTRY);
+			break;
 		case Opt_flush_merge:
 			set_opt(sbi, FLUSH_MERGE);
 			break;
+		case Opt_noflush_merge:
+			clear_opt(sbi, FLUSH_MERGE);
+			break;
 		case Opt_nobarrier:
 			set_opt(sbi, NOBARRIER);
 			break;
@@ -406,6 +456,136 @@ static int parse_options(struct super_block *sb, char *options)
 		case Opt_noinline_data:
 			clear_opt(sbi, INLINE_DATA);
 			break;
+		case Opt_data_flush:
+			set_opt(sbi, DATA_FLUSH);
+			break;
+		case Opt_mode:
+			name = match_strdup(&args[0]);
+
+			if (!name)
+				return -ENOMEM;
+			if (strlen(name) == 8 &&
+					!strncmp(name, "adaptive", 8)) {
+				if (f2fs_sb_mounted_blkzoned(sb)) {
+					f2fs_msg(sb, KERN_WARNING,
+						 "adaptive mode is not allowed with "
+						 "zoned block device feature");
+					kfree(name);
+					return -EINVAL;
+				}
+				set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
+			} else if (strlen(name) == 3 &&
+					!strncmp(name, "lfs", 3)) {
+				set_opt_mode(sbi, F2FS_MOUNT_LFS);
+			} else {
+				kfree(name);
+				return -EINVAL;
+			}
+			kfree(name);
+			break;
+		case Opt_io_size_bits:
+			if (args->from && match_int(args, &arg))
+				return -EINVAL;
+			if (arg > __ilog2_u32(BIO_MAX_PAGES)) {
+				f2fs_msg(sb, KERN_WARNING,
+					"Not support %d, larger than %d",
+					1 << arg, BIO_MAX_PAGES);
+				return -EINVAL;
+			}
+			sbi->write_io_size_bits = arg;
+			break;
+		case Opt_fault_injection:
+			if (args->from && match_int(args, &arg))
+				return -EINVAL;
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+			f2fs_build_fault_attr(sbi, arg);
+			set_opt(sbi, FAULT_INJECTION);
+#else
+			f2fs_msg(sb, KERN_INFO,
+				"FAULT_INJECTION was not selected");
+#endif
+			break;
+		case Opt_lazytime:
+			sb->s_flags |= MS_LAZYTIME;
+			break;
+		case Opt_nolazytime:
+			sb->s_flags &= ~MS_LAZYTIME;
+			break;
+#ifdef CONFIG_QUOTA
+		case Opt_quota:
+		case Opt_usrquota:
+			set_opt(sbi, USRQUOTA);
+			break;
+		case Opt_grpquota:
+			set_opt(sbi, GRPQUOTA);
+			break;
+		case Opt_prjquota:
+			set_opt(sbi, PRJQUOTA);
+			break;
+		case Opt_usrjquota:
+			ret = f2fs_set_qf_name(sb, USRQUOTA, &args[0]);
+			if (ret)
+				return ret;
+			break;
+		case Opt_grpjquota:
+			ret = f2fs_set_qf_name(sb, GRPQUOTA, &args[0]);
+			if (ret)
+				return ret;
+			break;
+		case Opt_prjjquota:
+			ret = f2fs_set_qf_name(sb, PRJQUOTA, &args[0]);
+			if (ret)
+				return ret;
+			break;
+		case Opt_offusrjquota:
+			ret = f2fs_clear_qf_name(sb, USRQUOTA);
+			if (ret)
+				return ret;
+			break;
+		case Opt_offgrpjquota:
+			ret = f2fs_clear_qf_name(sb, GRPQUOTA);
+			if (ret)
+				return ret;
+			break;
+		case Opt_offprjjquota:
+			ret = f2fs_clear_qf_name(sb, PRJQUOTA);
+			if (ret)
+				return ret;
+			break;
+		case Opt_jqfmt_vfsold:
+			sbi->s_jquota_fmt = QFMT_VFS_OLD;
+			break;
+		case Opt_jqfmt_vfsv0:
+			sbi->s_jquota_fmt = QFMT_VFS_V0;
+			break;
+		case Opt_jqfmt_vfsv1:
+			sbi->s_jquota_fmt = QFMT_VFS_V1;
+			break;
+		case Opt_noquota:
+			clear_opt(sbi, QUOTA);
+			clear_opt(sbi, USRQUOTA);
+			clear_opt(sbi, GRPQUOTA);
+			clear_opt(sbi, PRJQUOTA);
+			break;
+#else
+		case Opt_quota:
+		case Opt_usrquota:
+		case Opt_grpquota:
+		case Opt_prjquota:
+		case Opt_usrjquota:
+		case Opt_grpjquota:
+		case Opt_prjjquota:
+		case Opt_offusrjquota:
+		case Opt_offgrpjquota:
+		case Opt_offprjjquota:
+		case Opt_jqfmt_vfsold:
+		case Opt_jqfmt_vfsv0:
+		case Opt_jqfmt_vfsv1:
+		case Opt_noquota:
+			f2fs_msg(sb, KERN_INFO,
+					"quota operations not supported");
+			break;
+#endif
 		default:
 			f2fs_msg(sb, KERN_ERR,
 				"Unrecognized mount option \"%s\" or missing value",
@@ -413,6 +593,17 @@ static int parse_options(struct super_block *sb, char *options)
 			return -EINVAL;
 		}
 	}
+#ifdef CONFIG_QUOTA
+	if (f2fs_check_quota_options(sbi))
+		return -EINVAL;
+#endif
+
+	if (F2FS_IO_SIZE_BITS(sbi) && !test_opt(sbi, LFS)) {
+		f2fs_msg(sb, KERN_ERR,
+				"Should set mode=lfs with %uKB-sized IO",
+				F2FS_IO_SIZE_KB(sbi));
+		return -EINVAL;
+	}
 	return 0;
 }
 
@@ -432,25 +623,28 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb)
 	fi->i_current_depth = 1;
 	fi->i_advise = 0;
 	init_rwsem(&fi->i_sem);
+	INIT_LIST_HEAD(&fi->dirty_list);
+	INIT_LIST_HEAD(&fi->gdirty_list);
 	INIT_LIST_HEAD(&fi->inmem_pages);
 	mutex_init(&fi->inmem_lock);
-
-	set_inode_flag(fi, FI_NEW_INODE);
-
-	if (test_opt(F2FS_SB(sb), INLINE_XATTR))
-		set_inode_flag(fi, FI_INLINE_XATTR);
-
+	init_rwsem(&fi->dio_rwsem[READ]);
+	init_rwsem(&fi->dio_rwsem[WRITE]);
+	init_rwsem(&fi->i_mmap_sem);
+	init_rwsem(&fi->i_xattr_sem);
+
+#ifdef CONFIG_QUOTA
+	memset(&fi->i_dquot, 0, sizeof(fi->i_dquot));
+	fi->i_reserved_quota = 0;
+#endif
 	/* Will be used by directory only */
 	fi->i_dir_level = F2FS_SB(sb)->dir_level;
 
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-	fi->i_crypt_info = NULL;
-#endif
 	return &fi->vfs_inode;
 }
 
 static int f2fs_drop_inode(struct inode *inode)
 {
+	int ret;
 	/*
 	 * This is to avoid a deadlock condition like below.
 	 * writeback_single_inode(inode)
@@ -458,7 +652,7 @@ static int f2fs_drop_inode(struct inode *inode)
 	 *    - f2fs_gc -> iput -> evict
 	 *       - inode_wait_for_writeback(inode)
 	 */
-	if (!inode_unhashed(inode) && inode->i_state & I_SYNC) {
+	if ((!inode_unhashed(inode) && inode->i_state & I_SYNC)) {
 		if (!inode->i_nlink && !is_bad_inode(inode)) {
 			/* to avoid evict_inode call simultaneously */
 			atomic_inc(&inode->i_count);
@@ -466,30 +660,69 @@ static int f2fs_drop_inode(struct inode *inode)
 
 			/* some remained atomic pages should discarded */
 			if (f2fs_is_atomic_file(inode))
-				commit_inmem_pages(inode, true);
+				drop_inmem_pages(inode);
 
 			/* should remain fi->extent_tree for writepage */
 			f2fs_destroy_extent_node(inode);
 
 			sb_start_intwrite(inode->i_sb);
-			i_size_write(inode, 0);
+			f2fs_i_size_write(inode, 0);
 
 			if (F2FS_HAS_BLOCKS(inode))
-				f2fs_truncate(inode, true);
+				f2fs_truncate(inode);
 
 			sb_end_intwrite(inode->i_sb);
 
-#ifdef CONFIG_F2FS_FS_ENCRYPTION
-			if (F2FS_I(inode)->i_crypt_info)
-				f2fs_free_encryption_info(inode,
-					F2FS_I(inode)->i_crypt_info);
-#endif
+			fscrypt_put_encryption_info(inode, NULL);
 			spin_lock(&inode->i_lock);
 			atomic_dec(&inode->i_count);
 		}
+		trace_f2fs_drop_inode(inode, 0);
 		return 0;
 	}
-	return generic_drop_inode(inode);
+	ret = generic_drop_inode(inode);
+	trace_f2fs_drop_inode(inode, ret);
+	return ret;
+}
+
+int f2fs_inode_dirtied(struct inode *inode, bool sync)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	int ret = 0;
+
+	spin_lock(&sbi->inode_lock[DIRTY_META]);
+	if (is_inode_flag_set(inode, FI_DIRTY_INODE)) {
+		ret = 1;
+	} else {
+		set_inode_flag(inode, FI_DIRTY_INODE);
+		stat_inc_dirty_inode(sbi, DIRTY_META);
+	}
+	if (sync && list_empty(&F2FS_I(inode)->gdirty_list)) {
+		list_add_tail(&F2FS_I(inode)->gdirty_list,
+				&sbi->inode_list[DIRTY_META]);
+		inc_page_count(sbi, F2FS_DIRTY_IMETA);
+	}
+	spin_unlock(&sbi->inode_lock[DIRTY_META]);
+	return ret;
+}
+
+void f2fs_inode_synced(struct inode *inode)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+
+	spin_lock(&sbi->inode_lock[DIRTY_META]);
+	if (!is_inode_flag_set(inode, FI_DIRTY_INODE)) {
+		spin_unlock(&sbi->inode_lock[DIRTY_META]);
+		return;
+	}
+	if (!list_empty(&F2FS_I(inode)->gdirty_list)) {
+		list_del_init(&F2FS_I(inode)->gdirty_list);
+		dec_page_count(sbi, F2FS_DIRTY_IMETA);
+	}
+	clear_inode_flag(inode, FI_DIRTY_INODE);
+	clear_inode_flag(inode, FI_AUTO_RECOVER);
+	stat_dec_dirty_inode(F2FS_I_SB(inode), DIRTY_META);
+	spin_unlock(&sbi->inode_lock[DIRTY_META]);
 }
 
 /*
@@ -499,7 +732,19 @@ static int f2fs_drop_inode(struct inode *inode)
  */
 static void f2fs_dirty_inode(struct inode *inode, int flags)
 {
-	set_inode_flag(F2FS_I(inode), FI_DIRTY_INODE);
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+
+	if (inode->i_ino == F2FS_NODE_INO(sbi) ||
+			inode->i_ino == F2FS_META_INO(sbi))
+		return;
+
+	if (flags == I_DIRTY_TIME)
+		return;
+
+	if (is_inode_flag_set(inode, FI_AUTO_RECOVER))
+		clear_inode_flag(inode, FI_AUTO_RECOVER);
+
+	f2fs_inode_dirtied(inode, false);
 }
 
 static void f2fs_i_callback(struct rcu_head *head)
@@ -513,17 +758,31 @@ static void f2fs_destroy_inode(struct inode *inode)
 	call_rcu(&inode->i_rcu, f2fs_i_callback);
 }
 
-static void f2fs_put_super(struct super_block *sb)
+static void destroy_percpu_info(struct f2fs_sb_info *sbi)
 {
-	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	percpu_counter_destroy(&sbi->alloc_valid_block_count);
+	percpu_counter_destroy(&sbi->total_valid_inode_count);
+}
+
+static void destroy_device_list(struct f2fs_sb_info *sbi)
+{
+	int i;
 
-	if (sbi->s_proc) {
-		remove_proc_entry("segment_info", sbi->s_proc);
-		remove_proc_entry(sb->s_id, f2fs_proc_root);
+	for (i = 0; i < sbi->s_ndevs; i++) {
+		blkdev_put(FDEV(i).bdev, FMODE_EXCL);
+#ifdef CONFIG_BLK_DEV_ZONED
+		kfree(FDEV(i).blkz_type);
+#endif
 	}
-	kobject_del(&sbi->s_kobj);
+	kfree(sbi->devs);
+}
+
+static void f2fs_put_super(struct super_block *sb)
+{
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	int i;
 
-	stop_gc_thread(sbi);
+	f2fs_quota_off_umount(sb);
 
 	/* prevent remaining shrinker jobs */
 	mutex_lock(&sbi->umount_mutex);
@@ -534,13 +793,23 @@ static void f2fs_put_super(struct super_block *sb)
 	 * clean checkpoint again.
 	 */
 	if (is_sbi_flag_set(sbi, SBI_IS_DIRTY) ||
-			!is_set_ckpt_flags(F2FS_CKPT(sbi), CP_UMOUNT_FLAG)) {
+			!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
 		struct cp_control cpc = {
 			.reason = CP_UMOUNT,
 		};
 		write_checkpoint(sbi, &cpc);
 	}
 
+	/* be sure to wait for any on-going discard commands */
+	f2fs_wait_discard_bios(sbi, true);
+
+	if (f2fs_discard_en(sbi) && !sbi->discard_blks) {
+		struct cp_control cpc = {
+			.reason = CP_UMOUNT | CP_TRIMMED,
+		};
+		write_checkpoint(sbi, &cpc);
+	}
+
 	/* write_checkpoint can update stat informaion */
 	f2fs_destroy_stats(sbi);
 
@@ -548,12 +817,14 @@ static void f2fs_put_super(struct super_block *sb)
 	 * normally superblock is clean, so we need to release this.
 	 * In addition, EIO will skip do checkpoint, we need this as well.
 	 */
-	release_dirty_inode(sbi);
-	release_discard_addrs(sbi);
+	release_ino_entry(sbi, true);
 
 	f2fs_leave_shrinker(sbi);
 	mutex_unlock(&sbi->umount_mutex);
 
+	/* our cp_error case, we can wait for any writeback page */
+	f2fs_flush_merged_writes(sbi);
+
 	iput(sbi->node_inode);
 	iput(sbi->meta_inode);
 
@@ -562,45 +833,64 @@ static void f2fs_put_super(struct super_block *sb)
 	destroy_segment_manager(sbi);
 
 	kfree(sbi->ckpt);
-	kobject_put(&sbi->s_kobj);
-	wait_for_completion(&sbi->s_kobj_unregister);
+
+	f2fs_unregister_sysfs(sbi);
 
 	sb->s_fs_info = NULL;
-	brelse(sbi->raw_super_buf);
+	if (sbi->s_chksum_driver)
+		crypto_free_shash(sbi->s_chksum_driver);
+	kfree(sbi->raw_super);
+
+	destroy_device_list(sbi);
+	if (sbi->write_io_dummy)
+		mempool_destroy(sbi->write_io_dummy);
+#ifdef CONFIG_QUOTA
+	for (i = 0; i < MAXQUOTAS; i++)
+		kfree(sbi->s_qf_names[i]);
+#endif
+	destroy_percpu_info(sbi);
+	for (i = 0; i < NR_PAGE_TYPE; i++)
+		kfree(sbi->write_io[i]);
 	kfree(sbi);
 }
 
 int f2fs_sync_fs(struct super_block *sb, int sync)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	int err = 0;
 
 	trace_f2fs_sync_fs(sb, sync);
 
+	if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING)))
+		return -EAGAIN;
+
 	if (sync) {
 		struct cp_control cpc;
 
 		cpc.reason = __get_cp_reason(sbi);
 
 		mutex_lock(&sbi->gc_mutex);
-		write_checkpoint(sbi, &cpc);
+		err = write_checkpoint(sbi, &cpc);
 		mutex_unlock(&sbi->gc_mutex);
-	} else {
-		f2fs_balance_fs(sbi);
 	}
 	f2fs_trace_ios(NULL, 1);
 
-	return 0;
+	return err;
 }
 
 static int f2fs_freeze(struct super_block *sb)
 {
-	int err;
-
 	if (f2fs_readonly(sb))
 		return 0;
 
-	err = f2fs_sync_fs(sb, 1);
-	return err;
+	/* IO error happened before */
+	if (unlikely(f2fs_cp_error(F2FS_SB(sb))))
+		return -EIO;
+
+	/* must be clean, since sync_filesystem() was already called */
+	if (is_sbi_flag_set(F2FS_SB(sb), SBI_IS_DIRTY))
+		return -EINVAL;
+	return 0;
 }
 
 static int f2fs_unfreeze(struct super_block *sb)
@@ -608,12 +898,55 @@ static int f2fs_unfreeze(struct super_block *sb)
 	return 0;
 }
 
+#ifdef CONFIG_QUOTA
+static int f2fs_statfs_project(struct super_block *sb,
+				kprojid_t projid, struct kstatfs *buf)
+{
+	struct kqid qid;
+	struct dquot *dquot;
+	u64 limit;
+	u64 curblock;
+
+	qid = make_kqid_projid(projid);
+	dquot = dqget(sb, qid);
+	if (IS_ERR(dquot))
+		return PTR_ERR(dquot);
+	spin_lock(&dq_data_lock);
+
+	limit = (dquot->dq_dqb.dqb_bsoftlimit ?
+		 dquot->dq_dqb.dqb_bsoftlimit :
+		 dquot->dq_dqb.dqb_bhardlimit) >> sb->s_blocksize_bits;
+	if (limit && buf->f_blocks > limit) {
+		curblock = dquot->dq_dqb.dqb_curspace >> sb->s_blocksize_bits;
+		buf->f_blocks = limit;
+		buf->f_bfree = buf->f_bavail =
+			(buf->f_blocks > curblock) ?
+			 (buf->f_blocks - curblock) : 0;
+	}
+
+	limit = dquot->dq_dqb.dqb_isoftlimit ?
+		dquot->dq_dqb.dqb_isoftlimit :
+		dquot->dq_dqb.dqb_ihardlimit;
+	if (limit && buf->f_files > limit) {
+		buf->f_files = limit;
+		buf->f_ffree =
+			(buf->f_files > dquot->dq_dqb.dqb_curinodes) ?
+			 (buf->f_files - dquot->dq_dqb.dqb_curinodes) : 0;
+	}
+
+	spin_unlock(&dq_data_lock);
+	dqput(dquot);
+	return 0;
+}
+#endif
+
 static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf)
 {
 	struct super_block *sb = dentry->d_sb;
 	struct f2fs_sb_info *sbi = F2FS_SB(sb);
 	u64 id = huge_encode_dev(sb->s_bdev->bd_dev);
 	block_t total_count, user_block_count, start_count, ovp_count;
+	u64 avail_node_count;
 
 	total_count = le64_to_cpu(sbi->raw_super->block_count);
 	user_block_count = sbi->user_block_count;
@@ -623,19 +956,68 @@ static int f2fs_statfs(struct dentry *dentry, struct kstatfs *buf)
 	buf->f_bsize = sbi->blocksize;
 
 	buf->f_blocks = total_count - start_count;
-	buf->f_bfree = buf->f_blocks - valid_user_blocks(sbi) - ovp_count;
-	buf->f_bavail = user_block_count - valid_user_blocks(sbi);
+	buf->f_bfree = user_block_count - valid_user_blocks(sbi) + ovp_count;
+	buf->f_bavail = user_block_count - valid_user_blocks(sbi) -
+						sbi->reserved_blocks;
 
-	buf->f_files = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
-	buf->f_ffree = buf->f_files - valid_inode_count(sbi);
+	avail_node_count = sbi->total_node_count - F2FS_RESERVED_NODE_NUM;
+
+	if (avail_node_count > user_block_count) {
+		buf->f_files = user_block_count;
+		buf->f_ffree = buf->f_bavail;
+	} else {
+		buf->f_files = avail_node_count;
+		buf->f_ffree = min(avail_node_count - valid_node_count(sbi),
+					buf->f_bavail);
+	}
 
 	buf->f_namelen = F2FS_NAME_LEN;
 	buf->f_fsid.val[0] = (u32)id;
 	buf->f_fsid.val[1] = (u32)(id >> 32);
 
+#ifdef CONFIG_QUOTA
+	if (is_inode_flag_set(dentry->d_inode, FI_PROJ_INHERIT) &&
+			sb_has_quota_limits_enabled(sb, PRJQUOTA)) {
+		f2fs_statfs_project(sb, F2FS_I(dentry->d_inode)->i_projid, buf);
+	}
+#endif
 	return 0;
 }
 
+static inline void f2fs_show_quota_options(struct seq_file *seq,
+					   struct super_block *sb)
+{
+#ifdef CONFIG_QUOTA
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+
+	if (sbi->s_jquota_fmt) {
+		char *fmtname = "";
+
+		switch (sbi->s_jquota_fmt) {
+		case QFMT_VFS_OLD:
+			fmtname = "vfsold";
+			break;
+		case QFMT_VFS_V0:
+			fmtname = "vfsv0";
+			break;
+		case QFMT_VFS_V1:
+			fmtname = "vfsv1";
+			break;
+		}
+		seq_printf(seq, ",jqfmt=%s", fmtname);
+	}
+
+	if (sbi->s_qf_names[USRQUOTA])
+		seq_show_option(seq, "usrjquota", sbi->s_qf_names[USRQUOTA]);
+
+	if (sbi->s_qf_names[GRPQUOTA])
+		seq_show_option(seq, "grpjquota", sbi->s_qf_names[GRPQUOTA]);
+
+	if (sbi->s_qf_names[PRJQUOTA])
+		seq_show_option(seq, "prjjquota", sbi->s_qf_names[PRJQUOTA]);
+#endif
+}
+
 static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(root->d_sb);
@@ -653,7 +1035,9 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 	if (test_opt(sbi, DISCARD))
 		seq_puts(seq, ",discard");
 	if (test_opt(sbi, NOHEAP))
-		seq_puts(seq, ",no_heap_alloc");
+		seq_puts(seq, ",no_heap");
+	else
+		seq_puts(seq, ",heap");
 #ifdef CONFIG_F2FS_FS_XATTR
 	if (test_opt(sbi, XATTR_USER))
 		seq_puts(seq, ",user_xattr");
@@ -661,6 +1045,8 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 		seq_puts(seq, ",nouser_xattr");
 	if (test_opt(sbi, INLINE_XATTR))
 		seq_puts(seq, ",inline_xattr");
+	else
+		seq_puts(seq, ",noinline_xattr");
 #endif
 #ifdef CONFIG_F2FS_FS_POSIX_ACL
 	if (test_opt(sbi, POSIX_ACL))
@@ -676,6 +1062,8 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 		seq_puts(seq, ",noinline_data");
 	if (test_opt(sbi, INLINE_DENTRY))
 		seq_puts(seq, ",inline_dentry");
+	else
+		seq_puts(seq, ",noinline_dentry");
 	if (!f2fs_readonly(sbi->sb) && test_opt(sbi, FLUSH_MERGE))
 		seq_puts(seq, ",flush_merge");
 	if (test_opt(sbi, NOBARRIER))
@@ -686,59 +1074,56 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root)
 		seq_puts(seq, ",extent_cache");
 	else
 		seq_puts(seq, ",noextent_cache");
+	if (test_opt(sbi, DATA_FLUSH))
+		seq_puts(seq, ",data_flush");
+
+	seq_puts(seq, ",mode=");
+	if (test_opt(sbi, ADAPTIVE))
+		seq_puts(seq, "adaptive");
+	else if (test_opt(sbi, LFS))
+		seq_puts(seq, "lfs");
 	seq_printf(seq, ",active_logs=%u", sbi->active_logs);
+	if (F2FS_IO_SIZE_BITS(sbi))
+		seq_printf(seq, ",io_size=%uKB", F2FS_IO_SIZE_KB(sbi));
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (test_opt(sbi, FAULT_INJECTION))
+		seq_printf(seq, ",fault_injection=%u",
+				sbi->fault_info.inject_rate);
+#endif
+#ifdef CONFIG_QUOTA
+	if (test_opt(sbi, QUOTA))
+		seq_puts(seq, ",quota");
+	if (test_opt(sbi, USRQUOTA))
+		seq_puts(seq, ",usrquota");
+	if (test_opt(sbi, GRPQUOTA))
+		seq_puts(seq, ",grpquota");
+	if (test_opt(sbi, PRJQUOTA))
+		seq_puts(seq, ",prjquota");
+#endif
+	f2fs_show_quota_options(seq, sbi->sb);
 
 	return 0;
 }
 
-static int segment_info_seq_show(struct seq_file *seq, void *offset)
-{
-	struct super_block *sb = seq->private;
-	struct f2fs_sb_info *sbi = F2FS_SB(sb);
-	unsigned int total_segs =
-			le32_to_cpu(sbi->raw_super->segment_count_main);
-	int i;
-
-	seq_puts(seq, "format: segment_type|valid_blocks\n"
-		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");
-
-	for (i = 0; i < total_segs; i++) {
-		struct seg_entry *se = get_seg_entry(sbi, i);
-
-		if ((i % 10) == 0)
-			seq_printf(seq, "%-10d", i);
-		seq_printf(seq, "%d|%-3u", se->type,
-					get_valid_blocks(sbi, i, 1));
-		if ((i % 10) == 9 || i == (total_segs - 1))
-			seq_putc(seq, '\n');
-		else
-			seq_putc(seq, ' ');
-	}
-
-	return 0;
-}
-
-static int segment_info_open_fs(struct inode *inode, struct file *file)
-{
-	return single_open(file, segment_info_seq_show, PDE_DATA(inode));
-}
-
-static const struct file_operations f2fs_seq_segment_info_fops = {
-	.owner = THIS_MODULE,
-	.open = segment_info_open_fs,
-	.read = seq_read,
-	.llseek = seq_lseek,
-	.release = single_release,
-};
-
 static void default_options(struct f2fs_sb_info *sbi)
 {
 	/* init some FS parameters */
 	sbi->active_logs = NR_CURSEG_TYPE;
 
 	set_opt(sbi, BG_GC);
+	set_opt(sbi, INLINE_XATTR);
 	set_opt(sbi, INLINE_DATA);
+	set_opt(sbi, INLINE_DENTRY);
 	set_opt(sbi, EXTENT_CACHE);
+	set_opt(sbi, NOHEAP);
+	sbi->sb->s_flags |= MS_LAZYTIME;
+	set_opt(sbi, FLUSH_MERGE);
+	if (f2fs_sb_mounted_blkzoned(sbi->sb)) {
+		set_opt_mode(sbi, F2FS_MOUNT_LFS);
+		set_opt(sbi, DISCARD);
+	} else {
+		set_opt_mode(sbi, F2FS_MOUNT_ADAPTIVE);
+	}
 
 #ifdef CONFIG_F2FS_FS_XATTR
 	set_opt(sbi, XATTR_USER);
@@ -746,27 +1131,64 @@ static void default_options(struct f2fs_sb_info *sbi)
 #ifdef CONFIG_F2FS_FS_POSIX_ACL
 	set_opt(sbi, POSIX_ACL);
 #endif
+
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	f2fs_build_fault_attr(sbi, 0);
+#endif
 }
 
 static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 {
 	struct f2fs_sb_info *sbi = F2FS_SB(sb);
 	struct f2fs_mount_info org_mount_opt;
+	unsigned long old_sb_flags;
 	int err, active_logs;
 	bool need_restart_gc = false;
 	bool need_stop_gc = false;
 	bool no_extent_cache = !test_opt(sbi, EXTENT_CACHE);
-
-	sync_filesystem(sb);
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	struct f2fs_fault_info ffi = sbi->fault_info;
+#endif
+#ifdef CONFIG_QUOTA
+	int s_jquota_fmt;
+	char *s_qf_names[MAXQUOTAS];
+	int i, j;
+#endif
 
 	/*
 	 * Save the old mount options in case we
 	 * need to restore them.
 	 */
 	org_mount_opt = sbi->mount_opt;
+	old_sb_flags = sb->s_flags;
 	active_logs = sbi->active_logs;
 
-	sbi->mount_opt.opt = 0;
+#ifdef CONFIG_QUOTA
+	s_jquota_fmt = sbi->s_jquota_fmt;
+	for (i = 0; i < MAXQUOTAS; i++) {
+		if (sbi->s_qf_names[i]) {
+			s_qf_names[i] = kstrdup(sbi->s_qf_names[i],
+							 GFP_KERNEL);
+			if (!s_qf_names[i]) {
+				for (j = 0; j < i; j++)
+					kfree(s_qf_names[j]);
+				return -ENOMEM;
+			}
+		} else {
+			s_qf_names[i] = NULL;
+		}
+	}
+#endif
+
+	/* recover superblocks we couldn't write due to previous RO mount */
+	if (!(*flags & MS_RDONLY) && is_sbi_flag_set(sbi, SBI_NEED_SB_WRITE)) {
+		err = f2fs_commit_super(sbi, false);
+		f2fs_msg(sb, KERN_INFO,
+			"Try to recover all the superblocks, ret: %d", err);
+		if (!err)
+			clear_sbi_flag(sbi, SBI_NEED_SB_WRITE);
+	}
+
 	default_options(sbi);
 
 	/* parse mount options */
@@ -781,6 +1203,16 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 	if (f2fs_readonly(sb) && (*flags & MS_RDONLY))
 		goto skip;
 
+	if (!f2fs_readonly(sb) && (*flags & MS_RDONLY)) {
+		err = dquot_suspend(sb, -1);
+		if (err < 0)
+			goto restore_opts;
+	} else {
+		/* dquot_resume needs RW */
+		sb->s_flags &= ~MS_RDONLY;
+		dquot_resume(sb, -1);
+	}
+
 	/* disallow enable/disable extent_cache dynamically */
 	if (no_extent_cache == !!test_opt(sbi, EXTENT_CACHE)) {
 		err = -EINVAL;
@@ -797,7 +1229,6 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 	if ((*flags & MS_RDONLY) || !test_opt(sbi, BG_GC)) {
 		if (sbi->gc_thread) {
 			stop_gc_thread(sbi);
-			f2fs_sync_fs(sb, 1);
 			need_restart_gc = true;
 		}
 	} else if (!sbi->gc_thread) {
@@ -807,21 +1238,38 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data)
 		need_stop_gc = true;
 	}
 
+	if (*flags & MS_RDONLY) {
+		writeback_inodes_sb(sb, WB_REASON_SYNC);
+		sync_inodes_sb(sb);
+
+		set_sbi_flag(sbi, SBI_IS_DIRTY);
+		set_sbi_flag(sbi, SBI_IS_CLOSE);
+		f2fs_sync_fs(sb, 1);
+		clear_sbi_flag(sbi, SBI_IS_CLOSE);
+	}
+
 	/*
 	 * We stop issue flush thread if FS is mounted as RO
 	 * or if flush_merge is not passed in mount option.
 	 */
 	if ((*flags & MS_RDONLY) || !test_opt(sbi, FLUSH_MERGE)) {
-		destroy_flush_cmd_control(sbi);
-	} else if (!SM_I(sbi)->cmd_control_info) {
+		clear_opt(sbi, FLUSH_MERGE);
+		destroy_flush_cmd_control(sbi, false);
+	} else {
 		err = create_flush_cmd_control(sbi);
 		if (err)
 			goto restore_gc;
 	}
 skip:
+#ifdef CONFIG_QUOTA
+	/* Release old quota file names */
+	for (i = 0; i < MAXQUOTAS; i++)
+		kfree(s_qf_names[i]);
+#endif
 	/* Update the POSIXACL Flag */
-	 sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
+	sb->s_flags = (sb->s_flags & ~MS_POSIXACL) |
 		(test_opt(sbi, POSIX_ACL) ? MS_POSIXACL : 0);
+
 	return 0;
 restore_gc:
 	if (need_restart_gc) {
@@ -832,18 +1280,289 @@ restore_gc:
 		stop_gc_thread(sbi);
 	}
 restore_opts:
+#ifdef CONFIG_QUOTA
+	sbi->s_jquota_fmt = s_jquota_fmt;
+	for (i = 0; i < MAXQUOTAS; i++) {
+		kfree(sbi->s_qf_names[i]);
+		sbi->s_qf_names[i] = s_qf_names[i];
+	}
+#endif
 	sbi->mount_opt = org_mount_opt;
 	sbi->active_logs = active_logs;
+	sb->s_flags = old_sb_flags;
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	sbi->fault_info = ffi;
+#endif
+	return err;
+}
+
+#ifdef CONFIG_QUOTA
+/* Read data from quotafile */
+static ssize_t f2fs_quota_read(struct super_block *sb, int type, char *data,
+			       size_t len, loff_t off)
+{
+	struct inode *inode = sb_dqopt(sb)->files[type];
+	struct address_space *mapping = inode->i_mapping;
+	block_t blkidx = F2FS_BYTES_TO_BLK(off);
+	int offset = off & (sb->s_blocksize - 1);
+	int tocopy;
+	size_t toread;
+	loff_t i_size = i_size_read(inode);
+	struct page *page;
+	char *kaddr;
+
+	if (off > i_size)
+		return 0;
+
+	if (off + len > i_size)
+		len = i_size - off;
+	toread = len;
+	while (toread > 0) {
+		tocopy = min_t(unsigned long, sb->s_blocksize - offset, toread);
+repeat:
+		page = read_mapping_page(mapping, blkidx, NULL);
+		if (IS_ERR(page))
+			return PTR_ERR(page);
+
+		lock_page(page);
+
+		if (unlikely(page->mapping != mapping)) {
+			f2fs_put_page(page, 1);
+			goto repeat;
+		}
+		if (unlikely(!PageUptodate(page))) {
+			f2fs_put_page(page, 1);
+			return -EIO;
+		}
+
+		kaddr = kmap_atomic(page);
+		memcpy(data, kaddr + offset, tocopy);
+		kunmap_atomic(kaddr);
+		f2fs_put_page(page, 1);
+
+		offset = 0;
+		toread -= tocopy;
+		data += tocopy;
+		blkidx++;
+	}
+	return len;
+}
+
+/* Write to quotafile */
+static ssize_t f2fs_quota_write(struct super_block *sb, int type,
+				const char *data, size_t len, loff_t off)
+{
+	struct inode *inode = sb_dqopt(sb)->files[type];
+	struct address_space *mapping = inode->i_mapping;
+	const struct address_space_operations *a_ops = mapping->a_ops;
+	int offset = off & (sb->s_blocksize - 1);
+	size_t towrite = len;
+	struct page *page;
+	char *kaddr;
+	int err = 0;
+	int tocopy;
+
+	while (towrite > 0) {
+		tocopy = min_t(unsigned long, sb->s_blocksize - offset,
+								towrite);
+
+		err = a_ops->write_begin(NULL, mapping, off, tocopy, 0,
+							&page, NULL);
+		if (unlikely(err))
+			break;
+
+		kaddr = kmap_atomic(page);
+		memcpy(kaddr + offset, data, tocopy);
+		kunmap_atomic(kaddr);
+		flush_dcache_page(page);
+
+		a_ops->write_end(NULL, mapping, off, tocopy, tocopy,
+						page, NULL);
+		offset = 0;
+		towrite -= tocopy;
+		off += tocopy;
+		data += tocopy;
+		cond_resched();
+	}
+
+	if (len == towrite)
+		return 0;
+	inode->i_version++;
+	inode->i_mtime = inode->i_ctime = current_time(inode);
+	f2fs_mark_inode_dirty_sync(inode, false);
+	return len - towrite;
+}
+
+static struct dquot **f2fs_get_dquots(struct inode *inode)
+{
+	return F2FS_I(inode)->i_dquot;
+}
+
+static qsize_t *f2fs_get_reserved_space(struct inode *inode)
+{
+	return &F2FS_I(inode)->i_reserved_quota;
+}
+
+static int f2fs_quota_on_mount(struct f2fs_sb_info *sbi, int type)
+{
+	return dquot_quota_on_mount(sbi->sb, sbi->s_qf_names[type],
+						sbi->s_jquota_fmt, type);
+}
+
+void f2fs_enable_quota_files(struct f2fs_sb_info *sbi)
+{
+	int i, ret;
+
+	for (i = 0; i < MAXQUOTAS; i++) {
+		if (sbi->s_qf_names[i]) {
+			ret = f2fs_quota_on_mount(sbi, i);
+			if (ret < 0)
+				f2fs_msg(sbi->sb, KERN_ERR,
+					"Cannot turn on journaled "
+					"quota: error %d", ret);
+		}
+	}
+}
+
+static int f2fs_quota_sync(struct super_block *sb, int type)
+{
+	struct quota_info *dqopt = sb_dqopt(sb);
+	int cnt;
+	int ret;
+
+	ret = dquot_writeback_dquots(sb, type);
+	if (ret)
+		return ret;
+
+	/*
+	 * Now when everything is written we can discard the pagecache so
+	 * that userspace sees the changes.
+	 */
+	for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
+		if (type != -1 && cnt != type)
+			continue;
+		if (!sb_has_quota_active(sb, cnt))
+			continue;
+
+		ret = filemap_write_and_wait(dqopt->files[cnt]->i_mapping);
+		if (ret)
+			return ret;
+
+		inode_lock(dqopt->files[cnt]);
+		truncate_inode_pages(&dqopt->files[cnt]->i_data, 0);
+		inode_unlock(dqopt->files[cnt]);
+	}
+	return 0;
+}
+
+static int f2fs_quota_on(struct super_block *sb, int type, int format_id,
+							struct path *path)
+{
+	struct inode *inode;
+	int err;
+
+	err = f2fs_quota_sync(sb, type);
+	if (err)
+		return err;
+
+	err = dquot_quota_on(sb, type, format_id, path);
+	if (err)
+		return err;
+
+	inode = d_inode(path->dentry);
+
+	inode_lock(inode);
+	F2FS_I(inode)->i_flags |= FS_NOATIME_FL | FS_IMMUTABLE_FL;
+	inode_set_flags(inode, S_NOATIME | S_IMMUTABLE,
+					S_NOATIME | S_IMMUTABLE);
+	inode_unlock(inode);
+	f2fs_mark_inode_dirty_sync(inode, false);
+
+	return 0;
+}
+
+static int f2fs_quota_off(struct super_block *sb, int type)
+{
+	struct inode *inode = sb_dqopt(sb)->files[type];
+	int err;
+
+	if (!inode || !igrab(inode))
+		return dquot_quota_off(sb, type);
+
+	f2fs_quota_sync(sb, type);
+
+	err = dquot_quota_off(sb, type);
+	if (err)
+		goto out_put;
+
+	inode_lock(inode);
+	F2FS_I(inode)->i_flags &= ~(FS_NOATIME_FL | FS_IMMUTABLE_FL);
+	inode_set_flags(inode, 0, S_NOATIME | S_IMMUTABLE);
+	inode_unlock(inode);
+	f2fs_mark_inode_dirty_sync(inode, false);
+out_put:
+	iput(inode);
 	return err;
 }
 
-static struct super_operations f2fs_sops = {
+void f2fs_quota_off_umount(struct super_block *sb)
+{
+	int type;
+
+	for (type = 0; type < MAXQUOTAS; type++)
+		f2fs_quota_off(sb, type);
+}
+
+#if 0
+int f2fs_get_projid(struct inode *inode, kprojid_t *projid)
+{
+	*projid = F2FS_I(inode)->i_projid;
+	return 0;
+}
+#endif
+
+static const struct dquot_operations f2fs_quota_operations = {
+	.get_reserved_space = f2fs_get_reserved_space,
+	.write_dquot	= dquot_commit,
+	.acquire_dquot	= dquot_acquire,
+	.release_dquot	= dquot_release,
+	.mark_dirty	= dquot_mark_dquot_dirty,
+	.write_info	= dquot_commit_info,
+	.alloc_dquot	= dquot_alloc,
+	.destroy_dquot	= dquot_destroy,
+#if 0
+	.get_projid	= f2fs_get_projid,
+	.get_next_id	= dquot_get_next_id,
+#endif
+};
+
+static const struct quotactl_ops f2fs_quotactl_ops = {
+	.quota_on	= f2fs_quota_on,
+	.quota_off	= f2fs_quota_off,
+	.quota_sync	= f2fs_quota_sync,
+	.get_state	= dquot_get_state,
+	.set_info	= dquot_set_dqinfo,
+	.get_dqblk	= dquot_get_dqblk,
+	.set_dqblk	= dquot_set_dqblk,
+};
+#else
+void f2fs_quota_off_umount(struct super_block *sb)
+{
+}
+#endif
+
+static const struct super_operations f2fs_sops = {
 	.alloc_inode	= f2fs_alloc_inode,
 	.drop_inode	= f2fs_drop_inode,
 	.destroy_inode	= f2fs_destroy_inode,
 	.write_inode	= f2fs_write_inode,
 	.dirty_inode	= f2fs_dirty_inode,
 	.show_options	= f2fs_show_options,
+#ifdef CONFIG_QUOTA
+	.quota_read	= f2fs_quota_read,
+	.quota_write	= f2fs_quota_write,
+	.get_dquots	= f2fs_get_dquots,
+#endif
 	.evict_inode	= f2fs_evict_inode,
 	.put_super	= f2fs_put_super,
 	.sync_fs	= f2fs_sync_fs,
@@ -853,6 +1572,42 @@ static struct super_operations f2fs_sops = {
 	.remount_fs	= f2fs_remount,
 };
 
+#ifdef CONFIG_F2FS_FS_ENCRYPTION
+static int f2fs_get_context(struct inode *inode, void *ctx, size_t len)
+{
+	return f2fs_getxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
+				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
+				ctx, len, NULL);
+}
+
+static int f2fs_set_context(struct inode *inode, const void *ctx, size_t len,
+							void *fs_data)
+{
+	return f2fs_setxattr(inode, F2FS_XATTR_INDEX_ENCRYPTION,
+				F2FS_XATTR_NAME_ENCRYPTION_CONTEXT,
+				ctx, len, fs_data, XATTR_CREATE);
+}
+
+static unsigned f2fs_max_namelen(struct inode *inode)
+{
+	return S_ISLNK(inode->i_mode) ?
+			inode->i_sb->s_blocksize : F2FS_NAME_LEN;
+}
+
+static const struct fscrypt_operations f2fs_cryptops = {
+	.key_prefix	= "f2fs:",
+	.get_context	= f2fs_get_context,
+	.set_context	= f2fs_set_context,
+	.is_encrypted	= f2fs_encrypted_inode,
+	.empty_dir	= f2fs_empty_dir,
+	.max_namelen	= f2fs_max_namelen,
+};
+#else
+static const struct fscrypt_operations f2fs_cryptops = {
+	.is_encrypted	= f2fs_encrypted_inode,
+};
+#endif
+
 static struct inode *f2fs_nfs_get_inode(struct super_block *sb,
 		u64 ino, u32 generation)
 {
@@ -898,11 +1653,18 @@ static const struct export_operations f2fs_export_ops = {
 	.get_parent = f2fs_get_parent,
 };
 
-static loff_t max_file_size(unsigned bits)
+static loff_t max_file_blocks(void)
 {
-	loff_t result = (DEF_ADDRS_PER_INODE - F2FS_INLINE_XATTR_ADDRS);
+	loff_t result = 0;
 	loff_t leaf_count = ADDRS_PER_BLOCK;
 
+	/*
+	 * note: previously, result is equal to (DEF_ADDRS_PER_INODE -
+	 * F2FS_INLINE_XATTR_ADDRS), but now f2fs try to reserve more
+	 * space in inode.i_addr, it will be more safe to reassign
+	 * result as zero.
+	 */
+
 	/* two direct node blocks */
 	result += (leaf_count * 2);
 
@@ -914,13 +1676,29 @@ static loff_t max_file_size(unsigned bits)
 	leaf_count *= NIDS_PER_BLOCK;
 	result += leaf_count;
 
-	result <<= bits;
 	return result;
 }
 
-static inline bool sanity_check_area_boundary(struct super_block *sb,
-					struct f2fs_super_block *raw_super)
+static int __f2fs_commit_super(struct buffer_head *bh,
+			struct f2fs_super_block *super)
 {
+	lock_buffer(bh);
+	if (super)
+		memcpy(bh->b_data + F2FS_SUPER_OFFSET, super, sizeof(*super));
+	set_buffer_uptodate(bh);
+	set_buffer_dirty(bh);
+	unlock_buffer(bh);
+
+	/* it's rare case, we can do fua all the time */
+	return __sync_dirty_buffer(bh, WRITE_FLUSH_FUA);
+}
+
+static inline bool sanity_check_area_boundary(struct f2fs_sb_info *sbi,
+					struct buffer_head *bh)
+{
+	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
+					(bh->b_data + F2FS_SUPER_OFFSET);
+	struct super_block *sb = sbi->sb;
 	u32 segment0_blkaddr = le32_to_cpu(raw_super->segment0_blkaddr);
 	u32 cp_blkaddr = le32_to_cpu(raw_super->cp_blkaddr);
 	u32 sit_blkaddr = le32_to_cpu(raw_super->sit_blkaddr);
@@ -934,6 +1712,10 @@ static inline bool sanity_check_area_boundary(struct super_block *sb,
 	u32 segment_count_main = le32_to_cpu(raw_super->segment_count_main);
 	u32 segment_count = le32_to_cpu(raw_super->segment_count);
 	u32 log_blocks_per_seg = le32_to_cpu(raw_super->log_blocks_per_seg);
+	u64 main_end_blkaddr = main_blkaddr +
+				(segment_count_main << log_blocks_per_seg);
+	u64 seg_end_blkaddr = segment0_blkaddr +
+				(segment_count << log_blocks_per_seg);
 
 	if (segment0_blkaddr != cp_blkaddr) {
 		f2fs_msg(sb, KERN_INFO,
@@ -978,22 +1760,47 @@ static inline bool sanity_check_area_boundary(struct super_block *sb,
 		return true;
 	}
 
-	if (main_blkaddr + (segment_count_main << log_blocks_per_seg) !=
-		segment0_blkaddr + (segment_count << log_blocks_per_seg)) {
+	if (main_end_blkaddr > seg_end_blkaddr) {
 		f2fs_msg(sb, KERN_INFO,
-			"Wrong MAIN_AREA boundary, start(%u) end(%u) blocks(%u)",
+			"Wrong MAIN_AREA boundary, start(%u) end(%u) block(%u)",
 			main_blkaddr,
-			segment0_blkaddr + (segment_count << log_blocks_per_seg),
+			segment0_blkaddr +
+				(segment_count << log_blocks_per_seg),
 			segment_count_main << log_blocks_per_seg);
 		return true;
-	}
+	} else if (main_end_blkaddr < seg_end_blkaddr) {
+		int err = 0;
+		char *res;
 
+		/* fix in-memory information all the time */
+		raw_super->segment_count = cpu_to_le32((main_end_blkaddr -
+				segment0_blkaddr) >> log_blocks_per_seg);
+
+		if (f2fs_readonly(sb) || bdev_read_only(sb->s_bdev)) {
+			set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
+			res = "internally";
+		} else {
+			err = __f2fs_commit_super(bh, NULL);
+			res = err ? "failed" : "done";
+		}
+		f2fs_msg(sb, KERN_INFO,
+			"Fix alignment : %s, start(%u) end(%u) block(%u)",
+			res, main_blkaddr,
+			segment0_blkaddr +
+				(segment_count << log_blocks_per_seg),
+			segment_count_main << log_blocks_per_seg);
+		if (err)
+			return true;
+	}
 	return false;
 }
 
-static int sanity_check_raw_super(struct super_block *sb,
-			struct f2fs_super_block *raw_super)
+static int sanity_check_raw_super(struct f2fs_sb_info *sbi,
+				struct buffer_head *bh)
 {
+	struct f2fs_super_block *raw_super = (struct f2fs_super_block *)
+					(bh->b_data + F2FS_SUPER_OFFSET);
+	struct super_block *sb = sbi->sb;
 	unsigned int blocksize;
 
 	if (F2FS_SUPER_MAGIC != le32_to_cpu(raw_super->magic)) {
@@ -1004,10 +1811,10 @@ static int sanity_check_raw_super(struct super_block *sb,
 	}
 
 	/* Currently, support only 4KB page cache size */
-	if (F2FS_BLKSIZE != PAGE_CACHE_SIZE) {
+	if (F2FS_BLKSIZE != PAGE_SIZE) {
 		f2fs_msg(sb, KERN_INFO,
 			"Invalid page_cache_size (%lu), supports only 4KB\n",
-			PAGE_CACHE_SIZE);
+			PAGE_SIZE);
 		return 1;
 	}
 
@@ -1067,17 +1874,18 @@ static int sanity_check_raw_super(struct super_block *sb,
 	}
 
 	/* check CP/SIT/NAT/SSA/MAIN_AREA area boundary */
-	if (sanity_check_area_boundary(sb, raw_super))
+	if (sanity_check_area_boundary(sbi, bh))
 		return 1;
 
 	return 0;
 }
 
-static int sanity_check_ckpt(struct f2fs_sb_info *sbi)
+int sanity_check_ckpt(struct f2fs_sb_info *sbi)
 {
 	unsigned int total, fsmeta;
 	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
 	struct f2fs_checkpoint *ckpt = F2FS_CKPT(sbi);
+	unsigned int ovp_segments, reserved_segments;
 	unsigned int main_segs, blocks_per_seg;
 	int i;
 
@@ -1091,6 +1899,16 @@ static int sanity_check_ckpt(struct f2fs_sb_info *sbi)
 	if (unlikely(fsmeta >= total))
 		return 1;
 
+	ovp_segments = le32_to_cpu(ckpt->overprov_segment_count);
+	reserved_segments = le32_to_cpu(ckpt->rsvd_segment_count);
+
+	if (unlikely(fsmeta < F2FS_MIN_SEGMENTS ||
+			ovp_segments == 0 || reserved_segments == 0)) {
+		f2fs_msg(sbi->sb, KERN_ERR,
+			"Wrong layout: check mkfs.f2fs version");
+		return 1;
+	}
+
 	main_segs = le32_to_cpu(raw_super->segment_count_main);
 	blocks_per_seg = sbi->blocks_per_seg;
 
@@ -1115,7 +1933,7 @@ static int sanity_check_ckpt(struct f2fs_sb_info *sbi)
 static void init_sb_info(struct f2fs_sb_info *sbi)
 {
 	struct f2fs_super_block *raw_super = sbi->raw_super;
-	int i;
+	int i, j;
 
 	sbi->log_sectors_per_block =
 		le32_to_cpu(raw_super->log_sectors_per_block);
@@ -1135,129 +1953,306 @@ static void init_sb_info(struct f2fs_sb_info *sbi)
 	sbi->cur_victim_sec = NULL_SECNO;
 	sbi->max_victim_search = DEF_MAX_VICTIM_SEARCH;
 
+	sbi->dir_level = DEF_DIR_LEVEL;
+	sbi->interval_time[CP_TIME] = DEF_CP_INTERVAL;
+	sbi->interval_time[REQ_TIME] = DEF_IDLE_INTERVAL;
+	clear_sbi_flag(sbi, SBI_NEED_FSCK);
+
 	for (i = 0; i < NR_COUNT_TYPE; i++)
 		atomic_set(&sbi->nr_pages[i], 0);
 
-	sbi->dir_level = DEF_DIR_LEVEL;
-	sbi->cp_interval = DEF_CP_INTERVAL;
-	clear_sbi_flag(sbi, SBI_NEED_FSCK);
+	atomic_set(&sbi->wb_sync_req, 0);
 
 	INIT_LIST_HEAD(&sbi->s_list);
 	mutex_init(&sbi->umount_mutex);
+	for (i = 0; i < NR_PAGE_TYPE - 1; i++)
+		for (j = HOT; j < NR_TEMP_TYPE; j++)
+			mutex_init(&sbi->wio_mutex[i][j]);
+	spin_lock_init(&sbi->cp_lock);
+}
+
+static int init_percpu_info(struct f2fs_sb_info *sbi)
+{
+	int err;
+
+	err = percpu_counter_init(&sbi->alloc_valid_block_count, 0, GFP_KERNEL);
+	if (err)
+		return err;
+
+	return percpu_counter_init(&sbi->total_valid_inode_count, 0,
+								GFP_KERNEL);
+}
+
+#ifdef CONFIG_BLK_DEV_ZONED
+static int init_blkz_info(struct f2fs_sb_info *sbi, int devi)
+{
+	struct block_device *bdev = FDEV(devi).bdev;
+	sector_t nr_sectors = bdev->bd_part->nr_sects;
+	sector_t sector = 0;
+	struct blk_zone *zones;
+	unsigned int i, nr_zones;
+	unsigned int n = 0;
+	int err = -EIO;
+
+	if (!f2fs_sb_mounted_blkzoned(sbi->sb))
+		return 0;
+
+	if (sbi->blocks_per_blkz && sbi->blocks_per_blkz !=
+				SECTOR_TO_BLOCK(bdev_zone_sectors(bdev)))
+		return -EINVAL;
+	sbi->blocks_per_blkz = SECTOR_TO_BLOCK(bdev_zone_sectors(bdev));
+	if (sbi->log_blocks_per_blkz && sbi->log_blocks_per_blkz !=
+				__ilog2_u32(sbi->blocks_per_blkz))
+		return -EINVAL;
+	sbi->log_blocks_per_blkz = __ilog2_u32(sbi->blocks_per_blkz);
+	FDEV(devi).nr_blkz = SECTOR_TO_BLOCK(nr_sectors) >>
+					sbi->log_blocks_per_blkz;
+	if (nr_sectors & (bdev_zone_sectors(bdev) - 1))
+		FDEV(devi).nr_blkz++;
+
+	FDEV(devi).blkz_type = kmalloc(FDEV(devi).nr_blkz, GFP_KERNEL);
+	if (!FDEV(devi).blkz_type)
+		return -ENOMEM;
+
+#define F2FS_REPORT_NR_ZONES   4096
+
+	zones = kcalloc(F2FS_REPORT_NR_ZONES, sizeof(struct blk_zone),
+			GFP_KERNEL);
+	if (!zones)
+		return -ENOMEM;
+
+	/* Get block zones type */
+	while (zones && sector < nr_sectors) {
+
+		nr_zones = F2FS_REPORT_NR_ZONES;
+		err = blkdev_report_zones(bdev, sector,
+					  zones, &nr_zones,
+					  GFP_KERNEL);
+		if (err)
+			break;
+		if (!nr_zones) {
+			err = -EIO;
+			break;
+		}
+
+		for (i = 0; i < nr_zones; i++) {
+			FDEV(devi).blkz_type[n] = zones[i].type;
+			sector += zones[i].len;
+			n++;
+		}
+	}
+
+	kfree(zones);
+
+	return err;
 }
+#endif
 
 /*
  * Read f2fs raw super block.
- * Because we have two copies of super block, so read the first one at first,
- * if the first one is invalid, move to read the second one.
+ * Because we have two copies of super block, so read both of them
+ * to get the first valid one. If any one of them is broken, we pass
+ * them recovery flag back to the caller.
  */
-static int read_raw_super_block(struct super_block *sb,
+static int read_raw_super_block(struct f2fs_sb_info *sbi,
 			struct f2fs_super_block **raw_super,
-			struct buffer_head **raw_super_buf,
-			int *recovery)
+			int *valid_super_block, int *recovery)
 {
-	int block = 0;
-	struct buffer_head *buffer;
+	struct super_block *sb = sbi->sb;
+	int block;
+	struct buffer_head *bh;
 	struct f2fs_super_block *super;
 	int err = 0;
 
-retry:
-	buffer = sb_bread(sb, block);
-	if (!buffer) {
-		*recovery = 1;
-		f2fs_msg(sb, KERN_ERR, "Unable to read %dth superblock",
+	super = kzalloc(sizeof(struct f2fs_super_block), GFP_KERNEL);
+	if (!super)
+		return -ENOMEM;
+
+	for (block = 0; block < 2; block++) {
+		bh = sb_bread(sb, block);
+		if (!bh) {
+			f2fs_msg(sb, KERN_ERR, "Unable to read %dth superblock",
 				block + 1);
-		if (block == 0) {
-			block++;
-			goto retry;
-		} else {
 			err = -EIO;
-			goto out;
+			continue;
 		}
-	}
-
-	super = (struct f2fs_super_block *)
-		((char *)(buffer)->b_data + F2FS_SUPER_OFFSET);
 
-	/* sanity checking of raw super */
-	if (sanity_check_raw_super(sb, super)) {
-		brelse(buffer);
-		*recovery = 1;
-		f2fs_msg(sb, KERN_ERR,
-			"Can't find valid F2FS filesystem in %dth superblock",
-								block + 1);
-		if (block == 0) {
-			block++;
-			goto retry;
-		} else {
+		/* sanity checking of raw super */
+		if (sanity_check_raw_super(sbi, bh)) {
+			f2fs_msg(sb, KERN_ERR,
+				"Can't find valid F2FS filesystem in %dth superblock",
+				block + 1);
 			err = -EINVAL;
-			goto out;
+			brelse(bh);
+			continue;
 		}
-	}
 
-	if (!*raw_super) {
-		*raw_super_buf = buffer;
-		*raw_super = super;
-	} else {
-		/* already have a valid superblock */
-		brelse(buffer);
+		if (!*raw_super) {
+			memcpy(super, bh->b_data + F2FS_SUPER_OFFSET,
+							sizeof(*super));
+			*valid_super_block = block;
+			*raw_super = super;
+		}
+		brelse(bh);
 	}
 
-	/* check the validity of the second superblock */
-	if (block == 0) {
-		block++;
-		goto retry;
-	}
+	/* Fail to read any one of the superblocks*/
+	if (err < 0)
+		*recovery = 1;
 
-out:
 	/* No valid superblock */
 	if (!*raw_super)
-		return err;
+		kfree(super);
+	else
+		err = 0;
 
-	return 0;
+	return err;
 }
 
 int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover)
 {
-	struct buffer_head *sbh = sbi->raw_super_buf;
-	sector_t block = sbh->b_blocknr;
+	struct buffer_head *bh;
 	int err;
 
-	/* write back-up superblock first */
-	sbh->b_blocknr = block ? 0 : 1;
-	mark_buffer_dirty(sbh);
-	err = sync_dirty_buffer(sbh);
+	if ((recover && f2fs_readonly(sbi->sb)) ||
+				bdev_read_only(sbi->sb->s_bdev)) {
+		set_sbi_flag(sbi, SBI_NEED_SB_WRITE);
+		return -EROFS;
+	}
 
-	sbh->b_blocknr = block;
+	/* write back-up superblock first */
+	bh = sb_getblk(sbi->sb, sbi->valid_super_block ? 0: 1);
+	if (!bh)
+		return -EIO;
+	err = __f2fs_commit_super(bh, F2FS_RAW_SUPER(sbi));
+	brelse(bh);
 
 	/* if we are in recovery path, skip writing valid superblock */
 	if (recover || err)
-		goto out;
+		return err;
 
 	/* write current valid superblock */
-	mark_buffer_dirty(sbh);
-	err = sync_dirty_buffer(sbh);
-out:
-	clear_buffer_write_io_error(sbh);
-	set_buffer_uptodate(sbh);
+	bh = sb_getblk(sbi->sb, sbi->valid_super_block);
+	if (!bh)
+		return -EIO;
+	err = __f2fs_commit_super(bh, F2FS_RAW_SUPER(sbi));
+	brelse(bh);
 	return err;
 }
 
+static int f2fs_scan_devices(struct f2fs_sb_info *sbi)
+{
+	struct f2fs_super_block *raw_super = F2FS_RAW_SUPER(sbi);
+	unsigned int max_devices = MAX_DEVICES;
+	int i;
+
+	/* Initialize single device information */
+	if (!RDEV(0).path[0]) {
+#ifdef CONFIG_BLK_DEV_ZONED
+		if (!bdev_is_zoned(sbi->sb->s_bdev))
+			return 0;
+		max_devices = 1;
+#else
+		return 0;
+#endif
+	}
+
+	/*
+	 * Initialize multiple devices information, or single
+	 * zoned block device information.
+	 */
+	sbi->devs = kcalloc(max_devices, sizeof(struct f2fs_dev_info),
+				GFP_KERNEL);
+	if (!sbi->devs)
+		return -ENOMEM;
+
+	for (i = 0; i < max_devices; i++) {
+
+		if (i > 0 && !RDEV(i).path[0])
+			break;
+
+		if (max_devices == 1) {
+			/* Single zoned block device mount */
+			FDEV(0).bdev =
+				blkdev_get_by_dev(sbi->sb->s_bdev->bd_dev,
+					sbi->sb->s_mode, sbi->sb->s_type);
+		} else {
+			/* Multi-device mount */
+			memcpy(FDEV(i).path, RDEV(i).path, MAX_PATH_LEN);
+			FDEV(i).total_segments =
+				le32_to_cpu(RDEV(i).total_segments);
+			if (i == 0) {
+				FDEV(i).start_blk = 0;
+				FDEV(i).end_blk = FDEV(i).start_blk +
+				    (FDEV(i).total_segments <<
+				    sbi->log_blocks_per_seg) - 1 +
+				    le32_to_cpu(raw_super->segment0_blkaddr);
+			} else {
+				FDEV(i).start_blk = FDEV(i - 1).end_blk + 1;
+				FDEV(i).end_blk = FDEV(i).start_blk +
+					(FDEV(i).total_segments <<
+					sbi->log_blocks_per_seg) - 1;
+			}
+			FDEV(i).bdev = blkdev_get_by_path(FDEV(i).path,
+					sbi->sb->s_mode, sbi->sb->s_type);
+		}
+		if (IS_ERR(FDEV(i).bdev))
+			return PTR_ERR(FDEV(i).bdev);
+
+		/* to release errored devices */
+		sbi->s_ndevs = i + 1;
+
+#ifdef CONFIG_BLK_DEV_ZONED
+		if (bdev_zoned_model(FDEV(i).bdev) == BLK_ZONED_HM &&
+				!f2fs_sb_mounted_blkzoned(sbi->sb)) {
+			f2fs_msg(sbi->sb, KERN_ERR,
+				"Zoned block device feature not enabled\n");
+			return -EINVAL;
+		}
+		if (bdev_zoned_model(FDEV(i).bdev) != BLK_ZONED_NONE) {
+			if (init_blkz_info(sbi, i)) {
+				f2fs_msg(sbi->sb, KERN_ERR,
+					"Failed to initialize F2FS blkzone information");
+				return -EINVAL;
+			}
+			if (max_devices == 1)
+				break;
+			f2fs_msg(sbi->sb, KERN_INFO,
+				"Mount Device [%2d]: %20s, %8u, %8x - %8x (zone: %s)",
+				i, FDEV(i).path,
+				FDEV(i).total_segments,
+				FDEV(i).start_blk, FDEV(i).end_blk,
+				bdev_zoned_model(FDEV(i).bdev) == BLK_ZONED_HA ?
+				"Host-aware" : "Host-managed");
+			continue;
+		}
+#endif
+		f2fs_msg(sbi->sb, KERN_INFO,
+			"Mount Device [%2d]: %20s, %8u, %8x - %8x",
+				i, FDEV(i).path,
+				FDEV(i).total_segments,
+				FDEV(i).start_blk, FDEV(i).end_blk);
+	}
+	f2fs_msg(sbi->sb, KERN_INFO,
+			"IO Block Size: %8d KB", F2FS_IO_SIZE_KB(sbi));
+	return 0;
+}
+
 static int f2fs_fill_super(struct super_block *sb, void *data, int silent)
 {
 	struct f2fs_sb_info *sbi;
 	struct f2fs_super_block *raw_super;
-	struct buffer_head *raw_super_buf;
 	struct inode *root;
-	long err;
+	int err;
 	bool retry = true, need_fsck = false;
 	char *options = NULL;
-	int recovery, i;
+	int recovery, i, valid_super_block;
+	struct curseg_info *seg_i;
 
 try_onemore:
 	err = -EINVAL;
 	raw_super = NULL;
-	raw_super_buf = NULL;
+	valid_super_block = -1;
 	recovery = 0;
 
 	/* allocate memory for f2fs-specific super block info */
@@ -1265,17 +2260,49 @@ try_onemore:
 	if (!sbi)
 		return -ENOMEM;
 
+	sbi->sb = sb;
+
+	/* Load the checksum driver */
+	sbi->s_chksum_driver = crypto_alloc_shash("crc32", 0, 0);
+	if (IS_ERR(sbi->s_chksum_driver)) {
+		f2fs_msg(sb, KERN_ERR, "Cannot load crc32 driver.");
+		err = PTR_ERR(sbi->s_chksum_driver);
+		sbi->s_chksum_driver = NULL;
+		goto free_sbi;
+	}
+
 	/* set a block size */
 	if (unlikely(!sb_set_blocksize(sb, F2FS_BLKSIZE))) {
 		f2fs_msg(sb, KERN_ERR, "unable to set blocksize");
 		goto free_sbi;
 	}
 
-	err = read_raw_super_block(sb, &raw_super, &raw_super_buf, &recovery);
+	err = read_raw_super_block(sbi, &raw_super, &valid_super_block,
+								&recovery);
 	if (err)
 		goto free_sbi;
 
 	sb->s_fs_info = sbi;
+	sbi->raw_super = raw_super;
+
+	/* precompute checksum seed for metadata */
+	if (f2fs_sb_has_inode_chksum(sb))
+		sbi->s_chksum_seed = f2fs_chksum(sbi, ~0, raw_super->uuid,
+						sizeof(raw_super->uuid));
+
+	/*
+	 * The BLKZONED feature indicates that the drive was formatted with
+	 * zone alignment optimization. This is optional for host-aware
+	 * devices, but mandatory for host-managed zoned block devices.
+	 */
+#ifndef CONFIG_BLK_DEV_ZONED
+	if (f2fs_sb_mounted_blkzoned(sb)) {
+		f2fs_msg(sb, KERN_ERR,
+			 "Zoned block device support is not enabled\n");
+		err = -EOPNOTSUPP;
+		goto free_sb_buf;
+	}
+#endif
 	default_options(sbi);
 	/* parse mount options */
 	options = kstrdup((const char *)data, GFP_KERNEL);
@@ -1288,11 +2315,20 @@ try_onemore:
 	if (err)
 		goto free_options;
 
-	sb->s_maxbytes = max_file_size(le32_to_cpu(raw_super->log_blocksize));
+	sbi->max_file_blocks = max_file_blocks();
+	sb->s_maxbytes = sbi->max_file_blocks <<
+				le32_to_cpu(raw_super->log_blocksize);
 	sb->s_max_links = F2FS_LINK_MAX;
 	get_random_bytes(&sbi->s_next_generation, sizeof(u32));
 
+#ifdef CONFIG_QUOTA
+	sb->dq_op = &f2fs_quota_operations;
+	sb->s_qcop = &f2fs_quotactl_ops;
+	sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP | QTYPE_MASK_PRJ;
+#endif
+
 	sb->s_op = &f2fs_sops;
+	sb->s_cop = &f2fs_cryptops;
 	sb->s_xattr = f2fs_xattr_handlers;
 	sb->s_export_op = &f2fs_export_ops;
 	sb->s_magic = F2FS_SUPER_MAGIC;
@@ -1302,37 +2338,63 @@ try_onemore:
 	memcpy(sb->s_uuid, raw_super->uuid, sizeof(raw_super->uuid));
 
 	/* init f2fs-specific super block info */
-	sbi->sb = sb;
-	sbi->raw_super = raw_super;
-	sbi->raw_super_buf = raw_super_buf;
+	sbi->valid_super_block = valid_super_block;
 	mutex_init(&sbi->gc_mutex);
-	mutex_init(&sbi->writepages);
 	mutex_init(&sbi->cp_mutex);
 	init_rwsem(&sbi->node_write);
+	init_rwsem(&sbi->node_change);
 
 	/* disallow all the data/node/meta page writes */
 	set_sbi_flag(sbi, SBI_POR_DOING);
 	spin_lock_init(&sbi->stat_lock);
 
-	init_rwsem(&sbi->read_io.io_rwsem);
-	sbi->read_io.sbi = sbi;
-	sbi->read_io.bio = NULL;
+	/* init iostat info */
+	spin_lock_init(&sbi->iostat_lock);
+	sbi->iostat_enable = false;
+
 	for (i = 0; i < NR_PAGE_TYPE; i++) {
-		init_rwsem(&sbi->write_io[i].io_rwsem);
-		sbi->write_io[i].sbi = sbi;
-		sbi->write_io[i].bio = NULL;
+		int n = (i == META) ? 1: NR_TEMP_TYPE;
+		int j;
+
+		sbi->write_io[i] = kmalloc(n * sizeof(struct f2fs_bio_info),
+								GFP_KERNEL);
+		if (!sbi->write_io[i]) {
+			err = -ENOMEM;
+			goto free_options;
+		}
+
+		for (j = HOT; j < n; j++) {
+			init_rwsem(&sbi->write_io[i][j].io_rwsem);
+			sbi->write_io[i][j].sbi = sbi;
+			sbi->write_io[i][j].bio = NULL;
+			spin_lock_init(&sbi->write_io[i][j].io_lock);
+			INIT_LIST_HEAD(&sbi->write_io[i][j].io_list);
+		}
 	}
 
 	init_rwsem(&sbi->cp_rwsem);
 	init_waitqueue_head(&sbi->cp_wait);
 	init_sb_info(sbi);
 
+	err = init_percpu_info(sbi);
+	if (err)
+		goto free_options;
+
+	if (F2FS_IO_SIZE(sbi) > 1) {
+		sbi->write_io_dummy =
+			mempool_create_page_pool(2 * (F2FS_IO_SIZE(sbi) - 1), 0);
+		if (!sbi->write_io_dummy) {
+			err = -ENOMEM;
+			goto free_options;
+		}
+	}
+
 	/* get an inode for meta space */
 	sbi->meta_inode = f2fs_iget(sb, F2FS_META_INO(sbi));
 	if (IS_ERR(sbi->meta_inode)) {
 		f2fs_msg(sb, KERN_ERR, "Failed to read F2FS meta data inode");
 		err = PTR_ERR(sbi->meta_inode);
-		goto free_options;
+		goto free_io_dummy;
 	}
 
 	err = get_valid_checkpoint(sbi);
@@ -1341,24 +2403,27 @@ try_onemore:
 		goto free_meta_inode;
 	}
 
-	/* sanity checking of checkpoint */
-	err = -EINVAL;
-	if (sanity_check_ckpt(sbi)) {
-		f2fs_msg(sb, KERN_ERR, "Invalid F2FS checkpoint");
-		goto free_cp;
+	/* Initialize device list */
+	err = f2fs_scan_devices(sbi);
+	if (err) {
+		f2fs_msg(sb, KERN_ERR, "Failed to find devices");
+		goto free_devices;
 	}
 
 	sbi->total_valid_node_count =
 				le32_to_cpu(sbi->ckpt->valid_node_count);
-	sbi->total_valid_inode_count =
-				le32_to_cpu(sbi->ckpt->valid_inode_count);
+	percpu_counter_set(&sbi->total_valid_inode_count,
+				le32_to_cpu(sbi->ckpt->valid_inode_count));
 	sbi->user_block_count = le64_to_cpu(sbi->ckpt->user_block_count);
 	sbi->total_valid_block_count =
 				le64_to_cpu(sbi->ckpt->valid_block_count);
 	sbi->last_valid_block_count = sbi->total_valid_block_count;
-	sbi->alloc_valid_block_count = 0;
-	INIT_LIST_HEAD(&sbi->dir_inode_list);
-	spin_lock_init(&sbi->dir_inode_lock);
+	sbi->reserved_blocks = 0;
+
+	for (i = 0; i < NR_INODE_TYPE; i++) {
+		INIT_LIST_HEAD(&sbi->inode_list[i]);
+		spin_lock_init(&sbi->inode_lock[i]);
+	}
 
 	init_extent_cache_info(sbi);
 
@@ -1378,6 +2443,17 @@ try_onemore:
 		goto free_nm;
 	}
 
+	/* For write statistics */
+	if (sb->s_bdev->bd_part)
+		sbi->sectors_written_start =
+			(u64)part_stat_read(sb->s_bdev->bd_part, sectors[1]);
+
+	/* Read accumulated write IO statistics if exists */
+	seg_i = CURSEG_I(sbi, CURSEG_HOT_NODE);
+	if (__exist_node_summaries(sbi))
+		sbi->kbytes_written =
+			le64_to_cpu(seg_i->journal->info.kbytes_written);
+
 	build_gc_manager(sbi);
 
 	/* get an inode for node space */
@@ -1390,10 +2466,9 @@ try_onemore:
 
 	f2fs_join_shrinker(sbi);
 
-	/* if there are nt orphan nodes free them */
-	err = recover_orphan_inodes(sbi);
+	err = f2fs_build_stats(sbi);
 	if (err)
-		goto free_node_inode;
+		goto free_nm;
 
 	/* read root inode and dentry */
 	root = f2fs_iget(sb, F2FS_ROOT_INO(sbi));
@@ -1414,23 +2489,14 @@ try_onemore:
 		goto free_root_inode;
 	}
 
-	err = f2fs_build_stats(sbi);
+	err = f2fs_register_sysfs(sbi);
 	if (err)
 		goto free_root_inode;
 
-	if (f2fs_proc_root)
-		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);
-
-	if (sbi->s_proc)
-		proc_create_data("segment_info", S_IRUGO, sbi->s_proc,
-				 &f2fs_seq_segment_info_fops, sb);
-
-	sbi->s_kobj.kset = f2fs_kset;
-	init_completion(&sbi->s_kobj_unregister);
-	err = kobject_init_and_add(&sbi->s_kobj, &f2fs_ktype, NULL,
-							"%s", sb->s_id);
+	/* if there are nt orphan nodes free them */
+	err = recover_orphan_inodes(sbi);
 	if (err)
-		goto free_proc;
+		goto free_sysfs;
 
 	/* recover fsynced data */
 	if (!test_opt(sbi, DISABLE_ROLL_FORWARD)) {
@@ -1439,22 +2505,35 @@ try_onemore:
 		 * previous checkpoint was not done by clean system shutdown.
 		 */
 		if (bdev_read_only(sb->s_bdev) &&
-				!is_set_ckpt_flags(sbi->ckpt, CP_UMOUNT_FLAG)) {
+				!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) {
 			err = -EROFS;
-			goto free_kobj;
+			goto free_meta;
 		}
 
 		if (need_fsck)
 			set_sbi_flag(sbi, SBI_NEED_FSCK);
 
-		err = recover_fsync_data(sbi);
-		if (err) {
+		if (!retry)
+			goto skip_recovery;
+
+		err = recover_fsync_data(sbi, false);
+		if (err < 0) {
 			need_fsck = true;
 			f2fs_msg(sb, KERN_ERR,
-				"Cannot recover all fsync data errno=%ld", err);
-			goto free_kobj;
+				"Cannot recover all fsync data errno=%d", err);
+			goto free_meta;
+		}
+	} else {
+		err = recover_fsync_data(sbi, true);
+
+		if (!f2fs_readonly(sb) && err > 0) {
+			err = -EINVAL;
+			f2fs_msg(sb, KERN_ERR,
+				"Need to recover fsync data");
+			goto free_sysfs;
 		}
 	}
+skip_recovery:
 	/* recover_fsync_data() cleared this already */
 	clear_sbi_flag(sbi, SBI_POR_DOING);
 
@@ -1466,50 +2545,72 @@ try_onemore:
 		/* After POR, we can run background GC thread.*/
 		err = start_gc_thread(sbi);
 		if (err)
-			goto free_kobj;
+			goto free_meta;
 	}
 	kfree(options);
 
 	/* recover broken superblock */
-	if (recovery && !f2fs_readonly(sb) && !bdev_read_only(sb->s_bdev)) {
-		f2fs_msg(sb, KERN_INFO, "Recover invalid superblock");
-		f2fs_commit_super(sbi, true);
+	if (recovery) {
+		err = f2fs_commit_super(sbi, true);
+		f2fs_msg(sb, KERN_INFO,
+			"Try to recover %dth superblock, ret: %d",
+			sbi->valid_super_block ? 1 : 2, err);
 	}
 
-	sbi->cp_expires = round_jiffies_up(jiffies);
-
+	f2fs_msg(sbi->sb, KERN_NOTICE, "Mounted with checkpoint version = %llx",
+				cur_cp_version(F2FS_CKPT(sbi)));
+	f2fs_update_time(sbi, CP_TIME);
+	f2fs_update_time(sbi, REQ_TIME);
 	return 0;
 
-free_kobj:
-	kobject_del(&sbi->s_kobj);
-free_proc:
-	if (sbi->s_proc) {
-		remove_proc_entry("segment_info", sbi->s_proc);
-		remove_proc_entry(sb->s_id, f2fs_proc_root);
-	}
-	f2fs_destroy_stats(sbi);
+free_meta:
+	f2fs_sync_inode_meta(sbi);
+	/*
+	 * Some dirty meta pages can be produced by recover_orphan_inodes()
+	 * failed by EIO. Then, iput(node_inode) can trigger balance_fs_bg()
+	 * followed by write_checkpoint() through f2fs_write_node_pages(), which
+	 * falls into an infinite loop in sync_meta_pages().
+	 */
+	truncate_inode_pages_final(META_MAPPING(sbi));
+free_sysfs:
+	f2fs_unregister_sysfs(sbi);
 free_root_inode:
 	dput(sb->s_root);
 	sb->s_root = NULL;
 free_node_inode:
+	truncate_inode_pages_final(NODE_MAPPING(sbi));
 	mutex_lock(&sbi->umount_mutex);
+	release_ino_entry(sbi, true);
 	f2fs_leave_shrinker(sbi);
 	iput(sbi->node_inode);
 	mutex_unlock(&sbi->umount_mutex);
+	f2fs_destroy_stats(sbi);
 free_nm:
 	destroy_node_manager(sbi);
 free_sm:
 	destroy_segment_manager(sbi);
-free_cp:
+free_devices:
+	destroy_device_list(sbi);
 	kfree(sbi->ckpt);
 free_meta_inode:
 	make_bad_inode(sbi->meta_inode);
 	iput(sbi->meta_inode);
+free_io_dummy:
+	mempool_destroy(sbi->write_io_dummy);
 free_options:
+	for (i = 0; i < NR_PAGE_TYPE; i++)
+		kfree(sbi->write_io[i]);
+	destroy_percpu_info(sbi);
+#ifdef CONFIG_QUOTA
+	for (i = 0; i < MAXQUOTAS; i++)
+		kfree(sbi->s_qf_names[i]);
+#endif
 	kfree(options);
 free_sb_buf:
-	brelse(raw_super_buf);
+	kfree(raw_super);
 free_sbi:
+	if (sbi->s_chksum_driver)
+		crypto_free_shash(sbi->s_chksum_driver);
 	kfree(sbi);
 
 	/* give only one another chance */
@@ -1529,8 +2630,11 @@ static struct dentry *f2fs_mount(struct file_system_type *fs_type, int flags,
 
 static void kill_f2fs_super(struct super_block *sb)
 {
-	if (sb->s_root)
+	if (sb->s_root) {
 		set_sbi_flag(F2FS_SB(sb), SBI_IS_CLOSE);
+		stop_gc_thread(F2FS_SB(sb));
+		stop_discard_thread(F2FS_SB(sb));
+	}
 	kill_block_super(sb);
 }
 
@@ -1545,8 +2649,9 @@ MODULE_ALIAS_FS("f2fs");
 
 static int __init init_inodecache(void)
 {
-	f2fs_inode_cachep = f2fs_kmem_cache_create("f2fs_inode_cache",
-			sizeof(struct f2fs_inode_info));
+	f2fs_inode_cachep = kmem_cache_create("f2fs_inode_cache",
+			sizeof(struct f2fs_inode_info), 0,
+			SLAB_RECLAIM_ACCOUNT, NULL);
 	if (!f2fs_inode_cachep)
 		return -ENOMEM;
 	return 0;
@@ -1583,32 +2688,26 @@ static int __init init_f2fs_fs(void)
 	err = create_extent_cache();
 	if (err)
 		goto free_checkpoint_caches;
-	f2fs_kset = kset_create_and_add("f2fs", NULL, fs_kobj);
-	if (!f2fs_kset) {
-		err = -ENOMEM;
-		goto free_extent_cache;
-	}
-	err = f2fs_init_crypto();
+	err = f2fs_init_sysfs();
 	if (err)
-		goto free_kset;
-
+		goto free_extent_cache;
 	err = register_shrinker(&f2fs_shrinker_info);
 	if (err)
-		goto free_crypto;
-
+		goto free_sysfs;
 	err = register_filesystem(&f2fs_fs_type);
 	if (err)
 		goto free_shrinker;
-	f2fs_create_root_stats();
-	f2fs_proc_root = proc_mkdir("fs/f2fs", NULL);
+	err = f2fs_create_root_stats();
+	if (err)
+		goto free_filesystem;
 	return 0;
 
+free_filesystem:
+	unregister_filesystem(&f2fs_fs_type);
 free_shrinker:
 	unregister_shrinker(&f2fs_shrinker_info);
-free_crypto:
-	f2fs_exit_crypto();
-free_kset:
-	kset_unregister(f2fs_kset);
+free_sysfs:
+	f2fs_exit_sysfs();
 free_extent_cache:
 	destroy_extent_cache();
 free_checkpoint_caches:
@@ -1625,17 +2724,15 @@ fail:
 
 static void __exit exit_f2fs_fs(void)
 {
-	remove_proc_entry("fs/f2fs", NULL);
 	f2fs_destroy_root_stats();
-	unregister_shrinker(&f2fs_shrinker_info);
 	unregister_filesystem(&f2fs_fs_type);
-	f2fs_exit_crypto();
+	unregister_shrinker(&f2fs_shrinker_info);
+	f2fs_exit_sysfs();
 	destroy_extent_cache();
 	destroy_checkpoint_caches();
 	destroy_segment_manager_caches();
 	destroy_node_manager_caches();
 	destroy_inodecache();
-	kset_unregister(f2fs_kset);
 	f2fs_destroy_trace_ios();
 }
 
@@ -1645,3 +2742,4 @@ module_exit(exit_f2fs_fs)
 MODULE_AUTHOR("Samsung Electronics's Praesto Team");
 MODULE_DESCRIPTION("Flash Friendly File System");
 MODULE_LICENSE("GPL");
+
diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c
new file mode 100644
index 000000000000..e2c258f717cd
--- /dev/null
+++ b/fs/f2fs/sysfs.c
@@ -0,0 +1,556 @@
+/*
+ * f2fs sysfs interface
+ *
+ * Copyright (c) 2012 Samsung Electronics Co., Ltd.
+ *             http://www.samsung.com/
+ * Copyright (c) 2017 Chao Yu <chao@kernel.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#include <linux/proc_fs.h>
+#include <linux/f2fs_fs.h>
+#include <linux/seq_file.h>
+
+#include "f2fs.h"
+#include "segment.h"
+#include "gc.h"
+
+static struct proc_dir_entry *f2fs_proc_root;
+
+/* Sysfs support for f2fs */
+enum {
+	GC_THREAD,	/* struct f2fs_gc_thread */
+	SM_INFO,	/* struct f2fs_sm_info */
+	DCC_INFO,	/* struct discard_cmd_control */
+	NM_INFO,	/* struct f2fs_nm_info */
+	F2FS_SBI,	/* struct f2fs_sb_info */
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	FAULT_INFO_RATE,	/* struct f2fs_fault_info */
+	FAULT_INFO_TYPE,	/* struct f2fs_fault_info */
+#endif
+	RESERVED_BLOCKS,
+};
+
+struct f2fs_attr {
+	struct attribute attr;
+	ssize_t (*show)(struct f2fs_attr *, struct f2fs_sb_info *, char *);
+	ssize_t (*store)(struct f2fs_attr *, struct f2fs_sb_info *,
+			 const char *, size_t);
+	int struct_type;
+	int offset;
+	int id;
+};
+
+static unsigned char *__struct_ptr(struct f2fs_sb_info *sbi, int struct_type)
+{
+	if (struct_type == GC_THREAD)
+		return (unsigned char *)sbi->gc_thread;
+	else if (struct_type == SM_INFO)
+		return (unsigned char *)SM_I(sbi);
+	else if (struct_type == DCC_INFO)
+		return (unsigned char *)SM_I(sbi)->dcc_info;
+	else if (struct_type == NM_INFO)
+		return (unsigned char *)NM_I(sbi);
+	else if (struct_type == F2FS_SBI || struct_type == RESERVED_BLOCKS)
+		return (unsigned char *)sbi;
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	else if (struct_type == FAULT_INFO_RATE ||
+					struct_type == FAULT_INFO_TYPE)
+		return (unsigned char *)&sbi->fault_info;
+#endif
+	return NULL;
+}
+
+static ssize_t lifetime_write_kbytes_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+	struct super_block *sb = sbi->sb;
+
+	if (!sb->s_bdev->bd_part)
+		return snprintf(buf, PAGE_SIZE, "0\n");
+
+	return snprintf(buf, PAGE_SIZE, "%llu\n",
+		(unsigned long long)(sbi->kbytes_written +
+			BD_PART_WRITTEN(sbi)));
+}
+
+static ssize_t features_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+	struct super_block *sb = sbi->sb;
+	int len = 0;
+
+	if (!sb->s_bdev->bd_part)
+		return snprintf(buf, PAGE_SIZE, "0\n");
+
+	if (f2fs_sb_has_crypto(sb))
+		len += snprintf(buf, PAGE_SIZE - len, "%s",
+						"encryption");
+	if (f2fs_sb_mounted_blkzoned(sb))
+		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "blkzoned");
+	if (f2fs_sb_has_extra_attr(sb))
+		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "extra_attr");
+	if (f2fs_sb_has_project_quota(sb))
+		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "projquota");
+	if (f2fs_sb_has_inode_chksum(sb))
+		len += snprintf(buf + len, PAGE_SIZE - len, "%s%s",
+				len ? ", " : "", "inode_checksum");
+	len += snprintf(buf + len, PAGE_SIZE - len, "\n");
+	return len;
+}
+
+static ssize_t f2fs_sbi_show(struct f2fs_attr *a,
+			struct f2fs_sb_info *sbi, char *buf)
+{
+	unsigned char *ptr = NULL;
+	unsigned int *ui;
+
+	ptr = __struct_ptr(sbi, a->struct_type);
+	if (!ptr)
+		return -EINVAL;
+
+	ui = (unsigned int *)(ptr + a->offset);
+
+	return snprintf(buf, PAGE_SIZE, "%u\n", *ui);
+}
+
+static ssize_t f2fs_sbi_store(struct f2fs_attr *a,
+			struct f2fs_sb_info *sbi,
+			const char *buf, size_t count)
+{
+	unsigned char *ptr;
+	unsigned long t;
+	unsigned int *ui;
+	ssize_t ret;
+
+	ptr = __struct_ptr(sbi, a->struct_type);
+	if (!ptr)
+		return -EINVAL;
+
+	ui = (unsigned int *)(ptr + a->offset);
+
+	ret = kstrtoul(skip_spaces(buf), 0, &t);
+	if (ret < 0)
+		return ret;
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	if (a->struct_type == FAULT_INFO_TYPE && t >= (1 << FAULT_MAX))
+		return -EINVAL;
+#endif
+	if (a->struct_type == RESERVED_BLOCKS) {
+		spin_lock(&sbi->stat_lock);
+		if ((unsigned long)sbi->total_valid_block_count + t >
+				(unsigned long)sbi->user_block_count) {
+			spin_unlock(&sbi->stat_lock);
+			return -EINVAL;
+		}
+		*ui = t;
+		spin_unlock(&sbi->stat_lock);
+		return count;
+	}
+
+	if (!strcmp(a->attr.name, "discard_granularity")) {
+		struct discard_cmd_control *dcc = SM_I(sbi)->dcc_info;
+		int i;
+
+		if (t == 0 || t > MAX_PLIST_NUM)
+			return -EINVAL;
+		if (t == *ui)
+			return count;
+
+		mutex_lock(&dcc->cmd_lock);
+		for (i = 0; i < MAX_PLIST_NUM; i++) {
+			if (i >= t - 1)
+				dcc->pend_list_tag[i] |= P_ACTIVE;
+			else
+				dcc->pend_list_tag[i] &= (~P_ACTIVE);
+		}
+		mutex_unlock(&dcc->cmd_lock);
+
+		*ui = t;
+		return count;
+	}
+
+	*ui = t;
+
+	if (!strcmp(a->attr.name, "iostat_enable") && *ui == 0)
+		f2fs_reset_iostat(sbi);
+	if (!strcmp(a->attr.name, "gc_urgent") && t == 1 && sbi->gc_thread) {
+		sbi->gc_thread->gc_wake = 1;
+		wake_up_interruptible_all(&sbi->gc_thread->gc_wait_queue_head);
+		wake_up_discard_thread(sbi, true);
+	}
+
+	return count;
+}
+
+static ssize_t f2fs_attr_show(struct kobject *kobj,
+				struct attribute *attr, char *buf)
+{
+	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
+								s_kobj);
+	struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr);
+
+	return a->show ? a->show(a, sbi, buf) : 0;
+}
+
+static ssize_t f2fs_attr_store(struct kobject *kobj, struct attribute *attr,
+						const char *buf, size_t len)
+{
+	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
+									s_kobj);
+	struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr);
+
+	return a->store ? a->store(a, sbi, buf, len) : 0;
+}
+
+static void f2fs_sb_release(struct kobject *kobj)
+{
+	struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info,
+								s_kobj);
+	complete(&sbi->s_kobj_unregister);
+}
+
+enum feat_id {
+	FEAT_CRYPTO = 0,
+	FEAT_BLKZONED,
+	FEAT_ATOMIC_WRITE,
+	FEAT_EXTRA_ATTR,
+	FEAT_PROJECT_QUOTA,
+	FEAT_INODE_CHECKSUM,
+};
+
+static ssize_t f2fs_feature_show(struct f2fs_attr *a,
+		struct f2fs_sb_info *sbi, char *buf)
+{
+	switch (a->id) {
+	case FEAT_CRYPTO:
+	case FEAT_BLKZONED:
+	case FEAT_ATOMIC_WRITE:
+	case FEAT_EXTRA_ATTR:
+	case FEAT_PROJECT_QUOTA:
+	case FEAT_INODE_CHECKSUM:
+		return snprintf(buf, PAGE_SIZE, "supported\n");
+	}
+	return 0;
+}
+
+#define F2FS_ATTR_OFFSET(_struct_type, _name, _mode, _show, _store, _offset) \
+static struct f2fs_attr f2fs_attr_##_name = {			\
+	.attr = {.name = __stringify(_name), .mode = _mode },	\
+	.show	= _show,					\
+	.store	= _store,					\
+	.struct_type = _struct_type,				\
+	.offset = _offset					\
+}
+
+#define F2FS_RW_ATTR(struct_type, struct_name, name, elname)	\
+	F2FS_ATTR_OFFSET(struct_type, name, 0644,		\
+		f2fs_sbi_show, f2fs_sbi_store,			\
+		offsetof(struct struct_name, elname))
+
+#define F2FS_GENERAL_RO_ATTR(name) \
+static struct f2fs_attr f2fs_attr_##name = __ATTR(name, 0444, name##_show, NULL)
+
+#define F2FS_FEATURE_RO_ATTR(_name, _id)			\
+static struct f2fs_attr f2fs_attr_##_name = {			\
+	.attr = {.name = __stringify(_name), .mode = 0444 },	\
+	.show	= f2fs_feature_show,				\
+	.id	= _id,						\
+}
+
+F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_urgent_sleep_time,
+							urgent_sleep_time);
+F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_min_sleep_time, min_sleep_time);
+F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_max_sleep_time, max_sleep_time);
+F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_no_gc_sleep_time, no_gc_sleep_time);
+F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_idle, gc_idle);
+F2FS_RW_ATTR(GC_THREAD, f2fs_gc_kthread, gc_urgent, gc_urgent);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, reclaim_segments, rec_prefree_segments);
+F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, max_small_discards, max_discards);
+F2FS_RW_ATTR(DCC_INFO, discard_cmd_control, discard_granularity, discard_granularity);
+F2FS_RW_ATTR(RESERVED_BLOCKS, f2fs_sb_info, reserved_blocks, reserved_blocks);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, batched_trim_sections, trim_sections);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, ipu_policy, ipu_policy);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_ipu_util, min_ipu_util);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_fsync_blocks, min_fsync_blocks);
+F2FS_RW_ATTR(SM_INFO, f2fs_sm_info, min_hot_blocks, min_hot_blocks);
+F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ram_thresh, ram_thresh);
+F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, ra_nid_pages, ra_nid_pages);
+F2FS_RW_ATTR(NM_INFO, f2fs_nm_info, dirty_nats_ratio, dirty_nats_ratio);
+F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, max_victim_search, max_victim_search);
+F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, dir_level, dir_level);
+F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, cp_interval, interval_time[CP_TIME]);
+F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, idle_interval, interval_time[REQ_TIME]);
+F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, iostat_enable, iostat_enable);
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+F2FS_RW_ATTR(FAULT_INFO_RATE, f2fs_fault_info, inject_rate, inject_rate);
+F2FS_RW_ATTR(FAULT_INFO_TYPE, f2fs_fault_info, inject_type, inject_type);
+#endif
+F2FS_GENERAL_RO_ATTR(lifetime_write_kbytes);
+F2FS_GENERAL_RO_ATTR(features);
+
+#ifdef CONFIG_F2FS_FS_ENCRYPTION
+F2FS_FEATURE_RO_ATTR(encryption, FEAT_CRYPTO);
+#endif
+#ifdef CONFIG_BLK_DEV_ZONED
+F2FS_FEATURE_RO_ATTR(block_zoned, FEAT_BLKZONED);
+#endif
+F2FS_FEATURE_RO_ATTR(atomic_write, FEAT_ATOMIC_WRITE);
+F2FS_FEATURE_RO_ATTR(extra_attr, FEAT_EXTRA_ATTR);
+F2FS_FEATURE_RO_ATTR(project_quota, FEAT_PROJECT_QUOTA);
+F2FS_FEATURE_RO_ATTR(inode_checksum, FEAT_INODE_CHECKSUM);
+
+#define ATTR_LIST(name) (&f2fs_attr_##name.attr)
+static struct attribute *f2fs_attrs[] = {
+	ATTR_LIST(gc_urgent_sleep_time),
+	ATTR_LIST(gc_min_sleep_time),
+	ATTR_LIST(gc_max_sleep_time),
+	ATTR_LIST(gc_no_gc_sleep_time),
+	ATTR_LIST(gc_idle),
+	ATTR_LIST(gc_urgent),
+	ATTR_LIST(reclaim_segments),
+	ATTR_LIST(max_small_discards),
+	ATTR_LIST(discard_granularity),
+	ATTR_LIST(batched_trim_sections),
+	ATTR_LIST(ipu_policy),
+	ATTR_LIST(min_ipu_util),
+	ATTR_LIST(min_fsync_blocks),
+	ATTR_LIST(min_hot_blocks),
+	ATTR_LIST(max_victim_search),
+	ATTR_LIST(dir_level),
+	ATTR_LIST(ram_thresh),
+	ATTR_LIST(ra_nid_pages),
+	ATTR_LIST(dirty_nats_ratio),
+	ATTR_LIST(cp_interval),
+	ATTR_LIST(idle_interval),
+	ATTR_LIST(iostat_enable),
+#ifdef CONFIG_F2FS_FAULT_INJECTION
+	ATTR_LIST(inject_rate),
+	ATTR_LIST(inject_type),
+#endif
+	ATTR_LIST(lifetime_write_kbytes),
+	ATTR_LIST(features),
+	ATTR_LIST(reserved_blocks),
+	NULL,
+};
+
+static struct attribute *f2fs_feat_attrs[] = {
+#ifdef CONFIG_F2FS_FS_ENCRYPTION
+	ATTR_LIST(encryption),
+#endif
+#ifdef CONFIG_BLK_DEV_ZONED
+	ATTR_LIST(block_zoned),
+#endif
+	ATTR_LIST(atomic_write),
+	ATTR_LIST(extra_attr),
+	ATTR_LIST(project_quota),
+	ATTR_LIST(inode_checksum),
+	NULL,
+};
+
+static const struct sysfs_ops f2fs_attr_ops = {
+	.show	= f2fs_attr_show,
+	.store	= f2fs_attr_store,
+};
+
+static struct kobj_type f2fs_sb_ktype = {
+	.default_attrs	= f2fs_attrs,
+	.sysfs_ops	= &f2fs_attr_ops,
+	.release	= f2fs_sb_release,
+};
+
+static struct kobj_type f2fs_ktype = {
+	.sysfs_ops	= &f2fs_attr_ops,
+};
+
+static struct kset f2fs_kset = {
+	.kobj   = {.ktype = &f2fs_ktype},
+};
+
+static struct kobj_type f2fs_feat_ktype = {
+	.default_attrs	= f2fs_feat_attrs,
+	.sysfs_ops	= &f2fs_attr_ops,
+};
+
+static struct kobject f2fs_feat = {
+	.kset	= &f2fs_kset,
+};
+
+static int segment_info_seq_show(struct seq_file *seq, void *offset)
+{
+	struct super_block *sb = seq->private;
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	unsigned int total_segs =
+			le32_to_cpu(sbi->raw_super->segment_count_main);
+	int i;
+
+	seq_puts(seq, "format: segment_type|valid_blocks\n"
+		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");
+
+	for (i = 0; i < total_segs; i++) {
+		struct seg_entry *se = get_seg_entry(sbi, i);
+
+		if ((i % 10) == 0)
+			seq_printf(seq, "%-10d", i);
+		seq_printf(seq, "%d|%-3u", se->type,
+					get_valid_blocks(sbi, i, false));
+		if ((i % 10) == 9 || i == (total_segs - 1))
+			seq_putc(seq, '\n');
+		else
+			seq_putc(seq, ' ');
+	}
+
+	return 0;
+}
+
+static int segment_bits_seq_show(struct seq_file *seq, void *offset)
+{
+	struct super_block *sb = seq->private;
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	unsigned int total_segs =
+			le32_to_cpu(sbi->raw_super->segment_count_main);
+	int i, j;
+
+	seq_puts(seq, "format: segment_type|valid_blocks|bitmaps\n"
+		"segment_type(0:HD, 1:WD, 2:CD, 3:HN, 4:WN, 5:CN)\n");
+
+	for (i = 0; i < total_segs; i++) {
+		struct seg_entry *se = get_seg_entry(sbi, i);
+
+		seq_printf(seq, "%-10d", i);
+		seq_printf(seq, "%d|%-3u|", se->type,
+					get_valid_blocks(sbi, i, false));
+		for (j = 0; j < SIT_VBLOCK_MAP_SIZE; j++)
+			seq_printf(seq, " %.2x", se->cur_valid_map[j]);
+		seq_putc(seq, '\n');
+	}
+	return 0;
+}
+
+static int iostat_info_seq_show(struct seq_file *seq, void *offset)
+{
+	struct super_block *sb = seq->private;
+	struct f2fs_sb_info *sbi = F2FS_SB(sb);
+	time64_t now = ktime_get_real_seconds();
+
+	if (!sbi->iostat_enable)
+		return 0;
+
+	seq_printf(seq, "time:		%-16llu\n", now);
+
+	/* print app IOs */
+	seq_printf(seq, "app buffered:	%-16llu\n",
+				sbi->write_iostat[APP_BUFFERED_IO]);
+	seq_printf(seq, "app direct:	%-16llu\n",
+				sbi->write_iostat[APP_DIRECT_IO]);
+	seq_printf(seq, "app mapped:	%-16llu\n",
+				sbi->write_iostat[APP_MAPPED_IO]);
+
+	/* print fs IOs */
+	seq_printf(seq, "fs data:	%-16llu\n",
+				sbi->write_iostat[FS_DATA_IO]);
+	seq_printf(seq, "fs node:	%-16llu\n",
+				sbi->write_iostat[FS_NODE_IO]);
+	seq_printf(seq, "fs meta:	%-16llu\n",
+				sbi->write_iostat[FS_META_IO]);
+	seq_printf(seq, "fs gc data:	%-16llu\n",
+				sbi->write_iostat[FS_GC_DATA_IO]);
+	seq_printf(seq, "fs gc node:	%-16llu\n",
+				sbi->write_iostat[FS_GC_NODE_IO]);
+	seq_printf(seq, "fs cp data:	%-16llu\n",
+				sbi->write_iostat[FS_CP_DATA_IO]);
+	seq_printf(seq, "fs cp node:	%-16llu\n",
+				sbi->write_iostat[FS_CP_NODE_IO]);
+	seq_printf(seq, "fs cp meta:	%-16llu\n",
+				sbi->write_iostat[FS_CP_META_IO]);
+	seq_printf(seq, "fs discard:	%-16llu\n",
+				sbi->write_iostat[FS_DISCARD]);
+
+	return 0;
+}
+
+#define F2FS_PROC_FILE_DEF(_name)					\
+static int _name##_open_fs(struct inode *inode, struct file *file)	\
+{									\
+	return single_open(file, _name##_seq_show, PDE_DATA(inode));	\
+}									\
+									\
+static const struct file_operations f2fs_seq_##_name##_fops = {		\
+	.open = _name##_open_fs,					\
+	.read = seq_read,						\
+	.llseek = seq_lseek,						\
+	.release = single_release,					\
+};
+
+F2FS_PROC_FILE_DEF(segment_info);
+F2FS_PROC_FILE_DEF(segment_bits);
+F2FS_PROC_FILE_DEF(iostat_info);
+
+int __init f2fs_init_sysfs(void)
+{
+	int ret;
+
+	kobject_set_name(&f2fs_kset.kobj, "f2fs");
+	f2fs_kset.kobj.parent = fs_kobj;
+	ret = kset_register(&f2fs_kset);
+	if (ret)
+		return ret;
+
+	ret = kobject_init_and_add(&f2fs_feat, &f2fs_feat_ktype,
+				   NULL, "features");
+	if (ret)
+		kset_unregister(&f2fs_kset);
+	else
+		f2fs_proc_root = proc_mkdir("fs/f2fs", NULL);
+	return ret;
+}
+
+void f2fs_exit_sysfs(void)
+{
+	kobject_put(&f2fs_feat);
+	kset_unregister(&f2fs_kset);
+	remove_proc_entry("fs/f2fs", NULL);
+	f2fs_proc_root = NULL;
+}
+
+int f2fs_register_sysfs(struct f2fs_sb_info *sbi)
+{
+	struct super_block *sb = sbi->sb;
+	int err;
+
+	sbi->s_kobj.kset = &f2fs_kset;
+	init_completion(&sbi->s_kobj_unregister);
+	err = kobject_init_and_add(&sbi->s_kobj, &f2fs_sb_ktype, NULL,
+				"%s", sb->s_id);
+	if (err)
+		return err;
+
+	if (f2fs_proc_root)
+		sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root);
+
+	if (sbi->s_proc) {
+		proc_create_data("segment_info", S_IRUGO, sbi->s_proc,
+				 &f2fs_seq_segment_info_fops, sb);
+		proc_create_data("segment_bits", S_IRUGO, sbi->s_proc,
+				 &f2fs_seq_segment_bits_fops, sb);
+		proc_create_data("iostat_info", S_IRUGO, sbi->s_proc,
+				&f2fs_seq_iostat_info_fops, sb);
+	}
+	return 0;
+}
+
+void f2fs_unregister_sysfs(struct f2fs_sb_info *sbi)
+{
+	if (sbi->s_proc) {
+		remove_proc_entry("iostat_info", sbi->s_proc);
+		remove_proc_entry("segment_info", sbi->s_proc);
+		remove_proc_entry("segment_bits", sbi->s_proc);
+		remove_proc_entry(sbi->sb->s_id, f2fs_proc_root);
+	}
+	kobject_del(&sbi->s_kobj);
+}
diff --git a/fs/f2fs/trace.c b/fs/f2fs/trace.c
index 145fb659ad44..bccbbf2616d2 100644
--- a/fs/f2fs/trace.c
+++ b/fs/f2fs/trace.c
@@ -25,11 +25,12 @@ static inline void __print_last_io(void)
 	if (!last_io.len)
 		return;
 
-	trace_printk("%3x:%3x %4x %-16s %2x %5x %12x %4x\n",
+	trace_printk("%3x:%3x %4x %-16s %2x %5x %5x %12x %4x\n",
 			last_io.major, last_io.minor,
 			last_io.pid, "----------------",
 			last_io.type,
-			last_io.fio.rw, last_io.fio.blk_addr,
+			last_io.fio.op, last_io.fio.op_flags,
+			last_io.fio.new_blkaddr,
 			last_io.len);
 	memset(&last_io, 0, sizeof(last_io));
 }
@@ -58,7 +59,7 @@ void f2fs_trace_pid(struct page *page)
 	pid_t pid = task_pid_nr(current);
 	void *p;
 
-	page->private = pid;
+	set_page_private(page, (unsigned long)pid);
 
 	if (radix_tree_preload(GFP_NOFS))
 		return;
@@ -100,8 +101,10 @@ void f2fs_trace_ios(struct f2fs_io_info *fio, int flush)
 	if (last_io.major == major && last_io.minor == minor &&
 			last_io.pid == pid &&
 			last_io.type == __file_type(inode, pid) &&
-			last_io.fio.rw == fio->rw &&
-			last_io.fio.blk_addr + last_io.len == fio->blk_addr) {
+			last_io.fio.op == fio->op &&
+			last_io.fio.op_flags == fio->op_flags &&
+			last_io.fio.new_blkaddr + last_io.len ==
+							fio->new_blkaddr) {
 		last_io.len++;
 		return;
 	}
@@ -135,7 +138,7 @@ static unsigned int gang_lookup_pids(pid_t *results, unsigned long first_index,
 
 	radix_tree_for_each_slot(slot, &pids, &iter, first_index) {
 		results[ret] = iter.index;
-		if (++ret == PIDVEC_SIZE)
+		if (++ret == max_items)
 			break;
 	}
 	return ret;
diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c
index 862368a32e53..ab658419552b 100644
--- a/fs/f2fs/xattr.c
+++ b/fs/f2fs/xattr.c
@@ -151,7 +151,7 @@ static int f2fs_xattr_advise_set(const struct xattr_handler *handler,
 		return -EINVAL;
 
 	F2FS_I(inode)->i_advise |= *(char *)value;
-	mark_inode_dirty(inode);
+	f2fs_mark_inode_dirty_sync(inode, true);
 	return 0;
 }
 
@@ -264,18 +264,125 @@ static struct f2fs_xattr_entry *__find_xattr(void *base_addr, int index,
 	return entry;
 }
 
-static void *read_all_xattrs(struct inode *inode, struct page *ipage)
+static struct f2fs_xattr_entry *__find_inline_xattr(void *base_addr,
+					void **last_addr, int index,
+					size_t len, const char *name)
+{
+	struct f2fs_xattr_entry *entry;
+	unsigned int inline_size = F2FS_INLINE_XATTR_ADDRS << 2;
+
+	list_for_each_xattr(entry, base_addr) {
+		if ((void *)entry + sizeof(__u32) > base_addr + inline_size ||
+			(void *)XATTR_NEXT_ENTRY(entry) + sizeof(__u32) >
+			base_addr + inline_size) {
+			*last_addr = entry;
+			return NULL;
+		}
+		if (entry->e_name_index != index)
+			continue;
+		if (entry->e_name_len != len)
+			continue;
+		if (!memcmp(entry->e_name, name, len))
+			break;
+	}
+	return entry;
+}
+
+static int lookup_all_xattrs(struct inode *inode, struct page *ipage,
+				unsigned int index, unsigned int len,
+				const char *name, struct f2fs_xattr_entry **xe,
+				void **base_addr)
+{
+	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
+	void *cur_addr, *txattr_addr, *last_addr = NULL;
+	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
+	unsigned int size = xnid ? VALID_XATTR_BLOCK_SIZE : 0;
+	unsigned int inline_size = inline_xattr_size(inode);
+	int err = 0;
+
+	if (!size && !inline_size)
+		return -ENODATA;
+
+	txattr_addr = kzalloc(inline_size + size + XATTR_PADDING_SIZE,
+							GFP_F2FS_ZERO);
+	if (!txattr_addr)
+		return -ENOMEM;
+
+	/* read from inline xattr */
+	if (inline_size) {
+		struct page *page = NULL;
+		void *inline_addr;
+
+		if (ipage) {
+			inline_addr = inline_xattr_addr(ipage);
+		} else {
+			page = get_node_page(sbi, inode->i_ino);
+			if (IS_ERR(page)) {
+				err = PTR_ERR(page);
+				goto out;
+			}
+			inline_addr = inline_xattr_addr(page);
+		}
+		memcpy(txattr_addr, inline_addr, inline_size);
+		f2fs_put_page(page, 1);
+
+		*xe = __find_inline_xattr(txattr_addr, &last_addr,
+						index, len, name);
+		if (*xe)
+			goto check;
+	}
+
+	/* read from xattr node block */
+	if (xnid) {
+		struct page *xpage;
+		void *xattr_addr;
+
+		/* The inode already has an extended attribute block. */
+		xpage = get_node_page(sbi, xnid);
+		if (IS_ERR(xpage)) {
+			err = PTR_ERR(xpage);
+			goto out;
+		}
+
+		xattr_addr = page_address(xpage);
+		memcpy(txattr_addr + inline_size, xattr_addr, size);
+		f2fs_put_page(xpage, 1);
+	}
+
+	if (last_addr)
+		cur_addr = XATTR_HDR(last_addr) - 1;
+	else
+		cur_addr = txattr_addr;
+
+	*xe = __find_xattr(cur_addr, index, len, name);
+check:
+	if (IS_XATTR_LAST_ENTRY(*xe)) {
+		err = -ENODATA;
+		goto out;
+	}
+
+	*base_addr = txattr_addr;
+	return 0;
+out:
+	kzfree(txattr_addr);
+	return err;
+}
+
+static int read_all_xattrs(struct inode *inode, struct page *ipage,
+							void **base_addr)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
 	struct f2fs_xattr_header *header;
-	size_t size = PAGE_SIZE, inline_size = 0;
+	nid_t xnid = F2FS_I(inode)->i_xattr_nid;
+	unsigned int size = VALID_XATTR_BLOCK_SIZE;
+	unsigned int inline_size = inline_xattr_size(inode);
 	void *txattr_addr;
+	int err;
 
-	inline_size = inline_xattr_size(inode);
-
-	txattr_addr = kzalloc(inline_size + size, GFP_F2FS_ZERO);
+	txattr_addr = kzalloc(inline_size + size + XATTR_PADDING_SIZE,
+							GFP_F2FS_ZERO);
 	if (!txattr_addr)
-		return NULL;
+		return -ENOMEM;
 
 	/* read from inline xattr */
 	if (inline_size) {
@@ -286,8 +393,10 @@ static void *read_all_xattrs(struct inode *inode, struct page *ipage)
 			inline_addr = inline_xattr_addr(ipage);
 		} else {
 			page = get_node_page(sbi, inode->i_ino);
-			if (IS_ERR(page))
+			if (IS_ERR(page)) {
+				err = PTR_ERR(page);
 				goto fail;
+			}
 			inline_addr = inline_xattr_addr(page);
 		}
 		memcpy(txattr_addr, inline_addr, inline_size);
@@ -295,17 +404,19 @@ static void *read_all_xattrs(struct inode *inode, struct page *ipage)
 	}
 
 	/* read from xattr node block */
-	if (F2FS_I(inode)->i_xattr_nid) {
+	if (xnid) {
 		struct page *xpage;
 		void *xattr_addr;
 
 		/* The inode already has an extended attribute block. */
-		xpage = get_node_page(sbi, F2FS_I(inode)->i_xattr_nid);
-		if (IS_ERR(xpage))
+		xpage = get_node_page(sbi, xnid);
+		if (IS_ERR(xpage)) {
+			err = PTR_ERR(xpage);
 			goto fail;
+		}
 
 		xattr_addr = page_address(xpage);
-		memcpy(txattr_addr + inline_size, xattr_addr, PAGE_SIZE);
+		memcpy(txattr_addr + inline_size, xattr_addr, size);
 		f2fs_put_page(xpage, 1);
 	}
 
@@ -316,24 +427,23 @@ static void *read_all_xattrs(struct inode *inode, struct page *ipage)
 		header->h_magic = cpu_to_le32(F2FS_XATTR_MAGIC);
 		header->h_refcount = cpu_to_le32(1);
 	}
-	return txattr_addr;
+	*base_addr = txattr_addr;
+	return 0;
 fail:
 	kzfree(txattr_addr);
-	return NULL;
+	return err;
 }
 
 static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 				void *txattr_addr, struct page *ipage)
 {
 	struct f2fs_sb_info *sbi = F2FS_I_SB(inode);
-	size_t inline_size = 0;
+	size_t inline_size = inline_xattr_size(inode);
 	void *xattr_addr;
 	struct page *xpage;
 	nid_t new_nid = 0;
 	int err;
 
-	inline_size = inline_xattr_size(inode);
-
 	if (hsize > inline_size && !F2FS_I(inode)->i_xattr_nid)
 		if (!alloc_nid(sbi, &new_nid))
 			return -ENOSPC;
@@ -345,7 +455,8 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 
 		if (ipage) {
 			inline_addr = inline_xattr_addr(ipage);
-			f2fs_wait_on_page_writeback(ipage, NODE);
+			f2fs_wait_on_page_writeback(ipage, NODE, true);
+			set_page_dirty(ipage);
 		} else {
 			page = get_node_page(sbi, inode->i_ino);
 			if (IS_ERR(page)) {
@@ -353,7 +464,7 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 				return PTR_ERR(page);
 			}
 			inline_addr = inline_xattr_addr(page);
-			f2fs_wait_on_page_writeback(page, NODE);
+			f2fs_wait_on_page_writeback(page, NODE, true);
 		}
 		memcpy(inline_addr, txattr_addr, inline_size);
 		f2fs_put_page(page, 1);
@@ -374,11 +485,11 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 			return PTR_ERR(xpage);
 		}
 		f2fs_bug_on(sbi, new_nid);
-		f2fs_wait_on_page_writeback(xpage, NODE);
+		f2fs_wait_on_page_writeback(xpage, NODE, true);
 	} else {
 		struct dnode_of_data dn;
 		set_new_dnode(&dn, inode, NULL, NULL, new_nid);
-		xpage = new_node_page(&dn, XATTR_NODE_OFFSET, ipage);
+		xpage = new_node_page(&dn, XATTR_NODE_OFFSET);
 		if (IS_ERR(xpage)) {
 			alloc_nid_failed(sbi, new_nid);
 			return PTR_ERR(xpage);
@@ -387,23 +498,20 @@ static inline int write_all_xattrs(struct inode *inode, __u32 hsize,
 	}
 
 	xattr_addr = page_address(xpage);
-	memcpy(xattr_addr, txattr_addr + inline_size, PAGE_SIZE -
-						sizeof(struct node_footer));
+	memcpy(xattr_addr, txattr_addr + inline_size, VALID_XATTR_BLOCK_SIZE);
 	set_page_dirty(xpage);
 	f2fs_put_page(xpage, 1);
 
-	/* need to checkpoint during fsync */
-	F2FS_I(inode)->xattr_ver = cur_cp_version(F2FS_CKPT(sbi));
 	return 0;
 }
 
 int f2fs_getxattr(struct inode *inode, int index, const char *name,
 		void *buffer, size_t buffer_size, struct page *ipage)
 {
-	struct f2fs_xattr_entry *entry;
-	void *base_addr;
+	struct f2fs_xattr_entry *entry = NULL;
 	int error = 0;
-	size_t size, len;
+	unsigned int size, len;
+	void *base_addr = NULL;
 
 	if (name == NULL)
 		return -EINVAL;
@@ -412,21 +520,18 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
 	if (len > F2FS_NAME_LEN)
 		return -ERANGE;
 
-	base_addr = read_all_xattrs(inode, ipage);
-	if (!base_addr)
-		return -ENOMEM;
-
-	entry = __find_xattr(base_addr, index, len, name);
-	if (IS_XATTR_LAST_ENTRY(entry)) {
-		error = -ENODATA;
-		goto cleanup;
-	}
+	down_read(&F2FS_I(inode)->i_xattr_sem);
+	error = lookup_all_xattrs(inode, ipage, index, len, name,
+				&entry, &base_addr);
+	up_read(&F2FS_I(inode)->i_xattr_sem);
+	if (error)
+		return error;
 
 	size = le16_to_cpu(entry->e_value_size);
 
 	if (buffer && size > buffer_size) {
 		error = -ERANGE;
-		goto cleanup;
+		goto out;
 	}
 
 	if (buffer) {
@@ -434,8 +539,7 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name,
 		memcpy(buffer, pval, size);
 	}
 	error = size;
-
-cleanup:
+out:
 	kzfree(base_addr);
 	return error;
 }
@@ -448,9 +552,11 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size)
 	int error = 0;
 	size_t rest = buffer_size;
 
-	base_addr = read_all_xattrs(inode, NULL);
-	if (!base_addr)
-		return -ENOMEM;
+	down_read(&F2FS_I(inode)->i_xattr_sem);
+	error = read_all_xattrs(inode, NULL, &base_addr);
+	up_read(&F2FS_I(inode)->i_xattr_sem);
+	if (error)
+		return error;
 
 	list_for_each_xattr(entry, base_addr) {
 		const struct xattr_handler *handler =
@@ -477,17 +583,25 @@ cleanup:
 	return error;
 }
 
+static bool f2fs_xattr_value_same(struct f2fs_xattr_entry *entry,
+					const void *value, size_t size)
+{
+	void *pval = entry->e_name + entry->e_name_len;
+
+	return (le16_to_cpu(entry->e_value_size) == size) &&
+					!memcmp(pval, value, size);
+}
+
 static int __f2fs_setxattr(struct inode *inode, int index,
 			const char *name, const void *value, size_t size,
 			struct page *ipage, int flags)
 {
-	struct f2fs_inode_info *fi = F2FS_I(inode);
 	struct f2fs_xattr_entry *here, *last;
 	void *base_addr;
 	int found, newsize;
 	size_t len;
 	__u32 new_hsize;
-	int error = -ENOMEM;
+	int error = 0;
 
 	if (name == NULL)
 		return -EINVAL;
@@ -503,21 +617,26 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 	if (size > MAX_VALUE_LEN(inode))
 		return -E2BIG;
 
-	base_addr = read_all_xattrs(inode, ipage);
-	if (!base_addr)
-		goto exit;
+	error = read_all_xattrs(inode, ipage, &base_addr);
+	if (error)
+		return error;
 
 	/* find entry with wanted name. */
 	here = __find_xattr(base_addr, index, len, name);
 
 	found = IS_XATTR_LAST_ENTRY(here) ? 0 : 1;
 
-	if ((flags & XATTR_REPLACE) && !found) {
+	if (found) {
+		if ((flags & XATTR_CREATE)) {
+			error = -EEXIST;
+			goto exit;
+		}
+
+		if (f2fs_xattr_value_same(here, value, size))
+			goto exit;
+	} else if ((flags & XATTR_REPLACE)) {
 		error = -ENODATA;
 		goto exit;
-	} else if ((flags & XATTR_CREATE) && found) {
-		error = -EEXIST;
-		goto exit;
 	}
 
 	last = here;
@@ -538,7 +657,7 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 			free = free + ENTRY_SIZE(here);
 
 		if (unlikely(free < newsize)) {
-			error = -ENOSPC;
+			error = -E2BIG;
 			goto exit;
 		}
 	}
@@ -566,7 +685,6 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 		 * Before we come here, old entry is removed.
 		 * We just write new entry.
 		 */
-		memset(last, 0, newsize);
 		last->e_name_index = index;
 		last->e_name_len = len;
 		memcpy(last->e_name, name, len);
@@ -580,19 +698,17 @@ static int __f2fs_setxattr(struct inode *inode, int index,
 	if (error)
 		goto exit;
 
-	if (is_inode_flag_set(fi, FI_ACL_MODE)) {
-		inode->i_mode = fi->i_acl_mode;
-		inode->i_ctime = CURRENT_TIME;
-		clear_inode_flag(fi, FI_ACL_MODE);
+	if (is_inode_flag_set(inode, FI_ACL_MODE)) {
+		inode->i_mode = F2FS_I(inode)->i_acl_mode;
+		inode->i_ctime = current_time(inode);
+		clear_inode_flag(inode, FI_ACL_MODE);
 	}
 	if (index == F2FS_XATTR_INDEX_ENCRYPTION &&
 			!strcmp(name, F2FS_XATTR_NAME_ENCRYPTION_CONTEXT))
 		f2fs_set_encrypted_inode(inode);
-
-	if (ipage)
-		update_inode(inode, ipage);
-	else
-		update_inode_page(inode);
+	f2fs_mark_inode_dirty_sync(inode, true);
+	if (!error && S_ISDIR(inode->i_mode))
+		set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_CP);
 exit:
 	kzfree(base_addr);
 	return error;
@@ -609,14 +725,17 @@ int f2fs_setxattr(struct inode *inode, int index, const char *name,
 	if (ipage)
 		return __f2fs_setxattr(inode, index, name, value,
 						size, ipage, flags);
-	f2fs_balance_fs(sbi);
+	f2fs_balance_fs(sbi, true);
 
 	f2fs_lock_op(sbi);
 	/* protect xattr_ver */
 	down_write(&F2FS_I(inode)->i_sem);
+	down_write(&F2FS_I(inode)->i_xattr_sem);
 	err = __f2fs_setxattr(inode, index, name, value, size, ipage, flags);
+	up_write(&F2FS_I(inode)->i_xattr_sem);
 	up_write(&F2FS_I(inode)->i_sem);
 	f2fs_unlock_op(sbi);
 
+	f2fs_update_time(sbi, REQ_TIME);
 	return err;
 }
diff --git a/fs/f2fs/xattr.h b/fs/f2fs/xattr.h
index 71a7100d5492..08a4840d6d7d 100644
--- a/fs/f2fs/xattr.h
+++ b/fs/f2fs/xattr.h
@@ -58,10 +58,10 @@ struct f2fs_xattr_entry {
 #define XATTR_FIRST_ENTRY(ptr)	(XATTR_ENTRY(XATTR_HDR(ptr) + 1))
 #define XATTR_ROUND		(3)
 
-#define XATTR_ALIGN(size)	((size + XATTR_ROUND) & ~XATTR_ROUND)
+#define XATTR_ALIGN(size)	(((size) + XATTR_ROUND) & ~XATTR_ROUND)
 
 #define ENTRY_SIZE(entry) (XATTR_ALIGN(sizeof(struct f2fs_xattr_entry) + \
-			entry->e_name_len + le16_to_cpu(entry->e_value_size)))
+			(entry)->e_name_len + le16_to_cpu((entry)->e_value_size)))
 
 #define XATTR_NEXT_ENTRY(entry)	((struct f2fs_xattr_entry *)((char *)(entry) +\
 			ENTRY_SIZE(entry)))
@@ -72,9 +72,10 @@ struct f2fs_xattr_entry {
 		for (entry = XATTR_FIRST_ENTRY(addr);\
 				!IS_XATTR_LAST_ENTRY(entry);\
 				entry = XATTR_NEXT_ENTRY(entry))
-
-#define MIN_OFFSET(i)	XATTR_ALIGN(inline_xattr_size(i) + PAGE_SIZE -	\
-				sizeof(struct node_footer) - sizeof(__u32))
+#define VALID_XATTR_BLOCK_SIZE	(PAGE_SIZE - sizeof(struct node_footer))
+#define XATTR_PADDING_SIZE	(sizeof(__u32))
+#define MIN_OFFSET(i)		XATTR_ALIGN(inline_xattr_size(i) +	\
+						VALID_XATTR_BLOCK_SIZE)
 
 #define MAX_VALUE_LEN(i)	(MIN_OFFSET(i) -			\
 				sizeof(struct f2fs_xattr_header) -	\
@@ -126,7 +127,8 @@ extern ssize_t f2fs_listxattr(struct dentry *, char *, size_t);
 
 #define f2fs_xattr_handlers	NULL
 static inline int f2fs_setxattr(struct inode *inode, int index,
-		const char *name, const void *value, size_t size, int flags)
+		const char *name, const void *value, size_t size,
+		struct page *page, int flags)
 {
 	return -EOPNOTSUPP;
 }
diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 60d6fc2e0e4b..de11206dda63 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -2062,7 +2062,7 @@ void __mark_inode_dirty(struct inode *inode, int flags)
 	    (dirtytime && (inode->i_state & I_DIRTY_INODE)))
 		return;
 
-	if (unlikely(block_dump))
+	if (unlikely(block_dump > 1))
 		block_dump___mark_inode_dirty(inode);
 
 	spin_lock(&inode->i_lock);
diff --git a/fs/fs_struct.c b/fs/fs_struct.c
index 7dca743b2ce1..940c683561dd 100644
--- a/fs/fs_struct.c
+++ b/fs/fs_struct.c
@@ -44,6 +44,7 @@ void set_fs_pwd(struct fs_struct *fs, const struct path *path)
 	if (old_pwd.dentry)
 		path_put(&old_pwd);
 }
+EXPORT_SYMBOL(set_fs_pwd);
 
 static inline int replace_path(struct path *p, const struct path *old, const struct path *new)
 {
@@ -89,6 +90,7 @@ void free_fs_struct(struct fs_struct *fs)
 	path_put(&fs->pwd);
 	kmem_cache_free(fs_cachep, fs);
 }
+EXPORT_SYMBOL(free_fs_struct);
 
 void exit_fs(struct task_struct *tsk)
 {
@@ -127,6 +129,7 @@ struct fs_struct *copy_fs_struct(struct fs_struct *old)
 	}
 	return fs;
 }
+EXPORT_SYMBOL_GPL(copy_fs_struct);
 
 int unshare_fs_struct(void)
 {
diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c
index d0cf1f010fbe..fbfec06b054d 100644
--- a/fs/fuse/dev.c
+++ b/fs/fuse/dev.c
@@ -13,12 +13,14 @@
 #include <linux/poll.h>
 #include <linux/uio.h>
 #include <linux/miscdevice.h>
+#include <linux/namei.h>
 #include <linux/pagemap.h>
 #include <linux/file.h>
 #include <linux/slab.h>
 #include <linux/pipe_fs_i.h>
 #include <linux/swap.h>
 #include <linux/splice.h>
+#include <linux/freezer.h>
 
 MODULE_ALIAS_MISCDEV(FUSE_MINOR);
 MODULE_ALIAS("devname:fuse");
@@ -477,7 +479,9 @@ static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req)
 	 * Either request is already in userspace, or it was forced.
 	 * Wait it out.
 	 */
-	wait_event(req->waitq, test_bit(FR_FINISHED, &req->flags));
+	while (!test_bit(FR_FINISHED, &req->flags))
+		wait_event_freezable(req->waitq,
+				test_bit(FR_FINISHED, &req->flags));
 }
 
 static void __fuse_request_send(struct fuse_conn *fc, struct fuse_req *req)
@@ -1936,6 +1940,10 @@ static ssize_t fuse_dev_do_write(struct fuse_dev *fud,
 		cs->move_pages = 0;
 
 	err = copy_out_args(cs, &req->out, nbytes);
+	if (req->in.h.opcode == FUSE_CANONICAL_PATH) {
+		req->out.h.error = kern_path((char *)req->out.args[0].value, 0,
+							req->canonical_path);
+	}
 	fuse_copy_finish(cs);
 
 	spin_lock(&fpq->lock);
diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
index 5068dbf80ff8..bfa274c06666 100644
--- a/fs/fuse/dir.c
+++ b/fs/fuse/dir.c
@@ -267,6 +267,50 @@ invalid:
 	goto out;
 }
 
+/*
+ * Get the canonical path. Since we must translate to a path, this must be done
+ * in the context of the userspace daemon, however, the userspace daemon cannot
+ * look up paths on its own. Instead, we handle the lookup as a special case
+ * inside of the write request.
+ */
+static void fuse_dentry_canonical_path(const struct path *path, struct path *canonical_path) {
+	struct inode *inode = path->dentry->d_inode;
+	struct fuse_conn *fc = get_fuse_conn(inode);
+	struct fuse_req *req;
+	int err;
+	char *path_name;
+
+	req = fuse_get_req(fc, 1);
+	err = PTR_ERR(req);
+	if (IS_ERR(req))
+		goto default_path;
+
+	path_name = (char*)__get_free_page(GFP_KERNEL);
+	if (!path_name) {
+		fuse_put_request(fc, req);
+		goto default_path;
+	}
+
+	req->in.h.opcode = FUSE_CANONICAL_PATH;
+	req->in.h.nodeid = get_node_id(inode);
+	req->in.numargs = 0;
+	req->out.numargs = 1;
+	req->out.args[0].size = PATH_MAX;
+	req->out.args[0].value = path_name;
+	req->canonical_path = canonical_path;
+	req->out.argvar = 1;
+	fuse_request_send(fc, req);
+	err = req->out.h.error;
+	fuse_put_request(fc, req);
+	free_page((unsigned long)path_name);
+	if (!err)
+		return;
+default_path:
+	canonical_path->dentry = path->dentry;
+	canonical_path->mnt = path->mnt;
+	path_get(canonical_path);
+}
+
 static int invalid_nodeid(u64 nodeid)
 {
 	return !nodeid || nodeid == FUSE_ROOT_ID;
@@ -274,6 +318,7 @@ static int invalid_nodeid(u64 nodeid)
 
 const struct dentry_operations fuse_dentry_operations = {
 	.d_revalidate	= fuse_dentry_revalidate,
+	.d_canonical_path = fuse_dentry_canonical_path,
 };
 
 int fuse_valid_type(int m)
diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h
index 604cd42dafef..644687ae04bd 100644
--- a/fs/fuse/fuse_i.h
+++ b/fs/fuse/fuse_i.h
@@ -372,6 +372,9 @@ struct fuse_req {
 	/** Inode used in the request or NULL */
 	struct inode *inode;
 
+	/** Path used for completing d_canonical_path */
+	struct path *canonical_path;
+
 	/** AIO control block */
 	struct fuse_io_priv *io;
 
diff --git a/fs/inode.c b/fs/inode.c
index 2c16b758831d..6a7234f0afea 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -1721,7 +1721,7 @@ int dentry_needs_remove_privs(struct dentry *dentry)
 }
 EXPORT_SYMBOL(dentry_needs_remove_privs);
 
-static int __remove_privs(struct dentry *dentry, int kill)
+static int __remove_privs(struct vfsmount *mnt, struct dentry *dentry, int kill)
 {
 	struct iattr newattrs;
 
@@ -1730,7 +1730,7 @@ static int __remove_privs(struct dentry *dentry, int kill)
 	 * Note we call this on write, so notify_change will not
 	 * encounter any conflicting delegations:
 	 */
-	return notify_change(dentry, &newattrs, NULL);
+	return notify_change2(mnt, dentry, &newattrs, NULL);
 }
 
 /*
@@ -1752,7 +1752,7 @@ int file_remove_privs(struct file *file)
 	if (kill < 0)
 		return kill;
 	if (kill)
-		error = __remove_privs(dentry, kill);
+		error = __remove_privs(file->f_path.mnt, dentry, kill);
 	if (!error)
 		inode_has_no_xattr(inode);
 
diff --git a/fs/internal.h b/fs/internal.h
index 71859c4d0b41..6387b35a1c0d 100644
--- a/fs/internal.h
+++ b/fs/internal.h
@@ -84,9 +84,11 @@ extern struct file *get_empty_filp(void);
  * super.c
  */
 extern int do_remount_sb(struct super_block *, int, void *, int);
+extern int do_remount_sb2(struct vfsmount *, struct super_block *, int,
+								void *, int);
 extern bool trylock_super(struct super_block *sb);
 extern struct dentry *mount_fs(struct file_system_type *,
-			       int, const char *, void *);
+			       int, const char *, struct vfsmount *, void *);
 extern struct super_block *user_get_super(dev_t);
 
 /*
diff --git a/fs/mbcache2.c b/fs/mbcache2.c
new file mode 100644
index 000000000000..5c3e1a8c38f6
--- /dev/null
+++ b/fs/mbcache2.c
@@ -0,0 +1,359 @@
+#include <linux/spinlock.h>
+#include <linux/slab.h>
+#include <linux/list.h>
+#include <linux/list_bl.h>
+#include <linux/module.h>
+#include <linux/sched.h>
+#include <linux/mbcache2.h>
+
+/*
+ * Mbcache is a simple key-value store. Keys need not be unique, however
+ * key-value pairs are expected to be unique (we use this fact in
+ * mb2_cache_entry_delete_block()).
+ *
+ * Ext2 and ext4 use this cache for deduplication of extended attribute blocks.
+ * They use hash of a block contents as a key and block number as a value.
+ * That's why keys need not be unique (different xattr blocks may end up having
+ * the same hash). However block number always uniquely identifies a cache
+ * entry.
+ *
+ * We provide functions for creation and removal of entries, search by key,
+ * and a special "delete entry with given key-value pair" operation. Fixed
+ * size hash table is used for fast key lookups.
+ */
+
+struct mb2_cache {
+	/* Hash table of entries */
+	struct hlist_bl_head	*c_hash;
+	/* log2 of hash table size */
+	int			c_bucket_bits;
+	/* Protects c_lru_list, c_entry_count */
+	spinlock_t		c_lru_list_lock;
+	struct list_head	c_lru_list;
+	/* Number of entries in cache */
+	unsigned long		c_entry_count;
+	struct shrinker		c_shrink;
+};
+
+static struct kmem_cache *mb2_entry_cache;
+
+/*
+ * mb2_cache_entry_create - create entry in cache
+ * @cache - cache where the entry should be created
+ * @mask - gfp mask with which the entry should be allocated
+ * @key - key of the entry
+ * @block - block that contains data
+ *
+ * Creates entry in @cache with key @key and records that data is stored in
+ * block @block. The function returns -EBUSY if entry with the same key
+ * and for the same block already exists in cache. Otherwise 0 is returned.
+ */
+int mb2_cache_entry_create(struct mb2_cache *cache, gfp_t mask, u32 key,
+			   sector_t block)
+{
+	struct mb2_cache_entry *entry, *dup;
+	struct hlist_bl_node *dup_node;
+	struct hlist_bl_head *head;
+
+	entry = kmem_cache_alloc(mb2_entry_cache, mask);
+	if (!entry)
+		return -ENOMEM;
+
+	INIT_LIST_HEAD(&entry->e_lru_list);
+	/* One ref for hash, one ref returned */
+	atomic_set(&entry->e_refcnt, 1);
+	entry->e_key = key;
+	entry->e_block = block;
+	head = &cache->c_hash[hash_32(key, cache->c_bucket_bits)];
+	entry->e_hash_list_head = head;
+	hlist_bl_lock(head);
+	hlist_bl_for_each_entry(dup, dup_node, head, e_hash_list) {
+		if (dup->e_key == key && dup->e_block == block) {
+			hlist_bl_unlock(head);
+			kmem_cache_free(mb2_entry_cache, entry);
+			return -EBUSY;
+		}
+	}
+	hlist_bl_add_head(&entry->e_hash_list, head);
+	hlist_bl_unlock(head);
+
+	spin_lock(&cache->c_lru_list_lock);
+	list_add_tail(&entry->e_lru_list, &cache->c_lru_list);
+	/* Grab ref for LRU list */
+	atomic_inc(&entry->e_refcnt);
+	cache->c_entry_count++;
+	spin_unlock(&cache->c_lru_list_lock);
+
+	return 0;
+}
+EXPORT_SYMBOL(mb2_cache_entry_create);
+
+void __mb2_cache_entry_free(struct mb2_cache_entry *entry)
+{
+	kmem_cache_free(mb2_entry_cache, entry);
+}
+EXPORT_SYMBOL(__mb2_cache_entry_free);
+
+static struct mb2_cache_entry *__entry_find(struct mb2_cache *cache,
+					    struct mb2_cache_entry *entry,
+					    u32 key)
+{
+	struct mb2_cache_entry *old_entry = entry;
+	struct hlist_bl_node *node;
+	struct hlist_bl_head *head;
+
+	if (entry)
+		head = entry->e_hash_list_head;
+	else
+		head = &cache->c_hash[hash_32(key, cache->c_bucket_bits)];
+	hlist_bl_lock(head);
+	if (entry && !hlist_bl_unhashed(&entry->e_hash_list))
+		node = entry->e_hash_list.next;
+	else
+		node = hlist_bl_first(head);
+	while (node) {
+		entry = hlist_bl_entry(node, struct mb2_cache_entry,
+				       e_hash_list);
+		if (entry->e_key == key) {
+			atomic_inc(&entry->e_refcnt);
+			goto out;
+		}
+		node = node->next;
+	}
+	entry = NULL;
+out:
+	hlist_bl_unlock(head);
+	if (old_entry)
+		mb2_cache_entry_put(cache, old_entry);
+
+	return entry;
+}
+
+/*
+ * mb2_cache_entry_find_first - find the first entry in cache with given key
+ * @cache: cache where we should search
+ * @key: key to look for
+ *
+ * Search in @cache for entry with key @key. Grabs reference to the first
+ * entry found and returns the entry.
+ */
+struct mb2_cache_entry *mb2_cache_entry_find_first(struct mb2_cache *cache,
+						   u32 key)
+{
+	return __entry_find(cache, NULL, key);
+}
+EXPORT_SYMBOL(mb2_cache_entry_find_first);
+
+/*
+ * mb2_cache_entry_find_next - find next entry in cache with the same
+ * @cache: cache where we should search
+ * @entry: entry to start search from
+ *
+ * Finds next entry in the hash chain which has the same key as @entry.
+ * If @entry is unhashed (which can happen when deletion of entry races
+ * with the search), finds the first entry in the hash chain. The function
+ * drops reference to @entry and returns with a reference to the found entry.
+ */
+struct mb2_cache_entry *mb2_cache_entry_find_next(struct mb2_cache *cache,
+						  struct mb2_cache_entry *entry)
+{
+	return __entry_find(cache, entry, entry->e_key);
+}
+EXPORT_SYMBOL(mb2_cache_entry_find_next);
+
+/* mb2_cache_entry_delete_block - remove information about block from cache
+ * @cache - cache we work with
+ * @key - key of the entry to remove
+ * @block - block containing data for @key
+ *
+ * Remove entry from cache @cache with key @key with data stored in @block.
+ */
+void mb2_cache_entry_delete_block(struct mb2_cache *cache, u32 key,
+				  sector_t block)
+{
+	struct hlist_bl_node *node;
+	struct hlist_bl_head *head;
+	struct mb2_cache_entry *entry;
+
+	head = &cache->c_hash[hash_32(key, cache->c_bucket_bits)];
+	hlist_bl_lock(head);
+	hlist_bl_for_each_entry(entry, node, head, e_hash_list) {
+		if (entry->e_key == key && entry->e_block == block) {
+			/* We keep hash list reference to keep entry alive */
+			hlist_bl_del_init(&entry->e_hash_list);
+			hlist_bl_unlock(head);
+			spin_lock(&cache->c_lru_list_lock);
+			if (!list_empty(&entry->e_lru_list)) {
+				list_del_init(&entry->e_lru_list);
+				cache->c_entry_count--;
+				atomic_dec(&entry->e_refcnt);
+			}
+			spin_unlock(&cache->c_lru_list_lock);
+			mb2_cache_entry_put(cache, entry);
+			return;
+		}
+	}
+	hlist_bl_unlock(head);
+}
+EXPORT_SYMBOL(mb2_cache_entry_delete_block);
+
+/* mb2_cache_entry_touch - cache entry got used
+ * @cache - cache the entry belongs to
+ * @entry - entry that got used
+ *
+ * Move entry in lru list to reflect the fact that it was used.
+ */
+void mb2_cache_entry_touch(struct mb2_cache *cache,
+			   struct mb2_cache_entry *entry)
+{
+	spin_lock(&cache->c_lru_list_lock);
+	if (!list_empty(&entry->e_lru_list))
+		list_move_tail(&cache->c_lru_list, &entry->e_lru_list);
+	spin_unlock(&cache->c_lru_list_lock);
+}
+EXPORT_SYMBOL(mb2_cache_entry_touch);
+
+static unsigned long mb2_cache_count(struct shrinker *shrink,
+				     struct shrink_control *sc)
+{
+	struct mb2_cache *cache = container_of(shrink, struct mb2_cache,
+					       c_shrink);
+
+	return cache->c_entry_count;
+}
+
+/* Shrink number of entries in cache */
+static unsigned long mb2_cache_scan(struct shrinker *shrink,
+				    struct shrink_control *sc)
+{
+	int nr_to_scan = sc->nr_to_scan;
+	struct mb2_cache *cache = container_of(shrink, struct mb2_cache,
+					      c_shrink);
+	struct mb2_cache_entry *entry;
+	struct hlist_bl_head *head;
+	unsigned int shrunk = 0;
+
+	spin_lock(&cache->c_lru_list_lock);
+	while (nr_to_scan-- && !list_empty(&cache->c_lru_list)) {
+		entry = list_first_entry(&cache->c_lru_list,
+					 struct mb2_cache_entry, e_lru_list);
+		list_del_init(&entry->e_lru_list);
+		cache->c_entry_count--;
+		/*
+		 * We keep LRU list reference so that entry doesn't go away
+		 * from under us.
+		 */
+		spin_unlock(&cache->c_lru_list_lock);
+		head = entry->e_hash_list_head;
+		hlist_bl_lock(head);
+		if (!hlist_bl_unhashed(&entry->e_hash_list)) {
+			hlist_bl_del_init(&entry->e_hash_list);
+			atomic_dec(&entry->e_refcnt);
+		}
+		hlist_bl_unlock(head);
+		if (mb2_cache_entry_put(cache, entry))
+			shrunk++;
+		cond_resched();
+		spin_lock(&cache->c_lru_list_lock);
+	}
+	spin_unlock(&cache->c_lru_list_lock);
+
+	return shrunk;
+}
+
+/*
+ * mb2_cache_create - create cache
+ * @bucket_bits: log2 of the hash table size
+ *
+ * Create cache for keys with 2^bucket_bits hash entries.
+ */
+struct mb2_cache *mb2_cache_create(int bucket_bits)
+{
+	struct mb2_cache *cache;
+	int bucket_count = 1 << bucket_bits;
+	int i;
+
+	if (!try_module_get(THIS_MODULE))
+		return NULL;
+
+	cache = kzalloc(sizeof(struct mb2_cache), GFP_KERNEL);
+	if (!cache)
+		goto err_out;
+	cache->c_bucket_bits = bucket_bits;
+	INIT_LIST_HEAD(&cache->c_lru_list);
+	spin_lock_init(&cache->c_lru_list_lock);
+	cache->c_hash = kmalloc(bucket_count * sizeof(struct hlist_bl_head),
+				GFP_KERNEL);
+	if (!cache->c_hash) {
+		kfree(cache);
+		goto err_out;
+	}
+	for (i = 0; i < bucket_count; i++)
+		INIT_HLIST_BL_HEAD(&cache->c_hash[i]);
+
+	cache->c_shrink.count_objects = mb2_cache_count;
+	cache->c_shrink.scan_objects = mb2_cache_scan;
+	cache->c_shrink.seeks = DEFAULT_SEEKS;
+	register_shrinker(&cache->c_shrink);
+
+	return cache;
+
+err_out:
+	module_put(THIS_MODULE);
+	return NULL;
+}
+EXPORT_SYMBOL(mb2_cache_create);
+
+/*
+ * mb2_cache_destroy - destroy cache
+ * @cache: the cache to destroy
+ *
+ * Free all entries in cache and cache itself. Caller must make sure nobody
+ * (except shrinker) can reach @cache when calling this.
+ */
+void mb2_cache_destroy(struct mb2_cache *cache)
+{
+	struct mb2_cache_entry *entry, *next;
+
+	unregister_shrinker(&cache->c_shrink);
+
+	/*
+	 * We don't bother with any locking. Cache must not be used at this
+	 * point.
+	 */
+	list_for_each_entry_safe(entry, next, &cache->c_lru_list, e_lru_list) {
+		if (!hlist_bl_unhashed(&entry->e_hash_list)) {
+			hlist_bl_del_init(&entry->e_hash_list);
+			atomic_dec(&entry->e_refcnt);
+		} else
+			WARN_ON(1);
+		list_del(&entry->e_lru_list);
+		WARN_ON(atomic_read(&entry->e_refcnt) != 1);
+		mb2_cache_entry_put(cache, entry);
+	}
+	kfree(cache->c_hash);
+	kfree(cache);
+	module_put(THIS_MODULE);
+}
+EXPORT_SYMBOL(mb2_cache_destroy);
+
+static int __init mb2cache_init(void)
+{
+	mb2_entry_cache = kmem_cache_create("mbcache",
+				sizeof(struct mb2_cache_entry), 0,
+				SLAB_RECLAIM_ACCOUNT|SLAB_MEM_SPREAD, NULL);
+	BUG_ON(!mb2_entry_cache);
+	return 0;
+}
+
+static void __exit mb2cache_exit(void)
+{
+	kmem_cache_destroy(mb2_entry_cache);
+}
+
+module_init(mb2cache_init)
+module_exit(mb2cache_exit)
+
+MODULE_AUTHOR("Jan Kara <jack@suse.cz>");
+MODULE_DESCRIPTION("Meta block cache (for extended attributes)");
+MODULE_LICENSE("GPL");
diff --git a/fs/mpage.c b/fs/mpage.c
index 6ade29b19494..f37bb01a333b 100644
--- a/fs/mpage.c
+++ b/fs/mpage.c
@@ -30,6 +30,14 @@
 #include <linux/cleancache.h>
 #include "internal.h"
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/android_fs.h>
+
+EXPORT_TRACEPOINT_SYMBOL(android_fs_datawrite_start);
+EXPORT_TRACEPOINT_SYMBOL(android_fs_datawrite_end);
+EXPORT_TRACEPOINT_SYMBOL(android_fs_dataread_start);
+EXPORT_TRACEPOINT_SYMBOL(android_fs_dataread_end);
+
 /*
  * I/O completion handler for multipage BIOs.
  *
@@ -47,6 +55,16 @@ static void mpage_end_io(struct bio *bio)
 	struct bio_vec *bv;
 	int i;
 
+	if (trace_android_fs_dataread_end_enabled() &&
+	    (bio_data_dir(bio) == READ)) {
+		struct page *first_page = bio->bi_io_vec[0].bv_page;
+
+		if (first_page != NULL)
+			trace_android_fs_dataread_end(first_page->mapping->host,
+						      page_offset(first_page),
+						      bio->bi_iter.bi_size);
+	}
+
 	bio_for_each_segment_all(bv, bio, i) {
 		struct page *page = bv->bv_page;
 		page_endio(page, bio_data_dir(bio), bio->bi_error);
@@ -57,6 +75,24 @@ static void mpage_end_io(struct bio *bio)
 
 static struct bio *mpage_bio_submit(int rw, struct bio *bio)
 {
+	if (trace_android_fs_dataread_start_enabled() && (rw == READ)) {
+		struct page *first_page = bio->bi_io_vec[0].bv_page;
+
+		if (first_page != NULL) {
+			char *path, pathbuf[MAX_TRACE_PATHBUF_LEN];
+
+			path = android_fstrace_get_pathname(pathbuf,
+						    MAX_TRACE_PATHBUF_LEN,
+						    first_page->mapping->host);
+			trace_android_fs_dataread_start(
+				first_page->mapping->host,
+				page_offset(first_page),
+				bio->bi_iter.bi_size,
+				current->pid,
+				path,
+				current->comm);
+		}
+	}
 	bio->bi_end_io = mpage_end_io;
 	guard_bio_eod(rw, bio);
 	submit_bio(rw, bio);
diff --git a/fs/namei.c b/fs/namei.c
index 3f96ae087488..f8eeea956503 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -373,9 +373,11 @@ EXPORT_SYMBOL(generic_permission);
  * flag in inode->i_opflags, that says "this has not special
  * permission function, use the fast case".
  */
-static inline int do_inode_permission(struct inode *inode, int mask)
+static inline int do_inode_permission(struct vfsmount *mnt, struct inode *inode, int mask)
 {
 	if (unlikely(!(inode->i_opflags & IOP_FASTPERM))) {
+		if (likely(mnt && inode->i_op->permission2))
+			return inode->i_op->permission2(mnt, inode, mask);
 		if (likely(inode->i_op->permission))
 			return inode->i_op->permission(inode, mask);
 
@@ -399,7 +401,7 @@ static inline int do_inode_permission(struct inode *inode, int mask)
  * This does not check for a read-only file system.  You probably want
  * inode_permission().
  */
-int __inode_permission(struct inode *inode, int mask)
+int __inode_permission2(struct vfsmount *mnt, struct inode *inode, int mask)
 {
 	int retval;
 
@@ -411,7 +413,7 @@ int __inode_permission(struct inode *inode, int mask)
 			return -EACCES;
 	}
 
-	retval = do_inode_permission(inode, mask);
+	retval = do_inode_permission(mnt, inode, mask);
 	if (retval)
 		return retval;
 
@@ -419,7 +421,14 @@ int __inode_permission(struct inode *inode, int mask)
 	if (retval)
 		return retval;
 
-	return security_inode_permission(inode, mask);
+	retval = security_inode_permission(inode, mask);
+	return retval;
+}
+EXPORT_SYMBOL(__inode_permission2);
+
+int __inode_permission(struct inode *inode, int mask)
+{
+	return __inode_permission2(NULL, inode, mask);
 }
 EXPORT_SYMBOL(__inode_permission);
 
@@ -455,14 +464,20 @@ static int sb_permission(struct super_block *sb, struct inode *inode, int mask)
  *
  * When checking for MAY_APPEND, MAY_WRITE must also be set in @mask.
  */
-int inode_permission(struct inode *inode, int mask)
+int inode_permission2(struct vfsmount *mnt, struct inode *inode, int mask)
 {
 	int retval;
 
 	retval = sb_permission(inode->i_sb, inode, mask);
 	if (retval)
 		return retval;
-	return __inode_permission(inode, mask);
+	return __inode_permission2(mnt, inode, mask);
+}
+EXPORT_SYMBOL(inode_permission2);
+
+int inode_permission(struct inode *inode, int mask)
+{
+	return inode_permission2(NULL, inode, mask);
 }
 EXPORT_SYMBOL(inode_permission);
 
@@ -1645,13 +1660,13 @@ static int lookup_slow(struct nameidata *nd, struct path *path)
 static inline int may_lookup(struct nameidata *nd)
 {
 	if (nd->flags & LOOKUP_RCU) {
-		int err = inode_permission(nd->inode, MAY_EXEC|MAY_NOT_BLOCK);
+		int err = inode_permission2(nd->path.mnt, nd->inode, MAY_EXEC|MAY_NOT_BLOCK);
 		if (err != -ECHILD)
 			return err;
 		if (unlazy_walk(nd, NULL, 0))
 			return -ECHILD;
 	}
-	return inode_permission(nd->inode, MAY_EXEC);
+	return inode_permission2(nd->path.mnt, nd->inode, MAY_EXEC);
 }
 
 static inline int handle_dots(struct nameidata *nd, int type)
@@ -2005,11 +2020,12 @@ static const char *path_init(struct nameidata *nd, unsigned flags)
 	nd->depth = 0;
 	if (flags & LOOKUP_ROOT) {
 		struct dentry *root = nd->root.dentry;
+		struct vfsmount *mnt = nd->root.mnt;
 		struct inode *inode = root->d_inode;
 		if (*s) {
 			if (!d_can_lookup(root))
 				return ERR_PTR(-ENOTDIR);
-			retval = inode_permission(inode, MAY_EXEC);
+			retval = inode_permission2(mnt, inode, MAY_EXEC);
 			if (retval)
 				return ERR_PTR(retval);
 		}
@@ -2280,13 +2296,14 @@ EXPORT_SYMBOL(vfs_path_lookup);
 /**
  * lookup_one_len - filesystem helper to lookup single pathname component
  * @name:	pathname component to lookup
+ * @mnt:	mount we are looking up on
  * @base:	base directory to lookup from
  * @len:	maximum length @len should be interpreted to
  *
  * Note that this routine is purely a helper for filesystem usage and should
  * not be called by generic code.
  */
-struct dentry *lookup_one_len(const char *name, struct dentry *base, int len)
+struct dentry *lookup_one_len2(const char *name, struct vfsmount *mnt, struct dentry *base, int len)
 {
 	struct qstr this;
 	unsigned int c;
@@ -2320,12 +2337,18 @@ struct dentry *lookup_one_len(const char *name, struct dentry *base, int len)
 			return ERR_PTR(err);
 	}
 
-	err = inode_permission(base->d_inode, MAY_EXEC);
+	err = inode_permission2(mnt, base->d_inode, MAY_EXEC);
 	if (err)
 		return ERR_PTR(err);
 
 	return __lookup_hash(&this, base, 0);
 }
+EXPORT_SYMBOL(lookup_one_len2);
+
+struct dentry *lookup_one_len(const char *name, struct dentry *base, int len)
+{
+	return lookup_one_len2(name, NULL, base, len);
+}
 EXPORT_SYMBOL(lookup_one_len);
 
 int user_path_at_empty(int dfd, const char __user *name, unsigned flags,
@@ -2552,7 +2575,7 @@ EXPORT_SYMBOL(__check_sticky);
  * 10. We don't allow removal of NFS sillyrenamed files; it's handled by
  *     nfs_async_unlink().
  */
-static int may_delete(struct inode *dir, struct dentry *victim, bool isdir)
+static int may_delete(struct vfsmount *mnt, struct inode *dir, struct dentry *victim, bool isdir)
 {
 	struct inode *inode = d_backing_inode(victim);
 	int error;
@@ -2564,7 +2587,7 @@ static int may_delete(struct inode *dir, struct dentry *victim, bool isdir)
 	BUG_ON(victim->d_parent->d_inode != dir);
 	audit_inode_child(dir, victim, AUDIT_TYPE_CHILD_DELETE);
 
-	error = inode_permission(dir, MAY_WRITE | MAY_EXEC);
+	error = inode_permission2(mnt, dir, MAY_WRITE | MAY_EXEC);
 	if (error)
 		return error;
 	if (IS_APPEND(dir))
@@ -2595,14 +2618,14 @@ static int may_delete(struct inode *dir, struct dentry *victim, bool isdir)
  *  3. We should have write and exec permissions on dir
  *  4. We can't do it if dir is immutable (done in permission())
  */
-static inline int may_create(struct inode *dir, struct dentry *child)
+static inline int may_create(struct vfsmount *mnt, struct inode *dir, struct dentry *child)
 {
 	audit_inode_child(dir, child, AUDIT_TYPE_CHILD_CREATE);
 	if (child->d_inode)
 		return -EEXIST;
 	if (IS_DEADDIR(dir))
 		return -ENOENT;
-	return inode_permission(dir, MAY_WRITE | MAY_EXEC);
+	return inode_permission2(mnt, dir, MAY_WRITE | MAY_EXEC);
 }
 
 /*
@@ -2649,10 +2672,10 @@ void unlock_rename(struct dentry *p1, struct dentry *p2)
 }
 EXPORT_SYMBOL(unlock_rename);
 
-int vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
-		bool want_excl)
+int vfs_create2(struct vfsmount *mnt, struct inode *dir, struct dentry *dentry,
+		umode_t mode, bool want_excl)
 {
-	int error = may_create(dir, dentry);
+	int error = may_create(mnt, dir, dentry);
 	if (error)
 		return error;
 
@@ -2668,11 +2691,19 @@ int vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
 		fsnotify_create(dir, dentry);
 	return error;
 }
+EXPORT_SYMBOL(vfs_create2);
+
+int vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
+		bool want_excl)
+{
+	return vfs_create2(NULL, dir, dentry, mode, want_excl);
+}
 EXPORT_SYMBOL(vfs_create);
 
 static int may_open(struct path *path, int acc_mode, int flag)
 {
 	struct dentry *dentry = path->dentry;
+	struct vfsmount *mnt = path->mnt;
 	struct inode *inode = dentry->d_inode;
 	int error;
 
@@ -2701,7 +2732,7 @@ static int may_open(struct path *path, int acc_mode, int flag)
 		break;
 	}
 
-	error = inode_permission(inode, acc_mode);
+	error = inode_permission2(mnt, inode, acc_mode);
 	if (error)
 		return error;
 
@@ -2736,7 +2767,7 @@ static int handle_truncate(struct file *filp)
 	if (!error)
 		error = security_path_truncate(path);
 	if (!error) {
-		error = do_truncate(path->dentry, 0,
+		error = do_truncate2(path->mnt, path->dentry, 0,
 				    ATTR_MTIME|ATTR_CTIME|ATTR_OPEN,
 				    filp);
 	}
@@ -2757,7 +2788,7 @@ static int may_o_create(struct path *dir, struct dentry *dentry, umode_t mode)
 	if (error)
 		return error;
 
-	error = inode_permission(dir->dentry->d_inode, MAY_WRITE | MAY_EXEC);
+	error = inode_permission2(dir->mnt, dir->dentry->d_inode, MAY_WRITE | MAY_EXEC);
 	if (error)
 		return error;
 
@@ -2943,6 +2974,7 @@ static int lookup_open(struct nameidata *nd, struct path *path,
 			bool got_write, int *opened)
 {
 	struct dentry *dir = nd->path.dentry;
+	struct vfsmount *mnt = nd->path.mnt;
 	struct inode *dir_inode = dir->d_inode;
 	struct dentry *dentry;
 	int error;
@@ -2990,7 +3022,7 @@ static int lookup_open(struct nameidata *nd, struct path *path,
 		error = security_path_mknod(&nd->path, dentry, mode, 0);
 		if (error)
 			goto out_dput;
-		error = vfs_create(dir->d_inode, dentry, mode,
+		error = vfs_create2(mnt, dir->d_inode, dentry, mode,
 				   nd->flags & LOOKUP_EXCL);
 		if (error)
 			goto out_dput;
@@ -3252,7 +3284,7 @@ static int do_tmpfile(struct nameidata *nd, unsigned flags,
 		goto out;
 	dir = path.dentry->d_inode;
 	/* we want directory to be writable */
-	error = inode_permission(dir, MAY_WRITE | MAY_EXEC);
+	error = inode_permission2(path.mnt, dir, MAY_WRITE | MAY_EXEC);
 	if (error)
 		goto out2;
 	if (!dir->i_op->tmpfile) {
@@ -3486,9 +3518,9 @@ inline struct dentry *user_path_create(int dfd, const char __user *pathname,
 }
 EXPORT_SYMBOL(user_path_create);
 
-int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev)
+int vfs_mknod2(struct vfsmount *mnt, struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev)
 {
-	int error = may_create(dir, dentry);
+	int error = may_create(mnt, dir, dentry);
 
 	if (error)
 		return error;
@@ -3512,6 +3544,12 @@ int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev)
 		fsnotify_create(dir, dentry);
 	return error;
 }
+EXPORT_SYMBOL(vfs_mknod2);
+
+int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev)
+{
+	return vfs_mknod2(NULL, dir, dentry, mode, dev);
+}
 EXPORT_SYMBOL(vfs_mknod);
 
 static int may_mknod(umode_t mode)
@@ -3554,10 +3592,10 @@ retry:
 		goto out;
 	switch (mode & S_IFMT) {
 		case 0: case S_IFREG:
-			error = vfs_create(path.dentry->d_inode,dentry,mode,true);
+			error = vfs_create2(path.mnt, path.dentry->d_inode,dentry,mode,true);
 			break;
 		case S_IFCHR: case S_IFBLK:
-			error = vfs_mknod(path.dentry->d_inode,dentry,mode,
+			error = vfs_mknod2(path.mnt, path.dentry->d_inode,dentry,mode,
 					new_decode_dev(dev));
 			break;
 		case S_IFIFO: case S_IFSOCK:
@@ -3578,9 +3616,9 @@ SYSCALL_DEFINE3(mknod, const char __user *, filename, umode_t, mode, unsigned, d
 	return sys_mknodat(AT_FDCWD, filename, mode, dev);
 }
 
-int vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
+int vfs_mkdir2(struct vfsmount *mnt, struct inode *dir, struct dentry *dentry, umode_t mode)
 {
-	int error = may_create(dir, dentry);
+	int error = may_create(mnt, dir, dentry);
 	unsigned max_links = dir->i_sb->s_max_links;
 
 	if (error)
@@ -3602,6 +3640,12 @@ int vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
 		fsnotify_mkdir(dir, dentry);
 	return error;
 }
+EXPORT_SYMBOL(vfs_mkdir2);
+
+int vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
+{
+	return vfs_mkdir2(NULL, dir, dentry, mode);
+}
 EXPORT_SYMBOL(vfs_mkdir);
 
 SYSCALL_DEFINE3(mkdirat, int, dfd, const char __user *, pathname, umode_t, mode)
@@ -3620,7 +3664,7 @@ retry:
 		mode &= ~current_umask();
 	error = security_path_mkdir(&path, dentry, mode);
 	if (!error)
-		error = vfs_mkdir(path.dentry->d_inode, dentry, mode);
+		error = vfs_mkdir2(path.mnt, path.dentry->d_inode, dentry, mode);
 	done_path_create(&path, dentry);
 	if (retry_estale(error, lookup_flags)) {
 		lookup_flags |= LOOKUP_REVAL;
@@ -3659,9 +3703,9 @@ void dentry_unhash(struct dentry *dentry)
 }
 EXPORT_SYMBOL(dentry_unhash);
 
-int vfs_rmdir(struct inode *dir, struct dentry *dentry)
+int vfs_rmdir2(struct vfsmount *mnt, struct inode *dir, struct dentry *dentry)
 {
-	int error = may_delete(dir, dentry, 1);
+	int error = may_delete(mnt, dir, dentry, 1);
 
 	if (error)
 		return error;
@@ -3696,6 +3740,12 @@ out:
 		d_delete(dentry);
 	return error;
 }
+EXPORT_SYMBOL(vfs_rmdir2);
+
+int vfs_rmdir(struct inode *dir, struct dentry *dentry)
+{
+	return vfs_rmdir2(NULL, dir, dentry);
+}
 EXPORT_SYMBOL(vfs_rmdir);
 
 static long do_rmdir(int dfd, const char __user *pathname)
@@ -3741,7 +3791,7 @@ retry:
 	error = security_path_rmdir(&path, dentry);
 	if (error)
 		goto exit3;
-	error = vfs_rmdir(path.dentry->d_inode, dentry);
+	error = vfs_rmdir2(path.mnt, path.dentry->d_inode, dentry);
 exit3:
 	dput(dentry);
 exit2:
@@ -3780,10 +3830,10 @@ SYSCALL_DEFINE1(rmdir, const char __user *, pathname)
  * be appropriate for callers that expect the underlying filesystem not
  * to be NFS exported.
  */
-int vfs_unlink(struct inode *dir, struct dentry *dentry, struct inode **delegated_inode)
+int vfs_unlink2(struct vfsmount *mnt, struct inode *dir, struct dentry *dentry, struct inode **delegated_inode)
 {
 	struct inode *target = dentry->d_inode;
-	int error = may_delete(dir, dentry, 0);
+	int error = may_delete(mnt, dir, dentry, 0);
 
 	if (error)
 		return error;
@@ -3818,6 +3868,12 @@ out:
 
 	return error;
 }
+EXPORT_SYMBOL(vfs_unlink2);
+
+int vfs_unlink(struct inode *dir, struct dentry *dentry, struct inode **delegated_inode)
+{
+	return vfs_unlink2(NULL, dir, dentry, delegated_inode);
+}
 EXPORT_SYMBOL(vfs_unlink);
 
 /*
@@ -3865,7 +3921,7 @@ retry_deleg:
 		error = security_path_unlink(&path, dentry);
 		if (error)
 			goto exit2;
-		error = vfs_unlink(path.dentry->d_inode, dentry, &delegated_inode);
+		error = vfs_unlink2(path.mnt, path.dentry->d_inode, dentry, &delegated_inode);
 exit2:
 		dput(dentry);
 	}
@@ -3915,9 +3971,9 @@ SYSCALL_DEFINE1(unlink, const char __user *, pathname)
 	return do_unlinkat(AT_FDCWD, pathname);
 }
 
-int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname)
+int vfs_symlink2(struct vfsmount *mnt, struct inode *dir, struct dentry *dentry, const char *oldname)
 {
-	int error = may_create(dir, dentry);
+	int error = may_create(mnt, dir, dentry);
 
 	if (error)
 		return error;
@@ -3934,6 +3990,12 @@ int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname)
 		fsnotify_create(dir, dentry);
 	return error;
 }
+EXPORT_SYMBOL(vfs_symlink2);
+
+int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname)
+{
+	return vfs_symlink2(NULL, dir, dentry, oldname);
+}
 EXPORT_SYMBOL(vfs_symlink);
 
 SYSCALL_DEFINE3(symlinkat, const char __user *, oldname,
@@ -3956,7 +4018,7 @@ retry:
 
 	error = security_path_symlink(&path, dentry, from->name);
 	if (!error)
-		error = vfs_symlink(path.dentry->d_inode, dentry, from->name);
+		error = vfs_symlink2(path.mnt, path.dentry->d_inode, dentry, from->name);
 	done_path_create(&path, dentry);
 	if (retry_estale(error, lookup_flags)) {
 		lookup_flags |= LOOKUP_REVAL;
@@ -3991,7 +4053,7 @@ SYSCALL_DEFINE2(symlink, const char __user *, oldname, const char __user *, newn
  * be appropriate for callers that expect the underlying filesystem not
  * to be NFS exported.
  */
-int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry, struct inode **delegated_inode)
+int vfs_link2(struct vfsmount *mnt, struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry, struct inode **delegated_inode)
 {
 	struct inode *inode = old_dentry->d_inode;
 	unsigned max_links = dir->i_sb->s_max_links;
@@ -4000,7 +4062,7 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de
 	if (!inode)
 		return -ENOENT;
 
-	error = may_create(dir, new_dentry);
+	error = may_create(mnt, dir, new_dentry);
 	if (error)
 		return error;
 
@@ -4043,6 +4105,12 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de
 		fsnotify_link(dir, inode, new_dentry);
 	return error;
 }
+EXPORT_SYMBOL(vfs_link2);
+
+int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry, struct inode **delegated_inode)
+{
+	return vfs_link2(NULL, old_dentry, dir, new_dentry, delegated_inode);
+}
 EXPORT_SYMBOL(vfs_link);
 
 /*
@@ -4098,7 +4166,7 @@ retry:
 	error = security_path_link(old_path.dentry, &new_path, new_dentry);
 	if (error)
 		goto out_dput;
-	error = vfs_link(old_path.dentry, new_path.dentry->d_inode, new_dentry, &delegated_inode);
+	error = vfs_link2(old_path.mnt, old_path.dentry, new_path.dentry->d_inode, new_dentry, &delegated_inode);
 out_dput:
 	done_path_create(&new_path, new_dentry);
 	if (delegated_inode) {
@@ -4173,7 +4241,8 @@ SYSCALL_DEFINE2(link, const char __user *, oldname, const char __user *, newname
  *	   ->i_mutex on parents, which works but leads to some truly excessive
  *	   locking].
  */
-int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
+int vfs_rename2(struct vfsmount *mnt,
+	       struct inode *old_dir, struct dentry *old_dentry,
 	       struct inode *new_dir, struct dentry *new_dentry,
 	       struct inode **delegated_inode, unsigned int flags)
 {
@@ -4192,19 +4261,19 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	if (vfs_select_inode(old_dentry, 0) == vfs_select_inode(new_dentry, 0))
 		return 0;
 
-	error = may_delete(old_dir, old_dentry, is_dir);
+	error = may_delete(mnt, old_dir, old_dentry, is_dir);
 	if (error)
 		return error;
 
 	if (!target) {
-		error = may_create(new_dir, new_dentry);
+		error = may_create(mnt, new_dir, new_dentry);
 	} else {
 		new_is_dir = d_is_dir(new_dentry);
 
 		if (!(flags & RENAME_EXCHANGE))
-			error = may_delete(new_dir, new_dentry, is_dir);
+			error = may_delete(mnt, new_dir, new_dentry, is_dir);
 		else
-			error = may_delete(new_dir, new_dentry, new_is_dir);
+			error = may_delete(mnt, new_dir, new_dentry, new_is_dir);
 	}
 	if (error)
 		return error;
@@ -4221,12 +4290,12 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
 	 */
 	if (new_dir != old_dir) {
 		if (is_dir) {
-			error = inode_permission(source, MAY_WRITE);
+			error = inode_permission2(mnt, source, MAY_WRITE);
 			if (error)
 				return error;
 		}
 		if ((flags & RENAME_EXCHANGE) && new_is_dir) {
-			error = inode_permission(target, MAY_WRITE);
+			error = inode_permission2(mnt, target, MAY_WRITE);
 			if (error)
 				return error;
 		}
@@ -4309,6 +4378,14 @@ out:
 
 	return error;
 }
+EXPORT_SYMBOL(vfs_rename2);
+
+int vfs_rename(struct inode *old_dir, struct dentry *old_dentry,
+	       struct inode *new_dir, struct dentry *new_dentry,
+	       struct inode **delegated_inode, unsigned int flags)
+{
+	return vfs_rename2(NULL, old_dir, old_dentry, new_dir, new_dentry, delegated_inode, flags);
+}
 EXPORT_SYMBOL(vfs_rename);
 
 SYSCALL_DEFINE5(renameat2, int, olddfd, const char __user *, oldname,
@@ -4422,7 +4499,7 @@ retry_deleg:
 				     &new_path, new_dentry, flags);
 	if (error)
 		goto exit5;
-	error = vfs_rename(old_path.dentry->d_inode, old_dentry,
+	error = vfs_rename2(old_path.mnt, old_path.dentry->d_inode, old_dentry,
 			   new_path.dentry->d_inode, new_dentry,
 			   &delegated_inode, flags);
 exit5:
@@ -4467,7 +4544,7 @@ SYSCALL_DEFINE2(rename, const char __user *, oldname, const char __user *, newna
 
 int vfs_whiteout(struct inode *dir, struct dentry *dentry)
 {
-	int error = may_create(dir, dentry);
+	int error = may_create(NULL, dir, dentry);
 	if (error)
 		return error;
 
diff --git a/fs/namespace.c b/fs/namespace.c
index ec4078d16eb7..7e14bf1c851c 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
@@ -227,6 +227,7 @@ static struct mount *alloc_vfsmnt(const char *name)
 		mnt->mnt_count = 1;
 		mnt->mnt_writers = 0;
 #endif
+		mnt->mnt.data = NULL;
 
 		INIT_HLIST_NODE(&mnt->mnt_hash);
 		INIT_LIST_HEAD(&mnt->mnt_child);
@@ -581,6 +582,7 @@ int sb_prepare_remount_readonly(struct super_block *sb)
 
 static void free_vfsmnt(struct mount *mnt)
 {
+	kfree(mnt->mnt.data);
 	kfree_const(mnt->mnt_devname);
 #ifdef CONFIG_SMP
 	free_percpu(mnt->mnt_pcp);
@@ -975,10 +977,18 @@ vfs_kern_mount(struct file_system_type *type, int flags, const char *name, void
 	if (!mnt)
 		return ERR_PTR(-ENOMEM);
 
+	if (type->alloc_mnt_data) {
+		mnt->mnt.data = type->alloc_mnt_data();
+		if (!mnt->mnt.data) {
+			mnt_free_id(mnt);
+			free_vfsmnt(mnt);
+			return ERR_PTR(-ENOMEM);
+		}
+	}
 	if (flags & MS_KERNMOUNT)
 		mnt->mnt.mnt_flags = MNT_INTERNAL;
 
-	root = mount_fs(type, flags, name, data);
+	root = mount_fs(type, flags, name, &mnt->mnt, data);
 	if (IS_ERR(root)) {
 		mnt_free_id(mnt);
 		free_vfsmnt(mnt);
@@ -1007,6 +1017,14 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root,
 	if (!mnt)
 		return ERR_PTR(-ENOMEM);
 
+	if (sb->s_op->clone_mnt_data) {
+		mnt->mnt.data = sb->s_op->clone_mnt_data(old->mnt.data);
+		if (!mnt->mnt.data) {
+			err = -ENOMEM;
+			goto out_free;
+		}
+	}
+
 	if (flag & (CL_SLAVE | CL_PRIVATE | CL_SHARED_TO_SLAVE))
 		mnt->mnt_group_id = 0; /* not a peer of original */
 	else
@@ -2288,8 +2306,14 @@ static int do_remount(struct path *path, int flags, int mnt_flags,
 		err = change_mount_flags(path->mnt, flags);
 	else if (!capable(CAP_SYS_ADMIN))
 		err = -EPERM;
-	else
-		err = do_remount_sb(sb, flags, data, 0);
+	else {
+		err = do_remount_sb2(path->mnt, sb, flags, data, 0);
+		namespace_lock();
+		lock_mount_hash();
+		propagate_remount(mnt);
+		unlock_mount_hash();
+		namespace_unlock();
+	}
 	if (!err) {
 		lock_mount_hash();
 		mnt_flags |= mnt->mnt.mnt_flags & ~MNT_USER_SETTABLE_MASK;
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index a64313868d3a..2958e7a81f9c 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -488,7 +488,7 @@ static int fanotify_find_path(int dfd, const char __user *filename,
 	}
 
 	/* you can only watch an inode if you have read permissions on it */
-	ret = inode_permission(path->dentry->d_inode, MAY_READ);
+	ret = inode_permission2(path->mnt, path->dentry->d_inode, MAY_READ);
 	if (ret)
 		path_put(path);
 out:
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index b8d08d0d0a4d..4c5b43d15e6e 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -337,7 +337,7 @@ static int inotify_find_inode(const char __user *dirname, struct path *path, uns
 	if (error)
 		return error;
 	/* you can only watch an inode if you have read permissions on it */
-	error = inode_permission(path->dentry->d_inode, MAY_READ);
+	error = inode_permission2(path->mnt, path->dentry->d_inode, MAY_READ);
 	if (error)
 		path_put(path);
 	return error;
@@ -702,6 +702,8 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname,
 	struct fsnotify_group *group;
 	struct inode *inode;
 	struct path path;
+	struct path alteredpath;
+	struct path *canonical_path = &path;
 	struct fd f;
 	int ret;
 	unsigned flags = 0;
@@ -741,13 +743,22 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname,
 	if (ret)
 		goto fput_and_out;
 
+	/* support stacked filesystems */
+	if(path.dentry && path.dentry->d_op) {
+		if (path.dentry->d_op->d_canonical_path) {
+			path.dentry->d_op->d_canonical_path(&path, &alteredpath);
+			canonical_path = &alteredpath;
+			path_put(&path);
+		}
+	}
+
 	/* inode held in place by reference to path; group by fget on fd */
-	inode = path.dentry->d_inode;
+	inode = canonical_path->dentry->d_inode;
 	group = f.file->private_data;
 
 	/* create/update an inode mark */
 	ret = inotify_update_watch(group, inode, mask);
-	path_put(&path);
+	path_put(canonical_path);
 fput_and_out:
 	fdput(f);
 	return ret;
diff --git a/fs/open.c b/fs/open.c
index fbc5c7b230b3..1fd96c5d3895 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -34,8 +34,8 @@
 
 #include "internal.h"
 
-int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs,
-	struct file *filp)
+int do_truncate2(struct vfsmount *mnt, struct dentry *dentry, loff_t length,
+		unsigned int time_attrs, struct file *filp)
 {
 	int ret;
 	struct iattr newattrs;
@@ -60,17 +60,24 @@ int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs,
 
 	mutex_lock(&dentry->d_inode->i_mutex);
 	/* Note any delegations or leases have already been broken: */
-	ret = notify_change(dentry, &newattrs, NULL);
+	ret = notify_change2(mnt, dentry, &newattrs, NULL);
 	mutex_unlock(&dentry->d_inode->i_mutex);
 	return ret;
 }
+int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs,
+	struct file *filp)
+{
+	return do_truncate2(NULL, dentry, length, time_attrs, filp);
+}
 
 long vfs_truncate(struct path *path, loff_t length)
 {
 	struct inode *inode;
+	struct vfsmount *mnt;
 	long error;
 
 	inode = path->dentry->d_inode;
+	mnt = path->mnt;
 
 	/* For directories it's -EISDIR, for other non-regulars - -EINVAL */
 	if (S_ISDIR(inode->i_mode))
@@ -82,7 +89,7 @@ long vfs_truncate(struct path *path, loff_t length)
 	if (error)
 		goto out;
 
-	error = inode_permission(inode, MAY_WRITE);
+	error = inode_permission2(mnt, inode, MAY_WRITE);
 	if (error)
 		goto mnt_drop_write_and_out;
 
@@ -106,7 +113,7 @@ long vfs_truncate(struct path *path, loff_t length)
 	if (!error)
 		error = security_path_truncate(path);
 	if (!error)
-		error = do_truncate(path->dentry, length, 0, NULL);
+		error = do_truncate2(mnt, path->dentry, length, 0, NULL);
 
 put_write_and_out:
 	put_write_access(inode);
@@ -155,6 +162,7 @@ static long do_sys_ftruncate(unsigned int fd, loff_t length, int small)
 {
 	struct inode *inode;
 	struct dentry *dentry;
+	struct vfsmount *mnt;
 	struct fd f;
 	int error;
 
@@ -171,6 +179,7 @@ static long do_sys_ftruncate(unsigned int fd, loff_t length, int small)
 		small = 0;
 
 	dentry = f.file->f_path.dentry;
+	mnt = f.file->f_path.mnt;
 	inode = dentry->d_inode;
 	error = -EINVAL;
 	if (!S_ISREG(inode->i_mode) || !(f.file->f_mode & FMODE_WRITE))
@@ -190,7 +199,7 @@ static long do_sys_ftruncate(unsigned int fd, loff_t length, int small)
 	if (!error)
 		error = security_path_truncate(&f.file->f_path);
 	if (!error)
-		error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, f.file);
+		error = do_truncate2(mnt, dentry, length, ATTR_MTIME|ATTR_CTIME, f.file);
 	sb_end_write(inode->i_sb);
 out_putf:
 	fdput(f);
@@ -340,6 +349,7 @@ SYSCALL_DEFINE3(faccessat, int, dfd, const char __user *, filename, int, mode)
 	struct cred *override_cred;
 	struct path path;
 	struct inode *inode;
+	struct vfsmount *mnt;
 	int res;
 	unsigned int lookup_flags = LOOKUP_FOLLOW;
 
@@ -370,6 +380,7 @@ retry:
 		goto out;
 
 	inode = d_backing_inode(path.dentry);
+	mnt = path.mnt;
 
 	if ((mode & MAY_EXEC) && S_ISREG(inode->i_mode)) {
 		/*
@@ -381,7 +392,7 @@ retry:
 			goto out_path_release;
 	}
 
-	res = inode_permission(inode, mode | MAY_ACCESS);
+	res = inode_permission2(mnt, inode, mode | MAY_ACCESS);
 	/* SuS v2 requires we report a read only fs too */
 	if (res || !(mode & S_IWOTH) || special_file(inode->i_mode))
 		goto out_path_release;
@@ -425,7 +436,7 @@ retry:
 	if (error)
 		goto out;
 
-	error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
+	error = inode_permission2(path.mnt, path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
 	if (error)
 		goto dput_and_out;
 
@@ -445,6 +456,7 @@ SYSCALL_DEFINE1(fchdir, unsigned int, fd)
 {
 	struct fd f = fdget_raw(fd);
 	struct inode *inode;
+	struct vfsmount *mnt;
 	int error = -EBADF;
 
 	error = -EBADF;
@@ -452,12 +464,13 @@ SYSCALL_DEFINE1(fchdir, unsigned int, fd)
 		goto out;
 
 	inode = file_inode(f.file);
+	mnt = f.file->f_path.mnt;
 
 	error = -ENOTDIR;
 	if (!S_ISDIR(inode->i_mode))
 		goto out_putf;
 
-	error = inode_permission(inode, MAY_EXEC | MAY_CHDIR);
+	error = inode_permission2(mnt, inode, MAY_EXEC | MAY_CHDIR);
 	if (!error)
 		set_fs_pwd(current->fs, &f.file->f_path);
 out_putf:
@@ -476,7 +489,7 @@ retry:
 	if (error)
 		goto out;
 
-	error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
+	error = inode_permission2(path.mnt, path.dentry->d_inode, MAY_EXEC | MAY_CHDIR);
 	if (error)
 		goto dput_and_out;
 
@@ -516,7 +529,7 @@ retry_deleg:
 		goto out_unlock;
 	newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO);
 	newattrs.ia_valid = ATTR_MODE | ATTR_CTIME;
-	error = notify_change(path->dentry, &newattrs, &delegated_inode);
+	error = notify_change2(path->mnt, path->dentry, &newattrs, &delegated_inode);
 out_unlock:
 	mutex_unlock(&inode->i_mutex);
 	if (delegated_inode) {
@@ -596,7 +609,7 @@ retry_deleg:
 	mutex_lock(&inode->i_mutex);
 	error = security_path_chown(path, uid, gid);
 	if (!error)
-		error = notify_change(path->dentry, &newattrs, &delegated_inode);
+		error = notify_change2(path->mnt, path->dentry, &newattrs, &delegated_inode);
 	mutex_unlock(&inode->i_mutex);
 	if (delegated_inode) {
 		error = break_deleg_wait(&delegated_inode);
diff --git a/fs/pnode.c b/fs/pnode.c
index d15c63e97ef1..ddb846f878b8 100644
--- a/fs/pnode.c
+++ b/fs/pnode.c
@@ -609,3 +609,37 @@ int propagate_umount(struct list_head *list)
 
 	return 0;
 }
+
+/*
+ *  Iterates over all slaves, and slaves of slaves.
+ */
+static struct mount *next_descendent(struct mount *root, struct mount *cur)
+{
+	if (!IS_MNT_NEW(cur) && !list_empty(&cur->mnt_slave_list))
+		return first_slave(cur);
+	do {
+		struct mount *master = cur->mnt_master;
+
+		if (!master || cur->mnt_slave.next != &master->mnt_slave_list) {
+			struct mount *next = next_slave(cur);
+
+			return (next == root) ? NULL : next;
+		}
+		cur = master;
+	} while (cur != root);
+	return NULL;
+}
+
+void propagate_remount(struct mount *mnt)
+{
+	struct mount *m = mnt;
+	struct super_block *sb = mnt->mnt.mnt_sb;
+
+	if (sb->s_op->copy_mnt_data) {
+		m = next_descendent(mnt, m);
+		while (m) {
+			sb->s_op->copy_mnt_data(m->mnt.data, mnt->mnt.data);
+			m = next_descendent(mnt, m);
+		}
+	}
+}
diff --git a/fs/pnode.h b/fs/pnode.h
index dc87e65becd2..a9a6576540ad 100644
--- a/fs/pnode.h
+++ b/fs/pnode.h
@@ -44,6 +44,7 @@ int propagate_mnt(struct mount *, struct mountpoint *, struct mount *,
 int propagate_umount(struct list_head *);
 int propagate_mount_busy(struct mount *, int);
 void propagate_mount_unlock(struct mount *);
+void propagate_remount(struct mount *);
 void mnt_release_group_id(struct mount *);
 int get_dominating_id(struct mount *mnt, const struct path *root);
 unsigned int mnt_get_count(struct mount *mnt);
diff --git a/fs/proc/base.c b/fs/proc/base.c
index dd732400578e..deafb880368b 100644
--- a/fs/proc/base.c
+++ b/fs/proc/base.c
@@ -2240,6 +2240,92 @@ static const struct file_operations proc_timers_operations = {
 	.release	= seq_release_private,
 };
 
+static ssize_t timerslack_ns_write(struct file *file, const char __user *buf,
+					size_t count, loff_t *offset)
+{
+	struct inode *inode = file_inode(file);
+	struct task_struct *p;
+	u64 slack_ns;
+	int err;
+
+	err = kstrtoull_from_user(buf, count, 10, &slack_ns);
+	if (err < 0)
+		return err;
+
+	p = get_proc_task(inode);
+	if (!p)
+		return -ESRCH;
+
+	if (p != current) {
+		if (!capable(CAP_SYS_NICE)) {
+			count = -EPERM;
+			goto out;
+		}
+
+		err = security_task_setscheduler(p);
+		if (err) {
+			count = err;
+			goto out;
+		}
+	}
+
+	task_lock(p);
+	if (slack_ns == 0)
+		p->timer_slack_ns = p->default_timer_slack_ns;
+	else
+		p->timer_slack_ns = slack_ns;
+	task_unlock(p);
+
+out:
+	put_task_struct(p);
+
+	return count;
+}
+
+static int timerslack_ns_show(struct seq_file *m, void *v)
+{
+	struct inode *inode = m->private;
+	struct task_struct *p;
+	int err = 0;
+
+	p = get_proc_task(inode);
+	if (!p)
+		return -ESRCH;
+
+	if (p != current) {
+
+		if (!capable(CAP_SYS_NICE)) {
+			err = -EPERM;
+			goto out;
+		}
+		err = security_task_getscheduler(p);
+		if (err)
+			goto out;
+	}
+
+	task_lock(p);
+	seq_printf(m, "%llu\n", p->timer_slack_ns);
+	task_unlock(p);
+
+out:
+	put_task_struct(p);
+
+	return err;
+}
+
+static int timerslack_ns_open(struct inode *inode, struct file *filp)
+{
+	return single_open(filp, timerslack_ns_show, inode);
+}
+
+static const struct file_operations proc_pid_set_timerslack_ns_operations = {
+	.open		= timerslack_ns_open,
+	.read		= seq_read,
+	.write		= timerslack_ns_write,
+	.llseek		= seq_lseek,
+	.release	= single_release,
+};
+
 static int proc_pident_instantiate(struct inode *dir,
 	struct dentry *dentry, struct task_struct *task, const void *ptr)
 {
@@ -2790,8 +2876,8 @@ static const struct pid_entry tgid_base_stuff[] = {
 	ONE("cgroup",  S_IRUGO, proc_cgroup_show),
 #endif
 	ONE("oom_score",  S_IRUGO, proc_oom_score),
-	REG("oom_adj",    S_IRUGO|S_IWUSR, proc_oom_adj_operations),
-	REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations),
+	REG("oom_adj",    S_IRUSR, proc_oom_adj_operations),
+	REG("oom_score_adj", S_IRUSR, proc_oom_score_adj_operations),
 #ifdef CONFIG_AUDITSYSCALL
 	REG("loginuid",   S_IWUSR|S_IRUGO, proc_loginuid_operations),
 	REG("sessionid",  S_IRUGO, proc_sessionid_operations),
@@ -2817,6 +2903,7 @@ static const struct pid_entry tgid_base_stuff[] = {
 #ifdef CONFIG_CHECKPOINT_RESTORE
 	REG("timers",	  S_IRUGO, proc_timers_operations),
 #endif
+	REG("timerslack_ns", S_IRUGO|S_IWUGO, proc_pid_set_timerslack_ns_operations),
 };
 
 static int proc_tgid_base_readdir(struct file *file, struct dir_context *ctx)
@@ -3076,6 +3163,44 @@ int proc_pid_readdir(struct file *file, struct dir_context *ctx)
 }
 
 /*
+ * proc_tid_comm_permission is a special permission function exclusively
+ * used for the node /proc/<pid>/task/<tid>/comm.
+ * It bypasses generic permission checks in the case where a task of the same
+ * task group attempts to access the node.
+ * The rational behind this is that glibc and bionic access this node for
+ * cross thread naming (pthread_set/getname_np(!self)). However, if
+ * PR_SET_DUMPABLE gets set to 0 this node among others becomes uid=0 gid=0,
+ * which locks out the cross thread naming implementation.
+ * This function makes sure that the node is always accessible for members of
+ * same thread group.
+ */
+static int proc_tid_comm_permission(struct inode *inode, int mask)
+{
+	bool is_same_tgroup;
+	struct task_struct *task;
+
+	task = get_proc_task(inode);
+	if (!task)
+		return -ESRCH;
+	is_same_tgroup = same_thread_group(current, task);
+	put_task_struct(task);
+
+	if (likely(is_same_tgroup && !(mask & MAY_EXEC))) {
+		/* This file (/proc/<pid>/task/<tid>/comm) can always be
+		 * read or written by the members of the corresponding
+		 * thread group.
+		 */
+		return 0;
+	}
+
+	return generic_permission(inode, mask);
+}
+
+static const struct inode_operations proc_tid_comm_inode_operations = {
+		.permission = proc_tid_comm_permission,
+};
+
+/*
  * Tasks
  */
 static const struct pid_entry tid_base_stuff[] = {
@@ -3093,7 +3218,9 @@ static const struct pid_entry tid_base_stuff[] = {
 #ifdef CONFIG_SCHED_DEBUG
 	REG("sched",     S_IRUGO|S_IWUSR, proc_pid_sched_operations),
 #endif
-	REG("comm",      S_IRUGO|S_IWUSR, proc_pid_set_comm_operations),
+	NOD("comm",      S_IFREG|S_IRUGO|S_IWUSR,
+			 &proc_tid_comm_inode_operations,
+			 &proc_pid_set_comm_operations, {}),
 #ifdef CONFIG_HAVE_ARCH_TRACEHOOK
 	ONE("syscall",   S_IRUSR, proc_pid_syscall),
 #endif
@@ -3140,8 +3267,8 @@ static const struct pid_entry tid_base_stuff[] = {
 	ONE("cgroup",  S_IRUGO, proc_cgroup_show),
 #endif
 	ONE("oom_score", S_IRUGO, proc_oom_score),
-	REG("oom_adj",   S_IRUGO|S_IWUSR, proc_oom_adj_operations),
-	REG("oom_score_adj", S_IRUGO|S_IWUSR, proc_oom_score_adj_operations),
+	REG("oom_adj",   S_IRUSR, proc_oom_adj_operations),
+	REG("oom_score_adj", S_IRUSR, proc_oom_score_adj_operations),
 #ifdef CONFIG_AUDITSYSCALL
 	REG("loginuid",  S_IWUSR|S_IRUGO, proc_loginuid_operations),
 	REG("sessionid",  S_IRUGO, proc_sessionid_operations),
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
index 07ef85e19fbc..38be952b2fe7 100644
--- a/fs/proc/task_mmu.c
+++ b/fs/proc/task_mmu.c
@@ -116,6 +116,56 @@ static void release_task_mempolicy(struct proc_maps_private *priv)
 }
 #endif
 
+static void seq_print_vma_name(struct seq_file *m, struct vm_area_struct *vma)
+{
+	const char __user *name = vma_get_anon_name(vma);
+	struct mm_struct *mm = vma->vm_mm;
+
+	unsigned long page_start_vaddr;
+	unsigned long page_offset;
+	unsigned long num_pages;
+	unsigned long max_len = NAME_MAX;
+	int i;
+
+	page_start_vaddr = (unsigned long)name & PAGE_MASK;
+	page_offset = (unsigned long)name - page_start_vaddr;
+	num_pages = DIV_ROUND_UP(page_offset + max_len, PAGE_SIZE);
+
+	seq_puts(m, "[anon:");
+
+	for (i = 0; i < num_pages; i++) {
+		int len;
+		int write_len;
+		const char *kaddr;
+		long pages_pinned;
+		struct page *page;
+
+		pages_pinned = get_user_pages(current, mm, page_start_vaddr,
+				1, 0, 0, &page, NULL);
+		if (pages_pinned < 1) {
+			seq_puts(m, "<fault>]");
+			return;
+		}
+
+		kaddr = (const char *)kmap(page);
+		len = min(max_len, PAGE_SIZE - page_offset);
+		write_len = strnlen(kaddr + page_offset, len);
+		seq_write(m, kaddr + page_offset, write_len);
+		kunmap(page);
+		put_page(page);
+
+		/* if strnlen hit a null terminator then we're done */
+		if (write_len != len)
+			break;
+
+		max_len -= len;
+		page_offset = 0;
+		page_start_vaddr += PAGE_SIZE;
+	}
+
+	seq_putc(m, ']');
+}
+
 static void vma_stop(struct proc_maps_private *priv)
 {
 	struct mm_struct *mm = priv->mm;
@@ -337,8 +387,14 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
 			goto done;
 		}
 
-		if (is_stack(priv, vma, is_pid))
+		if (is_stack(priv, vma, is_pid)) {
 			name = "[stack]";
+			goto done;
+		}
+		if (vma_get_anon_name(vma)) {
+			seq_pad(m, ' ');
+			seq_print_vma_name(m, vma);
+		}
 	}
 
 done:
@@ -663,6 +719,12 @@ static int show_smap(struct seq_file *m, void *v, int is_pid)
 
 	show_map_vma(m, vma, is_pid);
 
+	if (vma_get_anon_name(vma)) {
+		seq_puts(m, "Name:           ");
+		seq_print_vma_name(m, vma);
+		seq_putc(m, '\n');
+	}
+
 	seq_printf(m,
 		   "Size:           %8lu kB\n"
 		   "Rss:            %8lu kB\n"
diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c
index 87645955990d..961e597acfc6 100644
--- a/fs/proc_namespace.c
+++ b/fs/proc_namespace.c
@@ -118,7 +118,9 @@ static int show_vfsmnt(struct seq_file *m, struct vfsmount *mnt)
 	if (err)
 		goto out;
 	show_mnt_opts(m, mnt);
-	if (sb->s_op->show_options)
+	if (sb->s_op->show_options2)
+			err = sb->s_op->show_options2(mnt, m, mnt_path.dentry);
+	else if (sb->s_op->show_options)
 		err = sb->s_op->show_options(m, mnt_path.dentry);
 	seq_puts(m, " 0 0\n");
 out:
@@ -178,7 +180,9 @@ static int show_mountinfo(struct seq_file *m, struct vfsmount *mnt)
 	err = show_sb_opts(m, sb);
 	if (err)
 		goto out;
-	if (sb->s_op->show_options)
+	if (sb->s_op->show_options2) {
+		err = sb->s_op->show_options2(mnt, m, mnt->mnt_root);
+	} else if (sb->s_op->show_options)
 		err = sb->s_op->show_options(m, mnt->mnt_root);
 	seq_putc(m, '\n');
 out:
diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c
index 588461bb2dd4..40a0fe0a4e05 100644
--- a/fs/pstore/platform.c
+++ b/fs/pstore/platform.c
@@ -431,6 +431,40 @@ static int pstore_write_compat(enum pstore_type_id type,
 			     size, psi);
 }
 
+static int pstore_write_buf_user_compat(enum pstore_type_id type,
+			       enum kmsg_dump_reason reason,
+			       u64 *id, unsigned int part,
+			       const char __user *buf,
+			       bool compressed, size_t size,
+			       struct pstore_info *psi)
+{
+	unsigned long flags = 0;
+	size_t i, bufsize = size;
+	long ret = 0;
+
+	if (unlikely(!access_ok(VERIFY_READ, buf, size)))
+		return -EFAULT;
+	if (bufsize > psinfo->bufsize)
+		bufsize = psinfo->bufsize;
+	spin_lock_irqsave(&psinfo->buf_lock, flags);
+	for (i = 0; i < size; ) {
+		size_t c = min(size - i, bufsize);
+
+		ret = __copy_from_user(psinfo->buf, buf + i, c);
+		if (unlikely(ret != 0)) {
+			ret = -EFAULT;
+			break;
+		}
+		ret = psi->write_buf(type, reason, id, part, psinfo->buf,
+				     compressed, c, psi);
+		if (unlikely(ret < 0))
+			break;
+		i += c;
+	}
+	spin_unlock_irqrestore(&psinfo->buf_lock, flags);
+	return unlikely(ret < 0) ? ret : size;
+}
+
 /*
  * platform specific persistent storage driver registers with
  * us here. If pstore is already mounted, call the platform
@@ -453,6 +487,8 @@ int pstore_register(struct pstore_info *psi)
 
 	if (!psi->write)
 		psi->write = pstore_write_compat;
+	if (!psi->write_buf_user)
+		psi->write_buf_user = pstore_write_buf_user_compat;
 	psinfo = psi;
 	mutex_init(&psinfo->read_mutex);
 	spin_unlock(&pstore_lock);
diff --git a/fs/pstore/pmsg.c b/fs/pstore/pmsg.c
index 7de20cd3797f..78f6176c020f 100644
--- a/fs/pstore/pmsg.c
+++ b/fs/pstore/pmsg.c
@@ -19,48 +19,25 @@
 #include "internal.h"
 
 static DEFINE_MUTEX(pmsg_lock);
-#define PMSG_MAX_BOUNCE_BUFFER_SIZE (2*PAGE_SIZE)
 
 static ssize_t write_pmsg(struct file *file, const char __user *buf,
 			  size_t count, loff_t *ppos)
 {
-	size_t i, buffer_size;
-	char *buffer;
+	u64 id;
+	int ret;
 
 	if (!count)
 		return 0;
 
+	/* check outside lock, page in any data. write_buf_user also checks */
 	if (!access_ok(VERIFY_READ, buf, count))
 		return -EFAULT;
 
-	buffer_size = count;
-	if (buffer_size > PMSG_MAX_BOUNCE_BUFFER_SIZE)
-		buffer_size = PMSG_MAX_BOUNCE_BUFFER_SIZE;
-	buffer = vmalloc(buffer_size);
-	if (!buffer)
-		return -ENOMEM;
-
 	mutex_lock(&pmsg_lock);
-	for (i = 0; i < count; ) {
-		size_t c = min(count - i, buffer_size);
-		u64 id;
-		long ret;
-
-		ret = __copy_from_user(buffer, buf + i, c);
-		if (unlikely(ret != 0)) {
-			mutex_unlock(&pmsg_lock);
-			vfree(buffer);
-			return -EFAULT;
-		}
-		psinfo->write_buf(PSTORE_TYPE_PMSG, 0, &id, 0, buffer, 0, c,
-				  psinfo);
-
-		i += c;
-	}
-
+	ret = psinfo->write_buf_user(PSTORE_TYPE_PMSG, 0, &id, 0, buf, 0, count,
+				     psinfo);
 	mutex_unlock(&pmsg_lock);
-	vfree(buffer);
-	return count;
+	return ret ? ret : count;
 }
 
 static const struct file_operations pmsg_fops = {
diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c
index 59d93acc29c7..c9e4bc47c79d 100644
--- a/fs/pstore/ram.c
+++ b/fs/pstore/ram.c
@@ -34,6 +34,8 @@
 #include <linux/slab.h>
 #include <linux/compiler.h>
 #include <linux/pstore_ram.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
 
 #define RAMOOPS_KERNMSG_HDR "===="
 #define MIN_MEM_SIZE 4096UL
@@ -329,6 +331,24 @@ static int notrace ramoops_pstore_write_buf(enum pstore_type_id type,
 	return 0;
 }
 
+static int notrace ramoops_pstore_write_buf_user(enum pstore_type_id type,
+						 enum kmsg_dump_reason reason,
+						 u64 *id, unsigned int part,
+						 const char __user *buf,
+						 bool compressed, size_t size,
+						 struct pstore_info *psi)
+{
+	if (type == PSTORE_TYPE_PMSG) {
+		struct ramoops_context *cxt = psi->data;
+
+		if (!cxt->mprz)
+			return -ENOMEM;
+		return persistent_ram_write_user(cxt->mprz, buf, size);
+	}
+
+	return -EINVAL;
+}
+
 static int ramoops_pstore_erase(enum pstore_type_id type, u64 id, int count,
 				struct timespec time, struct pstore_info *psi)
 {
@@ -367,6 +387,7 @@ static struct ramoops_context oops_cxt = {
 		.open	= ramoops_pstore_open,
 		.read	= ramoops_pstore_read,
 		.write_buf	= ramoops_pstore_write_buf,
+		.write_buf_user	= ramoops_pstore_write_buf_user,
 		.erase	= ramoops_pstore_erase,
 	},
 };
@@ -467,6 +488,97 @@ static int ramoops_init_prz(struct device *dev, struct ramoops_context *cxt,
 	return 0;
 }
 
+void notrace ramoops_console_write_buf(const char *buf, size_t size)
+{
+	struct ramoops_context *cxt = &oops_cxt;
+	persistent_ram_write(cxt->cprz, buf, size);
+}
+
+static int ramoops_parse_dt_size(struct platform_device *pdev,
+		const char *propname, unsigned long *val)
+{
+	u64 val64;
+	int ret;
+
+	ret = of_property_read_u64(pdev->dev.of_node, propname, &val64);
+	if (ret == -EINVAL) {
+		*val = 0;
+		return 0;
+	} else if (ret != 0) {
+		dev_err(&pdev->dev, "failed to parse property %s: %d\n",
+				propname, ret);
+		return ret;
+	}
+
+	if (val64 > ULONG_MAX) {
+		dev_err(&pdev->dev, "invalid %s %llu\n", propname, val64);
+		return -EOVERFLOW;
+	}
+
+	*val = val64;
+	return 0;
+}
+
+static int ramoops_parse_dt(struct platform_device *pdev,
+		struct ramoops_platform_data *pdata)
+{
+	struct device_node *of_node = pdev->dev.of_node;
+	struct device_node *mem_region;
+	struct resource res;
+	u32 ecc_size;
+	int ret;
+
+	dev_dbg(&pdev->dev, "using Device Tree\n");
+
+	mem_region = of_parse_phandle(of_node, "memory-region", 0);
+	if (!mem_region) {
+		dev_err(&pdev->dev, "no memory-region phandle\n");
+		return -ENODEV;
+	}
+
+	ret = of_address_to_resource(mem_region, 0, &res);
+	of_node_put(mem_region);
+	if (ret) {
+		dev_err(&pdev->dev, "failed to translate memory-region to resource: %d\n",
+				ret);
+		return ret;
+	}
+
+	pdata->mem_size = resource_size(&res);
+	pdata->mem_address = res.start;
+	pdata->mem_type = of_property_read_bool(of_node, "unbuffered");
+	pdata->dump_oops = !of_property_read_bool(of_node, "no-dump-oops");
+
+	ret = ramoops_parse_dt_size(pdev, "record-size", &pdata->record_size);
+	if (ret < 0)
+		return ret;
+
+	ret = ramoops_parse_dt_size(pdev, "console-size", &pdata->console_size);
+	if (ret < 0)
+		return ret;
+
+	ret = ramoops_parse_dt_size(pdev, "ftrace-size", &pdata->ftrace_size);
+	if (ret < 0)
+		return ret;
+
+	ret = ramoops_parse_dt_size(pdev, "pmsg-size", &pdata->pmsg_size);
+	if (ret < 0)
+		return ret;
+
+	ret = of_property_read_u32(of_node, "ecc-size", &ecc_size);
+	if (ret == 0) {
+		if (ecc_size > INT_MAX) {
+			dev_err(&pdev->dev, "invalid ecc-size %u\n", ecc_size);
+			return -EOVERFLOW;
+		}
+		pdata->ecc_info.ecc_size = ecc_size;
+	} else if (ret != -EINVAL) {
+		return ret;
+	}
+
+	return 0;
+}
+
 static int ramoops_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
@@ -476,6 +588,18 @@ static int ramoops_probe(struct platform_device *pdev)
 	phys_addr_t paddr;
 	int err = -EINVAL;
 
+	if (dev->of_node && !pdata) {
+		pdata = devm_kzalloc(&pdev->dev, sizeof(*pdata), GFP_KERNEL);
+		if (!pdata) {
+			err = -ENOMEM;
+			goto fail_out;
+		}
+
+		err = ramoops_parse_dt(pdev, pdata);
+		if (err < 0)
+			goto fail_out;
+	}
+
 	/* Only a single ramoops area allowed at a time, so fail extra
 	 * probes.
 	 */
@@ -604,11 +728,17 @@ static int ramoops_remove(struct platform_device *pdev)
 	return 0;
 }
 
+static const struct of_device_id dt_match[] = {
+	{ .compatible = "ramoops" },
+	{}
+};
+
 static struct platform_driver ramoops_driver = {
 	.probe		= ramoops_probe,
 	.remove		= ramoops_remove,
 	.driver		= {
-		.name	= "ramoops",
+		.name		= "ramoops",
+		.of_match_table	= dt_match,
 	},
 };
 
diff --git a/fs/pstore/ram_core.c b/fs/pstore/ram_core.c
index 27300533c2dd..e11672aa4575 100644
--- a/fs/pstore/ram_core.c
+++ b/fs/pstore/ram_core.c
@@ -17,15 +17,16 @@
 #include <linux/device.h>
 #include <linux/err.h>
 #include <linux/errno.h>
-#include <linux/kernel.h>
 #include <linux/init.h>
 #include <linux/io.h>
+#include <linux/kernel.h>
 #include <linux/list.h>
 #include <linux/memblock.h>
+#include <linux/pstore_ram.h>
 #include <linux/rslib.h>
 #include <linux/slab.h>
+#include <linux/uaccess.h>
 #include <linux/vmalloc.h>
-#include <linux/pstore_ram.h>
 #include <asm/page.h>
 
 struct persistent_ram_buffer {
@@ -269,6 +270,16 @@ static void notrace persistent_ram_update(struct persistent_ram_zone *prz,
 	persistent_ram_update_ecc(prz, start, count);
 }
 
+static int notrace persistent_ram_update_user(struct persistent_ram_zone *prz,
+	const void __user *s, unsigned int start, unsigned int count)
+{
+	struct persistent_ram_buffer *buffer = prz->buffer;
+	int ret = unlikely(__copy_from_user(buffer->data + start, s, count)) ?
+		-EFAULT : 0;
+	persistent_ram_update_ecc(prz, start, count);
+	return ret;
+}
+
 void persistent_ram_save_old(struct persistent_ram_zone *prz)
 {
 	struct persistent_ram_buffer *buffer = prz->buffer;
@@ -322,6 +333,38 @@ int notrace persistent_ram_write(struct persistent_ram_zone *prz,
 	return count;
 }
 
+int notrace persistent_ram_write_user(struct persistent_ram_zone *prz,
+	const void __user *s, unsigned int count)
+{
+	int rem, ret = 0, c = count;
+	size_t start;
+
+	if (unlikely(!access_ok(VERIFY_READ, s, count)))
+		return -EFAULT;
+	if (unlikely(c > prz->buffer_size)) {
+		s += c - prz->buffer_size;
+		c = prz->buffer_size;
+	}
+
+	buffer_size_add(prz, c);
+
+	start = buffer_start_add(prz, c);
+
+	rem = prz->buffer_size - start;
+	if (unlikely(rem < c)) {
+		ret = persistent_ram_update_user(prz, s, start, rem);
+		s += rem;
+		c -= rem;
+		start = 0;
+	}
+	if (likely(!ret))
+		ret = persistent_ram_update_user(prz, s, start, c);
+
+	persistent_ram_update_header_ecc(prz);
+
+	return unlikely(ret) ? ret : count;
+}
+
 size_t persistent_ram_old_size(struct persistent_ram_zone *prz)
 {
 	return prz->old_log_size;
diff --git a/fs/sdcardfs/Kconfig b/fs/sdcardfs/Kconfig
new file mode 100644
index 000000000000..a1c103316ac7
--- /dev/null
+++ b/fs/sdcardfs/Kconfig
@@ -0,0 +1,13 @@
+config SDCARD_FS
+	tristate "sdcard file system"
+	depends on CONFIGFS_FS
+	default n
+	help
+	  Sdcardfs is based on Wrapfs file system.
+
+config SDCARD_FS_FADV_NOACTIVE
+	bool "sdcardfs fadvise noactive support"
+	depends on FADV_NOACTIVE
+	default y
+	help
+	  Sdcardfs supports fadvise noactive mode.
diff --git a/fs/sdcardfs/Makefile b/fs/sdcardfs/Makefile
new file mode 100644
index 000000000000..b84fbb2b45a4
--- /dev/null
+++ b/fs/sdcardfs/Makefile
@@ -0,0 +1,7 @@
+SDCARDFS_VERSION="0.1"
+
+EXTRA_CFLAGS += -DSDCARDFS_VERSION=\"$(SDCARDFS_VERSION)\"
+
+obj-$(CONFIG_SDCARD_FS) += sdcardfs.o
+
+sdcardfs-y := dentry.o file.o inode.o main.o super.o lookup.o mmap.o packagelist.o derived_perm.o
diff --git a/fs/sdcardfs/dentry.c b/fs/sdcardfs/dentry.c
new file mode 100644
index 000000000000..13da7e5245bd
--- /dev/null
+++ b/fs/sdcardfs/dentry.c
@@ -0,0 +1,194 @@
+/*
+ * fs/sdcardfs/dentry.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+#include "linux/ctype.h"
+
+/*
+ * returns: -ERRNO if error (returned to user)
+ *          0: tell VFS to invalidate dentry
+ *          1: dentry is valid
+ */
+static int sdcardfs_d_revalidate(struct dentry *dentry, unsigned int flags)
+{
+	int err = 1;
+	struct path parent_lower_path, lower_path;
+	struct dentry *parent_dentry = NULL;
+	struct dentry *parent_lower_dentry = NULL;
+	struct dentry *lower_cur_parent_dentry = NULL;
+	struct dentry *lower_dentry = NULL;
+	struct inode *inode;
+	struct sdcardfs_inode_data *data;
+
+	if (flags & LOOKUP_RCU)
+		return -ECHILD;
+
+	spin_lock(&dentry->d_lock);
+	if (IS_ROOT(dentry)) {
+		spin_unlock(&dentry->d_lock);
+		return 1;
+	}
+	spin_unlock(&dentry->d_lock);
+
+	/* check uninitialized obb_dentry and
+	 * whether the base obbpath has been changed or not
+	 */
+	if (is_obbpath_invalid(dentry)) {
+		d_drop(dentry);
+		return 0;
+	}
+
+	parent_dentry = dget_parent(dentry);
+	sdcardfs_get_lower_path(parent_dentry, &parent_lower_path);
+	sdcardfs_get_real_lower(dentry, &lower_path);
+	parent_lower_dentry = parent_lower_path.dentry;
+	lower_dentry = lower_path.dentry;
+	lower_cur_parent_dentry = dget_parent(lower_dentry);
+
+	if ((lower_dentry->d_flags & DCACHE_OP_REVALIDATE)) {
+		err = lower_dentry->d_op->d_revalidate(lower_dentry, flags);
+		if (err == 0) {
+			d_drop(dentry);
+			goto out;
+		}
+	}
+
+	spin_lock(&lower_dentry->d_lock);
+	if (d_unhashed(lower_dentry)) {
+		spin_unlock(&lower_dentry->d_lock);
+		d_drop(dentry);
+		err = 0;
+		goto out;
+	}
+	spin_unlock(&lower_dentry->d_lock);
+
+	if (parent_lower_dentry != lower_cur_parent_dentry) {
+		d_drop(dentry);
+		err = 0;
+		goto out;
+	}
+
+	if (dentry < lower_dentry) {
+		spin_lock(&dentry->d_lock);
+		spin_lock_nested(&lower_dentry->d_lock, DENTRY_D_LOCK_NESTED);
+	} else {
+		spin_lock(&lower_dentry->d_lock);
+		spin_lock_nested(&dentry->d_lock, DENTRY_D_LOCK_NESTED);
+	}
+
+	if (!qstr_case_eq(&dentry->d_name, &lower_dentry->d_name)) {
+		__d_drop(dentry);
+		err = 0;
+	}
+
+	if (dentry < lower_dentry) {
+		spin_unlock(&lower_dentry->d_lock);
+		spin_unlock(&dentry->d_lock);
+	} else {
+		spin_unlock(&dentry->d_lock);
+		spin_unlock(&lower_dentry->d_lock);
+	}
+	if (!err)
+		goto out;
+
+	/* If our top's inode is gone, we may be out of date */
+	inode = igrab(d_inode(dentry));
+	if (inode) {
+		data = top_data_get(SDCARDFS_I(inode));
+		if (!data || data->abandoned) {
+			d_drop(dentry);
+			err = 0;
+		}
+		if (data)
+			data_put(data);
+		iput(inode);
+	}
+
+out:
+	dput(parent_dentry);
+	dput(lower_cur_parent_dentry);
+	sdcardfs_put_lower_path(parent_dentry, &parent_lower_path);
+	sdcardfs_put_real_lower(dentry, &lower_path);
+	return err;
+}
+
+static void sdcardfs_d_release(struct dentry *dentry)
+{
+	/* release and reset the lower paths */
+	if (has_graft_path(dentry))
+		sdcardfs_put_reset_orig_path(dentry);
+	sdcardfs_put_reset_lower_path(dentry);
+	free_dentry_private_data(dentry);
+}
+
+static int sdcardfs_hash_ci(const struct dentry *dentry,
+				struct qstr *qstr)
+{
+	/*
+	 * This function is copy of vfat_hashi.
+	 * FIXME Should we support national language?
+	 *       Refer to vfat_hashi()
+	 * struct nls_table *t = MSDOS_SB(dentry->d_sb)->nls_io;
+	 */
+	const unsigned char *name;
+	unsigned int len;
+	unsigned long hash;
+
+	name = qstr->name;
+	len = qstr->len;
+
+	hash = init_name_hash();
+	while (len--)
+		hash = partial_name_hash(tolower(*name++), hash);
+	qstr->hash = end_name_hash(hash);
+
+	return 0;
+}
+
+/*
+ * Case insensitive compare of two vfat names.
+ */
+static int sdcardfs_cmp_ci(const struct dentry *parent,
+		const struct dentry *dentry,
+		unsigned int len, const char *str, const struct qstr *name)
+{
+	/* FIXME Should we support national language? */
+
+	if (name->len == len) {
+		if (str_n_case_eq(name->name, str, len))
+			return 0;
+	}
+	return 1;
+}
+
+static void sdcardfs_canonical_path(const struct path *path,
+				struct path *actual_path)
+{
+	sdcardfs_get_real_lower(path->dentry, actual_path);
+}
+
+const struct dentry_operations sdcardfs_ci_dops = {
+	.d_revalidate	= sdcardfs_d_revalidate,
+	.d_release	= sdcardfs_d_release,
+	.d_hash	= sdcardfs_hash_ci,
+	.d_compare	= sdcardfs_cmp_ci,
+	.d_canonical_path = sdcardfs_canonical_path,
+};
+
diff --git a/fs/sdcardfs/derived_perm.c b/fs/sdcardfs/derived_perm.c
new file mode 100644
index 000000000000..e0b74dc77c5f
--- /dev/null
+++ b/fs/sdcardfs/derived_perm.c
@@ -0,0 +1,474 @@
+/*
+ * fs/sdcardfs/derived_perm.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+
+/* copy derived state from parent inode */
+static void inherit_derived_state(struct inode *parent, struct inode *child)
+{
+	struct sdcardfs_inode_info *pi = SDCARDFS_I(parent);
+	struct sdcardfs_inode_info *ci = SDCARDFS_I(child);
+
+	ci->data->perm = PERM_INHERIT;
+	ci->data->userid = pi->data->userid;
+	ci->data->d_uid = pi->data->d_uid;
+	ci->data->under_android = pi->data->under_android;
+	ci->data->under_cache = pi->data->under_cache;
+	ci->data->under_obb = pi->data->under_obb;
+	set_top(ci, pi->top_data);
+}
+
+/* helper function for derived state */
+void setup_derived_state(struct inode *inode, perm_t perm, userid_t userid,
+					uid_t uid, bool under_android,
+					struct sdcardfs_inode_data *top)
+{
+	struct sdcardfs_inode_info *info = SDCARDFS_I(inode);
+
+	info->data->perm = perm;
+	info->data->userid = userid;
+	info->data->d_uid = uid;
+	info->data->under_android = under_android;
+	info->data->under_cache = false;
+	info->data->under_obb = false;
+	set_top(info, top);
+}
+
+/* While renaming, there is a point where we want the path from dentry,
+ * but the name from newdentry
+ */
+void get_derived_permission_new(struct dentry *parent, struct dentry *dentry,
+				const struct qstr *name)
+{
+	struct sdcardfs_inode_info *info = SDCARDFS_I(d_inode(dentry));
+	struct sdcardfs_inode_data *parent_data =
+			SDCARDFS_I(d_inode(parent))->data;
+	appid_t appid;
+	unsigned long user_num;
+	int err;
+	struct qstr q_Android = QSTR_LITERAL("Android");
+	struct qstr q_data = QSTR_LITERAL("data");
+	struct qstr q_obb = QSTR_LITERAL("obb");
+	struct qstr q_media = QSTR_LITERAL("media");
+	struct qstr q_cache = QSTR_LITERAL("cache");
+
+	/* By default, each inode inherits from its parent.
+	 * the properties are maintained on its private fields
+	 * because the inode attributes will be modified with that of
+	 * its lower inode.
+	 * These values are used by our custom permission call instead
+	 * of using the inode permissions.
+	 */
+
+	inherit_derived_state(d_inode(parent), d_inode(dentry));
+
+	/* Files don't get special labels */
+	if (!S_ISDIR(d_inode(dentry)->i_mode))
+		return;
+	/* Derive custom permissions based on parent and current node */
+	switch (parent_data->perm) {
+	case PERM_INHERIT:
+	case PERM_ANDROID_PACKAGE_CACHE:
+		/* Already inherited above */
+		break;
+	case PERM_PRE_ROOT:
+		/* Legacy internal layout places users at top level */
+		info->data->perm = PERM_ROOT;
+		err = kstrtoul(name->name, 10, &user_num);
+		if (err)
+			info->data->userid = 0;
+		else
+			info->data->userid = user_num;
+		set_top(info, info->data);
+		break;
+	case PERM_ROOT:
+		/* Assume masked off by default. */
+		if (qstr_case_eq(name, &q_Android)) {
+			/* App-specific directories inside; let anyone traverse */
+			info->data->perm = PERM_ANDROID;
+			info->data->under_android = true;
+			set_top(info, info->data);
+		}
+		break;
+	case PERM_ANDROID:
+		if (qstr_case_eq(name, &q_data)) {
+			/* App-specific directories inside; let anyone traverse */
+			info->data->perm = PERM_ANDROID_DATA;
+			set_top(info, info->data);
+		} else if (qstr_case_eq(name, &q_obb)) {
+			/* App-specific directories inside; let anyone traverse */
+			info->data->perm = PERM_ANDROID_OBB;
+			info->data->under_obb = true;
+			set_top(info, info->data);
+			/* Single OBB directory is always shared */
+		} else if (qstr_case_eq(name, &q_media)) {
+			/* App-specific directories inside; let anyone traverse */
+			info->data->perm = PERM_ANDROID_MEDIA;
+			set_top(info, info->data);
+		}
+		break;
+	case PERM_ANDROID_OBB:
+	case PERM_ANDROID_DATA:
+	case PERM_ANDROID_MEDIA:
+		info->data->perm = PERM_ANDROID_PACKAGE;
+		appid = get_appid(name->name);
+		if (appid != 0 && !is_excluded(name->name, parent_data->userid))
+			info->data->d_uid =
+				multiuser_get_uid(parent_data->userid, appid);
+		set_top(info, info->data);
+		break;
+	case PERM_ANDROID_PACKAGE:
+		if (qstr_case_eq(name, &q_cache)) {
+			info->data->perm = PERM_ANDROID_PACKAGE_CACHE;
+			info->data->under_cache = true;
+		}
+		break;
+	}
+}
+
+void get_derived_permission(struct dentry *parent, struct dentry *dentry)
+{
+	get_derived_permission_new(parent, dentry, &dentry->d_name);
+}
+
+static appid_t get_type(const char *name)
+{
+	const char *ext = strrchr(name, '.');
+	appid_t id;
+
+	if (ext && ext[0]) {
+		ext = &ext[1];
+		id = get_ext_gid(ext);
+		return id?:AID_MEDIA_RW;
+	}
+	return AID_MEDIA_RW;
+}
+
+void fixup_lower_ownership(struct dentry *dentry, const char *name)
+{
+	struct path path;
+	struct inode *inode;
+	struct inode *delegated_inode = NULL;
+	int error;
+	struct sdcardfs_inode_info *info;
+	struct sdcardfs_inode_data *info_d;
+	struct sdcardfs_inode_data *info_top;
+	perm_t perm;
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+	uid_t uid = sbi->options.fs_low_uid;
+	gid_t gid = sbi->options.fs_low_gid;
+	struct iattr newattrs;
+
+	if (!sbi->options.gid_derivation)
+		return;
+
+	info = SDCARDFS_I(d_inode(dentry));
+	info_d = info->data;
+	perm = info_d->perm;
+	if (info_d->under_obb) {
+		perm = PERM_ANDROID_OBB;
+	} else if (info_d->under_cache) {
+		perm = PERM_ANDROID_PACKAGE_CACHE;
+	} else if (perm == PERM_INHERIT) {
+		info_top = top_data_get(info);
+		perm = info_top->perm;
+		data_put(info_top);
+	}
+
+	switch (perm) {
+	case PERM_ROOT:
+	case PERM_ANDROID:
+	case PERM_ANDROID_DATA:
+	case PERM_ANDROID_MEDIA:
+	case PERM_ANDROID_PACKAGE:
+	case PERM_ANDROID_PACKAGE_CACHE:
+		uid = multiuser_get_uid(info_d->userid, uid);
+		break;
+	case PERM_ANDROID_OBB:
+		uid = AID_MEDIA_OBB;
+		break;
+	case PERM_PRE_ROOT:
+	default:
+		break;
+	}
+	switch (perm) {
+	case PERM_ROOT:
+	case PERM_ANDROID:
+	case PERM_ANDROID_DATA:
+	case PERM_ANDROID_MEDIA:
+		if (S_ISDIR(d_inode(dentry)->i_mode))
+			gid = multiuser_get_uid(info_d->userid, AID_MEDIA_RW);
+		else
+			gid = multiuser_get_uid(info_d->userid, get_type(name));
+		break;
+	case PERM_ANDROID_OBB:
+		gid = AID_MEDIA_OBB;
+		break;
+	case PERM_ANDROID_PACKAGE:
+		if (uid_is_app(info_d->d_uid))
+			gid = multiuser_get_ext_gid(info_d->d_uid);
+		else
+			gid = multiuser_get_uid(info_d->userid, AID_MEDIA_RW);
+		break;
+	case PERM_ANDROID_PACKAGE_CACHE:
+		if (uid_is_app(info_d->d_uid))
+			gid = multiuser_get_ext_cache_gid(info_d->d_uid);
+		else
+			gid = multiuser_get_uid(info_d->userid, AID_MEDIA_RW);
+		break;
+	case PERM_PRE_ROOT:
+	default:
+		break;
+	}
+
+	sdcardfs_get_lower_path(dentry, &path);
+	inode = d_inode(path.dentry);
+	if (d_inode(path.dentry)->i_gid.val != gid || d_inode(path.dentry)->i_uid.val != uid) {
+retry_deleg:
+		newattrs.ia_valid = ATTR_GID | ATTR_UID | ATTR_FORCE;
+		newattrs.ia_uid = make_kuid(current_user_ns(), uid);
+		newattrs.ia_gid = make_kgid(current_user_ns(), gid);
+		if (!S_ISDIR(inode->i_mode))
+			newattrs.ia_valid |=
+				ATTR_KILL_SUID | ATTR_KILL_SGID | ATTR_KILL_PRIV;
+		mutex_lock(&inode->i_mutex);
+		error = security_path_chown(&path, newattrs.ia_uid, newattrs.ia_gid);
+		if (!error)
+			error = notify_change2(path.mnt, path.dentry, &newattrs, &delegated_inode);
+		mutex_unlock(&inode->i_mutex);
+		if (delegated_inode) {
+			error = break_deleg_wait(&delegated_inode);
+			if (!error)
+				goto retry_deleg;
+		}
+		if (error)
+			pr_debug("sdcardfs: Failed to touch up lower fs gid/uid for %s\n", name);
+	}
+	sdcardfs_put_lower_path(dentry, &path);
+}
+
+static int descendant_may_need_fixup(struct sdcardfs_inode_data *data,
+		struct limit_search *limit)
+{
+	if (data->perm == PERM_ROOT)
+		return (limit->flags & BY_USERID) ?
+				data->userid == limit->userid : 1;
+	if (data->perm == PERM_PRE_ROOT || data->perm == PERM_ANDROID)
+		return 1;
+	return 0;
+}
+
+static int needs_fixup(perm_t perm)
+{
+	if (perm == PERM_ANDROID_DATA || perm == PERM_ANDROID_OBB
+			|| perm == PERM_ANDROID_MEDIA)
+		return 1;
+	return 0;
+}
+
+static void __fixup_perms_recursive(struct dentry *dentry, struct limit_search *limit, int depth)
+{
+	struct dentry *child;
+	struct sdcardfs_inode_info *info;
+
+	/*
+	 * All paths will terminate their recursion on hitting PERM_ANDROID_OBB,
+	 * PERM_ANDROID_MEDIA, or PERM_ANDROID_DATA. This happens at a depth of
+	 * at most 3.
+	 */
+	WARN(depth > 3, "%s: Max expected depth exceeded!\n", __func__);
+	spin_lock_nested(&dentry->d_lock, depth);
+	if (!d_inode(dentry)) {
+		spin_unlock(&dentry->d_lock);
+		return;
+	}
+	info = SDCARDFS_I(d_inode(dentry));
+
+	if (needs_fixup(info->data->perm)) {
+		list_for_each_entry(child, &dentry->d_subdirs, d_child) {
+			spin_lock_nested(&child->d_lock, depth + 1);
+			if (!(limit->flags & BY_NAME) || qstr_case_eq(&child->d_name, &limit->name)) {
+				if (d_inode(child)) {
+					get_derived_permission(dentry, child);
+					fixup_tmp_permissions(d_inode(child));
+					spin_unlock(&child->d_lock);
+					break;
+				}
+			}
+			spin_unlock(&child->d_lock);
+		}
+	} else if (descendant_may_need_fixup(info->data, limit)) {
+		list_for_each_entry(child, &dentry->d_subdirs, d_child) {
+			__fixup_perms_recursive(child, limit, depth + 1);
+		}
+	}
+	spin_unlock(&dentry->d_lock);
+}
+
+void fixup_perms_recursive(struct dentry *dentry, struct limit_search *limit)
+{
+	__fixup_perms_recursive(dentry, limit, 0);
+}
+
+/* main function for updating derived permission */
+inline void update_derived_permission_lock(struct dentry *dentry)
+{
+	struct dentry *parent;
+
+	if (!dentry || !d_inode(dentry)) {
+		pr_err("sdcardfs: %s: invalid dentry\n", __func__);
+		return;
+	}
+	/* FIXME:
+	 * 1. need to check whether the dentry is updated or not
+	 * 2. remove the root dentry update
+	 */
+	if (!IS_ROOT(dentry)) {
+		parent = dget_parent(dentry);
+		if (parent) {
+			get_derived_permission(parent, dentry);
+			dput(parent);
+		}
+	}
+	fixup_tmp_permissions(d_inode(dentry));
+}
+
+int need_graft_path(struct dentry *dentry)
+{
+	int ret = 0;
+	struct dentry *parent = dget_parent(dentry);
+	struct sdcardfs_inode_info *parent_info = SDCARDFS_I(d_inode(parent));
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+	struct qstr obb = QSTR_LITERAL("obb");
+
+	if (parent_info->data->perm == PERM_ANDROID &&
+			qstr_case_eq(&dentry->d_name, &obb)) {
+
+		/* /Android/obb is the base obbpath of DERIVED_UNIFIED */
+		if (!(sbi->options.multiuser == false
+				&& parent_info->data->userid == 0)) {
+			ret = 1;
+		}
+	}
+	dput(parent);
+	return ret;
+}
+
+int is_obbpath_invalid(struct dentry *dent)
+{
+	int ret = 0;
+	struct sdcardfs_dentry_info *di = SDCARDFS_D(dent);
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dent->d_sb);
+	char *path_buf, *obbpath_s;
+	int need_put = 0;
+	struct path lower_path;
+
+	/* check the base obbpath has been changed.
+	 * this routine can check an uninitialized obb dentry as well.
+	 * regarding the uninitialized obb, refer to the sdcardfs_mkdir()
+	 */
+	spin_lock(&di->lock);
+	if (di->orig_path.dentry) {
+		if (!di->lower_path.dentry) {
+			ret = 1;
+		} else {
+			path_get(&di->lower_path);
+
+			path_buf = kmalloc(PATH_MAX, GFP_ATOMIC);
+			if (!path_buf) {
+				ret = 1;
+				pr_err("sdcardfs: fail to allocate path_buf in %s.\n", __func__);
+			} else {
+				obbpath_s = d_path(&di->lower_path, path_buf, PATH_MAX);
+				if (d_unhashed(di->lower_path.dentry) ||
+					!str_case_eq(sbi->obbpath_s, obbpath_s)) {
+					ret = 1;
+				}
+				kfree(path_buf);
+			}
+
+			pathcpy(&lower_path, &di->lower_path);
+			need_put = 1;
+		}
+	}
+	spin_unlock(&di->lock);
+	if (need_put)
+		path_put(&lower_path);
+	return ret;
+}
+
+int is_base_obbpath(struct dentry *dentry)
+{
+	int ret = 0;
+	struct dentry *parent = dget_parent(dentry);
+	struct sdcardfs_inode_info *parent_info = SDCARDFS_I(d_inode(parent));
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+	struct qstr q_obb = QSTR_LITERAL("obb");
+
+	spin_lock(&SDCARDFS_D(dentry)->lock);
+	if (sbi->options.multiuser) {
+		if (parent_info->data->perm == PERM_PRE_ROOT &&
+				qstr_case_eq(&dentry->d_name, &q_obb)) {
+			ret = 1;
+		}
+	} else  if (parent_info->data->perm == PERM_ANDROID &&
+			qstr_case_eq(&dentry->d_name, &q_obb)) {
+		ret = 1;
+	}
+	spin_unlock(&SDCARDFS_D(dentry)->lock);
+	return ret;
+}
+
+/* The lower_path will be stored to the dentry's orig_path
+ * and the base obbpath will be copyed to the lower_path variable.
+ * if an error returned, there's no change in the lower_path
+ * returns: -ERRNO if error (0: no error)
+ */
+int setup_obb_dentry(struct dentry *dentry, struct path *lower_path)
+{
+	int err = 0;
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+	struct path obbpath;
+
+	/* A local obb dentry must have its own orig_path to support rmdir
+	 * and mkdir of itself. Usually, we expect that the sbi->obbpath
+	 * is avaiable on this stage.
+	 */
+	sdcardfs_set_orig_path(dentry, lower_path);
+
+	err = kern_path(sbi->obbpath_s,
+			LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &obbpath);
+
+	if (!err) {
+		/* the obbpath base has been found */
+		pathcpy(lower_path, &obbpath);
+	} else {
+		/* if the sbi->obbpath is not available, we can optionally
+		 * setup the lower_path with its orig_path.
+		 * but, the current implementation just returns an error
+		 * because the sdcard daemon also regards this case as
+		 * a lookup fail.
+		 */
+		pr_info("sdcardfs: the sbi->obbpath is not available\n");
+	}
+	return err;
+}
+
+
diff --git a/fs/sdcardfs/file.c b/fs/sdcardfs/file.c
new file mode 100644
index 000000000000..5ac0b0bbb0ec
--- /dev/null
+++ b/fs/sdcardfs/file.c
@@ -0,0 +1,449 @@
+/*
+ * fs/sdcardfs/file.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+#ifdef CONFIG_SDCARD_FS_FADV_NOACTIVE
+#include <linux/backing-dev.h>
+#endif
+
+static ssize_t sdcardfs_read(struct file *file, char __user *buf,
+			   size_t count, loff_t *ppos)
+{
+	int err;
+	struct file *lower_file;
+	struct dentry *dentry = file->f_path.dentry;
+#ifdef CONFIG_SDCARD_FS_FADV_NOACTIVE
+	struct backing_dev_info *bdi;
+#endif
+
+	lower_file = sdcardfs_lower_file(file);
+
+#ifdef CONFIG_SDCARD_FS_FADV_NOACTIVE
+	if (file->f_mode & FMODE_NOACTIVE) {
+		if (!(lower_file->f_mode & FMODE_NOACTIVE)) {
+			bdi = lower_file->f_mapping->backing_dev_info;
+			lower_file->f_ra.ra_pages = bdi->ra_pages * 2;
+			spin_lock(&lower_file->f_lock);
+			lower_file->f_mode |= FMODE_NOACTIVE;
+			spin_unlock(&lower_file->f_lock);
+		}
+	}
+#endif
+
+	err = vfs_read(lower_file, buf, count, ppos);
+	/* update our inode atime upon a successful lower read */
+	if (err >= 0)
+		fsstack_copy_attr_atime(d_inode(dentry),
+					file_inode(lower_file));
+
+	return err;
+}
+
+static ssize_t sdcardfs_write(struct file *file, const char __user *buf,
+			    size_t count, loff_t *ppos)
+{
+	int err;
+	struct file *lower_file;
+	struct dentry *dentry = file->f_path.dentry;
+
+	/* check disk space */
+	if (!check_min_free_space(dentry, count, 0)) {
+		pr_err("No minimum free space.\n");
+		return -ENOSPC;
+	}
+
+	lower_file = sdcardfs_lower_file(file);
+	err = vfs_write(lower_file, buf, count, ppos);
+	/* update our inode times+sizes upon a successful lower write */
+	if (err >= 0) {
+		fsstack_copy_inode_size(d_inode(dentry),
+					file_inode(lower_file));
+		fsstack_copy_attr_times(d_inode(dentry),
+					file_inode(lower_file));
+	}
+
+	return err;
+}
+
+static int sdcardfs_readdir(struct file *file, struct dir_context *ctx)
+{
+	int err;
+	struct file *lower_file = NULL;
+	struct dentry *dentry = file->f_path.dentry;
+
+	lower_file = sdcardfs_lower_file(file);
+
+	lower_file->f_pos = file->f_pos;
+	err = iterate_dir(lower_file, ctx);
+	file->f_pos = lower_file->f_pos;
+	if (err >= 0)		/* copy the atime */
+		fsstack_copy_attr_atime(d_inode(dentry),
+					file_inode(lower_file));
+	return err;
+}
+
+static long sdcardfs_unlocked_ioctl(struct file *file, unsigned int cmd,
+				  unsigned long arg)
+{
+	long err = -ENOTTY;
+	struct file *lower_file;
+	const struct cred *saved_cred = NULL;
+	struct dentry *dentry = file->f_path.dentry;
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+
+	lower_file = sdcardfs_lower_file(file);
+
+	/* XXX: use vfs_ioctl if/when VFS exports it */
+	if (!lower_file || !lower_file->f_op)
+		goto out;
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(sbi, saved_cred, SDCARDFS_I(file_inode(file)));
+
+	if (lower_file->f_op->unlocked_ioctl)
+		err = lower_file->f_op->unlocked_ioctl(lower_file, cmd, arg);
+
+	/* some ioctls can change inode attributes (EXT2_IOC_SETFLAGS) */
+	if (!err)
+		sdcardfs_copy_and_fix_attrs(file_inode(file),
+				      file_inode(lower_file));
+	REVERT_CRED(saved_cred);
+out:
+	return err;
+}
+
+#ifdef CONFIG_COMPAT
+static long sdcardfs_compat_ioctl(struct file *file, unsigned int cmd,
+				unsigned long arg)
+{
+	long err = -ENOTTY;
+	struct file *lower_file;
+	const struct cred *saved_cred = NULL;
+	struct dentry *dentry = file->f_path.dentry;
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+
+	lower_file = sdcardfs_lower_file(file);
+
+	/* XXX: use vfs_ioctl if/when VFS exports it */
+	if (!lower_file || !lower_file->f_op)
+		goto out;
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(sbi, saved_cred, SDCARDFS_I(file_inode(file)));
+
+	if (lower_file->f_op->compat_ioctl)
+		err = lower_file->f_op->compat_ioctl(lower_file, cmd, arg);
+
+	REVERT_CRED(saved_cred);
+out:
+	return err;
+}
+#endif
+
+static int sdcardfs_mmap(struct file *file, struct vm_area_struct *vma)
+{
+	int err = 0;
+	bool willwrite;
+	struct file *lower_file;
+	const struct vm_operations_struct *saved_vm_ops = NULL;
+
+	/* this might be deferred to mmap's writepage */
+	willwrite = ((vma->vm_flags | VM_SHARED | VM_WRITE) == vma->vm_flags);
+
+	/*
+	 * File systems which do not implement ->writepage may use
+	 * generic_file_readonly_mmap as their ->mmap op.  If you call
+	 * generic_file_readonly_mmap with VM_WRITE, you'd get an -EINVAL.
+	 * But we cannot call the lower ->mmap op, so we can't tell that
+	 * writeable mappings won't work.  Therefore, our only choice is to
+	 * check if the lower file system supports the ->writepage, and if
+	 * not, return EINVAL (the same error that
+	 * generic_file_readonly_mmap returns in that case).
+	 */
+	lower_file = sdcardfs_lower_file(file);
+	if (willwrite && !lower_file->f_mapping->a_ops->writepage) {
+		err = -EINVAL;
+		pr_err("sdcardfs: lower file system does not support writeable mmap\n");
+		goto out;
+	}
+
+	/*
+	 * find and save lower vm_ops.
+	 *
+	 * XXX: the VFS should have a cleaner way of finding the lower vm_ops
+	 */
+	if (!SDCARDFS_F(file)->lower_vm_ops) {
+		err = lower_file->f_op->mmap(lower_file, vma);
+		if (err) {
+			pr_err("sdcardfs: lower mmap failed %d\n", err);
+			goto out;
+		}
+		saved_vm_ops = vma->vm_ops; /* save: came from lower ->mmap */
+	}
+
+	/*
+	 * Next 3 lines are all I need from generic_file_mmap.  I definitely
+	 * don't want its test for ->readpage which returns -ENOEXEC.
+	 */
+	file_accessed(file);
+	vma->vm_ops = &sdcardfs_vm_ops;
+
+	file->f_mapping->a_ops = &sdcardfs_aops; /* set our aops */
+	if (!SDCARDFS_F(file)->lower_vm_ops) /* save for our ->fault */
+		SDCARDFS_F(file)->lower_vm_ops = saved_vm_ops;
+	vma->vm_private_data = file;
+	get_file(lower_file);
+	vma->vm_file = lower_file;
+
+out:
+	return err;
+}
+
+static int sdcardfs_open(struct inode *inode, struct file *file)
+{
+	int err = 0;
+	struct file *lower_file = NULL;
+	struct path lower_path;
+	struct dentry *dentry = file->f_path.dentry;
+	struct dentry *parent = dget_parent(dentry);
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+	const struct cred *saved_cred = NULL;
+
+	/* don't open unhashed/deleted files */
+	if (d_unhashed(dentry)) {
+		err = -ENOENT;
+		goto out_err;
+	}
+
+	if (!check_caller_access_to_name(d_inode(parent), &dentry->d_name)) {
+		err = -EACCES;
+		goto out_err;
+	}
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(sbi, saved_cred, SDCARDFS_I(inode));
+
+	file->private_data =
+		kzalloc(sizeof(struct sdcardfs_file_info), GFP_KERNEL);
+	if (!SDCARDFS_F(file)) {
+		err = -ENOMEM;
+		goto out_revert_cred;
+	}
+
+	/* open lower object and link sdcardfs's file struct to lower's */
+	sdcardfs_get_lower_path(file->f_path.dentry, &lower_path);
+	lower_file = dentry_open(&lower_path, file->f_flags, current_cred());
+	path_put(&lower_path);
+	if (IS_ERR(lower_file)) {
+		err = PTR_ERR(lower_file);
+		lower_file = sdcardfs_lower_file(file);
+		if (lower_file) {
+			sdcardfs_set_lower_file(file, NULL);
+			fput(lower_file); /* fput calls dput for lower_dentry */
+		}
+	} else {
+		sdcardfs_set_lower_file(file, lower_file);
+	}
+
+	if (err)
+		kfree(SDCARDFS_F(file));
+	else
+		sdcardfs_copy_and_fix_attrs(inode, sdcardfs_lower_inode(inode));
+
+out_revert_cred:
+	REVERT_CRED(saved_cred);
+out_err:
+	dput(parent);
+	return err;
+}
+
+static int sdcardfs_flush(struct file *file, fl_owner_t id)
+{
+	int err = 0;
+	struct file *lower_file = NULL;
+
+	lower_file = sdcardfs_lower_file(file);
+	if (lower_file && lower_file->f_op && lower_file->f_op->flush) {
+		filemap_write_and_wait(file->f_mapping);
+		err = lower_file->f_op->flush(lower_file, id);
+	}
+
+	return err;
+}
+
+/* release all lower object references & free the file info structure */
+static int sdcardfs_file_release(struct inode *inode, struct file *file)
+{
+	struct file *lower_file;
+
+	lower_file = sdcardfs_lower_file(file);
+	if (lower_file) {
+		sdcardfs_set_lower_file(file, NULL);
+		fput(lower_file);
+	}
+
+	kfree(SDCARDFS_F(file));
+	return 0;
+}
+
+static int sdcardfs_fsync(struct file *file, loff_t start, loff_t end,
+			int datasync)
+{
+	int err;
+	struct file *lower_file;
+	struct path lower_path;
+	struct dentry *dentry = file->f_path.dentry;
+
+	err = __generic_file_fsync(file, start, end, datasync);
+	if (err)
+		goto out;
+
+	lower_file = sdcardfs_lower_file(file);
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	err = vfs_fsync_range(lower_file, start, end, datasync);
+	sdcardfs_put_lower_path(dentry, &lower_path);
+out:
+	return err;
+}
+
+static int sdcardfs_fasync(int fd, struct file *file, int flag)
+{
+	int err = 0;
+	struct file *lower_file = NULL;
+
+	lower_file = sdcardfs_lower_file(file);
+	if (lower_file->f_op && lower_file->f_op->fasync)
+		err = lower_file->f_op->fasync(fd, lower_file, flag);
+
+	return err;
+}
+
+/*
+ * Sdcardfs cannot use generic_file_llseek as ->llseek, because it would
+ * only set the offset of the upper file.  So we have to implement our
+ * own method to set both the upper and lower file offsets
+ * consistently.
+ */
+static loff_t sdcardfs_file_llseek(struct file *file, loff_t offset, int whence)
+{
+	int err;
+	struct file *lower_file;
+
+	err = generic_file_llseek(file, offset, whence);
+	if (err < 0)
+		goto out;
+
+	lower_file = sdcardfs_lower_file(file);
+	err = generic_file_llseek(lower_file, offset, whence);
+
+out:
+	return err;
+}
+
+/*
+ * Sdcardfs read_iter, redirect modified iocb to lower read_iter
+ */
+ssize_t sdcardfs_read_iter(struct kiocb *iocb, struct iov_iter *iter)
+{
+	int err;
+	struct file *file = iocb->ki_filp, *lower_file;
+
+	lower_file = sdcardfs_lower_file(file);
+	if (!lower_file->f_op->read_iter) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	get_file(lower_file); /* prevent lower_file from being released */
+	iocb->ki_filp = lower_file;
+	err = lower_file->f_op->read_iter(iocb, iter);
+	iocb->ki_filp = file;
+	fput(lower_file);
+	/* update upper inode atime as needed */
+	if (err >= 0 || err == -EIOCBQUEUED)
+		fsstack_copy_attr_atime(file->f_path.dentry->d_inode,
+					file_inode(lower_file));
+out:
+	return err;
+}
+
+/*
+ * Sdcardfs write_iter, redirect modified iocb to lower write_iter
+ */
+ssize_t sdcardfs_write_iter(struct kiocb *iocb, struct iov_iter *iter)
+{
+	int err;
+	struct file *file = iocb->ki_filp, *lower_file;
+
+	lower_file = sdcardfs_lower_file(file);
+	if (!lower_file->f_op->write_iter) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	get_file(lower_file); /* prevent lower_file from being released */
+	iocb->ki_filp = lower_file;
+	err = lower_file->f_op->write_iter(iocb, iter);
+	iocb->ki_filp = file;
+	fput(lower_file);
+	/* update upper inode times/sizes as needed */
+	if (err >= 0 || err == -EIOCBQUEUED) {
+		fsstack_copy_inode_size(file->f_path.dentry->d_inode,
+					file_inode(lower_file));
+		fsstack_copy_attr_times(file->f_path.dentry->d_inode,
+					file_inode(lower_file));
+	}
+out:
+	return err;
+}
+
+const struct file_operations sdcardfs_main_fops = {
+	.llseek		= generic_file_llseek,
+	.read		= sdcardfs_read,
+	.write		= sdcardfs_write,
+	.unlocked_ioctl	= sdcardfs_unlocked_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= sdcardfs_compat_ioctl,
+#endif
+	.mmap		= sdcardfs_mmap,
+	.open		= sdcardfs_open,
+	.flush		= sdcardfs_flush,
+	.release	= sdcardfs_file_release,
+	.fsync		= sdcardfs_fsync,
+	.fasync		= sdcardfs_fasync,
+	.read_iter	= sdcardfs_read_iter,
+	.write_iter	= sdcardfs_write_iter,
+};
+
+/* trimmed directory options */
+const struct file_operations sdcardfs_dir_fops = {
+	.llseek		= sdcardfs_file_llseek,
+	.read		= generic_read_dir,
+	.iterate	= sdcardfs_readdir,
+	.unlocked_ioctl	= sdcardfs_unlocked_ioctl,
+#ifdef CONFIG_COMPAT
+	.compat_ioctl	= sdcardfs_compat_ioctl,
+#endif
+	.open		= sdcardfs_open,
+	.release	= sdcardfs_file_release,
+	.flush		= sdcardfs_flush,
+	.fsync		= sdcardfs_fsync,
+	.fasync		= sdcardfs_fasync,
+};
diff --git a/fs/sdcardfs/inode.c b/fs/sdcardfs/inode.c
new file mode 100644
index 000000000000..85eb8ebb5372
--- /dev/null
+++ b/fs/sdcardfs/inode.c
@@ -0,0 +1,910 @@
+/*
+ * fs/sdcardfs/inode.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+#include <linux/fs_struct.h>
+#include <linux/ratelimit.h>
+
+/* Do not directly use this function. Use OVERRIDE_CRED() instead. */
+const struct cred *override_fsids(struct sdcardfs_sb_info *sbi,
+		struct sdcardfs_inode_data *data)
+{
+	struct cred *cred;
+	const struct cred *old_cred;
+	uid_t uid;
+
+	cred = prepare_creds();
+	if (!cred)
+		return NULL;
+
+	if (sbi->options.gid_derivation) {
+		if (data->under_obb)
+			uid = AID_MEDIA_OBB;
+		else
+			uid = multiuser_get_uid(data->userid, sbi->options.fs_low_uid);
+	} else {
+		uid = sbi->options.fs_low_uid;
+	}
+	cred->fsuid = make_kuid(&init_user_ns, uid);
+	cred->fsgid = make_kgid(&init_user_ns, sbi->options.fs_low_gid);
+
+	old_cred = override_creds(cred);
+
+	return old_cred;
+}
+
+/* Do not directly use this function, use REVERT_CRED() instead. */
+void revert_fsids(const struct cred *old_cred)
+{
+	const struct cred *cur_cred;
+
+	cur_cred = current->cred;
+	revert_creds(old_cred);
+	put_cred(cur_cred);
+}
+
+static int sdcardfs_create(struct inode *dir, struct dentry *dentry,
+			 umode_t mode, bool want_excl)
+{
+	int err;
+	struct dentry *lower_dentry;
+	struct vfsmount *lower_dentry_mnt;
+	struct dentry *lower_parent_dentry = NULL;
+	struct path lower_path;
+	const struct cred *saved_cred = NULL;
+	struct fs_struct *saved_fs;
+	struct fs_struct *copied_fs;
+
+	if (!check_caller_access_to_name(dir, &dentry->d_name)) {
+		err = -EACCES;
+		goto out_eacces;
+	}
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb), saved_cred, SDCARDFS_I(dir));
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	lower_dentry = lower_path.dentry;
+	lower_dentry_mnt = lower_path.mnt;
+	lower_parent_dentry = lock_parent(lower_dentry);
+
+	/* set last 16bytes of mode field to 0664 */
+	mode = (mode & S_IFMT) | 00664;
+
+	/* temporarily change umask for lower fs write */
+	saved_fs = current->fs;
+	copied_fs = copy_fs_struct(current->fs);
+	if (!copied_fs) {
+		err = -ENOMEM;
+		goto out_unlock;
+	}
+	current->fs = copied_fs;
+	current->fs->umask = 0;
+	err = vfs_create2(lower_dentry_mnt, d_inode(lower_parent_dentry), lower_dentry, mode, want_excl);
+	if (err)
+		goto out;
+
+	err = sdcardfs_interpose(dentry, dir->i_sb, &lower_path,
+			SDCARDFS_I(dir)->data->userid);
+	if (err)
+		goto out;
+	fsstack_copy_attr_times(dir, sdcardfs_lower_inode(dir));
+	fsstack_copy_inode_size(dir, d_inode(lower_parent_dentry));
+	fixup_lower_ownership(dentry, dentry->d_name.name);
+
+out:
+	current->fs = saved_fs;
+	free_fs_struct(copied_fs);
+out_unlock:
+	unlock_dir(lower_parent_dentry);
+	sdcardfs_put_lower_path(dentry, &lower_path);
+	REVERT_CRED(saved_cred);
+out_eacces:
+	return err;
+}
+
+#if 0
+static int sdcardfs_link(struct dentry *old_dentry, struct inode *dir,
+		       struct dentry *new_dentry)
+{
+	struct dentry *lower_old_dentry;
+	struct dentry *lower_new_dentry;
+	struct dentry *lower_dir_dentry;
+	u64 file_size_save;
+	int err;
+	struct path lower_old_path, lower_new_path;
+
+	OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb));
+
+	file_size_save = i_size_read(d_inode(old_dentry));
+	sdcardfs_get_lower_path(old_dentry, &lower_old_path);
+	sdcardfs_get_lower_path(new_dentry, &lower_new_path);
+	lower_old_dentry = lower_old_path.dentry;
+	lower_new_dentry = lower_new_path.dentry;
+	lower_dir_dentry = lock_parent(lower_new_dentry);
+
+	err = vfs_link(lower_old_dentry, d_inode(lower_dir_dentry),
+		       lower_new_dentry, NULL);
+	if (err || !d_inode(lower_new_dentry))
+		goto out;
+
+	err = sdcardfs_interpose(new_dentry, dir->i_sb, &lower_new_path);
+	if (err)
+		goto out;
+	fsstack_copy_attr_times(dir, d_inode(lower_new_dentry));
+	fsstack_copy_inode_size(dir, d_inode(lower_new_dentry));
+	set_nlink(d_inode(old_dentry),
+		  sdcardfs_lower_inode(d_inode(old_dentry))->i_nlink);
+	i_size_write(d_inode(new_dentry), file_size_save);
+out:
+	unlock_dir(lower_dir_dentry);
+	sdcardfs_put_lower_path(old_dentry, &lower_old_path);
+	sdcardfs_put_lower_path(new_dentry, &lower_new_path);
+	REVERT_CRED();
+	return err;
+}
+#endif
+
+static int sdcardfs_unlink(struct inode *dir, struct dentry *dentry)
+{
+	int err;
+	struct dentry *lower_dentry;
+	struct vfsmount *lower_mnt;
+	struct inode *lower_dir_inode = sdcardfs_lower_inode(dir);
+	struct dentry *lower_dir_dentry;
+	struct path lower_path;
+	const struct cred *saved_cred = NULL;
+
+	if (!check_caller_access_to_name(dir, &dentry->d_name)) {
+		err = -EACCES;
+		goto out_eacces;
+	}
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb), saved_cred, SDCARDFS_I(dir));
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	lower_dentry = lower_path.dentry;
+	lower_mnt = lower_path.mnt;
+	dget(lower_dentry);
+	lower_dir_dentry = lock_parent(lower_dentry);
+
+	err = vfs_unlink2(lower_mnt, lower_dir_inode, lower_dentry, NULL);
+
+	/*
+	 * Note: unlinking on top of NFS can cause silly-renamed files.
+	 * Trying to delete such files results in EBUSY from NFS
+	 * below.  Silly-renamed files will get deleted by NFS later on, so
+	 * we just need to detect them here and treat such EBUSY errors as
+	 * if the upper file was successfully deleted.
+	 */
+	if (err == -EBUSY && lower_dentry->d_flags & DCACHE_NFSFS_RENAMED)
+		err = 0;
+	if (err)
+		goto out;
+	fsstack_copy_attr_times(dir, lower_dir_inode);
+	fsstack_copy_inode_size(dir, lower_dir_inode);
+	set_nlink(d_inode(dentry),
+		  sdcardfs_lower_inode(d_inode(dentry))->i_nlink);
+	d_inode(dentry)->i_ctime = dir->i_ctime;
+	d_drop(dentry); /* this is needed, else LTP fails (VFS won't do it) */
+out:
+	unlock_dir(lower_dir_dentry);
+	dput(lower_dentry);
+	sdcardfs_put_lower_path(dentry, &lower_path);
+	REVERT_CRED(saved_cred);
+out_eacces:
+	return err;
+}
+
+#if 0
+static int sdcardfs_symlink(struct inode *dir, struct dentry *dentry,
+			  const char *symname)
+{
+	int err;
+	struct dentry *lower_dentry;
+	struct dentry *lower_parent_dentry = NULL;
+	struct path lower_path;
+
+	OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb));
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	lower_dentry = lower_path.dentry;
+	lower_parent_dentry = lock_parent(lower_dentry);
+
+	err = vfs_symlink(d_inode(lower_parent_dentry), lower_dentry, symname);
+	if (err)
+		goto out;
+	err = sdcardfs_interpose(dentry, dir->i_sb, &lower_path);
+	if (err)
+		goto out;
+	fsstack_copy_attr_times(dir, sdcardfs_lower_inode(dir));
+	fsstack_copy_inode_size(dir, d_inode(lower_parent_dentry));
+
+out:
+	unlock_dir(lower_parent_dentry);
+	sdcardfs_put_lower_path(dentry, &lower_path);
+	REVERT_CRED();
+	return err;
+}
+#endif
+
+static int touch(char *abs_path, mode_t mode)
+{
+	struct file *filp = filp_open(abs_path, O_RDWR|O_CREAT|O_EXCL|O_NOFOLLOW, mode);
+
+	if (IS_ERR(filp)) {
+		if (PTR_ERR(filp) == -EEXIST) {
+			return 0;
+		} else {
+			pr_err("sdcardfs: failed to open(%s): %ld\n",
+						abs_path, PTR_ERR(filp));
+			return PTR_ERR(filp);
+		}
+	}
+	filp_close(filp, current->files);
+	return 0;
+}
+
+static int sdcardfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
+{
+	int err;
+	int make_nomedia_in_obb = 0;
+	struct dentry *lower_dentry;
+	struct vfsmount *lower_mnt;
+	struct dentry *lower_parent_dentry = NULL;
+	struct path lower_path;
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+	const struct cred *saved_cred = NULL;
+	struct sdcardfs_inode_data *pd = SDCARDFS_I(dir)->data;
+	int touch_err = 0;
+	struct fs_struct *saved_fs;
+	struct fs_struct *copied_fs;
+	struct qstr q_obb = QSTR_LITERAL("obb");
+	struct qstr q_data = QSTR_LITERAL("data");
+
+	if (!check_caller_access_to_name(dir, &dentry->d_name)) {
+		err = -EACCES;
+		goto out_eacces;
+	}
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb), saved_cred, SDCARDFS_I(dir));
+
+	/* check disk space */
+	if (!check_min_free_space(dentry, 0, 1)) {
+		pr_err("sdcardfs: No minimum free space.\n");
+		err = -ENOSPC;
+		goto out_revert;
+	}
+
+	/* the lower_dentry is negative here */
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	lower_dentry = lower_path.dentry;
+	lower_mnt = lower_path.mnt;
+	lower_parent_dentry = lock_parent(lower_dentry);
+
+	/* set last 16bytes of mode field to 0775 */
+	mode = (mode & S_IFMT) | 00775;
+
+	/* temporarily change umask for lower fs write */
+	saved_fs = current->fs;
+	copied_fs = copy_fs_struct(current->fs);
+	if (!copied_fs) {
+		err = -ENOMEM;
+		unlock_dir(lower_parent_dentry);
+		goto out_unlock;
+	}
+	current->fs = copied_fs;
+	current->fs->umask = 0;
+	err = vfs_mkdir2(lower_mnt, d_inode(lower_parent_dentry), lower_dentry, mode);
+
+	if (err) {
+		unlock_dir(lower_parent_dentry);
+		goto out;
+	}
+
+	/* if it is a local obb dentry, setup it with the base obbpath */
+	if (need_graft_path(dentry)) {
+
+		err = setup_obb_dentry(dentry, &lower_path);
+		if (err) {
+			/* if the sbi->obbpath is not available, the lower_path won't be
+			 * changed by setup_obb_dentry() but the lower path is saved to
+			 * its orig_path. this dentry will be revalidated later.
+			 * but now, the lower_path should be NULL
+			 */
+			sdcardfs_put_reset_lower_path(dentry);
+
+			/* the newly created lower path which saved to its orig_path or
+			 * the lower_path is the base obbpath.
+			 * therefore, an additional path_get is required
+			 */
+			path_get(&lower_path);
+		} else
+			make_nomedia_in_obb = 1;
+	}
+
+	err = sdcardfs_interpose(dentry, dir->i_sb, &lower_path, pd->userid);
+	if (err) {
+		unlock_dir(lower_parent_dentry);
+		goto out;
+	}
+
+	fsstack_copy_attr_times(dir, sdcardfs_lower_inode(dir));
+	fsstack_copy_inode_size(dir, d_inode(lower_parent_dentry));
+	/* update number of links on parent directory */
+	set_nlink(dir, sdcardfs_lower_inode(dir)->i_nlink);
+	fixup_lower_ownership(dentry, dentry->d_name.name);
+	unlock_dir(lower_parent_dentry);
+	if ((!sbi->options.multiuser) && (qstr_case_eq(&dentry->d_name, &q_obb))
+		&& (pd->perm == PERM_ANDROID) && (pd->userid == 0))
+		make_nomedia_in_obb = 1;
+
+	/* When creating /Android/data and /Android/obb, mark them as .nomedia */
+	if (make_nomedia_in_obb ||
+		((pd->perm == PERM_ANDROID)
+				&& (qstr_case_eq(&dentry->d_name, &q_data)))) {
+		REVERT_CRED(saved_cred);
+		OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb), saved_cred, SDCARDFS_I(d_inode(dentry)));
+		set_fs_pwd(current->fs, &lower_path);
+		touch_err = touch(".nomedia", 0664);
+		if (touch_err) {
+			pr_err("sdcardfs: failed to create .nomedia in %s: %d\n",
+							lower_path.dentry->d_name.name, touch_err);
+			goto out;
+		}
+	}
+out:
+	current->fs = saved_fs;
+	free_fs_struct(copied_fs);
+out_unlock:
+	sdcardfs_put_lower_path(dentry, &lower_path);
+out_revert:
+	REVERT_CRED(saved_cred);
+out_eacces:
+	return err;
+}
+
+static int sdcardfs_rmdir(struct inode *dir, struct dentry *dentry)
+{
+	struct dentry *lower_dentry;
+	struct dentry *lower_dir_dentry;
+	struct vfsmount *lower_mnt;
+	int err;
+	struct path lower_path;
+	const struct cred *saved_cred = NULL;
+
+	if (!check_caller_access_to_name(dir, &dentry->d_name)) {
+		err = -EACCES;
+		goto out_eacces;
+	}
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb), saved_cred, SDCARDFS_I(dir));
+
+	/* sdcardfs_get_real_lower(): in case of remove an user's obb dentry
+	 * the dentry on the original path should be deleted.
+	 */
+	sdcardfs_get_real_lower(dentry, &lower_path);
+
+	lower_dentry = lower_path.dentry;
+	lower_mnt = lower_path.mnt;
+	lower_dir_dentry = lock_parent(lower_dentry);
+
+	err = vfs_rmdir2(lower_mnt, d_inode(lower_dir_dentry), lower_dentry);
+	if (err)
+		goto out;
+
+	d_drop(dentry);	/* drop our dentry on success (why not VFS's job?) */
+	if (d_inode(dentry))
+		clear_nlink(d_inode(dentry));
+	fsstack_copy_attr_times(dir, d_inode(lower_dir_dentry));
+	fsstack_copy_inode_size(dir, d_inode(lower_dir_dentry));
+	set_nlink(dir, d_inode(lower_dir_dentry)->i_nlink);
+
+out:
+	unlock_dir(lower_dir_dentry);
+	sdcardfs_put_real_lower(dentry, &lower_path);
+	REVERT_CRED(saved_cred);
+out_eacces:
+	return err;
+}
+
+#if 0
+static int sdcardfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode,
+			dev_t dev)
+{
+	int err;
+	struct dentry *lower_dentry;
+	struct dentry *lower_parent_dentry = NULL;
+	struct path lower_path;
+
+	OVERRIDE_CRED(SDCARDFS_SB(dir->i_sb));
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	lower_dentry = lower_path.dentry;
+	lower_parent_dentry = lock_parent(lower_dentry);
+
+	err = vfs_mknod(d_inode(lower_parent_dentry), lower_dentry, mode, dev);
+	if (err)
+		goto out;
+
+	err = sdcardfs_interpose(dentry, dir->i_sb, &lower_path);
+	if (err)
+		goto out;
+	fsstack_copy_attr_times(dir, sdcardfs_lower_inode(dir));
+	fsstack_copy_inode_size(dir, d_inode(lower_parent_dentry));
+
+out:
+	unlock_dir(lower_parent_dentry);
+	sdcardfs_put_lower_path(dentry, &lower_path);
+	REVERT_CRED();
+	return err;
+}
+#endif
+
+/*
+ * The locking rules in sdcardfs_rename are complex.  We could use a simpler
+ * superblock-level name-space lock for renames and copy-ups.
+ */
+static int sdcardfs_rename(struct inode *old_dir, struct dentry *old_dentry,
+			 struct inode *new_dir, struct dentry *new_dentry)
+{
+	int err = 0;
+	struct dentry *lower_old_dentry = NULL;
+	struct dentry *lower_new_dentry = NULL;
+	struct dentry *lower_old_dir_dentry = NULL;
+	struct dentry *lower_new_dir_dentry = NULL;
+	struct vfsmount *lower_mnt = NULL;
+	struct dentry *trap = NULL;
+	struct path lower_old_path, lower_new_path;
+	const struct cred *saved_cred = NULL;
+
+	if (!check_caller_access_to_name(old_dir, &old_dentry->d_name) ||
+		!check_caller_access_to_name(new_dir, &new_dentry->d_name)) {
+		err = -EACCES;
+		goto out_eacces;
+	}
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(SDCARDFS_SB(old_dir->i_sb), saved_cred, SDCARDFS_I(new_dir));
+
+	sdcardfs_get_real_lower(old_dentry, &lower_old_path);
+	sdcardfs_get_lower_path(new_dentry, &lower_new_path);
+	lower_old_dentry = lower_old_path.dentry;
+	lower_new_dentry = lower_new_path.dentry;
+	lower_mnt = lower_old_path.mnt;
+	lower_old_dir_dentry = dget_parent(lower_old_dentry);
+	lower_new_dir_dentry = dget_parent(lower_new_dentry);
+
+	trap = lock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
+	/* source should not be ancestor of target */
+	if (trap == lower_old_dentry) {
+		err = -EINVAL;
+		goto out;
+	}
+	/* target should not be ancestor of source */
+	if (trap == lower_new_dentry) {
+		err = -ENOTEMPTY;
+		goto out;
+	}
+
+	err = vfs_rename2(lower_mnt,
+			 d_inode(lower_old_dir_dentry), lower_old_dentry,
+			 d_inode(lower_new_dir_dentry), lower_new_dentry,
+			 NULL, 0);
+	if (err)
+		goto out;
+
+	/* Copy attrs from lower dir, but i_uid/i_gid */
+	sdcardfs_copy_and_fix_attrs(new_dir, d_inode(lower_new_dir_dentry));
+	fsstack_copy_inode_size(new_dir, d_inode(lower_new_dir_dentry));
+
+	if (new_dir != old_dir) {
+		sdcardfs_copy_and_fix_attrs(old_dir, d_inode(lower_old_dir_dentry));
+		fsstack_copy_inode_size(old_dir, d_inode(lower_old_dir_dentry));
+	}
+	get_derived_permission_new(new_dentry->d_parent, old_dentry, &new_dentry->d_name);
+	fixup_tmp_permissions(d_inode(old_dentry));
+	fixup_lower_ownership(old_dentry, new_dentry->d_name.name);
+	d_invalidate(old_dentry); /* Can't fixup ownership recursively :( */
+out:
+	unlock_rename(lower_old_dir_dentry, lower_new_dir_dentry);
+	dput(lower_old_dir_dentry);
+	dput(lower_new_dir_dentry);
+	sdcardfs_put_real_lower(old_dentry, &lower_old_path);
+	sdcardfs_put_lower_path(new_dentry, &lower_new_path);
+	REVERT_CRED(saved_cred);
+out_eacces:
+	return err;
+}
+
+#if 0
+static int sdcardfs_readlink(struct dentry *dentry, char __user *buf, int bufsiz)
+{
+	int err;
+	struct dentry *lower_dentry;
+	struct path lower_path;
+	/* XXX readlink does not requires overriding credential */
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	lower_dentry = lower_path.dentry;
+	if (!d_inode(lower_dentry)->i_op ||
+	    !d_inode(lower_dentry)->i_op->readlink) {
+		err = -EINVAL;
+		goto out;
+	}
+
+	err = d_inode(lower_dentry)->i_op->readlink(lower_dentry,
+						    buf, bufsiz);
+	if (err < 0)
+		goto out;
+	fsstack_copy_attr_atime(d_inode(dentry), d_inode(lower_dentry));
+
+out:
+	sdcardfs_put_lower_path(dentry, &lower_path);
+	return err;
+}
+#endif
+
+#if 0
+static const char *sdcardfs_follow_link(struct dentry *dentry, void **cookie)
+{
+	char *buf;
+	int len = PAGE_SIZE, err;
+	mm_segment_t old_fs;
+
+	/* This is freed by the put_link method assuming a successful call. */
+	buf = kmalloc(len, GFP_KERNEL);
+	if (!buf) {
+		buf = ERR_PTR(-ENOMEM);
+		return buf;
+	}
+
+	/* read the symlink, and then we will follow it */
+	old_fs = get_fs();
+	set_fs(KERNEL_DS);
+	err = sdcardfs_readlink(dentry, buf, len);
+	set_fs(old_fs);
+	if (err < 0) {
+		kfree(buf);
+		buf = ERR_PTR(err);
+	} else {
+		buf[err] = '\0';
+	}
+	return *cookie = buf;
+}
+#endif
+
+static int sdcardfs_permission_wrn(struct inode *inode, int mask)
+{
+	WARN_RATELIMIT(1, "sdcardfs does not support permission. Use permission2.\n");
+	return -EINVAL;
+}
+
+void copy_attrs(struct inode *dest, const struct inode *src)
+{
+	dest->i_mode = src->i_mode;
+	dest->i_uid = src->i_uid;
+	dest->i_gid = src->i_gid;
+	dest->i_rdev = src->i_rdev;
+	dest->i_atime = src->i_atime;
+	dest->i_mtime = src->i_mtime;
+	dest->i_ctime = src->i_ctime;
+	dest->i_blkbits = src->i_blkbits;
+	dest->i_flags = src->i_flags;
+#ifdef CONFIG_FS_POSIX_ACL
+	dest->i_acl = src->i_acl;
+#endif
+#ifdef CONFIG_SECURITY
+	dest->i_security = src->i_security;
+#endif
+}
+
+static int sdcardfs_permission(struct vfsmount *mnt, struct inode *inode, int mask)
+{
+	int err;
+	struct inode tmp;
+	struct sdcardfs_inode_data *top = top_data_get(SDCARDFS_I(inode));
+
+	if (!top)
+		return -EINVAL;
+
+	/*
+	 * Permission check on sdcardfs inode.
+	 * Calling process should have AID_SDCARD_RW permission
+	 * Since generic_permission only needs i_mode, i_uid,
+	 * i_gid, and i_sb, we can create a fake inode to pass
+	 * this information down in.
+	 *
+	 * The underlying code may attempt to take locks in some
+	 * cases for features we're not using, but if that changes,
+	 * locks must be dealt with to avoid undefined behavior.
+	 */
+	copy_attrs(&tmp, inode);
+	tmp.i_uid = make_kuid(&init_user_ns, top->d_uid);
+	tmp.i_gid = make_kgid(&init_user_ns, get_gid(mnt, top));
+	tmp.i_mode = (inode->i_mode & S_IFMT)
+			| get_mode(mnt, SDCARDFS_I(inode), top);
+	data_put(top);
+	tmp.i_sb = inode->i_sb;
+	if (IS_POSIXACL(inode))
+		pr_warn("%s: This may be undefined behavior...\n", __func__);
+	err = generic_permission(&tmp, mask);
+	/* XXX
+	 * Original sdcardfs code calls inode_permission(lower_inode,.. )
+	 * for checking inode permission. But doing such things here seems
+	 * duplicated work, because the functions called after this func,
+	 * such as vfs_create, vfs_unlink, vfs_rename, and etc,
+	 * does exactly same thing, i.e., they calls inode_permission().
+	 * So we just let they do the things.
+	 * If there are any security hole, just uncomment following if block.
+	 */
+#if 0
+	if (!err) {
+		/*
+		 * Permission check on lower_inode(=EXT4).
+		 * we check it with AID_MEDIA_RW permission
+		 */
+		struct inode *lower_inode;
+
+		OVERRIDE_CRED(SDCARDFS_SB(inode->sb));
+
+		lower_inode = sdcardfs_lower_inode(inode);
+		err = inode_permission(lower_inode, mask);
+
+		REVERT_CRED();
+	}
+#endif
+	return err;
+
+}
+
+static int sdcardfs_setattr_wrn(struct dentry *dentry, struct iattr *ia)
+{
+	WARN_RATELIMIT(1, "sdcardfs does not support setattr. User setattr2.\n");
+	return -EINVAL;
+}
+
+static int sdcardfs_setattr(struct vfsmount *mnt, struct dentry *dentry, struct iattr *ia)
+{
+	int err;
+	struct dentry *lower_dentry;
+	struct vfsmount *lower_mnt;
+	struct inode *inode;
+	struct inode *lower_inode;
+	struct path lower_path;
+	struct iattr lower_ia;
+	struct dentry *parent;
+	struct inode tmp;
+	struct sdcardfs_inode_data *top;
+	const struct cred *saved_cred = NULL;
+
+	inode = d_inode(dentry);
+	top = top_data_get(SDCARDFS_I(inode));
+
+	if (!top)
+		return -EINVAL;
+
+	/*
+	 * Permission check on sdcardfs inode.
+	 * Calling process should have AID_SDCARD_RW permission
+	 * Since generic_permission only needs i_mode, i_uid,
+	 * i_gid, and i_sb, we can create a fake inode to pass
+	 * this information down in.
+	 *
+	 * The underlying code may attempt to take locks in some
+	 * cases for features we're not using, but if that changes,
+	 * locks must be dealt with to avoid undefined behavior.
+	 *
+	 */
+	copy_attrs(&tmp, inode);
+	tmp.i_uid = make_kuid(&init_user_ns, top->d_uid);
+	tmp.i_gid = make_kgid(&init_user_ns, get_gid(mnt, top));
+	tmp.i_mode = (inode->i_mode & S_IFMT)
+			| get_mode(mnt, SDCARDFS_I(inode), top);
+	tmp.i_size = i_size_read(inode);
+	data_put(top);
+	tmp.i_sb = inode->i_sb;
+
+	/*
+	 * Check if user has permission to change inode.  We don't check if
+	 * this user can change the lower inode: that should happen when
+	 * calling notify_change on the lower inode.
+	 */
+	/* prepare our own lower struct iattr (with the lower file) */
+	memcpy(&lower_ia, ia, sizeof(lower_ia));
+	/* Allow touch updating timestamps. A previous permission check ensures
+	 * we have write access. Changes to mode, owner, and group are ignored
+	 */
+	ia->ia_valid |= ATTR_FORCE;
+	err = inode_change_ok(&tmp, ia);
+
+	if (!err) {
+		/* check the Android group ID */
+		parent = dget_parent(dentry);
+		if (!check_caller_access_to_name(d_inode(parent), &dentry->d_name))
+			err = -EACCES;
+		dput(parent);
+	}
+
+	if (err)
+		goto out_err;
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED(SDCARDFS_SB(dentry->d_sb), saved_cred, SDCARDFS_I(inode));
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	lower_dentry = lower_path.dentry;
+	lower_mnt = lower_path.mnt;
+	lower_inode = sdcardfs_lower_inode(inode);
+
+	if (ia->ia_valid & ATTR_FILE)
+		lower_ia.ia_file = sdcardfs_lower_file(ia->ia_file);
+
+	lower_ia.ia_valid &= ~(ATTR_UID | ATTR_GID | ATTR_MODE);
+
+	/*
+	 * If shrinking, first truncate upper level to cancel writing dirty
+	 * pages beyond the new eof; and also if its' maxbytes is more
+	 * limiting (fail with -EFBIG before making any change to the lower
+	 * level).  There is no need to vmtruncate the upper level
+	 * afterwards in the other cases: we fsstack_copy_inode_size from
+	 * the lower level.
+	 */
+	if (ia->ia_valid & ATTR_SIZE) {
+		err = inode_newsize_ok(&tmp, ia->ia_size);
+		if (err) {
+			goto out;
+		}
+		truncate_setsize(inode, ia->ia_size);
+	}
+
+	/*
+	 * mode change is for clearing setuid/setgid bits. Allow lower fs
+	 * to interpret this in its own way.
+	 */
+	if (lower_ia.ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
+		lower_ia.ia_valid &= ~ATTR_MODE;
+
+	/* notify the (possibly copied-up) lower inode */
+	/*
+	 * Note: we use d_inode(lower_dentry), because lower_inode may be
+	 * unlinked (no inode->i_sb and i_ino==0.  This happens if someone
+	 * tries to open(), unlink(), then ftruncate() a file.
+	 */
+	mutex_lock(&d_inode(lower_dentry)->i_mutex);
+	err = notify_change2(lower_mnt, lower_dentry, &lower_ia, /* note: lower_ia */
+			NULL);
+	mutex_unlock(&d_inode(lower_dentry)->i_mutex);
+	if (err)
+		goto out;
+
+	/* get attributes from the lower inode and update derived permissions */
+	sdcardfs_copy_and_fix_attrs(inode, lower_inode);
+
+	/*
+	 * Not running fsstack_copy_inode_size(inode, lower_inode), because
+	 * VFS should update our inode size, and notify_change on
+	 * lower_inode should update its size.
+	 */
+
+out:
+	sdcardfs_put_lower_path(dentry, &lower_path);
+	REVERT_CRED(saved_cred);
+out_err:
+	return err;
+}
+
+static int sdcardfs_fillattr(struct vfsmount *mnt,
+				struct inode *inode, struct kstat *stat)
+{
+	struct sdcardfs_inode_info *info = SDCARDFS_I(inode);
+	struct sdcardfs_inode_data *top = top_data_get(info);
+
+	if (!top)
+		return -EINVAL;
+
+	stat->dev = inode->i_sb->s_dev;
+	stat->ino = inode->i_ino;
+	stat->mode = (inode->i_mode  & S_IFMT) | get_mode(mnt, info, top);
+	stat->nlink = inode->i_nlink;
+	stat->uid = make_kuid(&init_user_ns, top->d_uid);
+	stat->gid = make_kgid(&init_user_ns, get_gid(mnt, top));
+	stat->rdev = inode->i_rdev;
+	stat->size = i_size_read(inode);
+	stat->atime = inode->i_atime;
+	stat->mtime = inode->i_mtime;
+	stat->ctime = inode->i_ctime;
+	stat->blksize = (1 << inode->i_blkbits);
+	stat->blocks = inode->i_blocks;
+	data_put(top);
+	return 0;
+}
+
+static int sdcardfs_getattr(struct vfsmount *mnt, struct dentry *dentry,
+		 struct kstat *stat)
+{
+	struct kstat lower_stat;
+	struct path lower_path;
+	struct dentry *parent;
+	int err;
+
+	parent = dget_parent(dentry);
+	if (!check_caller_access_to_name(d_inode(parent), &dentry->d_name)) {
+		dput(parent);
+		return -EACCES;
+	}
+	dput(parent);
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	err = vfs_getattr(&lower_path, &lower_stat);
+	if (err)
+		goto out;
+	sdcardfs_copy_and_fix_attrs(d_inode(dentry),
+			      d_inode(lower_path.dentry));
+	err = sdcardfs_fillattr(mnt, d_inode(dentry), stat);
+	stat->blocks = lower_stat.blocks;
+out:
+	sdcardfs_put_lower_path(dentry, &lower_path);
+	return err;
+}
+
+const struct inode_operations sdcardfs_symlink_iops = {
+	.permission2	= sdcardfs_permission,
+	.setattr2	= sdcardfs_setattr,
+	/* XXX Following operations are implemented,
+	 *     but FUSE(sdcard) or FAT does not support them
+	 *     These methods are *NOT* perfectly tested.
+	.readlink	= sdcardfs_readlink,
+	.follow_link	= sdcardfs_follow_link,
+	.put_link	= kfree_put_link,
+	 */
+};
+
+const struct inode_operations sdcardfs_dir_iops = {
+	.create		= sdcardfs_create,
+	.lookup		= sdcardfs_lookup,
+	.permission	= sdcardfs_permission_wrn,
+	.permission2	= sdcardfs_permission,
+	.unlink		= sdcardfs_unlink,
+	.mkdir		= sdcardfs_mkdir,
+	.rmdir		= sdcardfs_rmdir,
+	.rename		= sdcardfs_rename,
+	.setattr	= sdcardfs_setattr_wrn,
+	.setattr2	= sdcardfs_setattr,
+	.getattr	= sdcardfs_getattr,
+	/* XXX Following operations are implemented,
+	 *     but FUSE(sdcard) or FAT does not support them
+	 *     These methods are *NOT* perfectly tested.
+	.symlink	= sdcardfs_symlink,
+	.link		= sdcardfs_link,
+	.mknod		= sdcardfs_mknod,
+	 */
+};
+
+const struct inode_operations sdcardfs_main_iops = {
+	.permission	= sdcardfs_permission_wrn,
+	.permission2	= sdcardfs_permission,
+	.setattr	= sdcardfs_setattr_wrn,
+	.setattr2	= sdcardfs_setattr,
+	.getattr	= sdcardfs_getattr,
+};
diff --git a/fs/sdcardfs/lookup.c b/fs/sdcardfs/lookup.c
new file mode 100644
index 000000000000..676e394e07be
--- /dev/null
+++ b/fs/sdcardfs/lookup.c
@@ -0,0 +1,466 @@
+/*
+ * fs/sdcardfs/lookup.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+#include "linux/delay.h"
+
+/* The dentry cache is just so we have properly sized dentries */
+static struct kmem_cache *sdcardfs_dentry_cachep;
+
+int sdcardfs_init_dentry_cache(void)
+{
+	sdcardfs_dentry_cachep =
+		kmem_cache_create("sdcardfs_dentry",
+				  sizeof(struct sdcardfs_dentry_info),
+				  0, SLAB_RECLAIM_ACCOUNT, NULL);
+
+	return sdcardfs_dentry_cachep ? 0 : -ENOMEM;
+}
+
+void sdcardfs_destroy_dentry_cache(void)
+{
+	kmem_cache_destroy(sdcardfs_dentry_cachep);
+}
+
+void free_dentry_private_data(struct dentry *dentry)
+{
+	if (!dentry || !dentry->d_fsdata)
+		return;
+	kmem_cache_free(sdcardfs_dentry_cachep, dentry->d_fsdata);
+	dentry->d_fsdata = NULL;
+}
+
+/* allocate new dentry private data */
+int new_dentry_private_data(struct dentry *dentry)
+{
+	struct sdcardfs_dentry_info *info = SDCARDFS_D(dentry);
+
+	/* use zalloc to init dentry_info.lower_path */
+	info = kmem_cache_zalloc(sdcardfs_dentry_cachep, GFP_ATOMIC);
+	if (!info)
+		return -ENOMEM;
+
+	spin_lock_init(&info->lock);
+	dentry->d_fsdata = info;
+
+	return 0;
+}
+
+struct inode_data {
+	struct inode *lower_inode;
+	userid_t id;
+};
+
+static int sdcardfs_inode_test(struct inode *inode, void *candidate_data/*void *candidate_lower_inode*/)
+{
+	struct inode *current_lower_inode = sdcardfs_lower_inode(inode);
+	userid_t current_userid = SDCARDFS_I(inode)->data->userid;
+
+	if (current_lower_inode == ((struct inode_data *)candidate_data)->lower_inode &&
+			current_userid == ((struct inode_data *)candidate_data)->id)
+		return 1; /* found a match */
+	else
+		return 0; /* no match */
+}
+
+static int sdcardfs_inode_set(struct inode *inode, void *lower_inode)
+{
+	/* we do actual inode initialization in sdcardfs_iget */
+	return 0;
+}
+
+struct inode *sdcardfs_iget(struct super_block *sb, struct inode *lower_inode, userid_t id)
+{
+	struct sdcardfs_inode_info *info;
+	struct inode_data data;
+	struct inode *inode; /* the new inode to return */
+
+	if (!igrab(lower_inode))
+		return ERR_PTR(-ESTALE);
+
+	data.id = id;
+	data.lower_inode = lower_inode;
+	inode = iget5_locked(sb, /* our superblock */
+			     /*
+			      * hashval: we use inode number, but we can
+			      * also use "(unsigned long)lower_inode"
+			      * instead.
+			      */
+			     lower_inode->i_ino, /* hashval */
+			     sdcardfs_inode_test, /* inode comparison function */
+			     sdcardfs_inode_set, /* inode init function */
+			     &data); /* data passed to test+set fxns */
+	if (!inode) {
+		iput(lower_inode);
+		return ERR_PTR(-ENOMEM);
+	}
+	/* if found a cached inode, then just return it (after iput) */
+	if (!(inode->i_state & I_NEW)) {
+		iput(lower_inode);
+		return inode;
+	}
+
+	/* initialize new inode */
+	info = SDCARDFS_I(inode);
+
+	inode->i_ino = lower_inode->i_ino;
+	sdcardfs_set_lower_inode(inode, lower_inode);
+
+	inode->i_version++;
+
+	/* use different set of inode ops for symlinks & directories */
+	if (S_ISDIR(lower_inode->i_mode))
+		inode->i_op = &sdcardfs_dir_iops;
+	else if (S_ISLNK(lower_inode->i_mode))
+		inode->i_op = &sdcardfs_symlink_iops;
+	else
+		inode->i_op = &sdcardfs_main_iops;
+
+	/* use different set of file ops for directories */
+	if (S_ISDIR(lower_inode->i_mode))
+		inode->i_fop = &sdcardfs_dir_fops;
+	else
+		inode->i_fop = &sdcardfs_main_fops;
+
+	inode->i_mapping->a_ops = &sdcardfs_aops;
+
+	inode->i_atime.tv_sec = 0;
+	inode->i_atime.tv_nsec = 0;
+	inode->i_mtime.tv_sec = 0;
+	inode->i_mtime.tv_nsec = 0;
+	inode->i_ctime.tv_sec = 0;
+	inode->i_ctime.tv_nsec = 0;
+
+	/* properly initialize special inodes */
+	if (S_ISBLK(lower_inode->i_mode) || S_ISCHR(lower_inode->i_mode) ||
+	    S_ISFIFO(lower_inode->i_mode) || S_ISSOCK(lower_inode->i_mode))
+		init_special_inode(inode, lower_inode->i_mode,
+				   lower_inode->i_rdev);
+
+	/* all well, copy inode attributes */
+	sdcardfs_copy_and_fix_attrs(inode, lower_inode);
+	fsstack_copy_inode_size(inode, lower_inode);
+
+	unlock_new_inode(inode);
+	return inode;
+}
+
+/*
+ * Helper interpose routine, called directly by ->lookup to handle
+ * spliced dentries.
+ */
+static struct dentry *__sdcardfs_interpose(struct dentry *dentry,
+					 struct super_block *sb,
+					 struct path *lower_path,
+					 userid_t id)
+{
+	struct inode *inode;
+	struct inode *lower_inode;
+	struct super_block *lower_sb;
+	struct dentry *ret_dentry;
+
+	lower_inode = d_inode(lower_path->dentry);
+	lower_sb = sdcardfs_lower_super(sb);
+
+	/* check that the lower file system didn't cross a mount point */
+	if (lower_inode->i_sb != lower_sb) {
+		ret_dentry = ERR_PTR(-EXDEV);
+		goto out;
+	}
+
+	/*
+	 * We allocate our new inode below by calling sdcardfs_iget,
+	 * which will initialize some of the new inode's fields
+	 */
+
+	/* inherit lower inode number for sdcardfs's inode */
+	inode = sdcardfs_iget(sb, lower_inode, id);
+	if (IS_ERR(inode)) {
+		ret_dentry = ERR_CAST(inode);
+		goto out;
+	}
+
+	ret_dentry = d_splice_alias(inode, dentry);
+	dentry = ret_dentry ?: dentry;
+	if (!IS_ERR(dentry))
+		update_derived_permission_lock(dentry);
+out:
+	return ret_dentry;
+}
+
+/*
+ * Connect an sdcardfs inode dentry/inode with several lower ones.  This is
+ * the classic stackable file system "vnode interposition" action.
+ *
+ * @dentry: sdcardfs's dentry which interposes on lower one
+ * @sb: sdcardfs's super_block
+ * @lower_path: the lower path (caller does path_get/put)
+ */
+int sdcardfs_interpose(struct dentry *dentry, struct super_block *sb,
+		     struct path *lower_path, userid_t id)
+{
+	struct dentry *ret_dentry;
+
+	ret_dentry = __sdcardfs_interpose(dentry, sb, lower_path, id);
+	return PTR_ERR(ret_dentry);
+}
+
+struct sdcardfs_name_data {
+	struct dir_context ctx;
+	const struct qstr *to_find;
+	char *name;
+	bool found;
+};
+
+static int sdcardfs_name_match(struct dir_context *ctx, const char *name,
+		int namelen, loff_t offset, u64 ino, unsigned int d_type)
+{
+	struct sdcardfs_name_data *buf = container_of(ctx, struct sdcardfs_name_data, ctx);
+	struct qstr candidate = QSTR_INIT(name, namelen);
+
+	if (qstr_case_eq(buf->to_find, &candidate)) {
+		memcpy(buf->name, name, namelen);
+		buf->name[namelen] = 0;
+		buf->found = true;
+		return 1;
+	}
+	return 0;
+}
+
+/*
+ * Main driver function for sdcardfs's lookup.
+ *
+ * Returns: NULL (ok), ERR_PTR if an error occurred.
+ * Fills in lower_parent_path with <dentry,mnt> on success.
+ */
+static struct dentry *__sdcardfs_lookup(struct dentry *dentry,
+		unsigned int flags, struct path *lower_parent_path, userid_t id)
+{
+	int err = 0;
+	struct vfsmount *lower_dir_mnt;
+	struct dentry *lower_dir_dentry = NULL;
+	struct dentry *lower_dentry;
+	const struct qstr *name;
+	struct path lower_path;
+	struct qstr dname;
+	struct dentry *ret_dentry = NULL;
+	struct sdcardfs_sb_info *sbi;
+
+	sbi = SDCARDFS_SB(dentry->d_sb);
+	/* must initialize dentry operations */
+	d_set_d_op(dentry, &sdcardfs_ci_dops);
+
+	if (IS_ROOT(dentry))
+		goto out;
+
+	name = &dentry->d_name;
+
+	/* now start the actual lookup procedure */
+	lower_dir_dentry = lower_parent_path->dentry;
+	lower_dir_mnt = lower_parent_path->mnt;
+
+	/* Use vfs_path_lookup to check if the dentry exists or not */
+	err = vfs_path_lookup(lower_dir_dentry, lower_dir_mnt, name->name, 0,
+				&lower_path);
+	/* check for other cases */
+	if (err == -ENOENT) {
+		struct file *file;
+		const struct cred *cred = current_cred();
+
+		struct sdcardfs_name_data buffer = {
+			.ctx.actor = sdcardfs_name_match,
+			.to_find = name,
+			.name = __getname(),
+			.found = false,
+		};
+
+		if (!buffer.name) {
+			err = -ENOMEM;
+			goto out;
+		}
+		file = dentry_open(lower_parent_path, O_RDONLY, cred);
+		if (IS_ERR(file)) {
+			err = PTR_ERR(file);
+			goto put_name;
+		}
+		err = iterate_dir(file, &buffer.ctx);
+		fput(file);
+		if (err)
+			goto put_name;
+
+		if (buffer.found)
+			err = vfs_path_lookup(lower_dir_dentry,
+						lower_dir_mnt,
+						buffer.name, 0,
+						&lower_path);
+		else
+			err = -ENOENT;
+put_name:
+		__putname(buffer.name);
+	}
+
+	/* no error: handle positive dentries */
+	if (!err) {
+		/* check if the dentry is an obb dentry
+		 * if true, the lower_inode must be replaced with
+		 * the inode of the graft path
+		 */
+
+		if (need_graft_path(dentry)) {
+
+			/* setup_obb_dentry()
+			 * The lower_path will be stored to the dentry's orig_path
+			 * and the base obbpath will be copyed to the lower_path variable.
+			 * if an error returned, there's no change in the lower_path
+			 * returns: -ERRNO if error (0: no error)
+			 */
+			err = setup_obb_dentry(dentry, &lower_path);
+
+			if (err) {
+				/* if the sbi->obbpath is not available, we can optionally
+				 * setup the lower_path with its orig_path.
+				 * but, the current implementation just returns an error
+				 * because the sdcard daemon also regards this case as
+				 * a lookup fail.
+				 */
+				pr_info("sdcardfs: base obbpath is not available\n");
+				sdcardfs_put_reset_orig_path(dentry);
+				goto out;
+			}
+		}
+
+		sdcardfs_set_lower_path(dentry, &lower_path);
+		ret_dentry =
+			__sdcardfs_interpose(dentry, dentry->d_sb, &lower_path, id);
+		if (IS_ERR(ret_dentry)) {
+			err = PTR_ERR(ret_dentry);
+			 /* path_put underlying path on error */
+			sdcardfs_put_reset_lower_path(dentry);
+		}
+		goto out;
+	}
+
+	/*
+	 * We don't consider ENOENT an error, and we want to return a
+	 * negative dentry.
+	 */
+	if (err && err != -ENOENT)
+		goto out;
+
+	/* instatiate a new negative dentry */
+	dname.name = name->name;
+	dname.len = name->len;
+
+	/* See if the low-level filesystem might want
+	 * to use its own hash
+	 */
+	lower_dentry = d_hash_and_lookup(lower_dir_dentry, &dname);
+	if (IS_ERR(lower_dentry))
+		return lower_dentry;
+	if (!lower_dentry) {
+		/* We called vfs_path_lookup earlier, and did not get a negative
+		 * dentry then. Don't confuse the lower filesystem by forcing
+		 * one on it now...
+		 */
+		err = -ENOENT;
+		goto out;
+	}
+
+	lower_path.dentry = lower_dentry;
+	lower_path.mnt = mntget(lower_dir_mnt);
+	sdcardfs_set_lower_path(dentry, &lower_path);
+
+	/*
+	 * If the intent is to create a file, then don't return an error, so
+	 * the VFS will continue the process of making this negative dentry
+	 * into a positive one.
+	 */
+	if (flags & (LOOKUP_CREATE|LOOKUP_RENAME_TARGET))
+		err = 0;
+
+out:
+	if (err)
+		return ERR_PTR(err);
+	return ret_dentry;
+}
+
+/*
+ * On success:
+ * fills dentry object appropriate values and returns NULL.
+ * On fail (== error)
+ * returns error ptr
+ *
+ * @dir : Parent inode. It is locked (dir->i_mutex)
+ * @dentry : Target dentry to lookup. we should set each of fields.
+ *	     (dentry->d_name is initialized already)
+ * @nd : nameidata of parent inode
+ */
+struct dentry *sdcardfs_lookup(struct inode *dir, struct dentry *dentry,
+			     unsigned int flags)
+{
+	struct dentry *ret = NULL, *parent;
+	struct path lower_parent_path;
+	int err = 0;
+	const struct cred *saved_cred = NULL;
+
+	parent = dget_parent(dentry);
+
+	if (!check_caller_access_to_name(d_inode(parent), &dentry->d_name)) {
+		ret = ERR_PTR(-EACCES);
+		goto out_err;
+	}
+
+	/* save current_cred and override it */
+	OVERRIDE_CRED_PTR(SDCARDFS_SB(dir->i_sb), saved_cred, SDCARDFS_I(dir));
+
+	sdcardfs_get_lower_path(parent, &lower_parent_path);
+
+	/* allocate dentry private data.  We free it in ->d_release */
+	err = new_dentry_private_data(dentry);
+	if (err) {
+		ret = ERR_PTR(err);
+		goto out;
+	}
+
+	ret = __sdcardfs_lookup(dentry, flags, &lower_parent_path,
+				SDCARDFS_I(dir)->data->userid);
+	if (IS_ERR(ret))
+		goto out;
+	if (ret)
+		dentry = ret;
+	if (d_inode(dentry)) {
+		fsstack_copy_attr_times(d_inode(dentry),
+					sdcardfs_lower_inode(d_inode(dentry)));
+		/* get derived permission */
+		get_derived_permission(parent, dentry);
+		fixup_tmp_permissions(d_inode(dentry));
+		fixup_lower_ownership(dentry, dentry->d_name.name);
+	}
+	/* update parent directory's atime */
+	fsstack_copy_attr_atime(d_inode(parent),
+				sdcardfs_lower_inode(d_inode(parent)));
+
+out:
+	sdcardfs_put_lower_path(parent, &lower_parent_path);
+	REVERT_CRED(saved_cred);
+out_err:
+	dput(parent);
+	return ret;
+}
diff --git a/fs/sdcardfs/main.c b/fs/sdcardfs/main.c
new file mode 100644
index 000000000000..0a2b5167e9a2
--- /dev/null
+++ b/fs/sdcardfs/main.c
@@ -0,0 +1,486 @@
+/*
+ * fs/sdcardfs/main.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+#include <linux/module.h>
+#include <linux/types.h>
+#include <linux/parser.h>
+
+enum {
+	Opt_fsuid,
+	Opt_fsgid,
+	Opt_gid,
+	Opt_debug,
+	Opt_mask,
+	Opt_multiuser,
+	Opt_userid,
+	Opt_reserved_mb,
+	Opt_gid_derivation,
+	Opt_err,
+};
+
+static const match_table_t sdcardfs_tokens = {
+	{Opt_fsuid, "fsuid=%u"},
+	{Opt_fsgid, "fsgid=%u"},
+	{Opt_gid, "gid=%u"},
+	{Opt_debug, "debug"},
+	{Opt_mask, "mask=%u"},
+	{Opt_userid, "userid=%d"},
+	{Opt_multiuser, "multiuser"},
+	{Opt_gid_derivation, "derive_gid"},
+	{Opt_reserved_mb, "reserved_mb=%u"},
+	{Opt_err, NULL}
+};
+
+static int parse_options(struct super_block *sb, char *options, int silent,
+				int *debug, struct sdcardfs_vfsmount_options *vfsopts,
+				struct sdcardfs_mount_options *opts)
+{
+	char *p;
+	substring_t args[MAX_OPT_ARGS];
+	int option;
+
+	/* by default, we use AID_MEDIA_RW as uid, gid */
+	opts->fs_low_uid = AID_MEDIA_RW;
+	opts->fs_low_gid = AID_MEDIA_RW;
+	vfsopts->mask = 0;
+	opts->multiuser = false;
+	opts->fs_user_id = 0;
+	vfsopts->gid = 0;
+	/* by default, 0MB is reserved */
+	opts->reserved_mb = 0;
+	/* by default, gid derivation is off */
+	opts->gid_derivation = false;
+
+	*debug = 0;
+
+	if (!options)
+		return 0;
+
+	while ((p = strsep(&options, ",")) != NULL) {
+		int token;
+
+		if (!*p)
+			continue;
+
+		token = match_token(p, sdcardfs_tokens, args);
+
+		switch (token) {
+		case Opt_debug:
+			*debug = 1;
+			break;
+		case Opt_fsuid:
+			if (match_int(&args[0], &option))
+				return 0;
+			opts->fs_low_uid = option;
+			break;
+		case Opt_fsgid:
+			if (match_int(&args[0], &option))
+				return 0;
+			opts->fs_low_gid = option;
+			break;
+		case Opt_gid:
+			if (match_int(&args[0], &option))
+				return 0;
+			vfsopts->gid = option;
+			break;
+		case Opt_userid:
+			if (match_int(&args[0], &option))
+				return 0;
+			opts->fs_user_id = option;
+			break;
+		case Opt_mask:
+			if (match_int(&args[0], &option))
+				return 0;
+			vfsopts->mask = option;
+			break;
+		case Opt_multiuser:
+			opts->multiuser = true;
+			break;
+		case Opt_reserved_mb:
+			if (match_int(&args[0], &option))
+				return 0;
+			opts->reserved_mb = option;
+			break;
+		case Opt_gid_derivation:
+			opts->gid_derivation = true;
+			break;
+		/* unknown option */
+		default:
+			if (!silent)
+				pr_err("Unrecognized mount option \"%s\" or missing value", p);
+			return -EINVAL;
+		}
+	}
+
+	if (*debug) {
+		pr_info("sdcardfs : options - debug:%d\n", *debug);
+		pr_info("sdcardfs : options - uid:%d\n",
+							opts->fs_low_uid);
+		pr_info("sdcardfs : options - gid:%d\n",
+							opts->fs_low_gid);
+	}
+
+	return 0;
+}
+
+int parse_options_remount(struct super_block *sb, char *options, int silent,
+				struct sdcardfs_vfsmount_options *vfsopts)
+{
+	char *p;
+	substring_t args[MAX_OPT_ARGS];
+	int option;
+	int debug;
+
+	if (!options)
+		return 0;
+
+	while ((p = strsep(&options, ",")) != NULL) {
+		int token;
+
+		if (!*p)
+			continue;
+
+		token = match_token(p, sdcardfs_tokens, args);
+
+		switch (token) {
+		case Opt_debug:
+			debug = 1;
+			break;
+		case Opt_gid:
+			if (match_int(&args[0], &option))
+				return 0;
+			vfsopts->gid = option;
+
+			break;
+		case Opt_mask:
+			if (match_int(&args[0], &option))
+				return 0;
+			vfsopts->mask = option;
+			break;
+		case Opt_multiuser:
+		case Opt_userid:
+		case Opt_fsuid:
+		case Opt_fsgid:
+		case Opt_reserved_mb:
+			pr_warn("Option \"%s\" can't be changed during remount\n", p);
+			break;
+		/* unknown option */
+		default:
+			if (!silent)
+				pr_err("Unrecognized mount option \"%s\" or missing value", p);
+			return -EINVAL;
+		}
+	}
+
+	if (debug) {
+		pr_info("sdcardfs : options - debug:%d\n", debug);
+		pr_info("sdcardfs : options - gid:%d\n", vfsopts->gid);
+		pr_info("sdcardfs : options - mask:%d\n", vfsopts->mask);
+	}
+
+	return 0;
+}
+
+#if 0
+/*
+ * our custom d_alloc_root work-alike
+ *
+ * we can't use d_alloc_root if we want to use our own interpose function
+ * unchanged, so we simply call our own "fake" d_alloc_root
+ */
+static struct dentry *sdcardfs_d_alloc_root(struct super_block *sb)
+{
+	struct dentry *ret = NULL;
+
+	if (sb) {
+		static const struct qstr name = {
+			.name = "/",
+			.len = 1
+		};
+
+		ret = d_alloc(NULL, &name);
+		if (ret) {
+			d_set_d_op(ret, &sdcardfs_ci_dops);
+			ret->d_sb = sb;
+			ret->d_parent = ret;
+		}
+	}
+	return ret;
+}
+#endif
+
+DEFINE_MUTEX(sdcardfs_super_list_lock);
+EXPORT_SYMBOL_GPL(sdcardfs_super_list_lock);
+LIST_HEAD(sdcardfs_super_list);
+EXPORT_SYMBOL_GPL(sdcardfs_super_list);
+
+/*
+ * There is no need to lock the sdcardfs_super_info's rwsem as there is no
+ * way anyone can have a reference to the superblock at this point in time.
+ */
+static int sdcardfs_read_super(struct vfsmount *mnt, struct super_block *sb,
+		const char *dev_name, void *raw_data, int silent)
+{
+	int err = 0;
+	int debug;
+	struct super_block *lower_sb;
+	struct path lower_path;
+	struct sdcardfs_sb_info *sb_info;
+	struct sdcardfs_vfsmount_options *mnt_opt = mnt->data;
+	struct inode *inode;
+
+	pr_info("sdcardfs version 2.0\n");
+
+	if (!dev_name) {
+		pr_err("sdcardfs: read_super: missing dev_name argument\n");
+		err = -EINVAL;
+		goto out;
+	}
+
+	pr_info("sdcardfs: dev_name -> %s\n", dev_name);
+	pr_info("sdcardfs: options -> %s\n", (char *)raw_data);
+	pr_info("sdcardfs: mnt -> %p\n", mnt);
+
+	/* parse lower path */
+	err = kern_path(dev_name, LOOKUP_FOLLOW | LOOKUP_DIRECTORY,
+			&lower_path);
+	if (err) {
+		pr_err("sdcardfs: error accessing lower directory '%s'\n", dev_name);
+		goto out;
+	}
+
+	/* allocate superblock private data */
+	sb->s_fs_info = kzalloc(sizeof(struct sdcardfs_sb_info), GFP_KERNEL);
+	if (!SDCARDFS_SB(sb)) {
+		pr_crit("sdcardfs: read_super: out of memory\n");
+		err = -ENOMEM;
+		goto out_free;
+	}
+
+	sb_info = sb->s_fs_info;
+	/* parse options */
+	err = parse_options(sb, raw_data, silent, &debug, mnt_opt, &sb_info->options);
+	if (err) {
+		pr_err("sdcardfs: invalid options\n");
+		goto out_freesbi;
+	}
+
+	/* set the lower superblock field of upper superblock */
+	lower_sb = lower_path.dentry->d_sb;
+	atomic_inc(&lower_sb->s_active);
+	sdcardfs_set_lower_super(sb, lower_sb);
+
+	/* inherit maxbytes from lower file system */
+	sb->s_maxbytes = lower_sb->s_maxbytes;
+
+	/*
+	 * Our c/m/atime granularity is 1 ns because we may stack on file
+	 * systems whose granularity is as good.
+	 */
+	sb->s_time_gran = 1;
+
+	sb->s_magic = SDCARDFS_SUPER_MAGIC;
+	sb->s_op = &sdcardfs_sops;
+
+	/* get a new inode and allocate our root dentry */
+	inode = sdcardfs_iget(sb, d_inode(lower_path.dentry), 0);
+	if (IS_ERR(inode)) {
+		err = PTR_ERR(inode);
+		goto out_sput;
+	}
+	sb->s_root = d_make_root(inode);
+	if (!sb->s_root) {
+		err = -ENOMEM;
+		goto out_iput;
+	}
+	d_set_d_op(sb->s_root, &sdcardfs_ci_dops);
+
+	/* link the upper and lower dentries */
+	sb->s_root->d_fsdata = NULL;
+	err = new_dentry_private_data(sb->s_root);
+	if (err)
+		goto out_freeroot;
+
+	/* set the lower dentries for s_root */
+	sdcardfs_set_lower_path(sb->s_root, &lower_path);
+
+	/*
+	 * No need to call interpose because we already have a positive
+	 * dentry, which was instantiated by d_make_root.  Just need to
+	 * d_rehash it.
+	 */
+	d_rehash(sb->s_root);
+
+	/* setup permission policy */
+	sb_info->obbpath_s = kzalloc(PATH_MAX, GFP_KERNEL);
+	mutex_lock(&sdcardfs_super_list_lock);
+	if (sb_info->options.multiuser) {
+		setup_derived_state(d_inode(sb->s_root), PERM_PRE_ROOT,
+				sb_info->options.fs_user_id, AID_ROOT,
+				false, SDCARDFS_I(d_inode(sb->s_root))->data);
+		snprintf(sb_info->obbpath_s, PATH_MAX, "%s/obb", dev_name);
+	} else {
+		setup_derived_state(d_inode(sb->s_root), PERM_ROOT,
+				sb_info->options.fs_user_id, AID_ROOT,
+				false, SDCARDFS_I(d_inode(sb->s_root))->data);
+		snprintf(sb_info->obbpath_s, PATH_MAX, "%s/Android/obb", dev_name);
+	}
+	fixup_tmp_permissions(d_inode(sb->s_root));
+	sb_info->sb = sb;
+	list_add(&sb_info->list, &sdcardfs_super_list);
+	mutex_unlock(&sdcardfs_super_list_lock);
+
+	if (!silent)
+		pr_info("sdcardfs: mounted on top of %s type %s\n",
+				dev_name, lower_sb->s_type->name);
+	goto out; /* all is well */
+
+	/* no longer needed: free_dentry_private_data(sb->s_root); */
+out_freeroot:
+	dput(sb->s_root);
+out_iput:
+	iput(inode);
+out_sput:
+	/* drop refs we took earlier */
+	atomic_dec(&lower_sb->s_active);
+out_freesbi:
+	kfree(SDCARDFS_SB(sb));
+	sb->s_fs_info = NULL;
+out_free:
+	path_put(&lower_path);
+
+out:
+	return err;
+}
+
+struct sdcardfs_mount_private {
+	struct vfsmount *mnt;
+	const char *dev_name;
+	void *raw_data;
+};
+
+static int __sdcardfs_fill_super(
+	struct super_block *sb,
+	void *_priv, int silent)
+{
+	struct sdcardfs_mount_private *priv = _priv;
+
+	return sdcardfs_read_super(priv->mnt,
+		sb, priv->dev_name, priv->raw_data, silent);
+}
+
+static struct dentry *sdcardfs_mount(struct vfsmount *mnt,
+		struct file_system_type *fs_type, int flags,
+			    const char *dev_name, void *raw_data)
+{
+	struct sdcardfs_mount_private priv = {
+		.mnt = mnt,
+		.dev_name = dev_name,
+		.raw_data = raw_data
+	};
+
+	return mount_nodev(fs_type, flags,
+		&priv, __sdcardfs_fill_super);
+}
+
+static struct dentry *sdcardfs_mount_wrn(struct file_system_type *fs_type,
+		    int flags, const char *dev_name, void *raw_data)
+{
+	WARN(1, "sdcardfs does not support mount. Use mount2.\n");
+	return ERR_PTR(-EINVAL);
+}
+
+void *sdcardfs_alloc_mnt_data(void)
+{
+	return kmalloc(sizeof(struct sdcardfs_vfsmount_options), GFP_KERNEL);
+}
+
+void sdcardfs_kill_sb(struct super_block *sb)
+{
+	struct sdcardfs_sb_info *sbi;
+
+	if (sb->s_magic == SDCARDFS_SUPER_MAGIC) {
+		sbi = SDCARDFS_SB(sb);
+		mutex_lock(&sdcardfs_super_list_lock);
+		list_del(&sbi->list);
+		mutex_unlock(&sdcardfs_super_list_lock);
+	}
+	kill_anon_super(sb);
+}
+
+static struct file_system_type sdcardfs_fs_type = {
+	.owner		= THIS_MODULE,
+	.name		= SDCARDFS_NAME,
+	.mount		= sdcardfs_mount_wrn,
+	.mount2		= sdcardfs_mount,
+	.alloc_mnt_data = sdcardfs_alloc_mnt_data,
+	.kill_sb	= sdcardfs_kill_sb,
+	.fs_flags	= 0,
+};
+MODULE_ALIAS_FS(SDCARDFS_NAME);
+
+static int __init init_sdcardfs_fs(void)
+{
+	int err;
+
+	pr_info("Registering sdcardfs " SDCARDFS_VERSION "\n");
+
+	err = sdcardfs_init_inode_cache();
+	if (err)
+		goto out;
+	err = sdcardfs_init_dentry_cache();
+	if (err)
+		goto out;
+	err = packagelist_init();
+	if (err)
+		goto out;
+	err = register_filesystem(&sdcardfs_fs_type);
+out:
+	if (err) {
+		sdcardfs_destroy_inode_cache();
+		sdcardfs_destroy_dentry_cache();
+		packagelist_exit();
+	}
+	return err;
+}
+
+static void __exit exit_sdcardfs_fs(void)
+{
+	sdcardfs_destroy_inode_cache();
+	sdcardfs_destroy_dentry_cache();
+	packagelist_exit();
+	unregister_filesystem(&sdcardfs_fs_type);
+	pr_info("Completed sdcardfs module unload\n");
+}
+
+/* Original wrapfs authors */
+MODULE_AUTHOR("Erez Zadok, Filesystems and Storage Lab, Stony Brook University (http://www.fsl.cs.sunysb.edu/)");
+
+/* Original sdcardfs authors */
+MODULE_AUTHOR("Woojoong Lee, Daeho Jeong, Kitae Lee, Yeongjin Gil System Memory Lab., Samsung Electronics");
+
+/* Current maintainer */
+MODULE_AUTHOR("Daniel Rosenberg, Google");
+MODULE_DESCRIPTION("Sdcardfs " SDCARDFS_VERSION);
+MODULE_LICENSE("GPL");
+
+module_init(init_sdcardfs_fs);
+module_exit(exit_sdcardfs_fs);
diff --git a/fs/sdcardfs/mmap.c b/fs/sdcardfs/mmap.c
new file mode 100644
index 000000000000..b61f82275e7d
--- /dev/null
+++ b/fs/sdcardfs/mmap.c
@@ -0,0 +1,89 @@
+/*
+ * fs/sdcardfs/mmap.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+
+static int sdcardfs_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+	int err;
+	struct file *file;
+	const struct vm_operations_struct *lower_vm_ops;
+
+	file = (struct file *)vma->vm_private_data;
+	lower_vm_ops = SDCARDFS_F(file)->lower_vm_ops;
+	BUG_ON(!lower_vm_ops);
+
+	err = lower_vm_ops->fault(vma, vmf);
+	return err;
+}
+
+static void sdcardfs_vm_open(struct vm_area_struct *vma)
+{
+	struct file *file = (struct file *)vma->vm_private_data;
+
+	get_file(file);
+}
+
+static void sdcardfs_vm_close(struct vm_area_struct *vma)
+{
+	struct file *file = (struct file *)vma->vm_private_data;
+
+	fput(file);
+}
+
+static int sdcardfs_page_mkwrite(struct vm_area_struct *vma,
+			       struct vm_fault *vmf)
+{
+	int err = 0;
+	struct file *file;
+	const struct vm_operations_struct *lower_vm_ops;
+
+	file = (struct file *)vma->vm_private_data;
+	lower_vm_ops = SDCARDFS_F(file)->lower_vm_ops;
+	BUG_ON(!lower_vm_ops);
+	if (!lower_vm_ops->page_mkwrite)
+		goto out;
+
+	err = lower_vm_ops->page_mkwrite(vma, vmf);
+out:
+	return err;
+}
+
+static ssize_t sdcardfs_direct_IO(struct kiocb *iocb,
+		struct iov_iter *iter, loff_t pos)
+{
+	/*
+	 * This function should never be called directly.  We need it
+	 * to exist, to get past a check in open_check_o_direct(),
+	 * which is called from do_last().
+	 */
+	return -EINVAL;
+}
+
+const struct address_space_operations sdcardfs_aops = {
+	.direct_IO	= sdcardfs_direct_IO,
+};
+
+const struct vm_operations_struct sdcardfs_vm_ops = {
+	.fault		= sdcardfs_fault,
+	.page_mkwrite	= sdcardfs_page_mkwrite,
+	.open		= sdcardfs_vm_open,
+	.close		= sdcardfs_vm_close,
+};
diff --git a/fs/sdcardfs/multiuser.h b/fs/sdcardfs/multiuser.h
new file mode 100644
index 000000000000..85341e753f8c
--- /dev/null
+++ b/fs/sdcardfs/multiuser.h
@@ -0,0 +1,53 @@
+/*
+ * fs/sdcardfs/multiuser.h
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#define AID_USER_OFFSET     100000 /* offset for uid ranges for each user */
+#define AID_APP_START        10000 /* first app user */
+#define AID_APP_END          19999 /* last app user */
+#define AID_CACHE_GID_START  20000 /* start of gids for apps to mark cached data */
+#define AID_EXT_GID_START    30000 /* start of gids for apps to mark external data */
+#define AID_EXT_CACHE_GID_START 40000 /* start of gids for apps to mark external cached data */
+#define AID_EXT_CACHE_GID_END 49999   /* end of gids for apps to mark external cached data */
+#define AID_SHARED_GID_START 50000 /* start of gids for apps in each user to share */
+
+typedef uid_t userid_t;
+typedef uid_t appid_t;
+
+static inline uid_t multiuser_get_uid(userid_t user_id, appid_t app_id)
+{
+	return (user_id * AID_USER_OFFSET) + (app_id % AID_USER_OFFSET);
+}
+
+static inline bool uid_is_app(uid_t uid)
+{
+	appid_t appid = uid % AID_USER_OFFSET;
+
+	return appid >= AID_APP_START && appid <= AID_APP_END;
+}
+
+static inline gid_t multiuser_get_ext_cache_gid(uid_t uid)
+{
+	return uid - AID_APP_START + AID_EXT_CACHE_GID_START;
+}
+
+static inline gid_t multiuser_get_ext_gid(uid_t uid)
+{
+	return uid - AID_APP_START + AID_EXT_GID_START;
+}
diff --git a/fs/sdcardfs/packagelist.c b/fs/sdcardfs/packagelist.c
new file mode 100644
index 000000000000..66c6f8e72a02
--- /dev/null
+++ b/fs/sdcardfs/packagelist.c
@@ -0,0 +1,880 @@
+/*
+ * fs/sdcardfs/packagelist.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+#include <linux/hashtable.h>
+#include <linux/ctype.h>
+#include <linux/delay.h>
+#include <linux/radix-tree.h>
+#include <linux/dcache.h>
+
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/slab.h>
+
+#include <linux/configfs.h>
+
+struct hashtable_entry {
+	struct hlist_node hlist;
+	struct hlist_node dlist; /* for deletion cleanup */
+	struct qstr key;
+	atomic_t value;
+};
+
+static DEFINE_HASHTABLE(package_to_appid, 8);
+static DEFINE_HASHTABLE(package_to_userid, 8);
+static DEFINE_HASHTABLE(ext_to_groupid, 8);
+
+
+static struct kmem_cache *hashtable_entry_cachep;
+
+static unsigned int full_name_case_hash(const unsigned char *name, unsigned int len)
+{
+	unsigned long hash = init_name_hash();
+
+	while (len--)
+		hash = partial_name_hash(tolower(*name++), hash);
+	return end_name_hash(hash);
+}
+
+static inline void qstr_init(struct qstr *q, const char *name)
+{
+	q->name = name;
+	q->len = strlen(q->name);
+	q->hash = full_name_case_hash(q->name, q->len);
+}
+
+static inline int qstr_copy(const struct qstr *src, struct qstr *dest)
+{
+	dest->name = kstrdup(src->name, GFP_KERNEL);
+	dest->hash_len = src->hash_len;
+	return !!dest->name;
+}
+
+
+static appid_t __get_appid(const struct qstr *key)
+{
+	struct hashtable_entry *hash_cur;
+	unsigned int hash = key->hash;
+	appid_t ret_id;
+
+	rcu_read_lock();
+	hash_for_each_possible_rcu(package_to_appid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key)) {
+			ret_id = atomic_read(&hash_cur->value);
+			rcu_read_unlock();
+			return ret_id;
+		}
+	}
+	rcu_read_unlock();
+	return 0;
+}
+
+appid_t get_appid(const char *key)
+{
+	struct qstr q;
+
+	qstr_init(&q, key);
+	return __get_appid(&q);
+}
+
+static appid_t __get_ext_gid(const struct qstr *key)
+{
+	struct hashtable_entry *hash_cur;
+	unsigned int hash = key->hash;
+	appid_t ret_id;
+
+	rcu_read_lock();
+	hash_for_each_possible_rcu(ext_to_groupid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key)) {
+			ret_id = atomic_read(&hash_cur->value);
+			rcu_read_unlock();
+			return ret_id;
+		}
+	}
+	rcu_read_unlock();
+	return 0;
+}
+
+appid_t get_ext_gid(const char *key)
+{
+	struct qstr q;
+
+	qstr_init(&q, key);
+	return __get_ext_gid(&q);
+}
+
+static appid_t __is_excluded(const struct qstr *app_name, userid_t user)
+{
+	struct hashtable_entry *hash_cur;
+	unsigned int hash = app_name->hash;
+
+	rcu_read_lock();
+	hash_for_each_possible_rcu(package_to_userid, hash_cur, hlist, hash) {
+		if (atomic_read(&hash_cur->value) == user &&
+				qstr_case_eq(app_name, &hash_cur->key)) {
+			rcu_read_unlock();
+			return 1;
+		}
+	}
+	rcu_read_unlock();
+	return 0;
+}
+
+appid_t is_excluded(const char *key, userid_t user)
+{
+	struct qstr q;
+	qstr_init(&q, key);
+	return __is_excluded(&q, user);
+}
+
+/* Kernel has already enforced everything we returned through
+ * derive_permissions_locked(), so this is used to lock down access
+ * even further, such as enforcing that apps hold sdcard_rw.
+ */
+int check_caller_access_to_name(struct inode *parent_node, const struct qstr *name)
+{
+	struct qstr q_autorun = QSTR_LITERAL("autorun.inf");
+	struct qstr q__android_secure = QSTR_LITERAL(".android_secure");
+	struct qstr q_android_secure = QSTR_LITERAL("android_secure");
+
+	/* Always block security-sensitive files at root */
+	if (parent_node && SDCARDFS_I(parent_node)->data->perm == PERM_ROOT) {
+		if (qstr_case_eq(name, &q_autorun)
+			|| qstr_case_eq(name, &q__android_secure)
+			|| qstr_case_eq(name, &q_android_secure)) {
+			return 0;
+		}
+	}
+
+	/* Root always has access; access for any other UIDs should always
+	 * be controlled through packages.list.
+	 */
+	if (from_kuid(&init_user_ns, current_fsuid()) == 0)
+		return 1;
+
+	/* No extra permissions to enforce */
+	return 1;
+}
+
+static struct hashtable_entry *alloc_hashtable_entry(const struct qstr *key,
+		appid_t value)
+{
+	struct hashtable_entry *ret = kmem_cache_alloc(hashtable_entry_cachep,
+			GFP_KERNEL);
+	if (!ret)
+		return NULL;
+	INIT_HLIST_NODE(&ret->dlist);
+	INIT_HLIST_NODE(&ret->hlist);
+
+	if (!qstr_copy(key, &ret->key)) {
+		kmem_cache_free(hashtable_entry_cachep, ret);
+		return NULL;
+	}
+
+	atomic_set(&ret->value, value);
+	return ret;
+}
+
+static int insert_packagelist_appid_entry_locked(const struct qstr *key, appid_t value)
+{
+	struct hashtable_entry *hash_cur;
+	struct hashtable_entry *new_entry;
+	unsigned int hash = key->hash;
+
+	hash_for_each_possible_rcu(package_to_appid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key)) {
+			atomic_set(&hash_cur->value, value);
+			return 0;
+		}
+	}
+	new_entry = alloc_hashtable_entry(key, value);
+	if (!new_entry)
+		return -ENOMEM;
+	hash_add_rcu(package_to_appid, &new_entry->hlist, hash);
+	return 0;
+}
+
+static int insert_ext_gid_entry_locked(const struct qstr *key, appid_t value)
+{
+	struct hashtable_entry *hash_cur;
+	struct hashtable_entry *new_entry;
+	unsigned int hash = key->hash;
+
+	/* An extension can only belong to one gid */
+	hash_for_each_possible_rcu(ext_to_groupid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key))
+			return -EINVAL;
+	}
+	new_entry = alloc_hashtable_entry(key, value);
+	if (!new_entry)
+		return -ENOMEM;
+	hash_add_rcu(ext_to_groupid, &new_entry->hlist, hash);
+	return 0;
+}
+
+static int insert_userid_exclude_entry_locked(const struct qstr *key, userid_t value)
+{
+	struct hashtable_entry *hash_cur;
+	struct hashtable_entry *new_entry;
+	unsigned int hash = key->hash;
+
+	/* Only insert if not already present */
+	hash_for_each_possible_rcu(package_to_userid, hash_cur, hlist, hash) {
+		if (atomic_read(&hash_cur->value) == value &&
+				qstr_case_eq(key, &hash_cur->key))
+			return 0;
+	}
+	new_entry = alloc_hashtable_entry(key, value);
+	if (!new_entry)
+		return -ENOMEM;
+	hash_add_rcu(package_to_userid, &new_entry->hlist, hash);
+	return 0;
+}
+
+static void fixup_all_perms_name(const struct qstr *key)
+{
+	struct sdcardfs_sb_info *sbinfo;
+	struct limit_search limit = {
+		.flags = BY_NAME,
+		.name = QSTR_INIT(key->name, key->len),
+	};
+	list_for_each_entry(sbinfo, &sdcardfs_super_list, list) {
+		if (sbinfo_has_sdcard_magic(sbinfo))
+			fixup_perms_recursive(sbinfo->sb->s_root, &limit);
+	}
+}
+
+static void fixup_all_perms_name_userid(const struct qstr *key, userid_t userid)
+{
+	struct sdcardfs_sb_info *sbinfo;
+	struct limit_search limit = {
+		.flags = BY_NAME | BY_USERID,
+		.name = QSTR_INIT(key->name, key->len),
+		.userid = userid,
+	};
+	list_for_each_entry(sbinfo, &sdcardfs_super_list, list) {
+		if (sbinfo_has_sdcard_magic(sbinfo))
+			fixup_perms_recursive(sbinfo->sb->s_root, &limit);
+	}
+}
+
+static void fixup_all_perms_userid(userid_t userid)
+{
+	struct sdcardfs_sb_info *sbinfo;
+	struct limit_search limit = {
+		.flags = BY_USERID,
+		.userid = userid,
+	};
+	list_for_each_entry(sbinfo, &sdcardfs_super_list, list) {
+		if (sbinfo_has_sdcard_magic(sbinfo))
+			fixup_perms_recursive(sbinfo->sb->s_root, &limit);
+	}
+}
+
+static int insert_packagelist_entry(const struct qstr *key, appid_t value)
+{
+	int err;
+
+	mutex_lock(&sdcardfs_super_list_lock);
+	err = insert_packagelist_appid_entry_locked(key, value);
+	if (!err)
+		fixup_all_perms_name(key);
+	mutex_unlock(&sdcardfs_super_list_lock);
+
+	return err;
+}
+
+static int insert_ext_gid_entry(const struct qstr *key, appid_t value)
+{
+	int err;
+
+	mutex_lock(&sdcardfs_super_list_lock);
+	err = insert_ext_gid_entry_locked(key, value);
+	mutex_unlock(&sdcardfs_super_list_lock);
+
+	return err;
+}
+
+static int insert_userid_exclude_entry(const struct qstr *key, userid_t value)
+{
+	int err;
+
+	mutex_lock(&sdcardfs_super_list_lock);
+	err = insert_userid_exclude_entry_locked(key, value);
+	if (!err)
+		fixup_all_perms_name_userid(key, value);
+	mutex_unlock(&sdcardfs_super_list_lock);
+
+	return err;
+}
+
+static void free_hashtable_entry(struct hashtable_entry *entry)
+{
+	kfree(entry->key.name);
+	kmem_cache_free(hashtable_entry_cachep, entry);
+}
+
+static void remove_packagelist_entry_locked(const struct qstr *key)
+{
+	struct hashtable_entry *hash_cur;
+	unsigned int hash = key->hash;
+	struct hlist_node *h_t;
+	HLIST_HEAD(free_list);
+
+	hash_for_each_possible_rcu(package_to_userid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key)) {
+			hash_del_rcu(&hash_cur->hlist);
+			hlist_add_head(&hash_cur->dlist, &free_list);
+		}
+	}
+	hash_for_each_possible_rcu(package_to_appid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key)) {
+			hash_del_rcu(&hash_cur->hlist);
+			hlist_add_head(&hash_cur->dlist, &free_list);
+			break;
+		}
+	}
+	synchronize_rcu();
+	hlist_for_each_entry_safe(hash_cur, h_t, &free_list, dlist)
+		free_hashtable_entry(hash_cur);
+}
+
+static void remove_packagelist_entry(const struct qstr *key)
+{
+	mutex_lock(&sdcardfs_super_list_lock);
+	remove_packagelist_entry_locked(key);
+	fixup_all_perms_name(key);
+	mutex_unlock(&sdcardfs_super_list_lock);
+}
+
+static void remove_ext_gid_entry_locked(const struct qstr *key, gid_t group)
+{
+	struct hashtable_entry *hash_cur;
+	unsigned int hash = key->hash;
+
+	hash_for_each_possible_rcu(ext_to_groupid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key) && atomic_read(&hash_cur->value) == group) {
+			hash_del_rcu(&hash_cur->hlist);
+			synchronize_rcu();
+			free_hashtable_entry(hash_cur);
+			break;
+		}
+	}
+}
+
+static void remove_ext_gid_entry(const struct qstr *key, gid_t group)
+{
+	mutex_lock(&sdcardfs_super_list_lock);
+	remove_ext_gid_entry_locked(key, group);
+	mutex_unlock(&sdcardfs_super_list_lock);
+}
+
+static void remove_userid_all_entry_locked(userid_t userid)
+{
+	struct hashtable_entry *hash_cur;
+	struct hlist_node *h_t;
+	HLIST_HEAD(free_list);
+	int i;
+
+	hash_for_each_rcu(package_to_userid, i, hash_cur, hlist) {
+		if (atomic_read(&hash_cur->value) == userid) {
+			hash_del_rcu(&hash_cur->hlist);
+			hlist_add_head(&hash_cur->dlist, &free_list);
+		}
+	}
+	synchronize_rcu();
+	hlist_for_each_entry_safe(hash_cur, h_t, &free_list, dlist) {
+		free_hashtable_entry(hash_cur);
+	}
+}
+
+static void remove_userid_all_entry(userid_t userid)
+{
+	mutex_lock(&sdcardfs_super_list_lock);
+	remove_userid_all_entry_locked(userid);
+	fixup_all_perms_userid(userid);
+	mutex_unlock(&sdcardfs_super_list_lock);
+}
+
+static void remove_userid_exclude_entry_locked(const struct qstr *key, userid_t userid)
+{
+	struct hashtable_entry *hash_cur;
+	unsigned int hash = key->hash;
+
+	hash_for_each_possible_rcu(package_to_userid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(key, &hash_cur->key) &&
+				atomic_read(&hash_cur->value) == userid) {
+			hash_del_rcu(&hash_cur->hlist);
+			synchronize_rcu();
+			free_hashtable_entry(hash_cur);
+			break;
+		}
+	}
+}
+
+static void remove_userid_exclude_entry(const struct qstr *key, userid_t userid)
+{
+	mutex_lock(&sdcardfs_super_list_lock);
+	remove_userid_exclude_entry_locked(key, userid);
+	fixup_all_perms_name_userid(key, userid);
+	mutex_unlock(&sdcardfs_super_list_lock);
+}
+
+static void packagelist_destroy(void)
+{
+	struct hashtable_entry *hash_cur;
+	struct hlist_node *h_t;
+	HLIST_HEAD(free_list);
+	int i;
+
+	mutex_lock(&sdcardfs_super_list_lock);
+	hash_for_each_rcu(package_to_appid, i, hash_cur, hlist) {
+		hash_del_rcu(&hash_cur->hlist);
+		hlist_add_head(&hash_cur->dlist, &free_list);
+	}
+	hash_for_each_rcu(package_to_userid, i, hash_cur, hlist) {
+		hash_del_rcu(&hash_cur->hlist);
+		hlist_add_head(&hash_cur->dlist, &free_list);
+	}
+	synchronize_rcu();
+	hlist_for_each_entry_safe(hash_cur, h_t, &free_list, dlist)
+		free_hashtable_entry(hash_cur);
+	mutex_unlock(&sdcardfs_super_list_lock);
+	pr_info("sdcardfs: destroyed packagelist pkgld\n");
+}
+
+#define SDCARDFS_CONFIGFS_ATTR(_pfx, _name)			\
+static struct configfs_attribute _pfx##attr_##_name = {	\
+	.ca_name	= __stringify(_name),		\
+	.ca_mode	= S_IRUGO | S_IWUGO,		\
+	.ca_owner	= THIS_MODULE,			\
+	.show		= _pfx##_name##_show,		\
+	.store		= _pfx##_name##_store,		\
+}
+
+#define SDCARDFS_CONFIGFS_ATTR_RO(_pfx, _name)			\
+static struct configfs_attribute _pfx##attr_##_name = {	\
+	.ca_name	= __stringify(_name),		\
+	.ca_mode	= S_IRUGO,			\
+	.ca_owner	= THIS_MODULE,			\
+	.show		= _pfx##_name##_show,		\
+}
+
+#define SDCARDFS_CONFIGFS_ATTR_WO(_pfx, _name)			\
+static struct configfs_attribute _pfx##attr_##_name = {	\
+	.ca_name	= __stringify(_name),		\
+	.ca_mode	= S_IWUGO,			\
+	.ca_owner	= THIS_MODULE,			\
+	.store		= _pfx##_name##_store,		\
+}
+
+struct package_details {
+	struct config_item item;
+	struct qstr name;
+};
+
+static inline struct package_details *to_package_details(struct config_item *item)
+{
+	return item ? container_of(item, struct package_details, item) : NULL;
+}
+
+static ssize_t package_details_appid_show(struct config_item *item, char *page)
+{
+	return scnprintf(page, PAGE_SIZE, "%u\n", __get_appid(&to_package_details(item)->name));
+}
+
+static ssize_t package_details_appid_store(struct config_item *item,
+				       const char *page, size_t count)
+{
+	unsigned int tmp;
+	int ret;
+
+	ret = kstrtouint(page, 10, &tmp);
+	if (ret)
+		return ret;
+
+	ret = insert_packagelist_entry(&to_package_details(item)->name, tmp);
+
+	if (ret)
+		return ret;
+
+	return count;
+}
+
+static ssize_t package_details_excluded_userids_show(struct config_item *item,
+				      char *page)
+{
+	struct package_details *package_details = to_package_details(item);
+	struct hashtable_entry *hash_cur;
+	unsigned int hash = package_details->name.hash;
+	int count = 0;
+
+	rcu_read_lock();
+	hash_for_each_possible_rcu(package_to_userid, hash_cur, hlist, hash) {
+		if (qstr_case_eq(&package_details->name, &hash_cur->key))
+			count += scnprintf(page + count, PAGE_SIZE - count,
+					"%d ", atomic_read(&hash_cur->value));
+	}
+	rcu_read_unlock();
+	if (count)
+		count--;
+	count += scnprintf(page + count, PAGE_SIZE - count, "\n");
+	return count;
+}
+
+static ssize_t package_details_excluded_userids_store(struct config_item *item,
+				       const char *page, size_t count)
+{
+	unsigned int tmp;
+	int ret;
+
+	ret = kstrtouint(page, 10, &tmp);
+	if (ret)
+		return ret;
+
+	ret = insert_userid_exclude_entry(&to_package_details(item)->name, tmp);
+
+	if (ret)
+		return ret;
+
+	return count;
+}
+
+static ssize_t package_details_clear_userid_store(struct config_item *item,
+				       const char *page, size_t count)
+{
+	unsigned int tmp;
+	int ret;
+
+	ret = kstrtouint(page, 10, &tmp);
+	if (ret)
+		return ret;
+	remove_userid_exclude_entry(&to_package_details(item)->name, tmp);
+	return count;
+}
+
+static void package_details_release(struct config_item *item)
+{
+	struct package_details *package_details = to_package_details(item);
+
+	pr_info("sdcardfs: removing %s\n", package_details->name.name);
+	remove_packagelist_entry(&package_details->name);
+	kfree(package_details->name.name);
+	kfree(package_details);
+}
+
+SDCARDFS_CONFIGFS_ATTR(package_details_, appid);
+SDCARDFS_CONFIGFS_ATTR(package_details_, excluded_userids);
+SDCARDFS_CONFIGFS_ATTR_WO(package_details_, clear_userid);
+
+static struct configfs_attribute *package_details_attrs[] = {
+	&package_details_attr_appid,
+	&package_details_attr_excluded_userids,
+	&package_details_attr_clear_userid,
+	NULL,
+};
+
+static struct configfs_item_operations package_details_item_ops = {
+	.release = package_details_release,
+};
+
+static struct config_item_type package_appid_type = {
+	.ct_item_ops	= &package_details_item_ops,
+	.ct_attrs	= package_details_attrs,
+	.ct_owner	= THIS_MODULE,
+};
+
+struct extensions_value {
+	struct config_group group;
+	unsigned int num;
+};
+
+struct extension_details {
+	struct config_item item;
+	struct qstr name;
+	unsigned int num;
+};
+
+static inline struct extensions_value *to_extensions_value(struct config_item *item)
+{
+	return item ? container_of(to_config_group(item), struct extensions_value, group) : NULL;
+}
+
+static inline struct extension_details *to_extension_details(struct config_item *item)
+{
+	return item ? container_of(item, struct extension_details, item) : NULL;
+}
+
+static void extension_details_release(struct config_item *item)
+{
+	struct extension_details *extension_details = to_extension_details(item);
+
+	pr_info("sdcardfs: No longer mapping %s files to gid %d\n",
+			extension_details->name.name, extension_details->num);
+	remove_ext_gid_entry(&extension_details->name, extension_details->num);
+	kfree(extension_details->name.name);
+	kfree(extension_details);
+}
+
+static struct configfs_item_operations extension_details_item_ops = {
+	.release = extension_details_release,
+};
+
+static struct config_item_type extension_details_type = {
+	.ct_item_ops = &extension_details_item_ops,
+	.ct_owner = THIS_MODULE,
+};
+
+static struct config_item *extension_details_make_item(struct config_group *group, const char *name)
+{
+	struct extensions_value *extensions_value = to_extensions_value(&group->cg_item);
+	struct extension_details *extension_details = kzalloc(sizeof(struct extension_details), GFP_KERNEL);
+	const char *tmp;
+	int ret;
+
+	if (!extension_details)
+		return ERR_PTR(-ENOMEM);
+
+	tmp = kstrdup(name, GFP_KERNEL);
+	if (!tmp) {
+		kfree(extension_details);
+		return ERR_PTR(-ENOMEM);
+	}
+	qstr_init(&extension_details->name, tmp);
+	ret = insert_ext_gid_entry(&extension_details->name, extensions_value->num);
+
+	if (ret) {
+		kfree(extension_details->name.name);
+		kfree(extension_details);
+		return ERR_PTR(ret);
+	}
+	config_item_init_type_name(&extension_details->item, name, &extension_details_type);
+
+	return &extension_details->item;
+}
+
+static struct configfs_group_operations extensions_value_group_ops = {
+	.make_item = extension_details_make_item,
+};
+
+static struct config_item_type extensions_name_type = {
+	.ct_group_ops	= &extensions_value_group_ops,
+	.ct_owner	= THIS_MODULE,
+};
+
+static struct config_group *extensions_make_group(struct config_group *group, const char *name)
+{
+	struct extensions_value *extensions_value;
+	unsigned int tmp;
+	int ret;
+
+	extensions_value = kzalloc(sizeof(struct extensions_value), GFP_KERNEL);
+	if (!extensions_value)
+		return ERR_PTR(-ENOMEM);
+	ret = kstrtouint(name, 10, &tmp);
+	if (ret) {
+		kfree(extensions_value);
+		return ERR_PTR(ret);
+	}
+
+	extensions_value->num = tmp;
+	config_group_init_type_name(&extensions_value->group, name,
+						&extensions_name_type);
+	return &extensions_value->group;
+}
+
+static void extensions_drop_group(struct config_group *group, struct config_item *item)
+{
+	struct extensions_value *value = to_extensions_value(item);
+
+	pr_info("sdcardfs: No longer mapping any files to gid %d\n", value->num);
+	kfree(value);
+}
+
+static struct configfs_group_operations extensions_group_ops = {
+	.make_group	= extensions_make_group,
+	.drop_item	= extensions_drop_group,
+};
+
+static struct config_item_type extensions_type = {
+	.ct_group_ops	= &extensions_group_ops,
+	.ct_owner	= THIS_MODULE,
+};
+
+struct config_group extension_group = {
+	.cg_item = {
+		.ci_namebuf = "extensions",
+		.ci_type = &extensions_type,
+	},
+};
+
+static struct config_item *packages_make_item(struct config_group *group, const char *name)
+{
+	struct package_details *package_details;
+	const char *tmp;
+
+	package_details = kzalloc(sizeof(struct package_details), GFP_KERNEL);
+	if (!package_details)
+		return ERR_PTR(-ENOMEM);
+	tmp = kstrdup(name, GFP_KERNEL);
+	if (!tmp) {
+		kfree(package_details);
+		return ERR_PTR(-ENOMEM);
+	}
+	qstr_init(&package_details->name, tmp);
+	config_item_init_type_name(&package_details->item, name,
+						&package_appid_type);
+
+	return &package_details->item;
+}
+
+static ssize_t packages_list_show(struct config_item *item, char *page)
+{
+	struct hashtable_entry *hash_cur_app;
+	struct hashtable_entry *hash_cur_user;
+	int i;
+	int count = 0, written = 0;
+	const char errormsg[] = "<truncated>\n";
+	unsigned int hash;
+
+	rcu_read_lock();
+	hash_for_each_rcu(package_to_appid, i, hash_cur_app, hlist) {
+		written = scnprintf(page + count, PAGE_SIZE - sizeof(errormsg) - count, "%s %d\n",
+					hash_cur_app->key.name, atomic_read(&hash_cur_app->value));
+		hash = hash_cur_app->key.hash;
+		hash_for_each_possible_rcu(package_to_userid, hash_cur_user, hlist, hash) {
+			if (qstr_case_eq(&hash_cur_app->key, &hash_cur_user->key)) {
+				written += scnprintf(page + count + written - 1,
+					PAGE_SIZE - sizeof(errormsg) - count - written + 1,
+					" %d\n", atomic_read(&hash_cur_user->value)) - 1;
+			}
+		}
+		if (count + written == PAGE_SIZE - sizeof(errormsg) - 1) {
+			count += scnprintf(page + count, PAGE_SIZE - count, errormsg);
+			break;
+		}
+		count += written;
+	}
+	rcu_read_unlock();
+
+	return count;
+}
+
+static ssize_t packages_remove_userid_store(struct config_item *item,
+				       const char *page, size_t count)
+{
+	unsigned int tmp;
+	int ret;
+
+	ret = kstrtouint(page, 10, &tmp);
+	if (ret)
+		return ret;
+	remove_userid_all_entry(tmp);
+	return count;
+}
+
+static struct configfs_attribute packages_attr_packages_gid_list = {
+	.ca_name	= "packages_gid.list",
+	.ca_mode	= S_IRUGO,
+	.ca_owner	= THIS_MODULE,
+	.show		= packages_list_show,
+};
+
+SDCARDFS_CONFIGFS_ATTR_WO(packages_, remove_userid);
+
+static struct configfs_attribute *packages_attrs[] = {
+	&packages_attr_packages_gid_list,
+	&packages_attr_remove_userid,
+	NULL,
+};
+
+/*
+ * Note that, since no extra work is required on ->drop_item(),
+ * no ->drop_item() is provided.
+ */
+static struct configfs_group_operations packages_group_ops = {
+	.make_item	= packages_make_item,
+};
+
+static struct config_item_type packages_type = {
+	.ct_group_ops	= &packages_group_ops,
+	.ct_attrs	= packages_attrs,
+	.ct_owner	= THIS_MODULE,
+};
+
+struct config_group *sd_default_groups[] = {
+	&extension_group,
+	NULL,
+};
+
+static struct configfs_subsystem sdcardfs_packages = {
+	.su_group = {
+		.cg_item = {
+			.ci_namebuf = "sdcardfs",
+			.ci_type = &packages_type,
+		},
+		.default_groups = sd_default_groups,
+	},
+};
+
+static int configfs_sdcardfs_init(void)
+{
+	int ret, i;
+	struct configfs_subsystem *subsys = &sdcardfs_packages;
+
+	for (i = 0; sd_default_groups[i]; i++)
+		config_group_init(sd_default_groups[i]);
+	config_group_init(&subsys->su_group);
+	mutex_init(&subsys->su_mutex);
+	ret = configfs_register_subsystem(subsys);
+	if (ret) {
+		pr_err("Error %d while registering subsystem %s\n",
+		       ret,
+		       subsys->su_group.cg_item.ci_namebuf);
+	}
+	return ret;
+}
+
+static void configfs_sdcardfs_exit(void)
+{
+	configfs_unregister_subsystem(&sdcardfs_packages);
+}
+
+int packagelist_init(void)
+{
+	hashtable_entry_cachep =
+		kmem_cache_create("packagelist_hashtable_entry",
+					sizeof(struct hashtable_entry), 0, 0, NULL);
+	if (!hashtable_entry_cachep) {
+		pr_err("sdcardfs: failed creating pkgl_hashtable entry slab cache\n");
+		return -ENOMEM;
+	}
+
+	configfs_sdcardfs_init();
+	return 0;
+}
+
+void packagelist_exit(void)
+{
+	configfs_sdcardfs_exit();
+	packagelist_destroy();
+	kmem_cache_destroy(hashtable_entry_cachep);
+}
diff --git a/fs/sdcardfs/sdcardfs.h b/fs/sdcardfs/sdcardfs.h
new file mode 100644
index 000000000000..88b92b2f1872
--- /dev/null
+++ b/fs/sdcardfs/sdcardfs.h
@@ -0,0 +1,664 @@
+/*
+ * fs/sdcardfs/sdcardfs.h
+ *
+ * The sdcardfs v2.0
+ *   This file system replaces the sdcard daemon on Android
+ *   On version 2.0, some of the daemon functions have been ported
+ *   to support the multi-user concepts of Android 4.4
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#ifndef _SDCARDFS_H_
+#define _SDCARDFS_H_
+
+#include <linux/dcache.h>
+#include <linux/file.h>
+#include <linux/fs.h>
+#include <linux/aio.h>
+#include <linux/kref.h>
+#include <linux/mm.h>
+#include <linux/mount.h>
+#include <linux/namei.h>
+#include <linux/seq_file.h>
+#include <linux/statfs.h>
+#include <linux/fs_stack.h>
+#include <linux/magic.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+#include <linux/sched.h>
+#include <linux/types.h>
+#include <linux/security.h>
+#include <linux/string.h>
+#include <linux/list.h>
+#include "multiuser.h"
+
+/* the file system name */
+#define SDCARDFS_NAME "sdcardfs"
+
+/* sdcardfs root inode number */
+#define SDCARDFS_ROOT_INO     1
+
+/* useful for tracking code reachability */
+#define UDBG pr_default("DBG:%s:%s:%d\n", __FILE__, __func__, __LINE__)
+
+#define SDCARDFS_DIRENT_SIZE 256
+
+/* temporary static uid settings for development */
+#define AID_ROOT             0	/* uid for accessing /mnt/sdcard & extSdcard */
+#define AID_MEDIA_RW      1023	/* internal media storage write access */
+
+#define AID_SDCARD_RW     1015	/* external storage write access */
+#define AID_SDCARD_R      1028	/* external storage read access */
+#define AID_SDCARD_PICS   1033	/* external storage photos access */
+#define AID_SDCARD_AV     1034	/* external storage audio/video access */
+#define AID_SDCARD_ALL    1035	/* access all users external storage */
+#define AID_MEDIA_OBB     1059  /* obb files */
+
+#define AID_SDCARD_IMAGE  1057
+
+#define AID_PACKAGE_INFO  1027
+
+
+/*
+ * Permissions are handled by our permission function.
+ * We don't want anyone who happens to look at our inode value to prematurely
+ * block access, so store more permissive values. These are probably never
+ * used.
+ */
+#define fixup_tmp_permissions(x)	\
+	do {						\
+		(x)->i_uid = make_kuid(&init_user_ns,	\
+				SDCARDFS_I(x)->data->d_uid);	\
+		(x)->i_gid = make_kgid(&init_user_ns, AID_SDCARD_RW);	\
+		(x)->i_mode = ((x)->i_mode & S_IFMT) | 0775;\
+	} while (0)
+
+/* OVERRIDE_CRED() and REVERT_CRED()
+ *	OVERRIDE_CRED()
+ *		backup original task->cred
+ *		and modifies task->cred->fsuid/fsgid to specified value.
+ *	REVERT_CRED()
+ *		restore original task->cred->fsuid/fsgid.
+ * These two macro should be used in pair, and OVERRIDE_CRED() should be
+ * placed at the beginning of a function, right after variable declaration.
+ */
+#define OVERRIDE_CRED(sdcardfs_sbi, saved_cred, info)		\
+	do {	\
+		saved_cred = override_fsids(sdcardfs_sbi, info->data);	\
+		if (!saved_cred)	\
+			return -ENOMEM;	\
+	} while (0)
+
+#define OVERRIDE_CRED_PTR(sdcardfs_sbi, saved_cred, info)	\
+	do {	\
+		saved_cred = override_fsids(sdcardfs_sbi, info->data);	\
+		if (!saved_cred)	\
+			return ERR_PTR(-ENOMEM);	\
+	} while (0)
+
+#define REVERT_CRED(saved_cred)	revert_fsids(saved_cred)
+
+/* Android 5.0 support */
+
+/* Permission mode for a specific node. Controls how file permissions
+ * are derived for children nodes.
+ */
+typedef enum {
+	/* Nothing special; this node should just inherit from its parent. */
+	PERM_INHERIT,
+	/* This node is one level above a normal root; used for legacy layouts
+	 * which use the first level to represent user_id.
+	 */
+	PERM_PRE_ROOT,
+	/* This node is "/" */
+	PERM_ROOT,
+	/* This node is "/Android" */
+	PERM_ANDROID,
+	/* This node is "/Android/data" */
+	PERM_ANDROID_DATA,
+	/* This node is "/Android/obb" */
+	PERM_ANDROID_OBB,
+	/* This node is "/Android/media" */
+	PERM_ANDROID_MEDIA,
+	/* This node is "/Android/[data|media|obb]/[package]" */
+	PERM_ANDROID_PACKAGE,
+	/* This node is "/Android/[data|media|obb]/[package]/cache" */
+	PERM_ANDROID_PACKAGE_CACHE,
+} perm_t;
+
+struct sdcardfs_sb_info;
+struct sdcardfs_mount_options;
+struct sdcardfs_inode_info;
+struct sdcardfs_inode_data;
+
+/* Do not directly use this function. Use OVERRIDE_CRED() instead. */
+const struct cred *override_fsids(struct sdcardfs_sb_info *sbi,
+			struct sdcardfs_inode_data *data);
+/* Do not directly use this function, use REVERT_CRED() instead. */
+void revert_fsids(const struct cred *old_cred);
+
+/* operations vectors defined in specific files */
+extern const struct file_operations sdcardfs_main_fops;
+extern const struct file_operations sdcardfs_dir_fops;
+extern const struct inode_operations sdcardfs_main_iops;
+extern const struct inode_operations sdcardfs_dir_iops;
+extern const struct inode_operations sdcardfs_symlink_iops;
+extern const struct super_operations sdcardfs_sops;
+extern const struct dentry_operations sdcardfs_ci_dops;
+extern const struct address_space_operations sdcardfs_aops, sdcardfs_dummy_aops;
+extern const struct vm_operations_struct sdcardfs_vm_ops;
+
+extern int sdcardfs_init_inode_cache(void);
+extern void sdcardfs_destroy_inode_cache(void);
+extern int sdcardfs_init_dentry_cache(void);
+extern void sdcardfs_destroy_dentry_cache(void);
+extern int new_dentry_private_data(struct dentry *dentry);
+extern void free_dentry_private_data(struct dentry *dentry);
+extern struct dentry *sdcardfs_lookup(struct inode *dir, struct dentry *dentry,
+				unsigned int flags);
+extern struct inode *sdcardfs_iget(struct super_block *sb,
+				 struct inode *lower_inode, userid_t id);
+extern int sdcardfs_interpose(struct dentry *dentry, struct super_block *sb,
+			    struct path *lower_path, userid_t id);
+
+/* file private data */
+struct sdcardfs_file_info {
+	struct file *lower_file;
+	const struct vm_operations_struct *lower_vm_ops;
+};
+
+struct sdcardfs_inode_data {
+	struct kref refcount;
+	bool abandoned;
+
+	perm_t perm;
+	userid_t userid;
+	uid_t d_uid;
+	bool under_android;
+	bool under_cache;
+	bool under_obb;
+};
+
+/* sdcardfs inode data in memory */
+struct sdcardfs_inode_info {
+	struct inode *lower_inode;
+	/* state derived based on current position in hierarchy */
+	struct sdcardfs_inode_data *data;
+
+	/* top folder for ownership */
+	struct sdcardfs_inode_data *top_data;
+
+	struct inode vfs_inode;
+};
+
+
+/* sdcardfs dentry data in memory */
+struct sdcardfs_dentry_info {
+	spinlock_t lock;	/* protects lower_path */
+	struct path lower_path;
+	struct path orig_path;
+};
+
+struct sdcardfs_mount_options {
+	uid_t fs_low_uid;
+	gid_t fs_low_gid;
+	userid_t fs_user_id;
+	bool multiuser;
+	bool gid_derivation;
+	unsigned int reserved_mb;
+};
+
+struct sdcardfs_vfsmount_options {
+	gid_t gid;
+	mode_t mask;
+};
+
+extern int parse_options_remount(struct super_block *sb, char *options, int silent,
+		struct sdcardfs_vfsmount_options *vfsopts);
+
+/* sdcardfs super-block data in memory */
+struct sdcardfs_sb_info {
+	struct super_block *sb;
+	struct super_block *lower_sb;
+	/* derived perm policy : some of options have been added
+	 * to sdcardfs_mount_options (Android 4.4 support)
+	 */
+	struct sdcardfs_mount_options options;
+	spinlock_t lock;	/* protects obbpath */
+	char *obbpath_s;
+	struct path obbpath;
+	void *pkgl_id;
+	struct list_head list;
+};
+
+/*
+ * inode to private data
+ *
+ * Since we use containers and the struct inode is _inside_ the
+ * sdcardfs_inode_info structure, SDCARDFS_I will always (given a non-NULL
+ * inode pointer), return a valid non-NULL pointer.
+ */
+static inline struct sdcardfs_inode_info *SDCARDFS_I(const struct inode *inode)
+{
+	return container_of(inode, struct sdcardfs_inode_info, vfs_inode);
+}
+
+/* dentry to private data */
+#define SDCARDFS_D(dent) ((struct sdcardfs_dentry_info *)(dent)->d_fsdata)
+
+/* superblock to private data */
+#define SDCARDFS_SB(super) ((struct sdcardfs_sb_info *)(super)->s_fs_info)
+
+/* file to private Data */
+#define SDCARDFS_F(file) ((struct sdcardfs_file_info *)((file)->private_data))
+
+/* file to lower file */
+static inline struct file *sdcardfs_lower_file(const struct file *f)
+{
+	return SDCARDFS_F(f)->lower_file;
+}
+
+static inline void sdcardfs_set_lower_file(struct file *f, struct file *val)
+{
+	SDCARDFS_F(f)->lower_file = val;
+}
+
+/* inode to lower inode. */
+static inline struct inode *sdcardfs_lower_inode(const struct inode *i)
+{
+	return SDCARDFS_I(i)->lower_inode;
+}
+
+static inline void sdcardfs_set_lower_inode(struct inode *i, struct inode *val)
+{
+	SDCARDFS_I(i)->lower_inode = val;
+}
+
+/* superblock to lower superblock */
+static inline struct super_block *sdcardfs_lower_super(
+	const struct super_block *sb)
+{
+	return SDCARDFS_SB(sb)->lower_sb;
+}
+
+static inline void sdcardfs_set_lower_super(struct super_block *sb,
+					  struct super_block *val)
+{
+	SDCARDFS_SB(sb)->lower_sb = val;
+}
+
+/* path based (dentry/mnt) macros */
+static inline void pathcpy(struct path *dst, const struct path *src)
+{
+	dst->dentry = src->dentry;
+	dst->mnt = src->mnt;
+}
+
+/* sdcardfs_get_pname functions calls path_get()
+ * therefore, the caller must call "proper" path_put functions
+ */
+#define SDCARDFS_DENT_FUNC(pname) \
+static inline void sdcardfs_get_##pname(const struct dentry *dent, \
+					struct path *pname) \
+{ \
+	spin_lock(&SDCARDFS_D(dent)->lock); \
+	pathcpy(pname, &SDCARDFS_D(dent)->pname); \
+	path_get(pname); \
+	spin_unlock(&SDCARDFS_D(dent)->lock); \
+	return; \
+} \
+static inline void sdcardfs_put_##pname(const struct dentry *dent, \
+					struct path *pname) \
+{ \
+	path_put(pname); \
+	return; \
+} \
+static inline void sdcardfs_set_##pname(const struct dentry *dent, \
+					struct path *pname) \
+{ \
+	spin_lock(&SDCARDFS_D(dent)->lock); \
+	pathcpy(&SDCARDFS_D(dent)->pname, pname); \
+	spin_unlock(&SDCARDFS_D(dent)->lock); \
+	return; \
+} \
+static inline void sdcardfs_reset_##pname(const struct dentry *dent) \
+{ \
+	spin_lock(&SDCARDFS_D(dent)->lock); \
+	SDCARDFS_D(dent)->pname.dentry = NULL; \
+	SDCARDFS_D(dent)->pname.mnt = NULL; \
+	spin_unlock(&SDCARDFS_D(dent)->lock); \
+	return; \
+} \
+static inline void sdcardfs_put_reset_##pname(const struct dentry *dent) \
+{ \
+	struct path pname; \
+	spin_lock(&SDCARDFS_D(dent)->lock); \
+	if (SDCARDFS_D(dent)->pname.dentry) { \
+		pathcpy(&pname, &SDCARDFS_D(dent)->pname); \
+		SDCARDFS_D(dent)->pname.dentry = NULL; \
+		SDCARDFS_D(dent)->pname.mnt = NULL; \
+		spin_unlock(&SDCARDFS_D(dent)->lock); \
+		path_put(&pname); \
+	} else \
+		spin_unlock(&SDCARDFS_D(dent)->lock); \
+	return; \
+}
+
+SDCARDFS_DENT_FUNC(lower_path)
+SDCARDFS_DENT_FUNC(orig_path)
+
+static inline bool sbinfo_has_sdcard_magic(struct sdcardfs_sb_info *sbinfo)
+{
+	return sbinfo && sbinfo->sb
+			&& sbinfo->sb->s_magic == SDCARDFS_SUPER_MAGIC;
+}
+
+static inline struct sdcardfs_inode_data *data_get(
+		struct sdcardfs_inode_data *data)
+{
+	if (data)
+		kref_get(&data->refcount);
+	return data;
+}
+
+static inline struct sdcardfs_inode_data *top_data_get(
+		struct sdcardfs_inode_info *info)
+{
+	return data_get(info->top_data);
+}
+
+extern void data_release(struct kref *ref);
+
+static inline void data_put(struct sdcardfs_inode_data *data)
+{
+	kref_put(&data->refcount, data_release);
+}
+
+static inline void release_own_data(struct sdcardfs_inode_info *info)
+{
+	/*
+	 * This happens exactly once per inode. At this point, the inode that
+	 * originally held this data is about to be freed, and all references
+	 * to it are held as a top value, and will likely be released soon.
+	 */
+	info->data->abandoned = true;
+	data_put(info->data);
+}
+
+static inline void set_top(struct sdcardfs_inode_info *info,
+			struct sdcardfs_inode_data *top)
+{
+	struct sdcardfs_inode_data *old_top = info->top_data;
+
+	if (top)
+		data_get(top);
+	info->top_data = top;
+	if (old_top)
+		data_put(old_top);
+}
+
+static inline int get_gid(struct vfsmount *mnt,
+		struct sdcardfs_inode_data *data)
+{
+	struct sdcardfs_vfsmount_options *opts = mnt->data;
+
+	if (opts->gid == AID_SDCARD_RW)
+		/* As an optimization, certain trusted system components only run
+		 * as owner but operate across all users. Since we're now handing
+		 * out the sdcard_rw GID only to trusted apps, we're okay relaxing
+		 * the user boundary enforcement for the default view. The UIDs
+		 * assigned to app directories are still multiuser aware.
+		 */
+		return AID_SDCARD_RW;
+	else
+		return multiuser_get_uid(data->userid, opts->gid);
+}
+
+static inline int get_mode(struct vfsmount *mnt,
+		struct sdcardfs_inode_info *info,
+		struct sdcardfs_inode_data *data)
+{
+	int owner_mode;
+	int filtered_mode;
+	struct sdcardfs_vfsmount_options *opts = mnt->data;
+	int visible_mode = 0775 & ~opts->mask;
+
+
+	if (data->perm == PERM_PRE_ROOT) {
+		/* Top of multi-user view should always be visible to ensure
+		* secondary users can traverse inside.
+		*/
+		visible_mode = 0711;
+	} else if (data->under_android) {
+		/* Block "other" access to Android directories, since only apps
+		* belonging to a specific user should be in there; we still
+		* leave +x open for the default view.
+		*/
+		if (opts->gid == AID_SDCARD_RW)
+			visible_mode = visible_mode & ~0006;
+		else
+			visible_mode = visible_mode & ~0007;
+	}
+	owner_mode = info->lower_inode->i_mode & 0700;
+	filtered_mode = visible_mode & (owner_mode | (owner_mode >> 3) | (owner_mode >> 6));
+	return filtered_mode;
+}
+
+static inline int has_graft_path(const struct dentry *dent)
+{
+	int ret = 0;
+
+	spin_lock(&SDCARDFS_D(dent)->lock);
+	if (SDCARDFS_D(dent)->orig_path.dentry != NULL)
+		ret = 1;
+	spin_unlock(&SDCARDFS_D(dent)->lock);
+
+	return ret;
+}
+
+static inline void sdcardfs_get_real_lower(const struct dentry *dent,
+						struct path *real_lower)
+{
+	/* in case of a local obb dentry
+	 * the orig_path should be returned
+	 */
+	if (has_graft_path(dent))
+		sdcardfs_get_orig_path(dent, real_lower);
+	else
+		sdcardfs_get_lower_path(dent, real_lower);
+}
+
+static inline void sdcardfs_put_real_lower(const struct dentry *dent,
+						struct path *real_lower)
+{
+	if (has_graft_path(dent))
+		sdcardfs_put_orig_path(dent, real_lower);
+	else
+		sdcardfs_put_lower_path(dent, real_lower);
+}
+
+extern struct mutex sdcardfs_super_list_lock;
+extern struct list_head sdcardfs_super_list;
+
+/* for packagelist.c */
+extern appid_t get_appid(const char *app_name);
+extern appid_t get_ext_gid(const char *app_name);
+extern appid_t is_excluded(const char *app_name, userid_t userid);
+extern int check_caller_access_to_name(struct inode *parent_node, const struct qstr *name);
+extern int packagelist_init(void);
+extern void packagelist_exit(void);
+
+/* for derived_perm.c */
+#define BY_NAME		(1 << 0)
+#define BY_USERID	(1 << 1)
+struct limit_search {
+	unsigned int flags;
+	struct qstr name;
+	userid_t userid;
+};
+
+extern void setup_derived_state(struct inode *inode, perm_t perm,
+		userid_t userid, uid_t uid, bool under_android,
+		struct sdcardfs_inode_data *top);
+extern void get_derived_permission(struct dentry *parent, struct dentry *dentry);
+extern void get_derived_permission_new(struct dentry *parent, struct dentry *dentry, const struct qstr *name);
+extern void fixup_perms_recursive(struct dentry *dentry, struct limit_search *limit);
+
+extern void update_derived_permission_lock(struct dentry *dentry);
+void fixup_lower_ownership(struct dentry *dentry, const char *name);
+extern int need_graft_path(struct dentry *dentry);
+extern int is_base_obbpath(struct dentry *dentry);
+extern int is_obbpath_invalid(struct dentry *dentry);
+extern int setup_obb_dentry(struct dentry *dentry, struct path *lower_path);
+
+/* locking helpers */
+static inline struct dentry *lock_parent(struct dentry *dentry)
+{
+	struct dentry *dir = dget_parent(dentry);
+
+	mutex_lock_nested(&d_inode(dir)->i_mutex, I_MUTEX_PARENT);
+	return dir;
+}
+
+static inline void unlock_dir(struct dentry *dir)
+{
+	mutex_unlock(&d_inode(dir)->i_mutex);
+	dput(dir);
+}
+
+static inline int prepare_dir(const char *path_s, uid_t uid, gid_t gid, mode_t mode)
+{
+	int err;
+	struct dentry *dent;
+	struct iattr attrs;
+	struct path parent;
+
+	dent = kern_path_locked(path_s, &parent);
+	if (IS_ERR(dent)) {
+		err = PTR_ERR(dent);
+		if (err == -EEXIST)
+			err = 0;
+		goto out_unlock;
+	}
+
+	err = vfs_mkdir2(parent.mnt, d_inode(parent.dentry), dent, mode);
+	if (err) {
+		if (err == -EEXIST)
+			err = 0;
+		goto out_dput;
+	}
+
+	attrs.ia_uid = make_kuid(&init_user_ns, uid);
+	attrs.ia_gid = make_kgid(&init_user_ns, gid);
+	attrs.ia_valid = ATTR_UID | ATTR_GID;
+	mutex_lock(&d_inode(dent)->i_mutex);
+	notify_change2(parent.mnt, dent, &attrs, NULL);
+	mutex_unlock(&d_inode(dent)->i_mutex);
+
+out_dput:
+	dput(dent);
+
+out_unlock:
+	/* parent dentry locked by lookup_create */
+	mutex_unlock(&d_inode(parent.dentry)->i_mutex);
+	path_put(&parent);
+	return err;
+}
+
+/*
+ * Return 1, if a disk has enough free space, otherwise 0.
+ * We assume that any files can not be overwritten.
+ */
+static inline int check_min_free_space(struct dentry *dentry, size_t size, int dir)
+{
+	int err;
+	struct path lower_path;
+	struct kstatfs statfs;
+	u64 avail;
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+
+	if (sbi->options.reserved_mb) {
+		/* Get fs stat of lower filesystem. */
+		sdcardfs_get_lower_path(dentry, &lower_path);
+		err = vfs_statfs(&lower_path, &statfs);
+		sdcardfs_put_lower_path(dentry, &lower_path);
+
+		if (unlikely(err))
+			return 0;
+
+		/* Invalid statfs informations. */
+		if (unlikely(statfs.f_bsize == 0))
+			return 0;
+
+		/* if you are checking directory, set size to f_bsize. */
+		if (unlikely(dir))
+			size = statfs.f_bsize;
+
+		/* available size */
+		avail = statfs.f_bavail * statfs.f_bsize;
+
+		/* not enough space */
+		if ((u64)size > avail)
+			return 0;
+
+		/* enough space */
+		if ((avail - size) > (sbi->options.reserved_mb * 1024 * 1024))
+			return 1;
+
+		return 0;
+	} else
+		return 1;
+}
+
+/*
+ * Copies attrs and maintains sdcardfs managed attrs
+ * Since our permission check handles all special permissions, set those to be open
+ */
+static inline void sdcardfs_copy_and_fix_attrs(struct inode *dest, const struct inode *src)
+{
+	dest->i_mode = (src->i_mode  & S_IFMT) | S_IRWXU | S_IRWXG |
+			S_IROTH | S_IXOTH; /* 0775 */
+	dest->i_uid = make_kuid(&init_user_ns, SDCARDFS_I(dest)->data->d_uid);
+	dest->i_gid = make_kgid(&init_user_ns, AID_SDCARD_RW);
+	dest->i_rdev = src->i_rdev;
+	dest->i_atime = src->i_atime;
+	dest->i_mtime = src->i_mtime;
+	dest->i_ctime = src->i_ctime;
+	dest->i_blkbits = src->i_blkbits;
+	dest->i_flags = src->i_flags;
+	set_nlink(dest, src->i_nlink);
+}
+
+static inline bool str_case_eq(const char *s1, const char *s2)
+{
+	return !strcasecmp(s1, s2);
+}
+
+static inline bool str_n_case_eq(const char *s1, const char *s2, size_t len)
+{
+	return !strncasecmp(s1, s2, len);
+}
+
+static inline bool qstr_case_eq(const struct qstr *q1, const struct qstr *q2)
+{
+	return q1->len == q2->len && str_case_eq(q1->name, q2->name);
+}
+
+#define QSTR_LITERAL(string) QSTR_INIT(string, sizeof(string)-1)
+
+#endif	/* not _SDCARDFS_H_ */
diff --git a/fs/sdcardfs/super.c b/fs/sdcardfs/super.c
new file mode 100644
index 000000000000..b89947d878e3
--- /dev/null
+++ b/fs/sdcardfs/super.c
@@ -0,0 +1,326 @@
+/*
+ * fs/sdcardfs/super.c
+ *
+ * Copyright (c) 2013 Samsung Electronics Co. Ltd
+ *   Authors: Daeho Jeong, Woojoong Lee, Seunghwan Hyun,
+ *               Sunghwan Yun, Sungjong Seo
+ *
+ * This program has been developed as a stackable file system based on
+ * the WrapFS which written by
+ *
+ * Copyright (c) 1998-2011 Erez Zadok
+ * Copyright (c) 2009     Shrikar Archak
+ * Copyright (c) 2003-2011 Stony Brook University
+ * Copyright (c) 2003-2011 The Research Foundation of SUNY
+ *
+ * This file is dual licensed.  It may be redistributed and/or modified
+ * under the terms of the Apache 2.0 License OR version 2 of the GNU
+ * General Public License.
+ */
+
+#include "sdcardfs.h"
+
+/*
+ * The inode cache is used with alloc_inode for both our inode info and the
+ * vfs inode.
+ */
+static struct kmem_cache *sdcardfs_inode_cachep;
+
+/*
+ * To support the top references, we must track some data separately.
+ * An sdcardfs_inode_info always has a reference to its data, and once set up,
+ * also has a reference to its top. The top may be itself, in which case it
+ * holds two references to its data. When top is changed, it takes a ref to the
+ * new data and then drops the ref to the old data.
+ */
+static struct kmem_cache *sdcardfs_inode_data_cachep;
+
+void data_release(struct kref *ref)
+{
+	struct sdcardfs_inode_data *data =
+		container_of(ref, struct sdcardfs_inode_data, refcount);
+
+	kmem_cache_free(sdcardfs_inode_data_cachep, data);
+}
+
+/* final actions when unmounting a file system */
+static void sdcardfs_put_super(struct super_block *sb)
+{
+	struct sdcardfs_sb_info *spd;
+	struct super_block *s;
+
+	spd = SDCARDFS_SB(sb);
+	if (!spd)
+		return;
+
+	if (spd->obbpath_s) {
+		kfree(spd->obbpath_s);
+		path_put(&spd->obbpath);
+	}
+
+	/* decrement lower super references */
+	s = sdcardfs_lower_super(sb);
+	sdcardfs_set_lower_super(sb, NULL);
+	atomic_dec(&s->s_active);
+
+	kfree(spd);
+	sb->s_fs_info = NULL;
+}
+
+static int sdcardfs_statfs(struct dentry *dentry, struct kstatfs *buf)
+{
+	int err;
+	struct path lower_path;
+	u32 min_blocks;
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(dentry->d_sb);
+
+	sdcardfs_get_lower_path(dentry, &lower_path);
+	err = vfs_statfs(&lower_path, buf);
+	sdcardfs_put_lower_path(dentry, &lower_path);
+
+	if (sbi->options.reserved_mb) {
+		/* Invalid statfs informations. */
+		if (buf->f_bsize == 0) {
+			pr_err("Returned block size is zero.\n");
+			return -EINVAL;
+		}
+
+		min_blocks = ((sbi->options.reserved_mb * 1024 * 1024)/buf->f_bsize);
+		buf->f_blocks -= min_blocks;
+
+		if (buf->f_bavail > min_blocks)
+			buf->f_bavail -= min_blocks;
+		else
+			buf->f_bavail = 0;
+
+		/* Make reserved blocks invisiable to media storage */
+		buf->f_bfree = buf->f_bavail;
+	}
+
+	/* set return buf to our f/s to avoid confusing user-level utils */
+	buf->f_type = SDCARDFS_SUPER_MAGIC;
+
+	return err;
+}
+
+/*
+ * @flags: numeric mount options
+ * @options: mount options string
+ */
+static int sdcardfs_remount_fs(struct super_block *sb, int *flags, char *options)
+{
+	int err = 0;
+
+	/*
+	 * The VFS will take care of "ro" and "rw" flags among others.  We
+	 * can safely accept a few flags (RDONLY, MANDLOCK), and honor
+	 * SILENT, but anything else left over is an error.
+	 */
+	if ((*flags & ~(MS_RDONLY | MS_MANDLOCK | MS_SILENT)) != 0) {
+		pr_err("sdcardfs: remount flags 0x%x unsupported\n", *flags);
+		err = -EINVAL;
+	}
+
+	return err;
+}
+
+/*
+ * @mnt: mount point we are remounting
+ * @sb: superblock we are remounting
+ * @flags: numeric mount options
+ * @options: mount options string
+ */
+static int sdcardfs_remount_fs2(struct vfsmount *mnt, struct super_block *sb,
+						int *flags, char *options)
+{
+	int err = 0;
+
+	/*
+	 * The VFS will take care of "ro" and "rw" flags among others.  We
+	 * can safely accept a few flags (RDONLY, MANDLOCK), and honor
+	 * SILENT, but anything else left over is an error.
+	 */
+	if ((*flags & ~(MS_RDONLY | MS_MANDLOCK | MS_SILENT | MS_REMOUNT)) != 0) {
+		pr_err("sdcardfs: remount flags 0x%x unsupported\n", *flags);
+		err = -EINVAL;
+	}
+	pr_info("Remount options were %s for vfsmnt %p.\n", options, mnt);
+	err = parse_options_remount(sb, options, *flags & ~MS_SILENT, mnt->data);
+
+
+	return err;
+}
+
+static void *sdcardfs_clone_mnt_data(void *data)
+{
+	struct sdcardfs_vfsmount_options *opt = kmalloc(sizeof(struct sdcardfs_vfsmount_options), GFP_KERNEL);
+	struct sdcardfs_vfsmount_options *old = data;
+
+	if (!opt)
+		return NULL;
+	opt->gid = old->gid;
+	opt->mask = old->mask;
+	return opt;
+}
+
+static void sdcardfs_copy_mnt_data(void *data, void *newdata)
+{
+	struct sdcardfs_vfsmount_options *old = data;
+	struct sdcardfs_vfsmount_options *new = newdata;
+
+	old->gid = new->gid;
+	old->mask = new->mask;
+}
+
+/*
+ * Called by iput() when the inode reference count reached zero
+ * and the inode is not hashed anywhere.  Used to clear anything
+ * that needs to be, before the inode is completely destroyed and put
+ * on the inode free list.
+ */
+static void sdcardfs_evict_inode(struct inode *inode)
+{
+	struct inode *lower_inode;
+
+	truncate_inode_pages(&inode->i_data, 0);
+	set_top(SDCARDFS_I(inode), NULL);
+	clear_inode(inode);
+	/*
+	 * Decrement a reference to a lower_inode, which was incremented
+	 * by our read_inode when it was created initially.
+	 */
+	lower_inode = sdcardfs_lower_inode(inode);
+	sdcardfs_set_lower_inode(inode, NULL);
+	iput(lower_inode);
+}
+
+static struct inode *sdcardfs_alloc_inode(struct super_block *sb)
+{
+	struct sdcardfs_inode_info *i;
+	struct sdcardfs_inode_data *d;
+
+	i = kmem_cache_alloc(sdcardfs_inode_cachep, GFP_KERNEL);
+	if (!i)
+		return NULL;
+
+	/* memset everything up to the inode to 0 */
+	memset(i, 0, offsetof(struct sdcardfs_inode_info, vfs_inode));
+
+	d = kmem_cache_alloc(sdcardfs_inode_data_cachep,
+					GFP_KERNEL | __GFP_ZERO);
+	if (!d) {
+		kmem_cache_free(sdcardfs_inode_cachep, i);
+		return NULL;
+	}
+
+	i->data = d;
+	kref_init(&d->refcount);
+
+	i->vfs_inode.i_version = 1;
+	return &i->vfs_inode;
+}
+
+static void i_callback(struct rcu_head *head)
+{
+	struct inode *inode = container_of(head, struct inode, i_rcu);
+
+	release_own_data(SDCARDFS_I(inode));
+	kmem_cache_free(sdcardfs_inode_cachep, SDCARDFS_I(inode));
+}
+
+static void sdcardfs_destroy_inode(struct inode *inode)
+{
+	call_rcu(&inode->i_rcu, i_callback);
+}
+
+/* sdcardfs inode cache constructor */
+static void init_once(void *obj)
+{
+	struct sdcardfs_inode_info *i = obj;
+
+	inode_init_once(&i->vfs_inode);
+}
+
+int sdcardfs_init_inode_cache(void)
+{
+	sdcardfs_inode_cachep =
+		kmem_cache_create("sdcardfs_inode_cache",
+				  sizeof(struct sdcardfs_inode_info), 0,
+				  SLAB_RECLAIM_ACCOUNT, init_once);
+
+	if (!sdcardfs_inode_cachep)
+		return -ENOMEM;
+
+	sdcardfs_inode_data_cachep =
+		kmem_cache_create("sdcardfs_inode_data_cache",
+				  sizeof(struct sdcardfs_inode_data), 0,
+				  SLAB_RECLAIM_ACCOUNT, NULL);
+	if (!sdcardfs_inode_data_cachep) {
+		kmem_cache_destroy(sdcardfs_inode_cachep);
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+/* sdcardfs inode cache destructor */
+void sdcardfs_destroy_inode_cache(void)
+{
+	kmem_cache_destroy(sdcardfs_inode_data_cachep);
+	kmem_cache_destroy(sdcardfs_inode_cachep);
+}
+
+/*
+ * Used only in nfs, to kill any pending RPC tasks, so that subsequent
+ * code can actually succeed and won't leave tasks that need handling.
+ */
+static void sdcardfs_umount_begin(struct super_block *sb)
+{
+	struct super_block *lower_sb;
+
+	lower_sb = sdcardfs_lower_super(sb);
+	if (lower_sb && lower_sb->s_op && lower_sb->s_op->umount_begin)
+		lower_sb->s_op->umount_begin(lower_sb);
+}
+
+static int sdcardfs_show_options(struct vfsmount *mnt, struct seq_file *m,
+			struct dentry *root)
+{
+	struct sdcardfs_sb_info *sbi = SDCARDFS_SB(root->d_sb);
+	struct sdcardfs_mount_options *opts = &sbi->options;
+	struct sdcardfs_vfsmount_options *vfsopts = mnt->data;
+
+	if (opts->fs_low_uid != 0)
+		seq_printf(m, ",fsuid=%u", opts->fs_low_uid);
+	if (opts->fs_low_gid != 0)
+		seq_printf(m, ",fsgid=%u", opts->fs_low_gid);
+	if (vfsopts->gid != 0)
+		seq_printf(m, ",gid=%u", vfsopts->gid);
+	if (opts->multiuser)
+		seq_puts(m, ",multiuser");
+	if (vfsopts->mask)
+		seq_printf(m, ",mask=%u", vfsopts->mask);
+	if (opts->fs_user_id)
+		seq_printf(m, ",userid=%u", opts->fs_user_id);
+	if (opts->gid_derivation)
+		seq_puts(m, ",derive_gid");
+	if (opts->reserved_mb != 0)
+		seq_printf(m, ",reserved=%uMB", opts->reserved_mb);
+
+	return 0;
+};
+
+const struct super_operations sdcardfs_sops = {
+	.put_super	= sdcardfs_put_super,
+	.statfs		= sdcardfs_statfs,
+	.remount_fs	= sdcardfs_remount_fs,
+	.remount_fs2	= sdcardfs_remount_fs2,
+	.clone_mnt_data	= sdcardfs_clone_mnt_data,
+	.copy_mnt_data	= sdcardfs_copy_mnt_data,
+	.evict_inode	= sdcardfs_evict_inode,
+	.umount_begin	= sdcardfs_umount_begin,
+	.show_options2	= sdcardfs_show_options,
+	.alloc_inode	= sdcardfs_alloc_inode,
+	.destroy_inode	= sdcardfs_destroy_inode,
+	.drop_inode	= generic_delete_inode,
+};
diff --git a/fs/select.c b/fs/select.c
index 015547330e88..09e71a00a9b8 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -70,9 +70,9 @@ static long __estimate_accuracy(struct timespec *tv)
 	return slack;
 }
 
-long select_estimate_accuracy(struct timespec *tv)
+u64 select_estimate_accuracy(struct timespec *tv)
 {
-	unsigned long ret;
+	u64 ret;
 	struct timespec now;
 
 	/*
@@ -402,7 +402,7 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time)
 	struct poll_wqueues table;
 	poll_table *wait;
 	int retval, i, timed_out = 0;
-	unsigned long slack = 0;
+	u64 slack = 0;
 	unsigned int busy_flag = net_busy_loop_on() ? POLL_BUSY_LOOP : 0;
 	unsigned long busy_end = 0;
 
@@ -784,7 +784,7 @@ static int do_poll(unsigned int nfds,  struct poll_list *list,
 	poll_table* pt = &wait->pt;
 	ktime_t expire, *to = NULL;
 	int timed_out = 0, count = 0;
-	unsigned long slack = 0;
+	u64 slack = 0;
 	unsigned int busy_flag = net_busy_loop_on() ? POLL_BUSY_LOOP : 0;
 	unsigned long busy_end = 0;
 
diff --git a/fs/squashfs/Kconfig b/fs/squashfs/Kconfig
index ffb093e72b6c..6dd158a216f4 100644
--- a/fs/squashfs/Kconfig
+++ b/fs/squashfs/Kconfig
@@ -26,34 +26,6 @@ config SQUASHFS
 	  If unsure, say N.
 
 choice
-	prompt "File decompression options"
-	depends on SQUASHFS
-	help
-	  Squashfs now supports two options for decompressing file
-	  data.  Traditionally Squashfs has decompressed into an
-	  intermediate buffer and then memcopied it into the page cache.
-	  Squashfs now supports the ability to decompress directly into
-	  the page cache.
-
-	  If unsure, select "Decompress file data into an intermediate buffer"
-
-config SQUASHFS_FILE_CACHE
-	bool "Decompress file data into an intermediate buffer"
-	help
-	  Decompress file data into an intermediate buffer and then
-	  memcopy it into the page cache.
-
-config SQUASHFS_FILE_DIRECT
-	bool "Decompress files directly into the page cache"
-	help
-	  Directly decompress file data into the page cache.
-	  Doing so can significantly improve performance because
-	  it eliminates a memcpy and it also removes the lock contention
-	  on the single buffer.
-
-endchoice
-
-choice
 	prompt "Decompressor parallelisation options"
 	depends on SQUASHFS
 	help
diff --git a/fs/squashfs/Makefile b/fs/squashfs/Makefile
index 246a6f329d89..fe51f1507ed1 100644
--- a/fs/squashfs/Makefile
+++ b/fs/squashfs/Makefile
@@ -5,8 +5,7 @@
 obj-$(CONFIG_SQUASHFS) += squashfs.o
 squashfs-y += block.o cache.o dir.o export.o file.o fragment.o id.o inode.o
 squashfs-y += namei.o super.o symlink.o decompressor.o
-squashfs-$(CONFIG_SQUASHFS_FILE_CACHE) += file_cache.o
-squashfs-$(CONFIG_SQUASHFS_FILE_DIRECT) += file_direct.o page_actor.o
+squashfs-y += file_direct.o page_actor.o
 squashfs-$(CONFIG_SQUASHFS_DECOMP_SINGLE) += decompressor_single.o
 squashfs-$(CONFIG_SQUASHFS_DECOMP_MULTI) += decompressor_multi.o
 squashfs-$(CONFIG_SQUASHFS_DECOMP_MULTI_PERCPU) += decompressor_multi_percpu.o
diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c
index 0cea9b9236d0..b3b95e2ae2ff 100644
--- a/fs/squashfs/block.c
+++ b/fs/squashfs/block.c
@@ -28,9 +28,12 @@
 
 #include <linux/fs.h>
 #include <linux/vfs.h>
+#include <linux/bio.h>
 #include <linux/slab.h>
 #include <linux/string.h>
+#include <linux/pagemap.h>
 #include <linux/buffer_head.h>
+#include <linux/workqueue.h>
 
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
@@ -38,177 +41,435 @@
 #include "decompressor.h"
 #include "page_actor.h"
 
-/*
- * Read the metadata block length, this is stored in the first two
- * bytes of the metadata block.
- */
-static struct buffer_head *get_block_length(struct super_block *sb,
-			u64 *cur_index, int *offset, int *length)
+static struct workqueue_struct *squashfs_read_wq;
+
+struct squashfs_read_request {
+	struct super_block *sb;
+	u64 index;
+	int length;
+	int compressed;
+	int offset;
+	u64 read_end;
+	struct squashfs_page_actor *output;
+	enum {
+		SQUASHFS_COPY,
+		SQUASHFS_DECOMPRESS,
+		SQUASHFS_METADATA,
+	} data_processing;
+	bool synchronous;
+
+	/*
+	 * If the read is synchronous, it is possible to retrieve information
+	 * about the request by setting these pointers.
+	 */
+	int *res;
+	int *bytes_read;
+	int *bytes_uncompressed;
+
+	int nr_buffers;
+	struct buffer_head **bh;
+	struct work_struct offload;
+};
+
+struct squashfs_bio_request {
+	struct buffer_head **bh;
+	int nr_buffers;
+};
+
+static int squashfs_bio_submit(struct squashfs_read_request *req);
+
+int squashfs_init_read_wq(void)
 {
-	struct squashfs_sb_info *msblk = sb->s_fs_info;
-	struct buffer_head *bh;
+	squashfs_read_wq = create_workqueue("SquashFS read wq");
+	return !!squashfs_read_wq;
+}
+
+void squashfs_destroy_read_wq(void)
+{
+	flush_workqueue(squashfs_read_wq);
+	destroy_workqueue(squashfs_read_wq);
+}
+
+static void free_read_request(struct squashfs_read_request *req, int error)
+{
+	if (!req->synchronous)
+		squashfs_page_actor_free(req->output, error);
+	if (req->res)
+		*(req->res) = error;
+	kfree(req->bh);
+	kfree(req);
+}
+
+static void squashfs_process_blocks(struct squashfs_read_request *req)
+{
+	int error = 0;
+	int bytes, i, length;
+	struct squashfs_sb_info *msblk = req->sb->s_fs_info;
+	struct squashfs_page_actor *actor = req->output;
+	struct buffer_head **bh = req->bh;
+	int nr_buffers = req->nr_buffers;
+
+	for (i = 0; i < nr_buffers; ++i) {
+		if (!bh[i])
+			continue;
+		wait_on_buffer(bh[i]);
+		if (!buffer_uptodate(bh[i]))
+			error = -EIO;
+	}
+	if (error)
+		goto cleanup;
+
+	if (req->data_processing == SQUASHFS_METADATA) {
+		/* Extract the length of the metadata block */
+		if (req->offset != msblk->devblksize - 1) {
+			length = le16_to_cpup((__le16 *)
+					(bh[0]->b_data + req->offset));
+		} else {
+			length = (unsigned char)bh[0]->b_data[req->offset];
+			length |= (unsigned char)bh[1]->b_data[0] << 8;
+		}
+		req->compressed = SQUASHFS_COMPRESSED(length);
+		req->data_processing = req->compressed ? SQUASHFS_DECOMPRESS
+						       : SQUASHFS_COPY;
+		length = SQUASHFS_COMPRESSED_SIZE(length);
+		if (req->index + length + 2 > req->read_end) {
+			for (i = 0; i < nr_buffers; ++i)
+				put_bh(bh[i]);
+			kfree(bh);
+			req->length = length;
+			req->index += 2;
+			squashfs_bio_submit(req);
+			return;
+		}
+		req->length = length;
+		req->offset = (req->offset + 2) % PAGE_SIZE;
+		if (req->offset < 2) {
+			put_bh(bh[0]);
+			++bh;
+			--nr_buffers;
+		}
+	}
+	if (req->bytes_read)
+		*(req->bytes_read) = req->length;
 
-	bh = sb_bread(sb, *cur_index);
-	if (bh == NULL)
-		return NULL;
-
-	if (msblk->devblksize - *offset == 1) {
-		*length = (unsigned char) bh->b_data[*offset];
-		put_bh(bh);
-		bh = sb_bread(sb, ++(*cur_index));
-		if (bh == NULL)
-			return NULL;
-		*length |= (unsigned char) bh->b_data[0] << 8;
-		*offset = 1;
-	} else {
-		*length = (unsigned char) bh->b_data[*offset] |
-			(unsigned char) bh->b_data[*offset + 1] << 8;
-		*offset += 2;
-
-		if (*offset == msblk->devblksize) {
-			put_bh(bh);
-			bh = sb_bread(sb, ++(*cur_index));
-			if (bh == NULL)
-				return NULL;
-			*offset = 0;
+	if (req->data_processing == SQUASHFS_COPY) {
+		squashfs_bh_to_actor(bh, nr_buffers, req->output, req->offset,
+			req->length, msblk->devblksize);
+	} else if (req->data_processing == SQUASHFS_DECOMPRESS) {
+		req->length = squashfs_decompress(msblk, bh, nr_buffers,
+			req->offset, req->length, actor);
+		if (req->length < 0) {
+			error = -EIO;
+			goto cleanup;
 		}
 	}
 
-	return bh;
+	/* Last page may have trailing bytes not filled */
+	bytes = req->length % PAGE_SIZE;
+	if (bytes && actor->page[actor->pages - 1])
+		zero_user_segment(actor->page[actor->pages - 1], bytes,
+				  PAGE_SIZE);
+
+cleanup:
+	if (req->bytes_uncompressed)
+		*(req->bytes_uncompressed) = req->length;
+	if (error) {
+		for (i = 0; i < nr_buffers; ++i)
+			if (bh[i])
+				put_bh(bh[i]);
+	}
+	free_read_request(req, error);
 }
 
+static void read_wq_handler(struct work_struct *work)
+{
+	squashfs_process_blocks(container_of(work,
+		    struct squashfs_read_request, offload));
+}
 
-/*
- * Read and decompress a metadata block or datablock.  Length is non-zero
- * if a datablock is being read (the size is stored elsewhere in the
- * filesystem), otherwise the length is obtained from the first two bytes of
- * the metadata block.  A bit in the length field indicates if the block
- * is stored uncompressed in the filesystem (usually because compression
- * generated a larger block - this does occasionally happen with compression
- * algorithms).
- */
-int squashfs_read_data(struct super_block *sb, u64 index, int length,
-		u64 *next_index, struct squashfs_page_actor *output)
+static void squashfs_bio_end_io(struct bio *bio)
 {
-	struct squashfs_sb_info *msblk = sb->s_fs_info;
-	struct buffer_head **bh;
-	int offset = index & ((1 << msblk->devblksize_log2) - 1);
-	u64 cur_index = index >> msblk->devblksize_log2;
-	int bytes, compressed, b = 0, k = 0, avail, i;
+	int i;
+	int error = bio->bi_error;
+	struct squashfs_bio_request *bio_req = bio->bi_private;
+
+	bio_put(bio);
+
+	for (i = 0; i < bio_req->nr_buffers; ++i) {
+		if (!bio_req->bh[i])
+			continue;
+		if (!error)
+			set_buffer_uptodate(bio_req->bh[i]);
+		else
+			clear_buffer_uptodate(bio_req->bh[i]);
+		unlock_buffer(bio_req->bh[i]);
+	}
+	kfree(bio_req);
+}
+
+static int bh_is_optional(struct squashfs_read_request *req, int idx)
+{
+	int start_idx, end_idx;
+	struct squashfs_sb_info *msblk = req->sb->s_fs_info;
 
-	bh = kcalloc(((output->length + msblk->devblksize - 1)
-		>> msblk->devblksize_log2) + 1, sizeof(*bh), GFP_KERNEL);
-	if (bh == NULL)
+	start_idx = (idx * msblk->devblksize - req->offset) >> PAGE_SHIFT;
+	end_idx = ((idx + 1) * msblk->devblksize - req->offset + 1) >> PAGE_SHIFT;
+	if (start_idx >= req->output->pages)
+		return 1;
+	if (start_idx < 0)
+		start_idx = end_idx;
+	if (end_idx >= req->output->pages)
+		end_idx = start_idx;
+	return !req->output->page[start_idx] && !req->output->page[end_idx];
+}
+
+static int actor_getblks(struct squashfs_read_request *req, u64 block)
+{
+	int i;
+
+	req->bh = kmalloc_array(req->nr_buffers, sizeof(*(req->bh)), GFP_NOIO);
+	if (!req->bh)
 		return -ENOMEM;
 
-	if (length) {
+	for (i = 0; i < req->nr_buffers; ++i) {
 		/*
-		 * Datablock.
+		 * When dealing with an uncompressed block, the actor may
+		 * contains NULL pages. There's no need to read the buffers
+		 * associated with these pages.
 		 */
-		bytes = -offset;
-		compressed = SQUASHFS_COMPRESSED_BLOCK(length);
-		length = SQUASHFS_COMPRESSED_SIZE_BLOCK(length);
-		if (next_index)
-			*next_index = index + length;
-
-		TRACE("Block @ 0x%llx, %scompressed size %d, src size %d\n",
-			index, compressed ? "" : "un", length, output->length);
-
-		if (length < 0 || length > output->length ||
-				(index + length) > msblk->bytes_used)
-			goto read_failure;
-
-		for (b = 0; bytes < length; b++, cur_index++) {
-			bh[b] = sb_getblk(sb, cur_index);
-			if (bh[b] == NULL)
-				goto block_release;
-			bytes += msblk->devblksize;
+		if (!req->compressed && bh_is_optional(req, i)) {
+			req->bh[i] = NULL;
+			continue;
 		}
-		ll_rw_block(READ, b, bh);
-	} else {
-		/*
-		 * Metadata block.
-		 */
-		if ((index + 2) > msblk->bytes_used)
-			goto read_failure;
+		req->bh[i] = sb_getblk(req->sb, block + i);
+		if (!req->bh[i]) {
+			while (--i) {
+				if (req->bh[i])
+					put_bh(req->bh[i]);
+			}
+			return -1;
+		}
+	}
+	return 0;
+}
 
-		bh[0] = get_block_length(sb, &cur_index, &offset, &length);
-		if (bh[0] == NULL)
-			goto read_failure;
-		b = 1;
+static int squashfs_bio_submit(struct squashfs_read_request *req)
+{
+	struct bio *bio = NULL;
+	struct buffer_head *bh;
+	struct squashfs_bio_request *bio_req = NULL;
+	int b = 0, prev_block = 0;
+	struct squashfs_sb_info *msblk = req->sb->s_fs_info;
 
-		bytes = msblk->devblksize - offset;
-		compressed = SQUASHFS_COMPRESSED(length);
-		length = SQUASHFS_COMPRESSED_SIZE(length);
-		if (next_index)
-			*next_index = index + length + 2;
+	u64 read_start = round_down(req->index, msblk->devblksize);
+	u64 read_end = round_up(req->index + req->length, msblk->devblksize);
+	sector_t block = read_start >> msblk->devblksize_log2;
+	sector_t block_end = read_end >> msblk->devblksize_log2;
+	int offset = read_start - round_down(req->index, PAGE_SIZE);
+	int nr_buffers = block_end - block;
+	int blksz = msblk->devblksize;
+	int bio_max_pages = nr_buffers > BIO_MAX_PAGES ? BIO_MAX_PAGES
+						       : nr_buffers;
 
-		TRACE("Block @ 0x%llx, %scompressed size %d\n", index,
-				compressed ? "" : "un", length);
+	/* Setup the request */
+	req->read_end = read_end;
+	req->offset = req->index - read_start;
+	req->nr_buffers = nr_buffers;
+	if (actor_getblks(req, block) < 0)
+		goto getblk_failed;
 
-		if (length < 0 || length > output->length ||
-					(index + length) > msblk->bytes_used)
-			goto block_release;
+	/* Create and submit the BIOs */
+	for (b = 0; b < nr_buffers; ++b, offset += blksz) {
+		bh = req->bh[b];
+		if (!bh || !trylock_buffer(bh))
+			continue;
+		if (buffer_uptodate(bh)) {
+			unlock_buffer(bh);
+			continue;
+		}
+		offset %= PAGE_SIZE;
 
-		for (; bytes < length; b++) {
-			bh[b] = sb_getblk(sb, ++cur_index);
-			if (bh[b] == NULL)
-				goto block_release;
-			bytes += msblk->devblksize;
+		/* Append the buffer to the current BIO if it is contiguous */
+		if (bio && bio_req && prev_block + 1 == b) {
+			if (bio_add_page(bio, bh->b_page, blksz, offset)) {
+				bio_req->nr_buffers += 1;
+				prev_block = b;
+				continue;
+			}
 		}
-		ll_rw_block(READ, b - 1, bh + 1);
+
+		/* Otherwise, submit the current BIO and create a new one */
+		if (bio)
+			submit_bio(READ, bio);
+		bio_req = kcalloc(1, sizeof(struct squashfs_bio_request),
+				  GFP_NOIO);
+		if (!bio_req)
+			goto req_alloc_failed;
+		bio_req->bh = &req->bh[b];
+		bio = bio_alloc(GFP_NOIO, bio_max_pages);
+		if (!bio)
+			goto bio_alloc_failed;
+		bio->bi_bdev = req->sb->s_bdev;
+		bio->bi_iter.bi_sector = (block + b)
+				       << (msblk->devblksize_log2 - 9);
+		bio->bi_private = bio_req;
+		bio->bi_end_io = squashfs_bio_end_io;
+
+		bio_add_page(bio, bh->b_page, blksz, offset);
+		bio_req->nr_buffers += 1;
+		prev_block = b;
 	}
+	if (bio)
+		submit_bio(READ, bio);
 
-	for (i = 0; i < b; i++) {
-		wait_on_buffer(bh[i]);
-		if (!buffer_uptodate(bh[i]))
-			goto block_release;
+	if (req->synchronous)
+		squashfs_process_blocks(req);
+	else {
+		INIT_WORK(&req->offload, read_wq_handler);
+		schedule_work(&req->offload);
 	}
+	return 0;
 
-	if (compressed) {
-		length = squashfs_decompress(msblk, bh, b, offset, length,
-			output);
-		if (length < 0)
-			goto read_failure;
-	} else {
-		/*
-		 * Block is uncompressed.
-		 */
-		int in, pg_offset = 0;
-		void *data = squashfs_first_page(output);
-
-		for (bytes = length; k < b; k++) {
-			in = min(bytes, msblk->devblksize - offset);
-			bytes -= in;
-			while (in) {
-				if (pg_offset == PAGE_CACHE_SIZE) {
-					data = squashfs_next_page(output);
-					pg_offset = 0;
-				}
-				avail = min_t(int, in, PAGE_CACHE_SIZE -
-						pg_offset);
-				memcpy(data + pg_offset, bh[k]->b_data + offset,
-						avail);
-				in -= avail;
-				pg_offset += avail;
-				offset += avail;
-			}
-			offset = 0;
-			put_bh(bh[k]);
-		}
-		squashfs_finish_page(output);
+bio_alloc_failed:
+	kfree(bio_req);
+req_alloc_failed:
+	unlock_buffer(bh);
+	while (--nr_buffers >= b)
+		if (req->bh[nr_buffers])
+			put_bh(req->bh[nr_buffers]);
+	while (--b >= 0)
+		if (req->bh[b])
+			wait_on_buffer(req->bh[b]);
+getblk_failed:
+	free_read_request(req, -ENOMEM);
+	return -ENOMEM;
+}
+
+static int read_metadata_block(struct squashfs_read_request *req,
+			       u64 *next_index)
+{
+	int ret, error, bytes_read = 0, bytes_uncompressed = 0;
+	struct squashfs_sb_info *msblk = req->sb->s_fs_info;
+
+	if (req->index + 2 > msblk->bytes_used) {
+		free_read_request(req, -EINVAL);
+		return -EINVAL;
+	}
+	req->length = 2;
+
+	/* Do not read beyond the end of the device */
+	if (req->index + req->length > msblk->bytes_used)
+		req->length = msblk->bytes_used - req->index;
+	req->data_processing = SQUASHFS_METADATA;
+
+	/*
+	 * Reading metadata is always synchronous because we don't know the
+	 * length in advance and the function is expected to update
+	 * 'next_index' and return the length.
+	 */
+	req->synchronous = true;
+	req->res = &error;
+	req->bytes_read = &bytes_read;
+	req->bytes_uncompressed = &bytes_uncompressed;
+
+	TRACE("Metadata block @ 0x%llx, %scompressed size %d, src size %d\n",
+	      req->index, req->compressed ? "" : "un", bytes_read,
+	      req->output->length);
+
+	ret = squashfs_bio_submit(req);
+	if (ret)
+		return ret;
+	if (error)
+		return error;
+	if (next_index)
+		*next_index += 2 + bytes_read;
+	return bytes_uncompressed;
+}
+
+static int read_data_block(struct squashfs_read_request *req, int length,
+			   u64 *next_index, bool synchronous)
+{
+	int ret, error = 0, bytes_uncompressed = 0, bytes_read = 0;
+
+	req->compressed = SQUASHFS_COMPRESSED_BLOCK(length);
+	req->length = length = SQUASHFS_COMPRESSED_SIZE_BLOCK(length);
+	req->data_processing = req->compressed ? SQUASHFS_DECOMPRESS
+					       : SQUASHFS_COPY;
+
+	req->synchronous = synchronous;
+	if (synchronous) {
+		req->res = &error;
+		req->bytes_read = &bytes_read;
+		req->bytes_uncompressed = &bytes_uncompressed;
+	}
+
+	TRACE("Data block @ 0x%llx, %scompressed size %d, src size %d\n",
+	      req->index, req->compressed ? "" : "un", req->length,
+	      req->output->length);
+
+	ret = squashfs_bio_submit(req);
+	if (ret)
+		return ret;
+	if (synchronous)
+		ret = error ? error : bytes_uncompressed;
+	if (next_index)
+		*next_index += length;
+	return ret;
+}
+
+/*
+ * Read and decompress a metadata block or datablock.  Length is non-zero
+ * if a datablock is being read (the size is stored elsewhere in the
+ * filesystem), otherwise the length is obtained from the first two bytes of
+ * the metadata block.  A bit in the length field indicates if the block
+ * is stored uncompressed in the filesystem (usually because compression
+ * generated a larger block - this does occasionally happen with compression
+ * algorithms).
+ */
+static int __squashfs_read_data(struct super_block *sb, u64 index, int length,
+	u64 *next_index, struct squashfs_page_actor *output, bool sync)
+{
+	struct squashfs_read_request *req;
+
+	req = kcalloc(1, sizeof(struct squashfs_read_request), GFP_KERNEL);
+	if (!req) {
+		if (!sync)
+			squashfs_page_actor_free(output, -ENOMEM);
+		return -ENOMEM;
+	}
+
+	req->sb = sb;
+	req->index = index;
+	req->output = output;
+
+	if (next_index)
+		*next_index = index;
+
+	if (length)
+		length = read_data_block(req, length, next_index, sync);
+	else
+		length = read_metadata_block(req, next_index);
+
+	if (length < 0) {
+		ERROR("squashfs_read_data failed to read block 0x%llx\n",
+		      (unsigned long long)index);
+		return -EIO;
 	}
 
-	kfree(bh);
 	return length;
+}
 
-block_release:
-	for (; k < b; k++)
-		put_bh(bh[k]);
+int squashfs_read_data(struct super_block *sb, u64 index, int length,
+	u64 *next_index, struct squashfs_page_actor *output)
+{
+	return __squashfs_read_data(sb, index, length, next_index, output,
+				    true);
+}
+
+int squashfs_read_data_async(struct super_block *sb, u64 index, int length,
+	u64 *next_index, struct squashfs_page_actor *output)
+{
 
-read_failure:
-	ERROR("squashfs_read_data failed to read block 0x%llx\n",
-					(unsigned long long) index);
-	kfree(bh);
-	return -EIO;
+	return __squashfs_read_data(sb, index, length, next_index, output,
+				    false);
 }
diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c
index 1cb70a0b2168..6785d086ab38 100644
--- a/fs/squashfs/cache.c
+++ b/fs/squashfs/cache.c
@@ -209,17 +209,14 @@ void squashfs_cache_put(struct squashfs_cache_entry *entry)
  */
 void squashfs_cache_delete(struct squashfs_cache *cache)
 {
-	int i, j;
+	int i;
 
 	if (cache == NULL)
 		return;
 
 	for (i = 0; i < cache->entries; i++) {
-		if (cache->entry[i].data) {
-			for (j = 0; j < cache->pages; j++)
-				kfree(cache->entry[i].data[j]);
-			kfree(cache->entry[i].data);
-		}
+		if (cache->entry[i].page)
+			free_page_array(cache->entry[i].page, cache->pages);
 		kfree(cache->entry[i].actor);
 	}
 
@@ -236,7 +233,7 @@ void squashfs_cache_delete(struct squashfs_cache *cache)
 struct squashfs_cache *squashfs_cache_init(char *name, int entries,
 	int block_size)
 {
-	int i, j;
+	int i;
 	struct squashfs_cache *cache = kzalloc(sizeof(*cache), GFP_KERNEL);
 
 	if (cache == NULL) {
@@ -268,22 +265,13 @@ struct squashfs_cache *squashfs_cache_init(char *name, int entries,
 		init_waitqueue_head(&cache->entry[i].wait_queue);
 		entry->cache = cache;
 		entry->block = SQUASHFS_INVALID_BLK;
-		entry->data = kcalloc(cache->pages, sizeof(void *), GFP_KERNEL);
-		if (entry->data == NULL) {
+		entry->page = alloc_page_array(cache->pages, GFP_KERNEL);
+		if (!entry->page) {
 			ERROR("Failed to allocate %s cache entry\n", name);
 			goto cleanup;
 		}
-
-		for (j = 0; j < cache->pages; j++) {
-			entry->data[j] = kmalloc(PAGE_CACHE_SIZE, GFP_KERNEL);
-			if (entry->data[j] == NULL) {
-				ERROR("Failed to allocate %s buffer\n", name);
-				goto cleanup;
-			}
-		}
-
-		entry->actor = squashfs_page_actor_init(entry->data,
-						cache->pages, 0);
+		entry->actor = squashfs_page_actor_init(entry->page,
+			cache->pages, 0, NULL);
 		if (entry->actor == NULL) {
 			ERROR("Failed to allocate %s cache entry\n", name);
 			goto cleanup;
@@ -314,18 +302,20 @@ int squashfs_copy_data(void *buffer, struct squashfs_cache_entry *entry,
 		return min(length, entry->length - offset);
 
 	while (offset < entry->length) {
-		void *buff = entry->data[offset / PAGE_CACHE_SIZE]
-				+ (offset % PAGE_CACHE_SIZE);
+		void *buff = kmap_atomic(entry->page[offset / PAGE_CACHE_SIZE])
+			     + (offset % PAGE_CACHE_SIZE);
 		int bytes = min_t(int, entry->length - offset,
 				PAGE_CACHE_SIZE - (offset % PAGE_CACHE_SIZE));
 
 		if (bytes >= remaining) {
 			memcpy(buffer, buff, remaining);
+			kunmap_atomic(buff);
 			remaining = 0;
 			break;
 		}
 
 		memcpy(buffer, buff, bytes);
+		kunmap_atomic(buff);
 		buffer += bytes;
 		remaining -= bytes;
 		offset += bytes;
@@ -416,43 +406,38 @@ struct squashfs_cache_entry *squashfs_get_datablock(struct super_block *sb,
 void *squashfs_read_table(struct super_block *sb, u64 block, int length)
 {
 	int pages = (length + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
-	int i, res;
-	void *table, *buffer, **data;
+	struct page **page;
+	void *buff;
+	int res;
 	struct squashfs_page_actor *actor;
 
-	table = buffer = kmalloc(length, GFP_KERNEL);
-	if (table == NULL)
+	page = alloc_page_array(pages, GFP_KERNEL);
+	if (!page)
 		return ERR_PTR(-ENOMEM);
 
-	data = kcalloc(pages, sizeof(void *), GFP_KERNEL);
-	if (data == NULL) {
-		res = -ENOMEM;
-		goto failed;
-	}
-
-	actor = squashfs_page_actor_init(data, pages, length);
+	actor = squashfs_page_actor_init(page, pages, length, NULL);
 	if (actor == NULL) {
 		res = -ENOMEM;
-		goto failed2;
+		goto failed;
 	}
 
-	for (i = 0; i < pages; i++, buffer += PAGE_CACHE_SIZE)
-		data[i] = buffer;
-
 	res = squashfs_read_data(sb, block, length |
 		SQUASHFS_COMPRESSED_BIT_BLOCK, NULL, actor);
 
-	kfree(data);
-	kfree(actor);
-
 	if (res < 0)
-		goto failed;
+		goto failed2;
 
-	return table;
+	buff = kmalloc(length, GFP_KERNEL);
+	if (!buff)
+		goto failed2;
+	squashfs_actor_to_buf(actor, buff, length);
+	squashfs_page_actor_free(actor, 0);
+	free_page_array(page, pages);
+	return buff;
 
 failed2:
-	kfree(data);
+	squashfs_page_actor_free(actor, 0);
 failed:
-	kfree(table);
+	free_page_array(page, pages);
 	return ERR_PTR(res);
 }
diff --git a/fs/squashfs/decompressor.c b/fs/squashfs/decompressor.c
index e9034bf6e5ae..7de35bf297aa 100644
--- a/fs/squashfs/decompressor.c
+++ b/fs/squashfs/decompressor.c
@@ -24,7 +24,8 @@
 #include <linux/types.h>
 #include <linux/mutex.h>
 #include <linux/slab.h>
-#include <linux/buffer_head.h>
+#include <linux/highmem.h>
+#include <linux/fs.h>
 
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
@@ -94,40 +95,44 @@ const struct squashfs_decompressor *squashfs_lookup_decompressor(int id)
 static void *get_comp_opts(struct super_block *sb, unsigned short flags)
 {
 	struct squashfs_sb_info *msblk = sb->s_fs_info;
-	void *buffer = NULL, *comp_opts;
+	void *comp_opts, *buffer = NULL;
+	struct page *page;
 	struct squashfs_page_actor *actor = NULL;
 	int length = 0;
 
+	if (!SQUASHFS_COMP_OPTS(flags))
+		return squashfs_comp_opts(msblk, buffer, length);
+
 	/*
 	 * Read decompressor specific options from file system if present
 	 */
-	if (SQUASHFS_COMP_OPTS(flags)) {
-		buffer = kmalloc(PAGE_CACHE_SIZE, GFP_KERNEL);
-		if (buffer == NULL) {
-			comp_opts = ERR_PTR(-ENOMEM);
-			goto out;
-		}
-
-		actor = squashfs_page_actor_init(&buffer, 1, 0);
-		if (actor == NULL) {
-			comp_opts = ERR_PTR(-ENOMEM);
-			goto out;
-		}
-
-		length = squashfs_read_data(sb,
-			sizeof(struct squashfs_super_block), 0, NULL, actor);
-
-		if (length < 0) {
-			comp_opts = ERR_PTR(length);
-			goto out;
-		}
+
+	page = alloc_page(GFP_KERNEL);
+	if (!page)
+		return ERR_PTR(-ENOMEM);
+
+	actor = squashfs_page_actor_init(&page, 1, 0, NULL);
+	if (actor == NULL) {
+		comp_opts = ERR_PTR(-ENOMEM);
+		goto actor_error;
+	}
+
+	length = squashfs_read_data(sb,
+		sizeof(struct squashfs_super_block), 0, NULL, actor);
+
+	if (length < 0) {
+		comp_opts = ERR_PTR(length);
+		goto read_error;
 	}
 
+	buffer = kmap_atomic(page);
 	comp_opts = squashfs_comp_opts(msblk, buffer, length);
+	kunmap_atomic(buffer);
 
-out:
-	kfree(actor);
-	kfree(buffer);
+read_error:
+	squashfs_page_actor_free(actor, 0);
+actor_error:
+	__free_page(page);
 	return comp_opts;
 }
 
diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c
index e5c9689062ba..6f5ef8d7e55a 100644
--- a/fs/squashfs/file.c
+++ b/fs/squashfs/file.c
@@ -47,12 +47,16 @@
 #include <linux/string.h>
 #include <linux/pagemap.h>
 #include <linux/mutex.h>
+#include <linux/mm_inline.h>
 
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
 #include "squashfs_fs_i.h"
 #include "squashfs.h"
 
+// Backported from 4.5
+#define lru_to_page(head) (list_entry((head)->prev, struct page, lru))
+
 /*
  * Locate cache slot in range [offset, index] for specified inode.  If
  * there's more than one return the slot closest to index.
@@ -438,6 +442,21 @@ static int squashfs_readpage_fragment(struct page *page)
 	return res;
 }
 
+static int squashfs_readpages_fragment(struct page *page,
+	struct list_head *readahead_pages, struct address_space *mapping)
+{
+	if (!page) {
+		page = lru_to_page(readahead_pages);
+		list_del(&page->lru);
+		if (add_to_page_cache_lru(page, mapping, page->index,
+			mapping_gfp_constraint(mapping, GFP_KERNEL))) {
+			put_page(page);
+			return 0;
+		}
+	}
+	return squashfs_readpage_fragment(page);
+}
+
 static int squashfs_readpage_sparse(struct page *page, int index, int file_end)
 {
 	struct inode *inode = page->mapping->host;
@@ -450,54 +469,105 @@ static int squashfs_readpage_sparse(struct page *page, int index, int file_end)
 	return 0;
 }
 
-static int squashfs_readpage(struct file *file, struct page *page)
+static int squashfs_readpages_sparse(struct page *page,
+	struct list_head *readahead_pages, int index, int file_end,
+	struct address_space *mapping)
 {
-	struct inode *inode = page->mapping->host;
+	if (!page) {
+		page = lru_to_page(readahead_pages);
+		list_del(&page->lru);
+		if (add_to_page_cache_lru(page, mapping, page->index,
+			mapping_gfp_constraint(mapping, GFP_KERNEL))) {
+			put_page(page);
+			return 0;
+		}
+	}
+	return squashfs_readpage_sparse(page, index, file_end);
+}
+
+static int __squashfs_readpages(struct file *file, struct page *page,
+	struct list_head *readahead_pages, unsigned int nr_pages,
+	struct address_space *mapping)
+{
+	struct inode *inode = mapping->host;
 	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
-	int index = page->index >> (msblk->block_log - PAGE_CACHE_SHIFT);
 	int file_end = i_size_read(inode) >> msblk->block_log;
 	int res;
-	void *pageaddr;
 
-	TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
-				page->index, squashfs_i(inode)->start);
+	do {
+		struct page *cur_page = page ? page
+					     : lru_to_page(readahead_pages);
+		int page_index = cur_page->index;
+		int index = page_index >> (msblk->block_log - PAGE_CACHE_SHIFT);
+
+		if (page_index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >>
+						PAGE_CACHE_SHIFT))
+			return 1;
+
+		if (index < file_end || squashfs_i(inode)->fragment_block ==
+						SQUASHFS_INVALID_BLK) {
+			u64 block = 0;
+			int bsize = read_blocklist(inode, index, &block);
+
+			if (bsize < 0)
+				return -1;
+
+			if (bsize == 0) {
+				res = squashfs_readpages_sparse(page,
+					readahead_pages, index, file_end,
+					mapping);
+			} else {
+				res = squashfs_readpages_block(page,
+					readahead_pages, &nr_pages, mapping,
+					page_index, block, bsize);
+			}
+		} else {
+			res = squashfs_readpages_fragment(page,
+				readahead_pages, mapping);
+		}
+		if (res)
+			return 0;
+		page = NULL;
+	} while (readahead_pages && !list_empty(readahead_pages));
 
-	if (page->index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >>
-					PAGE_CACHE_SHIFT))
-		goto out;
+	return 0;
+}
 
-	if (index < file_end || squashfs_i(inode)->fragment_block ==
-					SQUASHFS_INVALID_BLK) {
-		u64 block = 0;
-		int bsize = read_blocklist(inode, index, &block);
-		if (bsize < 0)
-			goto error_out;
+static int squashfs_readpage(struct file *file, struct page *page)
+{
+	int ret;
 
-		if (bsize == 0)
-			res = squashfs_readpage_sparse(page, index, file_end);
+	TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n",
+	      page->index, squashfs_i(page->mapping->host)->start);
+
+	get_page(page);
+
+	ret = __squashfs_readpages(file, page, NULL, 1, page->mapping);
+	if (ret) {
+		flush_dcache_page(page);
+		if (ret < 0)
+			SetPageError(page);
 		else
-			res = squashfs_readpage_block(page, block, bsize);
-	} else
-		res = squashfs_readpage_fragment(page);
-
-	if (!res)
-		return 0;
-
-error_out:
-	SetPageError(page);
-out:
-	pageaddr = kmap_atomic(page);
-	memset(pageaddr, 0, PAGE_CACHE_SIZE);
-	kunmap_atomic(pageaddr);
-	flush_dcache_page(page);
-	if (!PageError(page))
-		SetPageUptodate(page);
-	unlock_page(page);
+			SetPageUptodate(page);
+		zero_user_segment(page, 0, PAGE_CACHE_SIZE);
+		unlock_page(page);
+		put_page(page);
+	}
 
 	return 0;
 }
 
+static int squashfs_readpages(struct file *file, struct address_space *mapping,
+			      struct list_head *pages, unsigned int nr_pages)
+{
+	TRACE("Entered squashfs_readpages, %u pages, first page index %lx\n",
+		nr_pages, lru_to_page(pages)->index);
+	__squashfs_readpages(file, NULL, pages, nr_pages, mapping);
+	return 0;
+}
+
 
 const struct address_space_operations squashfs_aops = {
-	.readpage = squashfs_readpage
+	.readpage = squashfs_readpage,
+	.readpages = squashfs_readpages,
 };
diff --git a/fs/squashfs/file_cache.c b/fs/squashfs/file_cache.c
deleted file mode 100644
index f2310d2a2019..000000000000
--- a/fs/squashfs/file_cache.c
+++ /dev/null
@@ -1,38 +0,0 @@
-/*
- * Copyright (c) 2013
- * Phillip Lougher <phillip@squashfs.org.uk>
- *
- * This work is licensed under the terms of the GNU GPL, version 2. See
- * the COPYING file in the top-level directory.
- */
-
-#include <linux/fs.h>
-#include <linux/vfs.h>
-#include <linux/kernel.h>
-#include <linux/slab.h>
-#include <linux/string.h>
-#include <linux/pagemap.h>
-#include <linux/mutex.h>
-
-#include "squashfs_fs.h"
-#include "squashfs_fs_sb.h"
-#include "squashfs_fs_i.h"
-#include "squashfs.h"
-
-/* Read separately compressed datablock and memcopy into page cache */
-int squashfs_readpage_block(struct page *page, u64 block, int bsize)
-{
-	struct inode *i = page->mapping->host;
-	struct squashfs_cache_entry *buffer = squashfs_get_datablock(i->i_sb,
-		block, bsize);
-	int res = buffer->error;
-
-	if (res)
-		ERROR("Unable to read page, block %llx, size %x\n", block,
-			bsize);
-	else
-		squashfs_copy_cache(page, buffer, buffer->length, 0);
-
-	squashfs_cache_put(buffer);
-	return res;
-}
diff --git a/fs/squashfs/file_direct.c b/fs/squashfs/file_direct.c
index 43e7a7eddac0..c97af4c6ccd0 100644
--- a/fs/squashfs/file_direct.c
+++ b/fs/squashfs/file_direct.c
@@ -13,6 +13,7 @@
 #include <linux/string.h>
 #include <linux/pagemap.h>
 #include <linux/mutex.h>
+#include <linux/mm_inline.h>
 
 #include "squashfs_fs.h"
 #include "squashfs_fs_sb.h"
@@ -20,157 +21,139 @@
 #include "squashfs.h"
 #include "page_actor.h"
 
-static int squashfs_read_cache(struct page *target_page, u64 block, int bsize,
-	int pages, struct page **page);
-
-/* Read separately compressed datablock directly into page cache */
-int squashfs_readpage_block(struct page *target_page, u64 block, int bsize)
+// Backported from 4.5
+#define lru_to_page(head) (list_entry((head)->prev, struct page, lru))
 
+static void release_actor_pages(struct page **page, int pages, int error)
 {
-	struct inode *inode = target_page->mapping->host;
-	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
+	int i;
 
-	int file_end = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT;
-	int mask = (1 << (msblk->block_log - PAGE_CACHE_SHIFT)) - 1;
-	int start_index = target_page->index & ~mask;
-	int end_index = start_index | mask;
-	int i, n, pages, missing_pages, bytes, res = -ENOMEM;
+	for (i = 0; i < pages; i++) {
+		if (!page[i])
+			continue;
+		flush_dcache_page(page[i]);
+		if (!error)
+			SetPageUptodate(page[i]);
+		else {
+			SetPageError(page[i]);
+			zero_user_segment(page[i], 0, PAGE_CACHE_SIZE);
+		}
+		unlock_page(page[i]);
+		put_page(page[i]);
+	}
+	kfree(page);
+}
+
+/*
+ * Create a "page actor" which will kmap and kunmap the
+ * page cache pages appropriately within the decompressor
+ */
+static struct squashfs_page_actor *actor_from_page_cache(
+	unsigned int actor_pages, struct page *target_page,
+	struct list_head *rpages, unsigned int *nr_pages, int start_index,
+	struct address_space *mapping)
+{
 	struct page **page;
 	struct squashfs_page_actor *actor;
-	void *pageaddr;
-
-	if (end_index > file_end)
-		end_index = file_end;
-
-	pages = end_index - start_index + 1;
-
-	page = kmalloc_array(pages, sizeof(void *), GFP_KERNEL);
-	if (page == NULL)
-		return res;
-
-	/*
-	 * Create a "page actor" which will kmap and kunmap the
-	 * page cache pages appropriately within the decompressor
-	 */
-	actor = squashfs_page_actor_init_special(page, pages, 0);
-	if (actor == NULL)
-		goto out;
-
-	/* Try to grab all the pages covered by the Squashfs block */
-	for (missing_pages = 0, i = 0, n = start_index; i < pages; i++, n++) {
-		page[i] = (n == target_page->index) ? target_page :
-			grab_cache_page_nowait(target_page->mapping, n);
+	int i, n;
+	gfp_t gfp = mapping_gfp_constraint(mapping, GFP_KERNEL);
+
+	page = kmalloc_array(actor_pages, sizeof(void *), GFP_KERNEL);
+	if (!page)
+		return NULL;
+
+	for (i = 0, n = start_index; i < actor_pages; i++, n++) {
+		if (target_page == NULL && rpages && !list_empty(rpages)) {
+			struct page *cur_page = lru_to_page(rpages);
+
+			if (cur_page->index < start_index + actor_pages) {
+				list_del(&cur_page->lru);
+				--(*nr_pages);
+				if (add_to_page_cache_lru(cur_page, mapping,
+							  cur_page->index, gfp))
+					put_page(cur_page);
+				else
+					target_page = cur_page;
+			} else
+				rpages = NULL;
+		}
 
-		if (page[i] == NULL) {
-			missing_pages++;
-			continue;
+		if (target_page && target_page->index == n) {
+			page[i] = target_page;
+			target_page = NULL;
+		} else {
+			page[i] = grab_cache_page_nowait(mapping, n);
+			if (page[i] == NULL)
+				continue;
 		}
 
 		if (PageUptodate(page[i])) {
 			unlock_page(page[i]);
-			page_cache_release(page[i]);
+			put_page(page[i]);
 			page[i] = NULL;
-			missing_pages++;
 		}
 	}
 
-	if (missing_pages) {
-		/*
-		 * Couldn't get one or more pages, this page has either
-		 * been VM reclaimed, but others are still in the page cache
-		 * and uptodate, or we're racing with another thread in
-		 * squashfs_readpage also trying to grab them.  Fall back to
-		 * using an intermediate buffer.
-		 */
-		res = squashfs_read_cache(target_page, block, bsize, pages,
-								page);
-		if (res < 0)
-			goto mark_errored;
-
-		goto out;
+	actor = squashfs_page_actor_init(page, actor_pages, 0,
+			release_actor_pages);
+	if (!actor) {
+		release_actor_pages(page, actor_pages, -ENOMEM);
+		kfree(page);
+		return NULL;
 	}
-
-	/* Decompress directly into the page cache buffers */
-	res = squashfs_read_data(inode->i_sb, block, bsize, NULL, actor);
-	if (res < 0)
-		goto mark_errored;
-
-	/* Last page may have trailing bytes not filled */
-	bytes = res % PAGE_CACHE_SIZE;
-	if (bytes) {
-		pageaddr = kmap_atomic(page[pages - 1]);
-		memset(pageaddr + bytes, 0, PAGE_CACHE_SIZE - bytes);
-		kunmap_atomic(pageaddr);
-	}
-
-	/* Mark pages as uptodate, unlock and release */
-	for (i = 0; i < pages; i++) {
-		flush_dcache_page(page[i]);
-		SetPageUptodate(page[i]);
-		unlock_page(page[i]);
-		if (page[i] != target_page)
-			page_cache_release(page[i]);
-	}
-
-	kfree(actor);
-	kfree(page);
-
-	return 0;
-
-mark_errored:
-	/* Decompression failed, mark pages as errored.  Target_page is
-	 * dealt with by the caller
-	 */
-	for (i = 0; i < pages; i++) {
-		if (page[i] == NULL || page[i] == target_page)
-			continue;
-		flush_dcache_page(page[i]);
-		SetPageError(page[i]);
-		unlock_page(page[i]);
-		page_cache_release(page[i]);
-	}
-
-out:
-	kfree(actor);
-	kfree(page);
-	return res;
+	return actor;
 }
 
+int squashfs_readpages_block(struct page *target_page,
+			     struct list_head *readahead_pages,
+			     unsigned int *nr_pages,
+			     struct address_space *mapping,
+			     int page_index, u64 block, int bsize)
 
-static int squashfs_read_cache(struct page *target_page, u64 block, int bsize,
-	int pages, struct page **page)
 {
-	struct inode *i = target_page->mapping->host;
-	struct squashfs_cache_entry *buffer = squashfs_get_datablock(i->i_sb,
-						 block, bsize);
-	int bytes = buffer->length, res = buffer->error, n, offset = 0;
-	void *pageaddr;
-
-	if (res) {
-		ERROR("Unable to read page, block %llx, size %x\n", block,
-			bsize);
-		goto out;
-	}
-
-	for (n = 0; n < pages && bytes > 0; n++,
-			bytes -= PAGE_CACHE_SIZE, offset += PAGE_CACHE_SIZE) {
-		int avail = min_t(int, bytes, PAGE_CACHE_SIZE);
-
-		if (page[n] == NULL)
-			continue;
+	struct squashfs_page_actor *actor;
+	struct inode *inode = mapping->host;
+	struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info;
+	int start_index, end_index, file_end, actor_pages, res;
+	int mask = (1 << (msblk->block_log - PAGE_CACHE_SHIFT)) - 1;
 
-		pageaddr = kmap_atomic(page[n]);
-		squashfs_copy_data(pageaddr, buffer, offset, avail);
-		memset(pageaddr + avail, 0, PAGE_CACHE_SIZE - avail);
-		kunmap_atomic(pageaddr);
-		flush_dcache_page(page[n]);
-		SetPageUptodate(page[n]);
-		unlock_page(page[n]);
-		if (page[n] != target_page)
-			page_cache_release(page[n]);
+	/*
+	 * If readpage() is called on an uncompressed datablock, we can just
+	 * read the pages instead of fetching the whole block.
+	 * This greatly improves the performance when a process keep doing
+	 * random reads because we only fetch the necessary data.
+	 * The readahead algorithm will take care of doing speculative reads
+	 * if necessary.
+	 * We can't read more than 1 block even if readahead provides use more
+	 * pages because we don't know yet if the next block is compressed or
+	 * not.
+	 */
+	if (bsize && !SQUASHFS_COMPRESSED_BLOCK(bsize)) {
+		u64 block_end = block + msblk->block_size;
+
+		block += (page_index & mask) * PAGE_CACHE_SIZE;
+		actor_pages = (block_end - block) / PAGE_CACHE_SIZE;
+		if (*nr_pages < actor_pages)
+			actor_pages = *nr_pages;
+		start_index = page_index;
+		bsize = min_t(int, bsize, (PAGE_CACHE_SIZE * actor_pages)
+					  | SQUASHFS_COMPRESSED_BIT_BLOCK);
+	} else {
+		file_end = (i_size_read(inode) - 1) >> PAGE_CACHE_SHIFT;
+		start_index = page_index & ~mask;
+		end_index = start_index | mask;
+		if (end_index > file_end)
+			end_index = file_end;
+		actor_pages = end_index - start_index + 1;
 	}
 
-out:
-	squashfs_cache_put(buffer);
-	return res;
+	actor = actor_from_page_cache(actor_pages, target_page,
+				      readahead_pages, nr_pages, start_index,
+				      mapping);
+	if (!actor)
+		return -ENOMEM;
+
+	res = squashfs_read_data_async(inode->i_sb, block, bsize, NULL,
+				       actor);
+	return res < 0 ? res : 0;
 }
diff --git a/fs/squashfs/lz4_wrapper.c b/fs/squashfs/lz4_wrapper.c
index c31e2bc9c081..df4fa3c7ddd0 100644
--- a/fs/squashfs/lz4_wrapper.c
+++ b/fs/squashfs/lz4_wrapper.c
@@ -94,39 +94,17 @@ static int lz4_uncompress(struct squashfs_sb_info *msblk, void *strm,
 	struct buffer_head **bh, int b, int offset, int length,
 	struct squashfs_page_actor *output)
 {
-	struct squashfs_lz4 *stream = strm;
-	void *buff = stream->input, *data;
-	int avail, i, bytes = length, res;
+	int res;
 	size_t dest_len = output->length;
+	struct squashfs_lz4 *stream = strm;
 
-	for (i = 0; i < b; i++) {
-		avail = min(bytes, msblk->devblksize - offset);
-		memcpy(buff, bh[i]->b_data + offset, avail);
-		buff += avail;
-		bytes -= avail;
-		offset = 0;
-		put_bh(bh[i]);
-	}
-
+	squashfs_bh_to_buf(bh, b, stream->input, offset, length,
+		msblk->devblksize);
 	res = lz4_decompress_unknownoutputsize(stream->input, length,
 					stream->output, &dest_len);
 	if (res)
 		return -EIO;
-
-	bytes = dest_len;
-	data = squashfs_first_page(output);
-	buff = stream->output;
-	while (data) {
-		if (bytes <= PAGE_CACHE_SIZE) {
-			memcpy(data, buff, bytes);
-			break;
-		}
-		memcpy(data, buff, PAGE_CACHE_SIZE);
-		buff += PAGE_CACHE_SIZE;
-		bytes -= PAGE_CACHE_SIZE;
-		data = squashfs_next_page(output);
-	}
-	squashfs_finish_page(output);
+	squashfs_buf_to_actor(stream->output, output, dest_len);
 
 	return dest_len;
 }
diff --git a/fs/squashfs/lzo_wrapper.c b/fs/squashfs/lzo_wrapper.c
index 244b9fbfff7b..2c844d53a59e 100644
--- a/fs/squashfs/lzo_wrapper.c
+++ b/fs/squashfs/lzo_wrapper.c
@@ -79,45 +79,19 @@ static int lzo_uncompress(struct squashfs_sb_info *msblk, void *strm,
 	struct buffer_head **bh, int b, int offset, int length,
 	struct squashfs_page_actor *output)
 {
-	struct squashfs_lzo *stream = strm;
-	void *buff = stream->input, *data;
-	int avail, i, bytes = length, res;
+	int res;
 	size_t out_len = output->length;
+	struct squashfs_lzo *stream = strm;
 
-	for (i = 0; i < b; i++) {
-		avail = min(bytes, msblk->devblksize - offset);
-		memcpy(buff, bh[i]->b_data + offset, avail);
-		buff += avail;
-		bytes -= avail;
-		offset = 0;
-		put_bh(bh[i]);
-	}
-
+	squashfs_bh_to_buf(bh, b, stream->input, offset, length,
+		msblk->devblksize);
 	res = lzo1x_decompress_safe(stream->input, (size_t)length,
 					stream->output, &out_len);
 	if (res != LZO_E_OK)
-		goto failed;
+		return -EIO;
+	squashfs_buf_to_actor(stream->output, output, out_len);
 
-	res = bytes = (int)out_len;
-	data = squashfs_first_page(output);
-	buff = stream->output;
-	while (data) {
-		if (bytes <= PAGE_CACHE_SIZE) {
-			memcpy(data, buff, bytes);
-			break;
-		} else {
-			memcpy(data, buff, PAGE_CACHE_SIZE);
-			buff += PAGE_CACHE_SIZE;
-			bytes -= PAGE_CACHE_SIZE;
-			data = squashfs_next_page(output);
-		}
-	}
-	squashfs_finish_page(output);
-
-	return res;
-
-failed:
-	return -EIO;
+	return out_len;
 }
 
 const struct squashfs_decompressor squashfs_lzo_comp_ops = {
diff --git a/fs/squashfs/page_actor.c b/fs/squashfs/page_actor.c
index 5a1c11f56441..53863508e400 100644
--- a/fs/squashfs/page_actor.c
+++ b/fs/squashfs/page_actor.c
@@ -9,39 +9,11 @@
 #include <linux/kernel.h>
 #include <linux/slab.h>
 #include <linux/pagemap.h>
+#include <linux/buffer_head.h>
 #include "page_actor.h"
 
-/*
- * This file contains implementations of page_actor for decompressing into
- * an intermediate buffer, and for decompressing directly into the
- * page cache.
- *
- * Calling code should avoid sleeping between calls to squashfs_first_page()
- * and squashfs_finish_page().
- */
-
-/* Implementation of page_actor for decompressing into intermediate buffer */
-static void *cache_first_page(struct squashfs_page_actor *actor)
-{
-	actor->next_page = 1;
-	return actor->buffer[0];
-}
-
-static void *cache_next_page(struct squashfs_page_actor *actor)
-{
-	if (actor->next_page == actor->pages)
-		return NULL;
-
-	return actor->buffer[actor->next_page++];
-}
-
-static void cache_finish_page(struct squashfs_page_actor *actor)
-{
-	/* empty */
-}
-
-struct squashfs_page_actor *squashfs_page_actor_init(void **buffer,
-	int pages, int length)
+struct squashfs_page_actor *squashfs_page_actor_init(struct page **page,
+	int pages, int length, void (*release_pages)(struct page **, int, int))
 {
 	struct squashfs_page_actor *actor = kmalloc(sizeof(*actor), GFP_KERNEL);
 
@@ -49,52 +21,133 @@ struct squashfs_page_actor *squashfs_page_actor_init(void **buffer,
 		return NULL;
 
 	actor->length = length ? : pages * PAGE_CACHE_SIZE;
-	actor->buffer = buffer;
+	actor->page = page;
 	actor->pages = pages;
 	actor->next_page = 0;
-	actor->squashfs_first_page = cache_first_page;
-	actor->squashfs_next_page = cache_next_page;
-	actor->squashfs_finish_page = cache_finish_page;
+	actor->pageaddr = NULL;
+	actor->release_pages = release_pages;
 	return actor;
 }
 
-/* Implementation of page_actor for decompressing directly into page cache. */
-static void *direct_first_page(struct squashfs_page_actor *actor)
+void squashfs_page_actor_free(struct squashfs_page_actor *actor, int error)
+{
+	if (!actor)
+		return;
+
+	if (actor->release_pages)
+		actor->release_pages(actor->page, actor->pages, error);
+	kfree(actor);
+}
+
+void squashfs_actor_to_buf(struct squashfs_page_actor *actor, void *buf,
+	int length)
 {
-	actor->next_page = 1;
-	return actor->pageaddr = kmap_atomic(actor->page[0]);
+	void *pageaddr;
+	int pos = 0, avail, i;
+
+	for (i = 0; i < actor->pages && pos < length; ++i) {
+		avail = min_t(int, length - pos, PAGE_CACHE_SIZE);
+		if (actor->page[i]) {
+			pageaddr = kmap_atomic(actor->page[i]);
+			memcpy(buf + pos, pageaddr, avail);
+			kunmap_atomic(pageaddr);
+		}
+		pos += avail;
+	}
 }
 
-static void *direct_next_page(struct squashfs_page_actor *actor)
+void squashfs_buf_to_actor(void *buf, struct squashfs_page_actor *actor,
+	int length)
 {
-	if (actor->pageaddr)
-		kunmap_atomic(actor->pageaddr);
+	void *pageaddr;
+	int pos = 0, avail, i;
+
+	for (i = 0; i < actor->pages && pos < length; ++i) {
+		avail = min_t(int, length - pos, PAGE_CACHE_SIZE);
+		if (actor->page[i]) {
+			pageaddr = kmap_atomic(actor->page[i]);
+			memcpy(pageaddr, buf + pos, avail);
+			kunmap_atomic(pageaddr);
+		}
+		pos += avail;
+	}
+}
 
-	return actor->pageaddr = actor->next_page == actor->pages ? NULL :
-		kmap_atomic(actor->page[actor->next_page++]);
+void squashfs_bh_to_actor(struct buffer_head **bh, int nr_buffers,
+	struct squashfs_page_actor *actor, int offset, int length, int blksz)
+{
+	void *kaddr = NULL;
+	int bytes = 0, pgoff = 0, b = 0, p = 0, avail, i;
+
+	while (bytes < length) {
+		if (actor->page[p]) {
+			kaddr = kmap_atomic(actor->page[p]);
+			while (pgoff < PAGE_CACHE_SIZE && bytes < length) {
+				avail = min_t(int, blksz - offset,
+						PAGE_CACHE_SIZE - pgoff);
+				memcpy(kaddr + pgoff, bh[b]->b_data + offset,
+				       avail);
+				pgoff += avail;
+				bytes += avail;
+				offset = (offset + avail) % blksz;
+				if (!offset) {
+					put_bh(bh[b]);
+					++b;
+				}
+			}
+			kunmap_atomic(kaddr);
+			pgoff = 0;
+		} else {
+			for (i = 0; i < PAGE_CACHE_SIZE / blksz; ++i) {
+				if (bh[b])
+					put_bh(bh[b]);
+				++b;
+			}
+			bytes += PAGE_CACHE_SIZE;
+		}
+		++p;
+	}
 }
 
-static void direct_finish_page(struct squashfs_page_actor *actor)
+void squashfs_bh_to_buf(struct buffer_head **bh, int nr_buffers, void *buf,
+	int offset, int length, int blksz)
 {
-	if (actor->pageaddr)
-		kunmap_atomic(actor->pageaddr);
+	int i, avail, bytes = 0;
+
+	for (i = 0; i < nr_buffers && bytes < length; ++i) {
+		avail = min_t(int, length - bytes, blksz - offset);
+		if (bh[i]) {
+			memcpy(buf + bytes, bh[i]->b_data + offset, avail);
+			put_bh(bh[i]);
+		}
+		bytes += avail;
+		offset = 0;
+	}
 }
 
-struct squashfs_page_actor *squashfs_page_actor_init_special(struct page **page,
-	int pages, int length)
+void free_page_array(struct page **page, int nr_pages)
 {
-	struct squashfs_page_actor *actor = kmalloc(sizeof(*actor), GFP_KERNEL);
+	int i;
 
-	if (actor == NULL)
-		return NULL;
+	for (i = 0; i < nr_pages; ++i)
+		__free_page(page[i]);
+	kfree(page);
+}
 
-	actor->length = length ? : pages * PAGE_CACHE_SIZE;
-	actor->page = page;
-	actor->pages = pages;
-	actor->next_page = 0;
-	actor->pageaddr = NULL;
-	actor->squashfs_first_page = direct_first_page;
-	actor->squashfs_next_page = direct_next_page;
-	actor->squashfs_finish_page = direct_finish_page;
-	return actor;
+struct page **alloc_page_array(int nr_pages, int gfp_mask)
+{
+	int i;
+	struct page **page;
+
+	page = kcalloc(nr_pages, sizeof(struct page *), gfp_mask);
+	if (!page)
+		return NULL;
+	for (i = 0; i < nr_pages; ++i) {
+		page[i] = alloc_page(gfp_mask);
+		if (!page[i]) {
+			free_page_array(page, i);
+			return NULL;
+		}
+	}
+	return page;
 }
diff --git a/fs/squashfs/page_actor.h b/fs/squashfs/page_actor.h
index 26dd82008b82..aa1ed790b5a3 100644
--- a/fs/squashfs/page_actor.h
+++ b/fs/squashfs/page_actor.h
@@ -5,77 +5,61 @@
  * Phillip Lougher <phillip@squashfs.org.uk>
  *
  * This work is licensed under the terms of the GNU GPL, version 2. See
- * the COPYING file in the top-level directory.
+ * the COPYING file in the top-level squashfsory.
  */
 
-#ifndef CONFIG_SQUASHFS_FILE_DIRECT
 struct squashfs_page_actor {
-	void	**page;
+	struct page	**page;
+	void	*pageaddr;
 	int	pages;
 	int	length;
 	int	next_page;
+	void	(*release_pages)(struct page **, int, int);
 };
 
-static inline struct squashfs_page_actor *squashfs_page_actor_init(void **page,
-	int pages, int length)
-{
-	struct squashfs_page_actor *actor = kmalloc(sizeof(*actor), GFP_KERNEL);
-
-	if (actor == NULL)
-		return NULL;
+extern struct squashfs_page_actor *squashfs_page_actor_init(struct page **,
+	int, int, void (*)(struct page **, int, int));
+extern void squashfs_page_actor_free(struct squashfs_page_actor *, int);
 
-	actor->length = length ? : pages * PAGE_CACHE_SIZE;
-	actor->page = page;
-	actor->pages = pages;
-	actor->next_page = 0;
-	return actor;
-}
+extern void squashfs_actor_to_buf(struct squashfs_page_actor *, void *, int);
+extern void squashfs_buf_to_actor(void *, struct squashfs_page_actor *, int);
+extern void squashfs_bh_to_actor(struct buffer_head **, int,
+	struct squashfs_page_actor *, int, int, int);
+extern void squashfs_bh_to_buf(struct buffer_head **, int, void *, int, int,
+	int);
 
+/*
+ * Calling code should avoid sleeping between calls to squashfs_first_page()
+ * and squashfs_finish_page().
+ */
 static inline void *squashfs_first_page(struct squashfs_page_actor *actor)
 {
 	actor->next_page = 1;
-	return actor->page[0];
+	return actor->pageaddr = actor->page[0] ? kmap_atomic(actor->page[0])
+						: NULL;
 }
 
 static inline void *squashfs_next_page(struct squashfs_page_actor *actor)
 {
-	return actor->next_page == actor->pages ? NULL :
-		actor->page[actor->next_page++];
-}
+	if (!IS_ERR_OR_NULL(actor->pageaddr))
+		kunmap_atomic(actor->pageaddr);
 
-static inline void squashfs_finish_page(struct squashfs_page_actor *actor)
-{
-	/* empty */
-}
-#else
-struct squashfs_page_actor {
-	union {
-		void		**buffer;
-		struct page	**page;
-	};
-	void	*pageaddr;
-	void    *(*squashfs_first_page)(struct squashfs_page_actor *);
-	void    *(*squashfs_next_page)(struct squashfs_page_actor *);
-	void    (*squashfs_finish_page)(struct squashfs_page_actor *);
-	int	pages;
-	int	length;
-	int	next_page;
-};
+	if (actor->next_page == actor->pages)
+		return actor->pageaddr = ERR_PTR(-ENODATA);
 
-extern struct squashfs_page_actor *squashfs_page_actor_init(void **, int, int);
-extern struct squashfs_page_actor *squashfs_page_actor_init_special(struct page
-							 **, int, int);
-static inline void *squashfs_first_page(struct squashfs_page_actor *actor)
-{
-	return actor->squashfs_first_page(actor);
-}
-static inline void *squashfs_next_page(struct squashfs_page_actor *actor)
-{
-	return actor->squashfs_next_page(actor);
+	actor->pageaddr = actor->page[actor->next_page] ?
+	    kmap_atomic(actor->page[actor->next_page]) : NULL;
+	++actor->next_page;
+	return actor->pageaddr;
 }
+
 static inline void squashfs_finish_page(struct squashfs_page_actor *actor)
 {
-	actor->squashfs_finish_page(actor);
+	if (!IS_ERR_OR_NULL(actor->pageaddr))
+		kunmap_atomic(actor->pageaddr);
 }
-#endif
+
+extern struct page **alloc_page_array(int, int);
+extern void free_page_array(struct page **, int);
+
 #endif
diff --git a/fs/squashfs/squashfs.h b/fs/squashfs/squashfs.h
index 887d6d270080..6093579c6c5d 100644
--- a/fs/squashfs/squashfs.h
+++ b/fs/squashfs/squashfs.h
@@ -28,8 +28,14 @@
 #define WARNING(s, args...)	pr_warn("SQUASHFS: "s, ## args)
 
 /* block.c */
+extern int squashfs_init_read_wq(void);
+extern void squashfs_destroy_read_wq(void);
 extern int squashfs_read_data(struct super_block *, u64, int, u64 *,
 				struct squashfs_page_actor *);
+extern int squashfs_read_data(struct super_block *, u64, int, u64 *,
+	struct squashfs_page_actor *);
+extern int squashfs_read_data_async(struct super_block *, u64, int, u64 *,
+	struct squashfs_page_actor *);
 
 /* cache.c */
 extern struct squashfs_cache *squashfs_cache_init(char *, int, int);
@@ -70,8 +76,9 @@ extern __le64 *squashfs_read_fragment_index_table(struct super_block *,
 void squashfs_copy_cache(struct page *, struct squashfs_cache_entry *, int,
 				int);
 
-/* file_xxx.c */
-extern int squashfs_readpage_block(struct page *, u64, int);
+/* file_direct.c */
+extern int squashfs_readpages_block(struct page *, struct list_head *,
+	unsigned int *, struct address_space *, int, u64, int);
 
 /* id.c */
 extern int squashfs_get_id(struct super_block *, unsigned int, unsigned int *);
diff --git a/fs/squashfs/squashfs_fs_sb.h b/fs/squashfs/squashfs_fs_sb.h
index 1da565cb50c3..8a6995de0277 100644
--- a/fs/squashfs/squashfs_fs_sb.h
+++ b/fs/squashfs/squashfs_fs_sb.h
@@ -49,7 +49,7 @@ struct squashfs_cache_entry {
 	int			num_waiters;
 	wait_queue_head_t	wait_queue;
 	struct squashfs_cache	*cache;
-	void			**data;
+	struct page		**page;
 	struct squashfs_page_actor	*actor;
 };
 
diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c
index 5056babe00df..61cd0b39ed0e 100644
--- a/fs/squashfs/super.c
+++ b/fs/squashfs/super.c
@@ -444,9 +444,15 @@ static int __init init_squashfs_fs(void)
 	if (err)
 		return err;
 
+	if (!squashfs_init_read_wq()) {
+		destroy_inodecache();
+		return -ENOMEM;
+        }
+
 	err = register_filesystem(&squashfs_fs_type);
 	if (err) {
 		destroy_inodecache();
+		squashfs_destroy_read_wq();
 		return err;
 	}
 
@@ -460,6 +466,7 @@ static void __exit exit_squashfs_fs(void)
 {
 	unregister_filesystem(&squashfs_fs_type);
 	destroy_inodecache();
+	squashfs_destroy_read_wq();
 }
 
 
diff --git a/fs/squashfs/xz_wrapper.c b/fs/squashfs/xz_wrapper.c
index c609624e4b8a..14cd373e1897 100644
--- a/fs/squashfs/xz_wrapper.c
+++ b/fs/squashfs/xz_wrapper.c
@@ -55,7 +55,7 @@ static void *squashfs_xz_comp_opts(struct squashfs_sb_info *msblk,
 	struct comp_opts *opts;
 	int err = 0, n;
 
-	opts = kmalloc(sizeof(*opts), GFP_KERNEL);
+	opts = kmalloc(sizeof(*opts), GFP_ATOMIC);
 	if (opts == NULL) {
 		err = -ENOMEM;
 		goto out2;
@@ -136,6 +136,7 @@ static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm,
 	enum xz_ret xz_err;
 	int avail, total = 0, k = 0;
 	struct squashfs_xz *stream = strm;
+	void *buf = NULL;
 
 	xz_dec_reset(stream->state);
 	stream->buf.in_pos = 0;
@@ -156,12 +157,20 @@ static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm,
 
 		if (stream->buf.out_pos == stream->buf.out_size) {
 			stream->buf.out = squashfs_next_page(output);
-			if (stream->buf.out != NULL) {
+			if (!IS_ERR(stream->buf.out)) {
 				stream->buf.out_pos = 0;
 				total += PAGE_CACHE_SIZE;
 			}
 		}
 
+		if (!stream->buf.out) {
+			if (!buf) {
+				buf = kmalloc(PAGE_CACHE_SIZE, GFP_ATOMIC);
+				if (!buf)
+					goto out;
+			}
+			stream->buf.out = buf;
+		}
 		xz_err = xz_dec_run(stream->state, &stream->buf);
 
 		if (stream->buf.in_pos == stream->buf.in_size && k < b)
@@ -173,11 +182,13 @@ static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm,
 	if (xz_err != XZ_STREAM_END || k < b)
 		goto out;
 
+	kfree(buf);
 	return total + stream->buf.out_pos;
 
 out:
 	for (; k < b; k++)
 		put_bh(bh[k]);
+	kfree(buf);
 
 	return -EIO;
 }
diff --git a/fs/squashfs/zlib_wrapper.c b/fs/squashfs/zlib_wrapper.c
index 8727caba6882..09c892b5308e 100644
--- a/fs/squashfs/zlib_wrapper.c
+++ b/fs/squashfs/zlib_wrapper.c
@@ -66,6 +66,7 @@ static int zlib_uncompress(struct squashfs_sb_info *msblk, void *strm,
 	struct buffer_head **bh, int b, int offset, int length,
 	struct squashfs_page_actor *output)
 {
+	void *buf = NULL;
 	int zlib_err, zlib_init = 0, k = 0;
 	z_stream *stream = strm;
 
@@ -84,10 +85,19 @@ static int zlib_uncompress(struct squashfs_sb_info *msblk, void *strm,
 
 		if (stream->avail_out == 0) {
 			stream->next_out = squashfs_next_page(output);
-			if (stream->next_out != NULL)
+			if (!IS_ERR(stream->next_out))
 				stream->avail_out = PAGE_CACHE_SIZE;
 		}
 
+		if (!stream->next_out) {
+			if (!buf) {
+				buf = kmalloc(PAGE_CACHE_SIZE, GFP_ATOMIC);
+				if (!buf)
+					goto out;
+			}
+			stream->next_out = buf;
+		}
+
 		if (!zlib_init) {
 			zlib_err = zlib_inflateInit(stream);
 			if (zlib_err != Z_OK) {
@@ -115,11 +125,13 @@ static int zlib_uncompress(struct squashfs_sb_info *msblk, void *strm,
 	if (k < b)
 		goto out;
 
+	kfree(buf);
 	return stream->total_out;
 
 out:
 	for (; k < b; k++)
 		put_bh(bh[k]);
+	kfree(buf);
 
 	return -EIO;
 }
diff --git a/fs/super.c b/fs/super.c
index d4d2591b77c8..c96434ea71e2 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -703,7 +703,8 @@ rescan:
 }
 
 /**
- *	do_remount_sb - asks filesystem to change mount options.
+ *	do_remount_sb2 - asks filesystem to change mount options.
+ *	@mnt:   mount we are looking at
  *	@sb:	superblock in question
  *	@flags:	numeric part of options
  *	@data:	the rest of options
@@ -711,7 +712,7 @@ rescan:
  *
  *	Alters the mount options of a mounted file system.
  */
-int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
+int do_remount_sb2(struct vfsmount *mnt, struct super_block *sb, int flags, void *data, int force)
 {
 	int retval;
 	int remount_ro;
@@ -753,7 +754,16 @@ int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
 		}
 	}
 
-	if (sb->s_op->remount_fs) {
+	if (mnt && sb->s_op->remount_fs2) {
+		retval = sb->s_op->remount_fs2(mnt, sb, &flags, data);
+		if (retval) {
+			if (!force)
+				goto cancel_readonly;
+			/* If forced remount, go ahead despite any errors */
+			WARN(1, "forced remount of a %s fs returned %i\n",
+			     sb->s_type->name, retval);
+		}
+	} else if (sb->s_op->remount_fs) {
 		retval = sb->s_op->remount_fs(sb, &flags, data);
 		if (retval) {
 			if (!force)
@@ -785,12 +795,17 @@ cancel_readonly:
 	return retval;
 }
 
+int do_remount_sb(struct super_block *sb, int flags, void *data, int force)
+{
+	return do_remount_sb2(NULL, sb, flags, data, force);
+}
+
 static void do_emergency_remount(struct work_struct *work)
 {
 	struct super_block *sb, *p = NULL;
 
 	spin_lock(&sb_lock);
-	list_for_each_entry(sb, &super_blocks, s_list) {
+	list_for_each_entry_reverse(sb, &super_blocks, s_list) {
 		if (hlist_unhashed(&sb->s_instances))
 			continue;
 		sb->s_count++;
@@ -1104,7 +1119,7 @@ struct dentry *mount_single(struct file_system_type *fs_type,
 EXPORT_SYMBOL(mount_single);
 
 struct dentry *
-mount_fs(struct file_system_type *type, int flags, const char *name, void *data)
+mount_fs(struct file_system_type *type, int flags, const char *name, struct vfsmount *mnt, void *data)
 {
 	struct dentry *root;
 	struct super_block *sb;
@@ -1121,7 +1136,10 @@ mount_fs(struct file_system_type *type, int flags, const char *name, void *data)
 			goto out_free_secdata;
 	}
 
-	root = type->mount(type, flags, name, data);
+	if (type->mount2)
+		root = type->mount2(mnt, type, flags, name, data);
+	else
+		root = type->mount(type, flags, name, data);
 	if (IS_ERR(root)) {
 		error = PTR_ERR(root);
 		goto out_free_secdata;
diff --git a/fs/sync.c b/fs/sync.c
index dd5d1711c7ac..452179e31c39 100644
--- a/fs/sync.c
+++ b/fs/sync.c
@@ -218,6 +218,7 @@ static int do_fsync(unsigned int fd, int datasync)
 	if (f.file) {
 		ret = vfs_fsync(f.file, datasync);
 		fdput(f);
+		inc_syscfs(current);
 	}
 	return ret;
 }
diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c
index 66cdb44616d5..d473e6e07a7e 100644
--- a/fs/userfaultfd.c
+++ b/fs/userfaultfd.c
@@ -457,7 +457,8 @@ static int userfaultfd_release(struct inode *inode, struct file *file)
 				 new_flags, vma->anon_vma,
 				 vma->vm_file, vma->vm_pgoff,
 				 vma_policy(vma),
-				 NULL_VM_UFFD_CTX);
+				 NULL_VM_UFFD_CTX,
+				 vma_get_anon_name(vma));
 		if (prev)
 			vma = prev;
 		else
@@ -833,7 +834,8 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx,
 		prev = vma_merge(mm, prev, start, vma_end, new_flags,
 				 vma->anon_vma, vma->vm_file, vma->vm_pgoff,
 				 vma_policy(vma),
-				 ((struct vm_userfaultfd_ctx){ ctx }));
+				 ((struct vm_userfaultfd_ctx){ ctx }),
+				 vma_get_anon_name(vma));
 		if (prev) {
 			vma = prev;
 			goto next;
@@ -967,7 +969,8 @@ static int userfaultfd_unregister(struct userfaultfd_ctx *ctx,
 		prev = vma_merge(mm, prev, start, vma_end, new_flags,
 				 vma->anon_vma, vma->vm_file, vma->vm_pgoff,
 				 vma_policy(vma),
-				 NULL_VM_UFFD_CTX);
+				 NULL_VM_UFFD_CTX,
+				 vma_get_anon_name(vma));
 		if (prev) {
 			vma = prev;
 			goto next;
diff --git a/fs/utimes.c b/fs/utimes.c
index cb771c30d102..a35e909cf8e3 100644
--- a/fs/utimes.c
+++ b/fs/utimes.c
@@ -91,7 +91,7 @@ static int utimes_common(struct path *path, struct timespec *times)
 	}
 retry_deleg:
 	mutex_lock(&inode->i_mutex);
-	error = notify_change(path->dentry, &newattrs, &delegated_inode);
+	error = notify_change2(path->mnt, path->dentry, &newattrs, &delegated_inode);
 	mutex_unlock(&inode->i_mutex);
 	if (delegated_inode) {
 		error = break_deleg_wait(&delegated_inode);
diff --git a/include/crypto/algapi.h b/include/crypto/algapi.h
index c9fe145f7dd3..04661e1fb625 100644
--- a/include/crypto/algapi.h
+++ b/include/crypto/algapi.h
@@ -202,6 +202,9 @@ int blkcipher_aead_walk_virt_block(struct blkcipher_desc *desc,
 				   struct blkcipher_walk *walk,
 				   struct crypto_aead *tfm,
 				   unsigned int blocksize);
+int blkcipher_ablkcipher_walk_virt(struct blkcipher_desc *desc,
+				   struct blkcipher_walk *walk,
+				   struct crypto_ablkcipher *tfm);
 
 int ablkcipher_walk_done(struct ablkcipher_request *req,
 			 struct ablkcipher_walk *walk, int err);
diff --git a/include/crypto/gf128mul.h b/include/crypto/gf128mul.h
index da2530e34b26..230760aef93b 100644
--- a/include/crypto/gf128mul.h
+++ b/include/crypto/gf128mul.h
@@ -43,7 +43,7 @@
  ---------------------------------------------------------------------------
  Issue Date: 31/01/2006
 
- An implementation of field multiplication in Galois Field GF(128)
+ An implementation of field multiplication in Galois Field GF(2^128)
 */
 
 #ifndef _CRYPTO_GF128MUL_H
@@ -65,7 +65,7 @@
  * are left and the lsb's are right. char b[16] is an array and b[0] is
  * the first octet.
  *
- * 80000000 00000000 00000000 00000000 .... 00000000 00000000 00000000
+ * 10000000 00000000 00000000 00000000 .... 00000000 00000000 00000000
  *   b[0]     b[1]     b[2]     b[3]          b[13]    b[14]    b[15]
  *
  * Every bit is a coefficient of some power of X. We can store the bits
@@ -99,21 +99,21 @@
  *
  * bbe on a little endian machine u32 x[4]:
  *
- *  MS            x[0]           LS  MS            x[1]		  LS
+ *  MS            x[0]           LS  MS            x[1]           LS
  *  ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
  *  103..96 111.104 119.112 127.120  71...64 79...72 87...80 95...88
  *
- *  MS            x[2]           LS  MS            x[3]		  LS
+ *  MS            x[2]           LS  MS            x[3]           LS
  *  ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
  *  39...32 47...40 55...48 63...56  07...00 15...08 23...16 31...24
  *
  * ble on a little endian machine
  *
- *  MS            x[0]           LS  MS            x[1]		  LS
+ *  MS            x[0]           LS  MS            x[1]           LS
  *  ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
  *  31...24 23...16 15...08 07...00  63...56 55...48 47...40 39...32
  *
- *  MS            x[2]           LS  MS            x[3]		  LS
+ *  MS            x[2]           LS  MS            x[3]           LS
  *  ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
  *  95...88 87...80 79...72 71...64  127.120 199.112 111.104 103..96
  *
@@ -127,7 +127,7 @@
  * machines this will automatically aligned to wordsize and on a 64-bit
  * machine also.
  */
-/*	Multiply a GF128 field element by x. Field elements are held in arrays
+/*  Multiply a GF128 field element by x. Field elements are held in arrays
     of bytes in which field bits 8n..8n + 7 are held in byte[n], with lower
     indexed bits placed in the more numerically significant bit positions
     within bytes.
@@ -135,62 +135,65 @@
     On little endian machines the bit indexes translate into the bit
     positions within four 32-bit words in the following way
 
-    MS            x[0]           LS  MS            x[1]		  LS
+    MS            x[0]           LS  MS            x[1]           LS
     ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
     24...31 16...23 08...15 00...07  56...63 48...55 40...47 32...39
 
-    MS            x[2]           LS  MS            x[3]		  LS
+    MS            x[2]           LS  MS            x[3]           LS
     ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
     88...95 80...87 72...79 64...71  120.127 112.119 104.111 96..103
 
     On big endian machines the bit indexes translate into the bit
     positions within four 32-bit words in the following way
 
-    MS            x[0]           LS  MS            x[1]		  LS
+    MS            x[0]           LS  MS            x[1]           LS
     ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
     00...07 08...15 16...23 24...31  32...39 40...47 48...55 56...63
 
-    MS            x[2]           LS  MS            x[3]		  LS
+    MS            x[2]           LS  MS            x[3]           LS
     ms   ls ms   ls ms   ls ms   ls  ms   ls ms   ls ms   ls ms   ls
     64...71 72...79 80...87 88...95  96..103 104.111 112.119 120.127
 */
 
-/*	A slow generic version of gf_mul, implemented for lle and bbe
- * 	It multiplies a and b and puts the result in a */
+/*  A slow generic version of gf_mul, implemented for lle, bbe, and ble.
+ *  It multiplies a and b and puts the result in a
+ */
 void gf128mul_lle(be128 *a, const be128 *b);
-
 void gf128mul_bbe(be128 *a, const be128 *b);
+void gf128mul_ble(be128 *a, const be128 *b);
 
-/* multiply by x in ble format, needed by XTS */
+/* multiply by x in ble format, needed by XTS and HEH */
 void gf128mul_x_ble(be128 *a, const be128 *b);
 
 /* 4k table optimization */
-
 struct gf128mul_4k {
 	be128 t[256];
 };
 
 struct gf128mul_4k *gf128mul_init_4k_lle(const be128 *g);
 struct gf128mul_4k *gf128mul_init_4k_bbe(const be128 *g);
+struct gf128mul_4k *gf128mul_init_4k_ble(const be128 *g);
 void gf128mul_4k_lle(be128 *a, struct gf128mul_4k *t);
 void gf128mul_4k_bbe(be128 *a, struct gf128mul_4k *t);
+void gf128mul_4k_ble(be128 *a, struct gf128mul_4k *t);
 
 static inline void gf128mul_free_4k(struct gf128mul_4k *t)
 {
-	kfree(t);
+	kzfree(t);
 }
 
 
-/* 64k table optimization, implemented for lle and bbe */
+/* 64k table optimization, implemented for lle, ble, and bbe */
 
 struct gf128mul_64k {
 	struct gf128mul_4k *t[16];
 };
 
-/* first initialize with the constant factor with which you
- * want to multiply and then call gf128_64k_lle with the other
- * factor in the first argument, the table in the second and a
- * scratch register in the third. Afterwards *a = *r. */
+/* First initialize with the constant factor with which you
+ * want to multiply and then call gf128mul_64k_bbe with the other
+ * factor in the first argument, and the table in the second.
+ * Afterwards, the result is stored in *a.
+ */
 struct gf128mul_64k *gf128mul_init_64k_lle(const be128 *g);
 struct gf128mul_64k *gf128mul_init_64k_bbe(const be128 *g);
 void gf128mul_free_64k(struct gf128mul_64k *t);
diff --git a/include/crypto/internal/hash.h b/include/crypto/internal/hash.h
index a25414ce2898..611b3d3bbab5 100644
--- a/include/crypto/internal/hash.h
+++ b/include/crypto/internal/hash.h
@@ -102,6 +102,8 @@ int shash_register_instance(struct crypto_template *tmpl,
 			    struct shash_instance *inst);
 void shash_free_instance(struct crypto_instance *inst);
 
+int crypto_grab_shash(struct crypto_shash_spawn *spawn,
+		      const char *name, u32 type, u32 mask);
 int crypto_init_shash_spawn(struct crypto_shash_spawn *spawn,
 			    struct shash_alg *alg,
 			    struct crypto_instance *inst);
@@ -111,6 +113,12 @@ static inline void crypto_drop_shash(struct crypto_shash_spawn *spawn)
 	crypto_drop_spawn(&spawn->base);
 }
 
+static inline struct shash_alg *crypto_spawn_shash_alg(
+	struct crypto_shash_spawn *spawn)
+{
+	return container_of(spawn->base.alg, struct shash_alg, base);
+}
+
 struct shash_alg *shash_attr_alg(struct rtattr *rta, u32 type, u32 mask);
 
 int shash_ahash_update(struct ahash_request *req, struct shash_desc *desc);
diff --git a/include/linux/Kbuild b/include/linux/Kbuild
new file mode 100644
index 000000000000..a4608897a5f3
--- /dev/null
+++ b/include/linux/Kbuild
@@ -0,0 +1,2 @@
+header-y += if_pppolac.h
+header-y += if_pppopns.h
diff --git a/include/linux/amba/mmci.h b/include/linux/amba/mmci.h
index 8c98113069ce..eff56cb0016a 100644
--- a/include/linux/amba/mmci.h
+++ b/include/linux/amba/mmci.h
@@ -5,6 +5,15 @@
 #define AMBA_MMCI_H
 
 #include <linux/mmc/host.h>
+#include <linux/mmc/card.h>
+#include <linux/mmc/sdio_func.h>
+
+struct embedded_sdio_data {
+        struct sdio_cis cis;
+        struct sdio_cccr cccr;
+        struct sdio_embedded_func *funcs;
+        int num_funcs;
+};
 
 /**
  * struct mmci_platform_data - platform configuration for the MMCI
@@ -31,6 +40,7 @@ struct mmci_platform_data {
 	int	gpio_wp;
 	int	gpio_cd;
 	bool	cd_invert;
+	struct embedded_sdio_data *embedded_sdio;
 };
 
 #endif
diff --git a/include/linux/android_aid.h b/include/linux/android_aid.h
new file mode 100644
index 000000000000..6f1fa1792dfc
--- /dev/null
+++ b/include/linux/android_aid.h
@@ -0,0 +1,28 @@
+/* include/linux/android_aid.h
+ *
+ * Copyright (C) 2008 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _LINUX_ANDROID_AID_H
+#define _LINUX_ANDROID_AID_H
+
+/* AIDs that the kernel treats differently */
+#define AID_OBSOLETE_000 KGIDT_INIT(3001)  /* was NET_BT_ADMIN */
+#define AID_OBSOLETE_001 KGIDT_INIT(3002)  /* was NET_BT */
+#define AID_INET         KGIDT_INIT(3003)
+#define AID_NET_RAW      KGIDT_INIT(3004)
+#define AID_NET_ADMIN    KGIDT_INIT(3005)
+#define AID_NET_BW_STATS KGIDT_INIT(3006)  /* read bandwidth statistics */
+#define AID_NET_BW_ACCT  KGIDT_INIT(3007)  /* change bandwidth statistics accounting */
+
+#endif
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index fe14382f9664..a9562bb029d0 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -197,6 +197,9 @@ struct request {
 
 	/* for bidi */
 	struct request *next_rq;
+
+	ktime_t			lat_hist_io_start;
+	int			lat_hist_enabled;
 };
 
 static inline unsigned short req_get_ioprio(struct request *req)
@@ -1656,6 +1659,79 @@ extern int bdev_write_page(struct block_device *, sector_t, struct page *,
 						struct writeback_control *);
 extern long bdev_direct_access(struct block_device *, sector_t,
 		void __pmem **addr, unsigned long *pfn, long size);
+
+/*
+ * X-axis for IO latency histogram support.
+ */
+static const u_int64_t latency_x_axis_us[] = {
+	100,
+	200,
+	300,
+	400,
+	500,
+	600,
+	700,
+	800,
+	900,
+	1000,
+	1200,
+	1400,
+	1600,
+	1800,
+	2000,
+	2500,
+	3000,
+	4000,
+	5000,
+	6000,
+	7000,
+	9000,
+	10000
+};
+
+#define BLK_IO_LAT_HIST_DISABLE         0
+#define BLK_IO_LAT_HIST_ENABLE          1
+#define BLK_IO_LAT_HIST_ZERO            2
+
+struct io_latency_state {
+	u_int64_t	latency_y_axis_read[ARRAY_SIZE(latency_x_axis_us) + 1];
+	u_int64_t	latency_reads_elems;
+	u_int64_t	latency_y_axis_write[ARRAY_SIZE(latency_x_axis_us) + 1];
+	u_int64_t	latency_writes_elems;
+};
+
+static inline void
+blk_update_latency_hist(struct io_latency_state *s,
+			int read,
+			u_int64_t delta_us)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(latency_x_axis_us); i++) {
+		if (delta_us < (u_int64_t)latency_x_axis_us[i]) {
+			if (read)
+				s->latency_y_axis_read[i]++;
+			else
+				s->latency_y_axis_write[i]++;
+			break;
+		}
+	}
+	if (i == ARRAY_SIZE(latency_x_axis_us)) {
+		/* Overflowed the histogram */
+		if (read)
+			s->latency_y_axis_read[i]++;
+		else
+			s->latency_y_axis_write[i]++;
+	}
+	if (read)
+		s->latency_reads_elems++;
+	else
+		s->latency_writes_elems++;
+}
+
+void blk_zero_latency_hist(struct io_latency_state *s);
+ssize_t blk_latency_hist_show(struct io_latency_state *s, char *buf);
+
 #else /* CONFIG_BLOCK */
 
 struct block_device;
diff --git a/include/linux/cgroup_subsys.h b/include/linux/cgroup_subsys.h
index 1a96fdaa33d5..e133705d794a 100644
--- a/include/linux/cgroup_subsys.h
+++ b/include/linux/cgroup_subsys.h
@@ -26,6 +26,10 @@ SUBSYS(cpu)
 SUBSYS(cpuacct)
 #endif
 
+#if IS_ENABLED(CONFIG_CGROUP_SCHEDTUNE)
+SUBSYS(schedtune)
+#endif
+
 #if IS_ENABLED(CONFIG_BLK_CGROUP)
 SUBSYS(io)
 #endif
diff --git a/include/linux/compiler-clang.h b/include/linux/compiler-clang.h
index d1e49d52b640..de179993e039 100644
--- a/include/linux/compiler-clang.h
+++ b/include/linux/compiler-clang.h
@@ -10,3 +10,8 @@
 #undef uninitialized_var
 #define uninitialized_var(x) x = *(&(x))
 #endif
+
+/* same as gcc, this was present in clang-2.6 so we can assume it works
+ * with any version that can compile the kernel
+ */
+#define __UNIQUE_ID(prefix) __PASTE(__PASTE(__UNIQUE_ID_, prefix), __COUNTER__)
diff --git a/include/linux/compiler-gcc.h b/include/linux/compiler-gcc.h
index 287e698c28de..557dae96ce74 100644
--- a/include/linux/compiler-gcc.h
+++ b/include/linux/compiler-gcc.h
@@ -66,18 +66,22 @@
 
 /*
  * Force always-inline if the user requests it so via the .config,
- * or if gcc is too old:
+ * or if gcc is too old.
+ * GCC does not warn about unused static inline functions for
+ * -Wunused-function.  This turns out to avoid the need for complex #ifdef
+ * directives.  Suppress the warning in clang as well by using "unused"
+ * function attribute, which is redundant but not harmful for gcc.
  */
 #if !defined(CONFIG_ARCH_SUPPORTS_OPTIMIZED_INLINING) ||		\
     !defined(CONFIG_OPTIMIZE_INLINING) || (__GNUC__ < 4)
-#define inline		inline		__attribute__((always_inline)) notrace
-#define __inline__	__inline__	__attribute__((always_inline)) notrace
-#define __inline	__inline	__attribute__((always_inline)) notrace
+#define inline inline		__attribute__((always_inline,unused)) notrace
+#define __inline__ __inline__	__attribute__((always_inline,unused)) notrace
+#define __inline __inline	__attribute__((always_inline,unused)) notrace
 #else
 /* A lot of inline functions can cause havoc with function tracing */
-#define inline		inline		notrace
-#define __inline__	__inline__	notrace
-#define __inline	__inline	notrace
+#define inline inline		__attribute__((unused)) notrace
+#define __inline__ __inline__	__attribute__((unused)) notrace
+#define __inline __inline	__attribute__((unused)) notrace
 #endif
 
 #define __always_inline	inline __attribute__((always_inline))
diff --git a/include/linux/cpu.h b/include/linux/cpu.h
index 3ea9aae2387d..1c95ef5a6ed9 100644
--- a/include/linux/cpu.h
+++ b/include/linux/cpu.h
@@ -283,4 +283,11 @@ bool cpu_wait_death(unsigned int cpu, int seconds);
 bool cpu_report_death(void);
 #endif /* #ifdef CONFIG_HOTPLUG_CPU */
 
+#define IDLE_START 1
+#define IDLE_END 2
+
+void idle_notifier_register(struct notifier_block *n);
+void idle_notifier_unregister(struct notifier_block *n);
+void idle_notifier_call_chain(unsigned long val);
+
 #endif /* _LINUX_CPU_H_ */
diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
index 177c7680c1a8..490d577aa144 100644
--- a/include/linux/cpufreq.h
+++ b/include/linux/cpufreq.h
@@ -107,6 +107,30 @@ struct cpufreq_policy {
 	 */
 	struct rw_semaphore	rwsem;
 
+
+	/*
+	 * Fast switch flags:
+	 * - fast_switch_possible should be set by the driver if it can
+	 *   guarantee that frequency can be changed on any CPU sharing the
+	 *   policy and that the change will affect all of the policy CPUs then.
+	 * - fast_switch_enabled is to be set by governors that support fast
+	 *   freqnency switching with the help of cpufreq_enable_fast_switch().
+	 */
+	bool                    fast_switch_possible;
+	bool                    fast_switch_enabled;
+
+	/*
+	 * Preferred average time interval between consecutive invocations of
+	 * the driver to set the frequency for this policy.  To be set by the
+	 * scaling driver (0, which is the default, means no preference).
+	 */
+	unsigned int		up_transition_delay_us;
+	unsigned int		down_transition_delay_us;
+
+	 /* Cached frequency lookup from cpufreq_driver_resolve_freq. */
+	unsigned int cached_target_freq;
+	int cached_resolved_idx;
+
 	/* Synchronization for frequency transitions */
 	bool			transition_ongoing; /* Tracks transition status */
 	spinlock_t		transition_lock;
@@ -160,6 +184,7 @@ u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy);
 int cpufreq_get_policy(struct cpufreq_policy *policy, unsigned int cpu);
 int cpufreq_update_policy(unsigned int cpu);
 bool have_governor_per_policy(void);
+bool cpufreq_driver_is_slow(void);
 struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy);
 #else
 static inline unsigned int cpufreq_get(unsigned int cpu)
@@ -317,6 +342,14 @@ struct cpufreq_driver {
  */
 #define CPUFREQ_NEED_INITIAL_FREQ_CHECK	(1 << 5)
 
+/*
+ * Indicates that it is safe to call cpufreq_driver_target from
+ * non-interruptable context in scheduler hot paths.  Drivers must
+ * opt-in to this flag, as the safe default is that they might sleep
+ * or be too slow for hot path use.
+ */
+#define CPUFREQ_DRIVER_FAST		(1 << 6)
+
 int cpufreq_register_driver(struct cpufreq_driver *driver_data);
 int cpufreq_unregister_driver(struct cpufreq_driver *driver_data);
 
@@ -462,6 +495,8 @@ int cpufreq_driver_target(struct cpufreq_policy *policy,
 int __cpufreq_driver_target(struct cpufreq_policy *policy,
 				   unsigned int target_freq,
 				   unsigned int relation);
+unsigned int cpufreq_driver_resolve_freq(struct cpufreq_policy *policy,
+					 unsigned int target_freq);
 int cpufreq_register_governor(struct cpufreq_governor *governor);
 void cpufreq_unregister_governor(struct cpufreq_governor *governor);
 
@@ -487,8 +522,48 @@ extern struct cpufreq_governor cpufreq_gov_ondemand;
 #elif defined(CONFIG_CPU_FREQ_DEFAULT_GOV_CONSERVATIVE)
 extern struct cpufreq_governor cpufreq_gov_conservative;
 #define CPUFREQ_DEFAULT_GOVERNOR	(&cpufreq_gov_conservative)
+#elif defined(CONFIG_CPU_FREQ_DEFAULT_GOV_INTERACTIVE)
+extern struct cpufreq_governor cpufreq_gov_interactive;
+#define CPUFREQ_DEFAULT_GOVERNOR	(&cpufreq_gov_interactive)
+#elif defined(CONFIG_CPU_FREQ_DEFAULT_GOV_SCHED)
+extern struct cpufreq_governor cpufreq_gov_sched;
+#define CPUFREQ_DEFAULT_GOVERNOR	(&cpufreq_gov_sched)
+#elif defined(CONFIG_CPU_FREQ_DEFAULT_GOV_SCHEDUTIL)
+extern struct cpufreq_governor cpufreq_gov_schedutil;
+#define CPUFREQ_DEFAULT_GOVERNOR	(&cpufreq_gov_schedutil)
 #endif
 
+static inline void cpufreq_policy_apply_limits(struct cpufreq_policy *policy)
+{
+	if (policy->max < policy->cur)
+		__cpufreq_driver_target(policy, policy->max, CPUFREQ_RELATION_H);
+	else if (policy->min > policy->cur)
+		__cpufreq_driver_target(policy, policy->min, CPUFREQ_RELATION_L);
+}
+
+/* Governor attribute set */
+struct gov_attr_set {
+	struct kobject kobj;
+	struct list_head policy_list;
+	struct mutex update_lock;
+	int usage_count;
+};
+
+/* sysfs ops for cpufreq governors */
+extern const struct sysfs_ops governor_sysfs_ops;
+
+void gov_attr_set_init(struct gov_attr_set *attr_set, struct list_head *list_node);
+void gov_attr_set_get(struct gov_attr_set *attr_set, struct list_head *list_node);
+unsigned int gov_attr_set_put(struct gov_attr_set *attr_set, struct list_head *list_node);
+
+/* Governor sysfs attribute */
+struct governor_attr {
+	struct attribute attr;
+	ssize_t (*show)(struct gov_attr_set *attr_set, char *buf);
+	ssize_t (*store)(struct gov_attr_set *attr_set, const char *buf,
+			 size_t count);
+};
+
 /*********************************************************************
  *                     FREQUENCY TABLE HELPERS                       *
  *********************************************************************/
@@ -616,4 +691,8 @@ unsigned int cpufreq_generic_get(unsigned int cpu);
 int cpufreq_generic_init(struct cpufreq_policy *policy,
 		struct cpufreq_frequency_table *table,
 		unsigned int transition_latency);
+
+struct sched_domain;
+unsigned long cpufreq_scale_freq_capacity(struct sched_domain *sd, int cpu);
+unsigned long cpufreq_scale_max_freq_capacity(int cpu);
 #endif /* _LINUX_CPUFREQ_H */
diff --git a/include/linux/cpuidle.h b/include/linux/cpuidle.h
index 786ad32631a6..6eae1576499e 100644
--- a/include/linux/cpuidle.h
+++ b/include/linux/cpuidle.h
@@ -204,7 +204,7 @@ static inline int cpuidle_enter_freeze(struct cpuidle_driver *drv,
 #endif
 
 /* kernel/sched/idle.c */
-extern void sched_idle_set_state(struct cpuidle_state *idle_state);
+extern void sched_idle_set_state(struct cpuidle_state *idle_state, int index);
 extern void default_idle_call(void);
 
 #ifdef CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED
diff --git a/include/linux/dcache.h b/include/linux/dcache.h
index d516847e0fae..51b47269e2fe 100644
--- a/include/linux/dcache.h
+++ b/include/linux/dcache.h
@@ -161,6 +161,7 @@ struct dentry_operations {
 	struct vfsmount *(*d_automount)(struct path *);
 	int (*d_manage)(struct dentry *, bool);
 	struct inode *(*d_select_inode)(struct dentry *, unsigned);
+	void (*d_canonical_path)(const struct path *, struct path *);
 	struct dentry *(*d_real)(struct dentry *, struct inode *);
 } ____cacheline_aligned;
 
@@ -228,6 +229,7 @@ struct dentry_operations {
 #define DCACHE_MAY_FREE			0x00800000
 #define DCACHE_FALLTHRU			0x01000000 /* Fall through to lower layer */
 #define DCACHE_OP_SELECT_INODE		0x02000000 /* Unioned entry: dcache op selects inode */
+#define DCACHE_ENCRYPTED_WITH_KEY	0x04000000 /* dir is encrypted with a valid key */
 #define DCACHE_OP_REAL			0x08000000
 
 extern seqlock_t rename_lock;
diff --git a/include/linux/device-mapper.h b/include/linux/device-mapper.h
index 899ab9f8549e..f3f87db34429 100644
--- a/include/linux/device-mapper.h
+++ b/include/linux/device-mapper.h
@@ -383,6 +383,12 @@ void dm_set_mdptr(struct mapped_device *md, void *ptr);
 void *dm_get_mdptr(struct mapped_device *md);
 
 /*
+ * Export the device via the ioctl interface (uses mdptr).
+ */
+int dm_ioctl_export(struct mapped_device *md, const char *name,
+		    const char *uuid);
+
+/*
  * A device can still be used while suspended, but I/O is deferred.
  */
 int dm_suspend(struct mapped_device *md, unsigned suspend_flags);
@@ -409,6 +415,13 @@ union map_info *dm_get_rq_mapinfo(struct request *rq);
 
 struct queue_limits *dm_get_queue_limits(struct mapped_device *md);
 
+void dm_lock_md_type(struct mapped_device *md);
+void dm_unlock_md_type(struct mapped_device *md);
+void dm_set_md_type(struct mapped_device *md, unsigned type);
+unsigned dm_get_md_type(struct mapped_device *md);
+int dm_setup_md_queue(struct mapped_device *md);
+unsigned dm_table_get_type(struct dm_table *t);
+
 /*
  * Geometry functions.
  */
diff --git a/include/linux/efi.h b/include/linux/efi.h
index 333d0ca6940f..516d83041206 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -1292,7 +1292,7 @@ efi_status_t handle_cmdline_files(efi_system_table_t *sys_table_arg,
 				  unsigned long *load_addr,
 				  unsigned long *load_size);
 
-efi_status_t efi_parse_options(char *cmdline);
+efi_status_t efi_parse_options(char const *cmdline);
 
 bool efi_runtime_disabled(void);
 #endif /* _LINUX_EFI_H */
diff --git a/include/linux/f2fs_fs.h b/include/linux/f2fs_fs.h
index 3d6e6ce44c5c..c2a975e4a711 100644
--- a/include/linux/f2fs_fs.h
+++ b/include/linux/f2fs_fs.h
@@ -21,7 +21,7 @@
 #define F2FS_BLKSIZE			4096	/* support only 4KB block */
 #define F2FS_BLKSIZE_BITS		12	/* bits for F2FS_BLKSIZE */
 #define F2FS_MAX_EXTENSION		64	/* # of extension entries */
-#define F2FS_BLK_ALIGN(x)	(((x) + F2FS_BLKSIZE - 1) / F2FS_BLKSIZE)
+#define F2FS_BLK_ALIGN(x)	(((x) + F2FS_BLKSIZE - 1) >> F2FS_BLKSIZE_BITS)
 
 #define NULL_ADDR		((block_t)0)	/* used as block_t addresses */
 #define NEW_ADDR		((block_t)-1)	/* used as block_t addresses */
@@ -36,6 +36,12 @@
 #define F2FS_NODE_INO(sbi)	(sbi->node_ino_num)
 #define F2FS_META_INO(sbi)	(sbi->meta_ino_num)
 
+#define F2FS_IO_SIZE(sbi)	(1 << (sbi)->write_io_size_bits) /* Blocks */
+#define F2FS_IO_SIZE_KB(sbi)	(1 << ((sbi)->write_io_size_bits + 2)) /* KB */
+#define F2FS_IO_SIZE_BYTES(sbi)	(1 << ((sbi)->write_io_size_bits + 12)) /* B */
+#define F2FS_IO_SIZE_BITS(sbi)	((sbi)->write_io_size_bits) /* power of 2 */
+#define F2FS_IO_SIZE_MASK(sbi)	(F2FS_IO_SIZE(sbi) - 1)
+
 /* This flag is used by node and meta inodes, and by recovery */
 #define GFP_F2FS_ZERO		(GFP_NOFS | __GFP_ZERO)
 #define GFP_F2FS_HIGH_ZERO	(GFP_NOFS | __GFP_ZERO | __GFP_HIGHMEM)
@@ -51,10 +57,18 @@
 #define MAX_ACTIVE_DATA_LOGS	8
 
 #define VERSION_LEN	256
+#define MAX_VOLUME_NAME		512
+#define MAX_PATH_LEN		64
+#define MAX_DEVICES		8
 
 /*
  * For superblock
  */
+struct f2fs_device {
+	__u8 path[MAX_PATH_LEN];
+	__le32 total_segments;
+} __packed;
+
 struct f2fs_super_block {
 	__le32 magic;			/* Magic Number */
 	__le16 major_ver;		/* Major Version */
@@ -84,7 +98,7 @@ struct f2fs_super_block {
 	__le32 node_ino;		/* node inode number */
 	__le32 meta_ino;		/* meta inode number */
 	__u8 uuid[16];			/* 128-bit uuid for volume */
-	__le16 volume_name[512];	/* volume name */
+	__le16 volume_name[MAX_VOLUME_NAME];	/* volume name */
 	__le32 extension_count;		/* # of extensions below */
 	__u8 extension_list[F2FS_MAX_EXTENSION][8];	/* extension array */
 	__le32 cp_payload;
@@ -93,12 +107,16 @@ struct f2fs_super_block {
 	__le32 feature;			/* defined features */
 	__u8 encryption_level;		/* versioning level for encryption */
 	__u8 encrypt_pw_salt[16];	/* Salt used for string2key algorithm */
-	__u8 reserved[871];		/* valid reserved region */
+	struct f2fs_device devs[MAX_DEVICES];	/* device list */
+	__u8 reserved[327];		/* valid reserved region */
 } __packed;
 
 /*
  * For checkpoint
  */
+#define CP_TRIMMED_FLAG		0x00000100
+#define CP_NAT_BITS_FLAG	0x00000080
+#define CP_CRC_RECOVERY_FLAG	0x00000040
 #define CP_FASTBOOT_FLAG	0x00000020
 #define CP_FSCK_FLAG		0x00000010
 #define CP_ERROR_FLAG		0x00000008
@@ -168,13 +186,15 @@ struct f2fs_extent {
 #define F2FS_NAME_LEN		255
 #define F2FS_INLINE_XATTR_ADDRS	50	/* 200 bytes for inline xattrs */
 #define DEF_ADDRS_PER_INODE	923	/* Address Pointers in an Inode */
+#define CUR_ADDRS_PER_INODE(inode)	(DEF_ADDRS_PER_INODE - \
+					get_extra_isize(inode))
 #define DEF_NIDS_PER_INODE	5	/* Node IDs in an Inode */
-#define ADDRS_PER_INODE(fi)	addrs_per_inode(fi)
+#define ADDRS_PER_INODE(inode)	addrs_per_inode(inode)
 #define ADDRS_PER_BLOCK		1018	/* Address Pointers in a Direct Block */
 #define NIDS_PER_BLOCK		1018	/* Node IDs in an Indirect Block */
 
-#define ADDRS_PER_PAGE(page, fi)	\
-	(IS_INODE(page) ? ADDRS_PER_INODE(fi) : ADDRS_PER_BLOCK)
+#define ADDRS_PER_PAGE(page, inode)	\
+	(IS_INODE(page) ? ADDRS_PER_INODE(inode) : ADDRS_PER_BLOCK)
 
 #define	NODE_DIR1_BLOCK		(DEF_ADDRS_PER_INODE + 1)
 #define	NODE_DIR2_BLOCK		(DEF_ADDRS_PER_INODE + 2)
@@ -187,9 +207,7 @@ struct f2fs_extent {
 #define F2FS_INLINE_DENTRY	0x04	/* file inline dentry flag */
 #define F2FS_DATA_EXIST		0x08	/* file inline data exist flag */
 #define F2FS_INLINE_DOTS	0x10	/* file having implicit dot dentries */
-
-#define MAX_INLINE_DATA		(sizeof(__le32) * (DEF_ADDRS_PER_INODE - \
-						F2FS_INLINE_XATTR_ADDRS - 1))
+#define F2FS_EXTRA_ATTR		0x20	/* file having extra attribute */
 
 struct f2fs_inode {
 	__le16 i_mode;			/* file mode */
@@ -217,8 +235,16 @@ struct f2fs_inode {
 
 	struct f2fs_extent i_ext;	/* caching a largest extent */
 
-	__le32 i_addr[DEF_ADDRS_PER_INODE];	/* Pointers to data blocks */
-
+	union {
+		struct {
+			__le16 i_extra_isize;	/* extra inode attribute size */
+			__le16 i_padding;	/* padding */
+			__le32 i_projid;	/* project id */
+			__le32 i_inode_checksum;/* inode meta checksum */
+			__le32 i_extra_end[0];	/* for attribute size calculation */
+		};
+		__le32 i_addr[DEF_ADDRS_PER_INODE];	/* Pointers to data blocks */
+	};
 	__le32 i_nid[DEF_NIDS_PER_INODE];	/* direct(2), indirect(2),
 						double_indirect(1) node id */
 } __packed;
@@ -261,7 +287,8 @@ struct f2fs_node {
 /*
  * For NAT entries
  */
-#define NAT_ENTRY_PER_BLOCK (PAGE_CACHE_SIZE / sizeof(struct f2fs_nat_entry))
+#define NAT_ENTRY_PER_BLOCK (PAGE_SIZE / sizeof(struct f2fs_nat_entry))
+#define NAT_ENTRY_BITMAP_SIZE	((NAT_ENTRY_PER_BLOCK + 7) / 8)
 
 struct f2fs_nat_entry {
 	__u8 version;		/* latest version of cached nat entry */
@@ -281,7 +308,7 @@ struct f2fs_nat_block {
  * Not allow to change this.
  */
 #define SIT_VBLOCK_MAP_SIZE 64
-#define SIT_ENTRY_PER_BLOCK (PAGE_CACHE_SIZE / sizeof(struct f2fs_sit_entry))
+#define SIT_ENTRY_PER_BLOCK (PAGE_SIZE / sizeof(struct f2fs_sit_entry))
 
 /*
  * F2FS uses 4 bytes to represent block address. As a result, supported size of
@@ -350,7 +377,7 @@ struct f2fs_summary {
 
 struct summary_footer {
 	unsigned char entry_type;	/* SUM_TYPE_XXX */
-	__u32 check_sum;		/* summary checksum */
+	__le32 check_sum;		/* summary checksum */
 } __packed;
 
 #define SUM_JOURNAL_SIZE	(F2FS_BLKSIZE - SUM_FOOTER_SIZE -\
@@ -363,6 +390,12 @@ struct summary_footer {
 				sizeof(struct sit_journal_entry))
 #define SIT_JOURNAL_RESERVED	((SUM_JOURNAL_SIZE - 2) %\
 				sizeof(struct sit_journal_entry))
+
+/* Reserved area should make size of f2fs_extra_info equals to
+ * that of nat_journal and sit_journal.
+ */
+#define EXTRA_INFO_RESERVED	(SUM_JOURNAL_SIZE - 2 - 8)
+
 /*
  * frequently updated NAT/SIT entries can be stored in the spare area in
  * summary blocks
@@ -392,18 +425,28 @@ struct sit_journal {
 	__u8 reserved[SIT_JOURNAL_RESERVED];
 } __packed;
 
-/* 4KB-sized summary block structure */
-struct f2fs_summary_block {
-	struct f2fs_summary entries[ENTRIES_IN_SUM];
+struct f2fs_extra_info {
+	__le64 kbytes_written;
+	__u8 reserved[EXTRA_INFO_RESERVED];
+} __packed;
+
+struct f2fs_journal {
 	union {
 		__le16 n_nats;
 		__le16 n_sits;
 	};
-	/* spare area is used by NAT or SIT journals */
+	/* spare area is used by NAT or SIT journals or extra info */
 	union {
 		struct nat_journal nat_j;
 		struct sit_journal sit_j;
+		struct f2fs_extra_info info;
 	};
+} __packed;
+
+/* 4KB-sized summary block structure */
+struct f2fs_summary_block {
+	struct f2fs_summary entries[ENTRIES_IN_SUM];
+	struct f2fs_journal journal;
 	struct summary_footer footer;
 } __packed;
 
@@ -430,7 +473,7 @@ typedef __le32	f2fs_hash_t;
 #define MAX_DIR_BUCKETS		(1 << ((MAX_DIR_HASH_DEPTH / 2) - 1))
 
 /*
- * space utilization of regular dentry and inline dentry
+ * space utilization of regular dentry and inline dentry (w/o extra reservation)
  *		regular dentry			inline dentry
  * bitmap	1 * 27 = 27			1 * 23 = 23
  * reserved	1 * 3 = 3			1 * 7 = 7
@@ -466,24 +509,6 @@ struct f2fs_dentry_block {
 	__u8 filename[NR_DENTRY_IN_BLOCK][F2FS_SLOT_LEN];
 } __packed;
 
-/* for inline dir */
-#define NR_INLINE_DENTRY	(MAX_INLINE_DATA * BITS_PER_BYTE / \
-				((SIZE_OF_DIR_ENTRY + F2FS_SLOT_LEN) * \
-				BITS_PER_BYTE + 1))
-#define INLINE_DENTRY_BITMAP_SIZE	((NR_INLINE_DENTRY + \
-					BITS_PER_BYTE - 1) / BITS_PER_BYTE)
-#define INLINE_RESERVED_SIZE	(MAX_INLINE_DATA - \
-				((SIZE_OF_DIR_ENTRY + F2FS_SLOT_LEN) * \
-				NR_INLINE_DENTRY + INLINE_DENTRY_BITMAP_SIZE))
-
-/* inline directory entry structure */
-struct f2fs_inline_dentry {
-	__u8 dentry_bitmap[INLINE_DENTRY_BITMAP_SIZE];
-	__u8 reserved[INLINE_RESERVED_SIZE];
-	struct f2fs_dir_entry dentry[NR_INLINE_DENTRY];
-	__u8 filename[NR_INLINE_DENTRY][F2FS_SLOT_LEN];
-} __packed;
-
 /* file types used in inode_info->flags */
 enum {
 	F2FS_FT_UNKNOWN,
@@ -497,4 +522,8 @@ enum {
 	F2FS_FT_MAX
 };
 
+#define S_SHIFT 12
+
+#define	F2FS_DEF_PROJID		0	/* default project ID */
+
 #endif  /* _LINUX_F2FS_FS_H */
diff --git a/include/linux/fence.h b/include/linux/fence.h
index bb522011383b..ce4434814b8b 100644
--- a/include/linux/fence.h
+++ b/include/linux/fence.h
@@ -107,6 +107,7 @@ struct fence_cb {
  * @get_driver_name: returns the driver name.
  * @get_timeline_name: return the name of the context this fence belongs to.
  * @enable_signaling: enable software signaling of fence.
+ * @disable_signaling: disable software signaling of fence (optional).
  * @signaled: [optional] peek whether the fence is signaled, can be null.
  * @wait: custom wait implementation, or fence_default_wait.
  * @release: [optional] called on destruction of fence, can be null
@@ -166,6 +167,7 @@ struct fence_ops {
 	const char * (*get_driver_name)(struct fence *fence);
 	const char * (*get_timeline_name)(struct fence *fence);
 	bool (*enable_signaling)(struct fence *fence);
+	void (*disable_signaling)(struct fence *fence);
 	bool (*signaled)(struct fence *fence);
 	signed long (*wait)(struct fence *fence, bool intr, signed long timeout);
 	void (*release)(struct fence *fence);
diff --git a/include/linux/freezer.h b/include/linux/freezer.h
index 6b7fd9cf5ea2..dd03e837ebb7 100644
--- a/include/linux/freezer.h
+++ b/include/linux/freezer.h
@@ -231,7 +231,7 @@ static inline long freezable_schedule_timeout_killable_unsafe(long timeout)
  * call this with locks held.
  */
 static inline int freezable_schedule_hrtimeout_range(ktime_t *expires,
-		unsigned long delta, const enum hrtimer_mode mode)
+		u64 delta, const enum hrtimer_mode mode)
 {
 	int __retval;
 	freezer_do_not_count();
diff --git a/include/linux/fs.h b/include/linux/fs.h
index c8decb7075d6..5b79adb9782e 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -52,6 +52,8 @@ struct swap_info_struct;
 struct seq_file;
 struct workqueue_struct;
 struct iov_iter;
+struct fscrypt_info;
+struct fscrypt_operations;
 
 extern void __init inode_init(void);
 extern void __init inode_init_early(void);
@@ -677,6 +679,9 @@ struct inode {
 	struct hlist_head	i_fsnotify_marks;
 #endif
 
+#if IS_ENABLED(CONFIG_FS_ENCRYPTION)
+	struct fscrypt_info	*i_crypt_info;
+#endif
 	void			*i_private; /* fs or device private pointer */
 };
 
@@ -1337,6 +1342,8 @@ struct super_block {
 #endif
 	const struct xattr_handler **s_xattr;
 
+	const struct fscrypt_operations	*s_cop;
+
 	struct hlist_bl_head	s_anon;		/* anonymous dentries for (nfs) exporting */
 	struct list_head	s_mounts;	/* list of mounts; _not_ for fs use */
 	struct block_device	*s_bdev;
@@ -1544,13 +1551,21 @@ extern bool inode_owner_or_capable(const struct inode *inode);
  * VFS helper functions..
  */
 extern int vfs_create(struct inode *, struct dentry *, umode_t, bool);
+extern int vfs_create2(struct vfsmount *, struct inode *, struct dentry *, umode_t, bool);
 extern int vfs_mkdir(struct inode *, struct dentry *, umode_t);
+extern int vfs_mkdir2(struct vfsmount *, struct inode *, struct dentry *, umode_t);
 extern int vfs_mknod(struct inode *, struct dentry *, umode_t, dev_t);
+extern int vfs_mknod2(struct vfsmount *, struct inode *, struct dentry *, umode_t, dev_t);
 extern int vfs_symlink(struct inode *, struct dentry *, const char *);
+extern int vfs_symlink2(struct vfsmount *, struct inode *, struct dentry *, const char *);
 extern int vfs_link(struct dentry *, struct inode *, struct dentry *, struct inode **);
+extern int vfs_link2(struct vfsmount *, struct dentry *, struct inode *, struct dentry *, struct inode **);
 extern int vfs_rmdir(struct inode *, struct dentry *);
+extern int vfs_rmdir2(struct vfsmount *, struct inode *, struct dentry *);
 extern int vfs_unlink(struct inode *, struct dentry *, struct inode **);
+extern int vfs_unlink2(struct vfsmount *, struct inode *, struct dentry *, struct inode **);
 extern int vfs_rename(struct inode *, struct dentry *, struct inode *, struct dentry *, struct inode **, unsigned int);
+extern int vfs_rename2(struct vfsmount *, struct inode *, struct dentry *, struct inode *, struct dentry *, struct inode **, unsigned int);
 extern int vfs_whiteout(struct inode *, struct dentry *);
 
 /*
@@ -1676,6 +1691,7 @@ struct inode_operations {
 	struct dentry * (*lookup) (struct inode *,struct dentry *, unsigned int);
 	const char * (*follow_link) (struct dentry *, void **);
 	int (*permission) (struct inode *, int);
+	int (*permission2) (struct vfsmount *, struct inode *, int);
 	struct posix_acl * (*get_acl)(struct inode *, int);
 
 	int (*readlink) (struct dentry *, char __user *,int);
@@ -1693,6 +1709,7 @@ struct inode_operations {
 	int (*rename2) (struct inode *, struct dentry *,
 			struct inode *, struct dentry *, unsigned int);
 	int (*setattr) (struct dentry *, struct iattr *);
+	int (*setattr2) (struct vfsmount *, struct dentry *, struct iattr *);
 	int (*getattr) (struct vfsmount *mnt, struct dentry *, struct kstat *);
 	int (*setxattr) (struct dentry *, const char *,const void *,size_t,int);
 	ssize_t (*getxattr) (struct dentry *, const char *, void *, size_t);
@@ -1738,9 +1755,13 @@ struct super_operations {
 	int (*unfreeze_fs) (struct super_block *);
 	int (*statfs) (struct dentry *, struct kstatfs *);
 	int (*remount_fs) (struct super_block *, int *, char *);
+	int (*remount_fs2) (struct vfsmount *, struct super_block *, int *, char *);
+	void *(*clone_mnt_data) (void *);
+	void (*copy_mnt_data) (void *, void *);
 	void (*umount_begin) (struct super_block *);
 
 	int (*show_options)(struct seq_file *, struct dentry *);
+	int (*show_options2)(struct vfsmount *,struct seq_file *, struct dentry *);
 	int (*show_devname)(struct seq_file *, struct dentry *);
 	int (*show_path)(struct seq_file *, struct dentry *);
 	int (*show_stats)(struct seq_file *, struct dentry *);
@@ -1972,6 +1993,9 @@ struct file_system_type {
 #define FS_RENAME_DOES_D_MOVE	32768	/* FS will handle d_move() during rename() internally. */
 	struct dentry *(*mount) (struct file_system_type *, int,
 		       const char *, void *);
+	struct dentry *(*mount2) (struct vfsmount *, struct file_system_type *, int,
+			       const char *, void *);
+	void *(*alloc_mnt_data) (void);
 	void (*kill_sb) (struct super_block *);
 	struct module *owner;
 	struct file_system_type * next;
@@ -2251,6 +2275,8 @@ struct filename {
 extern long vfs_truncate(struct path *, loff_t);
 extern int do_truncate(struct dentry *, loff_t start, unsigned int time_attrs,
 		       struct file *filp);
+extern int do_truncate2(struct vfsmount *, struct dentry *, loff_t start,
+			unsigned int time_attrs, struct file *filp);
 extern int vfs_fallocate(struct file *file, int mode, loff_t offset,
 			loff_t len);
 extern long do_sys_open(int dfd, const char __user *filename, int flags,
@@ -2475,8 +2501,11 @@ extern void emergency_remount(void);
 extern sector_t bmap(struct inode *, sector_t);
 #endif
 extern int notify_change(struct dentry *, struct iattr *, struct inode **);
+extern int notify_change2(struct vfsmount *, struct dentry *, struct iattr *, struct inode **);
 extern int inode_permission(struct inode *, int);
+extern int inode_permission2(struct vfsmount *, struct inode *, int);
 extern int __inode_permission(struct inode *, int);
+extern int __inode_permission2(struct vfsmount *, struct inode *, int);
 extern int generic_permission(struct inode *, int);
 extern int __check_sticky(struct inode *dir, struct inode *inode);
 
diff --git a/include/linux/fscrypt_common.h b/include/linux/fscrypt_common.h
new file mode 100644
index 000000000000..4022c61f7e9b
--- /dev/null
+++ b/include/linux/fscrypt_common.h
@@ -0,0 +1,138 @@
+/*
+ * fscrypt_common.h: common declarations for per-file encryption
+ *
+ * Copyright (C) 2015, Google, Inc.
+ *
+ * Written by Michael Halcrow, 2015.
+ * Modified by Jaegeuk Kim, 2015.
+ */
+
+#ifndef _LINUX_FSCRYPT_COMMON_H
+#define _LINUX_FSCRYPT_COMMON_H
+
+#include <linux/key.h>
+#include <linux/fs.h>
+#include <linux/mm.h>
+#include <linux/bio.h>
+#include <linux/dcache.h>
+#include <crypto/skcipher.h>
+#include <uapi/linux/fs.h>
+
+#define FS_CRYPTO_BLOCK_SIZE		16
+
+struct fscrypt_info;
+
+struct fscrypt_ctx {
+	union {
+		struct {
+			struct page *bounce_page;	/* Ciphertext page */
+			struct page *control_page;	/* Original page  */
+		} w;
+		struct {
+			struct bio *bio;
+			struct work_struct work;
+		} r;
+		struct list_head free_list;	/* Free list */
+	};
+	u8 flags;				/* Flags */
+};
+
+/**
+ * For encrypted symlinks, the ciphertext length is stored at the beginning
+ * of the string in little-endian format.
+ */
+struct fscrypt_symlink_data {
+	__le16 len;
+	char encrypted_path[1];
+} __packed;
+
+struct fscrypt_str {
+	unsigned char *name;
+	u32 len;
+};
+
+struct fscrypt_name {
+	const struct qstr *usr_fname;
+	struct fscrypt_str disk_name;
+	u32 hash;
+	u32 minor_hash;
+	struct fscrypt_str crypto_buf;
+};
+
+#define FSTR_INIT(n, l)		{ .name = n, .len = l }
+#define FSTR_TO_QSTR(f)		QSTR_INIT((f)->name, (f)->len)
+#define fname_name(p)		((p)->disk_name.name)
+#define fname_len(p)		((p)->disk_name.len)
+
+/*
+ * fscrypt superblock flags
+ */
+#define FS_CFLG_OWN_PAGES (1U << 1)
+
+/*
+ * crypto opertions for filesystems
+ */
+struct fscrypt_operations {
+	unsigned int flags;
+	const char *key_prefix;
+	int (*get_context)(struct inode *, void *, size_t);
+	int (*set_context)(struct inode *, const void *, size_t, void *);
+	int (*dummy_context)(struct inode *);
+	bool (*is_encrypted)(struct inode *);
+	bool (*empty_dir)(struct inode *);
+	unsigned (*max_namelen)(struct inode *);
+};
+
+static inline bool fscrypt_dummy_context_enabled(struct inode *inode)
+{
+	if (inode->i_sb->s_cop->dummy_context &&
+				inode->i_sb->s_cop->dummy_context(inode))
+		return true;
+	return false;
+}
+
+static inline bool fscrypt_valid_enc_modes(u32 contents_mode,
+					u32 filenames_mode)
+{
+	if (contents_mode == FS_ENCRYPTION_MODE_AES_128_CBC &&
+	    filenames_mode == FS_ENCRYPTION_MODE_AES_128_CTS)
+		return true;
+
+	if (contents_mode == FS_ENCRYPTION_MODE_AES_256_XTS &&
+	    filenames_mode == FS_ENCRYPTION_MODE_AES_256_CTS)
+		return true;
+
+	return false;
+}
+
+static inline bool fscrypt_is_dot_dotdot(const struct qstr *str)
+{
+	if (str->len == 1 && str->name[0] == '.')
+		return true;
+
+	if (str->len == 2 && str->name[0] == '.' && str->name[1] == '.')
+		return true;
+
+	return false;
+}
+
+static inline struct page *fscrypt_control_page(struct page *page)
+{
+#if IS_ENABLED(CONFIG_FS_ENCRYPTION)
+	return ((struct fscrypt_ctx *)page_private(page))->w.control_page;
+#else
+	WARN_ON_ONCE(1);
+	return ERR_PTR(-EINVAL);
+#endif
+}
+
+static inline int fscrypt_has_encryption_key(const struct inode *inode)
+{
+#if IS_ENABLED(CONFIG_FS_ENCRYPTION)
+	return (inode->i_crypt_info != NULL);
+#else
+	return 0;
+#endif
+}
+
+#endif	/* _LINUX_FSCRYPT_COMMON_H */
diff --git a/include/linux/fscrypt_notsupp.h b/include/linux/fscrypt_notsupp.h
new file mode 100644
index 000000000000..ec406aed2f2f
--- /dev/null
+++ b/include/linux/fscrypt_notsupp.h
@@ -0,0 +1,177 @@
+/*
+ * fscrypt_notsupp.h
+ *
+ * This stubs out the fscrypt functions for filesystems configured without
+ * encryption support.
+ */
+
+#ifndef _LINUX_FSCRYPT_NOTSUPP_H
+#define _LINUX_FSCRYPT_NOTSUPP_H
+
+#include <linux/fscrypt_common.h>
+
+/* crypto.c */
+static inline struct fscrypt_ctx *fscrypt_get_ctx(const struct inode *inode,
+						  gfp_t gfp_flags)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
+static inline void fscrypt_release_ctx(struct fscrypt_ctx *ctx)
+{
+	return;
+}
+
+static inline struct page *fscrypt_encrypt_page(const struct inode *inode,
+						struct page *page,
+						unsigned int len,
+						unsigned int offs,
+						u64 lblk_num, gfp_t gfp_flags)
+{
+	return ERR_PTR(-EOPNOTSUPP);
+}
+
+static inline int fscrypt_decrypt_page(const struct inode *inode,
+				       struct page *page,
+				       unsigned int len, unsigned int offs,
+				       u64 lblk_num)
+{
+	return -EOPNOTSUPP;
+}
+
+
+static inline void fscrypt_restore_control_page(struct page *page)
+{
+	return;
+}
+
+static inline void fscrypt_set_d_op(struct dentry *dentry)
+{
+	return;
+}
+
+static inline void fscrypt_set_encrypted_dentry(struct dentry *dentry)
+{
+	return;
+}
+
+/* policy.c */
+static inline int fscrypt_ioctl_set_policy(struct file *filp,
+					   const void __user *arg)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int fscrypt_ioctl_get_policy(struct file *filp, void __user *arg)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int fscrypt_has_permitted_context(struct inode *parent,
+						struct inode *child)
+{
+	return 0;
+}
+
+static inline int fscrypt_inherit_context(struct inode *parent,
+					  struct inode *child,
+					  void *fs_data, bool preload)
+{
+	return -EOPNOTSUPP;
+}
+
+/* keyinfo.c */
+static inline int fscrypt_get_encryption_info(struct inode *inode)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void fscrypt_put_encryption_info(struct inode *inode,
+					       struct fscrypt_info *ci)
+{
+	return;
+}
+
+ /* fname.c */
+static inline int fscrypt_setup_filename(struct inode *dir,
+					 const struct qstr *iname,
+					 int lookup, struct fscrypt_name *fname)
+{
+	if (dir->i_sb->s_cop->is_encrypted(dir))
+		return -EOPNOTSUPP;
+
+	memset(fname, 0, sizeof(struct fscrypt_name));
+	fname->usr_fname = iname;
+	fname->disk_name.name = (unsigned char *)iname->name;
+	fname->disk_name.len = iname->len;
+	return 0;
+}
+
+static inline void fscrypt_free_filename(struct fscrypt_name *fname)
+{
+	return;
+}
+
+static inline u32 fscrypt_fname_encrypted_size(const struct inode *inode,
+					       u32 ilen)
+{
+	/* never happens */
+	WARN_ON(1);
+	return 0;
+}
+
+static inline int fscrypt_fname_alloc_buffer(const struct inode *inode,
+					     u32 ilen,
+					     struct fscrypt_str *crypto_str)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline void fscrypt_fname_free_buffer(struct fscrypt_str *crypto_str)
+{
+	return;
+}
+
+static inline int fscrypt_fname_disk_to_usr(struct inode *inode,
+					    u32 hash, u32 minor_hash,
+					    const struct fscrypt_str *iname,
+					    struct fscrypt_str *oname)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline int fscrypt_fname_usr_to_disk(struct inode *inode,
+					    const struct qstr *iname,
+					    struct fscrypt_str *oname)
+{
+	return -EOPNOTSUPP;
+}
+
+static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
+				      const u8 *de_name, u32 de_name_len)
+{
+	/* Encryption support disabled; use standard comparison */
+	if (de_name_len != fname->disk_name.len)
+		return false;
+	return !memcmp(de_name, fname->disk_name.name, fname->disk_name.len);
+}
+
+/* bio.c */
+static inline void fscrypt_decrypt_bio_pages(struct fscrypt_ctx *ctx,
+					     struct bio *bio)
+{
+	return;
+}
+
+static inline void fscrypt_pullback_bio_page(struct page **page, bool restore)
+{
+	return;
+}
+
+static inline int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk,
+					sector_t pblk, unsigned int len)
+{
+	return -EOPNOTSUPP;
+}
+
+#endif	/* _LINUX_FSCRYPT_NOTSUPP_H */
diff --git a/include/linux/fscrypt_supp.h b/include/linux/fscrypt_supp.h
new file mode 100644
index 000000000000..32e2fcf13b01
--- /dev/null
+++ b/include/linux/fscrypt_supp.h
@@ -0,0 +1,145 @@
+/*
+ * fscrypt_supp.h
+ *
+ * This is included by filesystems configured with encryption support.
+ */
+
+#ifndef _LINUX_FSCRYPT_SUPP_H
+#define _LINUX_FSCRYPT_SUPP_H
+
+#include <linux/fscrypt_common.h>
+
+/* crypto.c */
+extern struct kmem_cache *fscrypt_info_cachep;
+extern struct fscrypt_ctx *fscrypt_get_ctx(const struct inode *, gfp_t);
+extern void fscrypt_release_ctx(struct fscrypt_ctx *);
+extern struct page *fscrypt_encrypt_page(const struct inode *, struct page *,
+						unsigned int, unsigned int,
+						u64, gfp_t);
+extern int fscrypt_decrypt_page(const struct inode *, struct page *, unsigned int,
+				unsigned int, u64);
+extern void fscrypt_restore_control_page(struct page *);
+
+extern const struct dentry_operations fscrypt_d_ops;
+
+static inline void fscrypt_set_d_op(struct dentry *dentry)
+{
+	d_set_d_op(dentry, &fscrypt_d_ops);
+}
+
+static inline void fscrypt_set_encrypted_dentry(struct dentry *dentry)
+{
+	spin_lock(&dentry->d_lock);
+	dentry->d_flags |= DCACHE_ENCRYPTED_WITH_KEY;
+	spin_unlock(&dentry->d_lock);
+}
+
+/* policy.c */
+extern int fscrypt_ioctl_set_policy(struct file *, const void __user *);
+extern int fscrypt_ioctl_get_policy(struct file *, void __user *);
+extern int fscrypt_has_permitted_context(struct inode *, struct inode *);
+extern int fscrypt_inherit_context(struct inode *, struct inode *,
+					void *, bool);
+/* keyinfo.c */
+extern int fscrypt_get_encryption_info(struct inode *);
+extern void fscrypt_put_encryption_info(struct inode *, struct fscrypt_info *);
+
+/* fname.c */
+extern int fscrypt_setup_filename(struct inode *, const struct qstr *,
+				int lookup, struct fscrypt_name *);
+
+static inline void fscrypt_free_filename(struct fscrypt_name *fname)
+{
+	kfree(fname->crypto_buf.name);
+}
+
+extern u32 fscrypt_fname_encrypted_size(const struct inode *, u32);
+extern int fscrypt_fname_alloc_buffer(const struct inode *, u32,
+				struct fscrypt_str *);
+extern void fscrypt_fname_free_buffer(struct fscrypt_str *);
+extern int fscrypt_fname_disk_to_usr(struct inode *, u32, u32,
+			const struct fscrypt_str *, struct fscrypt_str *);
+extern int fscrypt_fname_usr_to_disk(struct inode *, const struct qstr *,
+			struct fscrypt_str *);
+
+#define FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE	32
+
+/* Extracts the second-to-last ciphertext block; see explanation below */
+#define FSCRYPT_FNAME_DIGEST(name, len)	\
+	((name) + round_down((len) - FS_CRYPTO_BLOCK_SIZE - 1, \
+			     FS_CRYPTO_BLOCK_SIZE))
+
+#define FSCRYPT_FNAME_DIGEST_SIZE	FS_CRYPTO_BLOCK_SIZE
+
+/**
+ * fscrypt_digested_name - alternate identifier for an on-disk filename
+ *
+ * When userspace lists an encrypted directory without access to the key,
+ * filenames whose ciphertext is longer than FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE
+ * bytes are shown in this abbreviated form (base64-encoded) rather than as the
+ * full ciphertext (base64-encoded).  This is necessary to allow supporting
+ * filenames up to NAME_MAX bytes, since base64 encoding expands the length.
+ *
+ * To make it possible for filesystems to still find the correct directory entry
+ * despite not knowing the full on-disk name, we encode any filesystem-specific
+ * 'hash' and/or 'minor_hash' which the filesystem may need for its lookups,
+ * followed by the second-to-last ciphertext block of the filename.  Due to the
+ * use of the CBC-CTS encryption mode, the second-to-last ciphertext block
+ * depends on the full plaintext.  (Note that ciphertext stealing causes the
+ * last two blocks to appear "flipped".)  This makes accidental collisions very
+ * unlikely: just a 1 in 2^128 chance for two filenames to collide even if they
+ * share the same filesystem-specific hashes.
+ *
+ * However, this scheme isn't immune to intentional collisions, which can be
+ * created by anyone able to create arbitrary plaintext filenames and view them
+ * without the key.  Making the "digest" be a real cryptographic hash like
+ * SHA-256 over the full ciphertext would prevent this, although it would be
+ * less efficient and harder to implement, especially since the filesystem would
+ * need to calculate it for each directory entry examined during a search.
+ */
+struct fscrypt_digested_name {
+	u32 hash;
+	u32 minor_hash;
+	u8 digest[FSCRYPT_FNAME_DIGEST_SIZE];
+};
+
+/**
+ * fscrypt_match_name() - test whether the given name matches a directory entry
+ * @fname: the name being searched for
+ * @de_name: the name from the directory entry
+ * @de_name_len: the length of @de_name in bytes
+ *
+ * Normally @fname->disk_name will be set, and in that case we simply compare
+ * that to the name stored in the directory entry.  The only exception is that
+ * if we don't have the key for an encrypted directory and a filename in it is
+ * very long, then we won't have the full disk_name and we'll instead need to
+ * match against the fscrypt_digested_name.
+ *
+ * Return: %true if the name matches, otherwise %false.
+ */
+static inline bool fscrypt_match_name(const struct fscrypt_name *fname,
+				      const u8 *de_name, u32 de_name_len)
+{
+	if (unlikely(!fname->disk_name.name)) {
+		const struct fscrypt_digested_name *n =
+			(const void *)fname->crypto_buf.name;
+		if (WARN_ON_ONCE(fname->usr_fname->name[0] != '_'))
+			return false;
+		if (de_name_len <= FSCRYPT_FNAME_MAX_UNDIGESTED_SIZE)
+			return false;
+		return !memcmp(FSCRYPT_FNAME_DIGEST(de_name, de_name_len),
+			       n->digest, FSCRYPT_FNAME_DIGEST_SIZE);
+	}
+
+	if (de_name_len != fname->disk_name.len)
+		return false;
+	return !memcmp(de_name, fname->disk_name.name, fname->disk_name.len);
+}
+
+/* bio.c */
+extern void fscrypt_decrypt_bio_pages(struct fscrypt_ctx *, struct bio *);
+extern void fscrypt_pullback_bio_page(struct page **, bool);
+extern int fscrypt_zeroout_range(const struct inode *, pgoff_t, sector_t,
+				 unsigned int);
+
+#endif	/* _LINUX_FSCRYPT_SUPP_H */
diff --git a/include/linux/ftrace.h b/include/linux/ftrace.h
index 60048c50404e..ed94cea9eaff 100644
--- a/include/linux/ftrace.h
+++ b/include/linux/ftrace.h
@@ -702,7 +702,8 @@ static inline void __ftrace_enabled_restore(int enabled)
   static inline void time_hardirqs_off(unsigned long a0, unsigned long a1) { }
 #endif
 
-#ifdef CONFIG_PREEMPT_TRACER
+#if defined(CONFIG_PREEMPT_TRACER) || \
+	(defined(CONFIG_DEBUG_PREEMPT) && defined(CONFIG_PREEMPTIRQ_EVENTS))
   extern void trace_preempt_on(unsigned long a0, unsigned long a1);
   extern void trace_preempt_off(unsigned long a0, unsigned long a1);
 #else
diff --git a/include/linux/gpio_event.h b/include/linux/gpio_event.h
new file mode 100644
index 000000000000..2613fc5e4a93
--- /dev/null
+++ b/include/linux/gpio_event.h
@@ -0,0 +1,170 @@
+/* include/linux/gpio_event.h
+ *
+ * Copyright (C) 2007 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _LINUX_GPIO_EVENT_H
+#define _LINUX_GPIO_EVENT_H
+
+#include <linux/input.h>
+
+struct gpio_event_input_devs {
+	int count;
+	struct input_dev *dev[];
+};
+enum {
+	GPIO_EVENT_FUNC_UNINIT  = 0x0,
+	GPIO_EVENT_FUNC_INIT    = 0x1,
+	GPIO_EVENT_FUNC_SUSPEND = 0x2,
+	GPIO_EVENT_FUNC_RESUME  = 0x3,
+};
+struct gpio_event_info {
+	int (*func)(struct gpio_event_input_devs *input_devs,
+		    struct gpio_event_info *info,
+		    void **data, int func);
+	int (*event)(struct gpio_event_input_devs *input_devs,
+		     struct gpio_event_info *info,
+		     void **data, unsigned int dev, unsigned int type,
+		     unsigned int code, int value); /* out events */
+	bool no_suspend;
+};
+
+struct gpio_event_platform_data {
+	const char *name;
+	struct gpio_event_info **info;
+	size_t info_count;
+	int (*power)(const struct gpio_event_platform_data *pdata, bool on);
+	const char *names[]; /* If name is NULL, names contain a NULL */
+			     /* terminated list of input devices to create */
+};
+
+#define GPIO_EVENT_DEV_NAME "gpio-event"
+
+/* Key matrix */
+
+enum gpio_event_matrix_flags {
+	/* unset: drive active output low, set: drive active output high */
+	GPIOKPF_ACTIVE_HIGH              = 1U << 0,
+	GPIOKPF_DEBOUNCE                 = 1U << 1,
+	GPIOKPF_REMOVE_SOME_PHANTOM_KEYS = 1U << 2,
+	GPIOKPF_REMOVE_PHANTOM_KEYS      = GPIOKPF_REMOVE_SOME_PHANTOM_KEYS |
+					   GPIOKPF_DEBOUNCE,
+	GPIOKPF_DRIVE_INACTIVE           = 1U << 3,
+	GPIOKPF_LEVEL_TRIGGERED_IRQ      = 1U << 4,
+	GPIOKPF_PRINT_UNMAPPED_KEYS      = 1U << 16,
+	GPIOKPF_PRINT_MAPPED_KEYS        = 1U << 17,
+	GPIOKPF_PRINT_PHANTOM_KEYS       = 1U << 18,
+};
+
+#define MATRIX_CODE_BITS (10)
+#define MATRIX_KEY_MASK ((1U << MATRIX_CODE_BITS) - 1)
+#define MATRIX_KEY(dev, code) \
+	(((dev) << MATRIX_CODE_BITS) | (code & MATRIX_KEY_MASK))
+
+extern int gpio_event_matrix_func(struct gpio_event_input_devs *input_devs,
+			struct gpio_event_info *info, void **data, int func);
+struct gpio_event_matrix_info {
+	/* initialize to gpio_event_matrix_func */
+	struct gpio_event_info info;
+	/* size must be ninputs * noutputs */
+	const unsigned short *keymap;
+	unsigned int *input_gpios;
+	unsigned int *output_gpios;
+	unsigned int ninputs;
+	unsigned int noutputs;
+	/* time to wait before reading inputs after driving each output */
+	ktime_t settle_time;
+	/* time to wait before scanning the keypad a second time */
+	ktime_t debounce_delay;
+	ktime_t poll_time;
+	unsigned flags;
+};
+
+/* Directly connected inputs and outputs */
+
+enum gpio_event_direct_flags {
+	GPIOEDF_ACTIVE_HIGH         = 1U << 0,
+/*	GPIOEDF_USE_DOWN_IRQ        = 1U << 1, */
+/*	GPIOEDF_USE_IRQ             = (1U << 2) | GPIOIDF_USE_DOWN_IRQ, */
+	GPIOEDF_PRINT_KEYS          = 1U << 8,
+	GPIOEDF_PRINT_KEY_DEBOUNCE  = 1U << 9,
+	GPIOEDF_PRINT_KEY_UNSTABLE  = 1U << 10,
+};
+
+struct gpio_event_direct_entry {
+	uint32_t gpio:16;
+	uint32_t code:10;
+	uint32_t dev:6;
+};
+
+/* inputs */
+extern int gpio_event_input_func(struct gpio_event_input_devs *input_devs,
+			struct gpio_event_info *info, void **data, int func);
+struct gpio_event_input_info {
+	/* initialize to gpio_event_input_func */
+	struct gpio_event_info info;
+	ktime_t debounce_time;
+	ktime_t poll_time;
+	uint16_t flags;
+	uint16_t type;
+	const struct gpio_event_direct_entry *keymap;
+	size_t keymap_size;
+};
+
+/* outputs */
+extern int gpio_event_output_func(struct gpio_event_input_devs *input_devs,
+			struct gpio_event_info *info, void **data, int func);
+extern int gpio_event_output_event(struct gpio_event_input_devs *input_devs,
+			struct gpio_event_info *info, void **data,
+			unsigned int dev, unsigned int type,
+			unsigned int code, int value);
+struct gpio_event_output_info {
+	/* initialize to gpio_event_output_func and gpio_event_output_event */
+	struct gpio_event_info info;
+	uint16_t flags;
+	uint16_t type;
+	const struct gpio_event_direct_entry *keymap;
+	size_t keymap_size;
+};
+
+
+/* axes */
+
+enum gpio_event_axis_flags {
+	GPIOEAF_PRINT_UNKNOWN_DIRECTION  = 1U << 16,
+	GPIOEAF_PRINT_RAW                = 1U << 17,
+	GPIOEAF_PRINT_EVENT              = 1U << 18,
+};
+
+extern int gpio_event_axis_func(struct gpio_event_input_devs *input_devs,
+			struct gpio_event_info *info, void **data, int func);
+struct gpio_event_axis_info {
+	/* initialize to gpio_event_axis_func */
+	struct gpio_event_info info;
+	uint8_t  count; /* number of gpios for this axis */
+	uint8_t  dev; /* device index when using multiple input devices */
+	uint8_t  type; /* EV_REL or EV_ABS */
+	uint16_t code;
+	uint16_t decoded_size;
+	uint16_t (*map)(struct gpio_event_axis_info *info, uint16_t in);
+	uint32_t *gpio;
+	uint32_t flags;
+};
+#define gpio_axis_2bit_gray_map gpio_axis_4bit_gray_map
+#define gpio_axis_3bit_gray_map gpio_axis_4bit_gray_map
+uint16_t gpio_axis_4bit_gray_map(
+			struct gpio_event_axis_info *info, uint16_t in);
+uint16_t gpio_axis_5bit_singletrack_map(
+			struct gpio_event_axis_info *info, uint16_t in);
+
+#endif
diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
index 2ead22dd74a0..c98c6539e2c2 100644
--- a/include/linux/hrtimer.h
+++ b/include/linux/hrtimer.h
@@ -220,7 +220,7 @@ static inline void hrtimer_set_expires_range(struct hrtimer *timer, ktime_t time
 	timer->node.expires = ktime_add_safe(time, delta);
 }
 
-static inline void hrtimer_set_expires_range_ns(struct hrtimer *timer, ktime_t time, unsigned long delta)
+static inline void hrtimer_set_expires_range_ns(struct hrtimer *timer, ktime_t time, u64 delta)
 {
 	timer->_softexpires = time;
 	timer->node.expires = ktime_add_safe(time, ns_to_ktime(delta));
@@ -378,7 +378,7 @@ static inline void destroy_hrtimer_on_stack(struct hrtimer *timer) { }
 
 /* Basic timer operations: */
 extern void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
-			unsigned long range_ns, const enum hrtimer_mode mode);
+				   u64 range_ns, const enum hrtimer_mode mode);
 
 /**
  * hrtimer_start - (re)start an hrtimer on the current CPU
@@ -399,7 +399,7 @@ extern int hrtimer_try_to_cancel(struct hrtimer *timer);
 static inline void hrtimer_start_expires(struct hrtimer *timer,
 					 enum hrtimer_mode mode)
 {
-	unsigned long delta;
+	u64 delta;
 	ktime_t soft, hard;
 	soft = hrtimer_get_softexpires(timer);
 	hard = hrtimer_get_expires(timer);
@@ -477,10 +477,12 @@ extern long hrtimer_nanosleep_restart(struct restart_block *restart_block);
 extern void hrtimer_init_sleeper(struct hrtimer_sleeper *sl,
 				 struct task_struct *tsk);
 
-extern int schedule_hrtimeout_range(ktime_t *expires, unsigned long delta,
+extern int schedule_hrtimeout_range(ktime_t *expires, u64 delta,
 						const enum hrtimer_mode mode);
 extern int schedule_hrtimeout_range_clock(ktime_t *expires,
-		unsigned long delta, const enum hrtimer_mode mode, int clock);
+					  u64 delta,
+					  const enum hrtimer_mode mode,
+					  int clock);
 extern int schedule_hrtimeout(ktime_t *expires, const enum hrtimer_mode mode);
 
 /* Soft interrupt function to run the hrtimer queues: */
diff --git a/include/linux/if_pppolac.h b/include/linux/if_pppolac.h
new file mode 100644
index 000000000000..e40aa1075a30
--- /dev/null
+++ b/include/linux/if_pppolac.h
@@ -0,0 +1,23 @@
+/* include/linux/if_pppolac.h
+ *
+ * Header for PPP on L2TP Access Concentrator / PPPoLAC Socket (RFC 2661)
+ *
+ * Copyright (C) 2009 Google, Inc.
+ * Author: Chia-chi Yeh <chiachi@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __LINUX_IF_PPPOLAC_H
+#define __LINUX_IF_PPPOLAC_H
+
+#include <uapi/linux/if_pppolac.h>
+
+#endif /* __LINUX_IF_PPPOLAC_H */
diff --git a/include/linux/if_pppopns.h b/include/linux/if_pppopns.h
new file mode 100644
index 000000000000..4ac621a9ce7c
--- /dev/null
+++ b/include/linux/if_pppopns.h
@@ -0,0 +1,23 @@
+/* include/linux/if_pppopns.h
+ *
+ * Header for PPP on PPTP Network Server / PPPoPNS Socket (RFC 2637)
+ *
+ * Copyright (C) 2009 Google, Inc.
+ * Author: Chia-chi Yeh <chiachi@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __LINUX_IF_PPPOPNS_H
+#define __LINUX_IF_PPPOPNS_H
+
+#include <uapi/linux/if_pppopns.h>
+
+#endif /* __LINUX_IF_PPPOPNS_H */
diff --git a/include/linux/if_pppox.h b/include/linux/if_pppox.h
index b49cf923becc..63828a5870f1 100644
--- a/include/linux/if_pppox.h
+++ b/include/linux/if_pppox.h
@@ -43,6 +43,25 @@ struct pptp_opt {
 	u32 seq_sent, seq_recv;
 	int ppp_flags;
 };
+
+struct pppolac_opt {
+	__u32		local;
+	__u32		remote;
+	__u32		recv_sequence;
+	__u32		xmit_sequence;
+	atomic_t	sequencing;
+	int		(*backlog_rcv)(struct sock *sk_udp, struct sk_buff *skb);
+};
+
+struct pppopns_opt {
+	__u16		local;
+	__u16		remote;
+	__u32		recv_sequence;
+	__u32		xmit_sequence;
+	void		(*data_ready)(struct sock *sk_raw);
+	int		(*backlog_rcv)(struct sock *sk_raw, struct sk_buff *skb);
+};
+
 #include <net/sock.h>
 
 struct pppox_sock {
@@ -53,6 +72,8 @@ struct pppox_sock {
 	union {
 		struct pppoe_opt pppoe;
 		struct pptp_opt  pptp;
+		struct pppolac_opt lac;
+		struct pppopns_opt pns;
 	} proto;
 	__be16			num;
 };
diff --git a/include/linux/inet_diag.h b/include/linux/inet_diag.h
index 0e707f0c1a3e..795852dc3434 100644
--- a/include/linux/inet_diag.h
+++ b/include/linux/inet_diag.h
@@ -3,6 +3,7 @@
 
 #include <uapi/linux/inet_diag.h>
 
+struct net;
 struct sock;
 struct inet_hashinfo;
 struct nlattr;
@@ -23,6 +24,10 @@ struct inet_diag_handler {
 	void		(*idiag_get_info)(struct sock *sk,
 					  struct inet_diag_msg *r,
 					  void *info);
+
+	int		(*destroy)(struct sk_buff *in_skb,
+				   const struct inet_diag_req_v2 *req);
+
 	__u16		idiag_type;
 	__u16		idiag_info_size;
 };
@@ -32,7 +37,7 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 		      struct sk_buff *skb, const struct inet_diag_req_v2 *req,
 		      struct user_namespace *user_ns,
 		      u32 pid, u32 seq, u16 nlmsg_flags,
-		      const struct nlmsghdr *unlh);
+		      const struct nlmsghdr *unlh, bool net_admin);
 void inet_diag_dump_icsk(struct inet_hashinfo *h, struct sk_buff *skb,
 			 struct netlink_callback *cb,
 			 const struct inet_diag_req_v2 *r,
@@ -41,6 +46,10 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo,
 			    struct sk_buff *in_skb, const struct nlmsghdr *nlh,
 			    const struct inet_diag_req_v2 *req);
 
+struct sock *inet_diag_find_one_icsk(struct net *net,
+				     struct inet_hashinfo *hashinfo,
+				     const struct inet_diag_req_v2 *req);
+
 int inet_diag_bc_sk(const struct nlattr *_bc, struct sock *sk);
 
 extern int  inet_diag_register(const struct inet_diag_handler *handler);
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index 1c1ff7e4faa4..021b1e9ff6cd 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -15,6 +15,8 @@
 #include <net/net_namespace.h>
 #include <linux/sched/rt.h>
 
+#include <asm/thread_info.h>
+
 #ifdef CONFIG_SMP
 # define INIT_PUSHABLE_TASKS(tsk)					\
 	.pushable_tasks = PLIST_NODE_INIT(tsk.pushable_tasks, MAX_PRIO),
@@ -183,14 +185,21 @@ extern struct task_group root_task_group;
 # define INIT_KASAN(tsk)
 #endif
 
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+# define INIT_TASK_TI(tsk) .thread_info = INIT_THREAD_INFO(tsk),
+#else
+# define INIT_TASK_TI(tsk)
+#endif
+
 /*
  *  INIT_TASK is used to set up the first task table, touch at
  * your own risk!. Base=0, limit=0x1fffff (=2MB)
  */
 #define INIT_TASK(tsk)	\
 {									\
+	INIT_TASK_TI(tsk)						\
 	.state		= 0,						\
-	.stack		= &init_thread_info,				\
+	.stack		= init_stack,					\
 	.usage		= ATOMIC_INIT(2),				\
 	.flags		= PF_KTHREAD,					\
 	.prio		= MAX_PRIO-20,					\
diff --git a/include/linux/initramfs.h b/include/linux/initramfs.h
new file mode 100644
index 000000000000..fc7da63b125b
--- /dev/null
+++ b/include/linux/initramfs.h
@@ -0,0 +1,32 @@
+/*
+ * include/linux/initramfs.h
+ *
+ * Copyright (C) 2015, Google
+ * Rom Lemarchand <romlem@android.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ */
+
+#ifndef _LINUX_INITRAMFS_H
+#define _LINUX_INITRAMFS_H
+
+#include <linux/kconfig.h>
+
+#if IS_BUILTIN(CONFIG_BLK_DEV_INITRD)
+
+int __init default_rootfs(void);
+
+#endif
+
+#endif /* _LINUX_INITRAMFS_H */
diff --git a/include/linux/ipv6.h b/include/linux/ipv6.h
index 402753bccafa..a0fc3cf932af 100644
--- a/include/linux/ipv6.h
+++ b/include/linux/ipv6.h
@@ -18,6 +18,7 @@ struct ipv6_devconf {
 	__s32		dad_transmits;
 	__s32		rtr_solicits;
 	__s32		rtr_solicit_interval;
+	__s32		rtr_solicit_max_interval;
 	__s32		rtr_solicit_delay;
 	__s32		force_mld_version;
 	__s32		mldv1_unsolicited_report_interval;
@@ -36,9 +37,11 @@ struct ipv6_devconf {
 	__s32		accept_ra_rtr_pref;
 	__s32		rtr_probe_interval;
 #ifdef CONFIG_IPV6_ROUTE_INFO
+	__s32		accept_ra_rt_info_min_plen;
 	__s32		accept_ra_rt_info_max_plen;
 #endif
 #endif
+	__s32		accept_ra_rt_table;
 	__s32		proxy_ndp;
 	__s32		accept_source_route;
 	__s32		accept_ra_from_local;
diff --git a/include/linux/kbuild.h b/include/linux/kbuild.h
index 22a72198c14b..4e80f3a9ad58 100644
--- a/include/linux/kbuild.h
+++ b/include/linux/kbuild.h
@@ -2,14 +2,14 @@
 #define __LINUX_KBUILD_H
 
 #define DEFINE(sym, val) \
-        asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+	asm volatile("\n.ascii \"->" #sym " %0 " #val "\"" : : "i" (val))
 
-#define BLANK() asm volatile("\n->" : : )
+#define BLANK() asm volatile("\n.ascii \"->\"" : : )
 
 #define OFFSET(sym, str, mem) \
 	DEFINE(sym, offsetof(struct str, mem))
 
 #define COMMENT(x) \
-	asm volatile("\n->#" x)
+	asm volatile("\n.ascii \"->#" x "\"")
 
 #endif
diff --git a/include/linux/kdb.h b/include/linux/kdb.h
index a19bcf9e762e..410decacff8f 100644
--- a/include/linux/kdb.h
+++ b/include/linux/kdb.h
@@ -177,7 +177,7 @@ extern int kdb_get_kbd_char(void);
 static inline
 int kdb_process_cpu(const struct task_struct *p)
 {
-	unsigned int cpu = task_thread_info(p)->cpu;
+	unsigned int cpu = task_cpu(p);
 	if (cpu > num_possible_cpus())
 		cpu = 0;
 	return cpu;
diff --git a/include/linux/keychord.h b/include/linux/keychord.h
new file mode 100644
index 000000000000..08cf5402102c
--- /dev/null
+++ b/include/linux/keychord.h
@@ -0,0 +1,23 @@
+/*
+ *  Key chord input driver
+ *
+ * Copyright (C) 2008 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+*/
+
+#ifndef __LINUX_KEYCHORD_H_
+#define __LINUX_KEYCHORD_H_
+
+#include <uapi/linux/keychord.h>
+
+#endif	/* __LINUX_KEYCHORD_H_ */
diff --git a/include/linux/keycombo.h b/include/linux/keycombo.h
new file mode 100644
index 000000000000..c6db2626b0d3
--- /dev/null
+++ b/include/linux/keycombo.h
@@ -0,0 +1,36 @@
+/*
+ * include/linux/keycombo.h - platform data structure for keycombo driver
+ *
+ * Copyright (C) 2014 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _LINUX_KEYCOMBO_H
+#define _LINUX_KEYCOMBO_H
+
+#define KEYCOMBO_NAME "keycombo"
+
+/*
+ * if key_down_fn and key_up_fn are both present, you are guaranteed that
+ * key_down_fn will return before key_up_fn is called, and that key_up_fn
+ * is called iff key_down_fn is called.
+ */
+struct keycombo_platform_data {
+	void (*key_down_fn)(void *);
+	void (*key_up_fn)(void *);
+	void *priv;
+	int key_down_delay; /* Time in ms */
+	int *keys_up;
+	int keys_down[]; /* 0 terminated */
+};
+
+#endif /* _LINUX_KEYCOMBO_H */
diff --git a/include/linux/keyreset.h b/include/linux/keyreset.h
new file mode 100644
index 000000000000..2e34afab65e4
--- /dev/null
+++ b/include/linux/keyreset.h
@@ -0,0 +1,29 @@
+/*
+ * include/linux/keyreset.h - platform data structure for resetkeys driver
+ *
+ * Copyright (C) 2014 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _LINUX_KEYRESET_H
+#define _LINUX_KEYRESET_H
+
+#define KEYRESET_NAME "keyreset"
+
+struct keyreset_platform_data {
+	int (*reset_fn)(void);
+	int key_down_delay;
+	int *keys_up;
+	int keys_down[]; /* 0 terminated */
+};
+
+#endif /* _LINUX_KEYRESET_H */
diff --git a/include/linux/kthread.h b/include/linux/kthread.h
index e691b6a23f72..4289343ba1f9 100644
--- a/include/linux/kthread.h
+++ b/include/linux/kthread.h
@@ -75,6 +75,8 @@ struct kthread_work {
 	struct list_head	node;
 	kthread_work_func_t	func;
 	struct kthread_worker	*worker;
+	/* Number of canceling calls that are running at the moment. */
+	int			canceling;
 };
 
 #define KTHREAD_WORKER_INIT(worker)	{				\
@@ -129,4 +131,6 @@ bool queue_kthread_work(struct kthread_worker *worker,
 void flush_kthread_work(struct kthread_work *work);
 void flush_kthread_worker(struct kthread_worker *worker);
 
+bool kthread_cancel_work_sync(struct kthread_work *work);
+
 #endif /* _LINUX_KTHREAD_H */
diff --git a/include/linux/mbcache2.h b/include/linux/mbcache2.h
new file mode 100644
index 000000000000..b6f160ff2533
--- /dev/null
+++ b/include/linux/mbcache2.h
@@ -0,0 +1,50 @@
+#ifndef _LINUX_MB2CACHE_H
+#define _LINUX_MB2CACHE_H
+
+#include <linux/hash.h>
+#include <linux/list_bl.h>
+#include <linux/list.h>
+#include <linux/atomic.h>
+#include <linux/fs.h>
+
+struct mb2_cache;
+
+struct mb2_cache_entry {
+	/* LRU list - protected by cache->c_lru_list_lock */
+	struct list_head	e_lru_list;
+	/* Hash table list - protected by bitlock in e_hash_list_head */
+	struct hlist_bl_node	e_hash_list;
+	atomic_t		e_refcnt;
+	/* Key in hash - stable during lifetime of the entry */
+	u32			e_key;
+	/* Block number of hashed block - stable during lifetime of the entry */
+	sector_t		e_block;
+	/* Head of hash list (for list bit lock) - stable */
+	struct hlist_bl_head	*e_hash_list_head;
+};
+
+struct mb2_cache *mb2_cache_create(int bucket_bits);
+void mb2_cache_destroy(struct mb2_cache *cache);
+
+int mb2_cache_entry_create(struct mb2_cache *cache, gfp_t mask, u32 key,
+			   sector_t block);
+void __mb2_cache_entry_free(struct mb2_cache_entry *entry);
+static inline int mb2_cache_entry_put(struct mb2_cache *cache,
+				      struct mb2_cache_entry *entry)
+{
+	if (!atomic_dec_and_test(&entry->e_refcnt))
+		return 0;
+	__mb2_cache_entry_free(entry);
+	return 1;
+}
+
+void mb2_cache_entry_delete_block(struct mb2_cache *cache, u32 key,
+				  sector_t block);
+struct mb2_cache_entry *mb2_cache_entry_find_first(struct mb2_cache *cache,
+						   u32 key);
+struct mb2_cache_entry *mb2_cache_entry_find_next(struct mb2_cache *cache,
+						  struct mb2_cache_entry *entry);
+void mb2_cache_entry_touch(struct mb2_cache *cache,
+			   struct mb2_cache_entry *entry);
+
+#endif	/* _LINUX_MB2CACHE_H */
diff --git a/include/linux/memory-state-time.h b/include/linux/memory-state-time.h
new file mode 100644
index 000000000000..d2212b027866
--- /dev/null
+++ b/include/linux/memory-state-time.h
@@ -0,0 +1,42 @@
+/* include/linux/memory-state-time.h
+ *
+ * Copyright (C) 2016 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/workqueue.h>
+
+#define UPDATE_MEMORY_STATE(BLOCK, VALUE) BLOCK->update_call(BLOCK, VALUE)
+
+struct memory_state_update_block;
+
+typedef void (*memory_state_update_fn_t)(struct memory_state_update_block *ub,
+		int value);
+
+/* This struct is populated when you pass it to a memory_state_register*
+ * function. The update_call function is used for an update and defined in the
+ * typedef memory_state_update_fn_t
+ */
+struct memory_state_update_block {
+	memory_state_update_fn_t update_call;
+	int id;
+};
+
+/* Register a frequency struct memory_state_update_block to provide updates to
+ * memory_state_time about frequency changes using its update_call function.
+ */
+struct memory_state_update_block *memory_state_register_frequency_source(void);
+
+/* Register a bandwidth struct memory_state_update_block to provide updates to
+ * memory_state_time about bandwidth changes using its update_call function.
+ */
+struct memory_state_update_block *memory_state_register_bandwidth_source(void);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 5c2bf9096b7a..f29e8aa76e39 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -51,6 +51,17 @@ extern int sysctl_legacy_va_layout;
 #define sysctl_legacy_va_layout 0
 #endif
 
+#ifdef CONFIG_HAVE_ARCH_MMAP_RND_BITS
+extern const int mmap_rnd_bits_min;
+extern const int mmap_rnd_bits_max;
+extern int mmap_rnd_bits __read_mostly;
+#endif
+#ifdef CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS
+extern const int mmap_rnd_compat_bits_min;
+extern const int mmap_rnd_compat_bits_max;
+extern int mmap_rnd_compat_bits __read_mostly;
+#endif
+
 #include <asm/page.h>
 #include <asm/pgtable.h>
 #include <asm/processor.h>
@@ -1059,6 +1070,7 @@ extern void pagefault_out_of_memory(void);
 extern void show_free_areas(unsigned int flags);
 extern bool skip_free_areas_node(unsigned int flags, int nid);
 
+void shmem_set_file(struct vm_area_struct *vma, struct file *file);
 int shmem_zero_setup(struct vm_area_struct *);
 #ifdef CONFIG_SHMEM
 bool shmem_mapping(struct address_space *mapping);
@@ -1838,7 +1850,7 @@ extern int vma_adjust(struct vm_area_struct *vma, unsigned long start,
 extern struct vm_area_struct *vma_merge(struct mm_struct *,
 	struct vm_area_struct *prev, unsigned long addr, unsigned long end,
 	unsigned long vm_flags, struct anon_vma *, struct file *, pgoff_t,
-	struct mempolicy *, struct vm_userfaultfd_ctx);
+	struct mempolicy *, struct vm_userfaultfd_ctx, const char __user *);
 extern struct anon_vma *find_mergeable_anon_vma(struct vm_area_struct *);
 extern int split_vma(struct mm_struct *,
 	struct vm_area_struct *, unsigned long addr, int new_below);
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
index 36f4695aa604..907f94428ccc 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
@@ -12,6 +12,7 @@
 #include <linux/cpumask.h>
 #include <linux/uprobes.h>
 #include <linux/page-flags-layout.h>
+#include <linux/workqueue.h>
 #include <asm/page.h>
 #include <asm/mmu.h>
 
@@ -323,11 +324,18 @@ struct vm_area_struct {
 	/*
 	 * For areas with an address space and backing store,
 	 * linkage into the address_space->i_mmap interval tree.
+	 *
+	 * For private anonymous mappings, a pointer to a null terminated string
+	 * in the user process containing the name given to the vma, or NULL
+	 * if unnamed.
 	 */
-	struct {
-		struct rb_node rb;
-		unsigned long rb_subtree_last;
-	} shared;
+	union {
+		struct {
+			struct rb_node rb;
+			unsigned long rb_subtree_last;
+		} shared;
+		const char __user *anon_name;
+	};
 
 	/*
 	 * A file's MAP_PRIVATE vma can be in both i_mmap tree and anon_vma
@@ -516,6 +524,7 @@ struct mm_struct {
 #ifdef CONFIG_HUGETLB_PAGE
 	atomic_long_t hugetlb_usage;
 #endif
+	struct work_struct async_put_work;
 };
 
 static inline void mm_init_cpumask(struct mm_struct *mm)
@@ -596,4 +605,13 @@ typedef struct {
 	unsigned long val;
 } swp_entry_t;
 
+/* Return the name for an anonymous mapping or NULL for a file-backed mapping */
+static inline const char __user *vma_get_anon_name(struct vm_area_struct *vma)
+{
+	if (vma->vm_file)
+		return NULL;
+
+	return vma->anon_name;
+}
+
 #endif /* _LINUX_MM_TYPES_H */
diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h
index eb0151bac50c..8f23fb2c5ed2 100644
--- a/include/linux/mmc/card.h
+++ b/include/linux/mmc/card.h
@@ -118,6 +118,9 @@ struct mmc_ext_csd {
 	u8			raw_pwr_cl_ddr_200_360;	/* 253 */
 	u8			raw_bkops_status;	/* 246 */
 	u8			raw_sectors[4];		/* 212 - 4 bytes */
+	u8			pre_eol_info;		/* 267 */
+	u8			device_life_time_est_typ_a;	/* 268 */
+	u8			device_life_time_est_typ_b;	/* 269 */
 
 	unsigned int            feature_support;
 #define MMC_DISCARD_FEATURE	BIT(0)                  /* CMD38 feature */
diff --git a/include/linux/mmc/core.h b/include/linux/mmc/core.h
index 37967b6da03c..0860efd6e1be 100644
--- a/include/linux/mmc/core.h
+++ b/include/linux/mmc/core.h
@@ -136,6 +136,10 @@ struct mmc_request {
 	struct completion	completion;
 	void			(*done)(struct mmc_request *);/* completion function */
 	struct mmc_host		*host;
+	ktime_t			io_start;
+#ifdef CONFIG_BLOCK
+	int			lat_hist_enabled;
+#endif
 };
 
 struct mmc_card;
diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h
index 8673ffe3d86e..97b2b0b1f99d 100644
--- a/include/linux/mmc/host.h
+++ b/include/linux/mmc/host.h
@@ -16,6 +16,7 @@
 #include <linux/sched.h>
 #include <linux/device.h>
 #include <linux/fault-inject.h>
+#include <linux/blkdev.h>
 
 #include <linux/mmc/core.h>
 #include <linux/mmc/card.h>
@@ -370,6 +371,20 @@ struct mmc_host {
 	int			dsr_req;	/* DSR value is valid */
 	u32			dsr;	/* optional driver stage (DSR) value */
 
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+	struct {
+		struct sdio_cis			*cis;
+		struct sdio_cccr		*cccr;
+		struct sdio_embedded_func	*funcs;
+		int				num_funcs;
+	} embedded_sdio_data;
+#endif
+
+#ifdef CONFIG_BLOCK
+	int			latency_hist_enabled;
+	struct io_latency_state io_lat_s;
+#endif
+
 	unsigned long		private[0] ____cacheline_aligned;
 };
 
@@ -379,6 +394,14 @@ void mmc_remove_host(struct mmc_host *);
 void mmc_free_host(struct mmc_host *);
 int mmc_of_parse(struct mmc_host *host);
 
+#ifdef CONFIG_MMC_EMBEDDED_SDIO
+extern void mmc_set_embedded_sdio_data(struct mmc_host *host,
+				       struct sdio_cis *cis,
+				       struct sdio_cccr *cccr,
+				       struct sdio_embedded_func *funcs,
+				       int num_funcs);
+#endif
+
 static inline void *mmc_priv(struct mmc_host *host)
 {
 	return (void *)host->private;
diff --git a/include/linux/mmc/mmc.h b/include/linux/mmc/mmc.h
index 15f2c4a0a62c..2c6b1d45626e 100644
--- a/include/linux/mmc/mmc.h
+++ b/include/linux/mmc/mmc.h
@@ -330,6 +330,9 @@ struct _mmc_csd {
 #define EXT_CSD_CACHE_SIZE		249	/* RO, 4 bytes */
 #define EXT_CSD_PWR_CL_DDR_200_360	253	/* RO */
 #define EXT_CSD_FIRMWARE_VERSION	254	/* RO, 8 bytes */
+#define EXT_CSD_PRE_EOL_INFO		267	/* RO */
+#define EXT_CSD_DEVICE_LIFE_TIME_EST_TYP_A	268	/* RO */
+#define EXT_CSD_DEVICE_LIFE_TIME_EST_TYP_B	269	/* RO */
 #define EXT_CSD_SUPPORTED_MODE		493	/* RO */
 #define EXT_CSD_TAG_UNIT_SIZE		498	/* RO */
 #define EXT_CSD_DATA_TAG_SUPPORT	499	/* RO */
diff --git a/include/linux/mmc/pm.h b/include/linux/mmc/pm.h
index 4a139204c20c..6e2d6a135c7e 100644
--- a/include/linux/mmc/pm.h
+++ b/include/linux/mmc/pm.h
@@ -26,5 +26,6 @@ typedef unsigned int mmc_pm_flag_t;
 
 #define MMC_PM_KEEP_POWER	(1 << 0)	/* preserve card power during suspend */
 #define MMC_PM_WAKE_SDIO_IRQ	(1 << 1)	/* wake up host system on SDIO IRQ assertion */
+#define MMC_PM_IGNORE_PM_NOTIFY	(1 << 2)	/* ignore mmc pm notify */
 
 #endif /* LINUX_MMC_PM_H */
diff --git a/include/linux/mmc/sdio_func.h b/include/linux/mmc/sdio_func.h
index 97ca105347a6..f563bcf55e7f 100644
--- a/include/linux/mmc/sdio_func.h
+++ b/include/linux/mmc/sdio_func.h
@@ -23,6 +23,14 @@ struct sdio_func;
 typedef void (sdio_irq_handler_t)(struct sdio_func *);
 
 /*
+ * Structure used to hold embedded SDIO device data from platform layer
+ */
+struct sdio_embedded_func {
+	uint8_t f_class;
+	uint32_t f_maxblksize;
+};
+
+/*
  * SDIO function CIS tuple (unknown to the core)
  */
 struct sdio_func_tuple {
@@ -128,6 +136,8 @@ extern int sdio_release_irq(struct sdio_func *func);
 extern unsigned int sdio_align_size(struct sdio_func *func, unsigned int sz);
 
 extern u8 sdio_readb(struct sdio_func *func, unsigned int addr, int *err_ret);
+extern u8 sdio_readb_ext(struct sdio_func *func, unsigned int addr, int *err_ret,
+	unsigned in);
 extern u16 sdio_readw(struct sdio_func *func, unsigned int addr, int *err_ret);
 extern u32 sdio_readl(struct sdio_func *func, unsigned int addr, int *err_ret);
 
diff --git a/include/linux/module.h b/include/linux/module.h
index b229a9961d02..7fb9485bf0c9 100644
--- a/include/linux/module.h
+++ b/include/linux/module.h
@@ -125,13 +125,13 @@ extern void cleanup_module(void);
 
 /* Each module must use one module_init(). */
 #define module_init(initfn)					\
-	static inline initcall_t __inittest(void)		\
+	static inline initcall_t __maybe_unused __inittest(void)		\
 	{ return initfn; }					\
 	int init_module(void) __attribute__((alias(#initfn)));
 
 /* This is only required if you want to be unloadable. */
 #define module_exit(exitfn)					\
-	static inline exitcall_t __exittest(void)		\
+	static inline exitcall_t __maybe_unused __exittest(void)		\
 	{ return exitfn; }					\
 	void cleanup_module(void) __attribute__((alias(#exitfn)));
 
@@ -302,6 +302,28 @@ struct mod_tree_node {
 	struct latch_tree_node node;
 };
 
+struct module_layout {
+	/* The actual code + data. */
+	void *base;
+	/* Total size. */
+	unsigned int size;
+	/* The size of the executable code.  */
+	unsigned int text_size;
+	/* Size of RO section of the module (text+rodata) */
+	unsigned int ro_size;
+
+#ifdef CONFIG_MODULES_TREE_LOOKUP
+	struct mod_tree_node mtn;
+#endif
+};
+
+#ifdef CONFIG_MODULES_TREE_LOOKUP
+/* Only touch one cacheline for common rbtree-for-core-layout case. */
+#define __module_layout_align ____cacheline_aligned
+#else
+#define __module_layout_align
+#endif
+
 struct mod_kallsyms {
 	Elf_Sym *symtab;
 	unsigned int num_symtab;
@@ -372,37 +394,9 @@ struct module {
 	/* Startup function. */
 	int (*init)(void);
 
-	/*
-	 * If this is non-NULL, vfree() after init() returns.
-	 *
-	 * Cacheline align here, such that:
-	 *   module_init, module_core, init_size, core_size,
-	 *   init_text_size, core_text_size and mtn_core::{mod,node[0]}
-	 * are on the same cacheline.
-	 */
-	void *module_init	____cacheline_aligned;
-
-	/* Here is the actual code + data, vfree'd on unload. */
-	void *module_core;
-
-	/* Here are the sizes of the init and core sections */
-	unsigned int init_size, core_size;
-
-	/* The size of the executable code in each section.  */
-	unsigned int init_text_size, core_text_size;
-
-#ifdef CONFIG_MODULES_TREE_LOOKUP
-	/*
-	 * We want mtn_core::{mod,node[0]} to be in the same cacheline as the
-	 * above entries such that a regular lookup will only touch one
-	 * cacheline.
-	 */
-	struct mod_tree_node	mtn_core;
-	struct mod_tree_node	mtn_init;
-#endif
-
-	/* Size of RO sections of the module (text+rodata) */
-	unsigned int init_ro_size, core_ro_size;
+	/* Core layout: rbtree is accessed frequently, so keep together. */
+	struct module_layout core_layout __module_layout_align;
+	struct module_layout init_layout;
 
 	/* Arch-specific module values */
 	struct mod_arch_specific arch;
@@ -420,7 +414,7 @@ struct module {
 	/* Protected by RCU and/or module_mutex: use rcu_dereference() */
 	struct mod_kallsyms *kallsyms;
 	struct mod_kallsyms core_kallsyms;
-
+	
 	/* Section attributes */
 	struct module_sect_attrs *sect_attrs;
 
@@ -506,15 +500,15 @@ bool is_module_text_address(unsigned long addr);
 static inline bool within_module_core(unsigned long addr,
 				      const struct module *mod)
 {
-	return (unsigned long)mod->module_core <= addr &&
-	       addr < (unsigned long)mod->module_core + mod->core_size;
+	return (unsigned long)mod->core_layout.base <= addr &&
+	       addr < (unsigned long)mod->core_layout.base + mod->core_layout.size;
 }
 
 static inline bool within_module_init(unsigned long addr,
 				      const struct module *mod)
 {
-	return (unsigned long)mod->module_init <= addr &&
-	       addr < (unsigned long)mod->module_init + mod->init_size;
+	return (unsigned long)mod->init_layout.base <= addr &&
+	       addr < (unsigned long)mod->init_layout.base + mod->init_layout.size;
 }
 
 static inline bool within_module(unsigned long addr, const struct module *mod)
@@ -769,9 +763,13 @@ extern int module_sysfs_initialized;
 #ifdef CONFIG_DEBUG_SET_MODULE_RONX
 extern void set_all_modules_text_rw(void);
 extern void set_all_modules_text_ro(void);
+extern void module_enable_ro(const struct module *mod);
+extern void module_disable_ro(const struct module *mod);
 #else
 static inline void set_all_modules_text_rw(void) { }
 static inline void set_all_modules_text_ro(void) { }
+static inline void module_enable_ro(const struct module *mod) { }
+static inline void module_disable_ro(const struct module *mod) { }
 #endif
 
 #ifdef CONFIG_GENERIC_BUG
diff --git a/include/linux/mount.h b/include/linux/mount.h
index dc6cd800cd5d..b606d8f57adf 100644
--- a/include/linux/mount.h
+++ b/include/linux/mount.h
@@ -67,6 +67,7 @@ struct vfsmount {
 	struct dentry *mnt_root;	/* root of the mounted tree */
 	struct super_block *mnt_sb;	/* pointer to superblock */
 	int mnt_flags;
+	void *data;
 };
 
 struct file; /* forward dec */
diff --git a/include/linux/namei.h b/include/linux/namei.h
index d8c6334cd150..023359f18567 100644
--- a/include/linux/namei.h
+++ b/include/linux/namei.h
@@ -75,8 +75,11 @@ extern struct dentry *user_path_create(int, const char __user *, struct path *,
 extern void done_path_create(struct path *, struct dentry *);
 extern struct dentry *kern_path_locked(const char *, struct path *);
 extern int kern_path_mountpoint(int, const char *, struct path *, unsigned int);
+extern int vfs_path_lookup(struct dentry *, struct vfsmount *,
+		const char *, unsigned int, struct path *);
 
 extern struct dentry *lookup_one_len(const char *, struct dentry *, int);
+extern struct dentry *lookup_one_len2(const char *, struct vfsmount *mnt, struct dentry *, int);
 
 extern int follow_down_one(struct path *);
 extern int follow_down(struct path *);
diff --git a/include/linux/netfilter/xt_qtaguid.h b/include/linux/netfilter/xt_qtaguid.h
new file mode 100644
index 000000000000..1c671552ec37
--- /dev/null
+++ b/include/linux/netfilter/xt_qtaguid.h
@@ -0,0 +1,14 @@
+#ifndef _XT_QTAGUID_MATCH_H
+#define _XT_QTAGUID_MATCH_H
+
+/* For now we just replace the xt_owner.
+ * FIXME: make iptables aware of qtaguid. */
+#include <linux/netfilter/xt_owner.h>
+
+#define XT_QTAGUID_UID    XT_OWNER_UID
+#define XT_QTAGUID_GID    XT_OWNER_GID
+#define XT_QTAGUID_SOCKET XT_OWNER_SOCKET
+#define xt_qtaguid_match_info xt_owner_match_info
+
+int qtaguid_untag(struct socket *sock, bool kernel);
+#endif /* _XT_QTAGUID_MATCH_H */
diff --git a/include/linux/netfilter/xt_quota2.h b/include/linux/netfilter/xt_quota2.h
new file mode 100644
index 000000000000..eadc6903314e
--- /dev/null
+++ b/include/linux/netfilter/xt_quota2.h
@@ -0,0 +1,25 @@
+#ifndef _XT_QUOTA_H
+#define _XT_QUOTA_H
+
+enum xt_quota_flags {
+	XT_QUOTA_INVERT    = 1 << 0,
+	XT_QUOTA_GROW      = 1 << 1,
+	XT_QUOTA_PACKET    = 1 << 2,
+	XT_QUOTA_NO_CHANGE = 1 << 3,
+	XT_QUOTA_MASK      = 0x0F,
+};
+
+struct xt_quota_counter;
+
+struct xt_quota_mtinfo2 {
+	char name[15];
+	u_int8_t flags;
+
+	/* Comparison-invariant */
+	aligned_u64 quota;
+
+	/* Used internally by the kernel */
+	struct xt_quota_counter *master __attribute__((aligned(8)));
+};
+
+#endif /* _XT_QUOTA_H */
diff --git a/include/linux/nmi.h b/include/linux/nmi.h
index 7ec5b86735f3..74385351935d 100644
--- a/include/linux/nmi.h
+++ b/include/linux/nmi.h
@@ -14,8 +14,11 @@
  * may be used to reset the timeout - for code which intentionally
  * disables interrupts for a long time. This call is stateless.
  */
-#if defined(CONFIG_HAVE_NMI_WATCHDOG) || defined(CONFIG_HARDLOCKUP_DETECTOR)
+#if defined(CONFIG_HAVE_NMI_WATCHDOG) || defined(CONFIG_HARDLOCKUP_DETECTOR_NMI)
 #include <asm/nmi.h>
+#endif
+
+#if defined(CONFIG_HAVE_NMI_WATCHDOG) || defined(CONFIG_HARDLOCKUP_DETECTOR)
 extern void touch_nmi_watchdog(void);
 #else
 static inline void touch_nmi_watchdog(void)
diff --git a/include/linux/of_fdt.h b/include/linux/of_fdt.h
index df9ef3801812..fbbb89b69012 100644
--- a/include/linux/of_fdt.h
+++ b/include/linux/of_fdt.h
@@ -59,6 +59,27 @@ extern int of_flat_dt_match(unsigned long node, const char *const *matches);
 extern unsigned long of_get_flat_dt_root(void);
 extern int of_get_flat_dt_size(void);
 
+/*
+ * early_init_dt_scan_chosen - scan the device tree for ramdisk and bootargs
+ *
+ * The boot arguments will be placed into the memory pointed to by @data.
+ * That memory should be COMMAND_LINE_SIZE big and initialized to be a valid
+ * (possibly empty) string.  Logic for what will be in @data after this
+ * function finishes:
+ *
+ * - CONFIG_CMDLINE_FORCE=true
+ *     CONFIG_CMDLINE
+ * - CONFIG_CMDLINE_EXTEND=true, @data is non-empty string
+ *     @data + dt bootargs (even if dt bootargs are empty)
+ * - CONFIG_CMDLINE_EXTEND=true, @data is empty string
+ *     CONFIG_CMDLINE + dt bootargs (even if dt bootargs are empty)
+ * - CMDLINE_FROM_BOOTLOADER=true, dt bootargs=non-empty:
+ *     dt bootargs
+ * - CMDLINE_FROM_BOOTLOADER=true, dt bootargs=empty, @data is non-empty string
+ *     @data is left unchanged
+ * - CMDLINE_FROM_BOOTLOADER=true, dt bootargs=empty, @data is empty string
+ *     CONFIG_CMDLINE (or "" if that's not defined)
+ */
 extern int early_init_dt_scan_chosen(unsigned long node, const char *uname,
 				     int depth, void *data);
 extern int early_init_dt_scan_memory(unsigned long node, const char *uname,
diff --git a/include/linux/percpu-rwsem.h b/include/linux/percpu-rwsem.h
index c2fa3ecb0dce..146efefde2a1 100644
--- a/include/linux/percpu-rwsem.h
+++ b/include/linux/percpu-rwsem.h
@@ -10,30 +10,96 @@
 
 struct percpu_rw_semaphore {
 	struct rcu_sync		rss;
-	unsigned int __percpu	*fast_read_ctr;
+	unsigned int __percpu	*read_count;
 	struct rw_semaphore	rw_sem;
-	atomic_t		slow_read_ctr;
-	wait_queue_head_t	write_waitq;
+	wait_queue_head_t	writer;
+	int			readers_block;
 };
 
-extern void percpu_down_read(struct percpu_rw_semaphore *);
-extern int  percpu_down_read_trylock(struct percpu_rw_semaphore *);
-extern void percpu_up_read(struct percpu_rw_semaphore *);
+extern int __percpu_down_read(struct percpu_rw_semaphore *, int);
+extern void __percpu_up_read(struct percpu_rw_semaphore *);
+
+static inline void percpu_down_read(struct percpu_rw_semaphore *sem)
+{
+	might_sleep();
+
+	rwsem_acquire_read(&sem->rw_sem.dep_map, 0, 0, _RET_IP_);
+
+	preempt_disable();
+	/*
+	 * We are in an RCU-sched read-side critical section, so the writer
+	 * cannot both change sem->state from readers_fast and start checking
+	 * counters while we are here. So if we see !sem->state, we know that
+	 * the writer won't be checking until we're past the preempt_enable()
+	 * and that one the synchronize_sched() is done, the writer will see
+	 * anything we did within this RCU-sched read-size critical section.
+	 */
+	__this_cpu_inc(*sem->read_count);
+	if (unlikely(!rcu_sync_is_idle(&sem->rss)))
+		__percpu_down_read(sem, false); /* Unconditional memory barrier */
+	preempt_enable();
+	/*
+	 * The barrier() from preempt_enable() prevents the compiler from
+	 * bleeding the critical section out.
+	 */
+}
+
+static inline int percpu_down_read_trylock(struct percpu_rw_semaphore *sem)
+{
+	int ret = 1;
+
+	preempt_disable();
+	/*
+	 * Same as in percpu_down_read().
+	 */
+	__this_cpu_inc(*sem->read_count);
+	if (unlikely(!rcu_sync_is_idle(&sem->rss)))
+		ret = __percpu_down_read(sem, true); /* Unconditional memory barrier */
+	preempt_enable();
+	/*
+	 * The barrier() from preempt_enable() prevents the compiler from
+	 * bleeding the critical section out.
+	 */
+
+	if (ret)
+		rwsem_acquire_read(&sem->rw_sem.dep_map, 0, 1, _RET_IP_);
+
+	return ret;
+}
+
+static inline void percpu_up_read(struct percpu_rw_semaphore *sem)
+{
+	/*
+	 * The barrier() in preempt_disable() prevents the compiler from
+	 * bleeding the critical section out.
+	 */
+	preempt_disable();
+	/*
+	 * Same as in percpu_down_read().
+	 */
+	if (likely(rcu_sync_is_idle(&sem->rss)))
+		__this_cpu_dec(*sem->read_count);
+	else
+		__percpu_up_read(sem); /* Unconditional memory barrier */
+	preempt_enable();
+
+	rwsem_release(&sem->rw_sem.dep_map, 1, _RET_IP_);
+}
 
 extern void percpu_down_write(struct percpu_rw_semaphore *);
 extern void percpu_up_write(struct percpu_rw_semaphore *);
 
 extern int __percpu_init_rwsem(struct percpu_rw_semaphore *,
 				const char *, struct lock_class_key *);
+
 extern void percpu_free_rwsem(struct percpu_rw_semaphore *);
 
-#define percpu_init_rwsem(brw)	\
+#define percpu_init_rwsem(sem)					\
 ({								\
 	static struct lock_class_key rwsem_key;			\
-	__percpu_init_rwsem(brw, #brw, &rwsem_key);		\
+	__percpu_init_rwsem(sem, #sem, &rwsem_key);		\
 })
 
-
 #define percpu_rwsem_is_held(sem) lockdep_is_held(&(sem)->rw_sem)
 
 static inline void percpu_rwsem_release(struct percpu_rw_semaphore *sem,
diff --git a/include/linux/perf/arm_pmu.h b/include/linux/perf/arm_pmu.h
index bfa673bb822d..0e55e4016f49 100644
--- a/include/linux/perf/arm_pmu.h
+++ b/include/linux/perf/arm_pmu.h
@@ -107,6 +107,7 @@ struct arm_pmu {
 	struct platform_device	*plat_device;
 	struct pmu_hw_events	__percpu *hw_events;
 	struct notifier_block	hotplug_nb;
+	struct notifier_block	cpu_pm_nb;
 };
 
 #define to_arm_pmu(p) (container_of(p, struct arm_pmu, pmu))
diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index a288010667dc..9dcb00ca2ec1 100644
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -999,6 +999,11 @@ extern int perf_cpu_time_max_percent_handler(struct ctl_table *table, int write,
 		loff_t *ppos);
 
 
+static inline bool perf_paranoid_any(void)
+{
+	return sysctl_perf_event_paranoid > 2;
+}
+
 static inline bool perf_paranoid_tracepoint_raw(void)
 {
 	return sysctl_perf_event_paranoid > -1;
diff --git a/include/linux/platform_data/ds2482.h b/include/linux/platform_data/ds2482.h
new file mode 100644
index 000000000000..5a6879e2a09a
--- /dev/null
+++ b/include/linux/platform_data/ds2482.h
@@ -0,0 +1,21 @@
+/*
+ * Copyright (C) 2012 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __PLATFORM_DATA_DS2482__
+#define __PLATFORM_DATA_DS2482__
+
+struct ds2482_platform_data {
+	int		slpz_gpio;
+};
+
+#endif /* __PLATFORM_DATA_DS2482__ */
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 528be6787796..6a5d654f4447 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -573,6 +573,7 @@ struct dev_pm_info {
 	struct wakeup_source	*wakeup;
 	bool			wakeup_path:1;
 	bool			syscore:1;
+	bool			no_pm_callbacks:1;	/* Owned by the PM core */
 #else
 	unsigned int		should_wakeup:1;
 #endif
diff --git a/include/linux/poll.h b/include/linux/poll.h
index c08386fb3e08..9fb4f40d9a26 100644
--- a/include/linux/poll.h
+++ b/include/linux/poll.h
@@ -96,7 +96,7 @@ extern void poll_initwait(struct poll_wqueues *pwq);
 extern void poll_freewait(struct poll_wqueues *pwq);
 extern int poll_schedule_timeout(struct poll_wqueues *pwq, int state,
 				 ktime_t *expires, unsigned long slack);
-extern long select_estimate_accuracy(struct timespec *tv);
+extern u64 select_estimate_accuracy(struct timespec *tv);
 
 
 static inline int poll_schedule(struct poll_wqueues *pwq, int state)
diff --git a/include/linux/power_supply.h b/include/linux/power_supply.h
index ef9f1592185d..1c075892c6fd 100644
--- a/include/linux/power_supply.h
+++ b/include/linux/power_supply.h
@@ -18,6 +18,7 @@
 #include <linux/leds.h>
 #include <linux/spinlock.h>
 #include <linux/notifier.h>
+#include <linux/types.h>
 
 /*
  * All voltages, currents, charges, energies, time and temperatures in uV,
@@ -148,6 +149,12 @@ enum power_supply_property {
 	POWER_SUPPLY_PROP_SCOPE,
 	POWER_SUPPLY_PROP_CHARGE_TERM_CURRENT,
 	POWER_SUPPLY_PROP_CALIBRATE,
+	/* Local extensions */
+	POWER_SUPPLY_PROP_USB_HC,
+	POWER_SUPPLY_PROP_USB_OTG,
+	POWER_SUPPLY_PROP_CHARGE_ENABLED,
+	/* Local extensions of type int64_t */
+	POWER_SUPPLY_PROP_CHARGE_COUNTER_EXT,
 	/* Properties of type `const char *' */
 	POWER_SUPPLY_PROP_MODEL_NAME,
 	POWER_SUPPLY_PROP_MANUFACTURER,
@@ -172,6 +179,7 @@ enum power_supply_notifier_events {
 union power_supply_propval {
 	int intval;
 	const char *strval;
+	int64_t int64val;
 };
 
 struct device_node;
diff --git a/include/linux/pstore.h b/include/linux/pstore.h
index 831479f8df8f..5cae2c6c90ad 100644
--- a/include/linux/pstore.h
+++ b/include/linux/pstore.h
@@ -22,12 +22,13 @@
 #ifndef _LINUX_PSTORE_H
 #define _LINUX_PSTORE_H
 
-#include <linux/time.h>
+#include <linux/compiler.h>
+#include <linux/errno.h>
 #include <linux/kmsg_dump.h>
 #include <linux/mutex.h>
-#include <linux/types.h>
 #include <linux/spinlock.h>
-#include <linux/errno.h>
+#include <linux/time.h>
+#include <linux/types.h>
 
 /* types */
 enum pstore_type_id {
@@ -67,6 +68,10 @@ struct pstore_info {
 			enum kmsg_dump_reason reason, u64 *id,
 			unsigned int part, const char *buf, bool compressed,
 			size_t size, struct pstore_info *psi);
+	int		(*write_buf_user)(enum pstore_type_id type,
+			enum kmsg_dump_reason reason, u64 *id,
+			unsigned int part, const char __user *buf,
+			bool compressed, size_t size, struct pstore_info *psi);
 	int		(*erase)(enum pstore_type_id type, u64 id,
 			int count, struct timespec time,
 			struct pstore_info *psi);
diff --git a/include/linux/pstore_ram.h b/include/linux/pstore_ram.h
index 6f4520a82197..7097a45dbc25 100644
--- a/include/linux/pstore_ram.h
+++ b/include/linux/pstore_ram.h
@@ -17,11 +17,12 @@
 #ifndef __LINUX_PSTORE_RAM_H__
 #define __LINUX_PSTORE_RAM_H__
 
+#include <linux/compiler.h>
 #include <linux/device.h>
+#include <linux/init.h>
 #include <linux/kernel.h>
 #include <linux/list.h>
 #include <linux/types.h>
-#include <linux/init.h>
 
 /*
  * Choose whether access to the RAM zone requires locking or not.  If a zone
@@ -68,7 +69,9 @@ void persistent_ram_free(struct persistent_ram_zone *prz);
 void persistent_ram_zap(struct persistent_ram_zone *prz);
 
 int persistent_ram_write(struct persistent_ram_zone *prz, const void *s,
-	unsigned int count);
+			 unsigned int count);
+int persistent_ram_write_user(struct persistent_ram_zone *prz,
+			      const void __user *s, unsigned int count);
 
 void persistent_ram_save_old(struct persistent_ram_zone *prz);
 size_t persistent_ram_old_size(struct persistent_ram_zone *prz);
@@ -77,6 +80,8 @@ void persistent_ram_free_old(struct persistent_ram_zone *prz);
 ssize_t persistent_ram_ecc_string(struct persistent_ram_zone *prz,
 	char *str, size_t len);
 
+void ramoops_console_write_buf(const char *buf, size_t size);
+
 /*
  * Ramoops platform data
  * @mem_size	memory size for ramoops
diff --git a/include/linux/rcu_sync.h b/include/linux/rcu_sync.h
index a63a33e6196e..ece7ed9a4a70 100644
--- a/include/linux/rcu_sync.h
+++ b/include/linux/rcu_sync.h
@@ -59,6 +59,7 @@ static inline bool rcu_sync_is_idle(struct rcu_sync *rsp)
 }
 
 extern void rcu_sync_init(struct rcu_sync *, enum rcu_sync_type);
+extern void rcu_sync_enter_start(struct rcu_sync *);
 extern void rcu_sync_enter(struct rcu_sync *);
 extern void rcu_sync_exit(struct rcu_sync *);
 extern void rcu_sync_dtor(struct rcu_sync *);
diff --git a/include/linux/restart_block.h b/include/linux/restart_block.h
new file mode 100644
index 000000000000..0d905d8ec553
--- /dev/null
+++ b/include/linux/restart_block.h
@@ -0,0 +1,51 @@
+/*
+ * Common syscall restarting data
+ */
+#ifndef __LINUX_RESTART_BLOCK_H
+#define __LINUX_RESTART_BLOCK_H
+
+#include <linux/compiler.h>
+#include <linux/types.h>
+
+struct timespec;
+struct compat_timespec;
+struct pollfd;
+
+/*
+ * System call restart block.
+ */
+struct restart_block {
+	long (*fn)(struct restart_block *);
+	union {
+		/* For futex_wait and futex_wait_requeue_pi */
+		struct {
+			u32 __user *uaddr;
+			u32 val;
+			u32 flags;
+			u32 bitset;
+			u64 time;
+			u32 __user *uaddr2;
+		} futex;
+		/* For nanosleep */
+		struct {
+			clockid_t clockid;
+			struct timespec __user *rmtp;
+#ifdef CONFIG_COMPAT
+			struct compat_timespec __user *compat_rmtp;
+#endif
+			u64 expires;
+		} nanosleep;
+		/* For poll */
+		struct {
+			struct pollfd __user *ufds;
+			int nfds;
+			int has_timeout;
+			unsigned long tv_sec;
+			unsigned long tv_nsec;
+		} poll;
+	};
+};
+
+extern long do_no_restart_syscall(struct restart_block *parm);
+
+#endif /* __LINUX_RESTART_BLOCK_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index e887c8d6f395..60af9d2fa922 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -173,6 +173,9 @@ extern bool single_task_running(void);
 extern unsigned long nr_iowait(void);
 extern unsigned long nr_iowait_cpu(int cpu);
 extern void get_iowait_load(unsigned long *nr_waiters, unsigned long *load);
+#ifdef CONFIG_CPU_QUIET
+extern u64 nr_running_integral(unsigned int cpu);
+#endif
 
 extern void calc_global_load(unsigned long ticks);
 
@@ -219,9 +222,10 @@ extern void proc_sched_set_task(struct task_struct *p);
 #define TASK_WAKING		256
 #define TASK_PARKED		512
 #define TASK_NOLOAD		1024
-#define TASK_STATE_MAX		2048
+#define TASK_NEW		2048
+#define TASK_STATE_MAX		4096
 
-#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPN"
+#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPNn"
 
 extern char ___assert_task_state[1 - 2*!!(
 		sizeof(TASK_STATE_TO_CHAR_STR)-1 != ilog2(TASK_STATE_MAX)+1)];
@@ -314,6 +318,15 @@ extern char ___assert_task_state[1 - 2*!!(
 /* Task command name length */
 #define TASK_COMM_LEN 16
 
+enum task_event {
+	PUT_PREV_TASK   = 0,
+	PICK_NEXT_TASK  = 1,
+	TASK_WAKE       = 2,
+	TASK_MIGRATE    = 3,
+	TASK_UPDATE     = 4,
+	IRQ_UPDATE	= 5,
+};
+
 #include <linux/spinlock.h>
 
 /*
@@ -939,6 +952,14 @@ enum cpu_idle_type {
 #define SCHED_CAPACITY_SHIFT	10
 #define SCHED_CAPACITY_SCALE	(1L << SCHED_CAPACITY_SHIFT)
 
+struct sched_capacity_reqs {
+	unsigned long cfs;
+	unsigned long rt;
+	unsigned long dl;
+
+	unsigned long total;
+};
+
 /*
  * Wake-queues are lists of tasks with a pending wakeup, whose
  * callers have already marked the task as woken internally,
@@ -972,12 +993,13 @@ struct wake_q_node {
 struct wake_q_head {
 	struct wake_q_node *first;
 	struct wake_q_node **lastp;
+	int count;
 };
 
 #define WAKE_Q_TAIL ((struct wake_q_node *) 0x01)
 
 #define WAKE_Q(name)					\
-	struct wake_q_head name = { WAKE_Q_TAIL, &name.first }
+	struct wake_q_head name = { WAKE_Q_TAIL, &name.first, 0 }
 
 extern void wake_q_add(struct wake_q_head *head,
 		       struct task_struct *task);
@@ -993,7 +1015,8 @@ extern void wake_up_q(struct wake_q_head *head);
 #define SD_BALANCE_FORK		0x0008	/* Balance on fork, clone */
 #define SD_BALANCE_WAKE		0x0010  /* Balance on wakeup */
 #define SD_WAKE_AFFINE		0x0020	/* Wake task to waking CPU */
-#define SD_SHARE_CPUCAPACITY	0x0080	/* Domain members share cpu power */
+#define SD_ASYM_CPUCAPACITY	0x0040  /* Groups have different max cpu capacities */
+#define SD_SHARE_CPUCAPACITY	0x0080	/* Domain members share cpu capacity */
 #define SD_SHARE_POWERDOMAIN	0x0100	/* Domain members share power domain */
 #define SD_SHARE_PKG_RESOURCES	0x0200	/* Domain members share cpu pkg resources */
 #define SD_SERIALIZE		0x0400	/* Only a single load balancing instance */
@@ -1001,6 +1024,7 @@ extern void wake_up_q(struct wake_q_head *head);
 #define SD_PREFER_SIBLING	0x1000	/* Prefer to place tasks in a sibling domain */
 #define SD_OVERLAP		0x2000	/* sched_domains of this level overlap */
 #define SD_NUMA			0x4000	/* cross-node balancing */
+#define SD_SHARE_CAP_STATES	0x8000  /* Domain members share capacity state */
 
 #ifdef CONFIG_SCHED_SMT
 static inline int cpu_smt_flags(void)
@@ -1033,8 +1057,57 @@ struct sched_domain_attr {
 
 extern int sched_domain_level_max;
 
+struct capacity_state {
+	unsigned long cap;	/* compute capacity */
+	unsigned long power;	/* power consumption at this compute capacity */
+};
+
+struct idle_state {
+	unsigned long power;	 /* power consumption in this idle state */
+};
+
+struct sched_group_energy {
+	unsigned int nr_idle_states;	/* number of idle states */
+	struct idle_state *idle_states;	/* ptr to idle state array */
+	unsigned int nr_cap_states;	/* number of capacity states */
+	struct capacity_state *cap_states; /* ptr to capacity state array */
+};
+
+unsigned long capacity_curr_of(int cpu);
+
 struct sched_group;
 
+struct eas_stats {
+	/* select_idle_sibling() stats */
+	u64 sis_attempts;
+	u64 sis_idle;
+	u64 sis_cache_affine;
+	u64 sis_suff_cap;
+	u64 sis_idle_cpu;
+	u64 sis_count;
+
+	/* select_energy_cpu_brute() stats */
+	u64 secb_attempts;
+	u64 secb_sync;
+	u64 secb_idle_bt;
+	u64 secb_insuff_cap;
+	u64 secb_no_nrg_sav;
+	u64 secb_nrg_sav;
+	u64 secb_count;
+
+	/* find_best_target() stats */
+	u64 fbt_attempts;
+	u64 fbt_no_cpu;
+	u64 fbt_no_sd;
+	u64 fbt_pref_idle;
+	u64 fbt_count;
+
+	/* cas */
+	/* select_task_rq_fair() stats */
+	u64 cas_attempts;
+	u64 cas_count;
+};
+
 struct sched_domain {
 	/* These fields must be setup */
 	struct sched_domain *parent;	/* top domain must be null terminated */
@@ -1095,6 +1168,8 @@ struct sched_domain {
 	unsigned int ttwu_wake_remote;
 	unsigned int ttwu_move_affine;
 	unsigned int ttwu_move_balance;
+
+	struct eas_stats eas_stats;
 #endif
 #ifdef CONFIG_SCHED_DEBUG
 	char *name;
@@ -1131,6 +1206,8 @@ bool cpus_share_cache(int this_cpu, int that_cpu);
 
 typedef const struct cpumask *(*sched_domain_mask_f)(int cpu);
 typedef int (*sched_domain_flags_f)(void);
+typedef
+const struct sched_group_energy * const(*sched_domain_energy_f)(int cpu);
 
 #define SDTL_OVERLAP	0x01
 
@@ -1143,6 +1220,7 @@ struct sd_data {
 struct sched_domain_topology_level {
 	sched_domain_mask_f mask;
 	sched_domain_flags_f sd_flags;
+	sched_domain_energy_f energy;
 	int		    flags;
 	int		    numa_level;
 	struct sd_data      data;
@@ -1250,6 +1328,70 @@ struct sched_statistics {
 	u64			nr_wakeups_affine_attempts;
 	u64			nr_wakeups_passive;
 	u64			nr_wakeups_idle;
+
+	/* select_idle_sibling() */
+	u64			nr_wakeups_sis_attempts;
+	u64			nr_wakeups_sis_idle;
+	u64			nr_wakeups_sis_cache_affine;
+	u64			nr_wakeups_sis_suff_cap;
+	u64			nr_wakeups_sis_idle_cpu;
+	u64			nr_wakeups_sis_count;
+
+	/* energy_aware_wake_cpu() */
+	u64			nr_wakeups_secb_attempts;
+	u64			nr_wakeups_secb_sync;
+	u64			nr_wakeups_secb_idle_bt;
+	u64			nr_wakeups_secb_insuff_cap;
+	u64			nr_wakeups_secb_no_nrg_sav;
+	u64			nr_wakeups_secb_nrg_sav;
+	u64			nr_wakeups_secb_count;
+
+	/* find_best_target() */
+	u64			nr_wakeups_fbt_attempts;
+	u64			nr_wakeups_fbt_no_cpu;
+	u64			nr_wakeups_fbt_no_sd;
+	u64			nr_wakeups_fbt_pref_idle;
+	u64			nr_wakeups_fbt_count;
+
+	/* cas */
+	/* select_task_rq_fair() */
+	u64			nr_wakeups_cas_attempts;
+	u64			nr_wakeups_cas_count;
+};
+#endif
+
+#ifdef CONFIG_SCHED_WALT
+#define RAVG_HIST_SIZE_MAX  5
+
+/* ravg represents frequency scaled cpu-demand of tasks */
+struct ravg {
+	/*
+	 * 'mark_start' marks the beginning of an event (task waking up, task
+	 * starting to execute, task being preempted) within a window
+	 *
+	 * 'sum' represents how runnable a task has been within current
+	 * window. It incorporates both running time and wait time and is
+	 * frequency scaled.
+	 *
+	 * 'sum_history' keeps track of history of 'sum' seen over previous
+	 * RAVG_HIST_SIZE windows. Windows where task was entirely sleeping are
+	 * ignored.
+	 *
+	 * 'demand' represents maximum sum seen over previous
+	 * sysctl_sched_ravg_hist_size windows. 'demand' could drive frequency
+	 * demand for tasks.
+	 *
+	 * 'curr_window' represents task's contribution to cpu busy time
+	 * statistics (rq->curr_runnable_sum) in current window
+	 *
+	 * 'prev_window' represents task's contribution to cpu busy time
+	 * statistics (rq->prev_runnable_sum) in previous window
+	 */
+	u64 mark_start;
+	u32 sum, demand;
+	u32 sum_history[RAVG_HIST_SIZE_MAX];
+	u32 curr_window, prev_window;
+	u16 active_windows;
 };
 #endif
 
@@ -1388,6 +1530,13 @@ struct tlbflush_unmap_batch {
 };
 
 struct task_struct {
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	/*
+	 * For reasons of header soup (see current_thread_info()), this
+	 * must be the first element of task_struct.
+	 */
+	struct thread_info thread_info;
+#endif
 	volatile long state;	/* -1 unrunnable, 0 runnable, >0 stopped */
 	void *stack;
 	atomic_t usage;
@@ -1397,6 +1546,9 @@ struct task_struct {
 #ifdef CONFIG_SMP
 	struct llist_node wake_entry;
 	int on_cpu;
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	unsigned int cpu;	/* current CPU */
+#endif
 	unsigned int wakee_flips;
 	unsigned long wakee_flip_decay_ts;
 	struct task_struct *last_wakee;
@@ -1410,6 +1562,16 @@ struct task_struct {
 	const struct sched_class *sched_class;
 	struct sched_entity se;
 	struct sched_rt_entity rt;
+#ifdef CONFIG_SCHED_WALT
+	struct ravg ravg;
+	/*
+	 * 'init_load_pct' represents the initial task load assigned to children
+	 * of this task
+	 */
+	u32 init_load_pct;
+	u64 last_sleep_ts;
+#endif
+
 #ifdef CONFIG_CGROUP_SCHED
 	struct task_group *sched_task_group;
 #endif
@@ -1784,8 +1946,8 @@ struct task_struct {
 	 * time slack values; these are used to round up poll() and
 	 * select() etc timeout values. These are in nanoseconds.
 	 */
-	unsigned long timer_slack_ns;
-	unsigned long default_timer_slack_ns;
+	u64 timer_slack_ns;
+	u64 default_timer_slack_ns;
 
 #ifdef CONFIG_KASAN
 	unsigned int kasan_depth;
@@ -2421,7 +2583,9 @@ extern void set_curr_task(int cpu, struct task_struct *p);
 void yield(void);
 
 union thread_union {
+#ifndef CONFIG_THREAD_INFO_IN_TASK
 	struct thread_info thread_info;
+#endif
 	unsigned long stack[THREAD_SIZE/sizeof(long)];
 };
 
@@ -2603,6 +2767,11 @@ static inline void mmdrop(struct mm_struct * mm)
 
 /* mmput gets rid of the mappings and all user-space */
 extern void mmput(struct mm_struct *);
+/* same as above but performs the slow path from the async kontext. Can
+ * be called from the atomic context as well
+ */
+extern void mmput_async(struct mm_struct *);
+
 /* Grab a reference to a task's mm, if it is not already going away */
 extern struct mm_struct *get_task_mm(struct task_struct *task);
 /*
@@ -2811,10 +2980,34 @@ static inline void threadgroup_change_end(struct task_struct *tsk)
 	cgroup_threadgroup_change_end(tsk);
 }
 
-#ifndef __HAVE_THREAD_FUNCTIONS
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+
+static inline struct thread_info *task_thread_info(struct task_struct *task)
+{
+	return &task->thread_info;
+}
+
+/*
+ * When accessing the stack of a non-current task that might exit, use
+ * try_get_task_stack() instead.  task_stack_page will return a pointer
+ * that could get freed out from under you.
+ */
+static inline void *task_stack_page(const struct task_struct *task)
+{
+	return task->stack;
+}
+
+#define setup_thread_stack(new,old)	do { } while(0)
+
+static inline unsigned long *end_of_stack(const struct task_struct *task)
+{
+	return task->stack;
+}
+
+#elif !defined(__HAVE_THREAD_FUNCTIONS)
 
 #define task_thread_info(task)	((struct thread_info *)(task)->stack)
-#define task_stack_page(task)	((task)->stack)
+#define task_stack_page(task)	((void *)(task)->stack)
 
 static inline void setup_thread_stack(struct task_struct *p, struct task_struct *org)
 {
@@ -2841,6 +3034,14 @@ static inline unsigned long *end_of_stack(struct task_struct *p)
 }
 
 #endif
+
+static inline void *try_get_task_stack(struct task_struct *tsk)
+{
+	return task_stack_page(tsk);
+}
+
+static inline void put_task_stack(struct task_struct *tsk) {}
+
 #define task_stack_end_corrupted(task) \
 		(*(end_of_stack(task)) != STACK_END_MAGIC)
 
@@ -2851,7 +3052,7 @@ static inline int object_is_on_stack(void *obj)
 	return (obj >= stack) && (obj < (stack + THREAD_SIZE));
 }
 
-extern void thread_info_cache_init(void);
+extern void thread_stack_cache_init(void);
 
 #ifdef CONFIG_DEBUG_STACK_USAGE
 static inline unsigned long stack_not_used(struct task_struct *p)
@@ -3106,7 +3307,11 @@ static inline void ptrace_signal_wake_up(struct task_struct *t, bool resume)
 
 static inline unsigned int task_cpu(const struct task_struct *p)
 {
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	return p->cpu;
+#else
 	return task_thread_info(p)->cpu;
+#endif
 }
 
 static inline int task_node(const struct task_struct *p)
@@ -3159,6 +3364,11 @@ static inline void inc_syscw(struct task_struct *tsk)
 {
 	tsk->ioac.syscw++;
 }
+
+static inline void inc_syscfs(struct task_struct *tsk)
+{
+	tsk->ioac.syscfs++;
+}
 #else
 static inline void add_rchar(struct task_struct *tsk, ssize_t amt)
 {
@@ -3175,6 +3385,9 @@ static inline void inc_syscr(struct task_struct *tsk)
 static inline void inc_syscw(struct task_struct *tsk)
 {
 }
+static inline void inc_syscfs(struct task_struct *tsk)
+{
+}
 #endif
 
 #ifndef TASK_SIZE_OF
@@ -3211,4 +3424,19 @@ static inline unsigned long rlimit_max(unsigned int limit)
 	return task_rlimit_max(current, limit);
 }
 
+#define SCHED_CPUFREQ_RT        (1U << 0)
+#define SCHED_CPUFREQ_DL        (1U << 1)
+#define SCHED_CPUFREQ_IOWAIT    (1U << 2)
+
+#ifdef CONFIG_CPU_FREQ
+struct update_util_data {
+	void (*func)(struct update_util_data *data, u64 time, unsigned int flags);
+};
+
+void cpufreq_add_update_util_hook(int cpu, struct update_util_data *data,
+                       void (*func)(struct update_util_data *data, u64 time,
+                                    unsigned int flags));
+void cpufreq_remove_update_util_hook(int cpu);
+#endif /* CONFIG_CPU_FREQ */
+
 #endif
diff --git a/include/linux/sched/sysctl.h b/include/linux/sched/sysctl.h
index c9e4731cf10b..167279a4e24b 100644
--- a/include/linux/sched/sysctl.h
+++ b/include/linux/sched/sysctl.h
@@ -39,6 +39,14 @@ extern unsigned int sysctl_sched_latency;
 extern unsigned int sysctl_sched_min_granularity;
 extern unsigned int sysctl_sched_wakeup_granularity;
 extern unsigned int sysctl_sched_child_runs_first;
+extern unsigned int sysctl_sched_sync_hint_enable;
+extern unsigned int sysctl_sched_cstate_aware;
+#ifdef CONFIG_SCHED_WALT
+extern unsigned int sysctl_sched_use_walt_cpu_util;
+extern unsigned int sysctl_sched_use_walt_task_util;
+extern unsigned int sysctl_sched_walt_init_task_load_pct;
+extern unsigned int sysctl_sched_walt_cpu_high_irqload;
+#endif
 
 enum sched_tunable_scaling {
 	SCHED_TUNABLESCALING_NONE,
@@ -77,6 +85,22 @@ extern int sysctl_sched_rt_runtime;
 extern unsigned int sysctl_sched_cfs_bandwidth_slice;
 #endif
 
+#ifdef CONFIG_SCHED_TUNE
+extern unsigned int sysctl_sched_cfs_boost;
+int sysctl_sched_cfs_boost_handler(struct ctl_table *table, int write,
+				   void __user *buffer, size_t *length,
+				   loff_t *ppos);
+static inline unsigned int get_sysctl_sched_cfs_boost(void)
+{
+	return sysctl_sched_cfs_boost;
+}
+#else
+static inline unsigned int get_sysctl_sched_cfs_boost(void)
+{
+	return 0;
+}
+#endif
+
 #ifdef CONFIG_SCHED_AUTOGROUP
 extern unsigned int sysctl_sched_autogroup_enabled;
 #endif
diff --git a/include/linux/sched_energy.h b/include/linux/sched_energy.h
new file mode 100644
index 000000000000..1daf3e1f98a7
--- /dev/null
+++ b/include/linux/sched_energy.h
@@ -0,0 +1,44 @@
+#ifndef _LINUX_SCHED_ENERGY_H
+#define _LINUX_SCHED_ENERGY_H
+
+#include <linux/sched.h>
+#include <linux/slab.h>
+
+/*
+ * There doesn't seem to be an NR_CPUS style max number of sched domain
+ * levels so here's an arbitrary constant one for the moment.
+ *
+ * The levels alluded to here correspond to entries in struct
+ * sched_domain_topology_level that are meant to be populated by arch
+ * specific code (topology.c).
+ */
+#define NR_SD_LEVELS 8
+
+#define SD_LEVEL0   0
+#define SD_LEVEL1   1
+#define SD_LEVEL2   2
+#define SD_LEVEL3   3
+#define SD_LEVEL4   4
+#define SD_LEVEL5   5
+#define SD_LEVEL6   6
+#define SD_LEVEL7   7
+
+/*
+ * Convenience macro for iterating through said sd levels.
+ */
+#define for_each_possible_sd_level(level)		    \
+	for (level = 0; level < NR_SD_LEVELS; level++)
+
+#ifdef CONFIG_SMP
+
+extern struct sched_group_energy *sge_array[NR_CPUS][NR_SD_LEVELS];
+
+void init_sched_energy_costs(void);
+
+#else
+
+#define init_sched_energy_costs() do { } while (0)
+
+#endif /* CONFIG_SMP */
+
+#endif
diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h
index 297d4fa1cfe5..0afc11f8f300 100644
--- a/include/linux/serial_core.h
+++ b/include/linux/serial_core.h
@@ -66,6 +66,7 @@ struct uart_ops {
 	void		(*set_ldisc)(struct uart_port *, struct ktermios *);
 	void		(*pm)(struct uart_port *, unsigned int state,
 			      unsigned int oldstate);
+	void		(*wake_peer)(struct uart_port *);
 
 	/*
 	 * Return a string describing the type of the port
diff --git a/include/linux/sock_diag.h b/include/linux/sock_diag.h
index 8ff34ed1ae8a..a0596ca0e80a 100644
--- a/include/linux/sock_diag.h
+++ b/include/linux/sock_diag.h
@@ -15,6 +15,7 @@ struct sock_diag_handler {
 	__u8 family;
 	int (*dump)(struct sk_buff *skb, struct nlmsghdr *nlh);
 	int (*get_info)(struct sk_buff *skb, struct sock *sk);
+	int (*destroy)(struct sk_buff *skb, struct nlmsghdr *nlh);
 };
 
 int sock_diag_register(const struct sock_diag_handler *h);
@@ -74,4 +75,5 @@ bool sock_diag_has_destroy_listeners(const struct sock *sk)
 }
 void sock_diag_broadcast_destroy(struct sock *sk);
 
+int sock_diag_destroy(struct sock *sk, int err);
 #endif
diff --git a/include/linux/suspend.h b/include/linux/suspend.h
index 8b6ec7ef0854..c59803dc68de 100644
--- a/include/linux/suspend.h
+++ b/include/linux/suspend.h
@@ -433,6 +433,7 @@ extern bool pm_get_wakeup_count(unsigned int *count, bool block);
 extern bool pm_save_wakeup_count(unsigned int count);
 extern void pm_wakep_autosleep_enabled(bool set);
 extern void pm_print_active_wakeup_sources(void);
+extern void pm_get_active_wakeup_sources(char *pending_sources, size_t max);
 
 static inline void lock_system_sleep(void)
 {
diff --git a/include/linux/task_io_accounting.h b/include/linux/task_io_accounting.h
index bdf855c2856f..2dd338fdf881 100644
--- a/include/linux/task_io_accounting.h
+++ b/include/linux/task_io_accounting.h
@@ -18,6 +18,8 @@ struct task_io_accounting {
 	u64 syscr;
 	/* # of write syscalls */
 	u64 syscw;
+	/* # of fsync syscalls */
+	u64 syscfs;
 #endif /* CONFIG_TASK_XACCT */
 
 #ifdef CONFIG_TASK_IO_ACCOUNTING
diff --git a/include/linux/task_io_accounting_ops.h b/include/linux/task_io_accounting_ops.h
index 4d090f9ee608..1b505c804af3 100644
--- a/include/linux/task_io_accounting_ops.h
+++ b/include/linux/task_io_accounting_ops.h
@@ -96,6 +96,7 @@ static inline void task_chr_io_accounting_add(struct task_io_accounting *dst,
 	dst->wchar += src->wchar;
 	dst->syscr += src->syscr;
 	dst->syscw += src->syscw;
+	dst->syscfs += src->syscfs;
 }
 #else
 static inline void task_chr_io_accounting_add(struct task_io_accounting *dst,
diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
index eded095fe81e..8933ecc2bc9f 100644
--- a/include/linux/thread_info.h
+++ b/include/linux/thread_info.h
@@ -9,46 +9,17 @@
 
 #include <linux/types.h>
 #include <linux/bug.h>
+#include <linux/restart_block.h>
 
-struct timespec;
-struct compat_timespec;
-
+#ifdef CONFIG_THREAD_INFO_IN_TASK
 /*
- * System call restart block.
+ * For CONFIG_THREAD_INFO_IN_TASK kernels we need <asm/current.h> for the
+ * definition of current, but for !CONFIG_THREAD_INFO_IN_TASK kernels,
+ * including <asm/current.h> can cause a circular dependency on some platforms.
  */
-struct restart_block {
-	long (*fn)(struct restart_block *);
-	union {
-		/* For futex_wait and futex_wait_requeue_pi */
-		struct {
-			u32 __user *uaddr;
-			u32 val;
-			u32 flags;
-			u32 bitset;
-			u64 time;
-			u32 __user *uaddr2;
-		} futex;
-		/* For nanosleep */
-		struct {
-			clockid_t clockid;
-			struct timespec __user *rmtp;
-#ifdef CONFIG_COMPAT
-			struct compat_timespec __user *compat_rmtp;
+#include <asm/current.h>
+#define current_thread_info() ((struct thread_info *)current)
 #endif
-			u64 expires;
-		} nanosleep;
-		/* For poll */
-		struct {
-			struct pollfd __user *ufds;
-			int nfds;
-			int has_timeout;
-			unsigned long tv_sec;
-			unsigned long tv_nsec;
-		} poll;
-	};
-};
-
-extern long do_no_restart_syscall(struct restart_block *parm);
 
 #include <linux/bitops.h>
 #include <asm/thread_info.h>
@@ -158,8 +129,8 @@ static inline int arch_within_stack_frames(const void * const stack,
 extern void __check_object_size(const void *ptr, unsigned long n,
 					bool to_user);
 
-static inline void check_object_size(const void *ptr, unsigned long n,
-				     bool to_user)
+static __always_inline void check_object_size(const void *ptr, unsigned long n,
+					      bool to_user)
 {
 	if (!__builtin_constant_p(n))
 		__check_object_size(ptr, n, to_user);
diff --git a/include/linux/tick.h b/include/linux/tick.h
index e312219ff823..af5ac7f91a3b 100644
--- a/include/linux/tick.h
+++ b/include/linux/tick.h
@@ -103,6 +103,7 @@ extern void tick_nohz_idle_enter(void);
 extern void tick_nohz_idle_exit(void);
 extern void tick_nohz_irq_exit(void);
 extern ktime_t tick_nohz_get_sleep_length(void);
+extern unsigned long tick_nohz_get_idle_calls(void);
 extern u64 get_cpu_idle_time_us(int cpu, u64 *last_update_time);
 extern u64 get_cpu_iowait_time_us(int cpu, u64 *last_update_time);
 #else /* !CONFIG_NO_HZ_COMMON */
diff --git a/include/linux/timekeeping.h b/include/linux/timekeeping.h
index ec89d846324c..b7246d2ed7c9 100644
--- a/include/linux/timekeeping.h
+++ b/include/linux/timekeeping.h
@@ -233,6 +233,7 @@ static inline u64 ktime_get_raw_ns(void)
 
 extern u64 ktime_get_mono_fast_ns(void);
 extern u64 ktime_get_raw_fast_ns(void);
+extern u64 ktime_get_boot_fast_ns(void);
 
 /*
  * Timespec interfaces utilizing the ktime based ones
diff --git a/include/linux/usb/class-dual-role.h b/include/linux/usb/class-dual-role.h
new file mode 100644
index 000000000000..c6df2238012e
--- /dev/null
+++ b/include/linux/usb/class-dual-role.h
@@ -0,0 +1,129 @@
+#ifndef __LINUX_CLASS_DUAL_ROLE_H__
+#define __LINUX_CLASS_DUAL_ROLE_H__
+
+#include <linux/workqueue.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+
+struct device;
+
+enum dual_role_supported_modes {
+	DUAL_ROLE_SUPPORTED_MODES_DFP_AND_UFP = 0,
+	DUAL_ROLE_SUPPORTED_MODES_DFP,
+	DUAL_ROLE_SUPPORTED_MODES_UFP,
+/*The following should be the last element*/
+	DUAL_ROLE_PROP_SUPPORTED_MODES_TOTAL,
+};
+
+enum {
+	DUAL_ROLE_PROP_MODE_UFP = 0,
+	DUAL_ROLE_PROP_MODE_DFP,
+	DUAL_ROLE_PROP_MODE_NONE,
+/*The following should be the last element*/
+	DUAL_ROLE_PROP_MODE_TOTAL,
+};
+
+enum {
+	DUAL_ROLE_PROP_PR_SRC = 0,
+	DUAL_ROLE_PROP_PR_SNK,
+	DUAL_ROLE_PROP_PR_NONE,
+/*The following should be the last element*/
+	DUAL_ROLE_PROP_PR_TOTAL,
+
+};
+
+enum {
+	DUAL_ROLE_PROP_DR_HOST = 0,
+	DUAL_ROLE_PROP_DR_DEVICE,
+	DUAL_ROLE_PROP_DR_NONE,
+/*The following should be the last element*/
+	DUAL_ROLE_PROP_DR_TOTAL,
+};
+
+enum {
+	DUAL_ROLE_PROP_VCONN_SUPPLY_NO = 0,
+	DUAL_ROLE_PROP_VCONN_SUPPLY_YES,
+/*The following should be the last element*/
+	DUAL_ROLE_PROP_VCONN_SUPPLY_TOTAL,
+};
+
+enum dual_role_property {
+	DUAL_ROLE_PROP_SUPPORTED_MODES = 0,
+	DUAL_ROLE_PROP_MODE,
+	DUAL_ROLE_PROP_PR,
+	DUAL_ROLE_PROP_DR,
+	DUAL_ROLE_PROP_VCONN_SUPPLY,
+};
+
+struct dual_role_phy_instance;
+
+/* Description of typec port */
+struct dual_role_phy_desc {
+	/* /sys/class/dual_role_usb/<name>/ */
+	const char *name;
+	enum dual_role_supported_modes supported_modes;
+	enum dual_role_property *properties;
+	size_t num_properties;
+
+	/* Callback for "cat /sys/class/dual_role_usb/<name>/<property>" */
+	int (*get_property)(struct dual_role_phy_instance *dual_role,
+			     enum dual_role_property prop,
+			     unsigned int *val);
+	/* Callback for "echo <value> >
+	 *                      /sys/class/dual_role_usb/<name>/<property>" */
+	int (*set_property)(struct dual_role_phy_instance *dual_role,
+			     enum dual_role_property prop,
+			     const unsigned int *val);
+	/* Decides whether userspace can change a specific property */
+	int (*property_is_writeable)(struct dual_role_phy_instance *dual_role,
+				      enum dual_role_property prop);
+};
+
+struct dual_role_phy_instance {
+	const struct dual_role_phy_desc *desc;
+
+	/* Driver private data */
+	void *drv_data;
+
+	struct device dev;
+	struct work_struct changed_work;
+};
+
+#if IS_ENABLED(CONFIG_DUAL_ROLE_USB_INTF)
+extern void dual_role_instance_changed(struct dual_role_phy_instance
+				       *dual_role);
+extern struct dual_role_phy_instance *__must_check
+devm_dual_role_instance_register(struct device *parent,
+				 const struct dual_role_phy_desc *desc);
+extern void devm_dual_role_instance_unregister(struct device *dev,
+					       struct dual_role_phy_instance
+					       *dual_role);
+extern int dual_role_get_property(struct dual_role_phy_instance *dual_role,
+				  enum dual_role_property prop,
+				  unsigned int *val);
+extern int dual_role_set_property(struct dual_role_phy_instance *dual_role,
+				  enum dual_role_property prop,
+				  const unsigned int *val);
+extern int dual_role_property_is_writeable(struct dual_role_phy_instance
+					   *dual_role,
+					   enum dual_role_property prop);
+extern void *dual_role_get_drvdata(struct dual_role_phy_instance *dual_role);
+#else /* CONFIG_DUAL_ROLE_USB_INTF */
+static inline void dual_role_instance_changed(struct dual_role_phy_instance
+				       *dual_role){}
+static inline struct dual_role_phy_instance *__must_check
+devm_dual_role_instance_register(struct device *parent,
+				 const struct dual_role_phy_desc *desc)
+{
+	return ERR_PTR(-ENOSYS);
+}
+static inline void devm_dual_role_instance_unregister(struct device *dev,
+					       struct dual_role_phy_instance
+					       *dual_role){}
+static inline void *dual_role_get_drvdata(struct dual_role_phy_instance
+		*dual_role)
+{
+	return ERR_PTR(-ENOSYS);
+}
+#endif /* CONFIG_DUAL_ROLE_USB_INTF */
+#endif /* __LINUX_CLASS_DUAL_ROLE_H__ */
diff --git a/include/linux/usb/composite.h b/include/linux/usb/composite.h
index 1074b8921a5d..15d7c311e86e 100644
--- a/include/linux/usb/composite.h
+++ b/include/linux/usb/composite.h
@@ -574,6 +574,7 @@ struct usb_function_instance {
 	struct config_group group;
 	struct list_head cfs_list;
 	struct usb_function_driver *fd;
+	struct usb_function *f;
 	int (*set_inst_name)(struct usb_function_instance *inst,
 			      const char *name);
 	void (*free_func_inst)(struct usb_function_instance *inst);
diff --git a/include/linux/usb/f_accessory.h b/include/linux/usb/f_accessory.h
new file mode 100644
index 000000000000..ebe3c4d59309
--- /dev/null
+++ b/include/linux/usb/f_accessory.h
@@ -0,0 +1,23 @@
+/*
+ * Gadget Function Driver for Android USB accessories
+ *
+ * Copyright (C) 2011 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __LINUX_USB_F_ACCESSORY_H
+#define __LINUX_USB_F_ACCESSORY_H
+
+#include <uapi/linux/usb/f_accessory.h>
+
+#endif /* __LINUX_USB_F_ACCESSORY_H */
diff --git a/include/linux/usb/f_mtp.h b/include/linux/usb/f_mtp.h
new file mode 100644
index 000000000000..4e8417791bea
--- /dev/null
+++ b/include/linux/usb/f_mtp.h
@@ -0,0 +1,23 @@
+/*
+ * Gadget Function Driver for MTP
+ *
+ * Copyright (C) 2010 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef __LINUX_USB_F_MTP_H
+#define __LINUX_USB_F_MTP_H
+
+#include <uapi/linux/usb/f_mtp.h>
+
+#endif /* __LINUX_USB_F_MTP_H */
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index 3e5d9075960f..73fae8c4a5fb 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -189,6 +189,7 @@ extern void __inc_zone_state(struct zone *, enum zone_stat_item);
 extern void dec_zone_state(struct zone *, enum zone_stat_item);
 extern void __dec_zone_state(struct zone *, enum zone_stat_item);
 
+void quiet_vmstat(void);
 void cpu_vm_stats_fold(int cpu);
 void refresh_zone_stat_thresholds(void);
 
@@ -249,6 +250,7 @@ static inline void __dec_zone_page_state(struct page *page,
 
 static inline void refresh_zone_stat_thresholds(void) { }
 static inline void cpu_vm_stats_fold(int cpu) { }
+static inline void quiet_vmstat(void) { }
 
 static inline void drain_zonestat(struct zone *zone,
 			struct per_cpu_pageset *pset) { }
diff --git a/include/linux/wakelock.h b/include/linux/wakelock.h
new file mode 100644
index 000000000000..f4a698a22880
--- /dev/null
+++ b/include/linux/wakelock.h
@@ -0,0 +1,67 @@
+/* include/linux/wakelock.h
+ *
+ * Copyright (C) 2007-2012 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _LINUX_WAKELOCK_H
+#define _LINUX_WAKELOCK_H
+
+#include <linux/ktime.h>
+#include <linux/device.h>
+
+/* A wake_lock prevents the system from entering suspend or other low power
+ * states when active. If the type is set to WAKE_LOCK_SUSPEND, the wake_lock
+ * prevents a full system suspend.
+ */
+
+enum {
+	WAKE_LOCK_SUSPEND, /* Prevent suspend */
+	WAKE_LOCK_TYPE_COUNT
+};
+
+struct wake_lock {
+	struct wakeup_source ws;
+};
+
+static inline void wake_lock_init(struct wake_lock *lock, int type,
+				  const char *name)
+{
+	wakeup_source_init(&lock->ws, name);
+}
+
+static inline void wake_lock_destroy(struct wake_lock *lock)
+{
+	wakeup_source_trash(&lock->ws);
+}
+
+static inline void wake_lock(struct wake_lock *lock)
+{
+	__pm_stay_awake(&lock->ws);
+}
+
+static inline void wake_lock_timeout(struct wake_lock *lock, long timeout)
+{
+	__pm_wakeup_event(&lock->ws, jiffies_to_msecs(timeout));
+}
+
+static inline void wake_unlock(struct wake_lock *lock)
+{
+	__pm_relax(&lock->ws);
+}
+
+static inline int wake_lock_active(struct wake_lock *lock)
+{
+	return lock->ws.active;
+}
+
+#endif
diff --git a/include/linux/wakeup_reason.h b/include/linux/wakeup_reason.h
new file mode 100644
index 000000000000..d84d8c301546
--- /dev/null
+++ b/include/linux/wakeup_reason.h
@@ -0,0 +1,32 @@
+/*
+ * include/linux/wakeup_reason.h
+ *
+ * Logs the reason which caused the kernel to resume
+ * from the suspend mode.
+ *
+ * Copyright (C) 2014 Google, Inc.
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _LINUX_WAKEUP_REASON_H
+#define _LINUX_WAKEUP_REASON_H
+
+#define MAX_SUSPEND_ABORT_LEN 256
+
+void log_wakeup_reason(int irq);
+int check_wakeup_reason(int irq);
+
+#ifdef CONFIG_SUSPEND
+void log_suspend_abort_reason(const char *fmt, ...);
+#else
+static inline void log_suspend_abort_reason(const char *fmt, ...) { }
+#endif
+
+#endif /* _LINUX_WAKEUP_REASON_H */
diff --git a/include/linux/wlan_plat.h b/include/linux/wlan_plat.h
new file mode 100644
index 000000000000..8e8b06f1ba4a
--- /dev/null
+++ b/include/linux/wlan_plat.h
@@ -0,0 +1,30 @@
+/* include/linux/wlan_plat.h
+ *
+ * Copyright (C) 2010 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+#ifndef _LINUX_WLAN_PLAT_H_
+#define _LINUX_WLAN_PLAT_H_
+
+#define WLAN_PLAT_NODFS_FLAG	0x01
+
+struct wifi_platform_data {
+	int (*set_power)(int val);
+	int (*set_reset)(int val);
+	int (*set_carddetect)(int val);
+	void *(*mem_prealloc)(int section, unsigned long size);
+	int (*get_mac_addr)(unsigned char *buf);
+	int (*get_wake_irq)(void);
+	void *(*get_country_code)(char *ccode, u32 flags);
+};
+
+#endif
diff --git a/include/net/addrconf.h b/include/net/addrconf.h
index 18dd7a3caf2f..9cba4907695c 100644
--- a/include/net/addrconf.h
+++ b/include/net/addrconf.h
@@ -1,8 +1,9 @@
 #ifndef _ADDRCONF_H
 #define _ADDRCONF_H
 
-#define MAX_RTR_SOLICITATIONS		3
+#define MAX_RTR_SOLICITATIONS		-1		/* unlimited */
 #define RTR_SOLICITATION_INTERVAL	(4*HZ)
+#define RTR_SOLICITATION_MAX_INTERVAL	(3600*HZ)	/* 1 hour */
 
 #define MIN_VALID_LIFETIME		(2*3600)	/* 2 hours */
 
@@ -229,6 +230,8 @@ static inline bool ipv6_is_mld(struct sk_buff *skb, int nexthdr, int offset)
 void addrconf_prefix_rcv(struct net_device *dev,
 			 u8 *opt, int len, bool sllao);
 
+u32 addrconf_rt_table(const struct net_device *dev, u32 default_table);
+
 /*
  *	anycast prototypes (anycast.c)
  */
diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h
index 59160de702b6..bd2b5c007561 100644
--- a/include/net/fib_rules.h
+++ b/include/net/fib_rules.h
@@ -8,6 +8,11 @@
 #include <net/flow.h>
 #include <net/rtnetlink.h>
 
+struct fib_kuid_range {
+	kuid_t start;
+	kuid_t end;
+};
+
 struct fib_rule {
 	struct list_head	list;
 	int			iifindex;
@@ -29,6 +34,7 @@ struct fib_rule {
 	int			suppress_prefixlen;
 	char			iifname[IFNAMSIZ];
 	char			oifname[IFNAMSIZ];
+	struct fib_kuid_range	uid_range;
 	struct rcu_head		rcu;
 };
 
@@ -89,7 +95,8 @@ struct fib_rules_ops {
 	[FRA_TABLE]     = { .type = NLA_U32 }, \
 	[FRA_SUPPRESS_PREFIXLEN] = { .type = NLA_U32 }, \
 	[FRA_SUPPRESS_IFGROUP] = { .type = NLA_U32 }, \
-	[FRA_GOTO]	= { .type = NLA_U32 }
+	[FRA_GOTO]	= { .type = NLA_U32 }, \
+	[FRA_UID_RANGE]	= { .len = sizeof(struct fib_rule_uid_range) }
 
 static inline void fib_rule_get(struct fib_rule *rule)
 {
diff --git a/include/net/flow.h b/include/net/flow.h
index 83969eebebf3..833080732dec 100644
--- a/include/net/flow.h
+++ b/include/net/flow.h
@@ -11,6 +11,7 @@
 #include <linux/in6.h>
 #include <linux/atomic.h>
 #include <net/flow_dissector.h>
+#include <linux/uidgid.h>
 
 /*
  * ifindex generation is per-net namespace, and loopback is
@@ -38,6 +39,7 @@ struct flowi_common {
 #define FLOWI_FLAG_SKIP_NH_OIF		0x08
 	__u32	flowic_secid;
 	struct flowi_tunnel flowic_tun_key;
+	kuid_t  flowic_uid;
 };
 
 union flowi_uli {
@@ -75,6 +77,7 @@ struct flowi4 {
 #define flowi4_flags		__fl_common.flowic_flags
 #define flowi4_secid		__fl_common.flowic_secid
 #define flowi4_tun_key		__fl_common.flowic_tun_key
+#define flowi4_uid		__fl_common.flowic_uid
 
 	/* (saddr,daddr) must be grouped, same order as in IP header */
 	__be32			saddr;
@@ -94,7 +97,8 @@ static inline void flowi4_init_output(struct flowi4 *fl4, int oif,
 				      __u32 mark, __u8 tos, __u8 scope,
 				      __u8 proto, __u8 flags,
 				      __be32 daddr, __be32 saddr,
-				      __be16 dport, __be16 sport)
+				      __be16 dport, __be16 sport,
+				      kuid_t uid)
 {
 	fl4->flowi4_oif = oif;
 	fl4->flowi4_iif = LOOPBACK_IFINDEX;
@@ -105,6 +109,7 @@ static inline void flowi4_init_output(struct flowi4 *fl4, int oif,
 	fl4->flowi4_flags = flags;
 	fl4->flowi4_secid = 0;
 	fl4->flowi4_tun_key.tun_id = 0;
+	fl4->flowi4_uid = uid;
 	fl4->daddr = daddr;
 	fl4->saddr = saddr;
 	fl4->fl4_dport = dport;
@@ -133,6 +138,7 @@ struct flowi6 {
 #define flowi6_flags		__fl_common.flowic_flags
 #define flowi6_secid		__fl_common.flowic_secid
 #define flowi6_tun_key		__fl_common.flowic_tun_key
+#define flowi6_uid		__fl_common.flowic_uid
 	struct in6_addr		daddr;
 	struct in6_addr		saddr;
 	__be32			flowlabel;
@@ -177,6 +183,7 @@ struct flowi {
 #define flowi_flags	u.__fl_common.flowic_flags
 #define flowi_secid	u.__fl_common.flowic_secid
 #define flowi_tun_key	u.__fl_common.flowic_tun_key
+#define flowi_uid	u.__fl_common.flowic_uid
 } __attribute__((__aligned__(BITS_PER_LONG/8)));
 
 static inline struct flowi *flowi4_to_flowi(struct flowi4 *fl4)
diff --git a/include/net/if_inet6.h b/include/net/if_inet6.h
index 1c8b6820b694..515352c6280a 100644
--- a/include/net/if_inet6.h
+++ b/include/net/if_inet6.h
@@ -201,6 +201,7 @@ struct inet6_dev {
 	struct ipv6_devstat	stats;
 
 	struct timer_list	rs_timer;
+	__s32			rs_interval;	/* in jiffies */
 	__u8			rs_probes;
 
 	__u8			addr_gen_mode;
diff --git a/include/net/ip.h b/include/net/ip.h
index 7476bb10ff37..c5d8ee796b38 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h
@@ -170,6 +170,7 @@ struct ip_reply_arg {
 				/* -1 if not needed */ 
 	int	    bound_dev_if;
 	u8  	    tos;
+	kuid_t	    uid;
 }; 
 
 #define IP_REPLY_ARG_NOSRCCHECK 1
diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h
index b8a8d4239e85..f9bdfb096579 100644
--- a/include/net/ip6_route.h
+++ b/include/net/ip6_route.h
@@ -118,9 +118,10 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 		  const struct in6_addr *gwaddr);
 
 void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu, int oif,
-		     u32 mark);
+		     u32 mark, kuid_t uid);
 void ip6_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, __be32 mtu);
-void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark);
+void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark,
+		  kuid_t uid);
 void ip6_redirect_no_header(struct sk_buff *skb, struct net *net, int oif,
 			    u32 mark);
 void ip6_sk_redirect(struct sk_buff *skb, struct sock *sk);
diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h
index 5689a0c749f7..5567d46b3cff 100644
--- a/include/net/l3mdev.h
+++ b/include/net/l3mdev.h
@@ -51,6 +51,24 @@ static inline int l3mdev_master_ifindex(struct net_device *dev)
 	return ifindex;
 }
 
+static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex)
+{
+	struct net_device *dev;
+	int rc = 0;
+
+	if (likely(ifindex)) {
+		rcu_read_lock();
+
+		dev = dev_get_by_index_rcu(net, ifindex);
+		if (dev)
+			rc = l3mdev_master_ifindex_rcu(dev);
+
+		rcu_read_unlock();
+	}
+
+	return rc;
+}
+
 /* get index of an interface to use for FIB lookups. For devices
  * enslaved to an L3 master device FIB lookups are based on the
  * master index
@@ -170,6 +188,11 @@ static inline int l3mdev_master_ifindex(struct net_device *dev)
 	return 0;
 }
 
+static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex)
+{
+	return 0;
+}
+
 static inline int l3mdev_fib_oif_rcu(struct net_device *dev)
 {
 	return dev ? dev->ifindex : 0;
diff --git a/include/net/route.h b/include/net/route.h
index a3b9ef74a389..3adb9c724818 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -154,7 +154,7 @@ static inline struct rtable *ip_route_output_ports(struct net *net, struct flowi
 	flowi4_init_output(fl4, oif, sk ? sk->sk_mark : 0, tos,
 			   RT_SCOPE_UNIVERSE, proto,
 			   sk ? inet_sk_flowi_flags(sk) : 0,
-			   daddr, saddr, dport, sport);
+			   daddr, saddr, dport, sport, sock_net_uid(net, sk));
 	if (sk)
 		security_sk_classify_flow(sk, flowi4_to_flowi(fl4));
 	return ip_route_output_flow(net, fl4, sk);
@@ -267,7 +267,8 @@ static inline void ip_route_connect_init(struct flowi4 *fl4, __be32 dst, __be32
 		flow_flags |= FLOWI_FLAG_ANYSRC;
 
 	flowi4_init_output(fl4, oif, sk->sk_mark, tos, RT_SCOPE_UNIVERSE,
-			   protocol, flow_flags, dst, src, dport, sport);
+			   protocol, flow_flags, dst, src, dport, sport,
+			   sk->sk_uid);
 }
 
 static inline struct rtable *ip_route_connect(struct flowi4 *fl4,
diff --git a/include/net/sock.h b/include/net/sock.h
index 3d5ff7436f41..62c9e2268d63 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -446,6 +446,7 @@ struct sock {
 	void			*sk_security;
 #endif
 	__u32			sk_mark;
+	kuid_t			sk_uid;
 #ifdef CONFIG_CGROUP_NET_CLASSID
 	u32			sk_classid;
 #endif
@@ -1067,6 +1068,7 @@ struct proto {
 	void			(*destroy_cgroup)(struct mem_cgroup *memcg);
 	struct cg_proto		*(*proto_cgroup)(struct mem_cgroup *memcg);
 #endif
+	int			(*diag_destroy)(struct sock *sk, int err);
 };
 
 int proto_register(struct proto *prot, int alloc_slab);
@@ -1691,6 +1693,7 @@ static inline void sock_graft(struct sock *sk, struct socket *parent)
 	sk->sk_wq = parent->wq;
 	parent->sk = sk;
 	sk_set_socket(sk, parent);
+	sk->sk_uid = SOCK_INODE(parent)->i_uid;
 	security_sock_graft(sk, parent);
 	write_unlock_bh(&sk->sk_callback_lock);
 }
@@ -1698,6 +1701,11 @@ static inline void sock_graft(struct sock *sk, struct socket *parent)
 kuid_t sock_i_uid(struct sock *sk);
 unsigned long sock_i_ino(struct sock *sk);
 
+static inline kuid_t sock_net_uid(const struct net *net, const struct sock *sk)
+{
+	return sk ? sk->sk_uid : make_kuid(net->user_ns, 0);
+}
+
 static inline u32 net_tx_rndhash(void)
 {
 	u32 v = prandom_u32();
diff --git a/include/net/tcp.h b/include/net/tcp.h
index cecb0e0eff06..593253f2427e 100644
--- a/include/net/tcp.h
+++ b/include/net/tcp.h
@@ -284,6 +284,7 @@ extern int sysctl_tcp_autocorking;
 extern int sysctl_tcp_invalid_ratelimit;
 extern int sysctl_tcp_pacing_ss_ratio;
 extern int sysctl_tcp_pacing_ca_ratio;
+extern int sysctl_tcp_default_init_rwnd;
 
 extern atomic_long_t tcp_memory_allocated;
 extern struct percpu_counter tcp_sockets_allocated;
@@ -1171,6 +1172,8 @@ void tcp_set_state(struct sock *sk, int state);
 
 void tcp_done(struct sock *sk);
 
+int tcp_abort(struct sock *sk, int err);
+
 static inline void tcp_sack_reset(struct tcp_options_received *rx_opt)
 {
 	rx_opt->dsack = 0;
diff --git a/include/net/udp.h b/include/net/udp.h
index 6d4ed18e1427..e57f50258cda 100644
--- a/include/net/udp.h
+++ b/include/net/udp.h
@@ -238,6 +238,7 @@ int udp_get_port(struct sock *sk, unsigned short snum,
 		 int (*saddr_cmp)(const struct sock *,
 				  const struct sock *));
 void udp_err(struct sk_buff *, u32);
+int udp_abort(struct sock *sk, int err);
 int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len);
 int udp_push_pending_frames(struct sock *sk);
 void udp_flush_pending_frames(struct sock *sk);
diff --git a/include/net/xfrm.h b/include/net/xfrm.h
index 185fb037b332..9e1325e36415 100644
--- a/include/net/xfrm.h
+++ b/include/net/xfrm.h
@@ -159,6 +159,7 @@ struct xfrm_state {
 		int		header_len;
 		int		trailer_len;
 		u32		extra_flags;
+		u32		output_mark;
 	} props;
 
 	struct xfrm_lifetime_cfg lft;
@@ -288,10 +289,12 @@ struct xfrm_policy_afinfo {
 	struct dst_entry	*(*dst_lookup)(struct net *net,
 					       int tos, int oif,
 					       const xfrm_address_t *saddr,
-					       const xfrm_address_t *daddr);
+					       const xfrm_address_t *daddr,
+					       u32 mark);
 	int			(*get_saddr)(struct net *net, int oif,
 					     xfrm_address_t *saddr,
-					     xfrm_address_t *daddr);
+					     xfrm_address_t *daddr,
+					     u32 mark);
 	void			(*decode_session)(struct sk_buff *skb,
 						  struct flowi *fl,
 						  int reverse);
diff --git a/include/trace/events/android_fs.h b/include/trace/events/android_fs.h
new file mode 100644
index 000000000000..49509533d3fa
--- /dev/null
+++ b/include/trace/events/android_fs.h
@@ -0,0 +1,65 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM android_fs
+
+#if !defined(_TRACE_ANDROID_FS_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_ANDROID_FS_H
+
+#include <linux/tracepoint.h>
+#include <trace/events/android_fs_template.h>
+
+DEFINE_EVENT(android_fs_data_start_template, android_fs_dataread_start,
+	TP_PROTO(struct inode *inode, loff_t offset, int bytes,
+		 pid_t pid, char *pathname, char *command),
+	TP_ARGS(inode, offset, bytes, pid, pathname, command));
+
+DEFINE_EVENT(android_fs_data_end_template, android_fs_dataread_end,
+	TP_PROTO(struct inode *inode, loff_t offset, int bytes),
+	TP_ARGS(inode, offset, bytes));
+
+DEFINE_EVENT(android_fs_data_start_template, android_fs_datawrite_start,
+	TP_PROTO(struct inode *inode, loff_t offset, int bytes,
+		 pid_t pid, char *pathname, char *command),
+	TP_ARGS(inode, offset, bytes, pid, pathname, command));
+
+DEFINE_EVENT(android_fs_data_end_template, android_fs_datawrite_end,
+	TP_PROTO(struct inode *inode, loff_t offset, int bytes),
+	     TP_ARGS(inode, offset, bytes));
+
+#endif /* _TRACE_ANDROID_FS_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
+
+#ifndef ANDROID_FSTRACE_GET_PATHNAME
+#define ANDROID_FSTRACE_GET_PATHNAME
+
+/* Sizes an on-stack array, so careful if sizing this up ! */
+#define MAX_TRACE_PATHBUF_LEN	256
+
+static inline char *
+android_fstrace_get_pathname(char *buf, int buflen, struct inode *inode)
+{
+	char *path;
+	struct dentry *d;
+
+	/*
+	 * d_obtain_alias() will either iput() if it locates an existing
+	 * dentry or transfer the reference to the new dentry created.
+	 * So get an extra reference here.
+	 */
+	ihold(inode);
+	d = d_obtain_alias(inode);
+	if (likely(!IS_ERR(d))) {
+		path = dentry_path_raw(d, buf, buflen);
+		if (unlikely(IS_ERR(path))) {
+			strcpy(buf, "ERROR");
+			path = buf;
+		}
+		dput(d);
+	} else {
+		strcpy(buf, "ERROR");
+		path = buf;
+	}
+	return path;
+}
+#endif
diff --git a/include/trace/events/android_fs_template.h b/include/trace/events/android_fs_template.h
new file mode 100644
index 000000000000..b23d17b56c63
--- /dev/null
+++ b/include/trace/events/android_fs_template.h
@@ -0,0 +1,64 @@
+#if !defined(_TRACE_ANDROID_FS_TEMPLATE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_ANDROID_FS_TEMPLATE_H
+
+#include <linux/tracepoint.h>
+
+DECLARE_EVENT_CLASS(android_fs_data_start_template,
+	TP_PROTO(struct inode *inode, loff_t offset, int bytes,
+		 pid_t pid, char *pathname, char *command),
+	TP_ARGS(inode, offset, bytes, pid, pathname, command),
+	TP_STRUCT__entry(
+		__string(pathbuf, pathname);
+		__field(loff_t,	offset);
+		__field(int,	bytes);
+		__field(loff_t,	i_size);
+		__string(cmdline, command);
+		__field(pid_t,	pid);
+		__field(ino_t,	ino);
+	),
+	TP_fast_assign(
+		{
+			/*
+			 * Replace the spaces in filenames and cmdlines
+			 * because this screws up the tooling that parses
+			 * the traces.
+			 */
+			__assign_str(pathbuf, pathname);
+			(void)strreplace(__get_str(pathbuf), ' ', '_');
+			__entry->offset		= offset;
+			__entry->bytes		= bytes;
+			__entry->i_size		= i_size_read(inode);
+			__assign_str(cmdline, command);
+			(void)strreplace(__get_str(cmdline), ' ', '_');
+			__entry->pid		= pid;
+			__entry->ino		= inode->i_ino;
+		}
+	),
+	TP_printk("entry_name %s, offset %llu, bytes %d, cmdline %s,"
+		  " pid %d, i_size %llu, ino %lu",
+		  __get_str(pathbuf), __entry->offset, __entry->bytes,
+		  __get_str(cmdline), __entry->pid, __entry->i_size,
+		  (unsigned long) __entry->ino)
+);
+
+DECLARE_EVENT_CLASS(android_fs_data_end_template,
+	TP_PROTO(struct inode *inode, loff_t offset, int bytes),
+	TP_ARGS(inode, offset, bytes),
+	TP_STRUCT__entry(
+		__field(ino_t,	ino);
+		__field(loff_t,	offset);
+		__field(int,	bytes);
+	),
+	TP_fast_assign(
+		{
+			__entry->ino		= inode->i_ino;
+			__entry->offset		= offset;
+			__entry->bytes		= bytes;
+		}
+	),
+	TP_printk("ino %lu, offset %llu, bytes %d",
+		  (unsigned long) __entry->ino,
+		  __entry->offset, __entry->bytes)
+);
+
+#endif /* _TRACE_ANDROID_FS_TEMPLATE_H */
diff --git a/include/trace/events/cpufreq_interactive.h b/include/trace/events/cpufreq_interactive.h
new file mode 100644
index 000000000000..951e6ca12da8
--- /dev/null
+++ b/include/trace/events/cpufreq_interactive.h
@@ -0,0 +1,112 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM cpufreq_interactive
+
+#if !defined(_TRACE_CPUFREQ_INTERACTIVE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_CPUFREQ_INTERACTIVE_H
+
+#include <linux/tracepoint.h>
+
+DECLARE_EVENT_CLASS(set,
+	TP_PROTO(u32 cpu_id, unsigned long targfreq,
+	         unsigned long actualfreq),
+	TP_ARGS(cpu_id, targfreq, actualfreq),
+
+	TP_STRUCT__entry(
+	    __field(          u32, cpu_id    )
+	    __field(unsigned long, targfreq   )
+	    __field(unsigned long, actualfreq )
+	   ),
+
+	TP_fast_assign(
+	    __entry->cpu_id = (u32) cpu_id;
+	    __entry->targfreq = targfreq;
+	    __entry->actualfreq = actualfreq;
+	),
+
+	TP_printk("cpu=%u targ=%lu actual=%lu",
+	      __entry->cpu_id, __entry->targfreq,
+	      __entry->actualfreq)
+);
+
+DEFINE_EVENT(set, cpufreq_interactive_setspeed,
+	TP_PROTO(u32 cpu_id, unsigned long targfreq,
+	     unsigned long actualfreq),
+	TP_ARGS(cpu_id, targfreq, actualfreq)
+);
+
+DECLARE_EVENT_CLASS(loadeval,
+	    TP_PROTO(unsigned long cpu_id, unsigned long load,
+		     unsigned long curtarg, unsigned long curactual,
+		     unsigned long newtarg),
+		    TP_ARGS(cpu_id, load, curtarg, curactual, newtarg),
+
+	    TP_STRUCT__entry(
+		    __field(unsigned long, cpu_id    )
+		    __field(unsigned long, load      )
+		    __field(unsigned long, curtarg   )
+		    __field(unsigned long, curactual )
+		    __field(unsigned long, newtarg   )
+	    ),
+
+	    TP_fast_assign(
+		    __entry->cpu_id = cpu_id;
+		    __entry->load = load;
+		    __entry->curtarg = curtarg;
+		    __entry->curactual = curactual;
+		    __entry->newtarg = newtarg;
+	    ),
+
+	    TP_printk("cpu=%lu load=%lu cur=%lu actual=%lu targ=%lu",
+		      __entry->cpu_id, __entry->load, __entry->curtarg,
+		      __entry->curactual, __entry->newtarg)
+);
+
+DEFINE_EVENT(loadeval, cpufreq_interactive_target,
+	    TP_PROTO(unsigned long cpu_id, unsigned long load,
+		     unsigned long curtarg, unsigned long curactual,
+		     unsigned long newtarg),
+	    TP_ARGS(cpu_id, load, curtarg, curactual, newtarg)
+);
+
+DEFINE_EVENT(loadeval, cpufreq_interactive_already,
+	    TP_PROTO(unsigned long cpu_id, unsigned long load,
+		     unsigned long curtarg, unsigned long curactual,
+		     unsigned long newtarg),
+	    TP_ARGS(cpu_id, load, curtarg, curactual, newtarg)
+);
+
+DEFINE_EVENT(loadeval, cpufreq_interactive_notyet,
+	    TP_PROTO(unsigned long cpu_id, unsigned long load,
+		     unsigned long curtarg, unsigned long curactual,
+		     unsigned long newtarg),
+	    TP_ARGS(cpu_id, load, curtarg, curactual, newtarg)
+);
+
+TRACE_EVENT(cpufreq_interactive_boost,
+	    TP_PROTO(const char *s),
+	    TP_ARGS(s),
+	    TP_STRUCT__entry(
+		    __string(s, s)
+	    ),
+	    TP_fast_assign(
+		    __assign_str(s, s);
+	    ),
+	    TP_printk("%s", __get_str(s))
+);
+
+TRACE_EVENT(cpufreq_interactive_unboost,
+	    TP_PROTO(const char *s),
+	    TP_ARGS(s),
+	    TP_STRUCT__entry(
+		    __string(s, s)
+	    ),
+	    TP_fast_assign(
+		    __assign_str(s, s);
+	    ),
+	    TP_printk("%s", __get_str(s))
+);
+
+#endif /* _TRACE_CPUFREQ_INTERACTIVE_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/cpufreq_sched.h b/include/trace/events/cpufreq_sched.h
new file mode 100644
index 000000000000..a46cd088e969
--- /dev/null
+++ b/include/trace/events/cpufreq_sched.h
@@ -0,0 +1,87 @@
+/*
+ *  Copyright (C)  2015 Steve Muckle <smuckle@linaro.org>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM cpufreq_sched
+
+#if !defined(_TRACE_CPUFREQ_SCHED_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_CPUFREQ_SCHED_H
+
+#include <linux/sched.h>
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(cpufreq_sched_throttled,
+	    TP_PROTO(unsigned int rem),
+	    TP_ARGS(rem),
+	    TP_STRUCT__entry(
+		    __field(	unsigned int,	rem)
+	    ),
+	    TP_fast_assign(
+		    __entry->rem = rem;
+	    ),
+	    TP_printk("throttled - %d usec remaining", __entry->rem)
+);
+
+TRACE_EVENT(cpufreq_sched_request_opp,
+	    TP_PROTO(int cpu,
+		     unsigned long capacity,
+		     unsigned int freq_new,
+		     unsigned int requested_freq),
+	    TP_ARGS(cpu, capacity, freq_new, requested_freq),
+	    TP_STRUCT__entry(
+		    __field(	int,		cpu)
+		    __field(	unsigned long,	capacity)
+		    __field(	unsigned int,	freq_new)
+		    __field(	unsigned int,	requested_freq)
+		    ),
+	    TP_fast_assign(
+		    __entry->cpu = cpu;
+		    __entry->capacity = capacity;
+		    __entry->freq_new = freq_new;
+		    __entry->requested_freq = requested_freq;
+		    ),
+	    TP_printk("cpu %d cap change, cluster cap request %ld => OPP %d "
+		      "(cur %d)",
+		      __entry->cpu, __entry->capacity, __entry->freq_new,
+		      __entry->requested_freq)
+);
+
+TRACE_EVENT(cpufreq_sched_update_capacity,
+	    TP_PROTO(int cpu,
+		     bool request,
+		     struct sched_capacity_reqs *scr,
+		     unsigned long new_capacity),
+	    TP_ARGS(cpu, request, scr, new_capacity),
+	    TP_STRUCT__entry(
+		    __field(	int,		cpu)
+		    __field(	bool,		request)
+		    __field(	unsigned long,	cfs)
+		    __field(	unsigned long,	rt)
+		    __field(	unsigned long,	dl)
+		    __field(	unsigned long,	total)
+		    __field(	unsigned long,	new_total)
+	    ),
+	    TP_fast_assign(
+		    __entry->cpu = cpu;
+		    __entry->request = request;
+		    __entry->cfs = scr->cfs;
+		    __entry->rt = scr->rt;
+		    __entry->dl = scr->dl;
+		    __entry->total = scr->total;
+		    __entry->new_total = new_capacity;
+	    ),
+	    TP_printk("cpu=%d set_cap=%d cfs=%ld rt=%ld dl=%ld old_tot=%ld "
+		      "new_tot=%ld",
+		      __entry->cpu, __entry->request, __entry->cfs, __entry->rt,
+		      __entry->dl, __entry->total, __entry->new_total)
+);
+
+#endif /* _TRACE_CPUFREQ_SCHED_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/f2fs.h b/include/trace/events/f2fs.h
index 00b4a6308249..7063bbcca03b 100644
--- a/include/trace/events/f2fs.h
+++ b/include/trace/events/f2fs.h
@@ -6,8 +6,8 @@
 
 #include <linux/tracepoint.h>
 
-#define show_dev(entry)		MAJOR(entry->dev), MINOR(entry->dev)
-#define show_dev_ino(entry)	show_dev(entry), (unsigned long)entry->ino
+#define show_dev(dev)		MAJOR(dev), MINOR(dev)
+#define show_dev_ino(entry)	show_dev(entry->dev), (unsigned long)entry->ino
 
 TRACE_DEFINE_ENUM(NODE);
 TRACE_DEFINE_ENUM(DATA);
@@ -15,6 +15,7 @@ TRACE_DEFINE_ENUM(META);
 TRACE_DEFINE_ENUM(META_FLUSH);
 TRACE_DEFINE_ENUM(INMEM);
 TRACE_DEFINE_ENUM(INMEM_DROP);
+TRACE_DEFINE_ENUM(INMEM_INVALIDATE);
 TRACE_DEFINE_ENUM(IPU);
 TRACE_DEFINE_ENUM(OPU);
 TRACE_DEFINE_ENUM(CURSEG_HOT_DATA);
@@ -43,6 +44,7 @@ TRACE_DEFINE_ENUM(CP_FASTBOOT);
 TRACE_DEFINE_ENUM(CP_SYNC);
 TRACE_DEFINE_ENUM(CP_RECOVERY);
 TRACE_DEFINE_ENUM(CP_DISCARD);
+TRACE_DEFINE_ENUM(CP_TRIMMED);
 
 #define show_block_type(type)						\
 	__print_symbolic(type,						\
@@ -52,13 +54,16 @@ TRACE_DEFINE_ENUM(CP_DISCARD);
 		{ META_FLUSH,	"META_FLUSH" },				\
 		{ INMEM,	"INMEM" },				\
 		{ INMEM_DROP,	"INMEM_DROP" },				\
+		{ INMEM_INVALIDATE,	"INMEM_INVALIDATE" },		\
+		{ INMEM_REVOKE,	"INMEM_REVOKE" },			\
 		{ IPU,		"IN-PLACE" },				\
 		{ OPU,		"OUT-OF-PLACE" })
 
 #define F2FS_BIO_MASK(t)	(t & (READA | WRITE_FLUSH_FUA))
 #define F2FS_BIO_EXTRA_MASK(t)	(t & (REQ_META | REQ_PRIO))
 
-#define show_bio_type(type)	show_bio_base(type), show_bio_extra(type)
+#define show_bio_type(op, op_flags)					\
+		show_bio_base((op|op_flags)), show_bio_extra((op|op_flags))
 
 #define show_bio_base(type)						\
 	__print_symbolic(F2FS_BIO_MASK(type),				\
@@ -78,6 +83,12 @@ TRACE_DEFINE_ENUM(CP_DISCARD);
 		{ REQ_META | REQ_PRIO,	"(MP)" },			\
 		{ 0, " \b" })
 
+#define show_block_temp(temp)						\
+	__print_symbolic(temp,						\
+		{ HOT,		"HOT" },				\
+		{ WARM,		"WARM" },				\
+		{ COLD,		"COLD" })
+
 #define show_data_type(type)						\
 	__print_symbolic(type,						\
 		{ CURSEG_HOT_DATA, 	"Hot DATA" },			\
@@ -114,7 +125,8 @@ TRACE_DEFINE_ENUM(CP_DISCARD);
 		{ CP_FASTBOOT,	"Fastboot" },				\
 		{ CP_SYNC,	"Sync" },				\
 		{ CP_RECOVERY,	"Recovery" },				\
-		{ CP_DISCARD,	"Discard" })
+		{ CP_DISCARD,	"Discard" },				\
+		{ CP_UMOUNT | CP_TRIMMED,	"Umount,Trimmed" })
 
 struct victim_sel_policy;
 struct f2fs_map_blocks;
@@ -237,7 +249,7 @@ TRACE_EVENT(f2fs_sync_fs,
 	),
 
 	TP_printk("dev = (%d,%d), superblock is %s, wait = %d",
-		show_dev(__entry),
+		show_dev(__entry->dev),
 		__entry->dirty ? "dirty" : "not dirty",
 		__entry->wait)
 );
@@ -307,6 +319,13 @@ DEFINE_EVENT(f2fs__inode_exit, f2fs_unlink_exit,
 	TP_ARGS(inode, ret)
 );
 
+DEFINE_EVENT(f2fs__inode_exit, f2fs_drop_inode,
+
+	TP_PROTO(struct inode *inode, int ret),
+
+	TP_ARGS(inode, ret)
+);
+
 DEFINE_EVENT(f2fs__inode, f2fs_truncate,
 
 	TP_PROTO(struct inode *inode),
@@ -516,14 +535,14 @@ TRACE_EVENT(f2fs_map_blocks,
 
 TRACE_EVENT(f2fs_background_gc,
 
-	TP_PROTO(struct super_block *sb, long wait_ms,
+	TP_PROTO(struct super_block *sb, unsigned int wait_ms,
 			unsigned int prefree, unsigned int free),
 
 	TP_ARGS(sb, wait_ms, prefree, free),
 
 	TP_STRUCT__entry(
 		__field(dev_t,	dev)
-		__field(long,	wait_ms)
+		__field(unsigned int,	wait_ms)
 		__field(unsigned int,	prefree)
 		__field(unsigned int,	free)
 	),
@@ -535,13 +554,120 @@ TRACE_EVENT(f2fs_background_gc,
 		__entry->free		= free;
 	),
 
-	TP_printk("dev = (%d,%d), wait_ms = %ld, prefree = %u, free = %u",
-		show_dev(__entry),
+	TP_printk("dev = (%d,%d), wait_ms = %u, prefree = %u, free = %u",
+		show_dev(__entry->dev),
 		__entry->wait_ms,
 		__entry->prefree,
 		__entry->free)
 );
 
+TRACE_EVENT(f2fs_gc_begin,
+
+	TP_PROTO(struct super_block *sb, bool sync, bool background,
+			long long dirty_nodes, long long dirty_dents,
+			long long dirty_imeta, unsigned int free_sec,
+			unsigned int free_seg, int reserved_seg,
+			unsigned int prefree_seg),
+
+	TP_ARGS(sb, sync, background, dirty_nodes, dirty_dents, dirty_imeta,
+		free_sec, free_seg, reserved_seg, prefree_seg),
+
+	TP_STRUCT__entry(
+		__field(dev_t,		dev)
+		__field(bool,		sync)
+		__field(bool,		background)
+		__field(long long,	dirty_nodes)
+		__field(long long,	dirty_dents)
+		__field(long long,	dirty_imeta)
+		__field(unsigned int,	free_sec)
+		__field(unsigned int,	free_seg)
+		__field(int,		reserved_seg)
+		__field(unsigned int,	prefree_seg)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= sb->s_dev;
+		__entry->sync		= sync;
+		__entry->background	= background;
+		__entry->dirty_nodes	= dirty_nodes;
+		__entry->dirty_dents	= dirty_dents;
+		__entry->dirty_imeta	= dirty_imeta;
+		__entry->free_sec	= free_sec;
+		__entry->free_seg	= free_seg;
+		__entry->reserved_seg	= reserved_seg;
+		__entry->prefree_seg	= prefree_seg;
+	),
+
+	TP_printk("dev = (%d,%d), sync = %d, background = %d, nodes = %lld, "
+		"dents = %lld, imeta = %lld, free_sec:%u, free_seg:%u, "
+		"rsv_seg:%d, prefree_seg:%u",
+		show_dev(__entry->dev),
+		__entry->sync,
+		__entry->background,
+		__entry->dirty_nodes,
+		__entry->dirty_dents,
+		__entry->dirty_imeta,
+		__entry->free_sec,
+		__entry->free_seg,
+		__entry->reserved_seg,
+		__entry->prefree_seg)
+);
+
+TRACE_EVENT(f2fs_gc_end,
+
+	TP_PROTO(struct super_block *sb, int ret, int seg_freed,
+			int sec_freed, long long dirty_nodes,
+			long long dirty_dents, long long dirty_imeta,
+			unsigned int free_sec, unsigned int free_seg,
+			int reserved_seg, unsigned int prefree_seg),
+
+	TP_ARGS(sb, ret, seg_freed, sec_freed, dirty_nodes, dirty_dents,
+		dirty_imeta, free_sec, free_seg, reserved_seg, prefree_seg),
+
+	TP_STRUCT__entry(
+		__field(dev_t,		dev)
+		__field(int,		ret)
+		__field(int,		seg_freed)
+		__field(int,		sec_freed)
+		__field(long long,	dirty_nodes)
+		__field(long long,	dirty_dents)
+		__field(long long,	dirty_imeta)
+		__field(unsigned int,	free_sec)
+		__field(unsigned int,	free_seg)
+		__field(int,		reserved_seg)
+		__field(unsigned int,	prefree_seg)
+	),
+
+	TP_fast_assign(
+		__entry->dev		= sb->s_dev;
+		__entry->ret		= ret;
+		__entry->seg_freed	= seg_freed;
+		__entry->sec_freed	= sec_freed;
+		__entry->dirty_nodes	= dirty_nodes;
+		__entry->dirty_dents	= dirty_dents;
+		__entry->dirty_imeta	= dirty_imeta;
+		__entry->free_sec	= free_sec;
+		__entry->free_seg	= free_seg;
+		__entry->reserved_seg	= reserved_seg;
+		__entry->prefree_seg	= prefree_seg;
+	),
+
+	TP_printk("dev = (%d,%d), ret = %d, seg_freed = %d, sec_freed = %d, "
+		"nodes = %lld, dents = %lld, imeta = %lld, free_sec:%u, "
+		"free_seg:%u, rsv_seg:%d, prefree_seg:%u",
+		show_dev(__entry->dev),
+		__entry->ret,
+		__entry->seg_freed,
+		__entry->sec_freed,
+		__entry->dirty_nodes,
+		__entry->dirty_dents,
+		__entry->dirty_imeta,
+		__entry->free_sec,
+		__entry->free_seg,
+		__entry->reserved_seg,
+		__entry->prefree_seg)
+);
+
 TRACE_EVENT(f2fs_get_victim,
 
 	TP_PROTO(struct super_block *sb, int type, int gc_type,
@@ -578,7 +704,7 @@ TRACE_EVENT(f2fs_get_victim,
 
 	TP_printk("dev = (%d,%d), type = %s, policy = (%s, %s, %s), victim = %u "
 		"ofs_unit = %u, pre_victim_secno = %d, prefree = %u, free = %u",
-		show_dev(__entry),
+		show_dev(__entry->dev),
 		show_data_type(__entry->type),
 		show_gc_type(__entry->gc_type),
 		show_alloc_mode(__entry->alloc_mode),
@@ -693,28 +819,32 @@ TRACE_EVENT(f2fs_direct_IO_exit,
 		__entry->ret)
 );
 
-TRACE_EVENT(f2fs_reserve_new_block,
+TRACE_EVENT(f2fs_reserve_new_blocks,
 
-	TP_PROTO(struct inode *inode, nid_t nid, unsigned int ofs_in_node),
+	TP_PROTO(struct inode *inode, nid_t nid, unsigned int ofs_in_node,
+							blkcnt_t count),
 
-	TP_ARGS(inode, nid, ofs_in_node),
+	TP_ARGS(inode, nid, ofs_in_node, count),
 
 	TP_STRUCT__entry(
 		__field(dev_t,	dev)
 		__field(nid_t, nid)
 		__field(unsigned int, ofs_in_node)
+		__field(blkcnt_t, count)
 	),
 
 	TP_fast_assign(
 		__entry->dev	= inode->i_sb->s_dev;
 		__entry->nid	= nid;
 		__entry->ofs_in_node = ofs_in_node;
+		__entry->count = count;
 	),
 
-	TP_printk("dev = (%d,%d), nid = %u, ofs_in_node = %u",
-		show_dev(__entry),
+	TP_printk("dev = (%d,%d), nid = %u, ofs_in_node = %u, count = %llu",
+		show_dev(__entry->dev),
 		(unsigned int)__entry->nid,
-		__entry->ofs_in_node)
+		__entry->ofs_in_node,
+		(unsigned long long)__entry->count)
 );
 
 DECLARE_EVENT_CLASS(f2fs__submit_page_bio,
@@ -727,8 +857,11 @@ DECLARE_EVENT_CLASS(f2fs__submit_page_bio,
 		__field(dev_t, dev)
 		__field(ino_t, ino)
 		__field(pgoff_t, index)
-		__field(block_t, blkaddr)
-		__field(int, rw)
+		__field(block_t, old_blkaddr)
+		__field(block_t, new_blkaddr)
+		__field(int, op)
+		__field(int, op_flags)
+		__field(int, temp)
 		__field(int, type)
 	),
 
@@ -736,17 +869,22 @@ DECLARE_EVENT_CLASS(f2fs__submit_page_bio,
 		__entry->dev		= page->mapping->host->i_sb->s_dev;
 		__entry->ino		= page->mapping->host->i_ino;
 		__entry->index		= page->index;
-		__entry->blkaddr	= fio->blk_addr;
-		__entry->rw		= fio->rw;
+		__entry->old_blkaddr	= fio->old_blkaddr;
+		__entry->new_blkaddr	= fio->new_blkaddr;
+		__entry->op		= fio->op;
+		__entry->op_flags	= fio->op_flags;
+		__entry->temp		= fio->temp;
 		__entry->type		= fio->type;
 	),
 
 	TP_printk("dev = (%d,%d), ino = %lu, page_index = 0x%lx, "
-		"blkaddr = 0x%llx, rw = %s%s, type = %s",
+		"oldaddr = 0x%llx, newaddr = 0x%llx, rw = %s(%s), type = %s_%s",
 		show_dev_ino(__entry),
 		(unsigned long)__entry->index,
-		(unsigned long long)__entry->blkaddr,
-		show_bio_type(__entry->rw),
+		(unsigned long long)__entry->old_blkaddr,
+		(unsigned long long)__entry->new_blkaddr,
+		show_bio_type(__entry->op, __entry->op_flags),
+		show_block_temp(__entry->temp),
 		show_block_type(__entry->type))
 );
 
@@ -759,7 +897,7 @@ DEFINE_EVENT_CONDITION(f2fs__submit_page_bio, f2fs_submit_page_bio,
 	TP_CONDITION(page->mapping)
 );
 
-DEFINE_EVENT_CONDITION(f2fs__submit_page_bio, f2fs_submit_page_mbio,
+DEFINE_EVENT_CONDITION(f2fs__submit_page_bio, f2fs_submit_page_write,
 
 	TP_PROTO(struct page *page, struct f2fs_io_info *fio),
 
@@ -768,16 +906,17 @@ DEFINE_EVENT_CONDITION(f2fs__submit_page_bio, f2fs_submit_page_mbio,
 	TP_CONDITION(page->mapping)
 );
 
-DECLARE_EVENT_CLASS(f2fs__submit_bio,
+DECLARE_EVENT_CLASS(f2fs__bio,
 
-	TP_PROTO(struct super_block *sb, struct f2fs_io_info *fio,
-						struct bio *bio),
+	TP_PROTO(struct super_block *sb, int type, struct bio *bio),
 
-	TP_ARGS(sb, fio, bio),
+	TP_ARGS(sb, type, bio),
 
 	TP_STRUCT__entry(
 		__field(dev_t,	dev)
-		__field(int,	rw)
+		__field(dev_t,	target)
+		__field(int,	op)
+		__field(int,	op_flags)
 		__field(int,	type)
 		__field(sector_t,	sector)
 		__field(unsigned int,	size)
@@ -785,36 +924,55 @@ DECLARE_EVENT_CLASS(f2fs__submit_bio,
 
 	TP_fast_assign(
 		__entry->dev		= sb->s_dev;
-		__entry->rw		= fio->rw;
-		__entry->type		= fio->type;
+		__entry->target		= bio->bi_bdev->bd_dev;
+		__entry->op		= bio_op(bio);
+		__entry->op_flags	= bio->bi_rw;
+		__entry->type		= type;
 		__entry->sector		= bio->bi_iter.bi_sector;
 		__entry->size		= bio->bi_iter.bi_size;
 	),
 
-	TP_printk("dev = (%d,%d), %s%s, %s, sector = %lld, size = %u",
-		show_dev(__entry),
-		show_bio_type(__entry->rw),
+	TP_printk("dev = (%d,%d)/(%d,%d), rw = %s%s, %s, sector = %lld, size = %u",
+		show_dev(__entry->target),
+		show_dev(__entry->dev),
+		show_bio_type(__entry->op, __entry->op_flags),
 		show_block_type(__entry->type),
 		(unsigned long long)__entry->sector,
 		__entry->size)
 );
 
-DEFINE_EVENT_CONDITION(f2fs__submit_bio, f2fs_submit_write_bio,
+DEFINE_EVENT_CONDITION(f2fs__bio, f2fs_prepare_write_bio,
+
+	TP_PROTO(struct super_block *sb, int type, struct bio *bio),
+
+	TP_ARGS(sb, type, bio),
+
+	TP_CONDITION(bio)
+);
+
+DEFINE_EVENT_CONDITION(f2fs__bio, f2fs_prepare_read_bio,
+
+	TP_PROTO(struct super_block *sb, int type, struct bio *bio),
+
+	TP_ARGS(sb, type, bio),
+
+	TP_CONDITION(bio)
+);
+
+DEFINE_EVENT_CONDITION(f2fs__bio, f2fs_submit_read_bio,
 
-	TP_PROTO(struct super_block *sb, struct f2fs_io_info *fio,
-							struct bio *bio),
+	TP_PROTO(struct super_block *sb, int type, struct bio *bio),
 
-	TP_ARGS(sb, fio, bio),
+	TP_ARGS(sb, type, bio),
 
 	TP_CONDITION(bio)
 );
 
-DEFINE_EVENT_CONDITION(f2fs__submit_bio, f2fs_submit_read_bio,
+DEFINE_EVENT_CONDITION(f2fs__bio, f2fs_submit_write_bio,
 
-	TP_PROTO(struct super_block *sb, struct f2fs_io_info *fio,
-							struct bio *bio),
+	TP_PROTO(struct super_block *sb, int type, struct bio *bio),
 
-	TP_ARGS(sb, fio, bio),
+	TP_ARGS(sb, type, bio),
 
 	TP_CONDITION(bio)
 );
@@ -1073,16 +1231,16 @@ TRACE_EVENT(f2fs_write_checkpoint,
 	),
 
 	TP_printk("dev = (%d,%d), checkpoint for %s, state = %s",
-		show_dev(__entry),
+		show_dev(__entry->dev),
 		show_cpreason(__entry->reason),
 		__entry->msg)
 );
 
-TRACE_EVENT(f2fs_issue_discard,
+DECLARE_EVENT_CLASS(f2fs_discard,
 
-	TP_PROTO(struct super_block *sb, block_t blkstart, block_t blklen),
+	TP_PROTO(struct block_device *dev, block_t blkstart, block_t blklen),
 
-	TP_ARGS(sb, blkstart, blklen),
+	TP_ARGS(dev, blkstart, blklen),
 
 	TP_STRUCT__entry(
 		__field(dev_t,	dev)
@@ -1091,40 +1249,78 @@ TRACE_EVENT(f2fs_issue_discard,
 	),
 
 	TP_fast_assign(
-		__entry->dev	= sb->s_dev;
+		__entry->dev	= dev->bd_dev;
 		__entry->blkstart = blkstart;
 		__entry->blklen = blklen;
 	),
 
 	TP_printk("dev = (%d,%d), blkstart = 0x%llx, blklen = 0x%llx",
-		show_dev(__entry),
+		show_dev(__entry->dev),
 		(unsigned long long)__entry->blkstart,
 		(unsigned long long)__entry->blklen)
 );
 
+DEFINE_EVENT(f2fs_discard, f2fs_queue_discard,
+
+	TP_PROTO(struct block_device *dev, block_t blkstart, block_t blklen),
+
+	TP_ARGS(dev, blkstart, blklen)
+);
+
+DEFINE_EVENT(f2fs_discard, f2fs_issue_discard,
+
+	TP_PROTO(struct block_device *dev, block_t blkstart, block_t blklen),
+
+	TP_ARGS(dev, blkstart, blklen)
+);
+
+TRACE_EVENT(f2fs_issue_reset_zone,
+
+	TP_PROTO(struct block_device *dev, block_t blkstart),
+
+	TP_ARGS(dev, blkstart),
+
+	TP_STRUCT__entry(
+		__field(dev_t,	dev)
+		__field(block_t, blkstart)
+	),
+
+	TP_fast_assign(
+		__entry->dev	= dev->bd_dev;
+		__entry->blkstart = blkstart;
+	),
+
+	TP_printk("dev = (%d,%d), reset zone at block = 0x%llx",
+		show_dev(__entry->dev),
+		(unsigned long long)__entry->blkstart)
+);
+
 TRACE_EVENT(f2fs_issue_flush,
 
-	TP_PROTO(struct super_block *sb, unsigned int nobarrier,
-					unsigned int flush_merge),
+	TP_PROTO(struct block_device *dev, unsigned int nobarrier,
+				unsigned int flush_merge, int ret),
 
-	TP_ARGS(sb, nobarrier, flush_merge),
+	TP_ARGS(dev, nobarrier, flush_merge, ret),
 
 	TP_STRUCT__entry(
 		__field(dev_t,	dev)
 		__field(unsigned int, nobarrier)
 		__field(unsigned int, flush_merge)
+		__field(int,  ret)
 	),
 
 	TP_fast_assign(
-		__entry->dev	= sb->s_dev;
+		__entry->dev	= dev->bd_dev;
 		__entry->nobarrier = nobarrier;
 		__entry->flush_merge = flush_merge;
+		__entry->ret = ret;
 	),
 
-	TP_printk("dev = (%d,%d), %s %s",
-		show_dev(__entry),
+	TP_printk("dev = (%d,%d), %s %s, ret = %d",
+		show_dev(__entry->dev),
 		__entry->nobarrier ? "skip (nobarrier)" : "issue",
-		__entry->flush_merge ? " with flush_merge" : "")
+		__entry->flush_merge ? " with flush_merge" : "",
+		__entry->ret)
 );
 
 TRACE_EVENT(f2fs_lookup_extent_tree_start,
@@ -1237,7 +1433,7 @@ TRACE_EVENT(f2fs_shrink_extent_tree,
 	),
 
 	TP_printk("dev = (%d,%d), shrunk: node_cnt = %u, tree_cnt = %u",
-		show_dev(__entry),
+		show_dev(__entry->dev),
 		__entry->node_cnt,
 		__entry->tree_cnt)
 );
@@ -1265,6 +1461,44 @@ TRACE_EVENT(f2fs_destroy_extent_tree,
 		__entry->node_cnt)
 );
 
+DECLARE_EVENT_CLASS(f2fs_sync_dirty_inodes,
+
+	TP_PROTO(struct super_block *sb, int type, s64 count),
+
+	TP_ARGS(sb, type, count),
+
+	TP_STRUCT__entry(
+		__field(dev_t, dev)
+		__field(int, type)
+		__field(s64, count)
+	),
+
+	TP_fast_assign(
+		__entry->dev	= sb->s_dev;
+		__entry->type	= type;
+		__entry->count	= count;
+	),
+
+	TP_printk("dev = (%d,%d), %s, dirty count = %lld",
+		show_dev(__entry->dev),
+		show_file_type(__entry->type),
+		__entry->count)
+);
+
+DEFINE_EVENT(f2fs_sync_dirty_inodes, f2fs_sync_dirty_inodes_enter,
+
+	TP_PROTO(struct super_block *sb, int type, s64 count),
+
+	TP_ARGS(sb, type, count)
+);
+
+DEFINE_EVENT(f2fs_sync_dirty_inodes, f2fs_sync_dirty_inodes_exit,
+
+	TP_PROTO(struct super_block *sb, int type, s64 count),
+
+	TP_ARGS(sb, type, count)
+);
+
 #endif /* _TRACE_F2FS_H */
 
  /* This part must be outside protection */
diff --git a/include/trace/events/gpu.h b/include/trace/events/gpu.h
new file mode 100644
index 000000000000..7e15cdfafe5a
--- /dev/null
+++ b/include/trace/events/gpu.h
@@ -0,0 +1,143 @@
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM gpu
+
+#if !defined(_TRACE_GPU_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_GPU_H
+
+#include <linux/tracepoint.h>
+#include <linux/time.h>
+
+#define show_secs_from_ns(ns) \
+	({ \
+		u64 t = ns + (NSEC_PER_USEC / 2); \
+		do_div(t, NSEC_PER_SEC); \
+		t; \
+	})
+
+#define show_usecs_from_ns(ns) \
+	({ \
+		u64 t = ns + (NSEC_PER_USEC / 2) ; \
+		u32 rem; \
+		do_div(t, NSEC_PER_USEC); \
+		rem = do_div(t, USEC_PER_SEC); \
+	})
+
+/*
+ * The gpu_sched_switch event indicates that a switch from one GPU context to
+ * another occurred on one of the GPU hardware blocks.
+ *
+ * The gpu_name argument identifies the GPU hardware block.  Each independently
+ * scheduled GPU hardware block should have a different name.  This may be used
+ * in different ways for different GPUs.  For example, if a GPU includes
+ * multiple processing cores it may use names "GPU 0", "GPU 1", etc.  If a GPU
+ * includes a separately scheduled 2D and 3D hardware block, it might use the
+ * names "2D" and "3D".
+ *
+ * The timestamp argument is the timestamp at which the switch occurred on the
+ * GPU. These timestamps are in units of nanoseconds and must use
+ * approximately the same time as sched_clock, though they need not come from
+ * any CPU clock. The timestamps for a single hardware block must be
+ * monotonically nondecreasing.  This means that if a variable compensation
+ * offset is used to translate from some other clock to the sched_clock, then
+ * care must be taken when increasing that offset, and doing so may result in
+ * multiple events with the same timestamp.
+ *
+ * The next_ctx_id argument identifies the next context that was running on
+ * the GPU hardware block.  A value of 0 indicates that the hardware block
+ * will be idle.
+ *
+ * The next_prio argument indicates the priority of the next context at the
+ * time of the event.  The exact numeric values may mean different things for
+ * different GPUs, but they should follow the rule that lower values indicate a
+ * higher priority.
+ *
+ * The next_job_id argument identifies the batch of work that the GPU will be
+ * working on.  This should correspond to a job_id that was previously traced
+ * as a gpu_job_enqueue event when the batch of work was created.
+ */
+TRACE_EVENT(gpu_sched_switch,
+
+	TP_PROTO(const char *gpu_name, u64 timestamp,
+		u32 next_ctx_id, s32 next_prio, u32 next_job_id),
+
+	TP_ARGS(gpu_name, timestamp, next_ctx_id, next_prio, next_job_id),
+
+	TP_STRUCT__entry(
+		__string(       gpu_name,       gpu_name        )
+		__field(        u64,            timestamp       )
+		__field(        u32,            next_ctx_id     )
+		__field(        s32,            next_prio       )
+		__field(        u32,            next_job_id     )
+	),
+
+	TP_fast_assign(
+		__assign_str(gpu_name, gpu_name);
+		__entry->timestamp = timestamp;
+		__entry->next_ctx_id = next_ctx_id;
+		__entry->next_prio = next_prio;
+		__entry->next_job_id = next_job_id;
+	),
+
+	TP_printk("gpu_name=%s ts=%llu.%06lu next_ctx_id=%lu next_prio=%ld "
+		"next_job_id=%lu",
+		__get_str(gpu_name),
+		(unsigned long long)show_secs_from_ns(__entry->timestamp),
+		(unsigned long)show_usecs_from_ns(__entry->timestamp),
+		(unsigned long)__entry->next_ctx_id,
+		(long)__entry->next_prio,
+		(unsigned long)__entry->next_job_id)
+);
+
+/*
+ * The gpu_job_enqueue event indicates that a batch of work has been queued up
+ * to be processed by the GPU.  This event is not intended to indicate that
+ * the batch of work has been submitted to the GPU hardware, but rather that
+ * it has been submitted to the GPU kernel driver.
+ *
+ * This event should be traced on the thread that initiated the work being
+ * queued.  For example, if a batch of work is submitted to the kernel by a
+ * userland thread, the event should be traced on that thread.
+ *
+ * The ctx_id field identifies the GPU context in which the batch of work
+ * being queued is to be run.
+ *
+ * The job_id field identifies the batch of work being queued within the given
+ * GPU context.  The first batch of work submitted for a given GPU context
+ * should have a job_id of 0, and each subsequent batch of work should
+ * increment the job_id by 1.
+ *
+ * The type field identifies the type of the job being enqueued.  The job
+ * types may be different for different GPU hardware.  For example, a GPU may
+ * differentiate between "2D", "3D", and "compute" jobs.
+ */
+TRACE_EVENT(gpu_job_enqueue,
+
+	TP_PROTO(u32 ctx_id, u32 job_id, const char *type),
+
+	TP_ARGS(ctx_id, job_id, type),
+
+	TP_STRUCT__entry(
+		__field(        u32,            ctx_id          )
+		__field(        u32,            job_id          )
+		__string(       type,           type            )
+	),
+
+	TP_fast_assign(
+		__entry->ctx_id = ctx_id;
+		__entry->job_id = job_id;
+		__assign_str(type, type);
+	),
+
+	TP_printk("ctx_id=%lu job_id=%lu type=%s",
+		(unsigned long)__entry->ctx_id,
+		(unsigned long)__entry->job_id,
+		__get_str(type))
+);
+
+#undef show_secs_from_ns
+#undef show_usecs_from_ns
+
+#endif /* _TRACE_GPU_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/mmc.h b/include/trace/events/mmc.h
new file mode 100644
index 000000000000..82b368dbcefc
--- /dev/null
+++ b/include/trace/events/mmc.h
@@ -0,0 +1,91 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM mmc
+
+#if !defined(_TRACE_MMC_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_MMC_H
+
+#include <linux/tracepoint.h>
+#include <linux/mmc/mmc.h>
+#include <linux/mmc/core.h>
+
+/*
+ * Unconditional logging of mmc block erase operations,
+ * including cmd, address, size
+ */
+DECLARE_EVENT_CLASS(mmc_blk_erase_class,
+	TP_PROTO(unsigned int cmd, unsigned int addr, unsigned int size),
+	TP_ARGS(cmd, addr, size),
+	TP_STRUCT__entry(
+		__field(unsigned int, cmd)
+		__field(unsigned int, addr)
+		__field(unsigned int, size)
+	),
+	TP_fast_assign(
+		__entry->cmd = cmd;
+		__entry->addr = addr;
+		__entry->size = size;
+	),
+	TP_printk("cmd=%u,addr=0x%08x,size=0x%08x",
+		  __entry->cmd, __entry->addr, __entry->size)
+);
+
+DEFINE_EVENT(mmc_blk_erase_class, mmc_blk_erase_start,
+	TP_PROTO(unsigned int cmd, unsigned int addr, unsigned int size),
+	TP_ARGS(cmd, addr, size));
+
+DEFINE_EVENT(mmc_blk_erase_class, mmc_blk_erase_end,
+	TP_PROTO(unsigned int cmd, unsigned int addr, unsigned int size),
+	TP_ARGS(cmd, addr, size));
+
+/*
+ * Logging of start of read or write mmc block operation,
+ * including cmd, address, size
+ */
+DECLARE_EVENT_CLASS(mmc_blk_rw_class,
+	TP_PROTO(unsigned int cmd, unsigned int addr, struct mmc_data *data),
+	TP_ARGS(cmd, addr, data),
+	TP_STRUCT__entry(
+		__field(unsigned int, cmd)
+		__field(unsigned int, addr)
+		__field(unsigned int, size)
+	),
+	TP_fast_assign(
+		__entry->cmd = cmd;
+		__entry->addr = addr;
+		__entry->size = data->blocks;
+	),
+	TP_printk("cmd=%u,addr=0x%08x,size=0x%08x",
+		  __entry->cmd, __entry->addr, __entry->size)
+);
+
+DEFINE_EVENT_CONDITION(mmc_blk_rw_class, mmc_blk_rw_start,
+	TP_PROTO(unsigned int cmd, unsigned int addr, struct mmc_data *data),
+	TP_ARGS(cmd, addr, data),
+	TP_CONDITION(((cmd == MMC_READ_MULTIPLE_BLOCK) ||
+		      (cmd == MMC_WRITE_MULTIPLE_BLOCK)) &&
+		      data));
+
+DEFINE_EVENT_CONDITION(mmc_blk_rw_class, mmc_blk_rw_end,
+	TP_PROTO(unsigned int cmd, unsigned int addr, struct mmc_data *data),
+	TP_ARGS(cmd, addr, data),
+	TP_CONDITION(((cmd == MMC_READ_MULTIPLE_BLOCK) ||
+		      (cmd == MMC_WRITE_MULTIPLE_BLOCK)) &&
+		      data));
+#endif /* _TRACE_MMC_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>
diff --git a/include/trace/events/net.h b/include/trace/events/net.h
index 49cc7c3de252..89d009e10938 100644
--- a/include/trace/events/net.h
+++ b/include/trace/events/net.h
@@ -57,7 +57,7 @@ TRACE_EVENT(net_dev_start_xmit,
 		__entry->gso_type = skb_shinfo(skb)->gso_type;
 	),
 
-	TP_printk("dev=%s queue_mapping=%u skbaddr=%p vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d len=%u data_len=%u network_offset=%d transport_offset_valid=%d transport_offset=%d tx_flags=%d gso_size=%d gso_segs=%d gso_type=%#x",
+	TP_printk("dev=%s queue_mapping=%u skbaddr=%pK vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d len=%u data_len=%u network_offset=%d transport_offset_valid=%d transport_offset=%d tx_flags=%d gso_size=%d gso_segs=%d gso_type=%#x",
 		  __get_str(name), __entry->queue_mapping, __entry->skbaddr,
 		  __entry->vlan_tagged, __entry->vlan_proto, __entry->vlan_tci,
 		  __entry->protocol, __entry->ip_summed, __entry->len,
@@ -90,7 +90,7 @@ TRACE_EVENT(net_dev_xmit,
 		__assign_str(name, dev->name);
 	),
 
-	TP_printk("dev=%s skbaddr=%p len=%u rc=%d",
+	TP_printk("dev=%s skbaddr=%pK len=%u rc=%d",
 		__get_str(name), __entry->skbaddr, __entry->len, __entry->rc)
 );
 
@@ -112,7 +112,7 @@ DECLARE_EVENT_CLASS(net_dev_template,
 		__assign_str(name, skb->dev->name);
 	),
 
-	TP_printk("dev=%s skbaddr=%p len=%u",
+	TP_printk("dev=%s skbaddr=%pK len=%u",
 		__get_str(name), __entry->skbaddr, __entry->len)
 )
 
@@ -191,7 +191,7 @@ DECLARE_EVENT_CLASS(net_dev_rx_verbose_template,
 		__entry->gso_type = skb_shinfo(skb)->gso_type;
 	),
 
-	TP_printk("dev=%s napi_id=%#x queue_mapping=%u skbaddr=%p vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d hash=0x%08x l4_hash=%d len=%u data_len=%u truesize=%u mac_header_valid=%d mac_header=%d nr_frags=%d gso_size=%d gso_type=%#x",
+	TP_printk("dev=%s napi_id=%#x queue_mapping=%u skbaddr=%pK vlan_tagged=%d vlan_proto=0x%04x vlan_tci=0x%04x protocol=0x%04x ip_summed=%d hash=0x%08x l4_hash=%d len=%u data_len=%u truesize=%u mac_header_valid=%d mac_header=%d nr_frags=%d gso_size=%d gso_type=%#x",
 		  __get_str(name), __entry->napi_id, __entry->queue_mapping,
 		  __entry->skbaddr, __entry->vlan_tagged, __entry->vlan_proto,
 		  __entry->vlan_tci, __entry->protocol, __entry->ip_summed,
diff --git a/include/trace/events/power.h b/include/trace/events/power.h
index 284244ebfe8d..8924cc2b4ca8 100644
--- a/include/trace/events/power.h
+++ b/include/trace/events/power.h
@@ -120,6 +120,38 @@ DEFINE_EVENT(cpu, cpu_frequency,
 	TP_ARGS(frequency, cpu_id)
 );
 
+TRACE_EVENT(cpu_frequency_limits,
+
+	TP_PROTO(unsigned int max_freq, unsigned int min_freq,
+		unsigned int cpu_id),
+
+	TP_ARGS(max_freq, min_freq, cpu_id),
+
+	TP_STRUCT__entry(
+		__field(	u32,		min_freq	)
+		__field(	u32,		max_freq	)
+		__field(	u32,		cpu_id		)
+	),
+
+	TP_fast_assign(
+		__entry->min_freq = min_freq;
+		__entry->max_freq = max_freq;
+		__entry->cpu_id = cpu_id;
+	),
+
+	TP_printk("min=%lu max=%lu cpu_id=%lu",
+		  (unsigned long)__entry->min_freq,
+		  (unsigned long)__entry->max_freq,
+		  (unsigned long)__entry->cpu_id)
+);
+
+DEFINE_EVENT(cpu, cpu_capacity,
+
+	TP_PROTO(unsigned int capacity, unsigned int cpu_id),
+
+	TP_ARGS(capacity, cpu_id)
+);
+
 TRACE_EVENT(device_pm_callback_start,
 
 	TP_PROTO(struct device *dev, const char *pm_ops, int event),
@@ -273,6 +305,25 @@ DEFINE_EVENT(clock, clock_set_rate,
 	TP_ARGS(name, state, cpu_id)
 );
 
+TRACE_EVENT(clock_set_parent,
+
+	TP_PROTO(const char *name, const char *parent_name),
+
+	TP_ARGS(name, parent_name),
+
+	TP_STRUCT__entry(
+		__string(       name,           name            )
+		__string(       parent_name,    parent_name     )
+	),
+
+	TP_fast_assign(
+		__assign_str(name, name);
+		__assign_str(parent_name, parent_name);
+	),
+
+	TP_printk("%s parent=%s", __get_str(name), __get_str(parent_name))
+);
+
 /*
  * The power domain events are used for power domains transitions
  */
diff --git a/include/trace/events/preemptirq.h b/include/trace/events/preemptirq.h
new file mode 100644
index 000000000000..f5024c560d8f
--- /dev/null
+++ b/include/trace/events/preemptirq.h
@@ -0,0 +1,70 @@
+#ifdef CONFIG_PREEMPTIRQ_EVENTS
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM preemptirq
+
+#if !defined(_TRACE_PREEMPTIRQ_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_PREEMPTIRQ_H
+
+#include <linux/ktime.h>
+#include <linux/tracepoint.h>
+#include <linux/string.h>
+#include <asm/sections.h>
+
+DECLARE_EVENT_CLASS(preemptirq_template,
+
+	TP_PROTO(unsigned long ip, unsigned long parent_ip),
+
+	TP_ARGS(ip, parent_ip),
+
+	TP_STRUCT__entry(
+		__field(u32, caller_offs)
+		__field(u32, parent_offs)
+	),
+
+	TP_fast_assign(
+		__entry->caller_offs = (u32)(ip - (unsigned long)_stext);
+		__entry->parent_offs = (u32)(parent_ip - (unsigned long)_stext);
+	),
+
+	TP_printk("caller=%pF parent=%pF",
+		  (void *)((unsigned long)(_stext) + __entry->caller_offs),
+		  (void *)((unsigned long)(_stext) + __entry->parent_offs))
+);
+
+#ifndef CONFIG_PROVE_LOCKING
+DEFINE_EVENT(preemptirq_template, irq_disable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+
+DEFINE_EVENT(preemptirq_template, irq_enable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+#endif
+
+#ifdef CONFIG_DEBUG_PREEMPT
+DEFINE_EVENT(preemptirq_template, preempt_disable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+
+DEFINE_EVENT(preemptirq_template, preempt_enable,
+	     TP_PROTO(unsigned long ip, unsigned long parent_ip),
+	     TP_ARGS(ip, parent_ip));
+#endif
+
+#endif /* _TRACE_PREEMPTIRQ_H */
+
+#include <trace/define_trace.h>
+
+#else /* !CONFIG_PREEMPTIRQ_EVENTS */
+
+#define trace_irq_enable(...)
+#define trace_irq_disable(...)
+#define trace_preempt_enable(...)
+#define trace_preempt_disable(...)
+#define trace_irq_enable_rcuidle(...)
+#define trace_irq_disable_rcuidle(...)
+#define trace_preempt_enable_rcuidle(...)
+#define trace_preempt_disable_rcuidle(...)
+
+#endif
diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h
index 9b90c57517a9..9f7fd0c6662b 100644
--- a/include/trace/events/sched.h
+++ b/include/trace/events/sched.h
@@ -191,6 +191,31 @@ TRACE_EVENT(sched_migrate_task,
 		  __entry->orig_cpu, __entry->dest_cpu)
 );
 
+/*
+ * Tracepoint for a CPU going offline/online:
+ */
+TRACE_EVENT(sched_cpu_hotplug,
+
+	TP_PROTO(int affected_cpu, int error, int status),
+
+	TP_ARGS(affected_cpu, error, status),
+
+	TP_STRUCT__entry(
+		__field(	int,	affected_cpu		)
+		__field(	int,	error			)
+		__field(	int,	status			)
+	),
+
+	TP_fast_assign(
+		__entry->affected_cpu	= affected_cpu;
+		__entry->error		= error;
+		__entry->status		= status;
+	),
+
+	TP_printk("cpu %d %s error=%d", __entry->affected_cpu,
+		__entry->status ? "online" : "offline", __entry->error)
+);
+
 DECLARE_EVENT_CLASS(sched_process_template,
 
 	TP_PROTO(struct task_struct *p),
@@ -219,7 +244,7 @@ DECLARE_EVENT_CLASS(sched_process_template,
 DEFINE_EVENT(sched_process_template, sched_process_free,
 	     TP_PROTO(struct task_struct *p),
 	     TP_ARGS(p));
-	     
+
 
 /*
  * Tracepoint for a task exiting:
@@ -374,6 +399,30 @@ DEFINE_EVENT(sched_stat_template, sched_stat_blocked,
 	     TP_ARGS(tsk, delay));
 
 /*
+ * Tracepoint for recording the cause of uninterruptible sleep.
+ */
+TRACE_EVENT(sched_blocked_reason,
+
+	TP_PROTO(struct task_struct *tsk),
+
+	TP_ARGS(tsk),
+
+	TP_STRUCT__entry(
+		__field( pid_t,	pid	)
+		__field( void*, caller	)
+		__field( bool, io_wait	)
+	),
+
+	TP_fast_assign(
+		__entry->pid	= tsk->pid;
+		__entry->caller = (void*)get_wchan(tsk);
+		__entry->io_wait = tsk->in_iowait;
+	),
+
+	TP_printk("pid=%d iowait=%d caller=%pS", __entry->pid, __entry->io_wait, __entry->caller)
+);
+
+/*
  * Tracepoint for accounting runtime (time the task is executing
  * on a CPU).
  */
@@ -562,6 +611,577 @@ TRACE_EVENT(sched_wake_idle_without_ipi,
 
 	TP_printk("cpu=%d", __entry->cpu)
 );
+
+TRACE_EVENT(sched_contrib_scale_f,
+
+	TP_PROTO(int cpu, unsigned long freq_scale_factor,
+		 unsigned long cpu_scale_factor),
+
+	TP_ARGS(cpu, freq_scale_factor, cpu_scale_factor),
+
+	TP_STRUCT__entry(
+		__field(int, cpu)
+		__field(unsigned long, freq_scale_factor)
+		__field(unsigned long, cpu_scale_factor)
+	),
+
+	TP_fast_assign(
+		__entry->cpu = cpu;
+		__entry->freq_scale_factor = freq_scale_factor;
+		__entry->cpu_scale_factor = cpu_scale_factor;
+	),
+
+	TP_printk("cpu=%d freq_scale_factor=%lu cpu_scale_factor=%lu",
+		  __entry->cpu, __entry->freq_scale_factor,
+		  __entry->cpu_scale_factor)
+);
+
+#ifdef CONFIG_SMP
+
+#ifdef CONFIG_SCHED_WALT
+extern unsigned int sysctl_sched_use_walt_cpu_util;
+extern unsigned int sysctl_sched_use_walt_task_util;
+extern unsigned int walt_ravg_window;
+extern bool walt_disabled;
+#endif
+
+/*
+ * Tracepoint for accounting sched averages for tasks.
+ */
+TRACE_EVENT(sched_load_avg_task,
+
+	TP_PROTO(struct task_struct *tsk, struct sched_avg *avg, void *_ravg),
+
+	TP_ARGS(tsk, avg, _ravg),
+
+	TP_STRUCT__entry(
+		__array( char,	comm,	TASK_COMM_LEN		)
+		__field( pid_t,	pid				)
+		__field( int,	cpu				)
+		__field( unsigned long,	load_avg		)
+		__field( unsigned long,	util_avg		)
+		__field( unsigned long,	util_avg_pelt	)
+		__field( unsigned long,	util_avg_walt	)
+		__field( u64,		load_sum		)
+		__field( u32,		util_sum		)
+		__field( u32,		period_contrib		)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
+		__entry->pid			= tsk->pid;
+		__entry->cpu			= task_cpu(tsk);
+		__entry->load_avg		= avg->load_avg;
+		__entry->util_avg		= avg->util_avg;
+		__entry->load_sum		= avg->load_sum;
+		__entry->util_sum		= avg->util_sum;
+		__entry->period_contrib		= avg->period_contrib;
+		__entry->util_avg_pelt  = avg->util_avg;
+		__entry->util_avg_walt  = 0;
+#ifdef CONFIG_SCHED_WALT
+		__entry->util_avg_walt = (((unsigned long)((struct ravg*)_ravg)->demand) << SCHED_LOAD_SHIFT);
+		do_div(__entry->util_avg_walt, walt_ravg_window);
+		if (!walt_disabled && sysctl_sched_use_walt_task_util)
+			__entry->util_avg = __entry->util_avg_walt;
+#endif
+	),
+	TP_printk("comm=%s pid=%d cpu=%d load_avg=%lu util_avg=%lu "
+			"util_avg_pelt=%lu util_avg_walt=%lu load_sum=%llu"
+		  " util_sum=%u period_contrib=%u",
+		  __entry->comm,
+		  __entry->pid,
+		  __entry->cpu,
+		  __entry->load_avg,
+		  __entry->util_avg,
+		  __entry->util_avg_pelt,
+		  __entry->util_avg_walt,
+		  (u64)__entry->load_sum,
+		  (u32)__entry->util_sum,
+		  (u32)__entry->period_contrib)
+);
+
+/*
+ * Tracepoint for accounting sched averages for cpus.
+ */
+TRACE_EVENT(sched_load_avg_cpu,
+
+	TP_PROTO(int cpu, struct cfs_rq *cfs_rq),
+
+	TP_ARGS(cpu, cfs_rq),
+
+	TP_STRUCT__entry(
+		__field( int,	cpu				)
+		__field( unsigned long,	load_avg		)
+		__field( unsigned long,	util_avg		)
+		__field( unsigned long,	util_avg_pelt	)
+		__field( unsigned long,	util_avg_walt	)
+	),
+
+	TP_fast_assign(
+		__entry->cpu			= cpu;
+		__entry->load_avg		= cfs_rq->avg.load_avg;
+		__entry->util_avg		= cfs_rq->avg.util_avg;
+		__entry->util_avg_pelt	= cfs_rq->avg.util_avg;
+		__entry->util_avg_walt	= 0;
+#ifdef CONFIG_SCHED_WALT
+		__entry->util_avg_walt =
+				div64_u64(cpu_rq(cpu)->cumulative_runnable_avg,
+						  walt_ravg_window >> SCHED_LOAD_SHIFT);
+		if (!walt_disabled && sysctl_sched_use_walt_cpu_util)
+			__entry->util_avg		= __entry->util_avg_walt;
+#endif
+	),
+
+	TP_printk("cpu=%d load_avg=%lu util_avg=%lu "
+			  "util_avg_pelt=%lu util_avg_walt=%lu",
+		  __entry->cpu, __entry->load_avg, __entry->util_avg,
+		  __entry->util_avg_pelt, __entry->util_avg_walt)
+);
+
+/*
+ * Tracepoint for sched_tune_config settings
+ */
+TRACE_EVENT(sched_tune_config,
+
+	TP_PROTO(int boost),
+
+	TP_ARGS(boost),
+
+	TP_STRUCT__entry(
+		__field( int,	boost		)
+	),
+
+	TP_fast_assign(
+		__entry->boost 	= boost;
+	),
+
+	TP_printk("boost=%d ", __entry->boost)
+);
+
+/*
+ * Tracepoint for accounting CPU  boosted utilization
+ */
+TRACE_EVENT(sched_boost_cpu,
+
+	TP_PROTO(int cpu, unsigned long util, long margin),
+
+	TP_ARGS(cpu, util, margin),
+
+	TP_STRUCT__entry(
+		__field( int,		cpu			)
+		__field( unsigned long,	util			)
+		__field(long,		margin			)
+	),
+
+	TP_fast_assign(
+		__entry->cpu	= cpu;
+		__entry->util	= util;
+		__entry->margin	= margin;
+	),
+
+	TP_printk("cpu=%d util=%lu margin=%ld",
+		  __entry->cpu,
+		  __entry->util,
+		  __entry->margin)
+);
+
+/*
+ * Tracepoint for schedtune_tasks_update
+ */
+TRACE_EVENT(sched_tune_tasks_update,
+
+	TP_PROTO(struct task_struct *tsk, int cpu, int tasks, int idx,
+		int boost, int max_boost),
+
+	TP_ARGS(tsk, cpu, tasks, idx, boost, max_boost),
+
+	TP_STRUCT__entry(
+		__array( char,	comm,	TASK_COMM_LEN	)
+		__field( pid_t,		pid		)
+		__field( int,		cpu		)
+		__field( int,		tasks		)
+		__field( int,		idx		)
+		__field( int,		boost		)
+		__field( int,		max_boost	)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
+		__entry->pid		= tsk->pid;
+		__entry->cpu 		= cpu;
+		__entry->tasks		= tasks;
+		__entry->idx 		= idx;
+		__entry->boost		= boost;
+		__entry->max_boost	= max_boost;
+	),
+
+	TP_printk("pid=%d comm=%s "
+			"cpu=%d tasks=%d idx=%d boost=%d max_boost=%d",
+		__entry->pid, __entry->comm,
+		__entry->cpu, __entry->tasks, __entry->idx,
+		__entry->boost, __entry->max_boost)
+);
+
+/*
+ * Tracepoint for schedtune_boostgroup_update
+ */
+TRACE_EVENT(sched_tune_boostgroup_update,
+
+	TP_PROTO(int cpu, int variation, int max_boost),
+
+	TP_ARGS(cpu, variation, max_boost),
+
+	TP_STRUCT__entry(
+		__field( int,	cpu		)
+		__field( int,	variation	)
+		__field( int,	max_boost	)
+	),
+
+	TP_fast_assign(
+		__entry->cpu		= cpu;
+		__entry->variation	= variation;
+		__entry->max_boost	= max_boost;
+	),
+
+	TP_printk("cpu=%d variation=%d max_boost=%d",
+		__entry->cpu, __entry->variation, __entry->max_boost)
+);
+
+/*
+ * Tracepoint for accounting task boosted utilization
+ */
+TRACE_EVENT(sched_boost_task,
+
+	TP_PROTO(struct task_struct *tsk, unsigned long util, long margin),
+
+	TP_ARGS(tsk, util, margin),
+
+	TP_STRUCT__entry(
+		__array( char,	comm,	TASK_COMM_LEN		)
+		__field( pid_t,		pid			)
+		__field( unsigned long,	util			)
+		__field( long,		margin			)
+
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
+		__entry->pid	= tsk->pid;
+		__entry->util	= util;
+		__entry->margin	= margin;
+	),
+
+	TP_printk("comm=%s pid=%d util=%lu margin=%ld",
+		  __entry->comm, __entry->pid,
+		  __entry->util,
+		  __entry->margin)
+);
+
+/*
+ * Tracepoint for find_best_target
+ */
+TRACE_EVENT(sched_find_best_target,
+
+	TP_PROTO(struct task_struct *tsk, bool prefer_idle,
+		unsigned long min_util, int start_cpu,
+		int best_idle, int best_active, int target),
+
+	TP_ARGS(tsk, prefer_idle, min_util, start_cpu,
+		best_idle, best_active, target),
+
+	TP_STRUCT__entry(
+		__array( char,	comm,	TASK_COMM_LEN	)
+		__field( pid_t,	pid			)
+		__field( unsigned long,	min_util	)
+		__field( bool,	prefer_idle		)
+		__field( int,	start_cpu		)
+		__field( int,	best_idle		)
+		__field( int,	best_active		)
+		__field( int,	target			)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
+		__entry->pid		= tsk->pid;
+		__entry->min_util	= min_util;
+		__entry->prefer_idle	= prefer_idle;
+		__entry->start_cpu 	= start_cpu;
+		__entry->best_idle	= best_idle;
+		__entry->best_active	= best_active;
+		__entry->target		= target;
+	),
+
+	TP_printk("pid=%d comm=%s prefer_idle=%d start_cpu=%d "
+		  "best_idle=%d best_active=%d target=%d",
+		__entry->pid, __entry->comm,
+		__entry->prefer_idle, __entry->start_cpu,
+		__entry->best_idle, __entry->best_active,
+		__entry->target)
+);
+
+/*
+ * Tracepoint for accounting sched group energy
+ */
+TRACE_EVENT(sched_energy_diff,
+
+	TP_PROTO(struct task_struct *tsk, int scpu, int dcpu, int udelta,
+		int nrgb, int nrga, int nrgd, int capb, int capa, int capd,
+		int nrgn, int nrgp),
+
+	TP_ARGS(tsk, scpu, dcpu, udelta,
+		nrgb, nrga, nrgd, capb, capa, capd,
+		nrgn, nrgp),
+
+	TP_STRUCT__entry(
+		__array( char,	comm,	TASK_COMM_LEN	)
+		__field( pid_t,	pid	)
+		__field( int,	scpu	)
+		__field( int,	dcpu	)
+		__field( int,	udelta	)
+		__field( int,	nrgb	)
+		__field( int,	nrga	)
+		__field( int,	nrgd	)
+		__field( int,	capb	)
+		__field( int,	capa	)
+		__field( int,	capd	)
+		__field( int,	nrgn	)
+		__field( int,	nrgp	)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, tsk->comm, TASK_COMM_LEN);
+		__entry->pid		= tsk->pid;
+		__entry->scpu 		= scpu;
+		__entry->dcpu 		= dcpu;
+		__entry->udelta 	= udelta;
+		__entry->nrgb 		= nrgb;
+		__entry->nrga 		= nrga;
+		__entry->nrgd 		= nrgd;
+		__entry->capb 		= capb;
+		__entry->capa 		= capa;
+		__entry->capd 		= capd;
+		__entry->nrgn 		= nrgn;
+		__entry->nrgp 		= nrgp;
+	),
+
+	TP_printk("pid=%d comm=%s "
+			"src_cpu=%d dst_cpu=%d usage_delta=%d "
+			"nrg_before=%d nrg_after=%d nrg_diff=%d "
+			"cap_before=%d cap_after=%d cap_delta=%d "
+			"nrg_delta=%d nrg_payoff=%d",
+		__entry->pid, __entry->comm,
+		__entry->scpu, __entry->dcpu, __entry->udelta,
+		__entry->nrgb, __entry->nrga, __entry->nrgd,
+		__entry->capb, __entry->capa, __entry->capd,
+		__entry->nrgn, __entry->nrgp)
+);
+
+/*
+ * Tracepoint for schedtune_tasks_update
+ */
+TRACE_EVENT(sched_tune_filter,
+
+	TP_PROTO(int nrg_delta, int cap_delta,
+		 int nrg_gain,  int cap_gain,
+		 int payoff, int region),
+
+	TP_ARGS(nrg_delta, cap_delta, nrg_gain, cap_gain, payoff, region),
+
+	TP_STRUCT__entry(
+		__field( int,	nrg_delta	)
+		__field( int,	cap_delta	)
+		__field( int,	nrg_gain	)
+		__field( int,	cap_gain	)
+		__field( int,	payoff		)
+		__field( int,	region		)
+	),
+
+	TP_fast_assign(
+		__entry->nrg_delta	= nrg_delta;
+		__entry->cap_delta	= cap_delta;
+		__entry->nrg_gain	= nrg_gain;
+		__entry->cap_gain	= cap_gain;
+		__entry->payoff		= payoff;
+		__entry->region		= region;
+	),
+
+	TP_printk("nrg_delta=%d cap_delta=%d nrg_gain=%d cap_gain=%d payoff=%d region=%d",
+		__entry->nrg_delta, __entry->cap_delta,
+		__entry->nrg_gain, __entry->cap_gain,
+		__entry->payoff, __entry->region)
+);
+
+/*
+ * Tracepoint for system overutilized flag
+ */
+TRACE_EVENT(sched_overutilized,
+
+	TP_PROTO(bool overutilized),
+
+	TP_ARGS(overutilized),
+
+	TP_STRUCT__entry(
+		__field( bool,	overutilized	)
+	),
+
+	TP_fast_assign(
+		__entry->overutilized	= overutilized;
+	),
+
+	TP_printk("overutilized=%d",
+		__entry->overutilized ? 1 : 0)
+);
+#ifdef CONFIG_SCHED_WALT
+struct rq;
+
+TRACE_EVENT(walt_update_task_ravg,
+
+	TP_PROTO(struct task_struct *p, struct rq *rq, int evt,
+						u64 wallclock, u64 irqtime),
+
+	TP_ARGS(p, rq, evt, wallclock, irqtime),
+
+	TP_STRUCT__entry(
+		__array(	char,	comm,   TASK_COMM_LEN	)
+		__field(	pid_t,	pid			)
+		__field(	pid_t,	cur_pid			)
+		__field(	u64,	wallclock		)
+		__field(	u64,	mark_start		)
+		__field(	u64,	delta_m			)
+		__field(	u64,	win_start		)
+		__field(	u64,	delta			)
+		__field(	u64,	irqtime			)
+		__field(        int,    evt			)
+		__field(unsigned int,	demand			)
+		__field(unsigned int,	sum			)
+		__field(	 int,	cpu			)
+		__field(	u64,	cs			)
+		__field(	u64,	ps			)
+		__field(unsigned long,	util			)
+		__field(	u32,	curr_window		)
+		__field(	u32,	prev_window		)
+		__field(	u64,	nt_cs			)
+		__field(	u64,	nt_ps			)
+		__field(	u32,	active_windows		)
+	),
+
+	TP_fast_assign(
+		__entry->wallclock      = wallclock;
+		__entry->win_start      = rq->window_start;
+		__entry->delta          = (wallclock - rq->window_start);
+		__entry->evt            = evt;
+		__entry->cpu            = rq->cpu;
+		__entry->cur_pid        = rq->curr->pid;
+		memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
+		__entry->pid            = p->pid;
+		__entry->mark_start     = p->ravg.mark_start;
+		__entry->delta_m        = (wallclock - p->ravg.mark_start);
+		__entry->demand         = p->ravg.demand;
+		__entry->sum            = p->ravg.sum;
+		__entry->irqtime        = irqtime;
+		__entry->cs             = rq->curr_runnable_sum;
+		__entry->ps             = rq->prev_runnable_sum;
+		__entry->util           = rq->prev_runnable_sum << SCHED_LOAD_SHIFT;
+		do_div(__entry->util, walt_ravg_window);
+		__entry->curr_window	= p->ravg.curr_window;
+		__entry->prev_window	= p->ravg.prev_window;
+		__entry->nt_cs		= rq->nt_curr_runnable_sum;
+		__entry->nt_ps		= rq->nt_prev_runnable_sum;
+		__entry->active_windows	= p->ravg.active_windows;
+	),
+
+	TP_printk("wc %llu ws %llu delta %llu event %d cpu %d cur_pid %d task %d (%s) ms %llu delta %llu demand %u sum %u irqtime %llu"
+		" cs %llu ps %llu util %lu cur_window %u prev_window %u active_wins %u"
+		, __entry->wallclock, __entry->win_start, __entry->delta,
+		__entry->evt, __entry->cpu, __entry->cur_pid,
+		__entry->pid, __entry->comm, __entry->mark_start,
+		__entry->delta_m, __entry->demand,
+		__entry->sum, __entry->irqtime,
+		__entry->cs, __entry->ps, __entry->util,
+		__entry->curr_window, __entry->prev_window,
+		  __entry->active_windows
+		)
+);
+
+TRACE_EVENT(walt_update_history,
+
+	TP_PROTO(struct rq *rq, struct task_struct *p, u32 runtime, int samples,
+			int evt),
+
+	TP_ARGS(rq, p, runtime, samples, evt),
+
+	TP_STRUCT__entry(
+		__array(	char,	comm,   TASK_COMM_LEN	)
+		__field(	pid_t,	pid			)
+		__field(unsigned int,	runtime			)
+		__field(	 int,	samples			)
+		__field(	 int,	evt			)
+		__field(	 u64,	demand			)
+		__field(	 u64,	walt_avg		)
+		__field(unsigned int,	pelt_avg		)
+		__array(	 u32,	hist, RAVG_HIST_SIZE_MAX)
+		__field(	 int,	cpu			)
+	),
+
+	TP_fast_assign(
+		memcpy(__entry->comm, p->comm, TASK_COMM_LEN);
+		__entry->pid            = p->pid;
+		__entry->runtime        = runtime;
+		__entry->samples        = samples;
+		__entry->evt            = evt;
+		__entry->demand         = p->ravg.demand;
+		__entry->walt_avg	= (__entry->demand << 10);
+		do_div(__entry->walt_avg, walt_ravg_window);
+		__entry->pelt_avg	= p->se.avg.util_avg;
+		memcpy(__entry->hist, p->ravg.sum_history,
+					RAVG_HIST_SIZE_MAX * sizeof(u32));
+		__entry->cpu            = rq->cpu;
+	),
+
+	TP_printk("%d (%s): runtime %u samples %d event %d demand %llu"
+		" walt %llu pelt %u (hist: %u %u %u %u %u) cpu %d",
+		__entry->pid, __entry->comm,
+		__entry->runtime, __entry->samples, __entry->evt,
+		__entry->demand,
+		__entry->walt_avg,
+		__entry->pelt_avg,
+		__entry->hist[0], __entry->hist[1],
+		__entry->hist[2], __entry->hist[3],
+		__entry->hist[4], __entry->cpu)
+);
+
+TRACE_EVENT(walt_migration_update_sum,
+
+	TP_PROTO(struct rq *rq, struct task_struct *p),
+
+	TP_ARGS(rq, p),
+
+	TP_STRUCT__entry(
+		__field(int,		cpu			)
+		__field(int,		pid			)
+		__field(	u64,	cs			)
+		__field(	u64,	ps			)
+		__field(	s64,	nt_cs			)
+		__field(	s64,	nt_ps			)
+	),
+
+	TP_fast_assign(
+		__entry->cpu		= cpu_of(rq);
+		__entry->cs		= rq->curr_runnable_sum;
+		__entry->ps		= rq->prev_runnable_sum;
+		__entry->nt_cs		= (s64)rq->nt_curr_runnable_sum;
+		__entry->nt_ps		= (s64)rq->nt_prev_runnable_sum;
+		__entry->pid		= p->pid;
+	),
+
+	TP_printk("cpu %d: cs %llu ps %llu nt_cs %lld nt_ps %lld pid %d",
+		  __entry->cpu, __entry->cs, __entry->ps,
+		  __entry->nt_cs, __entry->nt_ps, __entry->pid)
+);
+#endif /* CONFIG_SCHED_WALT */
+
+#endif /* CONFIG_SMP */
+
 #endif /* _TRACE_SCHED_H */
 
 /* This part must be outside protection */
diff --git a/include/uapi/linux/android/binder.h b/include/uapi/linux/android/binder.h
index 41420e341e75..5539933b3491 100644
--- a/include/uapi/linux/android/binder.h
+++ b/include/uapi/linux/android/binder.h
@@ -33,11 +33,60 @@ enum {
 	BINDER_TYPE_HANDLE	= B_PACK_CHARS('s', 'h', '*', B_TYPE_LARGE),
 	BINDER_TYPE_WEAK_HANDLE	= B_PACK_CHARS('w', 'h', '*', B_TYPE_LARGE),
 	BINDER_TYPE_FD		= B_PACK_CHARS('f', 'd', '*', B_TYPE_LARGE),
+	BINDER_TYPE_FDA		= B_PACK_CHARS('f', 'd', 'a', B_TYPE_LARGE),
+	BINDER_TYPE_PTR		= B_PACK_CHARS('p', 't', '*', B_TYPE_LARGE),
 };
 
-enum {
+/**
+ * enum flat_binder_object_shifts: shift values for flat_binder_object_flags
+ * @FLAT_BINDER_FLAG_SCHED_POLICY_SHIFT: shift for getting scheduler policy.
+ *
+ */
+enum flat_binder_object_shifts {
+	FLAT_BINDER_FLAG_SCHED_POLICY_SHIFT = 9,
+};
+
+/**
+ * enum flat_binder_object_flags - flags for use in flat_binder_object.flags
+ */
+enum flat_binder_object_flags {
+	/**
+	 * @FLAT_BINDER_FLAG_PRIORITY_MASK: bit-mask for min scheduler priority
+	 *
+	 * These bits can be used to set the minimum scheduler priority
+	 * at which transactions into this node should run. Valid values
+	 * in these bits depend on the scheduler policy encoded in
+	 * @FLAT_BINDER_FLAG_SCHED_POLICY_MASK.
+	 *
+	 * For SCHED_NORMAL/SCHED_BATCH, the valid range is between [-20..19]
+	 * For SCHED_FIFO/SCHED_RR, the value can run between [1..99]
+	 */
 	FLAT_BINDER_FLAG_PRIORITY_MASK = 0xff,
+	/**
+	 * @FLAT_BINDER_FLAG_ACCEPTS_FDS: whether the node accepts fds.
+	 */
 	FLAT_BINDER_FLAG_ACCEPTS_FDS = 0x100,
+	/**
+	 * @FLAT_BINDER_FLAG_SCHED_POLICY_MASK: bit-mask for scheduling policy
+	 *
+	 * These two bits can be used to set the min scheduling policy at which
+	 * transactions on this node should run. These match the UAPI
+	 * scheduler policy values, eg:
+	 * 00b: SCHED_NORMAL
+	 * 01b: SCHED_FIFO
+	 * 10b: SCHED_RR
+	 * 11b: SCHED_BATCH
+	 */
+	FLAT_BINDER_FLAG_SCHED_POLICY_MASK =
+		3U << FLAT_BINDER_FLAG_SCHED_POLICY_SHIFT,
+
+	/**
+	 * @FLAT_BINDER_FLAG_INHERIT_RT: whether the node inherits RT policy
+	 *
+	 * Only when set, calls into this node will inherit a real-time
+	 * scheduling policy from the caller (for synchronous transactions).
+	 */
+	FLAT_BINDER_FLAG_INHERIT_RT = 0x800,
 };
 
 #ifdef BINDER_IPC_32BIT
@@ -48,6 +97,14 @@ typedef __u64 binder_size_t;
 typedef __u64 binder_uintptr_t;
 #endif
 
+/**
+ * struct binder_object_header - header shared by all binder metadata objects.
+ * @type:	type of the object
+ */
+struct binder_object_header {
+	__u32        type;
+};
+
 /*
  * This is the flattened representation of a Binder object for transfer
  * between processes.  The 'offsets' supplied as part of a binder transaction
@@ -56,9 +113,8 @@ typedef __u64 binder_uintptr_t;
  * between processes.
  */
 struct flat_binder_object {
-	/* 8 bytes for large_flat_header. */
-	__u32		type;
-	__u32		flags;
+	struct binder_object_header	hdr;
+	__u32				flags;
 
 	/* 8 bytes of data. */
 	union {
@@ -70,6 +126,86 @@ struct flat_binder_object {
 	binder_uintptr_t	cookie;
 };
 
+/**
+ * struct binder_fd_object - describes a filedescriptor to be fixed up.
+ * @hdr:	common header structure
+ * @pad_flags:	padding to remain compatible with old userspace code
+ * @pad_binder:	padding to remain compatible with old userspace code
+ * @fd:		file descriptor
+ * @cookie:	opaque data, used by user-space
+ */
+struct binder_fd_object {
+	struct binder_object_header	hdr;
+	__u32				pad_flags;
+	union {
+		binder_uintptr_t	pad_binder;
+		__u32			fd;
+	};
+
+	binder_uintptr_t		cookie;
+};
+
+/* struct binder_buffer_object - object describing a userspace buffer
+ * @hdr:		common header structure
+ * @flags:		one or more BINDER_BUFFER_* flags
+ * @buffer:		address of the buffer
+ * @length:		length of the buffer
+ * @parent:		index in offset array pointing to parent buffer
+ * @parent_offset:	offset in @parent pointing to this buffer
+ *
+ * A binder_buffer object represents an object that the
+ * binder kernel driver can copy verbatim to the target
+ * address space. A buffer itself may be pointed to from
+ * within another buffer, meaning that the pointer inside
+ * that other buffer needs to be fixed up as well. This
+ * can be done by setting the BINDER_BUFFER_FLAG_HAS_PARENT
+ * flag in @flags, by setting @parent buffer to the index
+ * in the offset array pointing to the parent binder_buffer_object,
+ * and by setting @parent_offset to the offset in the parent buffer
+ * at which the pointer to this buffer is located.
+ */
+struct binder_buffer_object {
+	struct binder_object_header	hdr;
+	__u32				flags;
+	binder_uintptr_t		buffer;
+	binder_size_t			length;
+	binder_size_t			parent;
+	binder_size_t			parent_offset;
+};
+
+enum {
+	BINDER_BUFFER_FLAG_HAS_PARENT = 0x01,
+};
+
+/* struct binder_fd_array_object - object describing an array of fds in a buffer
+ * @hdr:		common header structure
+ * @pad:		padding to ensure correct alignment
+ * @num_fds:		number of file descriptors in the buffer
+ * @parent:		index in offset array to buffer holding the fd array
+ * @parent_offset:	start offset of fd array in the buffer
+ *
+ * A binder_fd_array object represents an array of file
+ * descriptors embedded in a binder_buffer_object. It is
+ * different from a regular binder_buffer_object because it
+ * describes a list of file descriptors to fix up, not an opaque
+ * blob of memory, and hence the kernel needs to treat it differently.
+ *
+ * An example of how this would be used is with Android's
+ * native_handle_t object, which is a struct with a list of integers
+ * and a list of file descriptors. The native_handle_t struct itself
+ * will be represented by a struct binder_buffer_objct, whereas the
+ * embedded list of file descriptors is represented by a
+ * struct binder_fd_array_object with that binder_buffer_object as
+ * a parent.
+ */
+struct binder_fd_array_object {
+	struct binder_object_header	hdr;
+	__u32				pad;
+	binder_size_t			num_fds;
+	binder_size_t			parent;
+	binder_size_t			parent_offset;
+};
+
 /*
  * On 64-bit platforms where user code may run in 32-bits the driver must
  * translate the buffer (and local binder) addresses appropriately.
@@ -97,6 +233,19 @@ struct binder_version {
 #define BINDER_CURRENT_PROTOCOL_VERSION 8
 #endif
 
+/*
+ * Use with BINDER_GET_NODE_DEBUG_INFO, driver reads ptr, writes to all fields.
+ * Set ptr to NULL for the first call to get the info for the first node, and
+ * then repeat the call passing the previously returned value to get the next
+ * nodes.  ptr will be 0 when there are no more nodes.
+ */
+struct binder_node_debug_info {
+	binder_uintptr_t ptr;
+	binder_uintptr_t cookie;
+	__u32            has_strong_ref;
+	__u32            has_weak_ref;
+};
+
 #define BINDER_WRITE_READ		_IOWR('b', 1, struct binder_write_read)
 #define BINDER_SET_IDLE_TIMEOUT		_IOW('b', 3, __s64)
 #define BINDER_SET_MAX_THREADS		_IOW('b', 5, __u32)
@@ -104,6 +253,7 @@ struct binder_version {
 #define BINDER_SET_CONTEXT_MGR		_IOW('b', 7, __s32)
 #define BINDER_THREAD_EXIT		_IOW('b', 8, __s32)
 #define BINDER_VERSION			_IOWR('b', 9, struct binder_version)
+#define BINDER_GET_NODE_DEBUG_INFO	_IOWR('b', 11, struct binder_node_debug_info)
 
 /*
  * NOTE: Two special error codes you should check for when calling
@@ -162,6 +312,11 @@ struct binder_transaction_data {
 	} data;
 };
 
+struct binder_transaction_data_sg {
+	struct binder_transaction_data transaction_data;
+	binder_size_t buffers_size;
+};
+
 struct binder_ptr_cookie {
 	binder_uintptr_t ptr;
 	binder_uintptr_t cookie;
@@ -346,6 +501,12 @@ enum binder_driver_command_protocol {
 	/*
 	 * void *: cookie
 	 */
+
+	BC_TRANSACTION_SG = _IOW('c', 17, struct binder_transaction_data_sg),
+	BC_REPLY_SG = _IOW('c', 18, struct binder_transaction_data_sg),
+	/*
+	 * binder_transaction_data_sg: the sent command.
+	 */
 };
 
 #endif /* _UAPI_LINUX_BINDER_H */
diff --git a/include/uapi/linux/fib_rules.h b/include/uapi/linux/fib_rules.h
index 96161b8202b5..bbf02a63a011 100644
--- a/include/uapi/linux/fib_rules.h
+++ b/include/uapi/linux/fib_rules.h
@@ -29,6 +29,11 @@ struct fib_rule_hdr {
 	__u32		flags;
 };
 
+struct fib_rule_uid_range {
+	__u32		start;
+	__u32		end;
+};
+
 enum {
 	FRA_UNSPEC,
 	FRA_DST,	/* destination address */
@@ -49,6 +54,9 @@ enum {
 	FRA_TABLE,	/* Extended table id */
 	FRA_FWMASK,	/* mask for netfilter mark */
 	FRA_OIFNAME,
+	FRA_PAD,
+	FRA_L3MDEV,	/* iif or oif is l3mdev goto its table */
+	FRA_UID_RANGE,	/* UID range */
 	__FRA_MAX
 };
 
diff --git a/include/uapi/linux/fs.h b/include/uapi/linux/fs.h
index f15d980249b5..60d27496c328 100644
--- a/include/uapi/linux/fs.h
+++ b/include/uapi/linux/fs.h
@@ -160,6 +160,8 @@ struct inodes_stat_t {
 #define FITHAW		_IOWR('X', 120, int)	/* Thaw */
 #define FITRIM		_IOWR('X', 121, struct fstrim_range)	/* Trim */
 
+#define FIDTRIM	_IOWR('f', 128, struct fstrim_range)	/* Deep discard trim */
+
 #define	FS_IOC_GETFLAGS			_IOR('f', 1, long)
 #define	FS_IOC_SETFLAGS			_IOW('f', 2, long)
 #define	FS_IOC_GETVERSION		_IOR('v', 1, long)
@@ -171,6 +173,54 @@ struct inodes_stat_t {
 #define FS_IOC32_SETVERSION		_IOW('v', 2, int)
 
 /*
+ * File system encryption support
+ */
+/* Policy provided via an ioctl on the topmost directory */
+#define FS_KEY_DESCRIPTOR_SIZE	8
+
+#define FS_POLICY_FLAGS_PAD_4		0x00
+#define FS_POLICY_FLAGS_PAD_8		0x01
+#define FS_POLICY_FLAGS_PAD_16		0x02
+#define FS_POLICY_FLAGS_PAD_32		0x03
+#define FS_POLICY_FLAGS_PAD_MASK	0x03
+#define FS_POLICY_FLAGS_VALID		0x03
+
+/* Encryption algorithms */
+#define FS_ENCRYPTION_MODE_INVALID		0
+#define FS_ENCRYPTION_MODE_AES_256_XTS		1
+#define FS_ENCRYPTION_MODE_AES_256_GCM		2
+#define FS_ENCRYPTION_MODE_AES_256_CBC		3
+#define FS_ENCRYPTION_MODE_AES_256_CTS		4
+#define FS_ENCRYPTION_MODE_AES_128_CBC		5
+#define FS_ENCRYPTION_MODE_AES_128_CTS		6
+
+
+struct fscrypt_policy {
+	__u8 version;
+	__u8 contents_encryption_mode;
+	__u8 filenames_encryption_mode;
+	__u8 flags;
+	__u8 master_key_descriptor[FS_KEY_DESCRIPTOR_SIZE];
+} __packed;
+
+#define FS_IOC_SET_ENCRYPTION_POLICY	_IOR('f', 19, struct fscrypt_policy)
+#define FS_IOC_GET_ENCRYPTION_PWSALT	_IOW('f', 20, __u8[16])
+#define FS_IOC_GET_ENCRYPTION_POLICY	_IOW('f', 21, struct fscrypt_policy)
+
+/* Parameters for passing an encryption key into the kernel keyring */
+#define FS_KEY_DESC_PREFIX		"fscrypt:"
+#define FS_KEY_DESC_PREFIX_SIZE		8
+
+/* Structure that userspace passes to the kernel keyring */
+#define FS_MAX_KEY_SIZE			64
+
+struct fscrypt_key {
+	__u32 mode;
+	__u8 raw[FS_MAX_KEY_SIZE];
+	__u32 size;
+};
+
+/*
  * Inode flags (FS_IOC_GETFLAGS / FS_IOC_SETFLAGS)
  */
 #define	FS_SECRM_FL			0x00000001 /* Secure deletion */
diff --git a/include/uapi/linux/fuse.h b/include/uapi/linux/fuse.h
index c9aca042e61d..8485dd4300b8 100644
--- a/include/uapi/linux/fuse.h
+++ b/include/uapi/linux/fuse.h
@@ -358,6 +358,7 @@ enum fuse_opcode {
 	FUSE_FALLOCATE     = 43,
 	FUSE_READDIRPLUS   = 44,
 	FUSE_RENAME2       = 45,
+	FUSE_CANONICAL_PATH= 2016,
 
 	/* CUSE specific operations */
 	CUSE_INIT          = 4096,
diff --git a/include/uapi/linux/hw_breakpoint.h b/include/uapi/linux/hw_breakpoint.h
index b04000a2296a..2b65efd19a46 100644
--- a/include/uapi/linux/hw_breakpoint.h
+++ b/include/uapi/linux/hw_breakpoint.h
@@ -4,7 +4,11 @@
 enum {
 	HW_BREAKPOINT_LEN_1 = 1,
 	HW_BREAKPOINT_LEN_2 = 2,
+	HW_BREAKPOINT_LEN_3 = 3,
 	HW_BREAKPOINT_LEN_4 = 4,
+	HW_BREAKPOINT_LEN_5 = 5,
+	HW_BREAKPOINT_LEN_6 = 6,
+	HW_BREAKPOINT_LEN_7 = 7,
 	HW_BREAKPOINT_LEN_8 = 8,
 };
 
diff --git a/include/uapi/linux/if_pppolac.h b/include/uapi/linux/if_pppolac.h
new file mode 100644
index 000000000000..b7eb8153ef66
--- /dev/null
+++ b/include/uapi/linux/if_pppolac.h
@@ -0,0 +1,33 @@
+/* include/uapi/linux/if_pppolac.h
+ *
+ * Header for PPP on L2TP Access Concentrator / PPPoLAC Socket (RFC 2661)
+ *
+ * Copyright (C) 2009 Google, Inc.
+ * Author: Chia-chi Yeh <chiachi@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _UAPI_LINUX_IF_PPPOLAC_H
+#define _UAPI_LINUX_IF_PPPOLAC_H
+
+#include <linux/socket.h>
+#include <linux/types.h>
+
+struct sockaddr_pppolac {
+	sa_family_t	sa_family;	/* AF_PPPOX */
+	unsigned int	sa_protocol;	/* PX_PROTO_OLAC */
+	int		udp_socket;
+	struct __attribute__((packed)) {
+		__u16	tunnel, session;
+	} local, remote;
+} __attribute__((packed));
+
+#endif /* _UAPI_LINUX_IF_PPPOLAC_H */
diff --git a/include/uapi/linux/if_pppopns.h b/include/uapi/linux/if_pppopns.h
new file mode 100644
index 000000000000..a392b52ea6ec
--- /dev/null
+++ b/include/uapi/linux/if_pppopns.h
@@ -0,0 +1,32 @@
+/* include/uapi/linux/if_pppopns.h
+ *
+ * Header for PPP on PPTP Network Server / PPPoPNS Socket (RFC 2637)
+ *
+ * Copyright (C) 2009 Google, Inc.
+ * Author: Chia-chi Yeh <chiachi@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef _UAPI_LINUX_IF_PPPOPNS_H
+#define _UAPI_LINUX_IF_PPPOPNS_H
+
+#include <linux/socket.h>
+#include <linux/types.h>
+
+struct sockaddr_pppopns {
+	sa_family_t	sa_family;	/* AF_PPPOX */
+	unsigned int	sa_protocol;	/* PX_PROTO_OPNS */
+	int		tcp_socket;
+	__u16		local;
+	__u16		remote;
+} __attribute__((packed));
+
+#endif /* _UAPI_LINUX_IF_PPPOPNS_H */
diff --git a/include/uapi/linux/if_pppox.h b/include/uapi/linux/if_pppox.h
index e128769331b5..5861d45bc51d 100644
--- a/include/uapi/linux/if_pppox.h
+++ b/include/uapi/linux/if_pppox.h
@@ -23,6 +23,8 @@
 #include <linux/socket.h>
 #include <linux/if_ether.h>
 #include <linux/if_pppol2tp.h>
+#include <linux/if_pppolac.h>
+#include <linux/if_pppopns.h>
 
 /* For user-space programs to pick up these definitions
  * which they wouldn't get otherwise without defining __KERNEL__
@@ -56,7 +58,9 @@ struct pptp_addr {
 #define PX_PROTO_OE    0 /* Currently just PPPoE */
 #define PX_PROTO_OL2TP 1 /* Now L2TP also */
 #define PX_PROTO_PPTP  2
-#define PX_MAX_PROTO   3
+#define PX_PROTO_OLAC  3
+#define PX_PROTO_OPNS  4
+#define PX_MAX_PROTO   5
 
 struct sockaddr_pppox {
 	__kernel_sa_family_t sa_family;       /* address family, AF_PPPOX */
diff --git a/include/uapi/linux/inet_diag.h b/include/uapi/linux/inet_diag.h
index 68a1f71fde9f..c7f189bd5979 100644
--- a/include/uapi/linux/inet_diag.h
+++ b/include/uapi/linux/inet_diag.h
@@ -72,6 +72,8 @@ enum {
 	INET_DIAG_BC_AUTO,
 	INET_DIAG_BC_S_COND,
 	INET_DIAG_BC_D_COND,
+	INET_DIAG_BC_DEV_COND,   /* u32 ifindex */
+	INET_DIAG_BC_MARK_COND,
 };
 
 struct inet_diag_hostcond {
@@ -81,6 +83,11 @@ struct inet_diag_hostcond {
 	__be32	addr[0];
 };
 
+struct inet_diag_markcond {
+	__u32 mark;
+	__u32 mask;
+};
+
 /* Base info structure. It contains socket identity (addrs/ports/cookie)
  * and, alas, the information shown by netstat. */
 struct inet_diag_msg {
@@ -113,9 +120,13 @@ enum {
 	INET_DIAG_DCTCPINFO,
 	INET_DIAG_PROTOCOL,  /* response attribute only */
 	INET_DIAG_SKV6ONLY,
+	INET_DIAG_LOCALS,
+	INET_DIAG_PEERS,
+	INET_DIAG_PAD,
+	INET_DIAG_MARK,
 };
 
-#define INET_DIAG_MAX INET_DIAG_SKV6ONLY
+#define INET_DIAG_MAX INET_DIAG_MARK
 
 /* INET_DIAG_MEM */
 
diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h
index 38b4fef20219..c462f1dc175e 100644
--- a/include/uapi/linux/ipv6.h
+++ b/include/uapi/linux/ipv6.h
@@ -164,6 +164,7 @@ enum {
 	DEVCONF_ACCEPT_DAD,
 	DEVCONF_FORCE_TLLAO,
 	DEVCONF_NDISC_NOTIFY,
+	DEVCONF_ACCEPT_RA_RT_TABLE,
 	DEVCONF_MLDV1_UNSOLICITED_REPORT_INTERVAL,
 	DEVCONF_MLDV2_UNSOLICITED_REPORT_INTERVAL,
 	DEVCONF_SUPPRESS_FRAG_NDISC,
@@ -174,6 +175,16 @@ enum {
 	DEVCONF_USE_OIF_ADDRS_ONLY,
 	DEVCONF_ACCEPT_RA_MIN_HOP_LIMIT,
 	DEVCONF_IGNORE_ROUTES_WITH_LINKDOWN,
+	DEVCONF_DROP_UNICAST_IN_L2_MULTICAST,
+	DEVCONF_DROP_UNSOLICITED_NA,
+	DEVCONF_KEEP_ADDR_ON_DOWN,
+	DEVCONF_RTR_SOLICIT_MAX_INTERVAL,
+	DEVCONF_SEG6_ENABLED,
+	DEVCONF_SEG6_REQUIRE_HMAC,
+	DEVCONF_ENHANCED_DAD,
+	DEVCONF_ADDR_GEN_MODE,
+	DEVCONF_DISABLE_POLICY,
+	DEVCONF_ACCEPT_RA_RT_INFO_MIN_PLEN,
 	DEVCONF_MAX
 };
 
diff --git a/include/uapi/linux/keychord.h b/include/uapi/linux/keychord.h
new file mode 100644
index 000000000000..ea7cf4d27bbd
--- /dev/null
+++ b/include/uapi/linux/keychord.h
@@ -0,0 +1,52 @@
+/*
+ *  Key chord input driver
+ *
+ * Copyright (C) 2008 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+*/
+
+#ifndef _UAPI_LINUX_KEYCHORD_H_
+#define _UAPI_LINUX_KEYCHORD_H_
+
+#include <linux/input.h>
+
+#define KEYCHORD_VERSION		1
+
+/*
+ * One or more input_keychord structs are written to /dev/keychord
+ * at once to specify the list of keychords to monitor.
+ * Reading /dev/keychord returns the id of a keychord when the
+ * keychord combination is pressed.  A keychord is signalled when
+ * all of the keys in the keycode list are in the pressed state.
+ * The order in which the keys are pressed does not matter.
+ * The keychord will not be signalled if keys not in the keycode
+ * list are pressed.
+ * Keychords will not be signalled on key release events.
+ */
+struct input_keychord {
+	/* should be KEYCHORD_VERSION */
+	__u16 version;
+	/*
+	 * client specified ID, returned from read()
+	 * when this keychord is pressed.
+	 */
+	__u16 id;
+
+	/* number of keycodes in this keychord */
+	__u16 count;
+
+	/* variable length array of keycodes */
+	__u16 keycodes[];
+};
+
+#endif	/* _UAPI_LINUX_KEYCHORD_H_ */
diff --git a/include/uapi/linux/magic.h b/include/uapi/linux/magic.h
index b283d56c1db9..509897624fbc 100644
--- a/include/uapi/linux/magic.h
+++ b/include/uapi/linux/magic.h
@@ -52,6 +52,8 @@
 #define REISER2FS_SUPER_MAGIC_STRING	"ReIsEr2Fs"
 #define REISER2FS_JR_SUPER_MAGIC_STRING	"ReIsEr3Fs"
 
+#define SDCARDFS_SUPER_MAGIC	0x5dca2df5
+
 #define SMB_SUPER_MAGIC		0x517B
 #define CGROUP_SUPER_MAGIC	0x27e0eb
 #define CGROUP2_SUPER_MAGIC	0x63677270
diff --git a/include/uapi/linux/netfilter/xt_IDLETIMER.h b/include/uapi/linux/netfilter/xt_IDLETIMER.h
index 208ae9387331..faaa28b3d061 100644
--- a/include/uapi/linux/netfilter/xt_IDLETIMER.h
+++ b/include/uapi/linux/netfilter/xt_IDLETIMER.h
@@ -4,6 +4,7 @@
  * Header file for Xtables timer target module.
  *
  * Copyright (C) 2004, 2010 Nokia Corporation
+ *
  * Written by Timo Teras <ext-timo.teras@nokia.com>
  *
  * Converted to x_tables and forward-ported to 2.6.34
@@ -32,12 +33,19 @@
 #include <linux/types.h>
 
 #define MAX_IDLETIMER_LABEL_SIZE 28
+#define NLMSG_MAX_SIZE 64
+
+#define NL_EVENT_TYPE_INACTIVE 0
+#define NL_EVENT_TYPE_ACTIVE 1
 
 struct idletimer_tg_info {
 	__u32 timeout;
 
 	char label[MAX_IDLETIMER_LABEL_SIZE];
 
+	/* Use netlink messages for notification in addition to sysfs */
+	__u8 send_nl_msg;
+
 	/* for kernel module internal use only */
 	struct idletimer_tg *timer __attribute__((aligned(8)));
 };
diff --git a/include/uapi/linux/netfilter/xt_socket.h b/include/uapi/linux/netfilter/xt_socket.h
index 87644f832494..7f00df6cd897 100644
--- a/include/uapi/linux/netfilter/xt_socket.h
+++ b/include/uapi/linux/netfilter/xt_socket.h
@@ -26,4 +26,11 @@ struct xt_socket_mtinfo3 {
 			   | XT_SOCKET_NOWILDCARD \
 			   | XT_SOCKET_RESTORESKMARK)
 
+struct sock *xt_socket_lookup_slow_v4(struct net *net,
+				      const struct sk_buff *skb,
+				      const struct net_device *indev);
+struct sock *xt_socket_lookup_slow_v6(struct net *net,
+				      const struct sk_buff *skb,
+				      const struct net_device *indev);
+
 #endif /* _XT_SOCKET_H */
diff --git a/include/uapi/linux/prctl.h b/include/uapi/linux/prctl.h
index a8d0759a9e40..c1af9b3c27c4 100644
--- a/include/uapi/linux/prctl.h
+++ b/include/uapi/linux/prctl.h
@@ -197,4 +197,13 @@ struct prctl_mm_map {
 # define PR_CAP_AMBIENT_LOWER		3
 # define PR_CAP_AMBIENT_CLEAR_ALL	4
 
+/* Sets the timerslack for arbitrary threads
+ * arg2 slack value, 0 means "use default"
+ * arg3 pid of the thread whose timer slack needs to be set
+ */
+#define PR_SET_TIMERSLACK_PID	127
+
+#define PR_SET_VMA		0x53564d41
+# define PR_SET_VMA_ANON_NAME		0
+
 #endif /* _LINUX_PRCTL_H */
diff --git a/include/uapi/linux/rtnetlink.h b/include/uapi/linux/rtnetlink.h
index fa3b34365560..d7d7599b112a 100644
--- a/include/uapi/linux/rtnetlink.h
+++ b/include/uapi/linux/rtnetlink.h
@@ -311,6 +311,9 @@ enum rtattr_type_t {
 	RTA_PREF,
 	RTA_ENCAP_TYPE,
 	RTA_ENCAP,
+	RTA_EXPIRES,
+	RTA_PAD,
+	RTA_UID,
 	__RTA_MAX
 };
 
diff --git a/include/uapi/linux/sock_diag.h b/include/uapi/linux/sock_diag.h
index 49230d36f9ce..84e66ed670be 100644
--- a/include/uapi/linux/sock_diag.h
+++ b/include/uapi/linux/sock_diag.h
@@ -4,6 +4,7 @@
 #include <linux/types.h>
 
 #define SOCK_DIAG_BY_FAMILY 20
+#define SOCK_DESTROY_BACKPORT 21
 
 struct sock_diag_req {
 	__u8	sdiag_family;
diff --git a/include/uapi/linux/sysctl.h b/include/uapi/linux/sysctl.h
index 0956373b56db..d18980e74534 100644
--- a/include/uapi/linux/sysctl.h
+++ b/include/uapi/linux/sysctl.h
@@ -570,6 +570,7 @@ enum {
 	NET_IPV6_PROXY_NDP=23,
 	NET_IPV6_ACCEPT_SOURCE_ROUTE=25,
 	NET_IPV6_ACCEPT_RA_FROM_LOCAL=26,
+	NET_IPV6_ACCEPT_RA_RT_INFO_MIN_PLEN=27,
 	__NET_IPV6_MAX
 };
 
diff --git a/include/uapi/linux/usb/f_accessory.h b/include/uapi/linux/usb/f_accessory.h
new file mode 100644
index 000000000000..0baeb7d0d74c
--- /dev/null
+++ b/include/uapi/linux/usb/f_accessory.h
@@ -0,0 +1,146 @@
+/*
+ * Gadget Function Driver for Android USB accessories
+ *
+ * Copyright (C) 2011 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _UAPI_LINUX_USB_F_ACCESSORY_H
+#define _UAPI_LINUX_USB_F_ACCESSORY_H
+
+/* Use Google Vendor ID when in accessory mode */
+#define USB_ACCESSORY_VENDOR_ID 0x18D1
+
+
+/* Product ID to use when in accessory mode */
+#define USB_ACCESSORY_PRODUCT_ID 0x2D00
+
+/* Product ID to use when in accessory mode and adb is enabled */
+#define USB_ACCESSORY_ADB_PRODUCT_ID 0x2D01
+
+/* Indexes for strings sent by the host via ACCESSORY_SEND_STRING */
+#define ACCESSORY_STRING_MANUFACTURER   0
+#define ACCESSORY_STRING_MODEL          1
+#define ACCESSORY_STRING_DESCRIPTION    2
+#define ACCESSORY_STRING_VERSION        3
+#define ACCESSORY_STRING_URI            4
+#define ACCESSORY_STRING_SERIAL         5
+
+/* Control request for retrieving device's protocol version
+ *
+ *	requestType:    USB_DIR_IN | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_GET_PROTOCOL
+ *	value:          0
+ *	index:          0
+ *	data            version number (16 bits little endian)
+ *                     1 for original accessory support
+ *                     2 adds HID and device to host audio support
+ */
+#define ACCESSORY_GET_PROTOCOL  51
+
+/* Control request for host to send a string to the device
+ *
+ *	requestType:    USB_DIR_OUT | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_SEND_STRING
+ *	value:          0
+ *	index:          string ID
+ *	data            zero terminated UTF8 string
+ *
+ *  The device can later retrieve these strings via the
+ *  ACCESSORY_GET_STRING_* ioctls
+ */
+#define ACCESSORY_SEND_STRING   52
+
+/* Control request for starting device in accessory mode.
+ * The host sends this after setting all its strings to the device.
+ *
+ *	requestType:    USB_DIR_OUT | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_START
+ *	value:          0
+ *	index:          0
+ *	data            none
+ */
+#define ACCESSORY_START         53
+
+/* Control request for registering a HID device.
+ * Upon registering, a unique ID is sent by the accessory in the
+ * value parameter. This ID will be used for future commands for
+ * the device
+ *
+ *	requestType:    USB_DIR_OUT | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_REGISTER_HID_DEVICE
+ *	value:          Accessory assigned ID for the HID device
+ *	index:          total length of the HID report descriptor
+ *	data            none
+ */
+#define ACCESSORY_REGISTER_HID         54
+
+/* Control request for unregistering a HID device.
+ *
+ *	requestType:    USB_DIR_OUT | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_REGISTER_HID
+ *	value:          Accessory assigned ID for the HID device
+ *	index:          0
+ *	data            none
+ */
+#define ACCESSORY_UNREGISTER_HID         55
+
+/* Control request for sending the HID report descriptor.
+ * If the HID descriptor is longer than the endpoint zero max packet size,
+ * the descriptor will be sent in multiple ACCESSORY_SET_HID_REPORT_DESC
+ * commands. The data for the descriptor must be sent sequentially
+ * if multiple packets are needed.
+ *
+ *	requestType:    USB_DIR_OUT | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_SET_HID_REPORT_DESC
+ *	value:          Accessory assigned ID for the HID device
+ *	index:          offset of data in descriptor
+ *                      (needed when HID descriptor is too big for one packet)
+ *	data            the HID report descriptor
+ */
+#define ACCESSORY_SET_HID_REPORT_DESC         56
+
+/* Control request for sending HID events.
+ *
+ *	requestType:    USB_DIR_OUT | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_SEND_HID_EVENT
+ *	value:          Accessory assigned ID for the HID device
+ *	index:          0
+ *	data            the HID report for the event
+ */
+#define ACCESSORY_SEND_HID_EVENT         57
+
+/* Control request for setting the audio mode.
+ *
+ *	requestType:	USB_DIR_OUT | USB_TYPE_VENDOR
+ *	request:        ACCESSORY_SET_AUDIO_MODE
+ *	value:          0 - no audio
+ *                     1 - device to host, 44100 16-bit stereo PCM
+ *	index:          0
+ *	data            none
+ */
+#define ACCESSORY_SET_AUDIO_MODE         58
+
+/* ioctls for retrieving strings set by the host */
+#define ACCESSORY_GET_STRING_MANUFACTURER   _IOW('M', 1, char[256])
+#define ACCESSORY_GET_STRING_MODEL          _IOW('M', 2, char[256])
+#define ACCESSORY_GET_STRING_DESCRIPTION    _IOW('M', 3, char[256])
+#define ACCESSORY_GET_STRING_VERSION        _IOW('M', 4, char[256])
+#define ACCESSORY_GET_STRING_URI            _IOW('M', 5, char[256])
+#define ACCESSORY_GET_STRING_SERIAL         _IOW('M', 6, char[256])
+/* returns 1 if there is a start request pending */
+#define ACCESSORY_IS_START_REQUESTED        _IO('M', 7)
+/* returns audio mode (set via the ACCESSORY_SET_AUDIO_MODE control request) */
+#define ACCESSORY_GET_AUDIO_MODE            _IO('M', 8)
+
+#endif /* _UAPI_LINUX_USB_F_ACCESSORY_H */
diff --git a/include/uapi/linux/usb/f_mtp.h b/include/uapi/linux/usb/f_mtp.h
new file mode 100644
index 000000000000..503291855abd
--- /dev/null
+++ b/include/uapi/linux/usb/f_mtp.h
@@ -0,0 +1,61 @@
+/*
+ * Gadget Function Driver for MTP
+ *
+ * Copyright (C) 2010 Google, Inc.
+ * Author: Mike Lockwood <lockwood@android.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _UAPI_LINUX_USB_F_MTP_H
+#define _UAPI_LINUX_USB_F_MTP_H
+
+#include <linux/ioctl.h>
+#include <linux/types.h>
+
+struct mtp_file_range {
+	/* file descriptor for file to transfer */
+	int			fd;
+	/* offset in file for start of transfer */
+	loff_t		offset;
+	/* number of bytes to transfer */
+	int64_t		length;
+	/* MTP command ID for data header,
+	 * used only for MTP_SEND_FILE_WITH_HEADER
+	 */
+	uint16_t	command;
+	/* MTP transaction ID for data header,
+	 * used only for MTP_SEND_FILE_WITH_HEADER
+	 */
+	uint32_t	transaction_id;
+};
+
+struct mtp_event {
+	/* size of the event */
+	size_t		length;
+	/* event data to send */
+	void		*data;
+};
+
+/* Sends the specified file range to the host */
+#define MTP_SEND_FILE              _IOW('M', 0, struct mtp_file_range)
+/* Receives data from the host and writes it to a file.
+ * The file is created if it does not exist.
+ */
+#define MTP_RECEIVE_FILE           _IOW('M', 1, struct mtp_file_range)
+/* Sends an event to the host via the interrupt endpoint */
+#define MTP_SEND_EVENT             _IOW('M', 3, struct mtp_event)
+/* Sends the specified file range to the host,
+ * with a 12 byte MTP data packet header at the beginning.
+ */
+#define MTP_SEND_FILE_WITH_HEADER  _IOW('M', 4, struct mtp_file_range)
+
+#endif /* _UAPI_LINUX_USB_F_MTP_H */
diff --git a/include/uapi/linux/xfrm.h b/include/uapi/linux/xfrm.h
index 2cd9e608d0d1..8cb0f24a33f0 100644
--- a/include/uapi/linux/xfrm.h
+++ b/include/uapi/linux/xfrm.h
@@ -302,6 +302,9 @@ enum xfrm_attr_type_t {
 	XFRMA_SA_EXTRA_FLAGS,	/* __u32 */
 	XFRMA_PROTO,		/* __u8 */
 	XFRMA_ADDRESS_FILTER,	/* struct xfrm_address_filter */
+	XFRMA_PAD,
+	XFRMA_OFFLOAD_DEV,	/* struct xfrm_state_offload */
+	XFRMA_OUTPUT_MARK,	/* __u32 */
 	__XFRMA_MAX
 
 #define XFRMA_MAX (__XFRMA_MAX - 1)
diff --git a/include/uapi/video/adf.h b/include/uapi/video/adf.h
new file mode 100644
index 000000000000..c5d2e62cdb9b
--- /dev/null
+++ b/include/uapi/video/adf.h
@@ -0,0 +1,321 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _UAPI_VIDEO_ADF_H_
+#define _UAPI_VIDEO_ADF_H_
+
+#include <linux/ioctl.h>
+#include <linux/types.h>
+
+#include <drm/drm_fourcc.h>
+#include <drm/drm_mode.h>
+
+#define ADF_NAME_LEN 32
+#define ADF_MAX_CUSTOM_DATA_SIZE 4096
+
+enum adf_interface_type {
+	ADF_INTF_DSI = 0,
+	ADF_INTF_eDP = 1,
+	ADF_INTF_DPI = 2,
+	ADF_INTF_VGA = 3,
+	ADF_INTF_DVI = 4,
+	ADF_INTF_HDMI = 5,
+	ADF_INTF_MEMORY = 6,
+	ADF_INTF_TYPE_DEVICE_CUSTOM = 128,
+	ADF_INTF_TYPE_MAX = (~(__u32)0),
+};
+
+#define ADF_INTF_FLAG_PRIMARY (1 << 0)
+#define ADF_INTF_FLAG_EXTERNAL (1 << 1)
+
+enum adf_event_type {
+	ADF_EVENT_VSYNC = 0,
+	ADF_EVENT_HOTPLUG = 1,
+	ADF_EVENT_DEVICE_CUSTOM = 128,
+	ADF_EVENT_TYPE_MAX = 255,
+};
+
+/**
+ * struct adf_set_event - start or stop subscribing to ADF events
+ *
+ * @type: the type of event to (un)subscribe
+ * @enabled: subscribe or unsubscribe
+ *
+ * After subscribing to an event, userspace may poll() the ADF object's fd
+ * to wait for events or read() to consume the event's data.
+ *
+ * ADF reserves event types 0 to %ADF_EVENT_DEVICE_CUSTOM-1 for its own events.
+ * Devices may use event types %ADF_EVENT_DEVICE_CUSTOM to %ADF_EVENT_TYPE_MAX-1
+ * for driver-private events.
+ */
+struct adf_set_event {
+	__u8 type;
+	__u8 enabled;
+};
+
+/**
+ * struct adf_event - common header for ADF event data
+ *
+ * @type: event type
+ * @length: total size of event data, header inclusive
+ */
+struct adf_event {
+	__u8 type;
+	__u32 length;
+};
+
+/**
+ * struct adf_vsync_event - ADF vsync event
+ *
+ * @base: event header (see &struct adf_event)
+ * @timestamp: time of vsync event, in nanoseconds
+ */
+struct adf_vsync_event {
+	struct adf_event base;
+	__aligned_u64 timestamp;
+};
+
+/**
+ * struct adf_vsync_event - ADF display hotplug event
+ *
+ * @base: event header (see &struct adf_event)
+ * @connected: whether a display is now connected to the interface
+ */
+struct adf_hotplug_event {
+	struct adf_event base;
+	__u8 connected;
+};
+
+#define ADF_MAX_PLANES 4
+/**
+ * struct adf_buffer_config - description of buffer displayed by adf_post_config
+ *
+ * @overlay_engine: id of the target overlay engine
+ * @w: width of display region in pixels
+ * @h: height of display region in pixels
+ * @format: DRM-style fourcc, see drm_fourcc.h for standard formats
+ * @fd: dma_buf fd for each plane
+ * @offset: location of first pixel to scan out, in bytes
+ * @pitch: stride (i.e. length of a scanline including padding) in bytes
+ * @n_planes: number of planes in buffer
+ * @acquire_fence: sync_fence fd which will clear when the buffer is
+ *	ready for display, or <0 if the buffer is already ready
+ */
+struct adf_buffer_config {
+	__u32 overlay_engine;
+
+	__u32 w;
+	__u32 h;
+	__u32 format;
+
+	__s32 fd[ADF_MAX_PLANES];
+	__u32 offset[ADF_MAX_PLANES];
+	__u32 pitch[ADF_MAX_PLANES];
+	__u8 n_planes;
+
+	__s32 acquire_fence;
+};
+#define ADF_MAX_BUFFERS (4096 / sizeof(struct adf_buffer_config))
+
+/**
+ * struct adf_post_config - request to flip to a new set of buffers
+ *
+ * @n_interfaces: number of interfaces targeted by the flip (input)
+ * @interfaces: ids of interfaces targeted by the flip (input)
+ * @n_bufs: number of buffers displayed (input)
+ * @bufs: description of buffers displayed (input)
+ * @custom_data_size: size of driver-private data (input)
+ * @custom_data: driver-private data (input)
+ * @complete_fence: sync_fence fd which will clear when this
+ *	configuration has left the screen (output)
+ */
+struct adf_post_config {
+	size_t n_interfaces;
+	__u32 __user *interfaces;
+
+	size_t n_bufs;
+	struct adf_buffer_config __user *bufs;
+
+	size_t custom_data_size;
+	void __user *custom_data;
+
+	__s32 complete_fence;
+};
+#define ADF_MAX_INTERFACES (4096 / sizeof(__u32))
+
+/**
+ * struct adf_simple_buffer_allocate - request to allocate a "simple" buffer
+ *
+ * @w: width of buffer in pixels (input)
+ * @h: height of buffer in pixels (input)
+ * @format: DRM-style fourcc (input)
+ *
+ * @fd: dma_buf fd (output)
+ * @offset: location of first pixel, in bytes (output)
+ * @pitch: length of a scanline including padding, in bytes (output)
+ *
+ * Simple buffers are analogous to DRM's "dumb" buffers.  They have a single
+ * plane of linear RGB data which can be allocated and scanned out without
+ * any driver-private ioctls or data.
+ *
+ * @format must be a standard RGB format defined in drm_fourcc.h.
+ *
+ * ADF clients must NOT assume that an interface can scan out a simple buffer
+ * allocated by a different ADF interface, even if the two interfaces belong to
+ * the same ADF device.
+ */
+struct adf_simple_buffer_alloc {
+	__u16 w;
+	__u16 h;
+	__u32 format;
+
+	__s32 fd;
+	__u32 offset;
+	__u32 pitch;
+};
+
+/**
+ * struct adf_simple_post_config - request to flip to a single buffer without
+ * driver-private data
+ *
+ * @buf: description of buffer displayed (input)
+ * @complete_fence: sync_fence fd which will clear when this buffer has left the
+ * screen (output)
+ */
+struct adf_simple_post_config {
+	struct adf_buffer_config buf;
+	__s32 complete_fence;
+};
+
+/**
+ * struct adf_attachment_config - description of attachment between an overlay
+ * engine and an interface
+ *
+ * @overlay_engine: id of the overlay engine
+ * @interface: id of the interface
+ */
+struct adf_attachment_config {
+	__u32 overlay_engine;
+	__u32 interface;
+};
+
+/**
+ * struct adf_device_data - describes a display device
+ *
+ * @name: display device's name
+ * @n_attachments: the number of current attachments
+ * @attachments: list of current attachments
+ * @n_allowed_attachments: the number of allowed attachments
+ * @allowed_attachments: list of allowed attachments
+ * @custom_data_size: size of driver-private data
+ * @custom_data: driver-private data
+ */
+struct adf_device_data {
+	char name[ADF_NAME_LEN];
+
+	size_t n_attachments;
+	struct adf_attachment_config __user *attachments;
+
+	size_t n_allowed_attachments;
+	struct adf_attachment_config __user *allowed_attachments;
+
+	size_t custom_data_size;
+	void __user *custom_data;
+};
+#define ADF_MAX_ATTACHMENTS (4096 / sizeof(struct adf_attachment_config))
+
+/**
+ * struct adf_device_data - describes a display interface
+ *
+ * @name: display interface's name
+ * @type: interface type (see enum @adf_interface_type)
+ * @id: which interface of type @type;
+ *	e.g. interface DSI.1 -> @type=@ADF_INTF_TYPE_DSI, @id=1
+ * @flags: informational flags (bitmask of %ADF_INTF_FLAG_* values)
+ * @dpms_state: DPMS state (one of @DRM_MODE_DPMS_* defined in drm_mode.h)
+ * @hotplug_detect: whether a display is plugged in
+ * @width_mm: screen width in millimeters, or 0 if unknown
+ * @height_mm: screen height in millimeters, or 0 if unknown
+ * @current_mode: current display mode
+ * @n_available_modes: the number of hardware display modes
+ * @available_modes: list of hardware display modes
+ * @custom_data_size: size of driver-private data
+ * @custom_data: driver-private data
+ */
+struct adf_interface_data {
+	char name[ADF_NAME_LEN];
+
+	__u32 type;
+	__u32 id;
+	/* e.g. type=ADF_INTF_TYPE_DSI, id=1 => DSI.1 */
+	__u32 flags;
+
+	__u8 dpms_state;
+	__u8 hotplug_detect;
+	__u16 width_mm;
+	__u16 height_mm;
+
+	struct drm_mode_modeinfo current_mode;
+	size_t n_available_modes;
+	struct drm_mode_modeinfo __user *available_modes;
+
+	size_t custom_data_size;
+	void __user *custom_data;
+};
+#define ADF_MAX_MODES (4096 / sizeof(struct drm_mode_modeinfo))
+
+/**
+ * struct adf_overlay_engine_data - describes an overlay engine
+ *
+ * @name: overlay engine's name
+ * @n_supported_formats: number of supported formats
+ * @supported_formats: list of supported formats
+ * @custom_data_size: size of driver-private data
+ * @custom_data: driver-private data
+ */
+struct adf_overlay_engine_data {
+	char name[ADF_NAME_LEN];
+
+	size_t n_supported_formats;
+	__u32 __user *supported_formats;
+
+	size_t custom_data_size;
+	void __user *custom_data;
+};
+#define ADF_MAX_SUPPORTED_FORMATS (4096 / sizeof(__u32))
+
+#define ADF_IOCTL_TYPE		'D'
+#define ADF_IOCTL_NR_CUSTOM	128
+
+#define ADF_SET_EVENT		_IOW(ADF_IOCTL_TYPE, 0, struct adf_set_event)
+#define ADF_BLANK		_IOW(ADF_IOCTL_TYPE, 1, __u8)
+#define ADF_POST_CONFIG		_IOW(ADF_IOCTL_TYPE, 2, struct adf_post_config)
+#define ADF_SET_MODE		_IOW(ADF_IOCTL_TYPE, 3, \
+					struct drm_mode_modeinfo)
+#define ADF_GET_DEVICE_DATA	_IOR(ADF_IOCTL_TYPE, 4, struct adf_device_data)
+#define ADF_GET_INTERFACE_DATA	_IOR(ADF_IOCTL_TYPE, 5, \
+					struct adf_interface_data)
+#define ADF_GET_OVERLAY_ENGINE_DATA \
+				_IOR(ADF_IOCTL_TYPE, 6, \
+					struct adf_overlay_engine_data)
+#define ADF_SIMPLE_POST_CONFIG	_IOW(ADF_IOCTL_TYPE, 7, \
+					struct adf_simple_post_config)
+#define ADF_SIMPLE_BUFFER_ALLOC	_IOW(ADF_IOCTL_TYPE, 8, \
+					struct adf_simple_buffer_alloc)
+#define ADF_ATTACH		_IOW(ADF_IOCTL_TYPE, 9, \
+					struct adf_attachment_config)
+#define ADF_DETACH		_IOW(ADF_IOCTL_TYPE, 10, \
+					struct adf_attachment_config)
+
+#endif /* _UAPI_VIDEO_ADF_H_ */
diff --git a/include/video/adf.h b/include/video/adf.h
new file mode 100644
index 000000000000..34f10e538f9e
--- /dev/null
+++ b/include/video/adf.h
@@ -0,0 +1,502 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _VIDEO_ADF_H
+#define _VIDEO_ADF_H
+
+#include <linux/device.h>
+#include <linux/dma-buf.h>
+#include <linux/idr.h>
+#include <linux/kref.h>
+#include <linux/kthread.h>
+#include <linux/ktime.h>
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/scatterlist.h>
+#include <linux/sched.h>
+#include <linux/spinlock.h>
+#include <linux/wait.h>
+#include <linux/workqueue.h>
+#include <uapi/video/adf.h>
+#include "sync.h"
+
+struct adf_obj;
+struct adf_obj_ops;
+struct adf_device;
+struct adf_device_ops;
+struct adf_interface;
+struct adf_interface_ops;
+struct adf_overlay_engine;
+struct adf_overlay_engine_ops;
+
+/**
+ * struct adf_buffer - buffer displayed by adf_post
+ *
+ * @overlay_engine: target overlay engine
+ * @w: width of display region in pixels
+ * @h: height of display region in pixels
+ * @format: DRM-style fourcc, see drm_fourcc.h for standard formats
+ * @dma_bufs: dma_buf for each plane
+ * @offset: location of first pixel to scan out, in bytes
+ * @pitch: length of a scanline including padding, in bytes
+ * @n_planes: number of planes in buffer
+ * @acquire_fence: sync_fence which will clear when the buffer is
+ *	ready for display
+ *
+ * &struct adf_buffer is the in-kernel counterpart to the userspace-facing
+ * &struct adf_buffer_config.
+ */
+struct adf_buffer {
+	struct adf_overlay_engine *overlay_engine;
+
+	u32 w;
+	u32 h;
+	u32 format;
+
+	struct dma_buf *dma_bufs[ADF_MAX_PLANES];
+	u32 offset[ADF_MAX_PLANES];
+	u32 pitch[ADF_MAX_PLANES];
+	u8 n_planes;
+
+	struct sync_fence *acquire_fence;
+};
+
+/**
+ * struct adf_buffer_mapping - state for mapping a &struct adf_buffer into the
+ * display device
+ *
+ * @attachments: dma-buf attachment for each plane
+ * @sg_tables: SG tables for each plane
+ */
+struct adf_buffer_mapping {
+	struct dma_buf_attachment *attachments[ADF_MAX_PLANES];
+	struct sg_table *sg_tables[ADF_MAX_PLANES];
+};
+
+/**
+ * struct adf_post - request to flip to a new set of buffers
+ *
+ * @n_bufs: number of buffers displayed
+ * @bufs: buffers displayed
+ * @mappings: in-device mapping state for each buffer
+ * @custom_data_size: size of driver-private data
+ * @custom_data: driver-private data
+ *
+ * &struct adf_post is the in-kernel counterpart to the userspace-facing
+ * &struct adf_post_config.
+ */
+struct adf_post {
+	size_t n_bufs;
+	struct adf_buffer *bufs;
+	struct adf_buffer_mapping *mappings;
+
+	size_t custom_data_size;
+	void *custom_data;
+};
+
+/**
+ * struct adf_attachment - description of attachment between an overlay engine
+ * and an interface
+ *
+ * @overlay_engine: the overlay engine
+ * @interface: the interface
+ *
+ * &struct adf_attachment is the in-kernel counterpart to the userspace-facing
+ * &struct adf_attachment_config.
+ */
+struct adf_attachment {
+	struct adf_overlay_engine *overlay_engine;
+	struct adf_interface *interface;
+};
+
+struct adf_pending_post {
+	struct list_head head;
+	struct adf_post config;
+	void *state;
+};
+
+enum adf_obj_type {
+	ADF_OBJ_OVERLAY_ENGINE = 0,
+	ADF_OBJ_INTERFACE = 1,
+	ADF_OBJ_DEVICE = 2,
+};
+
+/**
+ * struct adf_obj_ops - common ADF object implementation ops
+ *
+ * @open: handle opening the object's device node
+ * @release: handle releasing an open file
+ * @ioctl: handle custom ioctls
+ *
+ * @supports_event: return whether the object supports generating events of type
+ *	@type
+ * @set_event: enable or disable events of type @type
+ * @event_type_str: return a string representation of custom event @type
+ *	(@type >= %ADF_EVENT_DEVICE_CUSTOM).
+ *
+ * @custom_data: copy up to %ADF_MAX_CUSTOM_DATA_SIZE bytes of driver-private
+ *	data into @data (allocated by ADF) and return the number of copied bytes
+ *	in @size.  Return 0 on success or an error code (<0) on failure.
+ */
+struct adf_obj_ops {
+	/* optional */
+	int (*open)(struct adf_obj *obj, struct inode *inode,
+			struct file *file);
+	/* optional */
+	void (*release)(struct adf_obj *obj, struct inode *inode,
+			struct file *file);
+	/* optional */
+	long (*ioctl)(struct adf_obj *obj, unsigned int cmd, unsigned long arg);
+
+	/* optional */
+	bool (*supports_event)(struct adf_obj *obj, enum adf_event_type type);
+	/* required if supports_event is implemented */
+	void (*set_event)(struct adf_obj *obj, enum adf_event_type type,
+			bool enabled);
+	/* optional */
+	const char *(*event_type_str)(struct adf_obj *obj,
+			enum adf_event_type type);
+
+	/* optional */
+	int (*custom_data)(struct adf_obj *obj, void *data, size_t *size);
+};
+
+struct adf_obj {
+	enum adf_obj_type type;
+	char name[ADF_NAME_LEN];
+	struct adf_device *parent;
+
+	const struct adf_obj_ops *ops;
+
+	struct device dev;
+
+	struct spinlock file_lock;
+	struct list_head file_list;
+
+	struct mutex event_lock;
+	struct rb_root event_refcount;
+
+	int id;
+	int minor;
+};
+
+/**
+ * struct adf_device_quirks - common display device quirks
+ *
+ * @buffer_padding: whether the last scanline of a buffer extends to the
+ * 	buffer's pitch (@ADF_BUFFER_PADDED_TO_PITCH) or just to the visible
+ * 	width (@ADF_BUFFER_UNPADDED)
+ */
+struct adf_device_quirks {
+	/* optional, defaults to ADF_BUFFER_PADDED_TO_PITCH */
+	enum {
+		ADF_BUFFER_PADDED_TO_PITCH = 0,
+		ADF_BUFFER_UNPADDED = 1,
+	} buffer_padding;
+};
+
+/**
+ * struct adf_device_ops - display device implementation ops
+ *
+ * @owner: device's module
+ * @base: common operations (see &struct adf_obj_ops)
+ * @quirks: device's quirks (see &struct adf_device_quirks)
+ *
+ * @attach: attach overlay engine @eng to interface @intf.  Return 0 on success
+ *	or error code (<0) on failure.
+ * @detach: detach overlay engine @eng from interface @intf.  Return 0 on
+ *	success or error code (<0) on failure.
+ *
+ * @validate_custom_format: validate the number and size of planes
+ *	in buffers with a custom format (i.e., not one of the @DRM_FORMAT_*
+ *	types defined in drm/drm_fourcc.h).  Return 0 if the buffer is valid or
+ *	an error code (<0) otherwise.
+ *
+ * @validate: validate that the proposed configuration @cfg is legal.  The
+ *	driver may optionally allocate and return some driver-private state in
+ *	@driver_state, which will be passed to the corresponding post().  The
+ *	driver may NOT commit any changes to hardware.  Return 0 if @cfg is
+ *	valid or an error code (<0) otherwise.
+ * @complete_fence: create a hardware-backed sync fence to be signaled when
+ *	@cfg is removed from the screen.  If unimplemented, ADF automatically
+ *	creates an sw_sync fence.  Return the sync fence on success or a
+ *	PTR_ERR() on failure.
+ * @post: flip @cfg onto the screen.  Wait for the display to begin scanning out
+ *	@cfg before returning.
+ * @advance_timeline: signal the sync fence for the last configuration to leave
+ *	the display.  If unimplemented, ADF automatically advances an sw_sync
+ *	timeline.
+ * @state_free: free driver-private state allocated during validate()
+ */
+struct adf_device_ops {
+	/* required */
+	struct module *owner;
+	const struct adf_obj_ops base;
+	/* optional */
+	const struct adf_device_quirks quirks;
+
+	/* optional */
+	int (*attach)(struct adf_device *dev, struct adf_overlay_engine *eng,
+			struct adf_interface *intf);
+	/* optional */
+	int (*detach)(struct adf_device *dev, struct adf_overlay_engine *eng,
+			struct adf_interface *intf);
+
+	/* required if any of the device's overlay engines supports at least one
+	   custom format */
+	int (*validate_custom_format)(struct adf_device *dev,
+			struct adf_buffer *buf);
+
+	/* required */
+	int (*validate)(struct adf_device *dev, struct adf_post *cfg,
+			void **driver_state);
+	/* optional */
+	struct sync_fence *(*complete_fence)(struct adf_device *dev,
+			struct adf_post *cfg, void *driver_state);
+	/* required */
+	void (*post)(struct adf_device *dev, struct adf_post *cfg,
+			void *driver_state);
+	/* required if complete_fence is implemented */
+	void (*advance_timeline)(struct adf_device *dev,
+			struct adf_post *cfg, void *driver_state);
+	/* required if validate allocates driver state */
+	void (*state_free)(struct adf_device *dev, void *driver_state);
+};
+
+struct adf_attachment_list {
+	struct adf_attachment attachment;
+	struct list_head head;
+};
+
+struct adf_device {
+	struct adf_obj base;
+	struct device *dev;
+
+	const struct adf_device_ops *ops;
+
+	struct mutex client_lock;
+
+	struct idr interfaces;
+	size_t n_interfaces;
+	struct idr overlay_engines;
+
+	struct list_head post_list;
+	struct mutex post_lock;
+	struct kthread_worker post_worker;
+	struct task_struct *post_thread;
+	struct kthread_work post_work;
+
+	struct list_head attached;
+	size_t n_attached;
+	struct list_head attach_allowed;
+	size_t n_attach_allowed;
+
+	struct adf_pending_post *onscreen;
+
+	struct sw_sync_timeline *timeline;
+	int timeline_max;
+};
+
+/**
+ * struct adf_interface_ops - display interface implementation ops
+ *
+ * @base: common operations (see &struct adf_obj_ops)
+ *
+ * @blank: change the display's DPMS state.  Return 0 on success or error
+ *	code (<0) on failure.
+ *
+ * @alloc_simple_buffer: allocate a buffer with the specified @w, @h, and
+ *	@format.  @format will be a standard RGB format (i.e.,
+ *	adf_format_is_rgb(@format) == true).  Return 0 on success or error code
+ *	(<0) on failure.  On success, return the buffer, offset, and pitch in
+ *	@dma_buf, @offset, and @pitch respectively.
+ * @describe_simple_post: provide driver-private data needed to post a single
+ *	buffer @buf.  Copy up to ADF_MAX_CUSTOM_DATA_SIZE bytes into @data
+ *	(allocated by ADF) and return the number of bytes in @size.  Return 0 on
+ *	success or error code (<0) on failure.
+ *
+ * @modeset: change the interface's mode.  @mode is not necessarily part of the
+ *	modelist passed to adf_hotplug_notify_connected(); the driver may
+ *	accept or reject custom modes at its discretion.  Return 0 on success or
+ *	error code (<0) if the mode could not be set.
+ *
+ * @screen_size: copy the screen dimensions in millimeters into @width_mm
+ *	and @height_mm.  Return 0 on success or error code (<0) if the display
+ *	dimensions are unknown.
+ *
+ * @type_str: return a string representation of custom @intf->type
+ *	(@intf->type >= @ADF_INTF_TYPE_DEVICE_CUSTOM).
+ */
+struct adf_interface_ops {
+	const struct adf_obj_ops base;
+
+	/* optional */
+	int (*blank)(struct adf_interface *intf, u8 state);
+
+	/* optional */
+	int (*alloc_simple_buffer)(struct adf_interface *intf,
+			u16 w, u16 h, u32 format,
+			struct dma_buf **dma_buf, u32 *offset, u32 *pitch);
+	/* optional */
+	int (*describe_simple_post)(struct adf_interface *intf,
+			struct adf_buffer *fb, void *data, size_t *size);
+
+	/* optional */
+	int (*modeset)(struct adf_interface *intf,
+			struct drm_mode_modeinfo *mode);
+
+	/* optional */
+	int (*screen_size)(struct adf_interface *intf, u16 *width_mm,
+			u16 *height_mm);
+
+	/* optional */
+	const char *(*type_str)(struct adf_interface *intf);
+};
+
+struct adf_interface {
+	struct adf_obj base;
+	const struct adf_interface_ops *ops;
+
+	struct drm_mode_modeinfo current_mode;
+
+	enum adf_interface_type type;
+	u32 idx;
+	u32 flags;
+
+	wait_queue_head_t vsync_wait;
+	ktime_t vsync_timestamp;
+	rwlock_t vsync_lock;
+
+	u8 dpms_state;
+
+	bool hotplug_detect;
+	struct drm_mode_modeinfo *modelist;
+	size_t n_modes;
+	rwlock_t hotplug_modelist_lock;
+};
+
+/**
+ * struct adf_interface_ops - overlay engine implementation ops
+ *
+ * @base: common operations (see &struct adf_obj_ops)
+ *
+ * @supported_formats: list of fourccs the overlay engine can scan out
+ * @n_supported_formats: length of supported_formats, up to
+ *	ADF_MAX_SUPPORTED_FORMATS
+ */
+struct adf_overlay_engine_ops {
+	const struct adf_obj_ops base;
+
+	/* required */
+	const u32 *supported_formats;
+	/* required */
+	const size_t n_supported_formats;
+};
+
+struct adf_overlay_engine {
+	struct adf_obj base;
+
+	const struct adf_overlay_engine_ops *ops;
+};
+
+#define adf_obj_to_device(ptr) \
+	container_of((ptr), struct adf_device, base)
+
+#define adf_obj_to_interface(ptr) \
+	container_of((ptr), struct adf_interface, base)
+
+#define adf_obj_to_overlay_engine(ptr) \
+	container_of((ptr), struct adf_overlay_engine, base)
+
+int __printf(4, 5) adf_device_init(struct adf_device *dev,
+		struct device *parent, const struct adf_device_ops *ops,
+		const char *fmt, ...);
+void adf_device_destroy(struct adf_device *dev);
+int __printf(7, 8) adf_interface_init(struct adf_interface *intf,
+		struct adf_device *dev, enum adf_interface_type type, u32 idx,
+		u32 flags, const struct adf_interface_ops *ops, const char *fmt,
+		...);
+void adf_interface_destroy(struct adf_interface *intf);
+static inline struct adf_device *adf_interface_parent(
+		struct adf_interface *intf)
+{
+	return intf->base.parent;
+}
+int __printf(4, 5) adf_overlay_engine_init(struct adf_overlay_engine *eng,
+		struct adf_device *dev,
+		const struct adf_overlay_engine_ops *ops, const char *fmt, ...);
+void adf_overlay_engine_destroy(struct adf_overlay_engine *eng);
+static inline struct adf_device *adf_overlay_engine_parent(
+		struct adf_overlay_engine *eng)
+{
+	return eng->base.parent;
+}
+
+int adf_attachment_allow(struct adf_device *dev, struct adf_overlay_engine *eng,
+		struct adf_interface *intf);
+
+const char *adf_obj_type_str(enum adf_obj_type type);
+const char *adf_interface_type_str(struct adf_interface *intf);
+const char *adf_event_type_str(struct adf_obj *obj, enum adf_event_type type);
+
+#define ADF_FORMAT_STR_SIZE 5
+void adf_format_str(u32 format, char buf[ADF_FORMAT_STR_SIZE]);
+int adf_format_validate_yuv(struct adf_device *dev, struct adf_buffer *buf,
+		u8 num_planes, u8 hsub, u8 vsub, u8 cpp[]);
+/**
+ * adf_format_validate_rgb - validate the number and size of planes in buffers
+ * with a custom RGB format.
+ *
+ * @dev: ADF device performing the validation
+ * @buf: buffer to validate
+ * @cpp: expected bytes per pixel
+ *
+ * adf_format_validate_rgb() is intended to be called as a helper from @dev's
+ * validate_custom_format() op.  @buf must have a single RGB plane.
+ *
+ * Returns 0 if @buf has a single plane with sufficient size, or -EINVAL
+ * otherwise.
+ */
+static inline int adf_format_validate_rgb(struct adf_device *dev,
+		struct adf_buffer *buf, u8 cpp)
+{
+	return adf_format_validate_yuv(dev, buf, 1, 1, 1, &cpp);
+}
+
+int adf_event_get(struct adf_obj *obj, enum adf_event_type type);
+int adf_event_put(struct adf_obj *obj, enum adf_event_type type);
+int adf_event_notify(struct adf_obj *obj, struct adf_event *event);
+
+static inline void adf_vsync_get(struct adf_interface *intf)
+{
+	adf_event_get(&intf->base, ADF_EVENT_VSYNC);
+}
+
+static inline void adf_vsync_put(struct adf_interface *intf)
+{
+	adf_event_put(&intf->base, ADF_EVENT_VSYNC);
+}
+
+int adf_vsync_wait(struct adf_interface *intf, long timeout);
+void adf_vsync_notify(struct adf_interface *intf, ktime_t timestamp);
+
+int adf_hotplug_notify_connected(struct adf_interface *intf,
+		struct drm_mode_modeinfo *modelist, size_t n_modes);
+void adf_hotplug_notify_disconnected(struct adf_interface *intf);
+
+void adf_modeinfo_set_name(struct drm_mode_modeinfo *mode);
+void adf_modeinfo_set_vrefresh(struct drm_mode_modeinfo *mode);
+
+#endif /* _VIDEO_ADF_H */
diff --git a/include/video/adf_client.h b/include/video/adf_client.h
new file mode 100644
index 000000000000..983f2b6a5890
--- /dev/null
+++ b/include/video/adf_client.h
@@ -0,0 +1,61 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _VIDEO_ADF_CLIENT_H_
+#define _VIDEO_ADF_CLIENT_H_
+
+#include <video/adf.h>
+
+int adf_interface_blank(struct adf_interface *intf, u8 state);
+u8 adf_interface_dpms_state(struct adf_interface *intf);
+
+void adf_interface_current_mode(struct adf_interface *intf,
+		struct drm_mode_modeinfo *mode);
+size_t adf_interface_modelist(struct adf_interface *intf,
+		struct drm_mode_modeinfo *modelist, size_t n_modes);
+int adf_interface_set_mode(struct adf_interface *intf,
+		struct drm_mode_modeinfo *mode);
+int adf_interface_get_screen_size(struct adf_interface *intf, u16 *width,
+		u16 *height);
+int adf_interface_simple_buffer_alloc(struct adf_interface *intf, u16 w, u16 h,
+		u32 format, struct dma_buf **dma_buf, u32 *offset, u32 *pitch);
+struct sync_fence *adf_interface_simple_post(struct adf_interface *intf,
+		struct adf_buffer *buf);
+
+bool adf_overlay_engine_supports_format(struct adf_overlay_engine *eng,
+		u32 format);
+
+size_t adf_device_attachments(struct adf_device *dev,
+		struct adf_attachment *attachments, size_t n_attachments);
+size_t adf_device_attachments_allowed(struct adf_device *dev,
+		struct adf_attachment *attachments, size_t n_attachments);
+bool adf_device_attached(struct adf_device *dev, struct adf_overlay_engine *eng,
+		struct adf_interface *intf);
+bool adf_device_attach_allowed(struct adf_device *dev,
+		struct adf_overlay_engine *eng, struct adf_interface *intf);
+int adf_device_attach(struct adf_device *dev, struct adf_overlay_engine *eng,
+		struct adf_interface *intf);
+int adf_device_detach(struct adf_device *dev, struct adf_overlay_engine *eng,
+		struct adf_interface *intf);
+
+struct sync_fence *adf_device_post(struct adf_device *dev,
+		struct adf_interface **intfs, size_t n_intfs,
+		struct adf_buffer *bufs, size_t n_bufs, void *custom_data,
+		size_t custom_data_size);
+struct sync_fence *adf_device_post_nocopy(struct adf_device *dev,
+		struct adf_interface **intfs, size_t n_intfs,
+		struct adf_buffer *bufs, size_t n_bufs, void *custom_data,
+		size_t custom_data_size);
+
+#endif /* _VIDEO_ADF_CLIENT_H_ */
diff --git a/include/video/adf_fbdev.h b/include/video/adf_fbdev.h
new file mode 100644
index 000000000000..b722c6b3ab02
--- /dev/null
+++ b/include/video/adf_fbdev.h
@@ -0,0 +1,124 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _VIDEO_ADF_FBDEV_H_
+#define _VIDEO_ADF_FBDEV_H_
+
+#include <linux/fb.h>
+#include <linux/mutex.h>
+#include <video/adf.h>
+
+struct adf_fbdev {
+	struct adf_interface *intf;
+	struct adf_overlay_engine *eng;
+	struct fb_info *info;
+	u32 pseudo_palette[16];
+
+	unsigned int refcount;
+	struct mutex refcount_lock;
+
+	struct dma_buf *dma_buf;
+	u32 offset;
+	u32 pitch;
+	void *vaddr;
+	u32 format;
+
+	u16 default_xres_virtual;
+	u16 default_yres_virtual;
+	u32 default_format;
+};
+
+#if IS_ENABLED(CONFIG_ADF_FBDEV)
+void adf_modeinfo_to_fb_videomode(const struct drm_mode_modeinfo *mode,
+		struct fb_videomode *vmode);
+void adf_modeinfo_from_fb_videomode(const struct fb_videomode *vmode,
+		struct drm_mode_modeinfo *mode);
+
+int adf_fbdev_init(struct adf_fbdev *fbdev, struct adf_interface *interface,
+		struct adf_overlay_engine *eng,
+		u16 xres_virtual, u16 yres_virtual, u32 format,
+		struct fb_ops *fbops, const char *fmt, ...);
+void adf_fbdev_destroy(struct adf_fbdev *fbdev);
+
+int adf_fbdev_open(struct fb_info *info, int user);
+int adf_fbdev_release(struct fb_info *info, int user);
+int adf_fbdev_check_var(struct fb_var_screeninfo *var, struct fb_info *info);
+int adf_fbdev_set_par(struct fb_info *info);
+int adf_fbdev_blank(int blank, struct fb_info *info);
+int adf_fbdev_pan_display(struct fb_var_screeninfo *var, struct fb_info *info);
+int adf_fbdev_mmap(struct fb_info *info, struct vm_area_struct *vma);
+#else
+static inline void adf_modeinfo_to_fb_videomode(const struct drm_mode_modeinfo *mode,
+		struct fb_videomode *vmode)
+{
+	WARN_ONCE(1, "%s: CONFIG_ADF_FBDEV is disabled\n", __func__);
+}
+
+static inline void adf_modeinfo_from_fb_videomode(const struct fb_videomode *vmode,
+		struct drm_mode_modeinfo *mode)
+{
+	WARN_ONCE(1, "%s: CONFIG_ADF_FBDEV is disabled\n", __func__);
+}
+
+static inline int adf_fbdev_init(struct adf_fbdev *fbdev,
+		struct adf_interface *interface,
+		struct adf_overlay_engine *eng,
+		u16 xres_virtual, u16 yres_virtual, u32 format,
+		struct fb_ops *fbops, const char *fmt, ...)
+{
+	return -ENODEV;
+}
+
+static inline void adf_fbdev_destroy(struct adf_fbdev *fbdev) { }
+
+static inline int adf_fbdev_open(struct fb_info *info, int user)
+{
+	return -ENODEV;
+}
+
+static inline int adf_fbdev_release(struct fb_info *info, int user)
+{
+	return -ENODEV;
+}
+
+static inline int adf_fbdev_check_var(struct fb_var_screeninfo *var,
+		struct fb_info *info)
+{
+	return -ENODEV;
+}
+
+static inline int adf_fbdev_set_par(struct fb_info *info)
+{
+	return -ENODEV;
+}
+
+static inline int adf_fbdev_blank(int blank, struct fb_info *info)
+{
+	return -ENODEV;
+}
+
+static inline int adf_fbdev_pan_display(struct fb_var_screeninfo *var,
+		struct fb_info *info)
+{
+	return -ENODEV;
+}
+
+static inline int adf_fbdev_mmap(struct fb_info *info,
+		struct vm_area_struct *vma)
+{
+	return -ENODEV;
+}
+#endif
+
+#endif /* _VIDEO_ADF_FBDEV_H_ */
diff --git a/include/video/adf_format.h b/include/video/adf_format.h
new file mode 100644
index 000000000000..e03182cdcb06
--- /dev/null
+++ b/include/video/adf_format.h
@@ -0,0 +1,26 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _VIDEO_ADF_FORMAT_H
+#define _VIDEO_ADF_FORMAT_H
+
+bool adf_format_is_standard(u32 format);
+bool adf_format_is_rgb(u32 format);
+u8 adf_format_num_planes(u32 format);
+u8 adf_format_bpp(u32 format);
+u8 adf_format_plane_cpp(u32 format, int plane);
+u8 adf_format_horz_chroma_subsampling(u32 format);
+u8 adf_format_vert_chroma_subsampling(u32 format);
+
+#endif /* _VIDEO_ADF_FORMAT_H */
diff --git a/include/video/adf_memblock.h b/include/video/adf_memblock.h
new file mode 100644
index 000000000000..6256e0eebcc9
--- /dev/null
+++ b/include/video/adf_memblock.h
@@ -0,0 +1,20 @@
+/*
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#ifndef _VIDEO_ADF_MEMBLOCK_H_
+#define _VIDEO_ADF_MEMBLOCK_H_
+
+struct dma_buf *adf_memblock_export(phys_addr_t base, size_t size, int flags);
+
+#endif /* _VIDEO_ADF_MEMBLOCK_H_ */
diff --git a/init/Kconfig b/init/Kconfig
index e1d1d6936f92..f5500e552254 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -26,6 +26,16 @@ config IRQ_WORK
 config BUILDTIME_EXTABLE_SORT
 	bool
 
+config THREAD_INFO_IN_TASK
+	bool
+	help
+	  Select this to move thread_info off the stack into task_struct.  To
+	  make this work, an arch will need to remove all thread_info fields
+	  except flags and fix any runtime bugs.
+
+	  One subtle change that will be needed is to use try_get_task_stack()
+	  and put_task_stack() in save_thread_stack_tsk() and get_wchan().
+
 menu "General setup"
 
 config BROKEN
@@ -392,6 +402,16 @@ config IRQ_TIME_ACCOUNTING
 
 endchoice
 
+config SCHED_WALT
+        bool "Support window based load tracking"
+        depends on SMP
+        depends on FAIR_GROUP_SCHED
+        help
+        This feature will allow the scheduler to maintain a tunable window
+	based set of metrics for tasks and runqueues. These metrics can be
+	used to guide task placement as well as task frequency requirements
+	for cpufreq governors.
+
 config BSD_PROCESS_ACCT
 	bool "BSD Process Accounting"
 	depends on MULTIUSER
@@ -999,6 +1019,23 @@ config CGROUP_CPUACCT
 	  Provides a simple Resource Controller for monitoring the
 	  total CPU consumed by the tasks in a cgroup.
 
+config CGROUP_SCHEDTUNE
+	bool "CFS tasks boosting cgroup subsystem (EXPERIMENTAL)"
+	depends on SCHED_TUNE
+	help
+	  This option provides the "schedtune" controller which improves the
+	  flexibility of the task boosting mechanism by introducing the support
+	  to define "per task" boost values.
+
+	  This new controller:
+	  1. allows only a two layers hierarchy, where the root defines the
+	     system-wide boost value and its direct childrens define each one a
+	     different "class of tasks" to be boosted with a different value
+	  2. supports up to 16 different task classes, each one which could be
+	     configured with a different boost value
+
+	  Say N if unsure.
+
 config PAGE_COUNTER
        bool
 
@@ -1237,6 +1274,43 @@ config SCHED_AUTOGROUP
 	  desktop applications.  Task group autogeneration is currently based
 	  upon task session.
 
+config SCHED_TUNE
+	bool "Boosting for CFS tasks (EXPERIMENTAL)"
+	depends on SMP
+	help
+	  This option enables the system-wide support for task boosting.
+	  When this support is enabled a new sysctl interface is exposed to
+	  userspace via:
+	     /proc/sys/kernel/sched_cfs_boost
+	  which allows to set a system-wide boost value in range [0..100].
+
+	  The currently boosting strategy is implemented in such a way that:
+	  - a 0% boost value requires to operate in "standard" mode by
+	    scheduling all tasks at the minimum capacities required by their
+	    workload demand
+	  - a 100% boost value requires to push at maximum the task
+	    performances, "regardless" of the incurred energy consumption
+
+	  A boost value in between these two boundaries is used to bias the
+	  power/performance trade-off, the higher the boost value the more the
+	  scheduler is biased toward performance boosting instead of energy
+	  efficiency.
+
+	  Since this support exposes a single system-wide knob, the specified
+	  boost value is applied to all (CFS) tasks in the system.
+
+	  If unsure, say N.
+
+config DEFAULT_USE_ENERGY_AWARE
+	bool "Default to enabling the Energy Aware Scheduler feature"
+	default n
+	help
+	  This option defaults the ENERGY_AWARE scheduling feature to true,
+	  as without SCHED_DEBUG set this feature can't be enabled or disabled
+	  via sysctl.
+
+	  Say N if unsure.
+
 config SYSFS_DEPRECATED
 	bool "Enable deprecated sysfs features to support old userspace tools"
 	depends on SYSFS
diff --git a/init/Makefile b/init/Makefile
index 7bc47ee31c36..243f61de2cba 100644
--- a/init/Makefile
+++ b/init/Makefile
@@ -3,11 +3,8 @@
 #
 
 obj-y                          := main.o version.o mounts.o
-ifneq ($(CONFIG_BLK_DEV_INITRD),y)
 obj-y                          += noinitramfs.o
-else
 obj-$(CONFIG_BLK_DEV_INITRD)   += initramfs.o
-endif
 obj-$(CONFIG_GENERIC_CALIBRATE_DELAY) += calibrate.o
 
 ifneq ($(CONFIG_ARCH_INIT_TASK),y)
@@ -18,6 +15,7 @@ mounts-y			:= do_mounts.o
 mounts-$(CONFIG_BLK_DEV_RAM)	+= do_mounts_rd.o
 mounts-$(CONFIG_BLK_DEV_INITRD)	+= do_mounts_initrd.o
 mounts-$(CONFIG_BLK_DEV_MD)	+= do_mounts_md.o
+mounts-$(CONFIG_BLK_DEV_DM)	+= do_mounts_dm.o
 
 # dependencies on generated files need to be listed explicitly
 $(obj)/version.o: include/generated/compile.h
diff --git a/init/do_mounts.c b/init/do_mounts.c
index dea5de95c2dd..1902a1c80831 100644
--- a/init/do_mounts.c
+++ b/init/do_mounts.c
@@ -566,6 +566,7 @@ void __init prepare_namespace(void)
 	wait_for_device_probe();
 
 	md_run_setup();
+	dm_run_setup();
 
 	if (saved_root_name[0]) {
 		root_device_name = saved_root_name;
diff --git a/init/do_mounts.h b/init/do_mounts.h
index f5b978a9bb92..09d22862e8c3 100644
--- a/init/do_mounts.h
+++ b/init/do_mounts.h
@@ -74,3 +74,13 @@ void md_run_setup(void);
 static inline void md_run_setup(void) {}
 
 #endif
+
+#ifdef CONFIG_BLK_DEV_DM
+
+void dm_run_setup(void);
+
+#else
+
+static inline void dm_run_setup(void) {}
+
+#endif
diff --git a/init/do_mounts_dm.c b/init/do_mounts_dm.c
new file mode 100644
index 000000000000..bce1c2fbb915
--- /dev/null
+++ b/init/do_mounts_dm.c
@@ -0,0 +1,470 @@
+/* do_mounts_dm.c
+ * Copyright (C) 2010 The Chromium OS Authors <chromium-os-dev@chromium.org>
+ *                    All Rights Reserved.
+ * Based on do_mounts_md.c
+ *
+ * This file is released under the GPL.
+ */
+#include <linux/async.h>
+#include <linux/ctype.h>
+#include <linux/device-mapper.h>
+#include <linux/fs.h>
+#include <linux/string.h>
+#include <linux/delay.h>
+
+#include "do_mounts.h"
+
+#define DM_MAX_DEVICES 256
+#define DM_MAX_TARGETS 256
+#define DM_MAX_NAME 32
+#define DM_MAX_UUID 129
+#define DM_NO_UUID "none"
+
+#define DM_MSG_PREFIX "init"
+
+/* Separators used for parsing the dm= argument. */
+#define DM_FIELD_SEP " "
+#define DM_LINE_SEP ","
+#define DM_ANY_SEP DM_FIELD_SEP DM_LINE_SEP
+
+/*
+ * When the device-mapper and any targets are compiled into the kernel
+ * (not a module), one or more device-mappers may be created and used
+ * as the root device at boot time with the parameters given with the
+ * boot line dm=...
+ *
+ * Multiple device-mappers can be stacked specifing the number of
+ * devices. A device can have multiple targets if the the number of
+ * targets is specified.
+ *
+ * TODO(taysom:defect 32847)
+ * In the future, the <num> field will be mandatory.
+ *
+ * <device>        ::= [<num>] <device-mapper>+
+ * <device-mapper> ::= <head> "," <target>+
+ * <head>          ::= <name> <uuid> <mode> [<num>]
+ * <target>        ::= <start> <length> <type> <options> ","
+ * <mode>          ::= "ro" | "rw"
+ * <uuid>          ::= xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx | "none"
+ * <type>          ::= "verity" | "bootcache" | ...
+ *
+ * Example:
+ * 2 vboot none ro 1,
+ *     0 1768000 bootcache
+ *       device=aa55b119-2a47-8c45-946a-5ac57765011f+1
+ *       signature=76e9be054b15884a9fa85973e9cb274c93afadb6
+ *       cache_start=1768000 max_blocks=100000 size_limit=23 max_trace=20000,
+ *   vroot none ro 1,
+ *     0 1740800 verity payload=254:0 hashtree=254:0 hashstart=1740800 alg=sha1
+ *       root_hexdigest=76e9be054b15884a9fa85973e9cb274c93afadb6
+ *       salt=5b3549d54d6c7a3837b9b81ed72e49463a64c03680c47835bef94d768e5646fe
+ *
+ * Notes:
+ *  1. uuid is a label for the device and we set it to "none".
+ *  2. The <num> field will be optional initially and assumed to be 1.
+ *     Once all the scripts that set these fields have been set, it will
+ *     be made mandatory.
+ */
+
+struct dm_setup_target {
+	sector_t begin;
+	sector_t length;
+	char *type;
+	char *params;
+	/* simple singly linked list */
+	struct dm_setup_target *next;
+};
+
+struct dm_device {
+	int minor;
+	int ro;
+	char name[DM_MAX_NAME];
+	char uuid[DM_MAX_UUID];
+	unsigned long num_targets;
+	struct dm_setup_target *target;
+	int target_count;
+	struct dm_device *next;
+};
+
+struct dm_option {
+	char *start;
+	char *next;
+	size_t len;
+	char delim;
+};
+
+static struct {
+	unsigned long num_devices;
+	char *str;
+} dm_setup_args __initdata;
+
+static __initdata int dm_early_setup;
+
+static int __init get_dm_option(struct dm_option *opt, const char *accept)
+{
+	char *str = opt->next;
+	char *endp;
+
+	if (!str)
+		return 0;
+
+	str = skip_spaces(str);
+	opt->start = str;
+	endp = strpbrk(str, accept);
+	if (!endp) {  /* act like strchrnul */
+		opt->len = strlen(str);
+		endp = str + opt->len;
+	} else {
+		opt->len = endp - str;
+	}
+	opt->delim = *endp;
+	if (*endp == 0) {
+		/* Don't advance past the nul. */
+		opt->next = endp;
+	} else {
+		opt->next = endp + 1;
+	}
+	return opt->len != 0;
+}
+
+static int __init dm_setup_cleanup(struct dm_device *devices)
+{
+	struct dm_device *dev = devices;
+
+	while (dev) {
+		struct dm_device *old_dev = dev;
+		struct dm_setup_target *target = dev->target;
+		while (target) {
+			struct dm_setup_target *old_target = target;
+			kfree(target->type);
+			kfree(target->params);
+			target = target->next;
+			kfree(old_target);
+			dev->target_count--;
+		}
+		BUG_ON(dev->target_count);
+		dev = dev->next;
+		kfree(old_dev);
+	}
+	return 0;
+}
+
+static char * __init dm_parse_device(struct dm_device *dev, char *str)
+{
+	struct dm_option opt;
+	size_t len;
+
+	/* Grab the logical name of the device to be exported to udev */
+	opt.next = str;
+	if (!get_dm_option(&opt, DM_FIELD_SEP)) {
+		DMERR("failed to parse device name");
+		goto parse_fail;
+	}
+	len = min(opt.len + 1, sizeof(dev->name));
+	strlcpy(dev->name, opt.start, len);  /* includes nul */
+
+	/* Grab the UUID value or "none" */
+	if (!get_dm_option(&opt, DM_FIELD_SEP)) {
+		DMERR("failed to parse device uuid");
+		goto parse_fail;
+	}
+	len = min(opt.len + 1, sizeof(dev->uuid));
+	strlcpy(dev->uuid, opt.start, len);
+
+	/* Determine if the table/device will be read only or read-write */
+	get_dm_option(&opt, DM_ANY_SEP);
+	if (!strncmp("ro", opt.start, opt.len)) {
+		dev->ro = 1;
+	} else if (!strncmp("rw", opt.start, opt.len)) {
+		dev->ro = 0;
+	} else {
+		DMERR("failed to parse table mode");
+		goto parse_fail;
+	}
+
+	/* Optional number field */
+	/* XXX: The <num> field will be mandatory in the next round */
+	if (opt.delim == DM_FIELD_SEP[0]) {
+		if (!get_dm_option(&opt, DM_LINE_SEP))
+			return NULL;
+		dev->num_targets = simple_strtoul(opt.start, NULL, 10);
+	} else {
+		dev->num_targets = 1;
+	}
+	if (dev->num_targets > DM_MAX_TARGETS) {
+		DMERR("too many targets %lu > %d",
+			dev->num_targets, DM_MAX_TARGETS);
+	}
+	return opt.next;
+
+parse_fail:
+	return NULL;
+}
+
+static char * __init dm_parse_targets(struct dm_device *dev, char *str)
+{
+	struct dm_option opt;
+	struct dm_setup_target **target = &dev->target;
+	unsigned long num_targets = dev->num_targets;
+	unsigned long i;
+
+	/* Targets are defined as per the table format but with a
+	 * comma as a newline separator. */
+	opt.next = str;
+	for (i = 0; i < num_targets; i++) {
+		*target = kzalloc(sizeof(struct dm_setup_target), GFP_KERNEL);
+		if (!*target) {
+			DMERR("failed to allocate memory for target %s<%ld>",
+				dev->name, i);
+			goto parse_fail;
+		}
+		dev->target_count++;
+
+		if (!get_dm_option(&opt, DM_FIELD_SEP)) {
+			DMERR("failed to parse starting sector"
+				" for target %s<%ld>", dev->name, i);
+			goto parse_fail;
+		}
+		(*target)->begin = simple_strtoull(opt.start, NULL, 10);
+
+		if (!get_dm_option(&opt, DM_FIELD_SEP)) {
+			DMERR("failed to parse length for target %s<%ld>",
+				dev->name, i);
+			goto parse_fail;
+		}
+		(*target)->length = simple_strtoull(opt.start, NULL, 10);
+
+		if (get_dm_option(&opt, DM_FIELD_SEP))
+			(*target)->type = kstrndup(opt.start, opt.len,
+							GFP_KERNEL);
+		if (!((*target)->type)) {
+			DMERR("failed to parse type for target %s<%ld>",
+				dev->name, i);
+			goto parse_fail;
+		}
+		if (get_dm_option(&opt, DM_LINE_SEP))
+			(*target)->params = kstrndup(opt.start, opt.len,
+							GFP_KERNEL);
+		if (!((*target)->params)) {
+			DMERR("failed to parse params for target %s<%ld>",
+				dev->name, i);
+			goto parse_fail;
+		}
+		target = &((*target)->next);
+	}
+	DMDEBUG("parsed %d targets", dev->target_count);
+
+	return opt.next;
+
+parse_fail:
+	return NULL;
+}
+
+static struct dm_device * __init dm_parse_args(void)
+{
+	struct dm_device *devices = NULL;
+	struct dm_device **tail = &devices;
+	struct dm_device *dev;
+	char *str = dm_setup_args.str;
+	unsigned long num_devices = dm_setup_args.num_devices;
+	unsigned long i;
+
+	if (!str)
+		return NULL;
+	for (i = 0; i < num_devices; i++) {
+		dev = kzalloc(sizeof(*dev), GFP_KERNEL);
+		if (!dev) {
+			DMERR("failed to allocated memory for dev");
+			goto error;
+		}
+		*tail = dev;
+		tail = &dev->next;
+		/*
+		 * devices are given minor numbers 0 - n-1
+		 * in the order they are found in the arg
+		 * string.
+		 */
+		dev->minor = i;
+		str = dm_parse_device(dev, str);
+		if (!str)	/* NULL indicates error in parsing, bail */
+			goto error;
+
+		str = dm_parse_targets(dev, str);
+		if (!str)
+			goto error;
+	}
+	return devices;
+error:
+	dm_setup_cleanup(devices);
+	return NULL;
+}
+
+/*
+ * Parse the command-line parameters given our kernel, but do not
+ * actually try to invoke the DM device now; that is handled by
+ * dm_setup_drives after the low-level disk drivers have initialised.
+ * dm format is described at the top of the file.
+ *
+ * Because dm minor numbers are assigned in assending order starting with 0,
+ * You can assume the first device is /dev/dm-0, the next device is /dev/dm-1,
+ * and so forth.
+ */
+static int __init dm_setup(char *str)
+{
+	struct dm_option opt;
+	unsigned long num_devices;
+
+	if (!str) {
+		DMDEBUG("str is NULL");
+		goto parse_fail;
+	}
+	opt.next = str;
+	if (!get_dm_option(&opt, DM_FIELD_SEP))
+		goto parse_fail;
+	if (isdigit(opt.start[0])) {	/* XXX: Optional number field */
+		num_devices = simple_strtoul(opt.start, NULL, 10);
+		str = opt.next;
+	} else {
+		num_devices = 1;
+		/* Don't advance str */
+	}
+	if (num_devices > DM_MAX_DEVICES) {
+		DMDEBUG("too many devices %lu > %d",
+			num_devices, DM_MAX_DEVICES);
+	}
+	dm_setup_args.str = str;
+	dm_setup_args.num_devices = num_devices;
+	DMINFO("will configure %lu devices", num_devices);
+	dm_early_setup = 1;
+	return 1;
+
+parse_fail:
+	DMWARN("Invalid arguments supplied to dm=.");
+	return 0;
+}
+
+static void __init dm_setup_drives(void)
+{
+	struct mapped_device *md = NULL;
+	struct dm_table *table = NULL;
+	struct dm_setup_target *target;
+	struct dm_device *dev;
+	char *uuid;
+	fmode_t fmode = FMODE_READ;
+	struct dm_device *devices;
+
+	devices = dm_parse_args();
+
+	for (dev = devices; dev; dev = dev->next) {
+		if (dm_create(dev->minor, &md)) {
+			DMDEBUG("failed to create the device");
+			goto dm_create_fail;
+		}
+		DMDEBUG("created device '%s'", dm_device_name(md));
+
+		/*
+		 * In addition to flagging the table below, the disk must be
+		 * set explicitly ro/rw.
+		 */
+		set_disk_ro(dm_disk(md), dev->ro);
+
+		if (!dev->ro)
+			fmode |= FMODE_WRITE;
+		if (dm_table_create(&table, fmode, dev->target_count, md)) {
+			DMDEBUG("failed to create the table");
+			goto dm_table_create_fail;
+		}
+
+		dm_lock_md_type(md);
+
+		for (target = dev->target; target; target = target->next) {
+			DMINFO("adding target '%llu %llu %s %s'",
+			       (unsigned long long) target->begin,
+			       (unsigned long long) target->length,
+			       target->type, target->params);
+			if (dm_table_add_target(table, target->type,
+						target->begin,
+						target->length,
+						target->params)) {
+				DMDEBUG("failed to add the target"
+					" to the table");
+				goto add_target_fail;
+			}
+		}
+		if (dm_table_complete(table)) {
+			DMDEBUG("failed to complete the table");
+			goto table_complete_fail;
+		}
+
+		/* Suspend the device so that we can bind it to the table. */
+		if (dm_suspend(md, 0)) {
+			DMDEBUG("failed to suspend the device pre-bind");
+			goto suspend_fail;
+		}
+
+		/* Initial table load: acquire type of table. */
+		dm_set_md_type(md, dm_table_get_type(table));
+
+		/* Setup md->queue to reflect md's type. */
+		if (dm_setup_md_queue(md)) {
+			DMWARN("unable to set up device queue for new table.");
+			goto setup_md_queue_fail;
+		}
+
+		/*
+		 * Bind the table to the device. This is the only way
+		 * to associate md->map with the table and set the disk
+		 * capacity directly.
+		 */
+		if (dm_swap_table(md, table)) {  /* should return NULL. */
+			DMDEBUG("failed to bind the device to the table");
+			goto table_bind_fail;
+		}
+
+		/* Finally, resume and the device should be ready. */
+		if (dm_resume(md)) {
+			DMDEBUG("failed to resume the device");
+			goto resume_fail;
+		}
+
+		/* Export the dm device via the ioctl interface */
+		if (!strcmp(DM_NO_UUID, dev->uuid))
+			uuid = NULL;
+		if (dm_ioctl_export(md, dev->name, uuid)) {
+			DMDEBUG("failed to export device with given"
+				" name and uuid");
+			goto export_fail;
+		}
+
+		dm_unlock_md_type(md);
+
+		DMINFO("dm-%d is ready", dev->minor);
+	}
+	dm_setup_cleanup(devices);
+	return;
+
+export_fail:
+resume_fail:
+table_bind_fail:
+setup_md_queue_fail:
+suspend_fail:
+table_complete_fail:
+add_target_fail:
+	dm_unlock_md_type(md);
+dm_table_create_fail:
+	dm_put(md);
+dm_create_fail:
+	DMWARN("starting dm-%d (%s) failed",
+	       dev->minor, dev->name);
+	dm_setup_cleanup(devices);
+}
+
+__setup("dm=", dm_setup);
+
+void __init dm_run_setup(void)
+{
+	if (!dm_early_setup)
+		return;
+	DMINFO("attempting early device configuration.");
+	dm_setup_drives();
+}
diff --git a/init/init_task.c b/init/init_task.c
index ba0a7f362d9e..11f83be1fa79 100644
--- a/init/init_task.c
+++ b/init/init_task.c
@@ -22,5 +22,8 @@ EXPORT_SYMBOL(init_task);
  * Initial thread structure. Alignment of this is handled by a special
  * linker map entry.
  */
-union thread_union init_thread_union __init_task_data =
-	{ INIT_THREAD_INFO(init_task) };
+union thread_union init_thread_union __init_task_data = {
+#ifndef CONFIG_THREAD_INFO_IN_TASK
+	INIT_THREAD_INFO(init_task)
+#endif
+};
diff --git a/init/initramfs.c b/init/initramfs.c
index b32ad7d97ac9..f8ce812ba43e 100644
--- a/init/initramfs.c
+++ b/init/initramfs.c
@@ -18,6 +18,7 @@
 #include <linux/dirent.h>
 #include <linux/syscalls.h>
 #include <linux/utime.h>
+#include <linux/initramfs.h>
 
 static ssize_t __init xwrite(int fd, const char *p, size_t count)
 {
@@ -605,9 +606,25 @@ static void __init clean_rootfs(void)
 }
 #endif
 
+static int __initdata do_skip_initramfs;
+
+static int __init skip_initramfs_param(char *str)
+{
+	if (*str)
+		return 0;
+	do_skip_initramfs = 1;
+	return 1;
+}
+__setup("skip_initramfs", skip_initramfs_param);
+
 static int __init populate_rootfs(void)
 {
-	char *err = unpack_to_rootfs(__initramfs_start, __initramfs_size);
+	char *err;
+
+	if (do_skip_initramfs)
+		return default_rootfs();
+
+	err = unpack_to_rootfs(__initramfs_start, __initramfs_size);
 	if (err)
 		panic("%s", err); /* Failed to decompress INTERNAL initramfs */
 	if (initrd_start) {
diff --git a/init/main.c b/init/main.c
index fbafa271531c..86f5ce9ede86 100644
--- a/init/main.c
+++ b/init/main.c
@@ -468,7 +468,7 @@ void __init __weak smp_setup_processor_id(void)
 }
 
 # if THREAD_SIZE >= PAGE_SIZE
-void __init __weak thread_info_cache_init(void)
+void __init __weak thread_stack_cache_init(void)
 {
 }
 #endif
@@ -645,7 +645,7 @@ asmlinkage __visible void __init start_kernel(void)
 	/* Should be run before the first non-init thread is created */
 	init_espfix_bsp();
 #endif
-	thread_info_cache_init();
+	thread_stack_cache_init();
 	cred_init();
 	fork_init();
 	proc_caches_init();
diff --git a/init/noinitramfs.c b/init/noinitramfs.c
index 267739d85179..bcc8bcb053ee 100644
--- a/init/noinitramfs.c
+++ b/init/noinitramfs.c
@@ -21,11 +21,16 @@
 #include <linux/stat.h>
 #include <linux/kdev_t.h>
 #include <linux/syscalls.h>
+#include <linux/kconfig.h>
+#include <linux/initramfs.h>
 
 /*
  * Create a simple rootfs that is similar to the default initramfs
  */
-static int __init default_rootfs(void)
+#if !IS_BUILTIN(CONFIG_BLK_DEV_INITRD)
+static
+#endif
+int __init default_rootfs(void)
 {
 	int err;
 
@@ -49,4 +54,6 @@ out:
 	printk(KERN_WARNING "Failed to create a rootfs\n");
 	return err;
 }
+#if !IS_BUILTIN(CONFIG_BLK_DEV_INITRD)
 rootfs_initcall(default_rootfs);
+#endif
diff --git a/ipc/mqueue.c b/ipc/mqueue.c
index 5e24eb0ab5dd..46436543ad28 100644
--- a/ipc/mqueue.c
+++ b/ipc/mqueue.c
@@ -747,7 +747,7 @@ static struct file *do_create(struct ipc_namespace *ipc_ns, struct inode *dir,
 	}
 
 	mode &= ~current_umask();
-	ret = vfs_create(dir, path->dentry, mode, true);
+	ret = vfs_create2(path->mnt, dir, path->dentry, mode, true);
 	path->dentry->d_fsdata = NULL;
 	if (ret)
 		return ERR_PTR(ret);
@@ -763,7 +763,7 @@ static struct file *do_open(struct path *path, int oflag)
 	if ((oflag & O_ACCMODE) == (O_RDWR | O_WRONLY))
 		return ERR_PTR(-EINVAL);
 	acc = oflag2acc[oflag & O_ACCMODE];
-	if (inode_permission(d_inode(path->dentry), acc))
+	if (inode_permission2(path->mnt, d_inode(path->dentry), acc))
 		return ERR_PTR(-EACCES);
 	return dentry_open(path, oflag, current_cred());
 }
@@ -796,7 +796,7 @@ SYSCALL_DEFINE4(mq_open, const char __user *, u_name, int, oflag, umode_t, mode,
 	ro = mnt_want_write(mnt);	/* we'll drop it in any case */
 	error = 0;
 	mutex_lock(&d_inode(root)->i_mutex);
-	path.dentry = lookup_one_len(name->name, root, strlen(name->name));
+	path.dentry = lookup_one_len2(name->name, mnt, root, strlen(name->name));
 	if (IS_ERR(path.dentry)) {
 		error = PTR_ERR(path.dentry);
 		goto out_putfd;
@@ -867,7 +867,7 @@ SYSCALL_DEFINE1(mq_unlink, const char __user *, u_name)
 	if (err)
 		goto out_name;
 	mutex_lock_nested(&d_inode(mnt->mnt_root)->i_mutex, I_MUTEX_PARENT);
-	dentry = lookup_one_len(name->name, mnt->mnt_root,
+	dentry = lookup_one_len2(name->name, mnt, mnt->mnt_root,
 				strlen(name->name));
 	if (IS_ERR(dentry)) {
 		err = PTR_ERR(dentry);
@@ -879,7 +879,7 @@ SYSCALL_DEFINE1(mq_unlink, const char __user *, u_name)
 		err = -ENOENT;
 	} else {
 		ihold(inode);
-		err = vfs_unlink(d_inode(dentry->d_parent), dentry, NULL);
+		err = vfs_unlink2(mnt, d_inode(dentry->d_parent), dentry, NULL);
 	}
 	dput(dentry);
 
diff --git a/kernel/audit.c b/kernel/audit.c
index 5ffcbd354a52..34f690b9213a 100644
--- a/kernel/audit.c
+++ b/kernel/audit.c
@@ -870,6 +870,12 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 				return err;
 		}
 		if (s.mask & AUDIT_STATUS_PID) {
+			/* NOTE: we are using task_tgid_vnr() below because
+			 *       the s.pid value is relative to the namespace
+			 *       of the caller; at present this doesn't matter
+			 *       much since you can really only run auditd
+			 *       from the initial pid namespace, but something
+			 *       to keep in mind if this changes */
 			int new_pid = s.pid;
 
 			if ((!new_pid) && (task_tgid_vnr(current) != audit_pid))
@@ -1896,7 +1902,7 @@ void audit_log_task_info(struct audit_buffer *ab, struct task_struct *tsk)
 			 " euid=%u suid=%u fsuid=%u"
 			 " egid=%u sgid=%u fsgid=%u tty=%s ses=%u",
 			 task_ppid_nr(tsk),
-			 task_pid_nr(tsk),
+			 task_tgid_nr(tsk),
 			 from_kuid(&init_user_ns, audit_get_loginuid(tsk)),
 			 from_kuid(&init_user_ns, cred->uid),
 			 from_kgid(&init_user_ns, cred->gid),
diff --git a/kernel/auditsc.c b/kernel/auditsc.c
index 48f45987dc6c..63f0e495f517 100644
--- a/kernel/auditsc.c
+++ b/kernel/auditsc.c
@@ -458,7 +458,7 @@ static int audit_filter_rules(struct task_struct *tsk,
 
 		switch (f->type) {
 		case AUDIT_PID:
-			pid = task_pid_nr(tsk);
+			pid = task_tgid_nr(tsk);
 			result = audit_comparator(pid, f->op, f->val);
 			break;
 		case AUDIT_PPID:
@@ -1987,7 +1987,7 @@ static void audit_log_set_loginuid(kuid_t koldloginuid, kuid_t kloginuid,
 	ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_LOGIN);
 	if (!ab)
 		return;
-	audit_log_format(ab, "pid=%d uid=%u", task_pid_nr(current), uid);
+	audit_log_format(ab, "pid=%d uid=%u", task_tgid_nr(current), uid);
 	audit_log_task_context(ab);
 	audit_log_format(ab, " old-auid=%u auid=%u old-ses=%u ses=%u res=%d",
 			 oldloginuid, loginuid, oldsessionid, sessionid, !rc);
@@ -2212,7 +2212,7 @@ void __audit_ptrace(struct task_struct *t)
 {
 	struct audit_context *context = current->audit_context;
 
-	context->target_pid = task_pid_nr(t);
+	context->target_pid = task_tgid_nr(t);
 	context->target_auid = audit_get_loginuid(t);
 	context->target_uid = task_uid(t);
 	context->target_sessionid = audit_get_sessionid(t);
@@ -2237,7 +2237,7 @@ int __audit_signal_info(int sig, struct task_struct *t)
 
 	if (audit_pid && t->tgid == audit_pid) {
 		if (sig == SIGTERM || sig == SIGHUP || sig == SIGUSR1 || sig == SIGUSR2) {
-			audit_sig_pid = task_pid_nr(tsk);
+			audit_sig_pid = task_tgid_nr(tsk);
 			if (uid_valid(tsk->loginuid))
 				audit_sig_uid = tsk->loginuid;
 			else
@@ -2337,7 +2337,7 @@ int __audit_log_bprm_fcaps(struct linux_binprm *bprm,
 void __audit_log_capset(const struct cred *new, const struct cred *old)
 {
 	struct audit_context *context = current->audit_context;
-	context->capset.pid = task_pid_nr(current);
+	context->capset.pid = task_tgid_nr(current);
 	context->capset.cap.effective   = new->cap_effective;
 	context->capset.cap.inheritable = new->cap_effective;
 	context->capset.cap.permitted   = new->cap_permitted;
@@ -2369,7 +2369,7 @@ static void audit_log_task(struct audit_buffer *ab)
 			 from_kgid(&init_user_ns, gid),
 			 sessionid);
 	audit_log_task_context(ab);
-	audit_log_format(ab, " pid=%d comm=", task_pid_nr(current));
+	audit_log_format(ab, " pid=%d comm=", task_tgid_nr(current));
 	audit_log_untrustedstring(ab, get_task_comm(comm, current));
 	audit_log_d_path_exe(ab, current->mm);
 }
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index ca961af986f7..16d5b7a46850 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -2686,7 +2686,8 @@ static int cgroup_procs_write_permission(struct task_struct *task,
 	 */
 	if (!uid_eq(cred->euid, GLOBAL_ROOT_UID) &&
 	    !uid_eq(cred->euid, tcred->uid) &&
-	    !uid_eq(cred->euid, tcred->suid))
+	    !uid_eq(cred->euid, tcred->suid) &&
+	    !ns_capable(tcred->user_ns, CAP_SYS_RESOURCE))
 		ret = -EACCES;
 
 	if (!ret && cgroup_on_dfl(dst_cgrp)) {
@@ -5327,6 +5328,12 @@ int __init cgroup_init(void)
 	BUG_ON(cgroup_init_cftypes(NULL, cgroup_dfl_base_files));
 	BUG_ON(cgroup_init_cftypes(NULL, cgroup_legacy_base_files));
 
+	/*
+	 * The latency of the synchronize_sched() is too high for cgroups,
+	 * avoid it at the cost of forcing all readers into the slow path.
+	 */
+	rcu_sync_enter_start(&cgroup_threadgroup_rwsem.rss);
+
 	mutex_lock(&cgroup_mutex);
 
 	/* Add init_css_set to the hash table */
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 40d20bf5de28..24ca2963754b 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -24,6 +24,8 @@
 #include <linux/irq.h>
 #include <trace/events/power.h>
 
+#include <trace/events/sched.h>
+
 #include "smpboot.h"
 
 #ifdef CONFIG_SMP
@@ -183,10 +185,17 @@ void cpu_hotplug_disable(void)
 }
 EXPORT_SYMBOL_GPL(cpu_hotplug_disable);
 
+static void __cpu_hotplug_enable(void)
+{
+	if (WARN_ONCE(!cpu_hotplug_disabled, "Unbalanced cpu hotplug enable\n"))
+		return;
+	cpu_hotplug_disabled--;
+}
+
 void cpu_hotplug_enable(void)
 {
 	cpu_maps_update_begin();
-	WARN_ON(--cpu_hotplug_disabled < 0);
+	__cpu_hotplug_enable();
 	cpu_maps_update_done();
 }
 EXPORT_SYMBOL_GPL(cpu_hotplug_enable);
@@ -425,6 +434,7 @@ static int _cpu_down(unsigned int cpu, int tasks_frozen)
 
 out_release:
 	cpu_hotplug_done();
+	trace_sched_cpu_hotplug(cpu, err, 0);
 	if (!err)
 		cpu_notify_nofail(CPU_POST_DEAD | mod, hcpu);
 	return err;
@@ -530,6 +540,7 @@ out_notify:
 		__cpu_notify(CPU_UP_CANCELED | mod, hcpu, nr_calls, NULL);
 out:
 	cpu_hotplug_done();
+	trace_sched_cpu_hotplug(cpu, ret, 1);
 
 	return ret;
 }
@@ -623,10 +634,11 @@ void __weak arch_enable_nonboot_cpus_end(void)
 void enable_nonboot_cpus(void)
 {
 	int cpu, error;
+	struct device *cpu_device;
 
 	/* Allow everyone to use the CPU hotplug again */
 	cpu_maps_update_begin();
-	WARN_ON(--cpu_hotplug_disabled < 0);
+	__cpu_hotplug_enable();
 	if (cpumask_empty(frozen_cpus))
 		goto out;
 
@@ -640,6 +652,12 @@ void enable_nonboot_cpus(void)
 		trace_suspend_resume(TPS("CPU_ON"), cpu, false);
 		if (!error) {
 			pr_info("CPU%d is up\n", cpu);
+			cpu_device = get_cpu_device(cpu);
+			if (!cpu_device)
+				pr_err("%s: failed to get cpu%d device\n",
+				       __func__, cpu);
+			else
+				kobject_uevent(&cpu_device->kobj, KOBJ_ONLINE);
 			continue;
 		}
 		pr_warn("Error taking CPU%d up: %d\n", cpu, error);
@@ -827,3 +845,23 @@ void init_cpu_online(const struct cpumask *src)
 {
 	cpumask_copy(to_cpumask(cpu_online_bits), src);
 }
+
+static ATOMIC_NOTIFIER_HEAD(idle_notifier);
+
+void idle_notifier_register(struct notifier_block *n)
+{
+	atomic_notifier_chain_register(&idle_notifier, n);
+}
+EXPORT_SYMBOL_GPL(idle_notifier_register);
+
+void idle_notifier_unregister(struct notifier_block *n)
+{
+	atomic_notifier_chain_unregister(&idle_notifier, n);
+}
+EXPORT_SYMBOL_GPL(idle_notifier_unregister);
+
+void idle_notifier_call_chain(unsigned long val)
+{
+	atomic_notifier_call_chain(&idle_notifier, val, NULL);
+}
+EXPORT_SYMBOL_GPL(idle_notifier_call_chain);
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index dd3ae6ee064d..db86d6b609b6 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -99,6 +99,7 @@ struct cpuset {
 
 	/* user-configured CPUs and Memory Nodes allow to tasks */
 	cpumask_var_t cpus_allowed;
+	cpumask_var_t cpus_requested;
 	nodemask_t mems_allowed;
 
 	/* effective CPUs and Memory Nodes allow to tasks */
@@ -398,7 +399,7 @@ static void cpuset_update_task_spread_flag(struct cpuset *cs,
 
 static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
 {
-	return	cpumask_subset(p->cpus_allowed, q->cpus_allowed) &&
+	return	cpumask_subset(p->cpus_requested, q->cpus_requested) &&
 		nodes_subset(p->mems_allowed, q->mems_allowed) &&
 		is_cpu_exclusive(p) <= is_cpu_exclusive(q) &&
 		is_mem_exclusive(p) <= is_mem_exclusive(q);
@@ -498,7 +499,7 @@ static int validate_change(struct cpuset *cur, struct cpuset *trial)
 	cpuset_for_each_child(c, css, par) {
 		if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) &&
 		    c != cur &&
-		    cpumask_intersects(trial->cpus_allowed, c->cpus_allowed))
+		    cpumask_intersects(trial->cpus_requested, c->cpus_requested))
 			goto out;
 		if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) &&
 		    c != cur &&
@@ -957,17 +958,18 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 	if (!*buf) {
 		cpumask_clear(trialcs->cpus_allowed);
 	} else {
-		retval = cpulist_parse(buf, trialcs->cpus_allowed);
+		retval = cpulist_parse(buf, trialcs->cpus_requested);
 		if (retval < 0)
 			return retval;
 
-		if (!cpumask_subset(trialcs->cpus_allowed,
-				    top_cpuset.cpus_allowed))
+		if (!cpumask_subset(trialcs->cpus_requested, cpu_present_mask))
 			return -EINVAL;
+
+		cpumask_and(trialcs->cpus_allowed, trialcs->cpus_requested, cpu_active_mask);
 	}
 
 	/* Nothing to do if the cpus didn't change */
-	if (cpumask_equal(cs->cpus_allowed, trialcs->cpus_allowed))
+	if (cpumask_equal(cs->cpus_requested, trialcs->cpus_requested))
 		return 0;
 
 	retval = validate_change(cs, trialcs);
@@ -976,6 +978,7 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
 
 	spin_lock_irq(&callback_lock);
 	cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed);
+	cpumask_copy(cs->cpus_requested, trialcs->cpus_requested);
 	spin_unlock_irq(&callback_lock);
 
 	/* use trialcs->cpus_allowed as a temp variable */
@@ -1766,7 +1769,7 @@ static int cpuset_common_seq_show(struct seq_file *sf, void *v)
 
 	switch (type) {
 	case FILE_CPULIST:
-		seq_printf(sf, "%*pbl\n", cpumask_pr_args(cs->cpus_allowed));
+		seq_printf(sf, "%*pbl\n", cpumask_pr_args(cs->cpus_requested));
 		break;
 	case FILE_MEMLIST:
 		seq_printf(sf, "%*pbl\n", nodemask_pr_args(&cs->mems_allowed));
@@ -1956,11 +1959,14 @@ cpuset_css_alloc(struct cgroup_subsys_state *parent_css)
 		return ERR_PTR(-ENOMEM);
 	if (!alloc_cpumask_var(&cs->cpus_allowed, GFP_KERNEL))
 		goto free_cs;
+	if (!alloc_cpumask_var(&cs->cpus_requested, GFP_KERNEL))
+		goto free_allowed;
 	if (!alloc_cpumask_var(&cs->effective_cpus, GFP_KERNEL))
-		goto free_cpus;
+		goto free_requested;
 
 	set_bit(CS_SCHED_LOAD_BALANCE, &cs->flags);
 	cpumask_clear(cs->cpus_allowed);
+	cpumask_clear(cs->cpus_requested);
 	nodes_clear(cs->mems_allowed);
 	cpumask_clear(cs->effective_cpus);
 	nodes_clear(cs->effective_mems);
@@ -1969,7 +1975,9 @@ cpuset_css_alloc(struct cgroup_subsys_state *parent_css)
 
 	return &cs->css;
 
-free_cpus:
+free_requested:
+	free_cpumask_var(cs->cpus_requested);
+free_allowed:
 	free_cpumask_var(cs->cpus_allowed);
 free_cs:
 	kfree(cs);
@@ -2032,6 +2040,7 @@ static int cpuset_css_online(struct cgroup_subsys_state *css)
 	cs->mems_allowed = parent->mems_allowed;
 	cs->effective_mems = parent->mems_allowed;
 	cpumask_copy(cs->cpus_allowed, parent->cpus_allowed);
+	cpumask_copy(cs->cpus_requested, parent->cpus_requested);
 	cpumask_copy(cs->effective_cpus, parent->cpus_allowed);
 	spin_unlock_irq(&callback_lock);
 out_unlock:
@@ -2066,6 +2075,7 @@ static void cpuset_css_free(struct cgroup_subsys_state *css)
 
 	free_cpumask_var(cs->effective_cpus);
 	free_cpumask_var(cs->cpus_allowed);
+	free_cpumask_var(cs->cpus_requested);
 	kfree(cs);
 }
 
@@ -2130,8 +2140,11 @@ int __init cpuset_init(void)
 		BUG();
 	if (!alloc_cpumask_var(&top_cpuset.effective_cpus, GFP_KERNEL))
 		BUG();
+	if (!alloc_cpumask_var(&top_cpuset.cpus_requested, GFP_KERNEL))
+		BUG();
 
 	cpumask_setall(top_cpuset.cpus_allowed);
+	cpumask_setall(top_cpuset.cpus_requested);
 	nodes_setall(top_cpuset.mems_allowed);
 	cpumask_setall(top_cpuset.effective_cpus);
 	nodes_setall(top_cpuset.effective_mems);
@@ -2265,7 +2278,7 @@ retry:
 		goto retry;
 	}
 
-	cpumask_and(&new_cpus, cs->cpus_allowed, parent_cs(cs)->effective_cpus);
+	cpumask_and(&new_cpus, cs->cpus_requested, parent_cs(cs)->effective_cpus);
 	nodes_and(new_mems, cs->mems_allowed, parent_cs(cs)->effective_mems);
 
 	cpus_updated = !cpumask_equal(&new_cpus, cs->effective_cpus);
diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
index fc1ef736253c..0b891286a150 100644
--- a/kernel/debug/kdb/kdb_io.c
+++ b/kernel/debug/kdb/kdb_io.c
@@ -216,7 +216,7 @@ static char *kdb_read(char *buffer, size_t bufsize)
 	int i;
 	int diag, dtab_count;
 	int key;
-
+	static int last_crlf;
 
 	diag = kdbgetintenv("DTABCOUNT", &dtab_count);
 	if (diag)
@@ -237,6 +237,9 @@ poll_again:
 		return buffer;
 	if (key != 9)
 		tab = 0;
+	if (key != 10 && key != 13)
+		last_crlf = 0;
+
 	switch (key) {
 	case 8: /* backspace */
 		if (cp > buffer) {
@@ -254,7 +257,12 @@ poll_again:
 			*cp = tmp;
 		}
 		break;
-	case 13: /* enter */
+	case 10: /* new line */
+	case 13: /* carriage return */
+		/* handle \n after \r */
+		if (last_crlf && last_crlf != key)
+			break;
+		last_crlf = key;
 		*lastchar++ = '\n';
 		*lastchar++ = '\0';
 		if (!KDB_STATE(KGDB_TRANS)) {
diff --git a/kernel/debug/kdb/kdb_main.c b/kernel/debug/kdb/kdb_main.c
index 4121345498e0..2a20c0dfdafc 100644
--- a/kernel/debug/kdb/kdb_main.c
+++ b/kernel/debug/kdb/kdb_main.c
@@ -2021,7 +2021,7 @@ static int kdb_lsmod(int argc, const char **argv)
 			continue;
 
 		kdb_printf("%-20s%8u  0x%p ", mod->name,
-			   mod->core_size, (void *)mod);
+			   mod->core_layout.size, (void *)mod);
 #ifdef CONFIG_MODULE_UNLOAD
 		kdb_printf("%4d ", module_refcount(mod));
 #endif
@@ -2031,7 +2031,7 @@ static int kdb_lsmod(int argc, const char **argv)
 			kdb_printf(" (Loading)");
 		else
 			kdb_printf(" (Live)");
-		kdb_printf(" 0x%p", mod->module_core);
+		kdb_printf(" 0x%p", mod->core_layout.base);
 
 #ifdef CONFIG_MODULE_UNLOAD
 		{
diff --git a/kernel/events/core.c b/kernel/events/core.c
index 936419f24652..2cd263d3ad14 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -175,8 +175,13 @@ static struct srcu_struct pmus_srcu;
  *   0 - disallow raw tracepoint access for unpriv
  *   1 - disallow cpu events for unpriv
  *   2 - disallow kernel profiling for unpriv
+ *   3 - disallow all unpriv perf event use
  */
+#ifdef CONFIG_SECURITY_PERF_EVENTS_RESTRICT
+int sysctl_perf_event_paranoid __read_mostly = 3;
+#else
 int sysctl_perf_event_paranoid __read_mostly = 1;
+#endif
 
 /* Minimum for 512 kiB + 1 user control page */
 int sysctl_perf_event_mlock __read_mostly = 512 + (PAGE_SIZE / 1024); /* 'free' kiB per user */
@@ -8447,6 +8452,9 @@ SYSCALL_DEFINE5(perf_event_open,
 	if (flags & ~PERF_FLAG_ALL)
 		return -EINVAL;
 
+	if (perf_paranoid_any() && !capable(CAP_SYS_ADMIN))
+		return -EACCES;
+
 	err = perf_copy_attr(attr_uptr, &attr);
 	if (err)
 		return err;
diff --git a/kernel/exit.c b/kernel/exit.c
index ffba5df4abd5..62c4bd4abd3a 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -54,6 +54,8 @@
 #include <linux/writeback.h>
 #include <linux/shm.h>
 
+#include "sched/tune.h"
+
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
 #include <asm/pgtable.h>
@@ -699,6 +701,9 @@ void do_exit(long code)
 	}
 
 	exit_signals(tsk);  /* sets PF_EXITING */
+
+	schedtune_exit_task(tsk);
+
 	/*
 	 * tsk->flags are checked in the futex code to protect against
 	 * an exiting task cleaning up the robust pi futexes.
diff --git a/kernel/fork.c b/kernel/fork.c
index 68cfda1c1800..2633d5f61d3c 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -147,18 +147,18 @@ static inline void free_task_struct(struct task_struct *tsk)
 }
 #endif
 
-void __weak arch_release_thread_info(struct thread_info *ti)
+void __weak arch_release_thread_stack(unsigned long *stack)
 {
 }
 
-#ifndef CONFIG_ARCH_THREAD_INFO_ALLOCATOR
+#ifndef CONFIG_ARCH_THREAD_STACK_ALLOCATOR
 
 /*
  * Allocate pages if THREAD_SIZE is >= PAGE_SIZE, otherwise use a
  * kmemcache based allocator.
  */
 # if THREAD_SIZE >= PAGE_SIZE
-static struct thread_info *alloc_thread_info_node(struct task_struct *tsk,
+static unsigned long *alloc_thread_stack_node(struct task_struct *tsk,
 						  int node)
 {
 	struct page *page = alloc_kmem_pages_node(node, THREADINFO_GFP,
@@ -167,29 +167,31 @@ static struct thread_info *alloc_thread_info_node(struct task_struct *tsk,
 	return page ? page_address(page) : NULL;
 }
 
-static inline void free_thread_info(struct thread_info *ti)
+static inline void free_thread_stack(unsigned long *stack)
 {
-	free_kmem_pages((unsigned long)ti, THREAD_SIZE_ORDER);
+	struct page *page = virt_to_page(stack);
+
+	__free_kmem_pages(page, THREAD_SIZE_ORDER);
 }
 # else
-static struct kmem_cache *thread_info_cache;
+static struct kmem_cache *thread_stack_cache;
 
-static struct thread_info *alloc_thread_info_node(struct task_struct *tsk,
+static unsigned long *alloc_thread_stack_node(struct task_struct *tsk,
 						  int node)
 {
-	return kmem_cache_alloc_node(thread_info_cache, THREADINFO_GFP, node);
+	return kmem_cache_alloc_node(thread_stack_cache, THREADINFO_GFP, node);
 }
 
-static void free_thread_info(struct thread_info *ti)
+static void free_thread_stack(unsigned long *stack)
 {
-	kmem_cache_free(thread_info_cache, ti);
+	kmem_cache_free(thread_stack_cache, stack);
 }
 
-void thread_info_cache_init(void)
+void thread_stack_cache_init(void)
 {
-	thread_info_cache = kmem_cache_create("thread_info", THREAD_SIZE,
+	thread_stack_cache = kmem_cache_create("thread_stack", THREAD_SIZE,
 					      THREAD_SIZE, 0, NULL);
-	BUG_ON(thread_info_cache == NULL);
+	BUG_ON(thread_stack_cache == NULL);
 }
 # endif
 #endif
@@ -212,9 +214,9 @@ struct kmem_cache *vm_area_cachep;
 /* SLAB cache for mm_struct structures (tsk->mm) */
 static struct kmem_cache *mm_cachep;
 
-static void account_kernel_stack(struct thread_info *ti, int account)
+static void account_kernel_stack(unsigned long *stack, int account)
 {
-	struct zone *zone = page_zone(virt_to_page(ti));
+	struct zone *zone = page_zone(virt_to_page(stack));
 
 	mod_zone_page_state(zone, NR_KERNEL_STACK, account);
 }
@@ -222,8 +224,8 @@ static void account_kernel_stack(struct thread_info *ti, int account)
 void free_task(struct task_struct *tsk)
 {
 	account_kernel_stack(tsk->stack, -1);
-	arch_release_thread_info(tsk->stack);
-	free_thread_info(tsk->stack);
+	arch_release_thread_stack(tsk->stack);
+	free_thread_stack(tsk->stack);
 	rt_mutex_debug_task_free(tsk);
 	ftrace_graph_exit_task(tsk);
 	put_seccomp_filter(tsk);
@@ -334,7 +336,7 @@ void set_task_stack_end_magic(struct task_struct *tsk)
 static struct task_struct *dup_task_struct(struct task_struct *orig, int node)
 {
 	struct task_struct *tsk;
-	struct thread_info *ti;
+	unsigned long *stack;
 	int err;
 
 	if (node == NUMA_NO_NODE)
@@ -343,15 +345,15 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node)
 	if (!tsk)
 		return NULL;
 
-	ti = alloc_thread_info_node(tsk, node);
-	if (!ti)
+	stack = alloc_thread_stack_node(tsk, node);
+	if (!stack)
 		goto free_tsk;
 
 	err = arch_dup_task_struct(tsk, orig);
 	if (err)
-		goto free_ti;
+		goto free_stack;
 
-	tsk->stack = ti;
+	tsk->stack = stack;
 #ifdef CONFIG_SECCOMP
 	/*
 	 * We must handle setting up seccomp filters once we're under
@@ -383,12 +385,12 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node)
 	tsk->task_frag.page = NULL;
 	tsk->wake_q.next = NULL;
 
-	account_kernel_stack(ti, 1);
+	account_kernel_stack(stack, 1);
 
 	return tsk;
 
-free_ti:
-	free_thread_info(ti);
+free_stack:
+	free_thread_stack(stack);
 free_tsk:
 	free_task_struct(tsk);
 	return NULL;
@@ -693,6 +695,26 @@ void __mmdrop(struct mm_struct *mm)
 }
 EXPORT_SYMBOL_GPL(__mmdrop);
 
+static inline void __mmput(struct mm_struct *mm)
+{
+	VM_BUG_ON(atomic_read(&mm->mm_users));
+
+	uprobe_clear_state(mm);
+	exit_aio(mm);
+	ksm_exit(mm);
+	khugepaged_exit(mm); /* must run before exit_mmap */
+	exit_mmap(mm);
+	set_mm_exe_file(mm, NULL);
+	if (!list_empty(&mm->mmlist)) {
+		spin_lock(&mmlist_lock);
+		list_del(&mm->mmlist);
+		spin_unlock(&mmlist_lock);
+	}
+	if (mm->binfmt)
+		module_put(mm->binfmt->module);
+	mmdrop(mm);
+}
+
 /*
  * Decrement the use count and release all resources for an mm.
  */
@@ -700,24 +722,24 @@ void mmput(struct mm_struct *mm)
 {
 	might_sleep();
 
+	if (atomic_dec_and_test(&mm->mm_users))
+		__mmput(mm);
+}
+EXPORT_SYMBOL_GPL(mmput);
+
+static void mmput_async_fn(struct work_struct *work)
+{
+	struct mm_struct *mm = container_of(work, struct mm_struct, async_put_work);
+	__mmput(mm);
+}
+
+void mmput_async(struct mm_struct *mm)
+{
 	if (atomic_dec_and_test(&mm->mm_users)) {
-		uprobe_clear_state(mm);
-		exit_aio(mm);
-		ksm_exit(mm);
-		khugepaged_exit(mm); /* must run before exit_mmap */
-		exit_mmap(mm);
-		set_mm_exe_file(mm, NULL);
-		if (!list_empty(&mm->mmlist)) {
-			spin_lock(&mmlist_lock);
-			list_del(&mm->mmlist);
-			spin_unlock(&mmlist_lock);
-		}
-		if (mm->binfmt)
-			module_put(mm->binfmt->module);
-		mmdrop(mm);
+		INIT_WORK(&mm->async_put_work, mmput_async_fn);
+		schedule_work(&mm->async_put_work);
 	}
 }
-EXPORT_SYMBOL_GPL(mmput);
 
 /**
  * set_mm_exe_file - change a reference to the mm's executable file
diff --git a/kernel/gcov/base.c b/kernel/gcov/base.c
index f850e906564b..c51a49c9be70 100644
--- a/kernel/gcov/base.c
+++ b/kernel/gcov/base.c
@@ -129,11 +129,6 @@ void gcov_enable_events(void)
 }
 
 #ifdef CONFIG_MODULES
-static inline int within(void *addr, void *start, unsigned long size)
-{
-	return ((addr >= start) && (addr < start + size));
-}
-
 /* Update list and generate events when modules are unloaded. */
 static int gcov_module_notifier(struct notifier_block *nb, unsigned long event,
 				void *data)
@@ -148,7 +143,7 @@ static int gcov_module_notifier(struct notifier_block *nb, unsigned long event,
 
 	/* Remove entries located in module from linked list. */
 	while ((info = gcov_info_next(info))) {
-		if (within(info, mod->module_core, mod->core_size)) {
+		if (within_module((unsigned long)info, mod)) {
 			gcov_info_unlink(prev, info);
 			if (gcov_events_enabled)
 				gcov_event(GCOV_REMOVE, info);
diff --git a/kernel/kthread.c b/kernel/kthread.c
index 850b255649a2..d9b0be5c6a5f 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -65,7 +65,7 @@ static inline struct kthread *to_kthread(struct task_struct *k)
 static struct kthread *to_live_kthread(struct task_struct *k)
 {
 	struct completion *vfork = ACCESS_ONCE(k->vfork_done);
-	if (likely(vfork))
+	if (likely(vfork) && try_get_task_stack(k))
 		return __to_kthread(vfork);
 	return NULL;
 }
@@ -427,8 +427,10 @@ void kthread_unpark(struct task_struct *k)
 {
 	struct kthread *kthread = to_live_kthread(k);
 
-	if (kthread)
+	if (kthread) {
 		__kthread_unpark(k, kthread);
+		put_task_stack(k);
+	}
 }
 EXPORT_SYMBOL_GPL(kthread_unpark);
 
@@ -457,6 +459,7 @@ int kthread_park(struct task_struct *k)
 				wait_for_completion(&kthread->parked);
 			}
 		}
+		put_task_stack(k);
 		ret = 0;
 	}
 	return ret;
@@ -492,6 +495,7 @@ int kthread_stop(struct task_struct *k)
 		__kthread_unpark(k, kthread);
 		wake_up_process(k);
 		wait_for_completion(&kthread->exited);
+		put_task_stack(k);
 	}
 	ret = k->exit_code;
 	put_task_struct(k);
@@ -604,6 +608,19 @@ repeat:
 }
 EXPORT_SYMBOL_GPL(kthread_worker_fn);
 
+/*
+ * Returns true when the work could not be queued at the moment.
+ * It happens when it is already pending in a worker list
+ * or when it is being cancelled.
+ */
+static inline bool queuing_blocked(struct kthread_worker *worker,
+				   struct kthread_work *work)
+{
+	lockdep_assert_held(&worker->lock);
+
+	return !list_empty(&work->node) || work->canceling;
+}
+
 /* insert @work before @pos in @worker */
 static void insert_kthread_work(struct kthread_worker *worker,
 			       struct kthread_work *work,
@@ -633,7 +650,7 @@ bool queue_kthread_work(struct kthread_worker *worker,
 	unsigned long flags;
 
 	spin_lock_irqsave(&worker->lock, flags);
-	if (list_empty(&work->node)) {
+	if (!queuing_blocked(worker, work)) {
 		insert_kthread_work(worker, work, &worker->work_list);
 		ret = true;
 	}
@@ -694,6 +711,87 @@ retry:
 }
 EXPORT_SYMBOL_GPL(flush_kthread_work);
 
+/*
+ * This function removes the work from the worker queue. Also it makes sure
+ * that it won't get queued later via the delayed work's timer.
+ *
+ * The work might still be in use when this function finishes. See the
+ * current_work proceed by the worker.
+ *
+ * Return: %true if @work was pending and successfully canceled,
+ *	%false if @work was not pending
+ */
+static bool __kthread_cancel_work(struct kthread_work *work,
+				  unsigned long *flags)
+{
+	/*
+	 * Try to remove the work from a worker list. It might either
+	 * be from worker->work_list or from worker->delayed_work_list.
+	 */
+	if (!list_empty(&work->node)) {
+		list_del_init(&work->node);
+		return true;
+	}
+
+	return false;
+}
+
+static bool __kthread_cancel_work_sync(struct kthread_work *work)
+{
+	struct kthread_worker *worker = work->worker;
+	unsigned long flags;
+	int ret = false;
+
+	if (!worker)
+		goto out;
+
+	spin_lock_irqsave(&worker->lock, flags);
+	/* Work must not be used with >1 worker, see kthread_queue_work(). */
+	WARN_ON_ONCE(work->worker != worker);
+
+	ret = __kthread_cancel_work(work, &flags);
+
+	if (worker->current_work != work)
+		goto out_fast;
+
+	/*
+	 * The work is in progress and we need to wait with the lock released.
+	 * In the meantime, block any queuing by setting the canceling counter.
+	 */
+	work->canceling++;
+	spin_unlock_irqrestore(&worker->lock, flags);
+	flush_kthread_work(work);
+	spin_lock_irqsave(&worker->lock, flags);
+	work->canceling--;
+
+out_fast:
+	spin_unlock_irqrestore(&worker->lock, flags);
+out:
+	return ret;
+}
+
+/**
+ * kthread_cancel_work_sync - cancel a kthread work and wait for it to finish
+ * @work: the kthread work to cancel
+ *
+ * Cancel @work and wait for its execution to finish.  This function
+ * can be used even if the work re-queues itself. On return from this
+ * function, @work is guaranteed to be not pending or executing on any CPU.
+ *
+ * kthread_cancel_work_sync(&delayed_work->work) must not be used for
+ * delayed_work's. Use kthread_cancel_delayed_work_sync() instead.
+ *
+ * The caller must ensure that the worker on which @work was last
+ * queued can't be destroyed before this function returns.
+ *
+ * Return: %true if @work was pending, %false otherwise.
+ */
+bool kthread_cancel_work_sync(struct kthread_work *work)
+{
+	return __kthread_cancel_work_sync(work);
+}
+EXPORT_SYMBOL_GPL(kthread_cancel_work_sync);
+
 /**
  * flush_kthread_worker - flush all current works on a kthread_worker
  * @worker: worker to flush
diff --git a/kernel/locking/mutex-debug.c b/kernel/locking/mutex-debug.c
index 3ef3736002d8..9c951fade415 100644
--- a/kernel/locking/mutex-debug.c
+++ b/kernel/locking/mutex-debug.c
@@ -49,21 +49,21 @@ void debug_mutex_free_waiter(struct mutex_waiter *waiter)
 }
 
 void debug_mutex_add_waiter(struct mutex *lock, struct mutex_waiter *waiter,
-			    struct thread_info *ti)
+			    struct task_struct *task)
 {
 	SMP_DEBUG_LOCKS_WARN_ON(!spin_is_locked(&lock->wait_lock));
 
 	/* Mark the current thread as blocked on the lock: */
-	ti->task->blocked_on = waiter;
+	task->blocked_on = waiter;
 }
 
 void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter,
-			 struct thread_info *ti)
+			 struct task_struct *task)
 {
 	DEBUG_LOCKS_WARN_ON(list_empty(&waiter->list));
-	DEBUG_LOCKS_WARN_ON(waiter->task != ti->task);
-	DEBUG_LOCKS_WARN_ON(ti->task->blocked_on != waiter);
-	ti->task->blocked_on = NULL;
+	DEBUG_LOCKS_WARN_ON(waiter->task != task);
+	DEBUG_LOCKS_WARN_ON(task->blocked_on != waiter);
+	task->blocked_on = NULL;
 
 	list_del_init(&waiter->list);
 	waiter->task = NULL;
diff --git a/kernel/locking/mutex-debug.h b/kernel/locking/mutex-debug.h
index 0799fd3e4cfa..d06ae3bb46c5 100644
--- a/kernel/locking/mutex-debug.h
+++ b/kernel/locking/mutex-debug.h
@@ -20,9 +20,9 @@ extern void debug_mutex_wake_waiter(struct mutex *lock,
 extern void debug_mutex_free_waiter(struct mutex_waiter *waiter);
 extern void debug_mutex_add_waiter(struct mutex *lock,
 				   struct mutex_waiter *waiter,
-				   struct thread_info *ti);
+				   struct task_struct *task);
 extern void mutex_remove_waiter(struct mutex *lock, struct mutex_waiter *waiter,
-				struct thread_info *ti);
+				struct task_struct *task);
 extern void debug_mutex_unlock(struct mutex *lock);
 extern void debug_mutex_init(struct mutex *lock, const char *name,
 			     struct lock_class_key *key);
diff --git a/kernel/locking/mutex.c b/kernel/locking/mutex.c
index 89350f924c85..f42f83a36506 100644
--- a/kernel/locking/mutex.c
+++ b/kernel/locking/mutex.c
@@ -537,7 +537,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
 		goto skip_wait;
 
 	debug_mutex_lock_common(lock, &waiter);
-	debug_mutex_add_waiter(lock, &waiter, task_thread_info(task));
+	debug_mutex_add_waiter(lock, &waiter, task);
 
 	/* add waiting tasks to the end of the waitqueue (FIFO): */
 	list_add_tail(&waiter.list, &lock->wait_list);
@@ -584,7 +584,7 @@ __mutex_lock_common(struct mutex *lock, long state, unsigned int subclass,
 	}
 	__set_task_state(task, TASK_RUNNING);
 
-	mutex_remove_waiter(lock, &waiter, current_thread_info());
+	mutex_remove_waiter(lock, &waiter, task);
 	/* set it to 0 if there are no waiters left: */
 	if (likely(list_empty(&lock->wait_list)))
 		atomic_set(&lock->count, 0);
@@ -605,7 +605,7 @@ skip_wait:
 	return 0;
 
 err:
-	mutex_remove_waiter(lock, &waiter, task_thread_info(task));
+	mutex_remove_waiter(lock, &waiter, task);
 	spin_unlock_mutex(&lock->wait_lock, flags);
 	debug_mutex_free_waiter(&waiter);
 	mutex_release(&lock->dep_map, 1, ip);
diff --git a/kernel/locking/mutex.h b/kernel/locking/mutex.h
index 5cda397607f2..a68bae5e852a 100644
--- a/kernel/locking/mutex.h
+++ b/kernel/locking/mutex.h
@@ -13,7 +13,7 @@
 		do { spin_lock(lock); (void)(flags); } while (0)
 #define spin_unlock_mutex(lock, flags) \
 		do { spin_unlock(lock); (void)(flags); } while (0)
-#define mutex_remove_waiter(lock, waiter, ti) \
+#define mutex_remove_waiter(lock, waiter, task) \
 		__list_del((waiter)->list.prev, (waiter)->list.next)
 
 #ifdef CONFIG_MUTEX_SPIN_ON_OWNER
diff --git a/kernel/locking/percpu-rwsem.c b/kernel/locking/percpu-rwsem.c
index f231e0bb311c..ce182599cf2e 100644
--- a/kernel/locking/percpu-rwsem.c
+++ b/kernel/locking/percpu-rwsem.c
@@ -8,151 +8,186 @@
 #include <linux/sched.h>
 #include <linux/errno.h>
 
-int __percpu_init_rwsem(struct percpu_rw_semaphore *brw,
+int __percpu_init_rwsem(struct percpu_rw_semaphore *sem,
 			const char *name, struct lock_class_key *rwsem_key)
 {
-	brw->fast_read_ctr = alloc_percpu(int);
-	if (unlikely(!brw->fast_read_ctr))
+	sem->read_count = alloc_percpu(int);
+	if (unlikely(!sem->read_count))
 		return -ENOMEM;
 
 	/* ->rw_sem represents the whole percpu_rw_semaphore for lockdep */
-	__init_rwsem(&brw->rw_sem, name, rwsem_key);
-	rcu_sync_init(&brw->rss, RCU_SCHED_SYNC);
-	atomic_set(&brw->slow_read_ctr, 0);
-	init_waitqueue_head(&brw->write_waitq);
+	rcu_sync_init(&sem->rss, RCU_SCHED_SYNC);
+	__init_rwsem(&sem->rw_sem, name, rwsem_key);
+	init_waitqueue_head(&sem->writer);
+	sem->readers_block = 0;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(__percpu_init_rwsem);
 
-void percpu_free_rwsem(struct percpu_rw_semaphore *brw)
+void percpu_free_rwsem(struct percpu_rw_semaphore *sem)
 {
 	/*
 	 * XXX: temporary kludge. The error path in alloc_super()
 	 * assumes that percpu_free_rwsem() is safe after kzalloc().
 	 */
-	if (!brw->fast_read_ctr)
+	if (!sem->read_count)
 		return;
 
-	rcu_sync_dtor(&brw->rss);
-	free_percpu(brw->fast_read_ctr);
-	brw->fast_read_ctr = NULL; /* catch use after free bugs */
+	rcu_sync_dtor(&sem->rss);
+	free_percpu(sem->read_count);
+	sem->read_count = NULL; /* catch use after free bugs */
 }
+EXPORT_SYMBOL_GPL(percpu_free_rwsem);
 
-/*
- * This is the fast-path for down_read/up_read. If it succeeds we rely
- * on the barriers provided by rcu_sync_enter/exit; see the comments in
- * percpu_down_write() and percpu_up_write().
- *
- * If this helper fails the callers rely on the normal rw_semaphore and
- * atomic_dec_and_test(), so in this case we have the necessary barriers.
- */
-static bool update_fast_ctr(struct percpu_rw_semaphore *brw, unsigned int val)
+int __percpu_down_read(struct percpu_rw_semaphore *sem, int try)
 {
-	bool success;
+	/*
+	 * Due to having preemption disabled the decrement happens on
+	 * the same CPU as the increment, avoiding the
+	 * increment-on-one-CPU-and-decrement-on-another problem.
+	 *
+	 * If the reader misses the writer's assignment of readers_block, then
+	 * the writer is guaranteed to see the reader's increment.
+	 *
+	 * Conversely, any readers that increment their sem->read_count after
+	 * the writer looks are guaranteed to see the readers_block value,
+	 * which in turn means that they are guaranteed to immediately
+	 * decrement their sem->read_count, so that it doesn't matter that the
+	 * writer missed them.
+	 */
 
-	preempt_disable();
-	success = rcu_sync_is_idle(&brw->rss);
-	if (likely(success))
-		__this_cpu_add(*brw->fast_read_ctr, val);
-	preempt_enable();
+	smp_mb(); /* A matches D */
 
-	return success;
-}
+	/*
+	 * If !readers_block the critical section starts here, matched by the
+	 * release in percpu_up_write().
+	 */
+	if (likely(!smp_load_acquire(&sem->readers_block)))
+		return 1;
 
-/*
- * Like the normal down_read() this is not recursive, the writer can
- * come after the first percpu_down_read() and create the deadlock.
- *
- * Note: returns with lock_is_held(brw->rw_sem) == T for lockdep,
- * percpu_up_read() does rwsem_release(). This pairs with the usage
- * of ->rw_sem in percpu_down/up_write().
- */
-void percpu_down_read(struct percpu_rw_semaphore *brw)
-{
-	might_sleep();
-	rwsem_acquire_read(&brw->rw_sem.dep_map, 0, 0, _RET_IP_);
+	/*
+	 * Per the above comment; we still have preemption disabled and
+	 * will thus decrement on the same CPU as we incremented.
+	 */
+	__percpu_up_read(sem);
 
-	if (likely(update_fast_ctr(brw, +1)))
-		return;
+	if (try)
+		return 0;
 
-	/* Avoid rwsem_acquire_read() and rwsem_release() */
-	__down_read(&brw->rw_sem);
-	atomic_inc(&brw->slow_read_ctr);
-	__up_read(&brw->rw_sem);
-}
-EXPORT_SYMBOL_GPL(percpu_down_read);
+	/*
+	 * We either call schedule() in the wait, or we'll fall through
+	 * and reschedule on the preempt_enable() in percpu_down_read().
+	 */
+	preempt_enable_no_resched();
 
-int percpu_down_read_trylock(struct percpu_rw_semaphore *brw)
-{
-	if (unlikely(!update_fast_ctr(brw, +1))) {
-		if (!__down_read_trylock(&brw->rw_sem))
-			return 0;
-		atomic_inc(&brw->slow_read_ctr);
-		__up_read(&brw->rw_sem);
-	}
-
-	rwsem_acquire_read(&brw->rw_sem.dep_map, 0, 1, _RET_IP_);
+	/*
+	 * Avoid lockdep for the down/up_read() we already have them.
+	 */
+	__down_read(&sem->rw_sem);
+	this_cpu_inc(*sem->read_count);
+	__up_read(&sem->rw_sem);
+
+	preempt_disable();
 	return 1;
 }
+EXPORT_SYMBOL_GPL(__percpu_down_read);
 
-void percpu_up_read(struct percpu_rw_semaphore *brw)
+void __percpu_up_read(struct percpu_rw_semaphore *sem)
 {
-	rwsem_release(&brw->rw_sem.dep_map, 1, _RET_IP_);
-
-	if (likely(update_fast_ctr(brw, -1)))
-		return;
+	smp_mb(); /* B matches C */
+	/*
+	 * In other words, if they see our decrement (presumably to aggregate
+	 * zero, as that is the only time it matters) they will also see our
+	 * critical section.
+	 */
+	__this_cpu_dec(*sem->read_count);
 
-	/* false-positive is possible but harmless */
-	if (atomic_dec_and_test(&brw->slow_read_ctr))
-		wake_up_all(&brw->write_waitq);
+	/* Prod writer to recheck readers_active */
+	wake_up(&sem->writer);
 }
-EXPORT_SYMBOL_GPL(percpu_up_read);
+EXPORT_SYMBOL_GPL(__percpu_up_read);
+
+#define per_cpu_sum(var)						\
+({									\
+	typeof(var) __sum = 0;						\
+	int cpu;							\
+	compiletime_assert_atomic_type(__sum);				\
+	for_each_possible_cpu(cpu)					\
+		__sum += per_cpu(var, cpu);				\
+	__sum;								\
+})
 
-static int clear_fast_ctr(struct percpu_rw_semaphore *brw)
+/*
+ * Return true if the modular sum of the sem->read_count per-CPU variable is
+ * zero.  If this sum is zero, then it is stable due to the fact that if any
+ * newly arriving readers increment a given counter, they will immediately
+ * decrement that same counter.
+ */
+static bool readers_active_check(struct percpu_rw_semaphore *sem)
 {
-	unsigned int sum = 0;
-	int cpu;
+	if (per_cpu_sum(*sem->read_count) != 0)
+		return false;
+
+	/*
+	 * If we observed the decrement; ensure we see the entire critical
+	 * section.
+	 */
 
-	for_each_possible_cpu(cpu) {
-		sum += per_cpu(*brw->fast_read_ctr, cpu);
-		per_cpu(*brw->fast_read_ctr, cpu) = 0;
-	}
+	smp_mb(); /* C matches B */
 
-	return sum;
+	return true;
 }
 
-void percpu_down_write(struct percpu_rw_semaphore *brw)
+void percpu_down_write(struct percpu_rw_semaphore *sem)
 {
+	/* Notify readers to take the slow path. */
+	rcu_sync_enter(&sem->rss);
+
+	down_write(&sem->rw_sem);
+
 	/*
-	 * Make rcu_sync_is_idle() == F and thus disable the fast-path in
-	 * percpu_down_read() and percpu_up_read(), and wait for gp pass.
-	 *
-	 * The latter synchronises us with the preceding readers which used
-	 * the fast-past, so we can not miss the result of __this_cpu_add()
-	 * or anything else inside their criticial sections.
+	 * Notify new readers to block; up until now, and thus throughout the
+	 * longish rcu_sync_enter() above, new readers could still come in.
 	 */
-	rcu_sync_enter(&brw->rss);
+	WRITE_ONCE(sem->readers_block, 1);
 
-	/* exclude other writers, and block the new readers completely */
-	down_write(&brw->rw_sem);
+	smp_mb(); /* D matches A */
 
-	/* nobody can use fast_read_ctr, move its sum into slow_read_ctr */
-	atomic_add(clear_fast_ctr(brw), &brw->slow_read_ctr);
+	/*
+	 * If they don't see our writer of readers_block, then we are
+	 * guaranteed to see their sem->read_count increment, and therefore
+	 * will wait for them.
+	 */
 
-	/* wait for all readers to complete their percpu_up_read() */
-	wait_event(brw->write_waitq, !atomic_read(&brw->slow_read_ctr));
+	/* Wait for all now active readers to complete. */
+	wait_event(sem->writer, readers_active_check(sem));
 }
 EXPORT_SYMBOL_GPL(percpu_down_write);
 
-void percpu_up_write(struct percpu_rw_semaphore *brw)
+void percpu_up_write(struct percpu_rw_semaphore *sem)
 {
-	/* release the lock, but the readers can't use the fast-path */
-	up_write(&brw->rw_sem);
 	/*
-	 * Enable the fast-path in percpu_down_read() and percpu_up_read()
-	 * but only after another gp pass; this adds the necessary barrier
-	 * to ensure the reader can't miss the changes done by us.
+	 * Signal the writer is done, no fast path yet.
+	 *
+	 * One reason that we cannot just immediately flip to readers_fast is
+	 * that new readers might fail to see the results of this writer's
+	 * critical section.
+	 *
+	 * Therefore we force it through the slow path which guarantees an
+	 * acquire and thereby guarantees the critical section's consistency.
+	 */
+	smp_store_release(&sem->readers_block, 0);
+
+	/*
+	 * Release the write lock, this will allow readers back in the game.
+	 */
+	up_write(&sem->rw_sem);
+
+	/*
+	 * Once this completes (at least one RCU-sched grace period hence) the
+	 * reader fast path will be available again. Safe to use outside the
+	 * exclusive write lock because its counting.
 	 */
-	rcu_sync_exit(&brw->rss);
+	rcu_sync_exit(&sem->rss);
 }
 EXPORT_SYMBOL_GPL(percpu_up_write);
diff --git a/kernel/module.c b/kernel/module.c
index b14a4f31221f..63c54d644bd4 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -80,15 +80,6 @@
 # define debug_align(X) (X)
 #endif
 
-/*
- * Given BASE and SIZE this macro calculates the number of pages the
- * memory regions occupies
- */
-#define MOD_NUMBER_OF_PAGES(BASE, SIZE) (((SIZE) > 0) ?		\
-		(PFN_DOWN((unsigned long)(BASE) + (SIZE) - 1) -	\
-			 PFN_DOWN((unsigned long)BASE) + 1)	\
-		: (0UL))
-
 /* If this is set, the section belongs in the init part of the module */
 #define INIT_OFFSET_MASK (1UL << (BITS_PER_LONG-1))
 
@@ -108,13 +99,6 @@ static LIST_HEAD(modules);
  * Use a latched RB-tree for __module_address(); this allows us to use
  * RCU-sched lookups of the address from any context.
  *
- * Because modules have two address ranges: init and core, we need two
- * latch_tree_nodes entries. Therefore we need the back-pointer from
- * mod_tree_node.
- *
- * Because init ranges are short lived we mark them unlikely and have placed
- * them outside the critical cacheline in struct module.
- *
  * This is conditional on PERF_EVENTS || TRACING because those can really hit
  * __module_address() hard by doing a lot of stack unwinding; potentially from
  * NMI context.
@@ -122,24 +106,16 @@ static LIST_HEAD(modules);
 
 static __always_inline unsigned long __mod_tree_val(struct latch_tree_node *n)
 {
-	struct mod_tree_node *mtn = container_of(n, struct mod_tree_node, node);
-	struct module *mod = mtn->mod;
+	struct module_layout *layout = container_of(n, struct module_layout, mtn.node);
 
-	if (unlikely(mtn == &mod->mtn_init))
-		return (unsigned long)mod->module_init;
-
-	return (unsigned long)mod->module_core;
+	return (unsigned long)layout->base;
 }
 
 static __always_inline unsigned long __mod_tree_size(struct latch_tree_node *n)
 {
-	struct mod_tree_node *mtn = container_of(n, struct mod_tree_node, node);
-	struct module *mod = mtn->mod;
-
-	if (unlikely(mtn == &mod->mtn_init))
-		return (unsigned long)mod->init_size;
+	struct module_layout *layout = container_of(n, struct module_layout, mtn.node);
 
-	return (unsigned long)mod->core_size;
+	return (unsigned long)layout->size;
 }
 
 static __always_inline bool
@@ -197,23 +173,23 @@ static void __mod_tree_remove(struct mod_tree_node *node)
  */
 static void mod_tree_insert(struct module *mod)
 {
-	mod->mtn_core.mod = mod;
-	mod->mtn_init.mod = mod;
+	mod->core_layout.mtn.mod = mod;
+	mod->init_layout.mtn.mod = mod;
 
-	__mod_tree_insert(&mod->mtn_core);
-	if (mod->init_size)
-		__mod_tree_insert(&mod->mtn_init);
+	__mod_tree_insert(&mod->core_layout.mtn);
+	if (mod->init_layout.size)
+		__mod_tree_insert(&mod->init_layout.mtn);
 }
 
 static void mod_tree_remove_init(struct module *mod)
 {
-	if (mod->init_size)
-		__mod_tree_remove(&mod->mtn_init);
+	if (mod->init_layout.size)
+		__mod_tree_remove(&mod->init_layout.mtn);
 }
 
 static void mod_tree_remove(struct module *mod)
 {
-	__mod_tree_remove(&mod->mtn_core);
+	__mod_tree_remove(&mod->core_layout.mtn);
 	mod_tree_remove_init(mod);
 }
 
@@ -267,9 +243,9 @@ static void __mod_update_bounds(void *base, unsigned int size)
 
 static void mod_update_bounds(struct module *mod)
 {
-	__mod_update_bounds(mod->module_core, mod->core_size);
-	if (mod->init_size)
-		__mod_update_bounds(mod->module_init, mod->init_size);
+	__mod_update_bounds(mod->core_layout.base, mod->core_layout.size);
+	if (mod->init_layout.size)
+		__mod_update_bounds(mod->init_layout.base, mod->init_layout.size);
 }
 
 #ifdef CONFIG_KGDB_KDB
@@ -1217,7 +1193,7 @@ struct module_attribute module_uevent =
 static ssize_t show_coresize(struct module_attribute *mattr,
 			     struct module_kobject *mk, char *buffer)
 {
-	return sprintf(buffer, "%u\n", mk->mod->core_size);
+	return sprintf(buffer, "%u\n", mk->mod->core_layout.size);
 }
 
 static struct module_attribute modinfo_coresize =
@@ -1226,7 +1202,7 @@ static struct module_attribute modinfo_coresize =
 static ssize_t show_initsize(struct module_attribute *mattr,
 			     struct module_kobject *mk, char *buffer)
 {
-	return sprintf(buffer, "%u\n", mk->mod->init_size);
+	return sprintf(buffer, "%u\n", mk->mod->init_layout.size);
 }
 
 static struct module_attribute modinfo_initsize =
@@ -1876,64 +1852,75 @@ static void mod_sysfs_teardown(struct module *mod)
 /*
  * LKM RO/NX protection: protect module's text/ro-data
  * from modification and any data from execution.
+ *
+ * General layout of module is:
+ *          [text] [read-only-data] [writable data]
+ * text_size -----^                ^               ^
+ * ro_size ------------------------|               |
+ * size -------------------------------------------|
+ *
+ * These values are always page-aligned (as is base)
  */
-void set_page_attributes(void *start, void *end, int (*set)(unsigned long start, int num_pages))
+static void frob_text(const struct module_layout *layout,
+		      int (*set_memory)(unsigned long start, int num_pages))
 {
-	unsigned long begin_pfn = PFN_DOWN((unsigned long)start);
-	unsigned long end_pfn = PFN_DOWN((unsigned long)end);
+	BUG_ON((unsigned long)layout->base & (PAGE_SIZE-1));
+	BUG_ON((unsigned long)layout->text_size & (PAGE_SIZE-1));
+	set_memory((unsigned long)layout->base,
+		   layout->text_size >> PAGE_SHIFT);
+}
 
-	if (end_pfn > begin_pfn)
-		set(begin_pfn << PAGE_SHIFT, end_pfn - begin_pfn);
+static void frob_rodata(const struct module_layout *layout,
+			int (*set_memory)(unsigned long start, int num_pages))
+{
+	BUG_ON((unsigned long)layout->base & (PAGE_SIZE-1));
+	BUG_ON((unsigned long)layout->text_size & (PAGE_SIZE-1));
+	BUG_ON((unsigned long)layout->ro_size & (PAGE_SIZE-1));
+	set_memory((unsigned long)layout->base + layout->text_size,
+		   (layout->ro_size - layout->text_size) >> PAGE_SHIFT);
 }
 
-static void set_section_ro_nx(void *base,
-			unsigned long text_size,
-			unsigned long ro_size,
-			unsigned long total_size)
+static void frob_writable_data(const struct module_layout *layout,
+			       int (*set_memory)(unsigned long start, int num_pages))
 {
-	/* begin and end PFNs of the current subsection */
-	unsigned long begin_pfn;
-	unsigned long end_pfn;
+	BUG_ON((unsigned long)layout->base & (PAGE_SIZE-1));
+	BUG_ON((unsigned long)layout->ro_size & (PAGE_SIZE-1));
+	BUG_ON((unsigned long)layout->size & (PAGE_SIZE-1));
+	set_memory((unsigned long)layout->base + layout->ro_size,
+		   (layout->size - layout->ro_size) >> PAGE_SHIFT);
+}
 
-	/*
-	 * Set RO for module text and RO-data:
-	 * - Always protect first page.
-	 * - Do not protect last partial page.
-	 */
-	if (ro_size > 0)
-		set_page_attributes(base, base + ro_size, set_memory_ro);
+/* livepatching wants to disable read-only so it can frob module. */
+void module_disable_ro(const struct module *mod)
+{
+	frob_text(&mod->core_layout, set_memory_rw);
+	frob_rodata(&mod->core_layout, set_memory_rw);
+	frob_text(&mod->init_layout, set_memory_rw);
+	frob_rodata(&mod->init_layout, set_memory_rw);
+}
 
-	/*
-	 * Set NX permissions for module data:
-	 * - Do not protect first partial page.
-	 * - Always protect last page.
-	 */
-	if (total_size > text_size) {
-		begin_pfn = PFN_UP((unsigned long)base + text_size);
-		end_pfn = PFN_UP((unsigned long)base + total_size);
-		if (end_pfn > begin_pfn)
-			set_memory_nx(begin_pfn << PAGE_SHIFT, end_pfn - begin_pfn);
-	}
+void module_enable_ro(const struct module *mod)
+{
+	frob_text(&mod->core_layout, set_memory_ro);
+	frob_rodata(&mod->core_layout, set_memory_ro);
+	frob_text(&mod->init_layout, set_memory_ro);
+	frob_rodata(&mod->init_layout, set_memory_ro);
 }
 
-static void unset_module_core_ro_nx(struct module *mod)
+static void module_enable_nx(const struct module *mod)
 {
-	set_page_attributes(mod->module_core + mod->core_text_size,
-		mod->module_core + mod->core_size,
-		set_memory_x);
-	set_page_attributes(mod->module_core,
-		mod->module_core + mod->core_ro_size,
-		set_memory_rw);
+	frob_rodata(&mod->core_layout, set_memory_nx);
+	frob_writable_data(&mod->core_layout, set_memory_nx);
+	frob_rodata(&mod->init_layout, set_memory_nx);
+	frob_writable_data(&mod->init_layout, set_memory_nx);
 }
 
-static void unset_module_init_ro_nx(struct module *mod)
+static void module_disable_nx(const struct module *mod)
 {
-	set_page_attributes(mod->module_init + mod->init_text_size,
-		mod->module_init + mod->init_size,
-		set_memory_x);
-	set_page_attributes(mod->module_init,
-		mod->module_init + mod->init_ro_size,
-		set_memory_rw);
+	frob_rodata(&mod->core_layout, set_memory_x);
+	frob_writable_data(&mod->core_layout, set_memory_x);
+	frob_rodata(&mod->init_layout, set_memory_x);
+	frob_writable_data(&mod->init_layout, set_memory_x);
 }
 
 /* Iterate through all modules and set each module's text as RW */
@@ -1945,16 +1932,9 @@ void set_all_modules_text_rw(void)
 	list_for_each_entry_rcu(mod, &modules, list) {
 		if (mod->state == MODULE_STATE_UNFORMED)
 			continue;
-		if ((mod->module_core) && (mod->core_text_size)) {
-			set_page_attributes(mod->module_core,
-						mod->module_core + mod->core_text_size,
-						set_memory_rw);
-		}
-		if ((mod->module_init) && (mod->init_text_size)) {
-			set_page_attributes(mod->module_init,
-						mod->module_init + mod->init_text_size,
-						set_memory_rw);
-		}
+
+		frob_text(&mod->core_layout, set_memory_rw);
+		frob_text(&mod->init_layout, set_memory_rw);
 	}
 	mutex_unlock(&module_mutex);
 }
@@ -1968,23 +1948,25 @@ void set_all_modules_text_ro(void)
 	list_for_each_entry_rcu(mod, &modules, list) {
 		if (mod->state == MODULE_STATE_UNFORMED)
 			continue;
-		if ((mod->module_core) && (mod->core_text_size)) {
-			set_page_attributes(mod->module_core,
-						mod->module_core + mod->core_text_size,
-						set_memory_ro);
-		}
-		if ((mod->module_init) && (mod->init_text_size)) {
-			set_page_attributes(mod->module_init,
-						mod->module_init + mod->init_text_size,
-						set_memory_ro);
-		}
+
+		frob_text(&mod->core_layout, set_memory_ro);
+		frob_text(&mod->init_layout, set_memory_ro);
 	}
 	mutex_unlock(&module_mutex);
 }
+
+static void disable_ro_nx(const struct module_layout *layout)
+{
+	frob_text(layout, set_memory_rw);
+	frob_rodata(layout, set_memory_rw);
+	frob_rodata(layout, set_memory_x);
+	frob_writable_data(layout, set_memory_x);
+}
+
 #else
-static inline void set_section_ro_nx(void *base, unsigned long text_size, unsigned long ro_size, unsigned long total_size) { }
-static void unset_module_core_ro_nx(struct module *mod) { }
-static void unset_module_init_ro_nx(struct module *mod) { }
+static void disable_ro_nx(const struct module_layout *layout) { }
+static void module_enable_nx(const struct module *mod) { }
+static void module_disable_nx(const struct module *mod) { }
 #endif
 
 void __weak module_memfree(void *module_region)
@@ -2036,19 +2018,19 @@ static void free_module(struct module *mod)
 	synchronize_sched();
 	mutex_unlock(&module_mutex);
 
-	/* This may be NULL, but that's OK */
-	unset_module_init_ro_nx(mod);
+	/* This may be empty, but that's OK */
+	disable_ro_nx(&mod->init_layout);
 	module_arch_freeing_init(mod);
-	module_memfree(mod->module_init);
+	module_memfree(mod->init_layout.base);
 	kfree(mod->args);
 	percpu_modfree(mod);
 
 	/* Free lock-classes; relies on the preceding sync_rcu(). */
-	lockdep_free_key_range(mod->module_core, mod->core_size);
+	lockdep_free_key_range(mod->core_layout.base, mod->core_layout.size);
 
 	/* Finally, free the core (containing the module structure) */
-	unset_module_core_ro_nx(mod);
-	module_memfree(mod->module_core);
+	disable_ro_nx(&mod->core_layout);
+	module_memfree(mod->core_layout.base);
 
 #ifdef CONFIG_MPU
 	update_protections(current->mm);
@@ -2251,20 +2233,20 @@ static void layout_sections(struct module *mod, struct load_info *info)
 			    || s->sh_entsize != ~0UL
 			    || strstarts(sname, ".init"))
 				continue;
-			s->sh_entsize = get_offset(mod, &mod->core_size, s, i);
+			s->sh_entsize = get_offset(mod, &mod->core_layout.size, s, i);
 			pr_debug("\t%s\n", sname);
 		}
 		switch (m) {
 		case 0: /* executable */
-			mod->core_size = debug_align(mod->core_size);
-			mod->core_text_size = mod->core_size;
+			mod->core_layout.size = debug_align(mod->core_layout.size);
+			mod->core_layout.text_size = mod->core_layout.size;
 			break;
 		case 1: /* RO: text and ro-data */
-			mod->core_size = debug_align(mod->core_size);
-			mod->core_ro_size = mod->core_size;
+			mod->core_layout.size = debug_align(mod->core_layout.size);
+			mod->core_layout.ro_size = mod->core_layout.size;
 			break;
 		case 3: /* whole core */
-			mod->core_size = debug_align(mod->core_size);
+			mod->core_layout.size = debug_align(mod->core_layout.size);
 			break;
 		}
 	}
@@ -2280,21 +2262,21 @@ static void layout_sections(struct module *mod, struct load_info *info)
 			    || s->sh_entsize != ~0UL
 			    || !strstarts(sname, ".init"))
 				continue;
-			s->sh_entsize = (get_offset(mod, &mod->init_size, s, i)
+			s->sh_entsize = (get_offset(mod, &mod->init_layout.size, s, i)
 					 | INIT_OFFSET_MASK);
 			pr_debug("\t%s\n", sname);
 		}
 		switch (m) {
 		case 0: /* executable */
-			mod->init_size = debug_align(mod->init_size);
-			mod->init_text_size = mod->init_size;
+			mod->init_layout.size = debug_align(mod->init_layout.size);
+			mod->init_layout.text_size = mod->init_layout.size;
 			break;
 		case 1: /* RO: text and ro-data */
-			mod->init_size = debug_align(mod->init_size);
-			mod->init_ro_size = mod->init_size;
+			mod->init_layout.size = debug_align(mod->init_layout.size);
+			mod->init_layout.ro_size = mod->init_layout.size;
 			break;
 		case 3: /* whole init */
-			mod->init_size = debug_align(mod->init_size);
+			mod->init_layout.size = debug_align(mod->init_layout.size);
 			break;
 		}
 	}
@@ -2404,7 +2386,7 @@ static char elf_type(const Elf_Sym *sym, const struct load_info *info)
 	}
 	if (sym->st_shndx == SHN_UNDEF)
 		return 'U';
-	if (sym->st_shndx == SHN_ABS)
+	if (sym->st_shndx == SHN_ABS || sym->st_shndx == info->index.pcpu)
 		return 'a';
 	if (sym->st_shndx >= SHN_LORESERVE)
 		return '?';
@@ -2433,7 +2415,7 @@ static char elf_type(const Elf_Sym *sym, const struct load_info *info)
 }
 
 static bool is_core_symbol(const Elf_Sym *src, const Elf_Shdr *sechdrs,
-			unsigned int shnum)
+			unsigned int shnum, unsigned int pcpundx)
 {
 	const Elf_Shdr *sec;
 
@@ -2442,6 +2424,11 @@ static bool is_core_symbol(const Elf_Sym *src, const Elf_Shdr *sechdrs,
 	    || !src->st_name)
 		return false;
 
+#ifdef CONFIG_KALLSYMS_ALL
+	if (src->st_shndx == pcpundx)
+		return true;
+#endif
+
 	sec = sechdrs + src->st_shndx;
 	if (!(sec->sh_flags & SHF_ALLOC)
 #ifndef CONFIG_KALLSYMS_ALL
@@ -2469,7 +2456,7 @@ static void layout_symtab(struct module *mod, struct load_info *info)
 
 	/* Put symbol section at end of init part of module. */
 	symsect->sh_flags |= SHF_ALLOC;
-	symsect->sh_entsize = get_offset(mod, &mod->init_size, symsect,
+	symsect->sh_entsize = get_offset(mod, &mod->init_layout.size, symsect,
 					 info->index.sym) | INIT_OFFSET_MASK;
 	pr_debug("\t%s\n", info->secstrings + symsect->sh_name);
 
@@ -2479,30 +2466,31 @@ static void layout_symtab(struct module *mod, struct load_info *info)
 	/* Compute total space required for the core symbols' strtab. */
 	for (ndst = i = 0; i < nsrc; i++) {
 		if (i == 0 ||
-		    is_core_symbol(src+i, info->sechdrs, info->hdr->e_shnum)) {
+		    is_core_symbol(src+i, info->sechdrs, info->hdr->e_shnum,
+				   info->index.pcpu)) {
 			strtab_size += strlen(&info->strtab[src[i].st_name])+1;
 			ndst++;
 		}
 	}
 
 	/* Append room for core symbols at end of core part. */
-	info->symoffs = ALIGN(mod->core_size, symsect->sh_addralign ?: 1);
-	info->stroffs = mod->core_size = info->symoffs + ndst * sizeof(Elf_Sym);
-	mod->core_size += strtab_size;
-	mod->core_size = debug_align(mod->core_size);
+	info->symoffs = ALIGN(mod->core_layout.size, symsect->sh_addralign ?: 1);
+	info->stroffs = mod->core_layout.size = info->symoffs + ndst * sizeof(Elf_Sym);
+	mod->core_layout.size += strtab_size;
+	mod->core_layout.size = debug_align(mod->core_layout.size);
 
 	/* Put string table section at end of init part of module. */
 	strsect->sh_flags |= SHF_ALLOC;
-	strsect->sh_entsize = get_offset(mod, &mod->init_size, strsect,
+	strsect->sh_entsize = get_offset(mod, &mod->init_layout.size, strsect,
 					 info->index.str) | INIT_OFFSET_MASK;
 	pr_debug("\t%s\n", info->secstrings + strsect->sh_name);
 
 	/* We'll tack temporary mod_kallsyms on the end. */
-	mod->init_size = ALIGN(mod->init_size,
-			       __alignof__(struct mod_kallsyms));
-	info->mod_kallsyms_init_off = mod->init_size;
-	mod->init_size += sizeof(struct mod_kallsyms);
-	mod->init_size = debug_align(mod->init_size);
+	mod->init_layout.size = ALIGN(mod->init_layout.size,
+				      __alignof__(struct mod_kallsyms));
+	info->mod_kallsyms_init_off = mod->init_layout.size;
+	mod->init_layout.size += sizeof(struct mod_kallsyms);
+	mod->init_layout.size = debug_align(mod->init_layout.size);
 }
 
 /*
@@ -2519,7 +2507,7 @@ static void add_kallsyms(struct module *mod, const struct load_info *info)
 	Elf_Shdr *symsec = &info->sechdrs[info->index.sym];
 
 	/* Set up to point into init section. */
-	mod->kallsyms = mod->module_init + info->mod_kallsyms_init_off;
+	mod->kallsyms = mod->init_layout.base + info->mod_kallsyms_init_off;
 
 	mod->kallsyms->symtab = (void *)symsec->sh_addr;
 	mod->kallsyms->num_symtab = symsec->sh_size / sizeof(Elf_Sym);
@@ -2532,12 +2520,13 @@ static void add_kallsyms(struct module *mod, const struct load_info *info)
 			= elf_type(&mod->kallsyms->symtab[i], info);
 
 	/* Now populate the cut down core kallsyms for after init. */
-	mod->core_kallsyms.symtab = dst = mod->module_core + info->symoffs;
-	mod->core_kallsyms.strtab = s = mod->module_core + info->stroffs;
+	mod->core_kallsyms.symtab = dst = mod->core_layout.base + info->symoffs;
+	mod->core_kallsyms.strtab = s = mod->core_layout.base + info->stroffs;
 	src = mod->kallsyms->symtab;
 	for (ndst = i = 0; i < mod->kallsyms->num_symtab; i++) {
 		if (i == 0 ||
-		    is_core_symbol(src+i, info->sechdrs, info->hdr->e_shnum)) {
+		    is_core_symbol(src+i, info->sechdrs, info->hdr->e_shnum,
+				   info->index.pcpu)) {
 			dst[ndst] = src[i];
 			dst[ndst++].st_name = s - mod->core_kallsyms.strtab;
 			s += strlcpy(s, &mod->kallsyms->strtab[src[i].st_name],
@@ -2988,7 +2977,7 @@ static int move_module(struct module *mod, struct load_info *info)
 	void *ptr;
 
 	/* Do the allocs. */
-	ptr = module_alloc(mod->core_size);
+	ptr = module_alloc(mod->core_layout.size);
 	/*
 	 * The pointer to this block is stored in the module structure
 	 * which is inside the block. Just mark it as not being a
@@ -2998,11 +2987,11 @@ static int move_module(struct module *mod, struct load_info *info)
 	if (!ptr)
 		return -ENOMEM;
 
-	memset(ptr, 0, mod->core_size);
-	mod->module_core = ptr;
+	memset(ptr, 0, mod->core_layout.size);
+	mod->core_layout.base = ptr;
 
-	if (mod->init_size) {
-		ptr = module_alloc(mod->init_size);
+	if (mod->init_layout.size) {
+		ptr = module_alloc(mod->init_layout.size);
 		/*
 		 * The pointer to this block is stored in the module structure
 		 * which is inside the block. This block doesn't need to be
@@ -3011,13 +3000,13 @@ static int move_module(struct module *mod, struct load_info *info)
 		 */
 		kmemleak_ignore(ptr);
 		if (!ptr) {
-			module_memfree(mod->module_core);
+			module_memfree(mod->core_layout.base);
 			return -ENOMEM;
 		}
-		memset(ptr, 0, mod->init_size);
-		mod->module_init = ptr;
+		memset(ptr, 0, mod->init_layout.size);
+		mod->init_layout.base = ptr;
 	} else
-		mod->module_init = NULL;
+		mod->init_layout.base = NULL;
 
 	/* Transfer each section which specifies SHF_ALLOC */
 	pr_debug("final section addresses:\n");
@@ -3029,10 +3018,10 @@ static int move_module(struct module *mod, struct load_info *info)
 			continue;
 
 		if (shdr->sh_entsize & INIT_OFFSET_MASK)
-			dest = mod->module_init
+			dest = mod->init_layout.base
 				+ (shdr->sh_entsize & ~INIT_OFFSET_MASK);
 		else
-			dest = mod->module_core + shdr->sh_entsize;
+			dest = mod->core_layout.base + shdr->sh_entsize;
 
 		if (shdr->sh_type != SHT_NOBITS)
 			memcpy(dest, (void *)shdr->sh_addr, shdr->sh_size);
@@ -3094,12 +3083,12 @@ static void flush_module_icache(const struct module *mod)
 	 * Do it before processing of module parameters, so the module
 	 * can provide parameter accessor functions of its own.
 	 */
-	if (mod->module_init)
-		flush_icache_range((unsigned long)mod->module_init,
-				   (unsigned long)mod->module_init
-				   + mod->init_size);
-	flush_icache_range((unsigned long)mod->module_core,
-			   (unsigned long)mod->module_core + mod->core_size);
+	if (mod->init_layout.base)
+		flush_icache_range((unsigned long)mod->init_layout.base,
+				   (unsigned long)mod->init_layout.base
+				   + mod->init_layout.size);
+	flush_icache_range((unsigned long)mod->core_layout.base,
+			   (unsigned long)mod->core_layout.base + mod->core_layout.size);
 
 	set_fs(old_fs);
 }
@@ -3157,8 +3146,8 @@ static void module_deallocate(struct module *mod, struct load_info *info)
 {
 	percpu_modfree(mod);
 	module_arch_freeing_init(mod);
-	module_memfree(mod->module_init);
-	module_memfree(mod->module_core);
+	module_memfree(mod->init_layout.base);
+	module_memfree(mod->core_layout.base);
 }
 
 int __weak module_finalize(const Elf_Ehdr *hdr,
@@ -3245,7 +3234,7 @@ static noinline int do_init_module(struct module *mod)
 		ret = -ENOMEM;
 		goto fail;
 	}
-	freeinit->module_init = mod->module_init;
+	freeinit->module_init = mod->init_layout.base;
 
 	/*
 	 * We want to find out whether @mod uses async during init.  Clear
@@ -3302,12 +3291,12 @@ static noinline int do_init_module(struct module *mod)
 	rcu_assign_pointer(mod->kallsyms, &mod->core_kallsyms);
 #endif
 	mod_tree_remove_init(mod);
-	unset_module_init_ro_nx(mod);
+	disable_ro_nx(&mod->init_layout);
 	module_arch_freeing_init(mod);
-	mod->module_init = NULL;
-	mod->init_size = 0;
-	mod->init_ro_size = 0;
-	mod->init_text_size = 0;
+	mod->init_layout.base = NULL;
+	mod->init_layout.size = 0;
+	mod->init_layout.ro_size = 0;
+	mod->init_layout.text_size = 0;
 	/*
 	 * We want to free module_init, but be aware that kallsyms may be
 	 * walking this with preempt disabled.  In all the failure paths, we
@@ -3396,17 +3385,9 @@ static int complete_formation(struct module *mod, struct load_info *info)
 	/* This relies on module_mutex for list integrity. */
 	module_bug_finalize(info->hdr, info->sechdrs, mod);
 
-	/* Set RO and NX regions for core */
-	set_section_ro_nx(mod->module_core,
-				mod->core_text_size,
-				mod->core_ro_size,
-				mod->core_size);
-
-	/* Set RO and NX regions for init */
-	set_section_ro_nx(mod->module_init,
-				mod->init_text_size,
-				mod->init_ro_size,
-				mod->init_size);
+	/* Set RO and NX regions */
+	module_enable_ro(mod);
+	module_enable_nx(mod);
 
 	/* Mark state as coming so strong_try_module_get() ignores us,
 	 * but kallsyms etc. can see us. */
@@ -3571,8 +3552,8 @@ static int load_module(struct load_info *info, const char __user *uargs,
 				     MODULE_STATE_GOING, mod);
 
 	/* we can't deallocate the module until we clear memory protection */
-	unset_module_init_ro_nx(mod);
-	unset_module_core_ro_nx(mod);
+	module_disable_ro(mod);
+	module_disable_nx(mod);
 
  ddebug_cleanup:
 	dynamic_debug_remove(info->debug);
@@ -3601,7 +3582,7 @@ static int load_module(struct load_info *info, const char __user *uargs,
 	 */
 	ftrace_release_mod(mod);
 	/* Free lock-classes; relies on the preceding sync_rcu() */
-	lockdep_free_key_range(mod->module_core, mod->core_size);
+	lockdep_free_key_range(mod->core_layout.base, mod->core_layout.size);
 
 	module_deallocate(mod, info);
  free_copy:
@@ -3685,9 +3666,9 @@ static const char *get_ksymbol(struct module *mod,
 
 	/* At worse, next value is at end of module */
 	if (within_module_init(addr, mod))
-		nextval = (unsigned long)mod->module_init+mod->init_text_size;
+		nextval = (unsigned long)mod->init_layout.base+mod->init_layout.text_size;
 	else
-		nextval = (unsigned long)mod->module_core+mod->core_text_size;
+		nextval = (unsigned long)mod->core_layout.base+mod->core_layout.text_size;
 
 	/* Scan for closest preceding symbol, and next symbol. (ELF
 	   starts real symbols at 1). */
@@ -3940,7 +3921,7 @@ static int m_show(struct seq_file *m, void *p)
 		return 0;
 
 	seq_printf(m, "%s %u",
-		   mod->name, mod->init_size + mod->core_size);
+		   mod->name, mod->init_layout.size + mod->core_layout.size);
 	print_unload_info(m, mod);
 
 	/* Informative for users. */
@@ -3949,7 +3930,7 @@ static int m_show(struct seq_file *m, void *p)
 		   mod->state == MODULE_STATE_COMING ? "Loading" :
 		   "Live");
 	/* Used by oprofile and other similar tools. */
-	seq_printf(m, " 0x%pK", mod->module_core);
+	seq_printf(m, " 0x%pK", mod->core_layout.base);
 
 	/* Taints info */
 	if (mod->taints)
@@ -4092,8 +4073,8 @@ struct module *__module_text_address(unsigned long addr)
 	struct module *mod = __module_address(addr);
 	if (mod) {
 		/* Make sure it's within the text section. */
-		if (!within(addr, mod->module_init, mod->init_text_size)
-		    && !within(addr, mod->module_core, mod->core_text_size))
+		if (!within(addr, mod->init_layout.base, mod->init_layout.text_size)
+		    && !within(addr, mod->core_layout.base, mod->core_layout.text_size))
 			mod = NULL;
 	}
 	return mod;
diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig
index 02e8dfaa1ce2..6d6f63be1f9b 100644
--- a/kernel/power/Kconfig
+++ b/kernel/power/Kconfig
@@ -1,6 +1,7 @@
 config SUSPEND
 	bool "Suspend to RAM and standby"
 	depends on ARCH_SUSPEND_POSSIBLE
+	select RTC_LIB
 	default y
 	---help---
 	  Allow the system to enter sleep states in which main memory is
@@ -28,6 +29,15 @@ config SUSPEND_SKIP_SYNC
 	  of suspend, or they are content with invoking sync() from
 	  user-space before invoking suspend.  Say Y if that's your case.
 
+config WAKELOCK
+	bool "Android's method of preventing suspend"
+	default y
+	---help---
+	  This allows applications to prevent the CPU from suspending while
+	  they need it.
+
+	  Say Y if you are running an android userspace.
+
 config HIBERNATE_CALLBACKS
 	bool
 
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index cb880a14cc39..22eb9ed879ad 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -12,3 +12,5 @@ obj-$(CONFIG_PM_AUTOSLEEP)	+= autosleep.o
 obj-$(CONFIG_PM_WAKELOCKS)	+= wakelock.o
 
 obj-$(CONFIG_MAGIC_SYSRQ)	+= poweroff.o
+
+obj-$(CONFIG_SUSPEND)	+= wakeup_reason.o
diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
index 3124cebaec31..797f19e2aaa9 100644
--- a/kernel/power/hibernate.c
+++ b/kernel/power/hibernate.c
@@ -647,7 +647,7 @@ static void power_down(void)
  */
 int hibernate(void)
 {
-	int error;
+	int error, nr_calls = 0;
 
 	if (!hibernation_available()) {
 		pr_debug("PM: Hibernation not available.\n");
@@ -662,9 +662,11 @@ int hibernate(void)
 	}
 
 	pm_prepare_console();
-	error = pm_notifier_call_chain(PM_HIBERNATION_PREPARE);
-	if (error)
+	error = __pm_notifier_call_chain(PM_HIBERNATION_PREPARE, -1, &nr_calls);
+	if (error) {
+		nr_calls--;
 		goto Exit;
+	}
 
 	printk(KERN_INFO "PM: Syncing filesystems ... ");
 	sys_sync();
@@ -714,7 +716,7 @@ int hibernate(void)
 	/* Don't bother checking whether freezer_test_done is true */
 	freezer_test_done = false;
  Exit:
-	pm_notifier_call_chain(PM_POST_HIBERNATION);
+	__pm_notifier_call_chain(PM_POST_HIBERNATION, nr_calls, NULL);
 	pm_restore_console();
 	atomic_inc(&snapshot_device_available);
  Unlock:
@@ -740,7 +742,7 @@ int hibernate(void)
  */
 static int software_resume(void)
 {
-	int error;
+	int error, nr_calls = 0;
 	unsigned int flags;
 
 	/*
@@ -827,9 +829,11 @@ static int software_resume(void)
 	}
 
 	pm_prepare_console();
-	error = pm_notifier_call_chain(PM_RESTORE_PREPARE);
-	if (error)
+	error = __pm_notifier_call_chain(PM_RESTORE_PREPARE, -1, &nr_calls);
+	if (error) {
+		nr_calls--;
 		goto Close_Finish;
+	}
 
 	pr_debug("PM: Preparing processes for restore.\n");
 	error = freeze_processes();
@@ -855,7 +859,7 @@ static int software_resume(void)
 	unlock_device_hotplug();
 	thaw_processes();
  Finish:
-	pm_notifier_call_chain(PM_POST_RESTORE);
+	__pm_notifier_call_chain(PM_POST_RESTORE, nr_calls, NULL);
 	pm_restore_console();
 	atomic_inc(&snapshot_device_available);
 	/* For success case, the suspend path will release the lock */
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 27946975eff0..5ea50b1b7595 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -38,12 +38,19 @@ int unregister_pm_notifier(struct notifier_block *nb)
 }
 EXPORT_SYMBOL_GPL(unregister_pm_notifier);
 
-int pm_notifier_call_chain(unsigned long val)
+int __pm_notifier_call_chain(unsigned long val, int nr_to_call, int *nr_calls)
 {
-	int ret = blocking_notifier_call_chain(&pm_chain_head, val, NULL);
+	int ret;
+
+	ret = __blocking_notifier_call_chain(&pm_chain_head, val, NULL,
+						nr_to_call, nr_calls);
 
 	return notifier_to_errno(ret);
 }
+int pm_notifier_call_chain(unsigned long val)
+{
+	return __pm_notifier_call_chain(val, -1, NULL);
+}
 
 /* If set, devices may be suspended and resumed asynchronously. */
 int pm_async_enabled = 1;
diff --git a/kernel/power/power.h b/kernel/power/power.h
index efe1b3b17c88..51f02ecaf125 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -200,6 +200,8 @@ static inline void suspend_test_finish(const char *label) {}
 
 #ifdef CONFIG_PM_SLEEP
 /* kernel/power/main.c */
+extern int __pm_notifier_call_chain(unsigned long val, int nr_to_call,
+				    int *nr_calls);
 extern int pm_notifier_call_chain(unsigned long val);
 #endif
 
diff --git a/kernel/power/process.c b/kernel/power/process.c
index ba2029a02259..cc177142a08f 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -18,6 +18,7 @@
 #include <linux/workqueue.h>
 #include <linux/kmod.h>
 #include <trace/events/power.h>
+#include <linux/wakeup_reason.h>
 #include <linux/cpuset.h>
 
 /*
@@ -36,6 +37,9 @@ static int try_to_freeze_tasks(bool user_only)
 	unsigned int elapsed_msecs;
 	bool wakeup = false;
 	int sleep_usecs = USEC_PER_MSEC;
+#ifdef CONFIG_PM_SLEEP
+	char suspend_abort[MAX_SUSPEND_ABORT_LEN];
+#endif
 
 	do_gettimeofday(&start);
 
@@ -65,6 +69,11 @@ static int try_to_freeze_tasks(bool user_only)
 			break;
 
 		if (pm_wakeup_pending()) {
+#ifdef CONFIG_PM_SLEEP
+			pm_get_active_wakeup_sources(suspend_abort,
+				MAX_SUSPEND_ABORT_LEN);
+			log_suspend_abort_reason(suspend_abort);
+#endif
 			wakeup = true;
 			break;
 		}
@@ -84,15 +93,17 @@ static int try_to_freeze_tasks(bool user_only)
 	do_div(elapsed_msecs64, NSEC_PER_MSEC);
 	elapsed_msecs = elapsed_msecs64;
 
-	if (todo) {
+	if (wakeup) {
 		pr_cont("\n");
-		pr_err("Freezing of tasks %s after %d.%03d seconds "
-		       "(%d tasks refusing to freeze, wq_busy=%d):\n",
-		       wakeup ? "aborted" : "failed",
+		pr_err("Freezing of tasks aborted after %d.%03d seconds",
+		       elapsed_msecs / 1000, elapsed_msecs % 1000);
+	} else if (todo) {
+		pr_cont("\n");
+		pr_err("Freezing of tasks failed after %d.%03d seconds"
+		       " (%d tasks refusing to freeze, wq_busy=%d):\n",
 		       elapsed_msecs / 1000, elapsed_msecs % 1000,
 		       todo - wq_busy, wq_busy);
 
-		if (!wakeup) {
 			read_lock(&tasklist_lock);
 			for_each_process_thread(g, p) {
 				if (p != current && !freezer_should_skip(p)
@@ -100,7 +111,6 @@ static int try_to_freeze_tasks(bool user_only)
 					sched_show_task(p);
 			}
 			read_unlock(&tasklist_lock);
-		}
 	} else {
 		pr_cont("(elapsed %d.%03d seconds) ", elapsed_msecs / 1000,
 			elapsed_msecs % 1000);
diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
index f9fe133c13e2..58209d8bfc56 100644
--- a/kernel/power/suspend.c
+++ b/kernel/power/suspend.c
@@ -26,9 +26,11 @@
 #include <linux/suspend.h>
 #include <linux/syscore_ops.h>
 #include <linux/ftrace.h>
+#include <linux/rtc.h>
 #include <trace/events/power.h>
 #include <linux/compiler.h>
 #include <linux/moduleparam.h>
+#include <linux/wakeup_reason.h>
 
 #include "power.h"
 
@@ -266,16 +268,18 @@ static int suspend_test(int level)
  */
 static int suspend_prepare(suspend_state_t state)
 {
-	int error;
+	int error, nr_calls = 0;
 
 	if (!sleep_state_supported(state))
 		return -EPERM;
 
 	pm_prepare_console();
 
-	error = pm_notifier_call_chain(PM_SUSPEND_PREPARE);
-	if (error)
+	error = __pm_notifier_call_chain(PM_SUSPEND_PREPARE, -1, &nr_calls);
+	if (error) {
+		nr_calls--;
 		goto Finish;
+	}
 
 	trace_suspend_resume(TPS("freeze_processes"), 0, true);
 	error = suspend_freeze_processes();
@@ -286,7 +290,7 @@ static int suspend_prepare(suspend_state_t state)
 	suspend_stats.failed_freeze++;
 	dpm_save_failed_step(SUSPEND_FREEZE);
  Finish:
-	pm_notifier_call_chain(PM_POST_SUSPEND);
+	__pm_notifier_call_chain(PM_POST_SUSPEND, nr_calls, NULL);
 	pm_restore_console();
 	return error;
 }
@@ -312,7 +316,8 @@ void __weak arch_suspend_enable_irqs(void)
  */
 static int suspend_enter(suspend_state_t state, bool *wakeup)
 {
-	int error;
+	char suspend_abort[MAX_SUSPEND_ABORT_LEN];
+	int error, last_dev;
 
 	error = platform_suspend_prepare(state);
 	if (error)
@@ -320,7 +325,11 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
 
 	error = dpm_suspend_late(PMSG_SUSPEND);
 	if (error) {
+		last_dev = suspend_stats.last_failed_dev + REC_FAILED_NUM - 1;
+		last_dev %= REC_FAILED_NUM;
 		printk(KERN_ERR "PM: late suspend of devices failed\n");
+		log_suspend_abort_reason("%s device failed to power down",
+			suspend_stats.failed_devs[last_dev]);
 		goto Platform_finish;
 	}
 	error = platform_suspend_prepare_late(state);
@@ -329,7 +338,11 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
 
 	error = dpm_suspend_noirq(PMSG_SUSPEND);
 	if (error) {
+		last_dev = suspend_stats.last_failed_dev + REC_FAILED_NUM - 1;
+		last_dev %= REC_FAILED_NUM;
 		printk(KERN_ERR "PM: noirq suspend of devices failed\n");
+		log_suspend_abort_reason("noirq suspend of %s device failed",
+			suspend_stats.failed_devs[last_dev]);
 		goto Platform_early_resume;
 	}
 	error = platform_suspend_prepare_noirq(state);
@@ -353,8 +366,10 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
 	}
 
 	error = disable_nonboot_cpus();
-	if (error || suspend_test(TEST_CPUS))
+	if (error || suspend_test(TEST_CPUS)) {
+		log_suspend_abort_reason("Disabling non-boot cpus failed");
 		goto Enable_cpus;
+	}
 
 	arch_suspend_disable_irqs();
 	BUG_ON(!irqs_disabled());
@@ -370,6 +385,9 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
 				state, false);
 			events_check_enabled = false;
 		} else if (*wakeup) {
+			pm_get_active_wakeup_sources(suspend_abort,
+				MAX_SUSPEND_ABORT_LEN);
+			log_suspend_abort_reason(suspend_abort);
 			error = -EBUSY;
 		}
 		syscore_resume();
@@ -417,6 +435,7 @@ int suspend_devices_and_enter(suspend_state_t state)
 	error = dpm_suspend_start(PMSG_SUSPEND);
 	if (error) {
 		pr_err("PM: Some devices failed to suspend, or early wake event detected\n");
+		log_suspend_abort_reason("Some devices failed to suspend, or early wake event detected");
 		goto Recover_platform;
 	}
 	suspend_test_finish("suspend devices");
@@ -518,6 +537,18 @@ static int enter_state(suspend_state_t state)
 	return error;
 }
 
+static void pm_suspend_marker(char *annotation)
+{
+	struct timespec ts;
+	struct rtc_time tm;
+
+	getnstimeofday(&ts);
+	rtc_time_to_tm(ts.tv_sec, &tm);
+	pr_info("PM: suspend %s %d-%02d-%02d %02d:%02d:%02d.%09lu UTC\n",
+		annotation, tm.tm_year + 1900, tm.tm_mon + 1, tm.tm_mday,
+		tm.tm_hour, tm.tm_min, tm.tm_sec, ts.tv_nsec);
+}
+
 /**
  * pm_suspend - Externally visible function for suspending the system.
  * @state: System sleep state to enter.
@@ -532,6 +563,7 @@ int pm_suspend(suspend_state_t state)
 	if (state <= PM_SUSPEND_ON || state >= PM_SUSPEND_MAX)
 		return -EINVAL;
 
+	pm_suspend_marker("entry");
 	error = enter_state(state);
 	if (error) {
 		suspend_stats.fail++;
@@ -539,6 +571,7 @@ int pm_suspend(suspend_state_t state)
 	} else {
 		suspend_stats.success++;
 	}
+	pm_suspend_marker("exit");
 	return error;
 }
 EXPORT_SYMBOL(pm_suspend);
diff --git a/kernel/power/user.c b/kernel/power/user.c
index 526e8911460a..35310b627388 100644
--- a/kernel/power/user.c
+++ b/kernel/power/user.c
@@ -47,7 +47,7 @@ atomic_t snapshot_device_available = ATOMIC_INIT(1);
 static int snapshot_open(struct inode *inode, struct file *filp)
 {
 	struct snapshot_data *data;
-	int error;
+	int error, nr_calls = 0;
 
 	if (!hibernation_available())
 		return -EPERM;
@@ -74,9 +74,9 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 			swap_type_of(swsusp_resume_device, 0, NULL) : -1;
 		data->mode = O_RDONLY;
 		data->free_bitmaps = false;
-		error = pm_notifier_call_chain(PM_HIBERNATION_PREPARE);
+		error = __pm_notifier_call_chain(PM_HIBERNATION_PREPARE, -1, &nr_calls);
 		if (error)
-			pm_notifier_call_chain(PM_POST_HIBERNATION);
+			__pm_notifier_call_chain(PM_POST_HIBERNATION, --nr_calls, NULL);
 	} else {
 		/*
 		 * Resuming.  We may need to wait for the image device to
@@ -86,13 +86,15 @@ static int snapshot_open(struct inode *inode, struct file *filp)
 
 		data->swap = -1;
 		data->mode = O_WRONLY;
-		error = pm_notifier_call_chain(PM_RESTORE_PREPARE);
+		error = __pm_notifier_call_chain(PM_RESTORE_PREPARE, -1, &nr_calls);
 		if (!error) {
 			error = create_basic_memory_bitmaps();
 			data->free_bitmaps = !error;
-		}
+		} else
+			nr_calls--;
+
 		if (error)
-			pm_notifier_call_chain(PM_POST_RESTORE);
+			__pm_notifier_call_chain(PM_POST_RESTORE, nr_calls, NULL);
 	}
 	if (error)
 		atomic_inc(&snapshot_device_available);
diff --git a/kernel/power/wakeup_reason.c b/kernel/power/wakeup_reason.c
new file mode 100644
index 000000000000..252611fad2fe
--- /dev/null
+++ b/kernel/power/wakeup_reason.c
@@ -0,0 +1,225 @@
+/*
+ * kernel/power/wakeup_reason.c
+ *
+ * Logs the reasons which caused the kernel to resume from
+ * the suspend mode.
+ *
+ * Copyright (C) 2014 Google, Inc.
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/wakeup_reason.h>
+#include <linux/kernel.h>
+#include <linux/irq.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/kobject.h>
+#include <linux/sysfs.h>
+#include <linux/init.h>
+#include <linux/spinlock.h>
+#include <linux/notifier.h>
+#include <linux/suspend.h>
+
+
+#define MAX_WAKEUP_REASON_IRQS 32
+static int irq_list[MAX_WAKEUP_REASON_IRQS];
+static int irqcount;
+static bool suspend_abort;
+static char abort_reason[MAX_SUSPEND_ABORT_LEN];
+static struct kobject *wakeup_reason;
+static DEFINE_SPINLOCK(resume_reason_lock);
+
+static ktime_t last_monotime; /* monotonic time before last suspend */
+static ktime_t curr_monotime; /* monotonic time after last suspend */
+static ktime_t last_stime; /* monotonic boottime offset before last suspend */
+static ktime_t curr_stime; /* monotonic boottime offset after last suspend */
+
+static ssize_t last_resume_reason_show(struct kobject *kobj, struct kobj_attribute *attr,
+		char *buf)
+{
+	int irq_no, buf_offset = 0;
+	struct irq_desc *desc;
+	spin_lock(&resume_reason_lock);
+	if (suspend_abort) {
+		buf_offset = sprintf(buf, "Abort: %s", abort_reason);
+	} else {
+		for (irq_no = 0; irq_no < irqcount; irq_no++) {
+			desc = irq_to_desc(irq_list[irq_no]);
+			if (desc && desc->action && desc->action->name)
+				buf_offset += sprintf(buf + buf_offset, "%d %s\n",
+						irq_list[irq_no], desc->action->name);
+			else
+				buf_offset += sprintf(buf + buf_offset, "%d\n",
+						irq_list[irq_no]);
+		}
+	}
+	spin_unlock(&resume_reason_lock);
+	return buf_offset;
+}
+
+static ssize_t last_suspend_time_show(struct kobject *kobj,
+			struct kobj_attribute *attr, char *buf)
+{
+	struct timespec sleep_time;
+	struct timespec total_time;
+	struct timespec suspend_resume_time;
+
+	/*
+	 * total_time is calculated from monotonic bootoffsets because
+	 * unlike CLOCK_MONOTONIC it include the time spent in suspend state.
+	 */
+	total_time = ktime_to_timespec(ktime_sub(curr_stime, last_stime));
+
+	/*
+	 * suspend_resume_time is calculated as monotonic (CLOCK_MONOTONIC)
+	 * time interval before entering suspend and post suspend.
+	 */
+	suspend_resume_time = ktime_to_timespec(ktime_sub(curr_monotime, last_monotime));
+
+	/* sleep_time = total_time - suspend_resume_time */
+	sleep_time = timespec_sub(total_time, suspend_resume_time);
+
+	/* Export suspend_resume_time and sleep_time in pair here. */
+	return sprintf(buf, "%lu.%09lu %lu.%09lu\n",
+				suspend_resume_time.tv_sec, suspend_resume_time.tv_nsec,
+				sleep_time.tv_sec, sleep_time.tv_nsec);
+}
+
+static struct kobj_attribute resume_reason = __ATTR_RO(last_resume_reason);
+static struct kobj_attribute suspend_time = __ATTR_RO(last_suspend_time);
+
+static struct attribute *attrs[] = {
+	&resume_reason.attr,
+	&suspend_time.attr,
+	NULL,
+};
+static struct attribute_group attr_group = {
+	.attrs = attrs,
+};
+
+/*
+ * logs all the wake up reasons to the kernel
+ * stores the irqs to expose them to the userspace via sysfs
+ */
+void log_wakeup_reason(int irq)
+{
+	struct irq_desc *desc;
+	desc = irq_to_desc(irq);
+	if (desc && desc->action && desc->action->name)
+		printk(KERN_INFO "Resume caused by IRQ %d, %s\n", irq,
+				desc->action->name);
+	else
+		printk(KERN_INFO "Resume caused by IRQ %d\n", irq);
+
+	spin_lock(&resume_reason_lock);
+	if (irqcount == MAX_WAKEUP_REASON_IRQS) {
+		spin_unlock(&resume_reason_lock);
+		printk(KERN_WARNING "Resume caused by more than %d IRQs\n",
+				MAX_WAKEUP_REASON_IRQS);
+		return;
+	}
+
+	irq_list[irqcount++] = irq;
+	spin_unlock(&resume_reason_lock);
+}
+
+int check_wakeup_reason(int irq)
+{
+	int irq_no;
+	int ret = false;
+
+	spin_lock(&resume_reason_lock);
+	for (irq_no = 0; irq_no < irqcount; irq_no++)
+		if (irq_list[irq_no] == irq) {
+			ret = true;
+			break;
+	}
+	spin_unlock(&resume_reason_lock);
+	return ret;
+}
+
+void log_suspend_abort_reason(const char *fmt, ...)
+{
+	va_list args;
+
+	spin_lock(&resume_reason_lock);
+
+	//Suspend abort reason has already been logged.
+	if (suspend_abort) {
+		spin_unlock(&resume_reason_lock);
+		return;
+	}
+
+	suspend_abort = true;
+	va_start(args, fmt);
+	vsnprintf(abort_reason, MAX_SUSPEND_ABORT_LEN, fmt, args);
+	va_end(args);
+	spin_unlock(&resume_reason_lock);
+}
+
+/* Detects a suspend and clears all the previous wake up reasons*/
+static int wakeup_reason_pm_event(struct notifier_block *notifier,
+		unsigned long pm_event, void *unused)
+{
+	switch (pm_event) {
+	case PM_SUSPEND_PREPARE:
+		spin_lock(&resume_reason_lock);
+		irqcount = 0;
+		suspend_abort = false;
+		spin_unlock(&resume_reason_lock);
+		/* monotonic time since boot */
+		last_monotime = ktime_get();
+		/* monotonic time since boot including the time spent in suspend */
+		last_stime = ktime_get_boottime();
+		break;
+	case PM_POST_SUSPEND:
+		/* monotonic time since boot */
+		curr_monotime = ktime_get();
+		/* monotonic time since boot including the time spent in suspend */
+		curr_stime = ktime_get_boottime();
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block wakeup_reason_pm_notifier_block = {
+	.notifier_call = wakeup_reason_pm_event,
+};
+
+/* Initializes the sysfs parameter
+ * registers the pm_event notifier
+ */
+int __init wakeup_reason_init(void)
+{
+	int retval;
+
+	retval = register_pm_notifier(&wakeup_reason_pm_notifier_block);
+	if (retval)
+		printk(KERN_WARNING "[%s] failed to register PM notifier %d\n",
+				__func__, retval);
+
+	wakeup_reason = kobject_create_and_add("wakeup_reasons", kernel_kobj);
+	if (!wakeup_reason) {
+		printk(KERN_WARNING "[%s] failed to create a sysfs kobject\n",
+				__func__);
+		return 1;
+	}
+	retval = sysfs_create_group(wakeup_reason, &attr_group);
+	if (retval) {
+		kobject_put(wakeup_reason);
+		printk(KERN_WARNING "[%s] failed to create a sysfs group %d\n",
+				__func__, retval);
+	}
+	return 0;
+}
+
+late_initcall(wakeup_reason_init);
diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
index 0b5613554769..b4573b55b435 100644
--- a/kernel/printk/printk.c
+++ b/kernel/printk/printk.c
@@ -55,6 +55,10 @@
 #include "console_cmdline.h"
 #include "braille.h"
 
+#ifdef CONFIG_EARLY_PRINTK_DIRECT
+extern void printascii(char *);
+#endif
+
 int console_printk[4] = {
 	CONSOLE_LOGLEVEL_DEFAULT,	/* console_loglevel */
 	MESSAGE_LOGLEVEL_DEFAULT,	/* default_message_loglevel */
@@ -1754,6 +1758,10 @@ asmlinkage int vprintk_emit(int facility, int level,
 		}
 	}
 
+#ifdef CONFIG_EARLY_PRINTK_DIRECT
+	printascii(text);
+#endif
+
 	if (level == LOGLEVEL_DEFAULT)
 		level = default_message_loglevel;
 
@@ -3168,9 +3176,8 @@ void show_regs_print_info(const char *log_lvl)
 {
 	dump_stack_print_info(log_lvl);
 
-	printk("%stask: %p ti: %p task.ti: %p\n",
-	       log_lvl, current, current_thread_info(),
-	       task_thread_info(current));
+	printk("%stask: %p task.stack: %p\n",
+	       log_lvl, current, task_stack_page(current));
 }
 
 #endif
diff --git a/kernel/rcu/sync.c b/kernel/rcu/sync.c
index be922c9f3d37..b49cf3ac2d47 100644
--- a/kernel/rcu/sync.c
+++ b/kernel/rcu/sync.c
@@ -68,6 +68,7 @@ void rcu_sync_lockdep_assert(struct rcu_sync *rsp)
 	RCU_LOCKDEP_WARN(!gp_ops[rsp->gp_type].held(),
 			 "suspicious rcu_sync_is_idle() usage");
 }
+EXPORT_SYMBOL_GPL(rcu_sync_lockdep_assert);
 #endif
 
 /**
@@ -83,6 +84,18 @@ void rcu_sync_init(struct rcu_sync *rsp, enum rcu_sync_type type)
 }
 
 /**
+ * Must be called after rcu_sync_init() and before first use.
+ *
+ * Ensures rcu_sync_is_idle() returns false and rcu_sync_{enter,exit}()
+ * pairs turn into NO-OPs.
+ */
+void rcu_sync_enter_start(struct rcu_sync *rsp)
+{
+	rsp->gp_count++;
+	rsp->gp_state = GP_PASSED;
+}
+
+/**
  * rcu_sync_enter() - Force readers onto slowpath
  * @rsp: Pointer to rcu_sync structure to use for synchronization
  *
diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile
index 67687973ce80..d7ec4f7dd0d9 100644
--- a/kernel/sched/Makefile
+++ b/kernel/sched/Makefile
@@ -14,8 +14,12 @@ endif
 obj-y += core.o loadavg.o clock.o cputime.o
 obj-y += idle_task.o fair.o rt.o deadline.o stop_task.o
 obj-y += wait.o completion.o idle.o
-obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o
+obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o energy.o
+obj-$(CONFIG_SCHED_WALT) += walt.o
 obj-$(CONFIG_SCHED_AUTOGROUP) += auto_group.o
 obj-$(CONFIG_SCHEDSTATS) += stats.o
 obj-$(CONFIG_SCHED_DEBUG) += debug.o
+obj-$(CONFIG_SCHED_TUNE) += tune.o
 obj-$(CONFIG_CGROUP_CPUACCT) += cpuacct.o
+obj-$(CONFIG_CPU_FREQ) += cpufreq.o
+obj-$(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) += cpufreq_schedutil.o
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index b5d372083624..1eb91a696069 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -89,6 +89,7 @@
 
 #define CREATE_TRACE_POINTS
 #include <trace/events/sched.h>
+#include "walt.h"
 
 DEFINE_MUTEX(sched_domains_mutex);
 DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues);
@@ -287,6 +288,18 @@ int sysctl_sched_rt_runtime = 950000;
 /* cpus with isolated domains */
 cpumask_var_t cpu_isolated_map;
 
+struct rq *
+lock_rq_of(struct task_struct *p, unsigned long *flags)
+{
+	return task_rq_lock(p, flags);
+}
+
+void
+unlock_rq_of(struct rq *rq, struct task_struct *p, unsigned long *flags)
+{
+	task_rq_unlock(rq, p, flags);
+}
+
 /*
  * this_rq_lock - lock this runqueue and disable interrupts.
  */
@@ -533,6 +546,8 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
 	if (cmpxchg(&node->next, NULL, WAKE_Q_TAIL))
 		return;
 
+	head->count++;
+
 	get_task_struct(task);
 
 	/*
@@ -542,6 +557,10 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
 	head->lastp = &node->next;
 }
 
+static int
+try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags,
+	       int sibling_count_hint);
+
 void wake_up_q(struct wake_q_head *head)
 {
 	struct wake_q_node *node = head->first;
@@ -556,10 +575,10 @@ void wake_up_q(struct wake_q_head *head)
 		task->wake_q.next = NULL;
 
 		/*
-		 * wake_up_process() implies a wmb() to pair with the queueing
+		 * try_to_wake_up() implies a wmb() to pair with the queueing
 		 * in wake_q_add() so as not to miss wakeups.
 		 */
-		wake_up_process(task);
+		try_to_wake_up(task, TASK_NORMAL, 0, head->count);
 		put_task_struct(task);
 	}
 }
@@ -1076,7 +1095,9 @@ static struct rq *move_queued_task(struct rq *rq, struct task_struct *p, int new
 
 	dequeue_task(rq, p, 0);
 	p->on_rq = TASK_ON_RQ_MIGRATING;
+	double_lock_balance(rq, cpu_rq(new_cpu));
 	set_task_cpu(p, new_cpu);
+	double_unlock_balance(rq, cpu_rq(new_cpu));
 	raw_spin_unlock(&rq->lock);
 
 	rq = cpu_rq(new_cpu);
@@ -1300,6 +1321,8 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu)
 			p->sched_class->migrate_task_rq(p);
 		p->se.nr_migrations++;
 		perf_event_task_migrate(p);
+
+		walt_fixup_busy_time(p, new_cpu);
 	}
 
 	__set_task_cpu(p, new_cpu);
@@ -1314,7 +1337,9 @@ static void __migrate_swap_task(struct task_struct *p, int cpu)
 		dst_rq = cpu_rq(cpu);
 
 		deactivate_task(src_rq, p, 0);
+		p->on_rq = TASK_ON_RQ_MIGRATING;
 		set_task_cpu(p, cpu);
+		p->on_rq = TASK_ON_RQ_QUEUED;
 		activate_task(dst_rq, p, 0);
 		check_preempt_curr(dst_rq, p, 0);
 	} else {
@@ -1623,12 +1648,14 @@ out:
  * The caller (fork, wakeup) owns p->pi_lock, ->cpus_allowed is stable.
  */
 static inline
-int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags)
+int select_task_rq(struct task_struct *p, int cpu, int sd_flags, int wake_flags,
+		   int sibling_count_hint)
 {
 	lockdep_assert_held(&p->pi_lock);
 
 	if (p->nr_cpus_allowed > 1)
-		cpu = p->sched_class->select_task_rq(p, cpu, sd_flags, wake_flags);
+		cpu = p->sched_class->select_task_rq(p, cpu, sd_flags, wake_flags,
+						     sibling_count_hint);
 
 	/*
 	 * In order not to call set_task_cpu() on a blocking task we need
@@ -1913,6 +1940,8 @@ static void ttwu_queue(struct task_struct *p, int cpu)
  * @p: the thread to be awakened
  * @state: the mask of task states that can be woken
  * @wake_flags: wake modifier flags (WF_*)
+ * @sibling_count_hint: A hint at the number of threads that are being woken up
+ *                      in this event.
  *
  * Put it on the run-queue if it's not already there. The "current"
  * thread is always on the run-queue (except when the actual
@@ -1924,10 +1953,15 @@ static void ttwu_queue(struct task_struct *p, int cpu)
  * or @state didn't match @p's state.
  */
 static int
-try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
+try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags,
+	       int sibling_count_hint)
 {
 	unsigned long flags;
 	int cpu, success = 0;
+#ifdef CONFIG_SMP
+	struct rq *rq;
+	u64 wallclock;
+#endif
 
 	/*
 	 * If we are going to wake up a thread waiting for CONDITION we
@@ -2007,17 +2041,27 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
 	 */
 	smp_rmb();
 
+	rq = cpu_rq(task_cpu(p));
+
+	raw_spin_lock(&rq->lock);
+	wallclock = walt_ktime_clock();
+	walt_update_task_ravg(rq->curr, rq, TASK_UPDATE, wallclock, 0);
+	walt_update_task_ravg(p, rq, TASK_WAKE, wallclock, 0);
+	raw_spin_unlock(&rq->lock);
+
 	p->sched_contributes_to_load = !!task_contributes_to_load(p);
 	p->state = TASK_WAKING;
 
 	if (p->sched_class->task_waking)
 		p->sched_class->task_waking(p);
 
-	cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags);
+	cpu = select_task_rq(p, p->wake_cpu, SD_BALANCE_WAKE, wake_flags,
+			     sibling_count_hint);
 	if (task_cpu(p) != cpu) {
 		wake_flags |= WF_MIGRATED;
 		set_task_cpu(p, cpu);
 	}
+
 #endif /* CONFIG_SMP */
 
 	ttwu_queue(p, cpu);
@@ -2066,8 +2110,13 @@ static void try_to_wake_up_local(struct task_struct *p)
 
 	trace_sched_waking(p);
 
-	if (!task_on_rq_queued(p))
+	if (!task_on_rq_queued(p)) {
+		u64 wallclock = walt_ktime_clock();
+
+		walt_update_task_ravg(rq->curr, rq, TASK_UPDATE, wallclock, 0);
+		walt_update_task_ravg(p, rq, TASK_WAKE, wallclock, 0);
 		ttwu_activate(rq, p, ENQUEUE_WAKEUP);
+	}
 
 	ttwu_do_wakeup(rq, p, 0);
 	ttwu_stat(p, smp_processor_id(), 0);
@@ -2089,13 +2138,13 @@ out:
  */
 int wake_up_process(struct task_struct *p)
 {
-	return try_to_wake_up(p, TASK_NORMAL, 0);
+	return try_to_wake_up(p, TASK_NORMAL, 0, 1);
 }
 EXPORT_SYMBOL(wake_up_process);
 
 int wake_up_state(struct task_struct *p, unsigned int state)
 {
-	return try_to_wake_up(p, state, 0);
+	return try_to_wake_up(p, state, 0, 1);
 }
 
 /*
@@ -2132,7 +2181,16 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p)
 	p->se.prev_sum_exec_runtime	= 0;
 	p->se.nr_migrations		= 0;
 	p->se.vruntime			= 0;
+#ifdef CONFIG_SCHED_WALT
+	p->last_sleep_ts		= 0;
+#endif
+
 	INIT_LIST_HEAD(&p->se.group_node);
+	walt_init_new_task_load(p);
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	p->se.cfs_rq			= NULL;
+#endif
 
 #ifdef CONFIG_SCHEDSTATS
 	memset(&p->se.statistics, 0, sizeof(p->se.statistics));
@@ -2216,11 +2274,11 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p)
 
 	__sched_fork(clone_flags, p);
 	/*
-	 * We mark the process as running here. This guarantees that
+	 * We mark the process as NEW here. This guarantees that
 	 * nobody will actually run it, and a signal or other external
 	 * event cannot wake it up and insert it on the runqueue either.
 	 */
-	p->state = TASK_RUNNING;
+	p->state = TASK_NEW;
 
 	/*
 	 * Make sure we do not leak PI boosting priority to the child.
@@ -2257,8 +2315,7 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p)
 		p->sched_class = &fair_sched_class;
 	}
 
-	if (p->sched_class->task_fork)
-		p->sched_class->task_fork(p);
+	init_entity_runnable_average(&p->se);
 
 	/*
 	 * The child is not yet in the pid-hash so no cgroup attach races,
@@ -2268,7 +2325,13 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p)
 	 * Silence PROVE_RCU.
 	 */
 	raw_spin_lock_irqsave(&p->pi_lock, flags);
-	set_task_cpu(p, cpu);
+	/*
+	 * We're setting the cpu for the first time, we don't migrate,
+	 * so use __set_task_cpu().
+	 */
+	__set_task_cpu(p, cpu);
+	if (p->sched_class->task_fork)
+		p->sched_class->task_fork(p);
 	raw_spin_unlock_irqrestore(&p->pi_lock, flags);
 
 #ifdef CONFIG_SCHED_INFO
@@ -2400,6 +2463,10 @@ void wake_up_new_task(struct task_struct *p)
 	struct rq *rq;
 
 	raw_spin_lock_irqsave(&p->pi_lock, flags);
+	p->state = TASK_RUNNING;
+
+	walt_init_new_task_load(p);
+
 	/* Initialize new task's runnable average */
 	init_entity_runnable_average(&p->se);
 #ifdef CONFIG_SMP
@@ -2407,12 +2474,18 @@ void wake_up_new_task(struct task_struct *p)
 	 * Fork balancing, do it here and not earlier because:
 	 *  - cpus_allowed can change in the fork path
 	 *  - any previously selected cpu might disappear through hotplug
+	 *
+	 * Use __set_task_cpu() to avoid calling sched_class::migrate_task_rq,
+	 * as we're not fully set-up yet.
 	 */
-	set_task_cpu(p, select_task_rq(p, task_cpu(p), SD_BALANCE_FORK, 0));
+	__set_task_cpu(p, select_task_rq(p, task_cpu(p), SD_BALANCE_FORK, 0, 1));
 #endif
-
 	rq = __task_rq_lock(p);
-	activate_task(rq, p, 0);
+	update_rq_clock(rq);
+	post_init_entity_util_avg(&p->se);
+
+	walt_mark_task_starting(p);
+	activate_task(rq, p, ENQUEUE_WAKEUP_NEW);
 	p->on_rq = TASK_ON_RQ_QUEUED;
 	trace_sched_wakeup_new(p);
 	check_preempt_curr(rq, p, WF_FORK);
@@ -2793,6 +2866,36 @@ unsigned long nr_iowait_cpu(int cpu)
 	return atomic_read(&this->nr_iowait);
 }
 
+#ifdef CONFIG_CPU_QUIET
+u64 nr_running_integral(unsigned int cpu)
+{
+	unsigned int seqcnt;
+	u64 integral;
+	struct rq *q;
+
+	if (cpu >= nr_cpu_ids)
+		return 0;
+
+	q = cpu_rq(cpu);
+
+	/*
+	 * Update average to avoid reading stalled value if there were
+	 * no run-queue changes for a long time. On the other hand if
+	 * the changes are happening right now, just read current value
+	 * directly.
+	 */
+
+	seqcnt = read_seqcount_begin(&q->ave_seqcnt);
+	integral = do_nr_running_integral(q);
+	if (read_seqcount_retry(&q->ave_seqcnt, seqcnt)) {
+		read_seqcount_begin(&q->ave_seqcnt);
+		integral = q->nr_running_integral;
+	}
+
+	return integral;
+}
+#endif
+
 void get_iowait_load(unsigned long *nr_waiters, unsigned long *load)
 {
 	struct rq *rq = this_rq();
@@ -2813,7 +2916,7 @@ void sched_exec(void)
 	int dest_cpu;
 
 	raw_spin_lock_irqsave(&p->pi_lock, flags);
-	dest_cpu = p->sched_class->select_task_rq(p, task_cpu(p), SD_BALANCE_EXEC, 0);
+	dest_cpu = p->sched_class->select_task_rq(p, task_cpu(p), SD_BALANCE_EXEC, 0, 1);
 	if (dest_cpu == smp_processor_id())
 		goto unlock;
 
@@ -2892,6 +2995,9 @@ void scheduler_tick(void)
 	sched_clock_tick();
 
 	raw_spin_lock(&rq->lock);
+	walt_set_window_start(rq);
+	walt_update_task_ravg(rq->curr, rq, TASK_UPDATE,
+			walt_ktime_clock(), 0);
 	update_rq_clock(rq);
 	curr->sched_class->task_tick(rq, curr, 0);
 	update_cpu_load_active(rq);
@@ -2905,6 +3011,9 @@ void scheduler_tick(void)
 	trigger_load_balance(rq);
 #endif
 	rq_last_tick_reset(rq);
+
+	if (curr->sched_class == &fair_sched_class)
+		check_for_migration(rq, curr);
 }
 
 #ifdef CONFIG_NO_HZ_FULL
@@ -3132,6 +3241,7 @@ static void __sched notrace __schedule(bool preempt)
 	unsigned long *switch_count;
 	struct rq *rq;
 	int cpu;
+	u64 wallclock;
 
 	cpu = smp_processor_id();
 	rq = cpu_rq(cpu);
@@ -3193,11 +3303,18 @@ static void __sched notrace __schedule(bool preempt)
 		update_rq_clock(rq);
 
 	next = pick_next_task(rq, prev);
+	wallclock = walt_ktime_clock();
+	walt_update_task_ravg(prev, rq, PUT_PREV_TASK, wallclock, 0);
+	walt_update_task_ravg(next, rq, PICK_NEXT_TASK, wallclock, 0);
 	clear_tsk_need_resched(prev);
 	clear_preempt_need_resched();
 	rq->clock_skip_update = 0;
 
 	if (likely(prev != next)) {
+#ifdef CONFIG_SCHED_WALT
+		if (!prev->on_rq)
+			prev->last_sleep_ts = wallclock;
+#endif
 		rq->nr_switches++;
 		rq->curr = next;
 		++*switch_count;
@@ -3371,7 +3488,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void)
 int default_wake_function(wait_queue_t *curr, unsigned mode, int wake_flags,
 			  void *key)
 {
-	return try_to_wake_up(curr->private, mode, wake_flags);
+	return try_to_wake_up(curr->private, mode, wake_flags, 1);
 }
 EXPORT_SYMBOL(default_wake_function);
 
@@ -3397,6 +3514,7 @@ void rt_mutex_setprio(struct task_struct *p, int prio)
 	BUG_ON(prio > MAX_PRIO);
 
 	rq = __task_rq_lock(p);
+	update_rq_clock(rq);
 
 	/*
 	 * Idle task boosting is a nono in general. There is one
@@ -3488,6 +3606,8 @@ void set_user_nice(struct task_struct *p, long nice)
 	 * the task might be in the middle of scheduling on another CPU.
 	 */
 	rq = task_rq_lock(p, &flags);
+	update_rq_clock(rq);
+
 	/*
 	 * The RT priorities are set via sched_setscheduler(), but we still
 	 * allow the 'normal' nice value to be set - but as expected
@@ -3914,6 +4034,7 @@ recheck:
 	 * runqueue lock must be held.
 	 */
 	rq = task_rq_lock(p, &flags);
+	update_rq_clock(rq);
 
 	/*
 	 * Changing the policy of the stop threads its a very bad idea
@@ -5019,6 +5140,7 @@ void init_idle(struct task_struct *idle, int cpu)
 	raw_spin_lock(&rq->lock);
 
 	__sched_fork(0, idle);
+
 	idle->state = TASK_RUNNING;
 	idle->se.exec_start = sched_clock();
 
@@ -5401,9 +5523,60 @@ set_table_entry(struct ctl_table *entry,
 }
 
 static struct ctl_table *
+sd_alloc_ctl_energy_table(struct sched_group_energy *sge)
+{
+	struct ctl_table *table = sd_alloc_ctl_entry(5);
+
+	if (table == NULL)
+		return NULL;
+
+	set_table_entry(&table[0], "nr_idle_states", &sge->nr_idle_states,
+			sizeof(int), 0644, proc_dointvec_minmax, false);
+	set_table_entry(&table[1], "idle_states", &sge->idle_states[0].power,
+			sge->nr_idle_states*sizeof(struct idle_state), 0644,
+			proc_doulongvec_minmax, false);
+	set_table_entry(&table[2], "nr_cap_states", &sge->nr_cap_states,
+			sizeof(int), 0644, proc_dointvec_minmax, false);
+	set_table_entry(&table[3], "cap_states", &sge->cap_states[0].cap,
+			sge->nr_cap_states*sizeof(struct capacity_state), 0644,
+			proc_doulongvec_minmax, false);
+
+	return table;
+}
+
+static struct ctl_table *
+sd_alloc_ctl_group_table(struct sched_group *sg)
+{
+	struct ctl_table *table = sd_alloc_ctl_entry(2);
+
+	if (table == NULL)
+		return NULL;
+
+	table->procname = kstrdup("energy", GFP_KERNEL);
+	table->mode = 0555;
+	table->child = sd_alloc_ctl_energy_table((struct sched_group_energy *)sg->sge);
+
+	return table;
+}
+
+static struct ctl_table *
 sd_alloc_ctl_domain_table(struct sched_domain *sd)
 {
-	struct ctl_table *table = sd_alloc_ctl_entry(14);
+	struct ctl_table *table;
+	unsigned int nr_entries = 14;
+
+	int i = 0;
+	struct sched_group *sg = sd->groups;
+
+	if (sg->sge) {
+		int nr_sgs = 0;
+
+		do {} while (nr_sgs++, sg = sg->next, sg != sd->groups);
+
+		nr_entries += nr_sgs;
+	}
+
+	table = sd_alloc_ctl_entry(nr_entries);
 
 	if (table == NULL)
 		return NULL;
@@ -5436,7 +5609,19 @@ sd_alloc_ctl_domain_table(struct sched_domain *sd)
 		sizeof(long), 0644, proc_doulongvec_minmax, false);
 	set_table_entry(&table[12], "name", sd->name,
 		CORENAME_MAX_SIZE, 0444, proc_dostring, false);
-	/* &table[13] is terminator */
+	sg = sd->groups;
+	if (sg->sge) {
+		char buf[32];
+		struct ctl_table *entry = &table[13];
+
+		do {
+			snprintf(buf, 32, "group%d", i);
+			entry->procname = kstrdup(buf, GFP_KERNEL);
+			entry->mode = 0555;
+			entry->child = sd_alloc_ctl_group_table(sg);
+		} while (entry++, i++, sg = sg->next, sg != sd->groups);
+	}
+	/* &table[nr_entries-1] is terminator */
 
 	return table;
 }
@@ -5552,6 +5737,9 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 	switch (action & ~CPU_TASKS_FROZEN) {
 
 	case CPU_UP_PREPARE:
+		raw_spin_lock_irqsave(&rq->lock, flags);
+		walt_set_window_start(rq);
+		raw_spin_unlock_irqrestore(&rq->lock, flags);
 		rq->calc_load_update = calc_load_update;
 		break;
 
@@ -5571,6 +5759,7 @@ migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu)
 		sched_ttwu_pending();
 		/* Update our root-domain */
 		raw_spin_lock_irqsave(&rq->lock, flags);
+		walt_migrate_sync_cpu(cpu);
 		if (rq->rd) {
 			BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span));
 			set_rq_offline(rq);
@@ -5698,9 +5887,6 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 
 	if (!(sd->flags & SD_LOAD_BALANCE)) {
 		printk("does not load-balance\n");
-		if (sd->parent)
-			printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain"
-					" has parent");
 		return -1;
 	}
 
@@ -5742,7 +5928,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
 		printk(KERN_CONT " %*pbl",
 		       cpumask_pr_args(sched_group_cpus(group)));
 		if (group->sgc->capacity != SCHED_CAPACITY_SCALE) {
-			printk(KERN_CONT " (cpu_capacity = %d)",
+			printk(KERN_CONT " (cpu_capacity = %lu)",
 				group->sgc->capacity);
 		}
 
@@ -5793,8 +5979,12 @@ static inline bool sched_debug(void)
 
 static int sd_degenerate(struct sched_domain *sd)
 {
-	if (cpumask_weight(sched_domain_span(sd)) == 1)
-		return 1;
+	if (cpumask_weight(sched_domain_span(sd)) == 1) {
+		if (sd->groups->sge)
+			sd->flags &= ~SD_LOAD_BALANCE;
+		else
+			return 1;
+	}
 
 	/* Following flags need at least 2 groups */
 	if (sd->flags & (SD_LOAD_BALANCE |
@@ -5802,8 +5992,10 @@ static int sd_degenerate(struct sched_domain *sd)
 			 SD_BALANCE_FORK |
 			 SD_BALANCE_EXEC |
 			 SD_SHARE_CPUCAPACITY |
+			 SD_ASYM_CPUCAPACITY |
 			 SD_SHARE_PKG_RESOURCES |
-			 SD_SHARE_POWERDOMAIN)) {
+			 SD_SHARE_POWERDOMAIN |
+			 SD_SHARE_CAP_STATES)) {
 		if (sd->groups != sd->groups->next)
 			return 0;
 	}
@@ -5832,10 +6024,16 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
 				SD_BALANCE_NEWIDLE |
 				SD_BALANCE_FORK |
 				SD_BALANCE_EXEC |
+				SD_ASYM_CPUCAPACITY |
 				SD_SHARE_CPUCAPACITY |
 				SD_SHARE_PKG_RESOURCES |
 				SD_PREFER_SIBLING |
-				SD_SHARE_POWERDOMAIN);
+				SD_SHARE_POWERDOMAIN |
+				SD_SHARE_CAP_STATES);
+		if (parent->groups->sge) {
+			parent->flags &= ~SD_LOAD_BALANCE;
+			return 0;
+		}
 		if (nr_node_ids == 1)
 			pflags &= ~SD_SERIALIZE;
 	}
@@ -5914,6 +6112,11 @@ static int init_rootdomain(struct root_domain *rd)
 
 	if (cpupri_init(&rd->cpupri) != 0)
 		goto free_rto_mask;
+
+	init_max_cpu_capacity(&rd->max_cpu_capacity);
+
+	rd->max_cap_orig_cpu = rd->min_cap_orig_cpu = -1;
+
 	return 0;
 
 free_rto_mask:
@@ -6019,11 +6222,13 @@ DEFINE_PER_CPU(int, sd_llc_id);
 DEFINE_PER_CPU(struct sched_domain *, sd_numa);
 DEFINE_PER_CPU(struct sched_domain *, sd_busy);
 DEFINE_PER_CPU(struct sched_domain *, sd_asym);
+DEFINE_PER_CPU(struct sched_domain *, sd_ea);
+DEFINE_PER_CPU(struct sched_domain *, sd_scs);
 
 static void update_top_cache_domain(int cpu)
 {
 	struct sched_domain *sd;
-	struct sched_domain *busy_sd = NULL;
+	struct sched_domain *busy_sd = NULL, *ea_sd = NULL;
 	int id = cpu;
 	int size = 1;
 
@@ -6044,6 +6249,17 @@ static void update_top_cache_domain(int cpu)
 
 	sd = highest_flag_domain(cpu, SD_ASYM_PACKING);
 	rcu_assign_pointer(per_cpu(sd_asym, cpu), sd);
+
+	for_each_domain(cpu, sd) {
+		if (sd->groups->sge)
+			ea_sd = sd;
+		else
+			break;
+	}
+	rcu_assign_pointer(per_cpu(sd_ea, cpu), ea_sd);
+
+	sd = highest_flag_domain(cpu, SD_SHARE_CAP_STATES);
+	rcu_assign_pointer(per_cpu(sd_scs, cpu), sd);
 }
 
 /*
@@ -6220,6 +6436,8 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu)
 		 * die on a /0 trap.
 		 */
 		sg->sgc->capacity = SCHED_CAPACITY_SCALE * cpumask_weight(sg_span);
+		sg->sgc->max_capacity = SCHED_CAPACITY_SCALE;
+		sg->sgc->min_capacity = SCHED_CAPACITY_SCALE;
 
 		/*
 		 * Make sure the first group of this domain contains the
@@ -6349,6 +6567,66 @@ static void init_sched_groups_capacity(int cpu, struct sched_domain *sd)
 }
 
 /*
+ * Check that the per-cpu provided sd energy data is consistent for all cpus
+ * within the mask.
+ */
+static inline void check_sched_energy_data(int cpu, sched_domain_energy_f fn,
+					   const struct cpumask *cpumask)
+{
+	const struct sched_group_energy * const sge = fn(cpu);
+	struct cpumask mask;
+	int i;
+
+	if (cpumask_weight(cpumask) <= 1)
+		return;
+
+	cpumask_xor(&mask, cpumask, get_cpu_mask(cpu));
+
+	for_each_cpu(i, &mask) {
+		const struct sched_group_energy * const e = fn(i);
+		int y;
+
+		BUG_ON(e->nr_idle_states != sge->nr_idle_states);
+
+		for (y = 0; y < (e->nr_idle_states); y++) {
+			BUG_ON(e->idle_states[y].power !=
+					sge->idle_states[y].power);
+		}
+
+		BUG_ON(e->nr_cap_states != sge->nr_cap_states);
+
+		for (y = 0; y < (e->nr_cap_states); y++) {
+			BUG_ON(e->cap_states[y].cap != sge->cap_states[y].cap);
+			BUG_ON(e->cap_states[y].power !=
+					sge->cap_states[y].power);
+		}
+	}
+}
+
+static void init_sched_energy(int cpu, struct sched_domain *sd,
+			      sched_domain_energy_f fn)
+{
+	if (!(fn && fn(cpu)))
+		return;
+
+	if (cpu != group_balance_cpu(sd->groups))
+		return;
+
+	if (sd->child && !sd->child->groups->sge) {
+		pr_err("BUG: EAS setup broken for CPU%d\n", cpu);
+#ifdef CONFIG_SCHED_DEBUG
+		pr_err("     energy data on %s but not on %s domain\n",
+			sd->name, sd->child->name);
+#endif
+		return;
+	}
+
+	check_sched_energy_data(cpu, fn, sched_group_cpus(sd->groups));
+
+	sd->groups->sge = fn(cpu);
+}
+
+/*
  * Initializers for schedule domains
  * Non-inlined to reduce accumulated stack pressure in build_sched_domains()
  */
@@ -6452,10 +6730,19 @@ static int sched_domains_curr_level;
 /*
  * SD_flags allowed in topology descriptions.
  *
- * SD_SHARE_CPUCAPACITY      - describes SMT topologies
- * SD_SHARE_PKG_RESOURCES - describes shared caches
- * SD_NUMA                - describes NUMA topologies
- * SD_SHARE_POWERDOMAIN   - describes shared power domain
+ * These flags are purely descriptive of the topology and do not prescribe
+ * behaviour. Behaviour is artificial and mapped in the below sd_init()
+ * function:
+ *
+ *   SD_SHARE_CPUCAPACITY   - describes SMT topologies
+ *   SD_SHARE_PKG_RESOURCES - describes shared caches
+ *   SD_NUMA                - describes NUMA topologies
+ *   SD_SHARE_POWERDOMAIN   - describes shared power domain
+ *   SD_ASYM_CPUCAPACITY    - describes mixed capacity topologies
+ *   SD_SHARE_CAP_STATES    - describes shared capacity states
+ *
+ * Odd one out, which beside describing the topology has a quirk also
+ * prescribes the desired behaviour that goes along with it:
  *
  * Odd one out:
  * SD_ASYM_PACKING        - describes SMT quirks
@@ -6465,10 +6752,13 @@ static int sched_domains_curr_level;
 	 SD_SHARE_PKG_RESOURCES |	\
 	 SD_NUMA |			\
 	 SD_ASYM_PACKING |		\
-	 SD_SHARE_POWERDOMAIN)
+	 SD_ASYM_CPUCAPACITY |		\
+	 SD_SHARE_POWERDOMAIN |		\
+	 SD_SHARE_CAP_STATES)
 
 static struct sched_domain *
-sd_init(struct sched_domain_topology_level *tl, int cpu)
+sd_init(struct sched_domain_topology_level *tl,
+	struct sched_domain *child, int cpu)
 {
 	struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu);
 	int sd_weight, sd_flags = 0;
@@ -6520,6 +6810,7 @@ sd_init(struct sched_domain_topology_level *tl, int cpu)
 		.smt_gain		= 0,
 		.max_newidle_lb_cost	= 0,
 		.next_decay_max_lb_cost	= jiffies,
+		.child			= child,
 #ifdef CONFIG_SCHED_DEBUG
 		.name			= tl->name,
 #endif
@@ -6529,6 +6820,13 @@ sd_init(struct sched_domain_topology_level *tl, int cpu)
 	 * Convert topological properties into behaviour.
 	 */
 
+	if (sd->flags & SD_ASYM_CPUCAPACITY) {
+		struct sched_domain *t = sd;
+
+		for_each_lower_domain(t)
+			t->flags |= SD_BALANCE_WAKE;
+	}
+
 	if (sd->flags & SD_SHARE_CPUCAPACITY) {
 		sd->flags |= SD_PREFER_SIBLING;
 		sd->imbalance_pct = 110;
@@ -6975,16 +7273,13 @@ struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl,
 		const struct cpumask *cpu_map, struct sched_domain_attr *attr,
 		struct sched_domain *child, int cpu)
 {
-	struct sched_domain *sd = sd_init(tl, cpu);
-	if (!sd)
-		return child;
+	struct sched_domain *sd = sd_init(tl, child, cpu);
 
 	cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu));
 	if (child) {
 		sd->level = child->level + 1;
 		sched_domain_level_max = max(sched_domain_level_max, sd->level);
 		child->parent = sd;
-		sd->child = child;
 
 		if (!cpumask_subset(sched_domain_span(child),
 				    sched_domain_span(sd))) {
@@ -7032,8 +7327,6 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 				*per_cpu_ptr(d.sd, i) = sd;
 			if (tl->flags & SDTL_OVERLAP || sched_feat(FORCE_SD_OVERLAP))
 				sd->flags |= SD_OVERLAP;
-			if (cpumask_equal(cpu_map, sched_domain_span(sd)))
-				break;
 		}
 	}
 
@@ -7053,10 +7346,13 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 
 	/* Calculate CPU capacity for physical packages and nodes */
 	for (i = nr_cpumask_bits-1; i >= 0; i--) {
+		struct sched_domain_topology_level *tl = sched_domain_topology;
+
 		if (!cpumask_test_cpu(i, cpu_map))
 			continue;
 
-		for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) {
+		for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent, tl++) {
+			init_sched_energy(i, sd, tl->energy);
 			claim_allocations(i, sd);
 			init_sched_groups_capacity(i, sd);
 		}
@@ -7065,7 +7361,19 @@ static int build_sched_domains(const struct cpumask *cpu_map,
 	/* Attach the domains */
 	rcu_read_lock();
 	for_each_cpu(i, cpu_map) {
+		int max_cpu = READ_ONCE(d.rd->max_cap_orig_cpu);
+		int min_cpu = READ_ONCE(d.rd->min_cap_orig_cpu);
+
+		if ((max_cpu < 0) || (cpu_rq(i)->cpu_capacity_orig >
+		    cpu_rq(max_cpu)->cpu_capacity_orig))
+			WRITE_ONCE(d.rd->max_cap_orig_cpu, i);
+
+		if ((min_cpu < 0) || (cpu_rq(i)->cpu_capacity_orig <
+		    cpu_rq(min_cpu)->cpu_capacity_orig))
+			WRITE_ONCE(d.rd->min_cap_orig_cpu, i);
+
 		sd = *per_cpu_ptr(d.sd, i);
+
 		cpu_attach_domain(sd, d.rd, i);
 	}
 	rcu_read_unlock();
@@ -7476,6 +7784,7 @@ void __init sched_init(void)
 #ifdef CONFIG_FAIR_GROUP_SCHED
 		root_task_group.shares = ROOT_TASK_GROUP_LOAD;
 		INIT_LIST_HEAD(&rq->leaf_cfs_rq_list);
+		rq->tmp_alone_branch = &rq->leaf_cfs_rq_list;
 		/*
 		 * How much cpu bandwidth does root_task_group get?
 		 *
@@ -7517,11 +7826,17 @@ void __init sched_init(void)
 		rq->active_balance = 0;
 		rq->next_balance = jiffies;
 		rq->push_cpu = 0;
+		rq->push_task = NULL;
 		rq->cpu = i;
 		rq->online = 0;
 		rq->idle_stamp = 0;
 		rq->avg_idle = 2*sysctl_sched_migration_cost;
 		rq->max_idle_balance_cost = sysctl_sched_migration_cost;
+#ifdef CONFIG_SCHED_WALT
+		rq->cur_irqload = 0;
+		rq->avg_irqload = 0;
+		rq->irqload_ts = 0;
+#endif
 
 		INIT_LIST_HEAD(&rq->cfs_tasks);
 
@@ -7585,6 +7900,14 @@ static inline int preempt_count_equals(int preempt_offset)
 	return (nested == preempt_offset);
 }
 
+static int __might_sleep_init_called;
+int __init __might_sleep_init(void)
+{
+	__might_sleep_init_called = 1;
+	return 0;
+}
+early_initcall(__might_sleep_init);
+
 void __might_sleep(const char *file, int line, int preempt_offset)
 {
 	/*
@@ -7609,8 +7932,10 @@ void ___might_sleep(const char *file, int line, int preempt_offset)
 
 	rcu_sleep_check(); /* WARN_ON_ONCE() by default, no rate limit reqd. */
 	if ((preempt_count_equals(preempt_offset) && !irqs_disabled() &&
-	     !is_idle_task(current)) ||
-	    system_state != SYSTEM_RUNNING || oops_in_progress)
+	     !is_idle_task(current)) || oops_in_progress)
+		return;
+	if (system_state != SYSTEM_RUNNING &&
+	    (!__might_sleep_init_called || system_state != SYSTEM_BOOTING))
 		return;
 	if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy)
 		return;
@@ -7808,27 +8133,9 @@ void sched_offline_group(struct task_group *tg)
 	spin_unlock_irqrestore(&task_group_lock, flags);
 }
 
-/* change task's runqueue when it moves between groups.
- *	The caller of this function should have put the task in its new group
- *	by now. This function just updates tsk->se.cfs_rq and tsk->se.parent to
- *	reflect its new group.
- */
-void sched_move_task(struct task_struct *tsk)
+static void sched_change_group(struct task_struct *tsk, int type)
 {
 	struct task_group *tg;
-	int queued, running;
-	unsigned long flags;
-	struct rq *rq;
-
-	rq = task_rq_lock(tsk, &flags);
-
-	running = task_current(rq, tsk);
-	queued = task_on_rq_queued(tsk);
-
-	if (queued)
-		dequeue_task(rq, tsk, DEQUEUE_SAVE);
-	if (unlikely(running))
-		put_prev_task(rq, tsk);
 
 	/*
 	 * All callers are synchronized by task_rq_lock(); we do not use RCU
@@ -7841,11 +8148,37 @@ void sched_move_task(struct task_struct *tsk)
 	tsk->sched_task_group = tg;
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
-	if (tsk->sched_class->task_move_group)
-		tsk->sched_class->task_move_group(tsk);
+	if (tsk->sched_class->task_change_group)
+		tsk->sched_class->task_change_group(tsk, type);
 	else
 #endif
 		set_task_rq(tsk, task_cpu(tsk));
+}
+
+/*
+ * Change task's runqueue when it moves between groups.
+ *
+ * The caller of this function should have put the task in its new group by
+ * now. This function just updates tsk->se.cfs_rq and tsk->se.parent to reflect
+ * its new group.
+ */
+void sched_move_task(struct task_struct *tsk)
+{
+	int queued, running;
+	unsigned long flags;
+	struct rq *rq;
+
+	rq = task_rq_lock(tsk, &flags);
+
+	running = task_current(rq, tsk);
+	queued = task_on_rq_queued(tsk);
+
+	if (queued)
+		dequeue_task(rq, tsk, DEQUEUE_SAVE);
+	if (unlikely(running))
+		put_prev_task(rq, tsk);
+
+	sched_change_group(tsk, TASK_MOVE_GROUP);
 
 	if (unlikely(running))
 		tsk->sched_class->set_curr_task(rq);
@@ -8282,15 +8615,28 @@ static void cpu_cgroup_css_free(struct cgroup_subsys_state *css)
 	sched_free_group(tg);
 }
 
+/*
+ * This is called before wake_up_new_task(), therefore we really only
+ * have to set its group bits, all the other stuff does not apply.
+ */
 static void cpu_cgroup_fork(struct task_struct *task, void *private)
 {
-	sched_move_task(task);
+	unsigned long flags;
+	struct rq *rq;
+
+	rq = task_rq_lock(task, &flags);
+
+	update_rq_clock(rq);
+	sched_change_group(task, TASK_SET_GROUP);
+
+	task_rq_unlock(rq, task, &flags);
 }
 
 static int cpu_cgroup_can_attach(struct cgroup_taskset *tset)
 {
 	struct task_struct *task;
 	struct cgroup_subsys_state *css;
+	int ret = 0;
 
 	cgroup_taskset_for_each(task, css, tset) {
 #ifdef CONFIG_RT_GROUP_SCHED
@@ -8301,8 +8647,24 @@ static int cpu_cgroup_can_attach(struct cgroup_taskset *tset)
 		if (task->sched_class != &fair_sched_class)
 			return -EINVAL;
 #endif
+		/*
+		 * Serialize against wake_up_new_task() such that if its
+		 * running, we're sure to observe its full state.
+		 */
+		raw_spin_lock_irq(&task->pi_lock);
+		/*
+		 * Avoid calling sched_move_task() before wake_up_new_task()
+		 * has happened. This would lead to problems with PELT, due to
+		 * move wanting to detach+attach while we're not attached yet.
+		 */
+		if (task->state == TASK_NEW)
+			ret = -EINVAL;
+		raw_spin_unlock_irq(&task->pi_lock);
+
+		if (ret)
+			break;
 	}
-	return 0;
+	return ret;
 }
 
 static void cpu_cgroup_attach(struct cgroup_taskset *tset)
diff --git a/kernel/sched/cpufreq.c b/kernel/sched/cpufreq.c
new file mode 100644
index 000000000000..dbc51442ecbc
--- /dev/null
+++ b/kernel/sched/cpufreq.c
@@ -0,0 +1,63 @@
+/*
+ * Scheduler code and data structures related to cpufreq.
+ *
+ * Copyright (C) 2016, Intel Corporation
+ * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include "sched.h"
+
+DEFINE_PER_CPU(struct update_util_data *, cpufreq_update_util_data);
+
+/**
+ * cpufreq_add_update_util_hook - Populate the CPU's update_util_data pointer.
+ * @cpu: The CPU to set the pointer for.
+ * @data: New pointer value.
+ * @func: Callback function to set for the CPU.
+ *
+ * Set and publish the update_util_data pointer for the given CPU.
+ *
+ * The update_util_data pointer of @cpu is set to @data and the callback
+ * function pointer in the target struct update_util_data is set to @func.
+ * That function will be called by cpufreq_update_util() from RCU-sched
+ * read-side critical sections, so it must not sleep.  @data will always be
+ * passed to it as the first argument which allows the function to get to the
+ * target update_util_data structure and its container.
+ *
+ * The update_util_data pointer of @cpu must be NULL when this function is
+ * called or it will WARN() and return with no effect.
+ */
+void cpufreq_add_update_util_hook(int cpu, struct update_util_data *data,
+			void (*func)(struct update_util_data *data, u64 time,
+				     unsigned int flags))
+{
+	if (WARN_ON(!data || !func))
+		return;
+
+	if (WARN_ON(per_cpu(cpufreq_update_util_data, cpu)))
+		return;
+
+	data->func = func;
+	rcu_assign_pointer(per_cpu(cpufreq_update_util_data, cpu), data);
+}
+EXPORT_SYMBOL_GPL(cpufreq_add_update_util_hook);
+
+/**
+ * cpufreq_remove_update_util_hook - Clear the CPU's update_util_data pointer.
+ * @cpu: The CPU to clear the pointer for.
+ *
+ * Clear the update_util_data pointer for the given CPU.
+ *
+ * Callers must use RCU-sched callbacks to free any memory that might be
+ * accessed via the old update_util_data pointer or invoke synchronize_sched()
+ * right after this function to avoid use-after-free.
+ */
+void cpufreq_remove_update_util_hook(int cpu)
+{
+	rcu_assign_pointer(per_cpu(cpufreq_update_util_data, cpu), NULL);
+}
+EXPORT_SYMBOL_GPL(cpufreq_remove_update_util_hook);
diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c
new file mode 100644
index 000000000000..d3765f0cb699
--- /dev/null
+++ b/kernel/sched/cpufreq_schedutil.c
@@ -0,0 +1,820 @@
+/*
+ * CPUFreq governor based on scheduler-provided CPU utilization data.
+ *
+ * Copyright (C) 2016, Intel Corporation
+ * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/cpufreq.h>
+#include <linux/kthread.h>
+#include <linux/slab.h>
+#include <trace/events/power.h>
+
+#include "sched.h"
+#include "tune.h"
+
+unsigned long boosted_cpu_util(int cpu);
+
+/* Stub out fast switch routines present on mainline to reduce the backport
+ * overhead. */
+#define cpufreq_driver_fast_switch(x, y) 0
+#define cpufreq_enable_fast_switch(x)
+#define cpufreq_disable_fast_switch(x)
+#define LATENCY_MULTIPLIER			(1000)
+#define SUGOV_KTHREAD_PRIORITY	50
+
+struct sugov_tunables {
+	struct gov_attr_set attr_set;
+	unsigned int up_rate_limit_us;
+	unsigned int down_rate_limit_us;
+};
+
+struct sugov_policy {
+	struct cpufreq_policy *policy;
+
+	struct sugov_tunables *tunables;
+	struct list_head tunables_hook;
+
+	raw_spinlock_t update_lock;  /* For shared policies */
+	u64 last_freq_update_time;
+	s64 min_rate_limit_ns;
+	s64 up_rate_delay_ns;
+	s64 down_rate_delay_ns;
+	unsigned int next_freq;
+	unsigned int cached_raw_freq;
+
+	/* The next fields are only needed if fast switch cannot be used. */
+	struct irq_work irq_work;
+	struct kthread_work work;
+	struct mutex work_lock;
+	struct kthread_worker worker;
+	struct task_struct *thread;
+	bool work_in_progress;
+
+	bool need_freq_update;
+};
+
+struct sugov_cpu {
+	struct update_util_data update_util;
+	struct sugov_policy *sg_policy;
+
+	bool iowait_boost_pending;
+	unsigned int iowait_boost;
+	unsigned int iowait_boost_max;
+	u64 last_update;
+
+	/* The fields below are only needed when sharing a policy. */
+	unsigned long util;
+	unsigned long max;
+	unsigned int flags;
+
+	/* The field below is for single-CPU policies only. */
+#ifdef CONFIG_NO_HZ_COMMON
+	unsigned long saved_idle_calls;
+#endif
+};
+
+static DEFINE_PER_CPU(struct sugov_cpu, sugov_cpu);
+
+/************************ Governor internals ***********************/
+
+static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time)
+{
+	s64 delta_ns;
+
+	if (sg_policy->work_in_progress)
+		return false;
+
+	if (unlikely(sg_policy->need_freq_update)) {
+		sg_policy->need_freq_update = false;
+		/*
+		 * This happens when limits change, so forget the previous
+		 * next_freq value and force an update.
+		 */
+		sg_policy->next_freq = UINT_MAX;
+		return true;
+	}
+
+	delta_ns = time - sg_policy->last_freq_update_time;
+
+	/* No need to recalculate next freq for min_rate_limit_us at least */
+	return delta_ns >= sg_policy->min_rate_limit_ns;
+}
+
+static bool sugov_up_down_rate_limit(struct sugov_policy *sg_policy, u64 time,
+				     unsigned int next_freq)
+{
+	s64 delta_ns;
+
+	delta_ns = time - sg_policy->last_freq_update_time;
+
+	if (next_freq > sg_policy->next_freq &&
+	    delta_ns < sg_policy->up_rate_delay_ns)
+			return true;
+
+	if (next_freq < sg_policy->next_freq &&
+	    delta_ns < sg_policy->down_rate_delay_ns)
+			return true;
+
+	return false;
+}
+
+static void sugov_update_commit(struct sugov_policy *sg_policy, u64 time,
+				unsigned int next_freq)
+{
+	struct cpufreq_policy *policy = sg_policy->policy;
+
+	if (sugov_up_down_rate_limit(sg_policy, time, next_freq))
+		return;
+
+	if (sg_policy->next_freq == next_freq)
+		return;
+
+	sg_policy->next_freq = next_freq;
+	sg_policy->last_freq_update_time = time;
+
+	if (policy->fast_switch_enabled) {
+		next_freq = cpufreq_driver_fast_switch(policy, next_freq);
+		if (next_freq == CPUFREQ_ENTRY_INVALID)
+			return;
+
+		policy->cur = next_freq;
+		trace_cpu_frequency(next_freq, smp_processor_id());
+	} else {
+		sg_policy->work_in_progress = true;
+		irq_work_queue(&sg_policy->irq_work);
+	}
+}
+
+/**
+ * get_next_freq - Compute a new frequency for a given cpufreq policy.
+ * @sg_policy: schedutil policy object to compute the new frequency for.
+ * @util: Current CPU utilization.
+ * @max: CPU capacity.
+ *
+ * If the utilization is frequency-invariant, choose the new frequency to be
+ * proportional to it, that is
+ *
+ * next_freq = C * max_freq * util / max
+ *
+ * Otherwise, approximate the would-be frequency-invariant utilization by
+ * util_raw * (curr_freq / max_freq) which leads to
+ *
+ * next_freq = C * curr_freq * util_raw / max
+ *
+ * Take C = 1.25 for the frequency tipping point at (util / max) = 0.8.
+ *
+ * The lowest driver-supported frequency which is equal or greater than the raw
+ * next_freq (as calculated above) is returned, subject to policy min/max and
+ * cpufreq driver limitations.
+ */
+static unsigned int get_next_freq(struct sugov_policy *sg_policy,
+				  unsigned long util, unsigned long max)
+{
+	struct cpufreq_policy *policy = sg_policy->policy;
+	unsigned int freq = arch_scale_freq_invariant() ?
+				policy->cpuinfo.max_freq : policy->cur;
+
+	freq = (freq + (freq >> 2)) * util / max;
+
+	if (freq == sg_policy->cached_raw_freq && sg_policy->next_freq != UINT_MAX)
+		return sg_policy->next_freq;
+	sg_policy->cached_raw_freq = freq;
+	return cpufreq_driver_resolve_freq(policy, freq);
+}
+
+static inline bool use_pelt(void)
+{
+#ifdef CONFIG_SCHED_WALT
+	return (!sysctl_sched_use_walt_cpu_util || walt_disabled);
+#else
+	return true;
+#endif
+}
+
+static void sugov_get_util(unsigned long *util, unsigned long *max, u64 time)
+{
+	int cpu = smp_processor_id();
+	struct rq *rq = cpu_rq(cpu);
+	unsigned long max_cap, rt;
+	s64 delta;
+
+	max_cap = arch_scale_cpu_capacity(NULL, cpu);
+
+	sched_avg_update(rq);
+	delta = time - rq->age_stamp;
+	if (unlikely(delta < 0))
+		delta = 0;
+	rt = div64_u64(rq->rt_avg, sched_avg_period() + delta);
+	rt = (rt * max_cap) >> SCHED_CAPACITY_SHIFT;
+
+	*util = boosted_cpu_util(cpu);
+	if (likely(use_pelt()))
+		*util = *util + rt;
+
+	*util = min(*util, max_cap);
+	*max = max_cap;
+}
+
+static void sugov_set_iowait_boost(struct sugov_cpu *sg_cpu, u64 time,
+				   unsigned int flags)
+{
+	if (flags & SCHED_CPUFREQ_IOWAIT) {
+		if (sg_cpu->iowait_boost_pending)
+			return;
+
+		sg_cpu->iowait_boost_pending = true;
+
+		if (sg_cpu->iowait_boost) {
+			sg_cpu->iowait_boost <<= 1;
+			if (sg_cpu->iowait_boost > sg_cpu->iowait_boost_max)
+				sg_cpu->iowait_boost = sg_cpu->iowait_boost_max;
+		} else {
+			sg_cpu->iowait_boost = sg_cpu->sg_policy->policy->min;
+		}
+	} else if (sg_cpu->iowait_boost) {
+		s64 delta_ns = time - sg_cpu->last_update;
+
+		/* Clear iowait_boost if the CPU apprears to have been idle. */
+		if (delta_ns > TICK_NSEC) {
+			sg_cpu->iowait_boost = 0;
+			sg_cpu->iowait_boost_pending = false;
+		}
+	}
+}
+
+static void sugov_iowait_boost(struct sugov_cpu *sg_cpu, unsigned long *util,
+			       unsigned long *max)
+{
+	unsigned int boost_util, boost_max;
+
+	if (!sg_cpu->iowait_boost)
+		return;
+
+	if (sg_cpu->iowait_boost_pending) {
+		sg_cpu->iowait_boost_pending = false;
+	} else {
+		sg_cpu->iowait_boost >>= 1;
+		if (sg_cpu->iowait_boost < sg_cpu->sg_policy->policy->min) {
+			sg_cpu->iowait_boost = 0;
+			return;
+		}
+	}
+
+	boost_util = sg_cpu->iowait_boost;
+	boost_max = sg_cpu->iowait_boost_max;
+
+	if (*util * boost_max < *max * boost_util) {
+		*util = boost_util;
+		*max = boost_max;
+	}
+}
+
+#ifdef CONFIG_NO_HZ_COMMON
+static bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu)
+{
+	unsigned long idle_calls = tick_nohz_get_idle_calls();
+	bool ret = idle_calls == sg_cpu->saved_idle_calls;
+
+	sg_cpu->saved_idle_calls = idle_calls;
+	return ret;
+}
+#else
+static inline bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu) { return false; }
+#endif /* CONFIG_NO_HZ_COMMON */
+
+static void sugov_update_single(struct update_util_data *hook, u64 time,
+				unsigned int flags)
+{
+	struct sugov_cpu *sg_cpu = container_of(hook, struct sugov_cpu, update_util);
+	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
+	struct cpufreq_policy *policy = sg_policy->policy;
+	unsigned long util, max;
+	unsigned int next_f;
+	bool busy;
+
+	sugov_set_iowait_boost(sg_cpu, time, flags);
+	sg_cpu->last_update = time;
+
+	if (!sugov_should_update_freq(sg_policy, time))
+		return;
+
+	busy = sugov_cpu_is_busy(sg_cpu);
+
+	if (flags & SCHED_CPUFREQ_DL) {
+		next_f = policy->cpuinfo.max_freq;
+	} else {
+		sugov_get_util(&util, &max, time);
+		sugov_iowait_boost(sg_cpu, &util, &max);
+		next_f = get_next_freq(sg_policy, util, max);
+		/*
+		 * Do not reduce the frequency if the CPU has not been idle
+		 * recently, as the reduction is likely to be premature then.
+		 */
+		if (busy && next_f < sg_policy->next_freq)
+			next_f = sg_policy->next_freq;
+	}
+	sugov_update_commit(sg_policy, time, next_f);
+}
+
+static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, u64 time)
+{
+	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
+	struct cpufreq_policy *policy = sg_policy->policy;
+	unsigned long util = 0, max = 1;
+	unsigned int j;
+
+	for_each_cpu(j, policy->cpus) {
+		struct sugov_cpu *j_sg_cpu = &per_cpu(sugov_cpu, j);
+		unsigned long j_util, j_max;
+		s64 delta_ns;
+
+		/*
+		 * If the CPU utilization was last updated before the previous
+		 * frequency update and the time elapsed between the last update
+		 * of the CPU utilization and the last frequency update is long
+		 * enough, don't take the CPU into account as it probably is
+		 * idle now (and clear iowait_boost for it).
+		 */
+		delta_ns = time - j_sg_cpu->last_update;
+		if (delta_ns > TICK_NSEC) {
+			j_sg_cpu->iowait_boost = 0;
+			j_sg_cpu->iowait_boost_pending = false;
+			continue;
+		}
+		if (j_sg_cpu->flags & SCHED_CPUFREQ_DL)
+			return policy->cpuinfo.max_freq;
+
+		j_util = j_sg_cpu->util;
+		j_max = j_sg_cpu->max;
+		if (j_util * max > j_max * util) {
+			util = j_util;
+			max = j_max;
+		}
+
+		sugov_iowait_boost(j_sg_cpu, &util, &max);
+	}
+
+	return get_next_freq(sg_policy, util, max);
+}
+
+static void sugov_update_shared(struct update_util_data *hook, u64 time,
+				unsigned int flags)
+{
+	struct sugov_cpu *sg_cpu = container_of(hook, struct sugov_cpu, update_util);
+	struct sugov_policy *sg_policy = sg_cpu->sg_policy;
+	unsigned long util, max;
+	unsigned int next_f;
+
+	sugov_get_util(&util, &max, time);
+
+	raw_spin_lock(&sg_policy->update_lock);
+
+	sg_cpu->util = util;
+	sg_cpu->max = max;
+	sg_cpu->flags = flags;
+
+	sugov_set_iowait_boost(sg_cpu, time, flags);
+	sg_cpu->last_update = time;
+
+	if (sugov_should_update_freq(sg_policy, time)) {
+		if (flags & SCHED_CPUFREQ_DL)
+			next_f = sg_policy->policy->cpuinfo.max_freq;
+		else
+			next_f = sugov_next_freq_shared(sg_cpu, time);
+
+		sugov_update_commit(sg_policy, time, next_f);
+	}
+
+	raw_spin_unlock(&sg_policy->update_lock);
+}
+
+static void sugov_work(struct kthread_work *work)
+{
+	struct sugov_policy *sg_policy = container_of(work, struct sugov_policy, work);
+
+	mutex_lock(&sg_policy->work_lock);
+	__cpufreq_driver_target(sg_policy->policy, sg_policy->next_freq,
+				CPUFREQ_RELATION_L);
+	mutex_unlock(&sg_policy->work_lock);
+
+	sg_policy->work_in_progress = false;
+}
+
+static void sugov_irq_work(struct irq_work *irq_work)
+{
+	struct sugov_policy *sg_policy;
+
+	sg_policy = container_of(irq_work, struct sugov_policy, irq_work);
+
+	/*
+	 * For RT and deadline tasks, the schedutil governor shoots the
+	 * frequency to maximum. Special care must be taken to ensure that this
+	 * kthread doesn't result in the same behavior.
+	 *
+	 * This is (mostly) guaranteed by the work_in_progress flag. The flag is
+	 * updated only at the end of the sugov_work() function and before that
+	 * the schedutil governor rejects all other frequency scaling requests.
+	 *
+	 * There is a very rare case though, where the RT thread yields right
+	 * after the work_in_progress flag is cleared. The effects of that are
+	 * neglected for now.
+	 */
+	queue_kthread_work(&sg_policy->worker, &sg_policy->work);
+}
+
+/************************** sysfs interface ************************/
+
+static struct sugov_tunables *global_tunables;
+static DEFINE_MUTEX(global_tunables_lock);
+
+static inline struct sugov_tunables *to_sugov_tunables(struct gov_attr_set *attr_set)
+{
+	return container_of(attr_set, struct sugov_tunables, attr_set);
+}
+
+static DEFINE_MUTEX(min_rate_lock);
+
+static void update_min_rate_limit_us(struct sugov_policy *sg_policy)
+{
+	mutex_lock(&min_rate_lock);
+	sg_policy->min_rate_limit_ns = min(sg_policy->up_rate_delay_ns,
+					   sg_policy->down_rate_delay_ns);
+	mutex_unlock(&min_rate_lock);
+}
+
+static ssize_t up_rate_limit_us_show(struct gov_attr_set *attr_set, char *buf)
+{
+	struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+
+	return sprintf(buf, "%u\n", tunables->up_rate_limit_us);
+}
+
+static ssize_t down_rate_limit_us_show(struct gov_attr_set *attr_set, char *buf)
+{
+	struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+
+	return sprintf(buf, "%u\n", tunables->down_rate_limit_us);
+}
+
+static ssize_t up_rate_limit_us_store(struct gov_attr_set *attr_set,
+				      const char *buf, size_t count)
+{
+	struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+	struct sugov_policy *sg_policy;
+	unsigned int rate_limit_us;
+
+	if (kstrtouint(buf, 10, &rate_limit_us))
+		return -EINVAL;
+
+	tunables->up_rate_limit_us = rate_limit_us;
+
+	list_for_each_entry(sg_policy, &attr_set->policy_list, tunables_hook) {
+		sg_policy->up_rate_delay_ns = rate_limit_us * NSEC_PER_USEC;
+		update_min_rate_limit_us(sg_policy);
+	}
+
+	return count;
+}
+
+static ssize_t down_rate_limit_us_store(struct gov_attr_set *attr_set,
+					const char *buf, size_t count)
+{
+	struct sugov_tunables *tunables = to_sugov_tunables(attr_set);
+	struct sugov_policy *sg_policy;
+	unsigned int rate_limit_us;
+
+	if (kstrtouint(buf, 10, &rate_limit_us))
+		return -EINVAL;
+
+	tunables->down_rate_limit_us = rate_limit_us;
+
+	list_for_each_entry(sg_policy, &attr_set->policy_list, tunables_hook) {
+		sg_policy->down_rate_delay_ns = rate_limit_us * NSEC_PER_USEC;
+		update_min_rate_limit_us(sg_policy);
+	}
+
+	return count;
+}
+
+static struct governor_attr up_rate_limit_us = __ATTR_RW(up_rate_limit_us);
+static struct governor_attr down_rate_limit_us = __ATTR_RW(down_rate_limit_us);
+
+static struct attribute *sugov_attributes[] = {
+	&up_rate_limit_us.attr,
+	&down_rate_limit_us.attr,
+	NULL
+};
+
+static struct kobj_type sugov_tunables_ktype = {
+	.default_attrs = sugov_attributes,
+	.sysfs_ops = &governor_sysfs_ops,
+};
+
+/********************** cpufreq governor interface *********************/
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_SCHEDUTIL
+static
+#endif
+struct cpufreq_governor cpufreq_gov_schedutil;
+
+static struct sugov_policy *sugov_policy_alloc(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy;
+
+	sg_policy = kzalloc(sizeof(*sg_policy), GFP_KERNEL);
+	if (!sg_policy)
+		return NULL;
+
+	sg_policy->policy = policy;
+	raw_spin_lock_init(&sg_policy->update_lock);
+	return sg_policy;
+}
+
+static void sugov_policy_free(struct sugov_policy *sg_policy)
+{
+	kfree(sg_policy);
+}
+
+static int sugov_kthread_create(struct sugov_policy *sg_policy)
+{
+	struct task_struct *thread;
+	struct sched_param param = { .sched_priority = MAX_USER_RT_PRIO / 2 };
+	struct cpufreq_policy *policy = sg_policy->policy;
+	int ret;
+
+	/* kthread only required for slow path */
+	if (policy->fast_switch_enabled)
+		return 0;
+
+	init_kthread_work(&sg_policy->work, sugov_work);
+	init_kthread_worker(&sg_policy->worker);
+	thread = kthread_create(kthread_worker_fn, &sg_policy->worker,
+				"sugov:%d",
+				cpumask_first(policy->related_cpus));
+	if (IS_ERR(thread)) {
+		pr_err("failed to create sugov thread: %ld\n", PTR_ERR(thread));
+		return PTR_ERR(thread);
+	}
+
+	ret = sched_setscheduler_nocheck(thread, SCHED_FIFO, &param);
+	if (ret) {
+		kthread_stop(thread);
+		pr_warn("%s: failed to set SCHED_FIFO\n", __func__);
+		return ret;
+	}
+
+	sg_policy->thread = thread;
+	kthread_bind_mask(thread, policy->related_cpus);
+	init_irq_work(&sg_policy->irq_work, sugov_irq_work);
+	mutex_init(&sg_policy->work_lock);
+
+	wake_up_process(thread);
+
+	return 0;
+}
+
+static void sugov_kthread_stop(struct sugov_policy *sg_policy)
+{
+	/* kthread only required for slow path */
+	if (sg_policy->policy->fast_switch_enabled)
+		return;
+
+	flush_kthread_worker(&sg_policy->worker);
+	kthread_stop(sg_policy->thread);
+	mutex_destroy(&sg_policy->work_lock);
+}
+
+static struct sugov_tunables *sugov_tunables_alloc(struct sugov_policy *sg_policy)
+{
+	struct sugov_tunables *tunables;
+
+	tunables = kzalloc(sizeof(*tunables), GFP_KERNEL);
+	if (tunables) {
+		gov_attr_set_init(&tunables->attr_set, &sg_policy->tunables_hook);
+		if (!have_governor_per_policy())
+			global_tunables = tunables;
+	}
+	return tunables;
+}
+
+static void sugov_tunables_free(struct sugov_tunables *tunables)
+{
+	if (!have_governor_per_policy())
+		global_tunables = NULL;
+
+	kfree(tunables);
+}
+
+static int sugov_init(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy;
+	struct sugov_tunables *tunables;
+	int ret = 0;
+
+	/* State should be equivalent to EXIT */
+	if (policy->governor_data)
+		return -EBUSY;
+
+	cpufreq_enable_fast_switch(policy);
+
+	sg_policy = sugov_policy_alloc(policy);
+	if (!sg_policy) {
+		ret = -ENOMEM;
+		goto disable_fast_switch;
+	}
+
+	ret = sugov_kthread_create(sg_policy);
+	if (ret)
+		goto free_sg_policy;
+
+	mutex_lock(&global_tunables_lock);
+
+	if (global_tunables) {
+		if (WARN_ON(have_governor_per_policy())) {
+			ret = -EINVAL;
+			goto stop_kthread;
+		}
+		policy->governor_data = sg_policy;
+		sg_policy->tunables = global_tunables;
+
+		gov_attr_set_get(&global_tunables->attr_set, &sg_policy->tunables_hook);
+		goto out;
+	}
+
+	tunables = sugov_tunables_alloc(sg_policy);
+	if (!tunables) {
+		ret = -ENOMEM;
+		goto stop_kthread;
+	}
+
+	if (policy->up_transition_delay_us && policy->down_transition_delay_us) {
+		tunables->up_rate_limit_us = policy->up_transition_delay_us;
+		tunables->down_rate_limit_us = policy->down_transition_delay_us;
+	} else {
+		unsigned int lat;
+
+                tunables->up_rate_limit_us = LATENCY_MULTIPLIER;
+                tunables->down_rate_limit_us = LATENCY_MULTIPLIER;
+		lat = policy->cpuinfo.transition_latency / NSEC_PER_USEC;
+		if (lat) {
+                        tunables->up_rate_limit_us *= lat;
+                        tunables->down_rate_limit_us *= lat;
+                }
+	}
+
+	policy->governor_data = sg_policy;
+	sg_policy->tunables = tunables;
+
+	ret = kobject_init_and_add(&tunables->attr_set.kobj, &sugov_tunables_ktype,
+				   get_governor_parent_kobj(policy), "%s",
+				   cpufreq_gov_schedutil.name);
+	if (ret)
+		goto fail;
+
+out:
+	mutex_unlock(&global_tunables_lock);
+	return 0;
+
+fail:
+	policy->governor_data = NULL;
+	sugov_tunables_free(tunables);
+
+stop_kthread:
+	sugov_kthread_stop(sg_policy);
+
+free_sg_policy:
+	mutex_unlock(&global_tunables_lock);
+
+	sugov_policy_free(sg_policy);
+
+disable_fast_switch:
+	cpufreq_disable_fast_switch(policy);
+
+	pr_err("initialization failed (error %d)\n", ret);
+	return ret;
+}
+
+static int sugov_exit(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+	struct sugov_tunables *tunables = sg_policy->tunables;
+	unsigned int count;
+
+	mutex_lock(&global_tunables_lock);
+
+	count = gov_attr_set_put(&tunables->attr_set, &sg_policy->tunables_hook);
+	policy->governor_data = NULL;
+	if (!count)
+		sugov_tunables_free(tunables);
+
+	mutex_unlock(&global_tunables_lock);
+
+	sugov_kthread_stop(sg_policy);
+	sugov_policy_free(sg_policy);
+
+	cpufreq_disable_fast_switch(policy);
+	return 0;
+}
+
+static int sugov_start(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+	unsigned int cpu;
+
+	sg_policy->up_rate_delay_ns =
+		sg_policy->tunables->up_rate_limit_us * NSEC_PER_USEC;
+	sg_policy->down_rate_delay_ns =
+		sg_policy->tunables->down_rate_limit_us * NSEC_PER_USEC;
+	update_min_rate_limit_us(sg_policy);
+	sg_policy->last_freq_update_time = 0;
+	sg_policy->next_freq = UINT_MAX;
+	sg_policy->work_in_progress = false;
+	sg_policy->need_freq_update = false;
+	sg_policy->cached_raw_freq = 0;
+
+	for_each_cpu(cpu, policy->cpus) {
+		struct sugov_cpu *sg_cpu = &per_cpu(sugov_cpu, cpu);
+
+		memset(sg_cpu, 0, sizeof(*sg_cpu));
+		sg_cpu->sg_policy = sg_policy;
+		sg_cpu->flags = SCHED_CPUFREQ_DL;
+		sg_cpu->iowait_boost_max = policy->cpuinfo.max_freq;
+		cpufreq_add_update_util_hook(cpu, &sg_cpu->update_util,
+					     policy_is_shared(policy) ?
+							sugov_update_shared :
+							sugov_update_single);
+	}
+	return 0;
+}
+
+static int sugov_stop(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+	unsigned int cpu;
+
+	for_each_cpu(cpu, policy->cpus)
+		cpufreq_remove_update_util_hook(cpu);
+
+	synchronize_sched();
+
+	if (!policy->fast_switch_enabled) {
+		irq_work_sync(&sg_policy->irq_work);
+		kthread_cancel_work_sync(&sg_policy->work);
+	}
+	return 0;
+}
+
+static int sugov_limits(struct cpufreq_policy *policy)
+{
+	struct sugov_policy *sg_policy = policy->governor_data;
+
+	if (!policy->fast_switch_enabled) {
+		mutex_lock(&sg_policy->work_lock);
+		cpufreq_policy_apply_limits(policy);
+		mutex_unlock(&sg_policy->work_lock);
+	}
+
+	sg_policy->need_freq_update = true;
+
+	return 0;
+}
+
+static int cpufreq_schedutil_cb(struct cpufreq_policy *policy,
+				unsigned int event)
+{
+	switch(event) {
+	case CPUFREQ_GOV_POLICY_INIT:
+		return sugov_init(policy);
+	case CPUFREQ_GOV_POLICY_EXIT:
+		return sugov_exit(policy);
+	case CPUFREQ_GOV_START:
+		return sugov_start(policy);
+	case CPUFREQ_GOV_STOP:
+		return sugov_stop(policy);
+	case CPUFREQ_GOV_LIMITS:
+		return sugov_limits(policy);
+	default:
+		BUG();
+	}
+}
+
+#ifndef CONFIG_CPU_FREQ_DEFAULT_GOV_SCHEDUTIL
+static
+#endif
+struct cpufreq_governor cpufreq_gov_schedutil = {
+	.name = "schedutil",
+	.governor = cpufreq_schedutil_cb,
+	.owner = THIS_MODULE,
+};
+
+static int __init sugov_register(void)
+{
+	return cpufreq_register_governor(&cpufreq_gov_schedutil);
+}
+fs_initcall(sugov_register);
diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c
index a1aecbedf5b1..acde1d7c763c 100644
--- a/kernel/sched/cputime.c
+++ b/kernel/sched/cputime.c
@@ -5,6 +5,7 @@
 #include <linux/static_key.h>
 #include <linux/context_tracking.h>
 #include "sched.h"
+#include "walt.h"
 
 
 #ifdef CONFIG_IRQ_TIME_ACCOUNTING
@@ -49,6 +50,10 @@ void irqtime_account_irq(struct task_struct *curr)
 	unsigned long flags;
 	s64 delta;
 	int cpu;
+#ifdef CONFIG_SCHED_WALT
+	u64 wallclock;
+	bool account = true;
+#endif
 
 	if (!sched_clock_irqtime)
 		return;
@@ -56,6 +61,9 @@ void irqtime_account_irq(struct task_struct *curr)
 	local_irq_save(flags);
 
 	cpu = smp_processor_id();
+#ifdef CONFIG_SCHED_WALT
+	wallclock = sched_clock_cpu(cpu);
+#endif
 	delta = sched_clock_cpu(cpu) - __this_cpu_read(irq_start_time);
 	__this_cpu_add(irq_start_time, delta);
 
@@ -70,8 +78,16 @@ void irqtime_account_irq(struct task_struct *curr)
 		__this_cpu_add(cpu_hardirq_time, delta);
 	else if (in_serving_softirq() && curr != this_cpu_ksoftirqd())
 		__this_cpu_add(cpu_softirq_time, delta);
+#ifdef CONFIG_SCHED_WALT
+	else
+		account = false;
+#endif
 
 	irq_time_write_end();
+#ifdef CONFIG_SCHED_WALT
+	if (account)
+		walt_account_irqtime(cpu, curr, delta, wallclock);
+#endif
 	local_irq_restore(flags);
 }
 EXPORT_SYMBOL_GPL(irqtime_account_irq);
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index e984f059e5fc..cdf2a0b97611 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -18,6 +18,8 @@
 
 #include <linux/slab.h>
 
+#include "walt.h"
+
 struct dl_bandwidth def_dl_bandwidth;
 
 static inline struct task_struct *dl_task_of(struct sched_dl_entity *dl_se)
@@ -43,6 +45,24 @@ static inline int on_dl_rq(struct sched_dl_entity *dl_se)
 	return !RB_EMPTY_NODE(&dl_se->rb_node);
 }
 
+static void add_average_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
+{
+	u64 se_bw = dl_se->dl_bw;
+
+	dl_rq->avg_bw += se_bw;
+}
+
+static void clear_average_bw(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
+{
+	u64 se_bw = dl_se->dl_bw;
+
+	dl_rq->avg_bw -= se_bw;
+	if (dl_rq->avg_bw < 0) {
+		WARN_ON(1);
+		dl_rq->avg_bw = 0;
+	}
+}
+
 static inline int is_leftmost(struct task_struct *p, struct dl_rq *dl_rq)
 {
 	struct sched_dl_entity *dl_se = &p->dl;
@@ -494,6 +514,9 @@ static void update_dl_entity(struct sched_dl_entity *dl_se,
 	struct dl_rq *dl_rq = dl_rq_of_se(dl_se);
 	struct rq *rq = rq_of_dl_rq(dl_rq);
 
+	if (dl_se->dl_new)
+		add_average_bw(dl_se, dl_rq);
+
 	/*
 	 * The arrival of a new instance needs special treatment, i.e.,
 	 * the actual scheduling parameters have to be "renewed".
@@ -732,6 +755,9 @@ static void update_curr_dl(struct rq *rq)
 	if (unlikely((s64)delta_exec <= 0))
 		return;
 
+	/* kick cpufreq (see the comment in kernel/sched/sched.h). */
+	cpufreq_update_this_cpu(rq, SCHED_CPUFREQ_DL);
+
 	schedstat_set(curr->se.statistics.exec_max,
 		      max(curr->se.statistics.exec_max, delta_exec));
 
@@ -741,8 +767,6 @@ static void update_curr_dl(struct rq *rq)
 	curr->se.exec_start = rq_clock_task(rq);
 	cpuacct_charge(curr, delta_exec);
 
-	sched_rt_avg_update(rq, delta_exec);
-
 	dl_se->runtime -= dl_se->dl_yielded ? 0 : delta_exec;
 	if (dl_runtime_exceeded(dl_se)) {
 		dl_se->dl_throttled = 1;
@@ -860,6 +884,7 @@ void inc_dl_tasks(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
 	WARN_ON(!dl_prio(prio));
 	dl_rq->dl_nr_running++;
 	add_nr_running(rq_of_dl_rq(dl_rq), 1);
+	walt_inc_cumulative_runnable_avg(rq_of_dl_rq(dl_rq), dl_task_of(dl_se));
 
 	inc_dl_deadline(dl_rq, deadline);
 	inc_dl_migration(dl_se, dl_rq);
@@ -874,6 +899,7 @@ void dec_dl_tasks(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq)
 	WARN_ON(!dl_rq->dl_nr_running);
 	dl_rq->dl_nr_running--;
 	sub_nr_running(rq_of_dl_rq(dl_rq), 1);
+	walt_dec_cumulative_runnable_avg(rq_of_dl_rq(dl_rq), dl_task_of(dl_se));
 
 	dec_dl_deadline(dl_rq, dl_se->deadline);
 	dec_dl_migration(dl_se, dl_rq);
@@ -1044,7 +1070,8 @@ static void yield_task_dl(struct rq *rq)
 static int find_later_rq(struct task_struct *task);
 
 static int
-select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_dl(struct task_struct *p, int cpu, int sd_flag, int flags,
+		  int sibling_count_hint)
 {
 	struct task_struct *curr;
 	struct rq *rq;
@@ -1241,6 +1268,8 @@ static void task_fork_dl(struct task_struct *p)
 static void task_dead_dl(struct task_struct *p)
 {
 	struct dl_bw *dl_b = dl_bw_of(task_cpu(p));
+	struct dl_rq *dl_rq = dl_rq_of_se(&p->dl);
+	struct rq *rq = rq_of_dl_rq(dl_rq);
 
 	/*
 	 * Since we are TASK_DEAD we won't slip out of the domain!
@@ -1249,6 +1278,8 @@ static void task_dead_dl(struct task_struct *p)
 	/* XXX we should retain the bw until 0-lag */
 	dl_b->total_bw -= p->dl.dl_bw;
 	raw_spin_unlock_irq(&dl_b->lock);
+
+	clear_average_bw(&p->dl, &rq->dl);
 }
 
 static void set_curr_task_dl(struct rq *rq)
@@ -1556,7 +1587,11 @@ retry:
 	}
 
 	deactivate_task(rq, next_task, 0);
+	clear_average_bw(&next_task->dl, &rq->dl);
+	next_task->on_rq = TASK_ON_RQ_MIGRATING;
 	set_task_cpu(next_task, later_rq->cpu);
+	next_task->on_rq = TASK_ON_RQ_QUEUED;
+	add_average_bw(&next_task->dl, &later_rq->dl);
 	activate_task(later_rq, next_task, 0);
 	ret = 1;
 
@@ -1644,7 +1679,11 @@ static void pull_dl_task(struct rq *this_rq)
 			resched = true;
 
 			deactivate_task(src_rq, p, 0);
+			clear_average_bw(&p->dl, &src_rq->dl);
+			p->on_rq = TASK_ON_RQ_MIGRATING;
 			set_task_cpu(p, this_cpu);
+			p->on_rq = TASK_ON_RQ_QUEUED;
+			add_average_bw(&p->dl, &this_rq->dl);
 			activate_task(this_rq, p, 0);
 			dmin = p->dl.deadline;
 
@@ -1750,6 +1789,8 @@ static void switched_from_dl(struct rq *rq, struct task_struct *p)
 	if (!start_dl_timer(p))
 		__dl_clear_params(p);
 
+	clear_average_bw(&p->dl, &rq->dl);
+
 	/*
 	 * Since this might be the only -deadline task on the rq,
 	 * this is the right place to try to pull some other one
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
index 641511771ae6..7f7116622631 100644
--- a/kernel/sched/debug.c
+++ b/kernel/sched/debug.c
@@ -597,6 +597,32 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
 	P(se.statistics.nr_wakeups_affine_attempts);
 	P(se.statistics.nr_wakeups_passive);
 	P(se.statistics.nr_wakeups_idle);
+	/* eas */
+	/* select_idle_sibling() */
+	P(se.statistics.nr_wakeups_sis_attempts);
+	P(se.statistics.nr_wakeups_sis_idle);
+	P(se.statistics.nr_wakeups_sis_cache_affine);
+	P(se.statistics.nr_wakeups_sis_suff_cap);
+	P(se.statistics.nr_wakeups_sis_idle_cpu);
+	P(se.statistics.nr_wakeups_sis_count);
+	/* select_energy_cpu_brute() */
+	P(se.statistics.nr_wakeups_secb_attempts);
+	P(se.statistics.nr_wakeups_secb_sync);
+	P(se.statistics.nr_wakeups_secb_idle_bt);
+	P(se.statistics.nr_wakeups_secb_insuff_cap);
+	P(se.statistics.nr_wakeups_secb_no_nrg_sav);
+	P(se.statistics.nr_wakeups_secb_nrg_sav);
+	P(se.statistics.nr_wakeups_secb_count);
+	/* find_best_target() */
+	P(se.statistics.nr_wakeups_fbt_attempts);
+	P(se.statistics.nr_wakeups_fbt_no_cpu);
+	P(se.statistics.nr_wakeups_fbt_no_sd);
+	P(se.statistics.nr_wakeups_fbt_pref_idle);
+	P(se.statistics.nr_wakeups_fbt_count);
+	/* cas */
+	/* select_task_rq_fair() */
+	P(se.statistics.nr_wakeups_cas_attempts);
+	P(se.statistics.nr_wakeups_cas_count);
 
 	{
 		u64 avg_atom, avg_per_cpu;
diff --git a/kernel/sched/energy.c b/kernel/sched/energy.c
new file mode 100644
index 000000000000..b0656b7a93e3
--- /dev/null
+++ b/kernel/sched/energy.c
@@ -0,0 +1,124 @@
+/*
+ * Obtain energy cost data from DT and populate relevant scheduler data
+ * structures.
+ *
+ * Copyright (C) 2015 ARM Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#define pr_fmt(fmt) "sched-energy: " fmt
+
+#define DEBUG
+
+#include <linux/gfp.h>
+#include <linux/of.h>
+#include <linux/printk.h>
+#include <linux/sched.h>
+#include <linux/sched_energy.h>
+#include <linux/stddef.h>
+
+struct sched_group_energy *sge_array[NR_CPUS][NR_SD_LEVELS];
+
+static void free_resources(void)
+{
+	int cpu, sd_level;
+	struct sched_group_energy *sge;
+
+	for_each_possible_cpu(cpu) {
+		for_each_possible_sd_level(sd_level) {
+			sge = sge_array[cpu][sd_level];
+			if (sge) {
+				kfree(sge->cap_states);
+				kfree(sge->idle_states);
+				kfree(sge);
+			}
+		}
+	}
+}
+
+void init_sched_energy_costs(void)
+{
+	struct device_node *cn, *cp;
+	struct capacity_state *cap_states;
+	struct idle_state *idle_states;
+	struct sched_group_energy *sge;
+	const struct property *prop;
+	int sd_level, i, nstates, cpu;
+	const __be32 *val;
+
+	for_each_possible_cpu(cpu) {
+		cn = of_get_cpu_node(cpu, NULL);
+		if (!cn) {
+			pr_warn("CPU device node missing for CPU %d\n", cpu);
+			return;
+		}
+
+		if (!of_find_property(cn, "sched-energy-costs", NULL)) {
+			pr_warn("CPU device node has no sched-energy-costs\n");
+			return;
+		}
+
+		for_each_possible_sd_level(sd_level) {
+			cp = of_parse_phandle(cn, "sched-energy-costs", sd_level);
+			if (!cp)
+				break;
+
+			prop = of_find_property(cp, "busy-cost-data", NULL);
+			if (!prop || !prop->value) {
+				pr_warn("No busy-cost data, skipping sched_energy init\n");
+				goto out;
+			}
+
+			sge = kcalloc(1, sizeof(struct sched_group_energy),
+				      GFP_NOWAIT);
+
+			nstates = (prop->length / sizeof(u32)) / 2;
+			cap_states = kcalloc(nstates,
+					     sizeof(struct capacity_state),
+					     GFP_NOWAIT);
+
+			for (i = 0, val = prop->value; i < nstates; i++) {
+				cap_states[i].cap = be32_to_cpup(val++);
+				cap_states[i].power = be32_to_cpup(val++);
+			}
+
+			sge->nr_cap_states = nstates;
+			sge->cap_states = cap_states;
+
+			prop = of_find_property(cp, "idle-cost-data", NULL);
+			if (!prop || !prop->value) {
+				pr_warn("No idle-cost data, skipping sched_energy init\n");
+				goto out;
+			}
+
+			nstates = (prop->length / sizeof(u32));
+			idle_states = kcalloc(nstates,
+					      sizeof(struct idle_state),
+					      GFP_NOWAIT);
+
+			for (i = 0, val = prop->value; i < nstates; i++)
+				idle_states[i].power = be32_to_cpup(val++);
+
+			sge->nr_idle_states = nstates;
+			sge->idle_states = idle_states;
+
+			sge_array[cpu][sd_level] = sge;
+		}
+	}
+
+	pr_info("Sched-energy-costs installed from DT\n");
+	return;
+
+out:
+	free_resources();
+}
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 812069b66f47..b5ea66e5551c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -30,10 +30,13 @@
 #include <linux/mempolicy.h>
 #include <linux/migrate.h>
 #include <linux/task_work.h>
+#include <linux/module.h>
 
 #include <trace/events/sched.h>
 
 #include "sched.h"
+#include "tune.h"
+#include "walt.h"
 
 /*
  * Targeted preemption latency for CPU-bound tasks:
@@ -50,6 +53,15 @@
 unsigned int sysctl_sched_latency = 6000000ULL;
 unsigned int normalized_sysctl_sched_latency = 6000000ULL;
 
+unsigned int sysctl_sched_sync_hint_enable = 1;
+unsigned int sysctl_sched_cstate_aware = 1;
+
+#ifdef CONFIG_SCHED_WALT
+unsigned int sysctl_sched_use_walt_cpu_util = 1;
+unsigned int sysctl_sched_use_walt_task_util = 1;
+__read_mostly unsigned int sysctl_sched_walt_cpu_high_irqload =
+    (10 * NSEC_PER_MSEC);
+#endif
 /*
  * The initial- and re-scaling of tunables is configurable
  * (default SCHED_TUNABLESCALING_LOG = *(1+ilog(ncpus))
@@ -114,6 +126,12 @@ unsigned int __read_mostly sysctl_sched_shares_window = 10000000UL;
 unsigned int sysctl_sched_cfs_bandwidth_slice = 5000UL;
 #endif
 
+/*
+ * The margin used when comparing utilization with CPU capacity:
+ * util * margin < capacity * 1024
+ */
+unsigned int capacity_margin = 1280; /* ~20% */
+
 static inline void update_load_add(struct load_weight *lw, unsigned long inc)
 {
 	lw->weight += inc;
@@ -286,19 +304,59 @@ static inline struct cfs_rq *group_cfs_rq(struct sched_entity *grp)
 static inline void list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq)
 {
 	if (!cfs_rq->on_list) {
+		struct rq *rq = rq_of(cfs_rq);
+		int cpu = cpu_of(rq);
 		/*
 		 * Ensure we either appear before our parent (if already
 		 * enqueued) or force our parent to appear after us when it is
-		 * enqueued.  The fact that we always enqueue bottom-up
-		 * reduces this to two cases.
+		 * enqueued. The fact that we always enqueue bottom-up
+		 * reduces this to two cases and a special case for the root
+		 * cfs_rq. Furthermore, it also means that we will always reset
+		 * tmp_alone_branch either when the branch is connected
+		 * to a tree or when we reach the beg of the tree
 		 */
 		if (cfs_rq->tg->parent &&
-		    cfs_rq->tg->parent->cfs_rq[cpu_of(rq_of(cfs_rq))]->on_list) {
-			list_add_rcu(&cfs_rq->leaf_cfs_rq_list,
-				&rq_of(cfs_rq)->leaf_cfs_rq_list);
-		} else {
+		    cfs_rq->tg->parent->cfs_rq[cpu]->on_list) {
+			/*
+			 * If parent is already on the list, we add the child
+			 * just before. Thanks to circular linked property of
+			 * the list, this means to put the child at the tail
+			 * of the list that starts by parent.
+			 */
+			list_add_tail_rcu(&cfs_rq->leaf_cfs_rq_list,
+				&(cfs_rq->tg->parent->cfs_rq[cpu]->leaf_cfs_rq_list));
+			/*
+			 * The branch is now connected to its tree so we can
+			 * reset tmp_alone_branch to the beginning of the
+			 * list.
+			 */
+			rq->tmp_alone_branch = &rq->leaf_cfs_rq_list;
+		} else if (!cfs_rq->tg->parent) {
+			/*
+			 * cfs rq without parent should be put
+			 * at the tail of the list.
+			 */
 			list_add_tail_rcu(&cfs_rq->leaf_cfs_rq_list,
-				&rq_of(cfs_rq)->leaf_cfs_rq_list);
+				&rq->leaf_cfs_rq_list);
+			/*
+			 * We have reach the beg of a tree so we can reset
+			 * tmp_alone_branch to the beginning of the list.
+			 */
+			rq->tmp_alone_branch = &rq->leaf_cfs_rq_list;
+		} else {
+			/*
+			 * The parent has not already been added so we want to
+			 * make sure that it will be put after us.
+			 * tmp_alone_branch points to the beg of the branch
+			 * where we will add parent.
+			 */
+			list_add_rcu(&cfs_rq->leaf_cfs_rq_list,
+				rq->tmp_alone_branch);
+			/*
+			 * update tmp_alone_branch to points to the new beg
+			 * of the branch
+			 */
+			rq->tmp_alone_branch = &cfs_rq->leaf_cfs_rq_list;
 		}
 
 		cfs_rq->on_list = 1;
@@ -656,7 +714,7 @@ static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se)
 }
 
 #ifdef CONFIG_SMP
-static int select_idle_sibling(struct task_struct *p, int cpu);
+static int select_idle_sibling(struct task_struct *p, int prev_cpu, int cpu);
 static unsigned long task_h_load(struct task_struct *p);
 
 /*
@@ -680,18 +738,115 @@ void init_entity_runnable_average(struct sched_entity *se)
 	 * will definitely be update (after enqueue).
 	 */
 	sa->period_contrib = 1023;
-	sa->load_avg = scale_load_down(se->load.weight);
+	/*
+	 * Tasks are intialized with full load to be seen as heavy tasks until
+	 * they get a chance to stabilize to their real load level.
+	 * Group entities are intialized with zero load to reflect the fact that
+	 * nothing has been attached to the task group yet.
+	 */
+	if (entity_is_task(se))
+		sa->load_avg = scale_load_down(se->load.weight);
 	sa->load_sum = sa->load_avg * LOAD_AVG_MAX;
-	sa->util_avg = scale_load_down(SCHED_LOAD_SCALE);
-	sa->util_sum = sa->util_avg * LOAD_AVG_MAX;
+	/*
+	 * In previous Android versions, we used to have:
+	 * 	sa->util_avg = scale_load_down(SCHED_LOAD_SCALE);
+	 * 	sa->util_sum = sa->util_avg * LOAD_AVG_MAX;
+	 * However, that functionality has been moved to enqueue.
+	 * It is unclear if we should restore this in enqueue.
+	 */
+	/*
+	 * At this point, util_avg won't be used in select_task_rq_fair anyway
+	 */
+	sa->util_avg = 0;
+	sa->util_sum = 0;
 	/* when this task enqueue'ed, it will contribute to its cfs_rq's load_avg */
 }
 
-#else
+static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq);
+static int update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq, bool update_freq);
+static void attach_entity_cfs_rq(struct sched_entity *se);
+static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se);
+
+/*
+ * With new tasks being created, their initial util_avgs are extrapolated
+ * based on the cfs_rq's current util_avg:
+ *
+ *   util_avg = cfs_rq->util_avg / (cfs_rq->load_avg + 1) * se.load.weight
+ *
+ * However, in many cases, the above util_avg does not give a desired
+ * value. Moreover, the sum of the util_avgs may be divergent, such
+ * as when the series is a harmonic series.
+ *
+ * To solve this problem, we also cap the util_avg of successive tasks to
+ * only 1/2 of the left utilization budget:
+ *
+ *   util_avg_cap = (1024 - cfs_rq->avg.util_avg) / 2^n
+ *
+ * where n denotes the nth task.
+ *
+ * For example, a simplest series from the beginning would be like:
+ *
+ *  task  util_avg: 512, 256, 128,  64,  32,   16,    8, ...
+ * cfs_rq util_avg: 512, 768, 896, 960, 992, 1008, 1016, ...
+ *
+ * Finally, that extrapolated util_avg is clamped to the cap (util_avg_cap)
+ * if util_avg > util_avg_cap.
+ */
+void post_init_entity_util_avg(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+	struct sched_avg *sa = &se->avg;
+	long cap = (long)(SCHED_CAPACITY_SCALE - cfs_rq->avg.util_avg) / 2;
+
+	if (cap > 0) {
+		if (cfs_rq->avg.util_avg != 0) {
+			sa->util_avg  = cfs_rq->avg.util_avg * se->load.weight;
+			sa->util_avg /= (cfs_rq->avg.load_avg + 1);
+
+			if (sa->util_avg > cap)
+				sa->util_avg = cap;
+		} else {
+			sa->util_avg = cap;
+		}
+		/*
+		 * If we wish to restore tuning via setting initial util,
+		 * this is where we should do it.
+		 */
+		sa->util_sum = sa->util_avg * LOAD_AVG_MAX;
+	}
+
+	if (entity_is_task(se)) {
+		struct task_struct *p = task_of(se);
+		if (p->sched_class != &fair_sched_class) {
+			/*
+			 * For !fair tasks do:
+			 *
+			update_cfs_rq_load_avg(now, cfs_rq, false);
+			attach_entity_load_avg(cfs_rq, se);
+			switched_from_fair(rq, p);
+			 *
+			 * such that the next switched_to_fair() has the
+			 * expected state.
+			 */
+			se->avg.last_update_time = cfs_rq_clock_task(cfs_rq);
+			return;
+		}
+	}
+
+	attach_entity_cfs_rq(se);
+}
+
+#else /* !CONFIG_SMP */
 void init_entity_runnable_average(struct sched_entity *se)
 {
 }
-#endif
+void post_init_entity_util_avg(struct sched_entity *se)
+{
+}
+static void update_tg_load_avg(struct cfs_rq *cfs_rq, int force)
+{
+}
+#endif /* CONFIG_SMP */
 
 /*
  * Update the current task's runtime statistics.
@@ -1388,7 +1543,8 @@ balance:
 	 * Call select_idle_sibling to maybe find a better one.
 	 */
 	if (!cur)
-		env->dst_cpu = select_idle_sibling(env->p, env->dst_cpu);
+		env->dst_cpu = select_idle_sibling(env->p, env->src_cpu,
+						   env->dst_cpu);
 
 assign:
 	assigned = true;
@@ -2373,28 +2529,22 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se)
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 # ifdef CONFIG_SMP
-static inline long calc_tg_weight(struct task_group *tg, struct cfs_rq *cfs_rq)
+static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
 {
-	long tg_weight;
+	long tg_weight, load, shares;
 
 	/*
-	 * Use this CPU's real-time load instead of the last load contribution
-	 * as the updating of the contribution is delayed, and we will use the
-	 * the real-time load to calc the share. See update_tg_load_avg().
+	 * This really should be: cfs_rq->avg.load_avg, but instead we use
+	 * cfs_rq->load.weight, which is its upper bound. This helps ramp up
+	 * the shares for small weight interactive tasks.
 	 */
-	tg_weight = atomic_long_read(&tg->load_avg);
-	tg_weight -= cfs_rq->tg_load_avg_contrib;
-	tg_weight += cfs_rq->load.weight;
-
-	return tg_weight;
-}
+	load = scale_load_down(cfs_rq->load.weight);
 
-static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
-{
-	long tg_weight, load, shares;
+	tg_weight = atomic_long_read(&tg->load_avg);
 
-	tg_weight = calc_tg_weight(tg, cfs_rq);
-	load = cfs_rq->load.weight;
+	/* Ensure tg_weight >= load */
+	tg_weight -= cfs_rq->tg_load_avg_contrib;
+	tg_weight += load;
 
 	shares = (tg->shares * load);
 	if (tg_weight)
@@ -2413,6 +2563,7 @@ static inline long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
 	return tg->shares;
 }
 # endif /* CONFIG_SMP */
+
 static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se,
 			    unsigned long weight)
 {
@@ -2431,16 +2582,20 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se,
 
 static inline int throttled_hierarchy(struct cfs_rq *cfs_rq);
 
-static void update_cfs_shares(struct cfs_rq *cfs_rq)
+static void update_cfs_shares(struct sched_entity *se)
 {
+	struct cfs_rq *cfs_rq = group_cfs_rq(se);
 	struct task_group *tg;
-	struct sched_entity *se;
 	long shares;
 
-	tg = cfs_rq->tg;
-	se = tg->se[cpu_of(rq_of(cfs_rq))];
-	if (!se || throttled_hierarchy(cfs_rq))
+	if (!cfs_rq)
 		return;
+
+	if (throttled_hierarchy(cfs_rq))
+		return;
+
+	tg = cfs_rq->tg;
+
 #ifndef CONFIG_SMP
 	if (likely(se->load.weight == tg->shares))
 		return;
@@ -2449,8 +2604,9 @@ static void update_cfs_shares(struct cfs_rq *cfs_rq)
 
 	reweight_entity(cfs_rq_of(se), se, shares);
 }
+
 #else /* CONFIG_FAIR_GROUP_SCHED */
-static inline void update_cfs_shares(struct cfs_rq *cfs_rq)
+static inline void update_cfs_shares(struct sched_entity *se)
 {
 }
 #endif /* CONFIG_FAIR_GROUP_SCHED */
@@ -2600,6 +2756,7 @@ __update_load_avg(u64 now, int cpu, struct sched_avg *sa,
 
 	scale_freq = arch_scale_freq_capacity(NULL, cpu);
 	scale_cpu = arch_scale_cpu_capacity(NULL, cpu);
+	trace_sched_contrib_scale_f(cpu, scale_freq, scale_cpu);
 
 	/* delta_w is the amount already accumulated against our next period */
 	delta_w = sa->period_contrib;
@@ -2675,25 +2832,262 @@ __update_load_avg(u64 now, int cpu, struct sched_avg *sa,
 	return decayed;
 }
 
-#ifdef CONFIG_FAIR_GROUP_SCHED
 /*
- * Updating tg's load_avg is necessary before update_cfs_share (which is done)
- * and effective_load (which is not done because it is too costly).
+ * Signed add and clamp on underflow.
+ *
+ * Explicitly do a load-store to ensure the intermediate value never hits
+ * memory. This allows lockless observations without ever seeing the negative
+ * values.
+ */
+#define add_positive(_ptr, _val) do {                           \
+	typeof(_ptr) ptr = (_ptr);                              \
+	typeof(_val) val = (_val);                              \
+	typeof(*ptr) res, var = READ_ONCE(*ptr);                \
+								\
+	res = var + val;                                        \
+								\
+	if (val < 0 && res > var)                               \
+		res = 0;                                        \
+								\
+	WRITE_ONCE(*ptr, res);                                  \
+} while (0)
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+/**
+ * update_tg_load_avg - update the tg's load avg
+ * @cfs_rq: the cfs_rq whose avg changed
+ * @force: update regardless of how small the difference
+ *
+ * This function 'ensures': tg->load_avg := \Sum tg->cfs_rq[]->avg.load.
+ * However, because tg->load_avg is a global value there are performance
+ * considerations.
+ *
+ * In order to avoid having to look at the other cfs_rq's, we use a
+ * differential update where we store the last value we propagated. This in
+ * turn allows skipping updates if the differential is 'small'.
+ *
+ * Updating tg's load_avg is necessary before update_cfs_share() (which is
+ * done) and effective_load() (which is not done because it is too costly).
  */
 static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force)
 {
 	long delta = cfs_rq->avg.load_avg - cfs_rq->tg_load_avg_contrib;
 
+	/*
+	 * No need to update load_avg for root_task_group as it is not used.
+	 */
+	if (cfs_rq->tg == &root_task_group)
+		return;
+
 	if (force || abs(delta) > cfs_rq->tg_load_avg_contrib / 64) {
 		atomic_long_add(delta, &cfs_rq->tg->load_avg);
 		cfs_rq->tg_load_avg_contrib = cfs_rq->avg.load_avg;
 	}
 }
 
+/*
+ * Called within set_task_rq() right before setting a task's cpu. The
+ * caller only guarantees p->pi_lock is held; no other assumptions,
+ * including the state of rq->lock, should be made.
+ */
+void set_task_rq_fair(struct sched_entity *se,
+		      struct cfs_rq *prev, struct cfs_rq *next)
+{
+	if (!sched_feat(ATTACH_AGE_LOAD))
+		return;
+
+	/*
+	 * We are supposed to update the task to "current" time, then its up to
+	 * date and ready to go to new CPU/cfs_rq. But we have difficulty in
+	 * getting what current time is, so simply throw away the out-of-date
+	 * time. This will result in the wakee task is less decayed, but giving
+	 * the wakee more load sounds not bad.
+	 */
+	if (se->avg.last_update_time && prev) {
+		u64 p_last_update_time;
+		u64 n_last_update_time;
+
+#ifndef CONFIG_64BIT
+		u64 p_last_update_time_copy;
+		u64 n_last_update_time_copy;
+
+		do {
+			p_last_update_time_copy = prev->load_last_update_time_copy;
+			n_last_update_time_copy = next->load_last_update_time_copy;
+
+			smp_rmb();
+
+			p_last_update_time = prev->avg.last_update_time;
+			n_last_update_time = next->avg.last_update_time;
+
+		} while (p_last_update_time != p_last_update_time_copy ||
+			 n_last_update_time != n_last_update_time_copy);
+#else
+		p_last_update_time = prev->avg.last_update_time;
+		n_last_update_time = next->avg.last_update_time;
+#endif
+		__update_load_avg(p_last_update_time, cpu_of(rq_of(prev)),
+				  &se->avg, 0, 0, NULL);
+		se->avg.last_update_time = n_last_update_time;
+	}
+}
+
+/* Take into account change of utilization of a child task group */
+static inline void
+update_tg_cfs_util(struct cfs_rq *cfs_rq, struct sched_entity *se)
+{
+	struct cfs_rq *gcfs_rq = group_cfs_rq(se);
+	long delta = gcfs_rq->avg.util_avg - se->avg.util_avg;
+
+	/* Nothing to update */
+	if (!delta)
+		return;
+
+	/* Set new sched_entity's utilization */
+	se->avg.util_avg = gcfs_rq->avg.util_avg;
+	se->avg.util_sum = se->avg.util_avg * LOAD_AVG_MAX;
+
+	/* Update parent cfs_rq utilization */
+	add_positive(&cfs_rq->avg.util_avg, delta);
+	cfs_rq->avg.util_sum = cfs_rq->avg.util_avg * LOAD_AVG_MAX;
+}
+
+/* Take into account change of load of a child task group */
+static inline void
+update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se)
+{
+	struct cfs_rq *gcfs_rq = group_cfs_rq(se);
+	long delta, load = gcfs_rq->avg.load_avg;
+
+	/*
+	 * If the load of group cfs_rq is null, the load of the
+	 * sched_entity will also be null so we can skip the formula
+	 */
+	if (load) {
+		long tg_load;
+
+		/* Get tg's load and ensure tg_load > 0 */
+		tg_load = atomic_long_read(&gcfs_rq->tg->load_avg) + 1;
+
+		/* Ensure tg_load >= load and updated with current load*/
+		tg_load -= gcfs_rq->tg_load_avg_contrib;
+		tg_load += load;
+
+		/*
+		 * We need to compute a correction term in the case that the
+		 * task group is consuming more CPU than a task of equal
+		 * weight. A task with a weight equals to tg->shares will have
+		 * a load less or equal to scale_load_down(tg->shares).
+		 * Similarly, the sched_entities that represent the task group
+		 * at parent level, can't have a load higher than
+		 * scale_load_down(tg->shares). And the Sum of sched_entities'
+		 * load must be <= scale_load_down(tg->shares).
+		 */
+		if (tg_load > scale_load_down(gcfs_rq->tg->shares)) {
+			/* scale gcfs_rq's load into tg's shares*/
+			load *= scale_load_down(gcfs_rq->tg->shares);
+			load /= tg_load;
+		}
+	}
+
+	delta = load - se->avg.load_avg;
+
+	/* Nothing to update */
+	if (!delta)
+		return;
+
+	/* Set new sched_entity's load */
+	se->avg.load_avg = load;
+	se->avg.load_sum = se->avg.load_avg * LOAD_AVG_MAX;
+
+	/* Update parent cfs_rq load */
+	add_positive(&cfs_rq->avg.load_avg, delta);
+	cfs_rq->avg.load_sum = cfs_rq->avg.load_avg * LOAD_AVG_MAX;
+
+	/*
+	 * If the sched_entity is already enqueued, we also have to update the
+	 * runnable load avg.
+	 */
+	if (se->on_rq) {
+		/* Update parent cfs_rq runnable_load_avg */
+		add_positive(&cfs_rq->runnable_load_avg, delta);
+		cfs_rq->runnable_load_sum = cfs_rq->runnable_load_avg * LOAD_AVG_MAX;
+	}
+}
+
+static inline void set_tg_cfs_propagate(struct cfs_rq *cfs_rq)
+{
+	cfs_rq->propagate_avg = 1;
+}
+
+static inline int test_and_clear_tg_cfs_propagate(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq = group_cfs_rq(se);
+
+	if (!cfs_rq->propagate_avg)
+		return 0;
+
+	cfs_rq->propagate_avg = 0;
+	return 1;
+}
+
+/* Update task and its cfs_rq load average */
+static inline int propagate_entity_load_avg(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq;
+
+	if (entity_is_task(se))
+		return 0;
+
+	if (!test_and_clear_tg_cfs_propagate(se))
+		return 0;
+
+	cfs_rq = cfs_rq_of(se);
+
+	set_tg_cfs_propagate(cfs_rq);
+
+	update_tg_cfs_util(cfs_rq, se);
+	update_tg_cfs_load(cfs_rq, se);
+
+	return 1;
+}
+
 #else /* CONFIG_FAIR_GROUP_SCHED */
+
 static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) {}
+
+static inline int propagate_entity_load_avg(struct sched_entity *se)
+{
+	return 0;
+}
+
+static inline void set_tg_cfs_propagate(struct cfs_rq *cfs_rq) {}
+
 #endif /* CONFIG_FAIR_GROUP_SCHED */
 
+static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq)
+{
+        if (&this_rq()->cfs == cfs_rq) {
+                /*
+                 * There are a few boundary cases this might miss but it should
+                 * get called often enough that that should (hopefully) not be
+                 * a real problem -- added to that it only calls on the local
+                 * CPU, so if we enqueue remotely we'll miss an update, but
+                 * the next tick/schedule should update.
+                 *
+                 * It will not get called when we go idle, because the idle
+                 * thread is a different class (!fair), nor will the utilization
+                 * number include things like RT tasks.
+                 *
+                 * As is, the util number is not freq-invariant (we'd have to
+                 * implement arch_scale_freq_capacity() for that).
+                 *
+                 * See cpu_util().
+                 */
+                cpufreq_update_util(rq_of(cfs_rq), 0);
+        }
+}
+
 static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq);
 
 /*
@@ -2713,23 +3107,43 @@ static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq);
 	WRITE_ONCE(*ptr, res);					\
 } while (0)
 
-/* Group cfs_rq's load_avg is used for task_h_load and update_cfs_share */
-static inline int update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq)
+/**
+ * update_cfs_rq_load_avg - update the cfs_rq's load/util averages
+ * @now: current time, as per cfs_rq_clock_task()
+ * @cfs_rq: cfs_rq to update
+ * @update_freq: should we call cfs_rq_util_change() or will the call do so
+ *
+ * The cfs_rq avg is the direct sum of all its entities (blocked and runnable)
+ * avg. The immediate corollary is that all (fair) tasks must be attached, see
+ * post_init_entity_util_avg().
+ *
+ * cfs_rq->avg is used for task_h_load() and update_cfs_share() for example.
+ *
+ * Returns true if the load decayed or we removed load.
+ *
+ * Since both these conditions indicate a changed cfs_rq->avg.load we should
+ * call update_tg_load_avg() when this function returns true.
+ */
+static inline int
+update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq, bool update_freq)
 {
 	struct sched_avg *sa = &cfs_rq->avg;
-	int decayed, removed = 0;
+	int decayed, removed = 0, removed_util = 0;
 
 	if (atomic_long_read(&cfs_rq->removed_load_avg)) {
 		s64 r = atomic_long_xchg(&cfs_rq->removed_load_avg, 0);
 		sub_positive(&sa->load_avg, r);
 		sub_positive(&sa->load_sum, r * LOAD_AVG_MAX);
 		removed = 1;
+		set_tg_cfs_propagate(cfs_rq);
 	}
 
 	if (atomic_long_read(&cfs_rq->removed_util_avg)) {
 		long r = atomic_long_xchg(&cfs_rq->removed_util_avg, 0);
 		sub_positive(&sa->util_avg, r);
 		sub_positive(&sa->util_sum, r * LOAD_AVG_MAX);
+		removed_util = 1;
+		set_tg_cfs_propagate(cfs_rq);
 	}
 
 	decayed = __update_load_avg(now, cpu_of(rq_of(cfs_rq)), sa,
@@ -2740,65 +3154,93 @@ static inline int update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq)
 	cfs_rq->load_last_update_time_copy = sa->last_update_time;
 #endif
 
+	/* Trace CPU load, unless cfs_rq belongs to a non-root task_group */
+	if (cfs_rq == &rq_of(cfs_rq)->cfs)
+		trace_sched_load_avg_cpu(cpu_of(rq_of(cfs_rq)), cfs_rq);
+
+	if (update_freq && (decayed || removed_util))
+		cfs_rq_util_change(cfs_rq);
+
 	return decayed || removed;
 }
 
+/*
+ * Optional action to be done while updating the load average
+ */
+#define UPDATE_TG	0x1
+#define SKIP_AGE_LOAD	0x2
+
 /* Update task and its cfs_rq load average */
-static inline void update_load_avg(struct sched_entity *se, int update_tg)
+static inline void update_load_avg(struct sched_entity *se, int flags)
 {
 	struct cfs_rq *cfs_rq = cfs_rq_of(se);
 	u64 now = cfs_rq_clock_task(cfs_rq);
 	int cpu = cpu_of(rq_of(cfs_rq));
+	int decayed;
+	void *ptr = NULL;
 
 	/*
 	 * Track task load average for carrying it to new CPU after migrated, and
 	 * track group sched_entity load average for task_h_load calc in migration
 	 */
-	__update_load_avg(now, cpu, &se->avg,
+	if (se->avg.last_update_time && !(flags & SKIP_AGE_LOAD)) {
+		__update_load_avg(now, cpu, &se->avg,
 			  se->on_rq * scale_load_down(se->load.weight),
 			  cfs_rq->curr == se, NULL);
+	}
 
-	if (update_cfs_rq_load_avg(now, cfs_rq) && update_tg)
+	decayed  = update_cfs_rq_load_avg(now, cfs_rq, true);
+	decayed |= propagate_entity_load_avg(se);
+
+	if (decayed && (flags & UPDATE_TG))
 		update_tg_load_avg(cfs_rq, 0);
+
+	if (entity_is_task(se)) {
+#ifdef CONFIG_SCHED_WALT
+		ptr = (void *)&(task_of(se)->ravg);
+#endif
+		trace_sched_load_avg_task(task_of(se), &se->avg, ptr);
+	}
 }
 
+/**
+ * attach_entity_load_avg - attach this entity to its cfs_rq load avg
+ * @cfs_rq: cfs_rq to attach to
+ * @se: sched_entity to attach
+ *
+ * Must call update_cfs_rq_load_avg() before this, since we rely on
+ * cfs_rq->avg.last_update_time being current.
+ */
 static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
-	if (!sched_feat(ATTACH_AGE_LOAD))
-		goto skip_aging;
-
-	/*
-	 * If we got migrated (either between CPUs or between cgroups) we'll
-	 * have aged the average right before clearing @last_update_time.
-	 */
-	if (se->avg.last_update_time) {
-		__update_load_avg(cfs_rq->avg.last_update_time, cpu_of(rq_of(cfs_rq)),
-				  &se->avg, 0, 0, NULL);
-
-		/*
-		 * XXX: we could have just aged the entire load away if we've been
-		 * absent from the fair class for too long.
-		 */
-	}
-
-skip_aging:
 	se->avg.last_update_time = cfs_rq->avg.last_update_time;
 	cfs_rq->avg.load_avg += se->avg.load_avg;
 	cfs_rq->avg.load_sum += se->avg.load_sum;
 	cfs_rq->avg.util_avg += se->avg.util_avg;
 	cfs_rq->avg.util_sum += se->avg.util_sum;
+	set_tg_cfs_propagate(cfs_rq);
+
+	cfs_rq_util_change(cfs_rq);
 }
 
+/**
+ * detach_entity_load_avg - detach this entity from its cfs_rq load avg
+ * @cfs_rq: cfs_rq to detach from
+ * @se: sched_entity to detach
+ *
+ * Must call update_cfs_rq_load_avg() before this, since we rely on
+ * cfs_rq->avg.last_update_time being current.
+ */
 static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
-	__update_load_avg(cfs_rq->avg.last_update_time, cpu_of(rq_of(cfs_rq)),
-			  &se->avg, se->on_rq * scale_load_down(se->load.weight),
-			  cfs_rq->curr == se, NULL);
 
 	sub_positive(&cfs_rq->avg.load_avg, se->avg.load_avg);
 	sub_positive(&cfs_rq->avg.load_sum, se->avg.load_sum);
 	sub_positive(&cfs_rq->avg.util_avg, se->avg.util_avg);
 	sub_positive(&cfs_rq->avg.util_sum, se->avg.util_sum);
+	set_tg_cfs_propagate(cfs_rq);
+
+	cfs_rq_util_change(cfs_rq);
 }
 
 /* Add the load generated by se into cfs_rq's load average */
@@ -2806,62 +3248,79 @@ static inline void
 enqueue_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
 	struct sched_avg *sa = &se->avg;
-	u64 now = cfs_rq_clock_task(cfs_rq);
-	int migrated, decayed;
-
-	migrated = !sa->last_update_time;
-	if (!migrated) {
-		__update_load_avg(now, cpu_of(rq_of(cfs_rq)), sa,
-			se->on_rq * scale_load_down(se->load.weight),
-			cfs_rq->curr == se, NULL);
-	}
-
-	decayed = update_cfs_rq_load_avg(now, cfs_rq);
 
 	cfs_rq->runnable_load_avg += sa->load_avg;
 	cfs_rq->runnable_load_sum += sa->load_sum;
 
-	if (migrated)
+	if (!sa->last_update_time) {
 		attach_entity_load_avg(cfs_rq, se);
-
-	if (decayed || migrated)
 		update_tg_load_avg(cfs_rq, 0);
+	}
 }
 
 /* Remove the runnable load generated by se from cfs_rq's runnable load average */
 static inline void
 dequeue_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
 {
-	update_load_avg(se, 1);
-
 	cfs_rq->runnable_load_avg =
 		max_t(long, cfs_rq->runnable_load_avg - se->avg.load_avg, 0);
 	cfs_rq->runnable_load_sum =
 		max_t(s64,  cfs_rq->runnable_load_sum - se->avg.load_sum, 0);
 }
 
-/*
- * Task first catches up with cfs_rq, and then subtract
- * itself from the cfs_rq (task must be off the queue now).
- */
-void remove_entity_load_avg(struct sched_entity *se)
-{
-	struct cfs_rq *cfs_rq = cfs_rq_of(se);
-	u64 last_update_time;
-
 #ifndef CONFIG_64BIT
+static inline u64 cfs_rq_last_update_time(struct cfs_rq *cfs_rq)
+{
 	u64 last_update_time_copy;
+	u64 last_update_time;
 
 	do {
 		last_update_time_copy = cfs_rq->load_last_update_time_copy;
 		smp_rmb();
 		last_update_time = cfs_rq->avg.last_update_time;
 	} while (last_update_time != last_update_time_copy);
+
+	return last_update_time;
+}
 #else
-	last_update_time = cfs_rq->avg.last_update_time;
+static inline u64 cfs_rq_last_update_time(struct cfs_rq *cfs_rq)
+{
+	return cfs_rq->avg.last_update_time;
+}
 #endif
 
+/*
+ * Synchronize entity load avg of dequeued entity without locking
+ * the previous rq.
+ */
+void sync_entity_load_avg(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+	u64 last_update_time;
+
+	last_update_time = cfs_rq_last_update_time(cfs_rq);
 	__update_load_avg(last_update_time, cpu_of(rq_of(cfs_rq)), &se->avg, 0, 0, NULL);
+}
+
+/*
+ * Task first catches up with cfs_rq, and then subtract
+ * itself from the cfs_rq (task must be off the queue now).
+ */
+void remove_entity_load_avg(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+	/*
+	 * tasks cannot exit without having gone through wake_up_new_task() ->
+	 * post_init_entity_util_avg() which will have added things to the
+	 * cfs_rq, so we can remove unconditionally.
+	 *
+	 * Similarly for groups, they will have passed through
+	 * post_init_entity_util_avg() before unregister_sched_fair_group()
+	 * calls this.
+	 */
+
+	sync_entity_load_avg(se);
 	atomic_long_add(se->avg.load_avg, &cfs_rq->removed_load_avg);
 	atomic_long_add(se->avg.util_avg, &cfs_rq->removed_util_avg);
 }
@@ -2898,7 +3357,16 @@ static int idle_balance(struct rq *this_rq);
 
 #else /* CONFIG_SMP */
 
-static inline void update_load_avg(struct sched_entity *se, int update_tg) {}
+static inline int
+update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq, bool update_freq)
+{
+	return 0;
+}
+
+#define UPDATE_TG	0x0
+#define SKIP_AGE_LOAD	0x0
+
+static inline void update_load_avg(struct sched_entity *se, int not_used1){}
 static inline void
 enqueue_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) {}
 static inline void
@@ -2962,6 +3430,7 @@ static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
 			}
 
 			trace_sched_stat_blocked(tsk, delta);
+			trace_sched_blocked_reason(tsk);
 
 			/*
 			 * Blocking time is in units of nanosecs, so shift by
@@ -3040,9 +3509,10 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
 	 * Update run-time statistics of the 'current'.
 	 */
 	update_curr(cfs_rq);
+	update_load_avg(se, UPDATE_TG);
 	enqueue_entity_load_avg(cfs_rq, se);
+	update_cfs_shares(se);
 	account_entity_enqueue(cfs_rq, se);
-	update_cfs_shares(cfs_rq);
 
 	if (flags & ENQUEUE_WAKEUP) {
 		place_entity(cfs_rq, se, 0);
@@ -3115,6 +3585,16 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
 	 * Update run-time statistics of the 'current'.
 	 */
 	update_curr(cfs_rq);
+
+	/*
+	 * When dequeuing a sched_entity, we must:
+	 *   - Update loads to have both entity and cfs_rq synced with now.
+	 *   - Substract its load from the cfs_rq->runnable_avg.
+	 *   - Substract its previous weight from cfs_rq->load.weight.
+	 *   - For group entity, update its weight to reflect the new share
+	 *     of its group cfs_rq.
+	 */
+	update_load_avg(se, UPDATE_TG);
 	dequeue_entity_load_avg(cfs_rq, se);
 
 	update_stats_dequeue(cfs_rq, se);
@@ -3150,7 +3630,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags)
 	return_cfs_rq_runtime(cfs_rq);
 
 	update_min_vruntime(cfs_rq);
-	update_cfs_shares(cfs_rq);
+	update_cfs_shares(se);
 }
 
 /*
@@ -3205,7 +3685,7 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se)
 		 */
 		update_stats_wait_end(cfs_rq, se);
 		__dequeue_entity(cfs_rq, se);
-		update_load_avg(se, 1);
+		update_load_avg(se, UPDATE_TG);
 	}
 
 	update_stats_curr_start(cfs_rq, se);
@@ -3321,8 +3801,8 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued)
 	/*
 	 * Ensure that runnable average is periodically updated.
 	 */
-	update_load_avg(curr, 1);
-	update_cfs_shares(cfs_rq);
+	update_load_avg(curr, UPDATE_TG);
+	update_cfs_shares(curr);
 
 #ifdef CONFIG_SCHED_HRTICK
 	/*
@@ -4177,6 +4657,14 @@ static inline void hrtick_update(struct rq *rq)
 }
 #endif
 
+#ifdef CONFIG_SMP
+static bool __cpu_overutilized(int cpu, int delta);
+static bool cpu_overutilized(int cpu);
+unsigned long boosted_cpu_util(int cpu);
+#else
+#define boosted_cpu_util(cpu) cpu_util_freq(cpu)
+#endif
+
 /*
  * The enqueue_task method is called before nr_running is
  * increased. Here we update the fair scheduling stats and
@@ -4187,6 +4675,17 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 {
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se = &p->se;
+#ifdef CONFIG_SMP
+	int task_new = flags & ENQUEUE_WAKEUP_NEW;
+#endif
+
+	/*
+	 * If in_iowait is set, the code below may not trigger any cpufreq
+	 * utilization updates, so do it here explicitly with the IOWAIT flag
+	 * passed.
+	 */
+	if (p->in_iowait)
+		cpufreq_update_this_cpu(rq, SCHED_CPUFREQ_IOWAIT);
 
 	for_each_sched_entity(se) {
 		if (se->on_rq)
@@ -4199,10 +4698,11 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 		 *
 		 * note: in the case of encountering a throttled cfs_rq we will
 		 * post the final h_nr_running increment below.
-		*/
+		 */
 		if (cfs_rq_throttled(cfs_rq))
 			break;
 		cfs_rq->h_nr_running++;
+		walt_inc_cfs_cumulative_runnable_avg(cfs_rq, p);
 
 		flags = ENQUEUE_WAKEUP;
 	}
@@ -4210,17 +4710,49 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 	for_each_sched_entity(se) {
 		cfs_rq = cfs_rq_of(se);
 		cfs_rq->h_nr_running++;
+		walt_inc_cfs_cumulative_runnable_avg(cfs_rq, p);
 
 		if (cfs_rq_throttled(cfs_rq))
 			break;
 
-		update_load_avg(se, 1);
-		update_cfs_shares(cfs_rq);
+		update_load_avg(se, UPDATE_TG);
+		update_cfs_shares(se);
 	}
 
 	if (!se)
 		add_nr_running(rq, 1);
 
+#ifdef CONFIG_SMP
+
+	/*
+	 * Update SchedTune accounting.
+	 *
+	 * We do it before updating the CPU capacity to ensure the
+	 * boost value of the current task is accounted for in the
+	 * selection of the OPP.
+	 *
+	 * We do it also in the case where we enqueue a throttled task;
+	 * we could argue that a throttled task should not boost a CPU,
+	 * however:
+	 * a) properly implementing CPU boosting considering throttled
+	 *    tasks will increase a lot the complexity of the solution
+	 * b) it's not easy to quantify the benefits introduced by
+	 *    such a more complex solution.
+	 * Thus, for the time being we go for the simple solution and boost
+	 * also for throttled RQs.
+	 */
+	schedtune_enqueue_task(p, cpu_of(rq));
+
+	if (!se) {
+		walt_inc_cumulative_runnable_avg(rq, p);
+		if (!task_new && !rq->rd->overutilized &&
+		    cpu_overutilized(rq->cpu)) {
+			rq->rd->overutilized = true;
+			trace_sched_overutilized(true);
+		}
+	}
+
+#endif /* CONFIG_SMP */
 	hrtick_update(rq);
 }
 
@@ -4250,6 +4782,7 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 		if (cfs_rq_throttled(cfs_rq))
 			break;
 		cfs_rq->h_nr_running--;
+		walt_dec_cfs_cumulative_runnable_avg(cfs_rq, p);
 
 		/* Don't dequeue parent if it has other entities besides us */
 		if (cfs_rq->load.weight) {
@@ -4269,17 +4802,33 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 	for_each_sched_entity(se) {
 		cfs_rq = cfs_rq_of(se);
 		cfs_rq->h_nr_running--;
+		walt_dec_cfs_cumulative_runnable_avg(cfs_rq, p);
 
 		if (cfs_rq_throttled(cfs_rq))
 			break;
 
-		update_load_avg(se, 1);
-		update_cfs_shares(cfs_rq);
+		update_load_avg(se, UPDATE_TG);
+		update_cfs_shares(se);
 	}
 
 	if (!se)
 		sub_nr_running(rq, 1);
 
+#ifdef CONFIG_SMP
+
+	/*
+	 * Update SchedTune accounting
+	 *
+	 * We do it before updating the CPU capacity to ensure the
+	 * boost value of the current task is accounted for in the
+	 * selection of the OPP.
+	 */
+	schedtune_dequeue_task(p, cpu_of(rq));
+
+	if (!se)
+		walt_dec_cumulative_runnable_avg(rq, p);
+#endif /* CONFIG_SMP */
+
 	hrtick_update(rq);
 }
 
@@ -4506,15 +5055,6 @@ static unsigned long target_load(int cpu, int type)
 	return max(rq->cpu_load[type-1], total);
 }
 
-static unsigned long capacity_of(int cpu)
-{
-	return cpu_rq(cpu)->cpu_capacity;
-}
-
-static unsigned long capacity_orig_of(int cpu)
-{
-	return cpu_rq(cpu)->cpu_capacity_orig;
-}
 
 static unsigned long cpu_avg_load_per_task(int cpu)
 {
@@ -4689,6 +5229,492 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg)
 #endif
 
 /*
+ * Returns the current capacity of cpu after applying both
+ * cpu and freq scaling.
+ */
+unsigned long capacity_curr_of(int cpu)
+{
+	return cpu_rq(cpu)->cpu_capacity_orig *
+	       arch_scale_freq_capacity(NULL, cpu)
+	       >> SCHED_CAPACITY_SHIFT;
+}
+
+static inline bool energy_aware(void)
+{
+	return sched_feat(ENERGY_AWARE);
+}
+
+struct energy_env {
+	struct sched_group	*sg_top;
+	struct sched_group	*sg_cap;
+	int			cap_idx;
+	int			util_delta;
+	int			src_cpu;
+	int			dst_cpu;
+	int			trg_cpu;
+	int			energy;
+	int			payoff;
+	struct task_struct	*task;
+	struct {
+		int before;
+		int after;
+		int delta;
+		int diff;
+	} nrg;
+	struct {
+		int before;
+		int after;
+		int delta;
+	} cap;
+};
+
+static int cpu_util_wake(int cpu, struct task_struct *p);
+
+/*
+ * __cpu_norm_util() returns the cpu util relative to a specific capacity,
+ * i.e. it's busy ratio, in the range [0..SCHED_LOAD_SCALE], which is useful for
+ * energy calculations.
+ *
+ * Since util is a scale-invariant utilization defined as:
+ *
+ *   util ~ (curr_freq/max_freq)*1024 * capacity_orig/1024 * running_time/time
+ *
+ * the normalized util can be found using the specific capacity.
+ *
+ *   capacity = capacity_orig * curr_freq/max_freq
+ *
+ *   norm_util = running_time/time ~ util/capacity
+ */
+static unsigned long __cpu_norm_util(unsigned long util, unsigned long capacity)
+{
+	if (util >= capacity)
+		return SCHED_CAPACITY_SCALE;
+
+	return (util << SCHED_CAPACITY_SHIFT)/capacity;
+}
+
+static unsigned long group_max_util(struct energy_env *eenv)
+{
+	unsigned long max_util = 0;
+	unsigned long util;
+	int cpu;
+
+	for_each_cpu(cpu, sched_group_cpus(eenv->sg_cap)) {
+		util = cpu_util_wake(cpu, eenv->task);
+
+		/*
+		 * If we are looking at the target CPU specified by the eenv,
+		 * then we should add the (estimated) utilization of the task
+		 * assuming we will wake it up on that CPU.
+		 */
+		if (unlikely(cpu == eenv->trg_cpu))
+			util += eenv->util_delta;
+
+		max_util = max(max_util, util);
+	}
+
+	return max_util;
+}
+
+/*
+ * group_norm_util() returns the approximated group util relative to it's
+ * current capacity (busy ratio), in the range [0..SCHED_LOAD_SCALE], for use
+ * in energy calculations.
+ *
+ * Since task executions may or may not overlap in time in the group the true
+ * normalized util is between MAX(cpu_norm_util(i)) and SUM(cpu_norm_util(i))
+ * when iterating over all CPUs in the group.
+ * The latter estimate is used as it leads to a more pessimistic energy
+ * estimate (more busy).
+ */
+static unsigned
+long group_norm_util(struct energy_env *eenv, struct sched_group *sg)
+{
+	unsigned long capacity = sg->sge->cap_states[eenv->cap_idx].cap;
+	unsigned long util, util_sum = 0;
+	int cpu;
+
+	for_each_cpu(cpu, sched_group_cpus(sg)) {
+		util = cpu_util_wake(cpu, eenv->task);
+
+		/*
+		 * If we are looking at the target CPU specified by the eenv,
+		 * then we should add the (estimated) utilization of the task
+		 * assuming we will wake it up on that CPU.
+		 */
+		if (unlikely(cpu == eenv->trg_cpu))
+			util += eenv->util_delta;
+
+		util_sum += __cpu_norm_util(util, capacity);
+	}
+
+	return min_t(unsigned long, util_sum, SCHED_CAPACITY_SCALE);
+}
+
+static int find_new_capacity(struct energy_env *eenv,
+	const struct sched_group_energy * const sge)
+{
+	int idx, max_idx = sge->nr_cap_states - 1;
+	unsigned long util = group_max_util(eenv);
+
+	/* default is max_cap if we don't find a match */
+	eenv->cap_idx = max_idx;
+
+	for (idx = 0; idx < sge->nr_cap_states; idx++) {
+		if (sge->cap_states[idx].cap >= util) {
+			eenv->cap_idx = idx;
+			break;
+		}
+	}
+
+	return eenv->cap_idx;
+}
+
+static int group_idle_state(struct energy_env *eenv, struct sched_group *sg)
+{
+	int i, state = INT_MAX;
+	int src_in_grp, dst_in_grp;
+	long grp_util = 0;
+
+	/* Find the shallowest idle state in the sched group. */
+	for_each_cpu(i, sched_group_cpus(sg))
+		state = min(state, idle_get_state_idx(cpu_rq(i)));
+
+	/* Take non-cpuidle idling into account (active idle/arch_cpu_idle()) */
+	state++;
+
+	src_in_grp = cpumask_test_cpu(eenv->src_cpu, sched_group_cpus(sg));
+	dst_in_grp = cpumask_test_cpu(eenv->dst_cpu, sched_group_cpus(sg));
+	if (src_in_grp == dst_in_grp) {
+		/* both CPUs under consideration are in the same group or not in
+		 * either group, migration should leave idle state the same.
+		 */
+		goto end;
+	}
+
+	/*
+	 * Try to estimate if a deeper idle state is
+	 * achievable when we move the task.
+	 */
+	for_each_cpu(i, sched_group_cpus(sg)) {
+		grp_util += cpu_util_wake(i, eenv->task);
+		if (unlikely(i == eenv->trg_cpu))
+			grp_util += eenv->util_delta;
+	}
+
+	if (grp_util <=
+		((long)sg->sgc->max_capacity * (int)sg->group_weight)) {
+		/* after moving, this group is at most partly
+		 * occupied, so it should have some idle time.
+		 */
+		int max_idle_state_idx = sg->sge->nr_idle_states - 2;
+		int new_state = grp_util * max_idle_state_idx;
+		if (grp_util <= 0)
+			/* group will have no util, use lowest state */
+			new_state = max_idle_state_idx + 1;
+		else {
+			/* for partially idle, linearly map util to idle
+			 * states, excluding the lowest one. This does not
+			 * correspond to the state we expect to enter in
+			 * reality, but an indication of what might happen.
+			 */
+			new_state = min(max_idle_state_idx, (int)
+					(new_state / sg->sgc->max_capacity));
+			new_state = max_idle_state_idx - new_state;
+		}
+		state = new_state;
+	} else {
+		/* After moving, the group will be fully occupied
+		 * so assume it will not be idle at all.
+		 */
+		state = 0;
+	}
+end:
+	return state;
+}
+
+/*
+ * sched_group_energy(): Computes the absolute energy consumption of cpus
+ * belonging to the sched_group including shared resources shared only by
+ * members of the group. Iterates over all cpus in the hierarchy below the
+ * sched_group starting from the bottom working it's way up before going to
+ * the next cpu until all cpus are covered at all levels. The current
+ * implementation is likely to gather the same util statistics multiple times.
+ * This can probably be done in a faster but more complex way.
+ * Note: sched_group_energy() may fail when racing with sched_domain updates.
+ */
+static int sched_group_energy(struct energy_env *eenv)
+{
+	struct cpumask visit_cpus;
+	u64 total_energy = 0;
+
+	WARN_ON(!eenv->sg_top->sge);
+
+	cpumask_copy(&visit_cpus, sched_group_cpus(eenv->sg_top));
+
+	while (!cpumask_empty(&visit_cpus)) {
+		struct sched_group *sg_shared_cap = NULL;
+		int cpu = cpumask_first(&visit_cpus);
+		struct sched_domain *sd;
+
+		/*
+		 * Is the group utilization affected by cpus outside this
+		 * sched_group?
+		 */
+		sd = rcu_dereference(per_cpu(sd_scs, cpu));
+
+		if (sd && sd->parent)
+			sg_shared_cap = sd->parent->groups;
+
+		for_each_domain(cpu, sd) {
+			struct sched_group *sg = sd->groups;
+
+			/* Has this sched_domain already been visited? */
+			if (sd->child && group_first_cpu(sg) != cpu)
+				break;
+
+			do {
+				unsigned long group_util;
+				int sg_busy_energy, sg_idle_energy;
+				int cap_idx, idle_idx;
+
+				if (sg_shared_cap && sg_shared_cap->group_weight >= sg->group_weight)
+					eenv->sg_cap = sg_shared_cap;
+				else
+					eenv->sg_cap = sg;
+
+				cap_idx = find_new_capacity(eenv, sg->sge);
+
+				if (sg->group_weight == 1) {
+					/* Remove capacity of src CPU (before task move) */
+					if (eenv->trg_cpu == eenv->src_cpu &&
+					    cpumask_test_cpu(eenv->src_cpu, sched_group_cpus(sg))) {
+						eenv->cap.before = sg->sge->cap_states[cap_idx].cap;
+						eenv->cap.delta -= eenv->cap.before;
+					}
+					/* Add capacity of dst CPU  (after task move) */
+					if (eenv->trg_cpu == eenv->dst_cpu &&
+					    cpumask_test_cpu(eenv->dst_cpu, sched_group_cpus(sg))) {
+						eenv->cap.after = sg->sge->cap_states[cap_idx].cap;
+						eenv->cap.delta += eenv->cap.after;
+					}
+				}
+
+				idle_idx = group_idle_state(eenv, sg);
+				group_util = group_norm_util(eenv, sg);
+
+				sg_busy_energy = (group_util * sg->sge->cap_states[cap_idx].power);
+				sg_idle_energy = ((SCHED_LOAD_SCALE-group_util)
+								* sg->sge->idle_states[idle_idx].power);
+
+				total_energy += sg_busy_energy + sg_idle_energy;
+
+				if (!sd->child)
+					cpumask_xor(&visit_cpus, &visit_cpus, sched_group_cpus(sg));
+
+				if (cpumask_equal(sched_group_cpus(sg), sched_group_cpus(eenv->sg_top)))
+					goto next_cpu;
+
+			} while (sg = sg->next, sg != sd->groups);
+		}
+
+		/*
+		 * If we raced with hotplug and got an sd NULL-pointer;
+		 * returning a wrong energy estimation is better than
+		 * entering an infinite loop.
+		 */
+		if (cpumask_test_cpu(cpu, &visit_cpus))
+			return -EINVAL;
+next_cpu:
+		cpumask_clear_cpu(cpu, &visit_cpus);
+		continue;
+	}
+
+	eenv->energy = total_energy >> SCHED_CAPACITY_SHIFT;
+	return 0;
+}
+
+static inline bool cpu_in_sg(struct sched_group *sg, int cpu)
+{
+	return cpu != -1 && cpumask_test_cpu(cpu, sched_group_cpus(sg));
+}
+
+static inline unsigned long task_util(struct task_struct *p);
+
+/*
+ * energy_diff(): Estimate the energy impact of changing the utilization
+ * distribution. eenv specifies the change: utilisation amount, source, and
+ * destination cpu. Source or destination cpu may be -1 in which case the
+ * utilization is removed from or added to the system (e.g. task wake-up). If
+ * both are specified, the utilization is migrated.
+ */
+static inline int __energy_diff(struct energy_env *eenv)
+{
+	struct sched_domain *sd;
+	struct sched_group *sg;
+	int sd_cpu = -1, energy_before = 0, energy_after = 0;
+	int diff, margin;
+
+	struct energy_env eenv_before = {
+		.util_delta	= task_util(eenv->task),
+		.src_cpu	= eenv->src_cpu,
+		.dst_cpu	= eenv->dst_cpu,
+		.trg_cpu	= eenv->src_cpu,
+		.nrg		= { 0, 0, 0, 0},
+		.cap		= { 0, 0, 0 },
+		.task		= eenv->task,
+	};
+
+	if (eenv->src_cpu == eenv->dst_cpu)
+		return 0;
+
+	sd_cpu = (eenv->src_cpu != -1) ? eenv->src_cpu : eenv->dst_cpu;
+	sd = rcu_dereference(per_cpu(sd_ea, sd_cpu));
+
+	if (!sd)
+		return 0; /* Error */
+
+	sg = sd->groups;
+
+	do {
+		if (cpu_in_sg(sg, eenv->src_cpu) || cpu_in_sg(sg, eenv->dst_cpu)) {
+			eenv_before.sg_top = eenv->sg_top = sg;
+
+			if (sched_group_energy(&eenv_before))
+				return 0; /* Invalid result abort */
+			energy_before += eenv_before.energy;
+
+			/* Keep track of SRC cpu (before) capacity */
+			eenv->cap.before = eenv_before.cap.before;
+			eenv->cap.delta = eenv_before.cap.delta;
+
+			if (sched_group_energy(eenv))
+				return 0; /* Invalid result abort */
+			energy_after += eenv->energy;
+		}
+	} while (sg = sg->next, sg != sd->groups);
+
+	eenv->nrg.before = energy_before;
+	eenv->nrg.after = energy_after;
+	eenv->nrg.diff = eenv->nrg.after - eenv->nrg.before;
+	eenv->payoff = 0;
+#ifndef CONFIG_SCHED_TUNE
+	trace_sched_energy_diff(eenv->task,
+			eenv->src_cpu, eenv->dst_cpu, eenv->util_delta,
+			eenv->nrg.before, eenv->nrg.after, eenv->nrg.diff,
+			eenv->cap.before, eenv->cap.after, eenv->cap.delta,
+			eenv->nrg.delta, eenv->payoff);
+#endif
+	/*
+	 * Dead-zone margin preventing too many migrations.
+	 */
+
+	margin = eenv->nrg.before >> 6; /* ~1.56% */
+
+	diff = eenv->nrg.after - eenv->nrg.before;
+
+	eenv->nrg.diff = (abs(diff) < margin) ? 0 : eenv->nrg.diff;
+
+	return eenv->nrg.diff;
+}
+
+#ifdef CONFIG_SCHED_TUNE
+
+struct target_nrg schedtune_target_nrg;
+
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+extern bool schedtune_initialized;
+#endif /* CONFIG_CGROUP_SCHEDTUNE */
+
+/*
+ * System energy normalization
+ * Returns the normalized value, in the range [0..SCHED_CAPACITY_SCALE],
+ * corresponding to the specified energy variation.
+ */
+static inline int
+normalize_energy(int energy_diff)
+{
+	u32 normalized_nrg;
+
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+	/* during early setup, we don't know the extents */
+	if (unlikely(!schedtune_initialized))
+		return energy_diff < 0 ? -1 : 1 ;
+#endif /* CONFIG_CGROUP_SCHEDTUNE */
+
+#ifdef CONFIG_SCHED_DEBUG
+	{
+	int max_delta;
+
+	/* Check for boundaries */
+	max_delta  = schedtune_target_nrg.max_power;
+	max_delta -= schedtune_target_nrg.min_power;
+	WARN_ON(abs(energy_diff) >= max_delta);
+	}
+#endif
+
+	/* Do scaling using positive numbers to increase the range */
+	normalized_nrg = (energy_diff < 0) ? -energy_diff : energy_diff;
+
+	/* Scale by energy magnitude */
+	normalized_nrg <<= SCHED_CAPACITY_SHIFT;
+
+	/* Normalize on max energy for target platform */
+	normalized_nrg = reciprocal_divide(
+			normalized_nrg, schedtune_target_nrg.rdiv);
+
+	return (energy_diff < 0) ? -normalized_nrg : normalized_nrg;
+}
+
+static inline int
+energy_diff(struct energy_env *eenv)
+{
+	int boost = schedtune_task_boost(eenv->task);
+	int nrg_delta;
+
+	/* Conpute "absolute" energy diff */
+	__energy_diff(eenv);
+
+	/* Return energy diff when boost margin is 0 */
+	if (boost == 0) {
+		trace_sched_energy_diff(eenv->task,
+				eenv->src_cpu, eenv->dst_cpu, eenv->util_delta,
+				eenv->nrg.before, eenv->nrg.after, eenv->nrg.diff,
+				eenv->cap.before, eenv->cap.after, eenv->cap.delta,
+				0, -eenv->nrg.diff);
+		return eenv->nrg.diff;
+	}
+
+	/* Compute normalized energy diff */
+	nrg_delta = normalize_energy(eenv->nrg.diff);
+	eenv->nrg.delta = nrg_delta;
+
+	eenv->payoff = schedtune_accept_deltas(
+			eenv->nrg.delta,
+			eenv->cap.delta,
+			eenv->task);
+
+	trace_sched_energy_diff(eenv->task,
+			eenv->src_cpu, eenv->dst_cpu, eenv->util_delta,
+			eenv->nrg.before, eenv->nrg.after, eenv->nrg.diff,
+			eenv->cap.before, eenv->cap.after, eenv->cap.delta,
+			eenv->nrg.delta, eenv->payoff);
+
+	/*
+	 * When SchedTune is enabled, the energy_diff() function will return
+	 * the computed energy payoff value. Since the energy_diff() return
+	 * value is expected to be negative by its callers, this evaluation
+	 * function return a negative value each time the evaluation return a
+	 * positive payoff, which is the condition for the acceptance of
+	 * a scheduling decision
+	 */
+	return -eenv->payoff;
+}
+#else /* CONFIG_SCHED_TUNE */
+#define energy_diff(eenv) __energy_diff(eenv)
+#endif
+
+/*
  * Detect M:N waker/wakee relationships via a switching-frequency heuristic.
  * A waker of many should wake a different task than the one last awakened
  * at a frequency roughly N times higher than one of its wakees.  In order
@@ -4700,31 +5726,34 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg)
  * being client/server, worker/dispatcher, interrupt source or whatever is
  * irrelevant, spread criteria is apparent partner count exceeds socket size.
  */
-static int wake_wide(struct task_struct *p)
+static int wake_wide(struct task_struct *p, int sibling_count_hint)
 {
 	unsigned int master = current->wakee_flips;
 	unsigned int slave = p->wakee_flips;
-	int factor = this_cpu_read(sd_llc_size);
+	int llc_size = this_cpu_read(sd_llc_size);
+
+	if (sibling_count_hint >= llc_size)
+		return 1;
 
 	if (master < slave)
 		swap(master, slave);
-	if (slave < factor || master < slave * factor)
+	if (slave < llc_size || master < slave * llc_size)
 		return 0;
 	return 1;
 }
 
-static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
+static int wake_affine(struct sched_domain *sd, struct task_struct *p,
+		       int prev_cpu, int sync)
 {
 	s64 this_load, load;
 	s64 this_eff_load, prev_eff_load;
-	int idx, this_cpu, prev_cpu;
+	int idx, this_cpu;
 	struct task_group *tg;
 	unsigned long weight;
 	int balanced;
 
 	idx	  = sd->wake_idx;
 	this_cpu  = smp_processor_id();
-	prev_cpu  = task_cpu(p);
 	load	  = source_load(prev_cpu, idx);
 	this_load = target_load(this_cpu, idx);
 
@@ -4779,16 +5808,166 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
 	return 1;
 }
 
+static inline unsigned long task_util(struct task_struct *p)
+{
+#ifdef CONFIG_SCHED_WALT
+	if (!walt_disabled && sysctl_sched_use_walt_task_util) {
+		unsigned long demand = p->ravg.demand;
+		return (demand << 10) / walt_ravg_window;
+	}
+#endif
+	return p->se.avg.util_avg;
+}
+
+static inline unsigned long boosted_task_util(struct task_struct *task);
+
+static inline bool __task_fits(struct task_struct *p, int cpu, int util)
+{
+	unsigned long capacity = capacity_of(cpu);
+
+	util += boosted_task_util(p);
+
+	return (capacity * 1024) > (util * capacity_margin);
+}
+
+static inline bool task_fits_max(struct task_struct *p, int cpu)
+{
+	unsigned long capacity = capacity_of(cpu);
+	unsigned long max_capacity = cpu_rq(cpu)->rd->max_cpu_capacity.val;
+
+	if (capacity == max_capacity)
+		return true;
+
+	if (capacity * capacity_margin > max_capacity * 1024)
+		return true;
+
+	return __task_fits(p, cpu, 0);
+}
+
+static bool __cpu_overutilized(int cpu, int delta)
+{
+	return (capacity_of(cpu) * 1024) < ((cpu_util(cpu) + delta) * capacity_margin);
+}
+
+static bool cpu_overutilized(int cpu)
+{
+	return __cpu_overutilized(cpu, 0);
+}
+
+#ifdef CONFIG_SCHED_TUNE
+
+struct reciprocal_value schedtune_spc_rdiv;
+
+static long
+schedtune_margin(unsigned long signal, long boost)
+{
+	long long margin = 0;
+
+	/*
+	 * Signal proportional compensation (SPC)
+	 *
+	 * The Boost (B) value is used to compute a Margin (M) which is
+	 * proportional to the complement of the original Signal (S):
+	 *   M = B * (SCHED_CAPACITY_SCALE - S)
+	 * The obtained M could be used by the caller to "boost" S.
+	 */
+	if (boost >= 0) {
+		margin  = SCHED_CAPACITY_SCALE - signal;
+		margin *= boost;
+	} else
+		margin = -signal * boost;
+
+	margin  = reciprocal_divide(margin, schedtune_spc_rdiv);
+
+	if (boost < 0)
+		margin *= -1;
+	return margin;
+}
+
+static inline int
+schedtune_cpu_margin(unsigned long util, int cpu)
+{
+	int boost = schedtune_cpu_boost(cpu);
+
+	if (boost == 0)
+		return 0;
+
+	return schedtune_margin(util, boost);
+}
+
+static inline long
+schedtune_task_margin(struct task_struct *task)
+{
+	int boost = schedtune_task_boost(task);
+	unsigned long util;
+	long margin;
+
+	if (boost == 0)
+		return 0;
+
+	util = task_util(task);
+	margin = schedtune_margin(util, boost);
+
+	return margin;
+}
+
+#else /* CONFIG_SCHED_TUNE */
+
+static inline int
+schedtune_cpu_margin(unsigned long util, int cpu)
+{
+	return 0;
+}
+
+static inline int
+schedtune_task_margin(struct task_struct *task)
+{
+	return 0;
+}
+
+#endif /* CONFIG_SCHED_TUNE */
+
+unsigned long
+boosted_cpu_util(int cpu)
+{
+	unsigned long util = cpu_util_freq(cpu);
+	long margin = schedtune_cpu_margin(util, cpu);
+
+	trace_sched_boost_cpu(cpu, util, margin);
+
+	return util + margin;
+}
+
+static inline unsigned long
+boosted_task_util(struct task_struct *task)
+{
+	unsigned long util = task_util(task);
+	long margin = schedtune_task_margin(task);
+
+	trace_sched_boost_task(task, util, margin);
+
+	return util + margin;
+}
+
+static unsigned long capacity_spare_wake(int cpu, struct task_struct *p)
+{
+	return capacity_orig_of(cpu) - cpu_util_wake(cpu, p);
+}
+
 /*
  * find_idlest_group finds and returns the least busy CPU group within the
  * domain.
+ *
+ * Assumes p is allowed on at least one CPU in sd.
  */
 static struct sched_group *
 find_idlest_group(struct sched_domain *sd, struct task_struct *p,
 		  int this_cpu, int sd_flag)
 {
 	struct sched_group *idlest = NULL, *group = sd->groups;
-	unsigned long min_load = ULONG_MAX, this_load = 0;
+	struct sched_group *most_spare_sg = NULL;
+	unsigned long min_load = ULONG_MAX, this_load = ULONG_MAX;
+	unsigned long most_spare = 0, this_spare = 0;
 	int load_idx = sd->forkexec_idx;
 	int imbalance = 100 + (sd->imbalance_pct-100)/2;
 
@@ -4796,7 +5975,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p,
 		load_idx = sd->wake_idx;
 
 	do {
-		unsigned long load, avg_load;
+		unsigned long load, avg_load, spare_cap, max_spare_cap;
 		int local_group;
 		int i;
 
@@ -4808,8 +5987,12 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p,
 		local_group = cpumask_test_cpu(this_cpu,
 					       sched_group_cpus(group));
 
-		/* Tally up the load of all CPUs in the group */
+		/*
+		 * Tally up the load of all CPUs in the group and find
+		 * the group containing the CPU with most spare capacity.
+		 */
 		avg_load = 0;
+		max_spare_cap = 0;
 
 		for_each_cpu(i, sched_group_cpus(group)) {
 			/* Bias balancing toward cpus of our domain */
@@ -4819,6 +6002,11 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p,
 				load = target_load(i, load_idx);
 
 			avg_load += load;
+
+			spare_cap = capacity_spare_wake(i, p);
+
+			if (spare_cap > max_spare_cap)
+				max_spare_cap = spare_cap;
 		}
 
 		/* Adjust by relative CPU capacity of the group */
@@ -4826,22 +6014,51 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p,
 
 		if (local_group) {
 			this_load = avg_load;
-		} else if (avg_load < min_load) {
-			min_load = avg_load;
-			idlest = group;
+			this_spare = max_spare_cap;
+		} else {
+			if (avg_load < min_load) {
+				min_load = avg_load;
+				idlest = group;
+			}
+
+			if (most_spare < max_spare_cap) {
+				most_spare = max_spare_cap;
+				most_spare_sg = group;
+			}
 		}
 	} while (group = group->next, group != sd->groups);
 
+	/*
+	 * The cross-over point between using spare capacity or least load
+	 * is too conservative for high utilization tasks on partially
+	 * utilized systems if we require spare_capacity > task_util(p),
+	 * so we allow for some task stuffing by using
+	 * spare_capacity > task_util(p)/2.
+	 *
+	 * Spare capacity can't be used for fork because the utilization has
+	 * not been set yet, we must first select a rq to compute the initial
+	 * utilization.
+	 */
+	if (sd_flag & SD_BALANCE_FORK)
+		goto skip_spare;
+
+	if (this_spare > task_util(p) / 2 &&
+	    imbalance*this_spare > 100*most_spare)
+		return NULL;
+	else if (most_spare > task_util(p) / 2)
+		return most_spare_sg;
+
+skip_spare:
 	if (!idlest || 100*this_load < imbalance*min_load)
 		return NULL;
 	return idlest;
 }
 
 /*
- * find_idlest_cpu - find the idlest cpu among the cpus in group.
+ * find_idlest_group_cpu - find the idlest cpu among the cpus in group.
  */
 static int
-find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
+find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 {
 	unsigned long load, min_load = ULONG_MAX;
 	unsigned int min_exit_latency = UINT_MAX;
@@ -4850,6 +6067,10 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 	int shallowest_idle_cpu = -1;
 	int i;
 
+	/* Check if we have any choice: */
+	if (group->group_weight == 1)
+		return cpumask_first(sched_group_cpus(group));
+
 	/* Traverse only the allowed CPUs */
 	for_each_cpu_and(i, sched_group_cpus(group), tsk_cpus_allowed(p)) {
 		if (idle_cpu(i)) {
@@ -4884,25 +6105,100 @@ find_idlest_cpu(struct sched_group *group, struct task_struct *p, int this_cpu)
 	}
 
 	return shallowest_idle_cpu != -1 ? shallowest_idle_cpu : least_loaded_cpu;
+ }
+
+static inline int find_idlest_cpu(struct sched_domain *sd, struct task_struct *p,
+				  int cpu, int prev_cpu, int sd_flag)
+{
+	int new_cpu = cpu;
+	int wu = sd_flag & SD_BALANCE_WAKE;
+	int cas_cpu = -1;
+
+	if (wu) {
+		schedstat_inc(p, se.statistics.nr_wakeups_cas_attempts);
+		schedstat_inc(this_rq(), eas_stats.cas_attempts);
+	}
+
+	if (!cpumask_intersects(sched_domain_span(sd), &p->cpus_allowed))
+		return prev_cpu;
+
+	while (sd) {
+		struct sched_group *group;
+		struct sched_domain *tmp;
+		int weight;
+
+		if (wu)
+			schedstat_inc(sd, eas_stats.cas_attempts);
+
+		if (!(sd->flags & sd_flag)) {
+			sd = sd->child;
+			continue;
+		}
+
+		group = find_idlest_group(sd, p, cpu, sd_flag);
+		if (!group) {
+			sd = sd->child;
+			continue;
+		}
+
+		new_cpu = find_idlest_group_cpu(group, p, cpu);
+		if (new_cpu == cpu) {
+			/* Now try balancing at a lower domain level of cpu */
+			sd = sd->child;
+			continue;
+		}
+
+		/* Now try balancing at a lower domain level of new_cpu */
+		cpu = cas_cpu = new_cpu;
+		weight = sd->span_weight;
+		sd = NULL;
+		for_each_domain(cpu, tmp) {
+			if (weight <= tmp->span_weight)
+				break;
+			if (tmp->flags & sd_flag)
+				sd = tmp;
+		}
+		/* while loop will break here if sd == NULL */
+	}
+
+	if (wu && (cas_cpu >= 0)) {
+		schedstat_inc(p, se.statistics.nr_wakeups_cas_count);
+		schedstat_inc(this_rq(), eas_stats.cas_count);
+	}
+
+	return new_cpu;
 }
 
 /*
  * Try and locate an idle CPU in the sched_domain.
  */
-static int select_idle_sibling(struct task_struct *p, int target)
+static int select_idle_sibling(struct task_struct *p, int prev, int target)
 {
 	struct sched_domain *sd;
 	struct sched_group *sg;
-	int i = task_cpu(p);
-
-	if (idle_cpu(target))
-		return target;
+	int best_idle_cpu = -1;
+	int best_idle_cstate = INT_MAX;
+	unsigned long best_idle_capacity = ULONG_MAX;
+
+	schedstat_inc(p, se.statistics.nr_wakeups_sis_attempts);
+	schedstat_inc(this_rq(), eas_stats.sis_attempts);
+
+	if (!sysctl_sched_cstate_aware) {
+		if (idle_cpu(target)) {
+			schedstat_inc(p, se.statistics.nr_wakeups_sis_idle);
+			schedstat_inc(this_rq(), eas_stats.sis_idle);
+			return target;
+		}
 
-	/*
-	 * If the prevous cpu is cache affine and idle, don't be stupid.
-	 */
-	if (i != target && cpus_share_cache(i, target) && idle_cpu(i))
-		return i;
+		/*
+		 * If the prevous cpu is cache affine and idle, don't be stupid.
+		 */
+		if (prev != target && cpus_share_cache(prev, target) && idle_cpu(prev)) {
+			schedstat_inc(p, se.statistics.nr_wakeups_sis_cache_affine);
+			schedstat_inc(this_rq(), eas_stats.sis_cache_affine);
+			return prev;
+		}
+	}
 
 	/*
 	 * Otherwise, iterate the domains and find an elegible idle cpu.
@@ -4911,60 +6207,500 @@ static int select_idle_sibling(struct task_struct *p, int target)
 	for_each_lower_domain(sd) {
 		sg = sd->groups;
 		do {
+			int i;
 			if (!cpumask_intersects(sched_group_cpus(sg),
 						tsk_cpus_allowed(p)))
 				goto next;
 
-			for_each_cpu(i, sched_group_cpus(sg)) {
-				if (i == target || !idle_cpu(i))
-					goto next;
-			}
+			if (sysctl_sched_cstate_aware) {
+				for_each_cpu_and(i, tsk_cpus_allowed(p), sched_group_cpus(sg)) {
+					int idle_idx = idle_get_state_idx(cpu_rq(i));
+					unsigned long new_usage = boosted_task_util(p);
+					unsigned long capacity_orig = capacity_orig_of(i);
+
+					if (new_usage > capacity_orig || !idle_cpu(i))
+						goto next;
+
+					if (i == target && new_usage <= capacity_curr_of(target)) {
+						schedstat_inc(p, se.statistics.nr_wakeups_sis_suff_cap);
+						schedstat_inc(this_rq(), eas_stats.sis_suff_cap);
+						schedstat_inc(sd, eas_stats.sis_suff_cap);
+						return target;
+					}
+
+					if (idle_idx < best_idle_cstate &&
+					    capacity_orig <= best_idle_capacity) {
+						best_idle_cpu = i;
+						best_idle_cstate = idle_idx;
+						best_idle_capacity = capacity_orig;
+					}
+				}
+			} else {
+				for_each_cpu(i, sched_group_cpus(sg)) {
+					if (i == target || !idle_cpu(i))
+						goto next;
+				}
 
-			target = cpumask_first_and(sched_group_cpus(sg),
+				target = cpumask_first_and(sched_group_cpus(sg),
 					tsk_cpus_allowed(p));
-			goto done;
+				schedstat_inc(p, se.statistics.nr_wakeups_sis_idle_cpu);
+				schedstat_inc(this_rq(), eas_stats.sis_idle_cpu);
+				schedstat_inc(sd, eas_stats.sis_idle_cpu);
+				goto done;
+			}
 next:
 			sg = sg->next;
 		} while (sg != sd->groups);
 	}
+
+	if (best_idle_cpu >= 0)
+		target = best_idle_cpu;
+
 done:
+	schedstat_inc(p, se.statistics.nr_wakeups_sis_count);
+	schedstat_inc(this_rq(), eas_stats.sis_count);
+
 	return target;
 }
 
 /*
- * cpu_util returns the amount of capacity of a CPU that is used by CFS
- * tasks. The unit of the return value must be the one of capacity so we can
- * compare the utilization with the capacity of the CPU that is available for
- * CFS task (ie cpu_capacity).
- *
- * cfs_rq.avg.util_avg is the sum of running time of runnable tasks plus the
- * recent utilization of currently non-runnable tasks on a CPU. It represents
- * the amount of utilization of a CPU in the range [0..capacity_orig] where
- * capacity_orig is the cpu_capacity available at the highest frequency
- * (arch_scale_freq_capacity()).
- * The utilization of a CPU converges towards a sum equal to or less than the
- * current capacity (capacity_curr <= capacity_orig) of the CPU because it is
- * the running time on this CPU scaled by capacity_curr.
- *
- * Nevertheless, cfs_rq.avg.util_avg can be higher than capacity_curr or even
- * higher than capacity_orig because of unfortunate rounding in
- * cfs.avg.util_avg or just after migrating tasks and new task wakeups until
- * the average stabilizes with the new running time. We need to check that the
- * utilization stays within the range of [0..capacity_orig] and cap it if
- * necessary. Without utilization capping, a group could be seen as overloaded
- * (CPU0 utilization at 121% + CPU1 utilization at 80%) whereas CPU1 has 20% of
- * available capacity. We allow utilization to overshoot capacity_curr (but not
- * capacity_orig) as it useful for predicting the capacity required after task
- * migrations (scheduler-driven DVFS).
- */
-static int cpu_util(int cpu)
-{
-	unsigned long util = cpu_rq(cpu)->cfs.avg.util_avg;
-	unsigned long capacity = capacity_orig_of(cpu);
+ * cpu_util_wake: Compute cpu utilization with any contributions from
+ * the waking task p removed.  check_for_migration() looks for a better CPU of
+ * rq->curr. For that case we should return cpu util with contributions from
+ * currently running task p removed.
+ */
+static int cpu_util_wake(int cpu, struct task_struct *p)
+{
+	unsigned long util, capacity;
+
+#ifdef CONFIG_SCHED_WALT
+	/*
+	 * WALT does not decay idle tasks in the same manner
+	 * as PELT, so it makes little sense to subtract task
+	 * utilization from cpu utilization. Instead just use
+	 * cpu_util for this case.
+	 */
+	if (!walt_disabled && sysctl_sched_use_walt_cpu_util &&
+	    p->state == TASK_WAKING)
+		return cpu_util(cpu);
+#endif
+	/* Task has no contribution or is new */
+	if (cpu != task_cpu(p) || !p->se.avg.last_update_time)
+		return cpu_util(cpu);
+
+	capacity = capacity_orig_of(cpu);
+	util = max_t(long, cpu_util(cpu) - task_util(p), 0);
 
 	return (util >= capacity) ? capacity : util;
 }
 
+static int start_cpu(bool boosted)
+{
+	struct root_domain *rd = cpu_rq(smp_processor_id())->rd;
+
+	return boosted ? rd->max_cap_orig_cpu : rd->min_cap_orig_cpu;
+}
+
+static inline int find_best_target(struct task_struct *p, int *backup_cpu,
+				   bool boosted, bool prefer_idle)
+{
+	unsigned long best_idle_min_cap_orig = ULONG_MAX;
+	unsigned long min_util = boosted_task_util(p);
+	unsigned long target_capacity = ULONG_MAX;
+	unsigned long min_wake_util = ULONG_MAX;
+	unsigned long target_max_spare_cap = 0;
+	unsigned long best_active_util = ULONG_MAX;
+	int best_idle_cstate = INT_MAX;
+	struct sched_domain *sd;
+	struct sched_group *sg;
+	int best_active_cpu = -1;
+	int best_idle_cpu = -1;
+	int target_cpu = -1;
+	int cpu, i;
+
+	*backup_cpu = -1;
+
+	schedstat_inc(p, se.statistics.nr_wakeups_fbt_attempts);
+	schedstat_inc(this_rq(), eas_stats.fbt_attempts);
+
+	/* Find start CPU based on boost value */
+	cpu = start_cpu(boosted);
+	if (cpu < 0) {
+		schedstat_inc(p, se.statistics.nr_wakeups_fbt_no_cpu);
+		schedstat_inc(this_rq(), eas_stats.fbt_no_cpu);
+		return -1;
+	}
+
+	/* Find SD for the start CPU */
+	sd = rcu_dereference(per_cpu(sd_ea, cpu));
+	if (!sd) {
+		schedstat_inc(p, se.statistics.nr_wakeups_fbt_no_sd);
+		schedstat_inc(this_rq(), eas_stats.fbt_no_sd);
+		return -1;
+	}
+
+	/* Scan CPUs in all SDs */
+	sg = sd->groups;
+	do {
+		for_each_cpu_and(i, tsk_cpus_allowed(p), sched_group_cpus(sg)) {
+			unsigned long capacity_curr = capacity_curr_of(i);
+			unsigned long capacity_orig = capacity_orig_of(i);
+			unsigned long wake_util, new_util;
+
+			if (!cpu_online(i))
+				continue;
+
+			if (walt_cpu_high_irqload(i))
+				continue;
+
+			/*
+			 * p's blocked utilization is still accounted for on prev_cpu
+			 * so prev_cpu will receive a negative bias due to the double
+			 * accounting. However, the blocked utilization may be zero.
+			 */
+			wake_util = cpu_util_wake(i, p);
+			new_util = wake_util + task_util(p);
+
+			/*
+			 * Ensure minimum capacity to grant the required boost.
+			 * The target CPU can be already at a capacity level higher
+			 * than the one required to boost the task.
+			 */
+			new_util = max(min_util, new_util);
+			if (new_util > capacity_orig)
+				continue;
+
+			/*
+			 * Case A) Latency sensitive tasks
+			 *
+			 * Unconditionally favoring tasks that prefer idle CPU to
+			 * improve latency.
+			 *
+			 * Looking for:
+			 * - an idle CPU, whatever its idle_state is, since
+			 *   the first CPUs we explore are more likely to be
+			 *   reserved for latency sensitive tasks.
+			 * - a non idle CPU where the task fits in its current
+			 *   capacity and has the maximum spare capacity.
+			 * - a non idle CPU with lower contention from other
+			 *   tasks and running at the lowest possible OPP.
+			 *
+			 * The last two goals tries to favor a non idle CPU
+			 * where the task can run as if it is "almost alone".
+			 * A maximum spare capacity CPU is favoured since
+			 * the task already fits into that CPU's capacity
+			 * without waiting for an OPP chance.
+			 *
+			 * The following code path is the only one in the CPUs
+			 * exploration loop which is always used by
+			 * prefer_idle tasks. It exits the loop with wither a
+			 * best_active_cpu or a target_cpu which should
+			 * represent an optimal choice for latency sensitive
+			 * tasks.
+			 */
+			if (prefer_idle) {
+
+				/*
+				 * Case A.1: IDLE CPU
+				 * Return the first IDLE CPU we find.
+				 */
+				if (idle_cpu(i)) {
+					schedstat_inc(p, se.statistics.nr_wakeups_fbt_pref_idle);
+					schedstat_inc(this_rq(), eas_stats.fbt_pref_idle);
+
+					trace_sched_find_best_target(p,
+							prefer_idle, min_util,
+							cpu, best_idle_cpu,
+							best_active_cpu, i);
+
+					return i;
+				}
+
+				/*
+				 * Case A.2: Target ACTIVE CPU
+				 * Favor CPUs with max spare capacity.
+				 */
+				if ((capacity_curr > new_util) &&
+					(capacity_orig - new_util > target_max_spare_cap)) {
+					target_max_spare_cap = capacity_orig - new_util;
+					target_cpu = i;
+					continue;
+				}
+				if (target_cpu != -1)
+					continue;
+
+
+				/*
+				 * Case A.3: Backup ACTIVE CPU
+				 * Favor CPUs with:
+				 * - lower utilization due to other tasks
+				 * - lower utilization with the task in
+				 */
+				if (wake_util > min_wake_util)
+					continue;
+				if (new_util > best_active_util)
+					continue;
+				min_wake_util = wake_util;
+				best_active_util = new_util;
+				best_active_cpu = i;
+				continue;
+			}
+
+			/*
+			 * Enforce EAS mode
+			 *
+			 * For non latency sensitive tasks, skip CPUs that
+			 * will be overutilized by moving the task there.
+			 *
+			 * The goal here is to remain in EAS mode as long as
+			 * possible at least for !prefer_idle tasks.
+			 */
+			if ((new_util * capacity_margin) >
+			    (capacity_orig * SCHED_CAPACITY_SCALE))
+				continue;
+
+			/*
+			 * Case B) Non latency sensitive tasks on IDLE CPUs.
+			 *
+			 * Find an optimal backup IDLE CPU for non latency
+			 * sensitive tasks.
+			 *
+			 * Looking for:
+			 * - minimizing the capacity_orig,
+			 *   i.e. preferring LITTLE CPUs
+			 * - favoring shallowest idle states
+			 *   i.e. avoid to wakeup deep-idle CPUs
+			 *
+			 * The following code path is used by non latency
+			 * sensitive tasks if IDLE CPUs are available. If at
+			 * least one of such CPUs are available it sets the
+			 * best_idle_cpu to the most suitable idle CPU to be
+			 * selected.
+			 *
+			 * If idle CPUs are available, favour these CPUs to
+			 * improve performances by spreading tasks.
+			 * Indeed, the energy_diff() computed by the caller
+			 * will take care to ensure the minimization of energy
+			 * consumptions without affecting performance.
+			 */
+			if (idle_cpu(i)) {
+				int idle_idx = idle_get_state_idx(cpu_rq(i));
+
+				/* Select idle CPU with lower cap_orig */
+				if (capacity_orig > best_idle_min_cap_orig)
+					continue;
+
+				/*
+				 * Skip CPUs in deeper idle state, but only
+				 * if they are also less energy efficient.
+				 * IOW, prefer a deep IDLE LITTLE CPU vs a
+				 * shallow idle big CPU.
+				 */
+				if (sysctl_sched_cstate_aware &&
+				    best_idle_cstate <= idle_idx)
+					continue;
+
+				/* Keep track of best idle CPU */
+				best_idle_min_cap_orig = capacity_orig;
+				best_idle_cstate = idle_idx;
+				best_idle_cpu = i;
+				continue;
+			}
+
+			/*
+			 * Case C) Non latency sensitive tasks on ACTIVE CPUs.
+			 *
+			 * Pack tasks in the most energy efficient capacities.
+			 *
+			 * This task packing strategy prefers more energy
+			 * efficient CPUs (i.e. pack on smaller maximum
+			 * capacity CPUs) while also trying to spread tasks to
+			 * run them all at the lower OPP.
+			 *
+			 * This assumes for example that it's more energy
+			 * efficient to run two tasks on two CPUs at a lower
+			 * OPP than packing both on a single CPU but running
+			 * that CPU at an higher OPP.
+			 *
+			 * Thus, this case keep track of the CPU with the
+			 * smallest maximum capacity and highest spare maximum
+			 * capacity.
+			 */
+
+			/* Favor CPUs with smaller capacity */
+			if (capacity_orig > target_capacity)
+				continue;
+
+			/* Favor CPUs with maximum spare capacity */
+			if ((capacity_orig - new_util) < target_max_spare_cap)
+				continue;
+
+			target_max_spare_cap = capacity_orig - new_util;
+			target_capacity = capacity_orig;
+			target_cpu = i;
+		}
+
+	} while (sg = sg->next, sg != sd->groups);
+
+	/*
+	 * For non latency sensitive tasks, cases B and C in the previous loop,
+	 * we pick the best IDLE CPU only if we was not able to find a target
+	 * ACTIVE CPU.
+	 *
+	 * Policies priorities:
+	 *
+	 * - prefer_idle tasks:
+	 *
+	 *   a) IDLE CPU available, we return immediately
+	 *   b) ACTIVE CPU where task fits and has the bigger maximum spare
+	 *      capacity (i.e. target_cpu)
+	 *   c) ACTIVE CPU with less contention due to other tasks
+	 *      (i.e. best_active_cpu)
+	 *
+	 * - NON prefer_idle tasks:
+	 *
+	 *   a) ACTIVE CPU: target_cpu
+	 *   b) IDLE CPU: best_idle_cpu
+	 */
+	if (target_cpu == -1)
+		target_cpu = prefer_idle
+			? best_active_cpu
+			: best_idle_cpu;
+	else
+		*backup_cpu = prefer_idle
+		? best_active_cpu
+		: best_idle_cpu;
+
+	trace_sched_find_best_target(p, prefer_idle, min_util, cpu,
+				     best_idle_cpu, best_active_cpu,
+				     target_cpu);
+
+	schedstat_inc(p, se.statistics.nr_wakeups_fbt_count);
+	schedstat_inc(this_rq(), eas_stats.fbt_count);
+
+	return target_cpu;
+}
+
+/*
+ * Disable WAKE_AFFINE in the case where task @p doesn't fit in the
+ * capacity of either the waking CPU @cpu or the previous CPU @prev_cpu.
+ *
+ * In that case WAKE_AFFINE doesn't make sense and we'll let
+ * BALANCE_WAKE sort things out.
+ */
+static int wake_cap(struct task_struct *p, int cpu, int prev_cpu)
+{
+	long min_cap, max_cap;
+
+	min_cap = min(capacity_orig_of(prev_cpu), capacity_orig_of(cpu));
+	max_cap = cpu_rq(cpu)->rd->max_cpu_capacity.val;
+
+	/* Minimum capacity is close to max, no need to abort wake_affine */
+	if (max_cap - min_cap < max_cap >> 3)
+		return 0;
+
+	/* Bring task utilization in sync with prev_cpu */
+	sync_entity_load_avg(&p->se);
+
+	return min_cap * 1024 < task_util(p) * capacity_margin;
+}
+
+static int select_energy_cpu_brute(struct task_struct *p, int prev_cpu, int sync)
+{
+	struct sched_domain *sd;
+	int target_cpu = prev_cpu, tmp_target, tmp_backup;
+	bool boosted, prefer_idle;
+
+	schedstat_inc(p, se.statistics.nr_wakeups_secb_attempts);
+	schedstat_inc(this_rq(), eas_stats.secb_attempts);
+
+	if (sysctl_sched_sync_hint_enable && sync) {
+		int cpu = smp_processor_id();
+
+		if (cpumask_test_cpu(cpu, tsk_cpus_allowed(p))) {
+			schedstat_inc(p, se.statistics.nr_wakeups_secb_sync);
+			schedstat_inc(this_rq(), eas_stats.secb_sync);
+			return cpu;
+		}
+	}
+
+	rcu_read_lock();
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+	boosted = schedtune_task_boost(p) > 0;
+	prefer_idle = schedtune_prefer_idle(p) > 0;
+#else
+	boosted = get_sysctl_sched_cfs_boost() > 0;
+	prefer_idle = 0;
+#endif
+
+	sync_entity_load_avg(&p->se);
+
+	sd = rcu_dereference(per_cpu(sd_ea, prev_cpu));
+	/* Find a cpu with sufficient capacity */
+	tmp_target = find_best_target(p, &tmp_backup, boosted, prefer_idle);
+
+	if (!sd)
+		goto unlock;
+	if (tmp_target >= 0) {
+		target_cpu = tmp_target;
+		if ((boosted || prefer_idle) && idle_cpu(target_cpu)) {
+			schedstat_inc(p, se.statistics.nr_wakeups_secb_idle_bt);
+			schedstat_inc(this_rq(), eas_stats.secb_idle_bt);
+			goto unlock;
+		}
+	}
+
+	if (target_cpu != prev_cpu) {
+		int delta = 0;
+		struct energy_env eenv = {
+			.util_delta     = task_util(p),
+			.src_cpu        = prev_cpu,
+			.dst_cpu        = target_cpu,
+			.task           = p,
+			.trg_cpu	= target_cpu,
+		};
+
+
+#ifdef CONFIG_SCHED_WALT
+		if (!walt_disabled && sysctl_sched_use_walt_cpu_util)
+			delta = task_util(p);
+#endif
+		/* Not enough spare capacity on previous cpu */
+		if (__cpu_overutilized(prev_cpu, delta)) {
+			schedstat_inc(p, se.statistics.nr_wakeups_secb_insuff_cap);
+			schedstat_inc(this_rq(), eas_stats.secb_insuff_cap);
+			goto unlock;
+		}
+
+		if (energy_diff(&eenv) >= 0) {
+			/* No energy saving for target_cpu, try backup */
+			target_cpu = tmp_backup;
+			eenv.dst_cpu = target_cpu;
+			eenv.trg_cpu = target_cpu;
+			if (tmp_backup < 0 ||
+			    tmp_backup == prev_cpu ||
+			    energy_diff(&eenv) >= 0) {
+				schedstat_inc(p, se.statistics.nr_wakeups_secb_no_nrg_sav);
+				schedstat_inc(this_rq(), eas_stats.secb_no_nrg_sav);
+				target_cpu = prev_cpu;
+				goto unlock;
+			}
+		}
+
+		schedstat_inc(p, se.statistics.nr_wakeups_secb_nrg_sav);
+		schedstat_inc(this_rq(), eas_stats.secb_nrg_sav);
+		goto unlock;
+	}
+
+	schedstat_inc(p, se.statistics.nr_wakeups_secb_count);
+	schedstat_inc(this_rq(), eas_stats.secb_count);
+
+unlock:
+	rcu_read_unlock();
+
+	return target_cpu;
+}
+
 /*
  * select_task_rq_fair: Select target runqueue for the waking task in domains
  * that have the 'sd_flag' flag set. In practice, this is SD_BALANCE_WAKE,
@@ -4978,7 +6714,8 @@ static int cpu_util(int cpu)
  * preempt must be disabled.
  */
 static int
-select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_flags)
+select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_flags,
+		    int sibling_count_hint)
 {
 	struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL;
 	int cpu = smp_processor_id();
@@ -4986,8 +6723,15 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f
 	int want_affine = 0;
 	int sync = wake_flags & WF_SYNC;
 
-	if (sd_flag & SD_BALANCE_WAKE)
-		want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));
+	if (sd_flag & SD_BALANCE_WAKE) {
+		record_wakee(p);
+		want_affine = !wake_wide(p, sibling_count_hint) &&
+			      !wake_cap(p, cpu, prev_cpu) &&
+			      cpumask_test_cpu(cpu, &p->cpus_allowed);
+	}
+
+	if (energy_aware() && !(cpu_rq(prev_cpu)->rd->overutilized))
+		return select_energy_cpu_brute(p, prev_cpu, sync);
 
 	rcu_read_lock();
 	for_each_domain(cpu, tmp) {
@@ -5012,47 +6756,25 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f
 
 	if (affine_sd) {
 		sd = NULL; /* Prefer wake_affine over balance flags */
-		if (cpu != prev_cpu && wake_affine(affine_sd, p, sync))
+		if (cpu != prev_cpu && wake_affine(affine_sd, p, prev_cpu, sync))
 			new_cpu = cpu;
 	}
 
+	if (sd && !(sd_flag & SD_BALANCE_FORK)) {
+		/*
+		 * We're going to need the task's util for capacity_spare_wake
+		 * in find_idlest_group. Sync it up to prev_cpu's
+		 * last_update_time.
+		 */
+		sync_entity_load_avg(&p->se);
+	}
+
 	if (!sd) {
 		if (sd_flag & SD_BALANCE_WAKE) /* XXX always ? */
-			new_cpu = select_idle_sibling(p, new_cpu);
-
-	} else while (sd) {
-		struct sched_group *group;
-		int weight;
+			new_cpu = select_idle_sibling(p, prev_cpu, new_cpu);
 
-		if (!(sd->flags & sd_flag)) {
-			sd = sd->child;
-			continue;
-		}
-
-		group = find_idlest_group(sd, p, cpu, sd_flag);
-		if (!group) {
-			sd = sd->child;
-			continue;
-		}
-
-		new_cpu = find_idlest_cpu(group, p, cpu);
-		if (new_cpu == -1 || new_cpu == cpu) {
-			/* Now try balancing at a lower domain level of cpu */
-			sd = sd->child;
-			continue;
-		}
-
-		/* Now try balancing at a lower domain level of new_cpu */
-		cpu = new_cpu;
-		weight = sd->span_weight;
-		sd = NULL;
-		for_each_domain(cpu, tmp) {
-			if (weight <= tmp->span_weight)
-				break;
-			if (tmp->flags & sd_flag)
-				sd = tmp;
-		}
-		/* while loop will break here if sd == NULL */
+	} else {
+		new_cpu = find_idlest_cpu(sd, p, cpu, prev_cpu, sd_flag);
 	}
 	rcu_read_unlock();
 
@@ -5087,6 +6809,8 @@ static void task_dead_fair(struct task_struct *p)
 {
 	remove_entity_load_avg(&p->se);
 }
+#else
+#define task_fits_max(p, cpu) true
 #endif /* CONFIG_SMP */
 
 static unsigned long
@@ -5333,6 +7057,8 @@ again:
 	if (hrtick_enabled(rq))
 		hrtick_start_fair(rq, p);
 
+	rq->misfit_task = !task_fits_max(p, rq->cpu);
+
 	return p;
 simple:
 	cfs_rq = &rq->cfs;
@@ -5354,9 +7080,12 @@ simple:
 	if (hrtick_enabled(rq))
 		hrtick_start_fair(rq, p);
 
+	rq->misfit_task = !task_fits_max(p, rq->cpu);
+
 	return p;
 
 idle:
+	rq->misfit_task = 0;
 	/*
 	 * This is OK, because current is on_cpu, which avoids it being picked
 	 * for load-balance and preemption/IRQs are still disabled avoiding
@@ -5569,6 +7298,13 @@ static unsigned long __read_mostly max_load_balance_interval = HZ/10;
 
 enum fbq_type { regular, remote, all };
 
+enum group_type {
+	group_other = 0,
+	group_misfit_task,
+	group_imbalanced,
+	group_overloaded,
+};
+
 #define LBF_ALL_PINNED	0x01
 #define LBF_NEED_BREAK	0x02
 #define LBF_DST_PINNED  0x04
@@ -5587,6 +7323,7 @@ struct lb_env {
 	int			new_dst_cpu;
 	enum cpu_idle_type	idle;
 	long			imbalance;
+	unsigned int		src_grp_nr_running;
 	/* The set of CPUs under consideration for load-balancing */
 	struct cpumask		*cpus;
 
@@ -5597,6 +7334,7 @@ struct lb_env {
 	unsigned int		loop_max;
 
 	enum fbq_type		fbq_type;
+	enum group_type		busiest_group_type;
 	struct list_head	tasks;
 };
 
@@ -5778,7 +7516,9 @@ static void detach_task(struct task_struct *p, struct lb_env *env)
 
 	deactivate_task(env->src_rq, p, 0);
 	p->on_rq = TASK_ON_RQ_MIGRATING;
+	double_lock_balance(env->src_rq, env->dst_rq);
 	set_task_cpu(p, env->dst_cpu);
+	double_unlock_balance(env->src_rq, env->dst_rq);
 }
 
 /*
@@ -5966,8 +7706,13 @@ static void update_blocked_averages(int cpu)
 		if (throttled_hierarchy(cfs_rq))
 			continue;
 
-		if (update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq))
+		if (update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq,
+					   true))
 			update_tg_load_avg(cfs_rq, 0);
+
+		/* Propagate pending load changes to the parent */
+		if (cfs_rq->tg->se[cpu])
+			update_load_avg(cfs_rq->tg->se[cpu], 0);
 	}
 	raw_spin_unlock_irqrestore(&rq->lock, flags);
 }
@@ -6027,7 +7772,7 @@ static inline void update_blocked_averages(int cpu)
 
 	raw_spin_lock_irqsave(&rq->lock, flags);
 	update_rq_clock(rq);
-	update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq);
+	update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq, true);
 	raw_spin_unlock_irqrestore(&rq->lock, flags);
 }
 
@@ -6039,12 +7784,6 @@ static unsigned long task_h_load(struct task_struct *p)
 
 /********** Helpers for find_busiest_group ************************/
 
-enum group_type {
-	group_other = 0,
-	group_imbalanced,
-	group_overloaded,
-};
-
 /*
  * sg_lb_stats - stats of a sched_group required for load_balancing
  */
@@ -6060,6 +7799,7 @@ struct sg_lb_stats {
 	unsigned int group_weight;
 	enum group_type group_type;
 	int group_no_capacity;
+	int group_misfit_task; /* A cpu has a task too big for its capacity */
 #ifdef CONFIG_NUMA_BALANCING
 	unsigned int nr_numa_running;
 	unsigned int nr_preferred_running;
@@ -6151,19 +7891,58 @@ static unsigned long scale_rt_capacity(int cpu)
 
 	used = div_u64(avg, total);
 
+	/*
+	 * deadline bandwidth is defined at system level so we must
+	 * weight this bandwidth with the max capacity of the system.
+	 * As a reminder, avg_bw is 20bits width and
+	 * scale_cpu_capacity is 10 bits width
+	 */
+	used += div_u64(rq->dl.avg_bw, arch_scale_cpu_capacity(NULL, cpu));
+
 	if (likely(used < SCHED_CAPACITY_SCALE))
 		return SCHED_CAPACITY_SCALE - used;
 
 	return 1;
 }
 
+void init_max_cpu_capacity(struct max_cpu_capacity *mcc)
+{
+	raw_spin_lock_init(&mcc->lock);
+	mcc->val = 0;
+	mcc->cpu = -1;
+}
+
 static void update_cpu_capacity(struct sched_domain *sd, int cpu)
 {
 	unsigned long capacity = arch_scale_cpu_capacity(sd, cpu);
 	struct sched_group *sdg = sd->groups;
+	struct max_cpu_capacity *mcc;
+	unsigned long max_capacity;
+	int max_cap_cpu;
+	unsigned long flags;
 
 	cpu_rq(cpu)->cpu_capacity_orig = capacity;
 
+	mcc = &cpu_rq(cpu)->rd->max_cpu_capacity;
+
+	raw_spin_lock_irqsave(&mcc->lock, flags);
+	max_capacity = mcc->val;
+	max_cap_cpu = mcc->cpu;
+
+	if ((max_capacity > capacity && max_cap_cpu == cpu) ||
+	    (max_capacity < capacity)) {
+		mcc->val = capacity;
+		mcc->cpu = cpu;
+#ifdef CONFIG_SCHED_DEBUG
+		raw_spin_unlock_irqrestore(&mcc->lock, flags);
+		printk_deferred(KERN_INFO "CPU%d: update max cpu_capacity %lu\n",
+				cpu, capacity);
+		goto skip_unlock;
+#endif
+	}
+	raw_spin_unlock_irqrestore(&mcc->lock, flags);
+
+skip_unlock: __attribute__ ((unused));
 	capacity *= scale_rt_capacity(cpu);
 	capacity >>= SCHED_CAPACITY_SHIFT;
 
@@ -6172,13 +7951,15 @@ static void update_cpu_capacity(struct sched_domain *sd, int cpu)
 
 	cpu_rq(cpu)->cpu_capacity = capacity;
 	sdg->sgc->capacity = capacity;
+	sdg->sgc->max_capacity = capacity;
+	sdg->sgc->min_capacity = capacity;
 }
 
 void update_group_capacity(struct sched_domain *sd, int cpu)
 {
 	struct sched_domain *child = sd->child;
 	struct sched_group *group, *sdg = sd->groups;
-	unsigned long capacity;
+	unsigned long capacity, max_capacity, min_capacity;
 	unsigned long interval;
 
 	interval = msecs_to_jiffies(sd->balance_interval);
@@ -6191,6 +7972,8 @@ void update_group_capacity(struct sched_domain *sd, int cpu)
 	}
 
 	capacity = 0;
+	max_capacity = 0;
+	min_capacity = ULONG_MAX;
 
 	if (child->flags & SD_OVERLAP) {
 		/*
@@ -6215,11 +7998,13 @@ void update_group_capacity(struct sched_domain *sd, int cpu)
 			 */
 			if (unlikely(!rq->sd)) {
 				capacity += capacity_of(cpu);
-				continue;
+			} else {
+				sgc = rq->sd->groups->sgc;
+				capacity += sgc->capacity;
 			}
 
-			sgc = rq->sd->groups->sgc;
-			capacity += sgc->capacity;
+			max_capacity = max(capacity, max_capacity);
+			min_capacity = min(capacity, min_capacity);
 		}
 	} else  {
 		/*
@@ -6229,12 +8014,18 @@ void update_group_capacity(struct sched_domain *sd, int cpu)
 
 		group = child->groups;
 		do {
-			capacity += group->sgc->capacity;
+			struct sched_group_capacity *sgc = group->sgc;
+
+			capacity += sgc->capacity;
+			max_capacity = max(sgc->max_capacity, max_capacity);
+			min_capacity = min(sgc->min_capacity, min_capacity);
 			group = group->next;
 		} while (group != child->groups);
 	}
 
 	sdg->sgc->capacity = capacity;
+	sdg->sgc->max_capacity = max_capacity;
+	sdg->sgc->min_capacity = min_capacity;
 }
 
 /*
@@ -6329,6 +8120,18 @@ group_is_overloaded(struct lb_env *env, struct sg_lb_stats *sgs)
 	return false;
 }
 
+
+/*
+ * group_smaller_cpu_capacity: Returns true if sched_group sg has smaller
+ * per-cpu capacity than sched_group ref.
+ */
+static inline bool
+group_smaller_cpu_capacity(struct sched_group *sg, struct sched_group *ref)
+{
+	return sg->sgc->max_capacity + capacity_margin - SCHED_LOAD_SCALE <
+							ref->sgc->max_capacity;
+}
+
 static inline enum
 group_type group_classify(struct sched_group *group,
 			  struct sg_lb_stats *sgs)
@@ -6339,9 +8142,44 @@ group_type group_classify(struct sched_group *group,
 	if (sg_imbalanced(group))
 		return group_imbalanced;
 
+	if (sgs->group_misfit_task)
+		return group_misfit_task;
+
 	return group_other;
 }
 
+#ifdef CONFIG_NO_HZ_COMMON
+/*
+ * idle load balancing data
+ *  - used by the nohz balance, but we want it available here
+ *    so that we can see which CPUs have no tick.
+ */
+static struct {
+	cpumask_var_t idle_cpus_mask;
+	atomic_t nr_cpus;
+	unsigned long next_balance;     /* in jiffy units */
+} nohz ____cacheline_aligned;
+
+static inline void update_cpu_stats_if_tickless(struct rq *rq)
+{
+	/* only called from update_sg_lb_stats when irqs are disabled */
+	if (cpumask_test_cpu(rq->cpu, nohz.idle_cpus_mask)) {
+		/* rate limit updates to once-per-jiffie at most */
+		if (READ_ONCE(jiffies) <= rq->last_load_update_tick)
+			return;
+
+		raw_spin_lock(&rq->lock);
+		update_rq_clock(rq);
+		update_idle_cpu_load(rq);
+		update_cfs_rq_load_avg(rq->clock_task, &rq->cfs, false);
+		raw_spin_unlock(&rq->lock);
+	}
+}
+
+#else
+static inline void update_cpu_stats_if_tickless(struct rq *rq) { }
+#endif
+
 /**
  * update_sg_lb_stats - Update sched_group's statistics for load balancing.
  * @env: The load balancing environment.
@@ -6350,20 +8188,27 @@ group_type group_classify(struct sched_group *group,
  * @local_group: Does group contain this_cpu.
  * @sgs: variable to hold the statistics for this group.
  * @overload: Indicate more than one runnable task for any CPU.
+ * @overutilized: Indicate overutilization for any CPU.
  */
 static inline void update_sg_lb_stats(struct lb_env *env,
 			struct sched_group *group, int load_idx,
 			int local_group, struct sg_lb_stats *sgs,
-			bool *overload)
+			bool *overload, bool *overutilized)
 {
 	unsigned long load;
-	int i;
+	int i, nr_running;
 
 	memset(sgs, 0, sizeof(*sgs));
 
 	for_each_cpu_and(i, sched_group_cpus(group), env->cpus) {
 		struct rq *rq = cpu_rq(i);
 
+		/* if we are entering idle and there are CPUs with
+		 * their tick stopped, do an update for them
+		 */
+		if (env->idle == CPU_NEWLY_IDLE)
+			update_cpu_stats_if_tickless(rq);
+
 		/* Bias balancing toward cpus of our domain */
 		if (local_group)
 			load = target_load(i, load_idx);
@@ -6374,7 +8219,8 @@ static inline void update_sg_lb_stats(struct lb_env *env,
 		sgs->group_util += cpu_util(i);
 		sgs->sum_nr_running += rq->cfs.h_nr_running;
 
-		if (rq->nr_running > 1)
+		nr_running = rq->nr_running;
+		if (nr_running > 1)
 			*overload = true;
 
 #ifdef CONFIG_NUMA_BALANCING
@@ -6382,8 +8228,17 @@ static inline void update_sg_lb_stats(struct lb_env *env,
 		sgs->nr_preferred_running += rq->nr_preferred_running;
 #endif
 		sgs->sum_weighted_load += weighted_cpuload(i);
-		if (idle_cpu(i))
+		/*
+		 * No need to call idle_cpu() if nr_running is not 0
+		 */
+		if (!nr_running && idle_cpu(i))
 			sgs->idle_cpus++;
+
+		if (cpu_overutilized(i)) {
+			*overutilized = true;
+			if (!sgs->group_misfit_task && rq->misfit_task)
+				sgs->group_misfit_task = capacity_of(i);
+		}
 	}
 
 	/* Adjust by relative CPU capacity of the group */
@@ -6425,9 +8280,31 @@ static bool update_sd_pick_busiest(struct lb_env *env,
 	if (sgs->group_type < busiest->group_type)
 		return false;
 
+	/*
+	 * Candidate sg doesn't face any serious load-balance problems
+	 * so don't pick it if the local sg is already filled up.
+	 */
+	if (sgs->group_type == group_other &&
+	    !group_has_capacity(env, &sds->local_stat))
+		return false;
+
 	if (sgs->avg_load <= busiest->avg_load)
 		return false;
 
+	if (!(env->sd->flags & SD_ASYM_CPUCAPACITY))
+		goto asym_packing;
+
+	/*
+	 * Candidate sg has no more than one task per CPU and
+	 * has higher per-CPU capacity. Migrating tasks to less
+	 * capable CPUs may harm throughput. Maximize throughput,
+	 * power/energy consequences are not considered.
+	 */
+	if (sgs->sum_nr_running <= sgs->group_weight &&
+	    group_smaller_cpu_capacity(sds->local, sg))
+		return false;
+
+asym_packing:
 	/* This is the busiest node in its class. */
 	if (!(env->sd->flags & SD_ASYM_PACKING))
 		return true;
@@ -6478,6 +8355,9 @@ static inline enum fbq_type fbq_classify_rq(struct rq *rq)
 }
 #endif /* CONFIG_NUMA_BALANCING */
 
+#define lb_sd_parent(sd) \
+	(sd->parent && sd->parent->groups != sd->parent->groups->next)
+
 /**
  * update_sd_lb_stats - Update sched_domain's statistics for load balancing.
  * @env: The load balancing environment.
@@ -6489,7 +8369,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 	struct sched_group *sg = env->sd->groups;
 	struct sg_lb_stats tmp_sgs;
 	int load_idx, prefer_sibling = 0;
-	bool overload = false;
+	bool overload = false, overutilized = false;
 
 	if (child && child->flags & SD_PREFER_SIBLING)
 		prefer_sibling = 1;
@@ -6511,7 +8391,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 		}
 
 		update_sg_lb_stats(env, sg, load_idx, local_group, sgs,
-						&overload);
+						&overload, &overutilized);
 
 		if (local_group)
 			goto next_group;
@@ -6533,6 +8413,15 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 			sgs->group_type = group_classify(sg, sgs);
 		}
 
+		/*
+		 * Ignore task groups with misfit tasks if local group has no
+		 * capacity or if per-cpu capacity isn't higher.
+		 */
+		if (sgs->group_type == group_misfit_task &&
+		    (!group_has_capacity(env, &sds->local_stat) ||
+		     !group_smaller_cpu_capacity(sg, sds->local)))
+			sgs->group_type = group_other;
+
 		if (update_sd_pick_busiest(env, sds, sg, sgs)) {
 			sds->busiest = sg;
 			sds->busiest_stat = *sgs;
@@ -6549,10 +8438,23 @@ next_group:
 	if (env->sd->flags & SD_NUMA)
 		env->fbq_type = fbq_classify_group(&sds->busiest_stat);
 
-	if (!env->sd->parent) {
+	env->src_grp_nr_running = sds->busiest_stat.sum_nr_running;
+
+	if (!lb_sd_parent(env->sd)) {
 		/* update overload indicator if we are at root domain */
 		if (env->dst_rq->rd->overload != overload)
 			env->dst_rq->rd->overload = overload;
+
+		/* Update over-utilization (tipping point, U >= 0) indicator */
+		if (env->dst_rq->rd->overutilized != overutilized) {
+			env->dst_rq->rd->overutilized = overutilized;
+			trace_sched_overutilized(overutilized);
+		}
+	} else {
+		if (!env->dst_rq->rd->overutilized && overutilized) {
+			env->dst_rq->rd->overutilized = true;
+			trace_sched_overutilized(true);
+		}
 	}
 
 }
@@ -6701,6 +8603,22 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
 	 */
 	if (busiest->avg_load <= sds->avg_load ||
 	    local->avg_load >= sds->avg_load) {
+		/* Misfitting tasks should be migrated in any case */
+		if (busiest->group_type == group_misfit_task) {
+			env->imbalance = busiest->group_misfit_task;
+			return;
+		}
+
+		/*
+		 * Busiest group is overloaded, local is not, use the spare
+		 * cycles to maximize throughput
+		 */
+		if (busiest->group_type == group_overloaded &&
+		    local->group_type <= group_misfit_task) {
+			env->imbalance = busiest->load_per_task;
+			return;
+		}
+
 		env->imbalance = 0;
 		return fix_small_imbalance(env, sds);
 	}
@@ -6734,6 +8652,11 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
 		(sds->avg_load - local->avg_load) * local->group_capacity
 	) / SCHED_CAPACITY_SCALE;
 
+	/* Boost imbalance to allow misfit task to be balanced. */
+	if (busiest->group_type == group_misfit_task)
+		env->imbalance = max_t(long, env->imbalance,
+				     busiest->group_misfit_task);
+
 	/*
 	 * if *imbalance is less than the average load per runnable task
 	 * there is no guarantee that any tasks will be moved so we'll have
@@ -6775,6 +8698,10 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 	 * this level.
 	 */
 	update_sd_lb_stats(env, &sds);
+
+	if (energy_aware() && !env->dst_rq->rd->overutilized)
+		goto out_balanced;
+
 	local = &sds.local_stat;
 	busiest = &sds.busiest_stat;
 
@@ -6798,11 +8725,19 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 	if (busiest->group_type == group_imbalanced)
 		goto force_balance;
 
-	/* SD_BALANCE_NEWIDLE trumps SMP nice when underutilized */
-	if (env->idle == CPU_NEWLY_IDLE && group_has_capacity(env, local) &&
+	/*
+	 * When dst_cpu is idle, prevent SMP nice and/or asymmetric group
+	 * capacities from resulting in underutilization due to avg_load.
+	 */
+	if (env->idle != CPU_NOT_IDLE && group_has_capacity(env, local) &&
 	    busiest->group_no_capacity)
 		goto force_balance;
 
+	/* Misfitting tasks should be dealt with regardless of the avg load */
+	if (busiest->group_type == group_misfit_task) {
+		goto force_balance;
+	}
+
 	/*
 	 * If the local group is busier than the selected busiest group
 	 * don't try and pull any tasks.
@@ -6826,7 +8761,8 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 		 * might end up to just move the imbalance on another group
 		 */
 		if ((busiest->group_type != group_overloaded) &&
-				(local->idle_cpus <= (busiest->idle_cpus + 1)))
+		    (local->idle_cpus <= (busiest->idle_cpus + 1)) &&
+		    !group_smaller_cpu_capacity(sds.busiest, sds.local))
 			goto out_balanced;
 	} else {
 		/*
@@ -6839,6 +8775,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 	}
 
 force_balance:
+	env->busiest_group_type = busiest->group_type;
 	/* Looks like there is an imbalance. Compute it */
 	calculate_imbalance(env, &sds);
 	return sds.busiest;
@@ -6897,7 +8834,8 @@ static struct rq *find_busiest_queue(struct lb_env *env,
 		 */
 
 		if (rq->nr_running == 1 && wl > env->imbalance &&
-		    !check_cpu_capacity(rq, env->sd))
+		    !check_cpu_capacity(rq, env->sd) &&
+		    env->busiest_group_type != group_misfit_task)
 			continue;
 
 		/*
@@ -6958,6 +8896,14 @@ static int need_active_balance(struct lb_env *env)
 			return 1;
 	}
 
+	if ((capacity_of(env->src_cpu) < capacity_of(env->dst_cpu)) &&
+	    ((capacity_orig_of(env->src_cpu) < capacity_orig_of(env->dst_cpu))) &&
+				env->src_rq->cfs.h_nr_running == 1 &&
+				cpu_overutilized(env->src_cpu) &&
+				!cpu_overutilized(env->dst_cpu)) {
+			return 1;
+	}
+
 	return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2);
 }
 
@@ -7006,7 +8952,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
 			int *continue_balancing)
 {
 	int ld_moved, cur_ld_moved, active_balance = 0;
-	struct sched_domain *sd_parent = sd->parent;
+	struct sched_domain *sd_parent = lb_sd_parent(sd) ? sd->parent : NULL;
 	struct sched_group *group;
 	struct rq *busiest;
 	unsigned long flags;
@@ -7073,6 +9019,7 @@ redo:
 
 more_balance:
 		raw_spin_lock_irqsave(&busiest->lock, flags);
+		update_rq_clock(busiest);
 
 		/*
 		 * cur_ld_moved - load moved in current iteration
@@ -7170,7 +9117,8 @@ more_balance:
 		 * excessive cache_hot migrations and active balances.
 		 */
 		if (idle != CPU_NEWLY_IDLE)
-			sd->nr_balance_failed++;
+			if (env.src_grp_nr_running > 1)
+				sd->nr_balance_failed++;
 
 		if (need_active_balance(&env)) {
 			raw_spin_lock_irqsave(&busiest->lock, flags);
@@ -7311,8 +9259,9 @@ static int idle_balance(struct rq *this_rq)
 	 */
 	this_rq->idle_stamp = rq_clock(this_rq);
 
-	if (this_rq->avg_idle < sysctl_sched_migration_cost ||
-	    !this_rq->rd->overload) {
+	if (!energy_aware() &&
+	    (this_rq->avg_idle < sysctl_sched_migration_cost ||
+	     !this_rq->rd->overload)) {
 		rcu_read_lock();
 		sd = rcu_dereference_check_sched_domain(this_rq->sd);
 		if (sd)
@@ -7405,8 +9354,18 @@ static int active_load_balance_cpu_stop(void *data)
 	int busiest_cpu = cpu_of(busiest_rq);
 	int target_cpu = busiest_rq->push_cpu;
 	struct rq *target_rq = cpu_rq(target_cpu);
-	struct sched_domain *sd;
+	struct sched_domain *sd = NULL;
 	struct task_struct *p = NULL;
+	struct task_struct *push_task;
+	int push_task_detached = 0;
+	struct lb_env env = {
+		.sd		= sd,
+		.dst_cpu	= target_cpu,
+		.dst_rq		= target_rq,
+		.src_cpu	= busiest_rq->cpu,
+		.src_rq		= busiest_rq,
+		.idle		= CPU_IDLE,
+	};
 
 	raw_spin_lock_irq(&busiest_rq->lock);
 
@@ -7426,6 +9385,17 @@ static int active_load_balance_cpu_stop(void *data)
 	 */
 	BUG_ON(busiest_rq == target_rq);
 
+	push_task = busiest_rq->push_task;
+	if (push_task) {
+		if (task_on_rq_queued(push_task) &&
+			task_cpu(push_task) == busiest_cpu &&
+					cpu_online(target_cpu)) {
+			detach_task(push_task, &env);
+			push_task_detached = 1;
+		}
+		goto out_unlock;
+	}
+
 	/* Search for an sd spanning us and the target CPU. */
 	rcu_read_lock();
 	for_each_domain(target_cpu, sd) {
@@ -7435,16 +9405,9 @@ static int active_load_balance_cpu_stop(void *data)
 	}
 
 	if (likely(sd)) {
-		struct lb_env env = {
-			.sd		= sd,
-			.dst_cpu	= target_cpu,
-			.dst_rq		= target_rq,
-			.src_cpu	= busiest_rq->cpu,
-			.src_rq		= busiest_rq,
-			.idle		= CPU_IDLE,
-		};
-
+		env.sd = sd;
 		schedstat_inc(sd, alb_count);
+		update_rq_clock(busiest_rq);
 
 		p = detach_one_task(&env);
 		if (p)
@@ -7455,8 +9418,18 @@ static int active_load_balance_cpu_stop(void *data)
 	rcu_read_unlock();
 out_unlock:
 	busiest_rq->active_balance = 0;
+
+	if (push_task)
+		busiest_rq->push_task = NULL;
+
 	raw_spin_unlock(&busiest_rq->lock);
 
+	if (push_task) {
+		if (push_task_detached)
+			attach_one_task(target_rq, push_task);
+		put_task_struct(push_task);
+	}
+
 	if (p)
 		attach_one_task(target_rq, p);
 
@@ -7477,12 +9450,6 @@ static inline int on_null_domain(struct rq *rq)
  *   needed, they will kick the idle load balancer, which then does idle
  *   load balancing for all the idle CPUs.
  */
-static struct {
-	cpumask_var_t idle_cpus_mask;
-	atomic_t nr_cpus;
-	unsigned long next_balance;     /* in jiffy units */
-} nohz ____cacheline_aligned;
-
 static inline int find_new_ilb(void)
 {
 	int ilb = cpumask_first(nohz.idle_cpus_mask);
@@ -7828,9 +9795,14 @@ static inline bool nohz_kick_needed(struct rq *rq)
 	if (time_before(now, nohz.next_balance))
 		return false;
 
-	if (rq->nr_running >= 2)
+	if (rq->nr_running >= 2 &&
+	    (!energy_aware() || cpu_overutilized(cpu)))
 		return true;
 
+	/* Do idle load balance if there have misfit task */
+	if (energy_aware())
+		return rq->misfit_task;
+
 	rcu_read_lock();
 	sd = rcu_dereference(per_cpu(sd_busy, cpu));
 	if (sd) {
@@ -7922,6 +9894,47 @@ static void rq_offline_fair(struct rq *rq)
 	unthrottle_offline_cfs_rqs(rq);
 }
 
+static inline int
+kick_active_balance(struct rq *rq, struct task_struct *p, int new_cpu)
+{
+	int rc = 0;
+
+	/* Invoke active balance to force migrate currently running task */
+	raw_spin_lock(&rq->lock);
+	if (!rq->active_balance) {
+		rq->active_balance = 1;
+		rq->push_cpu = new_cpu;
+		get_task_struct(p);
+		rq->push_task = p;
+		rc = 1;
+	}
+	raw_spin_unlock(&rq->lock);
+
+	return rc;
+}
+
+void check_for_migration(struct rq *rq, struct task_struct *p)
+{
+	int new_cpu;
+	int active_balance;
+	int cpu = task_cpu(p);
+
+	if (rq->misfit_task) {
+		if (rq->curr->state != TASK_RUNNING ||
+		    rq->curr->nr_cpus_allowed == 1)
+			return;
+
+		new_cpu = select_energy_cpu_brute(p, cpu, 0);
+		if (capacity_orig_of(new_cpu) > capacity_orig_of(cpu)) {
+			active_balance = kick_active_balance(rq, p, new_cpu);
+			if (active_balance)
+				stop_one_cpu_nowait(cpu,
+						active_load_balance_cpu_stop,
+						rq, &rq->active_balance_work);
+		}
+	}
+}
+
 #endif /* CONFIG_SMP */
 
 /*
@@ -7939,6 +9952,16 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 
 	if (static_branch_unlikely(&sched_numa_balancing))
 		task_tick_numa(rq, curr);
+
+#ifdef CONFIG_SMP
+	if (!rq->rd->overutilized && cpu_overutilized(task_cpu(curr))) {
+		rq->rd->overutilized = true;
+		trace_sched_overutilized(true);
+	}
+
+	rq->misfit_task = !task_fits_max(curr, rq->cpu);
+#endif
+
 }
 
 /*
@@ -7950,31 +9973,17 @@ static void task_fork_fair(struct task_struct *p)
 {
 	struct cfs_rq *cfs_rq;
 	struct sched_entity *se = &p->se, *curr;
-	int this_cpu = smp_processor_id();
 	struct rq *rq = this_rq();
-	unsigned long flags;
-
-	raw_spin_lock_irqsave(&rq->lock, flags);
 
+	raw_spin_lock(&rq->lock);
 	update_rq_clock(rq);
 
 	cfs_rq = task_cfs_rq(current);
 	curr = cfs_rq->curr;
-
-	/*
-	 * Not only the cpu but also the task_group of the parent might have
-	 * been changed after parent->se.parent,cfs_rq were copied to
-	 * child->se.parent,cfs_rq. So call __set_task_cpu() to make those
-	 * of child point to valid ones.
-	 */
-	rcu_read_lock();
-	__set_task_cpu(p, this_cpu);
-	rcu_read_unlock();
-
-	update_curr(cfs_rq);
-
-	if (curr)
+	if (curr) {
+		update_curr(cfs_rq);
 		se->vruntime = curr->vruntime;
+	}
 	place_entity(cfs_rq, se, 1);
 
 	if (sysctl_sched_child_runs_first && curr && entity_before(curr, se)) {
@@ -7987,8 +9996,7 @@ static void task_fork_fair(struct task_struct *p)
 	}
 
 	se->vruntime -= cfs_rq->min_vruntime;
-
-	raw_spin_unlock_irqrestore(&rq->lock, flags);
+	raw_spin_unlock(&rq->lock);
 }
 
 /*
@@ -8040,6 +10048,61 @@ static inline bool vruntime_normalized(struct task_struct *p)
 	return false;
 }
 
+#ifdef CONFIG_FAIR_GROUP_SCHED
+/*
+ * Propagate the changes of the sched_entity across the tg tree to make it
+ * visible to the root
+ */
+static void propagate_entity_cfs_rq(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq;
+
+	/* Start to propagate at parent */
+	se = se->parent;
+
+	for_each_sched_entity(se) {
+		cfs_rq = cfs_rq_of(se);
+
+		if (cfs_rq_throttled(cfs_rq))
+			break;
+
+		update_load_avg(se, UPDATE_TG);
+	}
+}
+#else
+static void propagate_entity_cfs_rq(struct sched_entity *se) { }
+#endif
+
+static void detach_entity_cfs_rq(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+	/* Catch up with the cfs_rq and remove our load when we leave */
+	update_load_avg(se, 0);
+	detach_entity_load_avg(cfs_rq, se);
+	update_tg_load_avg(cfs_rq, false);
+	propagate_entity_cfs_rq(se);
+}
+
+static void attach_entity_cfs_rq(struct sched_entity *se)
+{
+	struct cfs_rq *cfs_rq = cfs_rq_of(se);
+
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	/*
+	 * Since the real-depth could have been changed (only FAIR
+	 * class maintain depth value), reset depth properly.
+	 */
+	se->depth = se->parent ? se->parent->depth + 1 : 0;
+#endif
+
+	/* Synchronize entity with its cfs_rq */
+	update_load_avg(se, sched_feat(ATTACH_AGE_LOAD) ? 0 : SKIP_AGE_LOAD);
+	attach_entity_load_avg(cfs_rq, se);
+	update_tg_load_avg(cfs_rq, false);
+	propagate_entity_cfs_rq(se);
+}
+
 static void detach_task_cfs_rq(struct task_struct *p)
 {
 	struct sched_entity *se = &p->se;
@@ -8054,8 +10117,7 @@ static void detach_task_cfs_rq(struct task_struct *p)
 		se->vruntime -= cfs_rq->min_vruntime;
 	}
 
-	/* Catch up with the cfs_rq and remove our load when we leave */
-	detach_entity_load_avg(cfs_rq, se);
+	detach_entity_cfs_rq(se);
 }
 
 static void attach_task_cfs_rq(struct task_struct *p)
@@ -8063,16 +10125,7 @@ static void attach_task_cfs_rq(struct task_struct *p)
 	struct sched_entity *se = &p->se;
 	struct cfs_rq *cfs_rq = cfs_rq_of(se);
 
-#ifdef CONFIG_FAIR_GROUP_SCHED
-	/*
-	 * Since the real-depth could have been changed (only FAIR
-	 * class maintain depth value), reset depth properly.
-	 */
-	se->depth = se->parent ? se->parent->depth + 1 : 0;
-#endif
-
-	/* Synchronize task with its cfs_rq */
-	attach_entity_load_avg(cfs_rq, se);
+	attach_entity_cfs_rq(se);
 
 	if (!vruntime_normalized(p))
 		se->vruntime += cfs_rq->min_vruntime;
@@ -8126,12 +10179,23 @@ void init_cfs_rq(struct cfs_rq *cfs_rq)
 	cfs_rq->min_vruntime_copy = cfs_rq->min_vruntime;
 #endif
 #ifdef CONFIG_SMP
+#ifdef CONFIG_FAIR_GROUP_SCHED
+	cfs_rq->propagate_avg = 0;
+#endif
 	atomic_long_set(&cfs_rq->removed_load_avg, 0);
 	atomic_long_set(&cfs_rq->removed_util_avg, 0);
 #endif
 }
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
+static void task_set_group_fair(struct task_struct *p)
+{
+	struct sched_entity *se = &p->se;
+
+	set_task_rq(p, task_cpu(p));
+	se->depth = se->parent ? se->parent->depth + 1 : 0;
+}
+
 static void task_move_group_fair(struct task_struct *p)
 {
 	detach_task_cfs_rq(p);
@@ -8144,6 +10208,19 @@ static void task_move_group_fair(struct task_struct *p)
 	attach_task_cfs_rq(p);
 }
 
+static void task_change_group_fair(struct task_struct *p, int type)
+{
+	switch (type) {
+	case TASK_SET_GROUP:
+		task_set_group_fair(p);
+		break;
+
+	case TASK_MOVE_GROUP:
+		task_move_group_fair(p);
+		break;
+	}
+}
+
 void free_fair_sched_group(struct task_group *tg)
 {
 	int i;
@@ -8166,8 +10243,9 @@ void free_fair_sched_group(struct task_group *tg)
 
 int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent)
 {
-	struct cfs_rq *cfs_rq;
 	struct sched_entity *se;
+	struct cfs_rq *cfs_rq;
+	struct rq *rq;
 	int i;
 
 	tg->cfs_rq = kzalloc(sizeof(cfs_rq) * nr_cpu_ids, GFP_KERNEL);
@@ -8182,6 +10260,8 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent)
 	init_cfs_bandwidth(tg_cfs_bandwidth(tg));
 
 	for_each_possible_cpu(i) {
+		rq = cpu_rq(i);
+
 		cfs_rq = kzalloc_node(sizeof(struct cfs_rq),
 				      GFP_KERNEL, cpu_to_node(i));
 		if (!cfs_rq)
@@ -8195,6 +10275,10 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent)
 		init_cfs_rq(cfs_rq);
 		init_tg_cfs_entry(tg, cfs_rq, se, i, parent->se[i]);
 		init_entity_runnable_average(se);
+
+		raw_spin_lock_irq(&rq->lock);
+		post_init_entity_util_avg(se);
+		raw_spin_unlock_irq(&rq->lock);
 	}
 
 	return 1;
@@ -8283,8 +10367,10 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares)
 
 		/* Possible calls to update_curr() need rq clock */
 		update_rq_clock(rq);
-		for_each_sched_entity(se)
-			update_cfs_shares(group_cfs_rq(se));
+		for_each_sched_entity(se) {
+			update_load_avg(se, UPDATE_TG);
+			update_cfs_shares(se);
+		}
 		raw_spin_unlock_irqrestore(&rq->lock, flags);
 	}
 
@@ -8361,7 +10447,7 @@ const struct sched_class fair_sched_class = {
 	.update_curr		= update_curr_fair,
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
-	.task_move_group	= task_move_group_fair,
+	.task_change_group	= task_change_group_fair,
 #endif
 };
 
diff --git a/kernel/sched/features.h b/kernel/sched/features.h
index 69631fa46c2f..55e461055332 100644
--- a/kernel/sched/features.h
+++ b/kernel/sched/features.h
@@ -69,3 +69,12 @@ SCHED_FEAT(RT_RUNTIME_SHARE, true)
 SCHED_FEAT(LB_MIN, false)
 SCHED_FEAT(ATTACH_AGE_LOAD, true)
 
+/*
+ * Energy aware scheduling. Use platform energy model to guide scheduling
+ * decisions optimizing for energy efficiency.
+ */
+#ifdef CONFIG_DEFAULT_USE_ENERGY_AWARE
+SCHED_FEAT(ENERGY_AWARE, true)
+#else
+SCHED_FEAT(ENERGY_AWARE, false)
+#endif
diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c
index 4a2ef5a02fd3..917c94abf5bb 100644
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -19,9 +19,10 @@
  * sched_idle_set_state - Record idle state for the current CPU.
  * @idle_state: State to record.
  */
-void sched_idle_set_state(struct cpuidle_state *idle_state)
+void sched_idle_set_state(struct cpuidle_state *idle_state, int index)
 {
 	idle_set_state(this_rq(), idle_state);
+	idle_set_state_idx(this_rq(), index);
 }
 
 static int __read_mostly cpu_idle_force_poll;
@@ -219,6 +220,7 @@ static void cpu_idle_loop(void)
 		 */
 
 		__current_set_polling();
+		quiet_vmstat();
 		tick_nohz_idle_enter();
 
 		while (!need_resched()) {
diff --git a/kernel/sched/idle_task.c b/kernel/sched/idle_task.c
index c4ae0f1fdf9b..33d7003fa1b8 100644
--- a/kernel/sched/idle_task.c
+++ b/kernel/sched/idle_task.c
@@ -9,7 +9,8 @@
 
 #ifdef CONFIG_SMP
 static int
-select_task_rq_idle(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_idle(struct task_struct *p, int cpu, int sd_flag, int flags,
+		    int sibling_count_hint)
 {
 	return task_cpu(p); /* IDLE tasks as never migrated */
 }
diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
index 78ae5c1d9412..c8322ab130eb 100644
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -8,6 +8,9 @@
 #include <linux/slab.h>
 #include <linux/irq_work.h>
 
+#include "walt.h"
+#include "tune.h"
+
 int sched_rr_timeslice = RR_TIMESLICE;
 
 static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun);
@@ -889,6 +892,51 @@ static inline int rt_se_prio(struct sched_rt_entity *rt_se)
 	return rt_task_of(rt_se)->prio;
 }
 
+static void dump_throttled_rt_tasks(struct rt_rq *rt_rq)
+{
+	struct rt_prio_array *array = &rt_rq->active;
+	struct sched_rt_entity *rt_se;
+	char buf[500];
+	char *pos = buf;
+	char *end = buf + sizeof(buf);
+	int idx;
+
+	pos += snprintf(pos, sizeof(buf),
+		"sched: RT throttling activated for rt_rq %p (cpu %d)\n",
+		rt_rq, cpu_of(rq_of_rt_rq(rt_rq)));
+
+	if (bitmap_empty(array->bitmap, MAX_RT_PRIO))
+		goto out;
+
+	pos += snprintf(pos, end - pos, "potential CPU hogs:\n");
+	idx = sched_find_first_bit(array->bitmap);
+	while (idx < MAX_RT_PRIO) {
+		list_for_each_entry(rt_se, array->queue + idx, run_list) {
+			struct task_struct *p;
+
+			if (!rt_entity_is_task(rt_se))
+				continue;
+
+			p = rt_task_of(rt_se);
+			if (pos < end)
+				pos += snprintf(pos, end - pos, "\t%s (%d)\n",
+					p->comm, p->pid);
+		}
+		idx = find_next_bit(array->bitmap, MAX_RT_PRIO, idx + 1);
+	}
+out:
+#ifdef CONFIG_PANIC_ON_RT_THROTTLING
+	/*
+	 * Use pr_err() in the BUG() case since printk_sched() will
+	 * not get flushed and deadlock is not a concern.
+	 */
+	pr_err("%s", buf);
+	BUG();
+#else
+	printk_deferred("%s", buf);
+#endif
+}
+
 static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
 {
 	u64 runtime = sched_rt_runtime(rt_rq);
@@ -912,8 +960,14 @@ static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq)
 		 * but accrue some time due to boosting.
 		 */
 		if (likely(rt_b->rt_runtime)) {
+			static bool once = false;
+
 			rt_rq->rt_throttled = 1;
-			printk_deferred_once("sched: RT throttling activated\n");
+
+			if (!once) {
+				once = true;
+				dump_throttled_rt_tasks(rt_rq);
+			}
 		} else {
 			/*
 			 * In case we did anyway, make it go away,
@@ -949,6 +1003,9 @@ static void update_curr_rt(struct rq *rq)
 	if (unlikely((s64)delta_exec <= 0))
 		return;
 
+	/* Kick cpufreq (see the comment in kernel/sched/sched.h). */
+	cpufreq_update_this_cpu(rq, SCHED_CPUFREQ_RT);
+
 	schedstat_set(curr->se.statistics.exec_max,
 		      max(curr->se.statistics.exec_max, delta_exec));
 
@@ -1261,9 +1318,12 @@ enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags)
 		rt_se->timeout = 0;
 
 	enqueue_rt_entity(rt_se, flags & ENQUEUE_HEAD);
+	walt_inc_cumulative_runnable_avg(rq, p);
 
 	if (!task_current(rq, p) && p->nr_cpus_allowed > 1)
 		enqueue_pushable_task(rq, p);
+
+	schedtune_enqueue_task(p, cpu_of(rq));
 }
 
 static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int flags)
@@ -1272,8 +1332,10 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int flags)
 
 	update_curr_rt(rq);
 	dequeue_rt_entity(rt_se);
+	walt_dec_cumulative_runnable_avg(rq, p);
 
 	dequeue_pushable_task(rq, p);
+	schedtune_dequeue_task(p, cpu_of(rq));
 }
 
 /*
@@ -1314,7 +1376,8 @@ static void yield_task_rt(struct rq *rq)
 static int find_lowest_rq(struct task_struct *task);
 
 static int
-select_task_rq_rt(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_rt(struct task_struct *p, int cpu, int sd_flag, int flags,
+		  int sibling_count_hint)
 {
 	struct task_struct *curr;
 	struct rq *rq;
@@ -1780,7 +1843,9 @@ retry:
 	}
 
 	deactivate_task(rq, next_task, 0);
+	next_task->on_rq = TASK_ON_RQ_MIGRATING;
 	set_task_cpu(next_task, lowest_rq->cpu);
+	next_task->on_rq = TASK_ON_RQ_QUEUED;
 	activate_task(lowest_rq, next_task, 0);
 	ret = 1;
 
@@ -2034,7 +2099,9 @@ static void pull_rt_task(struct rq *this_rq)
 			resched = true;
 
 			deactivate_task(src_rq, p, 0);
+			p->on_rq = TASK_ON_RQ_MIGRATING;
 			set_task_cpu(p, this_cpu);
+			p->on_rq = TASK_ON_RQ_QUEUED;
 			activate_task(this_rq, p, 0);
 			/*
 			 * We continue with the search, just in
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 55d92a1ca070..203d64a0c947 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -31,8 +31,10 @@ extern long calc_load_fold_active(struct rq *this_rq);
 
 #ifdef CONFIG_SMP
 extern void update_cpu_load_active(struct rq *this_rq);
+extern void check_for_migration(struct rq *rq, struct task_struct *p);
 #else
 static inline void update_cpu_load_active(struct rq *this_rq) { }
+static inline void check_for_migration(struct rq *rq, struct task_struct *p) { }
 #endif
 
 /*
@@ -335,7 +337,15 @@ extern void sched_move_task(struct task_struct *tsk);
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
 extern int sched_group_set_shares(struct task_group *tg, unsigned long shares);
-#endif
+
+#ifdef CONFIG_SMP
+extern void set_task_rq_fair(struct sched_entity *se,
+			     struct cfs_rq *prev, struct cfs_rq *next);
+#else /* !CONFIG_SMP */
+static inline void set_task_rq_fair(struct sched_entity *se,
+			     struct cfs_rq *prev, struct cfs_rq *next) { }
+#endif /* CONFIG_SMP */
+#endif /* CONFIG_FAIR_GROUP_SCHED */
 
 #else /* CONFIG_CGROUP_SCHED */
 
@@ -376,6 +386,7 @@ struct cfs_rq {
 	unsigned long runnable_load_avg;
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	unsigned long tg_load_avg_contrib;
+	unsigned long propagate_avg;
 #endif
 	atomic_long_t removed_load_avg, removed_util_avg;
 #ifndef CONFIG_64BIT
@@ -410,6 +421,10 @@ struct cfs_rq {
 	struct list_head leaf_cfs_rq_list;
 	struct task_group *tg;	/* group that "owns" this runqueue */
 
+#ifdef CONFIG_SCHED_WALT
+	u64 cumulative_runnable_avg;
+#endif
+
 #ifdef CONFIG_CFS_BANDWIDTH
 	int runtime_enabled;
 	u64 runtime_expires;
@@ -506,10 +521,18 @@ struct dl_rq {
 #else
 	struct dl_bw dl_bw;
 #endif
+	/* This is the "average utilization" for this runqueue */
+	s64 avg_bw;
 };
 
 #ifdef CONFIG_SMP
 
+struct max_cpu_capacity {
+	raw_spinlock_t lock;
+	unsigned long val;
+	int cpu;
+};
+
 /*
  * We add the notion of a root-domain which will be used to define per-domain
  * variables. Each exclusive cpuset essentially defines an island domain by
@@ -528,6 +551,9 @@ struct root_domain {
 	/* Indicate more than one runnable task for any CPU */
 	bool overload;
 
+	/* Indicate one or more cpus over-utilized (tipping point) */
+	bool overutilized;
+
 	/*
 	 * The bit corresponding to a CPU gets set here if such CPU has more
 	 * than one runnable -deadline task (as it is below for RT tasks).
@@ -543,6 +569,12 @@ struct root_domain {
 	 */
 	cpumask_var_t rto_mask;
 	struct cpupri cpupri;
+
+	/* Maximum cpu capacity in the system. */
+	struct max_cpu_capacity max_cpu_capacity;
+
+	/* First cpu with maximum and minimum original capacity */
+	int max_cap_orig_cpu, min_cap_orig_cpu;
 };
 
 extern struct root_domain def_root_domain;
@@ -572,6 +604,7 @@ struct rq {
 	#define CPU_LOAD_IDX_MAX 5
 	unsigned long cpu_load[CPU_LOAD_IDX_MAX];
 	unsigned long last_load_update_tick;
+	unsigned int misfit_task;
 #ifdef CONFIG_NO_HZ_COMMON
 	u64 nohz_stamp;
 	unsigned long nohz_flags;
@@ -579,6 +612,14 @@ struct rq {
 #ifdef CONFIG_NO_HZ_FULL
 	unsigned long last_sched_tick;
 #endif
+
+#ifdef CONFIG_CPU_QUIET
+	/* time-based average load */
+	u64 nr_last_stamp;
+	u64 nr_running_integral;
+	seqcount_t ave_seqcnt;
+#endif
+
 	/* capture load from *all* tasks on this cpu: */
 	struct load_weight load;
 	unsigned long nr_load_updates;
@@ -591,6 +632,7 @@ struct rq {
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	/* list of leaf cfs_rq on this cpu: */
 	struct list_head leaf_cfs_rq_list;
+	struct list_head *tmp_alone_branch;
 #endif /* CONFIG_FAIR_GROUP_SCHED */
 
 	/*
@@ -624,6 +666,7 @@ struct rq {
 	/* For active balancing */
 	int active_balance;
 	int push_cpu;
+	struct task_struct *push_task;
 	struct cpu_stop_work active_balance_work;
 	/* cpu of this runqueue: */
 	int cpu;
@@ -640,6 +683,20 @@ struct rq {
 	u64 max_idle_balance_cost;
 #endif
 
+#ifdef CONFIG_SCHED_WALT
+	u64 cumulative_runnable_avg;
+	u64 window_start;
+	u64 curr_runnable_sum;
+	u64 prev_runnable_sum;
+	u64 nt_curr_runnable_sum;
+	u64 nt_prev_runnable_sum;
+	u64 cur_irqload;
+	u64 avg_irqload;
+	u64 irqload_ts;
+	u64 cum_window_demand;
+#endif /* CONFIG_SCHED_WALT */
+
+
 #ifdef CONFIG_IRQ_TIME_ACCOUNTING
 	u64 prev_irq_time;
 #endif
@@ -678,6 +735,9 @@ struct rq {
 	/* try_to_wake_up() stats */
 	unsigned int ttwu_count;
 	unsigned int ttwu_local;
+#ifdef CONFIG_SMP
+	struct eas_stats eas_stats;
+#endif
 #endif
 
 #ifdef CONFIG_SMP
@@ -687,6 +747,7 @@ struct rq {
 #ifdef CONFIG_CPU_IDLE
 	/* Must be inspected within a rcu lock section */
 	struct cpuidle_state *idle_state;
+	int idle_state_idx;
 #endif
 };
 
@@ -836,6 +897,8 @@ DECLARE_PER_CPU(int, sd_llc_id);
 DECLARE_PER_CPU(struct sched_domain *, sd_numa);
 DECLARE_PER_CPU(struct sched_domain *, sd_busy);
 DECLARE_PER_CPU(struct sched_domain *, sd_asym);
+DECLARE_PER_CPU(struct sched_domain *, sd_ea);
+DECLARE_PER_CPU(struct sched_domain *, sd_scs);
 
 struct sched_group_capacity {
 	atomic_t ref;
@@ -843,7 +906,9 @@ struct sched_group_capacity {
 	 * CPU capacity of this group, SCHED_LOAD_SCALE being max capacity
 	 * for a single CPU.
 	 */
-	unsigned int capacity;
+	unsigned long capacity;
+	unsigned long max_capacity; /* Max per-cpu capacity in group */
+	unsigned long min_capacity; /* Min per-CPU capacity in group */
 	unsigned long next_update;
 	int imbalance; /* XXX unrelated to capacity but shared group state */
 	/*
@@ -860,6 +925,7 @@ struct sched_group {
 
 	unsigned int group_weight;
 	struct sched_group_capacity *sgc;
+	const struct sched_group_energy *sge;
 
 	/*
 	 * The CPUs this group covers.
@@ -933,6 +999,7 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu)
 #endif
 
 #ifdef CONFIG_FAIR_GROUP_SCHED
+	set_task_rq_fair(&p->se, p->se.cfs_rq, tg->cfs_rq[cpu]);
 	p->se.cfs_rq = tg->cfs_rq[cpu];
 	p->se.parent = tg->se[cpu];
 #endif
@@ -963,7 +1030,11 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
 	 * per-task data have been completed by this moment.
 	 */
 	smp_wmb();
+#ifdef CONFIG_THREAD_INFO_IN_TASK
+	p->cpu = cpu;
+#else
 	task_thread_info(p)->cpu = cpu;
+#endif
 	p->wake_cpu = cpu;
 #endif
 }
@@ -1163,6 +1234,7 @@ static const u32 prio_to_wmult[40] = {
 #endif
 #define ENQUEUE_REPLENISH	0x08
 #define ENQUEUE_RESTORE	0x10
+#define ENQUEUE_WAKEUP_NEW	0x20
 
 #define DEQUEUE_SLEEP		0x01
 #define DEQUEUE_SAVE		0x02
@@ -1192,7 +1264,8 @@ struct sched_class {
 	void (*put_prev_task) (struct rq *rq, struct task_struct *p);
 
 #ifdef CONFIG_SMP
-	int  (*select_task_rq)(struct task_struct *p, int task_cpu, int sd_flag, int flags);
+	int  (*select_task_rq)(struct task_struct *p, int task_cpu, int sd_flag, int flags,
+			       int subling_count_hint);
 	void (*migrate_task_rq)(struct task_struct *p);
 
 	void (*task_waking) (struct task_struct *task);
@@ -1225,8 +1298,11 @@ struct sched_class {
 
 	void (*update_curr) (struct rq *rq);
 
+#define TASK_SET_GROUP  0
+#define TASK_MOVE_GROUP	1
+
 #ifdef CONFIG_FAIR_GROUP_SCHED
-	void (*task_move_group) (struct task_struct *p);
+	void (*task_change_group)(struct task_struct *p, int type);
 #endif
 };
 
@@ -1248,6 +1324,7 @@ extern const struct sched_class idle_sched_class;
 
 #ifdef CONFIG_SMP
 
+extern void init_max_cpu_capacity(struct max_cpu_capacity *mcc);
 extern void update_group_capacity(struct sched_domain *sd, int cpu);
 
 extern void trigger_load_balance(struct rq *rq);
@@ -1276,6 +1353,17 @@ static inline struct cpuidle_state *idle_get_state(struct rq *rq)
 	WARN_ON(!rcu_read_lock_held());
 	return rq->idle_state;
 }
+
+static inline void idle_set_state_idx(struct rq *rq, int idle_state_idx)
+{
+	rq->idle_state_idx = idle_state_idx;
+}
+
+static inline int idle_get_state_idx(struct rq *rq)
+{
+	WARN_ON(!rcu_read_lock_held());
+	return rq->idle_state_idx;
+}
 #else
 static inline void idle_set_state(struct rq *rq,
 				  struct cpuidle_state *idle_state)
@@ -1286,6 +1374,15 @@ static inline struct cpuidle_state *idle_get_state(struct rq *rq)
 {
 	return NULL;
 }
+
+static inline void idle_set_state_idx(struct rq *rq, int idle_state_idx)
+{
+}
+
+static inline int idle_get_state_idx(struct rq *rq)
+{
+	return -1;
+}
 #endif
 
 extern void sysrq_sched_debug_show(void);
@@ -1309,8 +1406,9 @@ extern void init_dl_task_timer(struct sched_dl_entity *dl_se);
 unsigned long to_ratio(u64 period, u64 runtime);
 
 extern void init_entity_runnable_average(struct sched_entity *se);
+extern void post_init_entity_util_avg(struct sched_entity *se);
 
-static inline void add_nr_running(struct rq *rq, unsigned count)
+static inline void __add_nr_running(struct rq *rq, unsigned count)
 {
 	unsigned prev_nr = rq->nr_running;
 
@@ -1338,11 +1436,48 @@ static inline void add_nr_running(struct rq *rq, unsigned count)
 	}
 }
 
-static inline void sub_nr_running(struct rq *rq, unsigned count)
+static inline void __sub_nr_running(struct rq *rq, unsigned count)
 {
 	rq->nr_running -= count;
 }
 
+#ifdef CONFIG_CPU_QUIET
+#define NR_AVE_SCALE(x)		((x) << FSHIFT)
+static inline u64 do_nr_running_integral(struct rq *rq)
+{
+	s64 nr, deltax;
+	u64 nr_running_integral = rq->nr_running_integral;
+
+	deltax = rq->clock_task - rq->nr_last_stamp;
+	nr = NR_AVE_SCALE(rq->nr_running);
+
+	nr_running_integral += nr * deltax;
+
+	return nr_running_integral;
+}
+
+static inline void add_nr_running(struct rq *rq, unsigned count)
+{
+	write_seqcount_begin(&rq->ave_seqcnt);
+	rq->nr_running_integral = do_nr_running_integral(rq);
+	rq->nr_last_stamp = rq->clock_task;
+	__add_nr_running(rq, count);
+	write_seqcount_end(&rq->ave_seqcnt);
+}
+
+static inline void sub_nr_running(struct rq *rq, unsigned count)
+{
+	write_seqcount_begin(&rq->ave_seqcnt);
+	rq->nr_running_integral = do_nr_running_integral(rq);
+	rq->nr_last_stamp = rq->clock_task;
+	__sub_nr_running(rq, count);
+	write_seqcount_end(&rq->ave_seqcnt);
+}
+#else
+#define add_nr_running __add_nr_running
+#define sub_nr_running __sub_nr_running
+#endif
+
 static inline void rq_last_tick_reset(struct rq *rq)
 {
 #ifdef CONFIG_NO_HZ_FULL
@@ -1415,10 +1550,87 @@ unsigned long arch_scale_cpu_capacity(struct sched_domain *sd, int cpu)
 }
 #endif
 
+#ifdef CONFIG_SMP
+static inline unsigned long capacity_of(int cpu)
+{
+	return cpu_rq(cpu)->cpu_capacity;
+}
+
+static inline unsigned long capacity_orig_of(int cpu)
+{
+	return cpu_rq(cpu)->cpu_capacity_orig;
+}
+
+extern unsigned int sysctl_sched_use_walt_cpu_util;
+extern unsigned int walt_ravg_window;
+extern bool walt_disabled;
+
+/*
+ * cpu_util returns the amount of capacity of a CPU that is used by CFS
+ * tasks. The unit of the return value must be the one of capacity so we can
+ * compare the utilization with the capacity of the CPU that is available for
+ * CFS task (ie cpu_capacity).
+ *
+ * cfs_rq.avg.util_avg is the sum of running time of runnable tasks plus the
+ * recent utilization of currently non-runnable tasks on a CPU. It represents
+ * the amount of utilization of a CPU in the range [0..capacity_orig] where
+ * capacity_orig is the cpu_capacity available at the highest frequency
+ * (arch_scale_freq_capacity()).
+ * The utilization of a CPU converges towards a sum equal to or less than the
+ * current capacity (capacity_curr <= capacity_orig) of the CPU because it is
+ * the running time on this CPU scaled by capacity_curr.
+ *
+ * Nevertheless, cfs_rq.avg.util_avg can be higher than capacity_curr or even
+ * higher than capacity_orig because of unfortunate rounding in
+ * cfs.avg.util_avg or just after migrating tasks and new task wakeups until
+ * the average stabilizes with the new running time. We need to check that the
+ * utilization stays within the range of [0..capacity_orig] and cap it if
+ * necessary. Without utilization capping, a group could be seen as overloaded
+ * (CPU0 utilization at 121% + CPU1 utilization at 80%) whereas CPU1 has 20% of
+ * available capacity. We allow utilization to overshoot capacity_curr (but not
+ * capacity_orig) as it useful for predicting the capacity required after task
+ * migrations (scheduler-driven DVFS).
+ */
+static inline unsigned long __cpu_util(int cpu, int delta)
+{
+	unsigned long util = cpu_rq(cpu)->cfs.avg.util_avg;
+	unsigned long capacity = capacity_orig_of(cpu);
+
+#ifdef CONFIG_SCHED_WALT
+	if (!walt_disabled && sysctl_sched_use_walt_cpu_util)
+		util = div64_u64(cpu_rq(cpu)->cumulative_runnable_avg,
+				 walt_ravg_window >> SCHED_LOAD_SHIFT);
+#endif
+	delta += util;
+	if (delta < 0)
+		return 0;
+
+	return (delta >= capacity) ? capacity : delta;
+}
+
+static inline unsigned long cpu_util(int cpu)
+{
+	return __cpu_util(cpu, 0);
+}
+
+static inline unsigned long cpu_util_freq(int cpu)
+{
+	unsigned long util = cpu_rq(cpu)->cfs.avg.util_avg;
+	unsigned long capacity = capacity_orig_of(cpu);
+
+#ifdef CONFIG_SCHED_WALT
+	if (!walt_disabled && sysctl_sched_use_walt_cpu_util)
+		util = div64_u64(cpu_rq(cpu)->prev_runnable_sum,
+				 walt_ravg_window >> SCHED_LOAD_SHIFT);
+#endif
+	return (util >= capacity) ? capacity : util;
+}
+
+#endif
+
 static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta)
 {
 	rq->rt_avg += rt_delta * arch_scale_freq_capacity(NULL, cpu_of(rq));
-	sched_avg_update(rq);
 }
 #else
 static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { }
@@ -1507,6 +1719,9 @@ task_rq_unlock(struct rq *rq, struct task_struct *p, unsigned long *flags)
 	raw_spin_unlock_irqrestore(&p->pi_lock, *flags);
 }
 
+extern struct rq *lock_rq_of(struct task_struct *p, unsigned long *flags);
+extern void unlock_rq_of(struct rq *rq, struct task_struct *p, unsigned long *flags);
+
 #ifdef CONFIG_SMP
 #ifdef CONFIG_PREEMPT
 
@@ -1579,7 +1794,8 @@ static inline int double_lock_balance(struct rq *this_rq, struct rq *busiest)
 static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest)
 	__releases(busiest->lock)
 {
-	raw_spin_unlock(&busiest->lock);
+	if (this_rq != busiest)
+		raw_spin_unlock(&busiest->lock);
 	lock_set_subclass(&this_rq->lock.dep_map, 0, _RET_IP_);
 }
 
@@ -1770,3 +1986,66 @@ static inline u64 irq_time_read(int cpu)
 }
 #endif /* CONFIG_64BIT */
 #endif /* CONFIG_IRQ_TIME_ACCOUNTING */
+
+#ifdef CONFIG_CPU_FREQ
+DECLARE_PER_CPU(struct update_util_data *, cpufreq_update_util_data);
+
+/**
+ * cpufreq_update_util - Take a note about CPU utilization changes.
+ * @rq: Runqueue to carry out the update for.
+ * @flags: Update reason flags.
+ *
+ * This function is called by the scheduler on the CPU whose utilization is
+ * being updated.
+ *
+ * It can only be called from RCU-sched read-side critical sections.
+ *
+ * The way cpufreq is currently arranged requires it to evaluate the CPU
+ * performance state (frequency/voltage) on a regular basis to prevent it from
+ * being stuck in a completely inadequate performance level for too long.
+ * That is not guaranteed to happen if the updates are only triggered from CFS,
+ * though, because they may not be coming in if RT or deadline tasks are active
+ * all the time (or there are RT and DL tasks only).
+ *
+ * As a workaround for that issue, this function is called by the RT and DL
+ * sched classes to trigger extra cpufreq updates to prevent it from stalling,
+ * but that really is a band-aid.  Going forward it should be replaced with
+ * solutions targeted more specifically at RT and DL tasks.
+ */
+static inline void cpufreq_update_util(struct rq *rq, unsigned int flags)
+{
+        struct update_util_data *data;
+
+        data = rcu_dereference_sched(*this_cpu_ptr(&cpufreq_update_util_data));
+        if (data)
+                data->func(data, rq_clock(rq), flags);
+}
+
+static inline void cpufreq_update_this_cpu(struct rq *rq, unsigned int flags)
+{
+        if (cpu_of(rq) == smp_processor_id())
+                cpufreq_update_util(rq, flags);
+}
+#else
+static inline void cpufreq_update_util(struct rq *rq, unsigned int flags) {}
+static inline void cpufreq_update_this_cpu(struct rq *rq, unsigned int flags) {}
+#endif /* CONFIG_CPU_FREQ */
+
+#ifdef CONFIG_SCHED_WALT
+
+static inline bool
+walt_task_in_cum_window_demand(struct rq *rq, struct task_struct *p)
+{
+	return cpu_of(rq) == task_cpu(p) &&
+	       (p->on_rq || p->last_sleep_ts >= rq->window_start);
+}
+
+#endif /* CONFIG_SCHED_WALT */
+
+#ifdef arch_scale_freq_capacity
+#ifndef arch_scale_freq_invariant
+#define arch_scale_freq_invariant()     (true)
+#endif
+#else /* arch_scale_freq_capacity */
+#define arch_scale_freq_invariant()     (false)
+#endif
diff --git a/kernel/sched/stats.c b/kernel/sched/stats.c
index 87e2c9f0c33e..6d74a7c77c8c 100644
--- a/kernel/sched/stats.c
+++ b/kernel/sched/stats.c
@@ -12,6 +12,28 @@
  */
 #define SCHEDSTAT_VERSION 15
 
+#ifdef CONFIG_SMP
+static inline void show_easstat(struct seq_file *seq, struct eas_stats *stats)
+{
+	/* eas-specific runqueue stats */
+	seq_printf(seq, "eas %llu %llu %llu %llu %llu %llu ",
+	    stats->sis_attempts, stats->sis_idle, stats->sis_cache_affine,
+	    stats->sis_suff_cap, stats->sis_idle_cpu, stats->sis_count);
+
+	seq_printf(seq, "%llu %llu %llu %llu %llu %llu %llu ",
+	    stats->secb_attempts, stats->secb_sync, stats->secb_idle_bt,
+	    stats->secb_insuff_cap, stats->secb_no_nrg_sav,
+	    stats->secb_nrg_sav, stats->secb_count);
+
+	seq_printf(seq, "%llu %llu %llu %llu %llu ",
+	    stats->fbt_attempts, stats->fbt_no_cpu, stats->fbt_no_sd,
+	    stats->fbt_pref_idle, stats->fbt_count);
+
+	seq_printf(seq, "%llu %llu\n",
+	    stats->cas_attempts, stats->cas_count);
+}
+#endif
+
 static int show_schedstat(struct seq_file *seq, void *v)
 {
 	int cpu;
@@ -40,6 +62,8 @@ static int show_schedstat(struct seq_file *seq, void *v)
 		seq_printf(seq, "\n");
 
 #ifdef CONFIG_SMP
+		show_easstat(seq, &rq->eas_stats);
+
 		/* domain-specific stats */
 		rcu_read_lock();
 		for_each_domain(cpu, sd) {
@@ -66,6 +90,8 @@ static int show_schedstat(struct seq_file *seq, void *v)
 			    sd->sbf_count, sd->sbf_balanced, sd->sbf_pushed,
 			    sd->ttwu_wake_remote, sd->ttwu_move_affine,
 			    sd->ttwu_move_balance);
+
+			show_easstat(seq, &sd->eas_stats);
 		}
 		rcu_read_unlock();
 #endif
diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c
index cbc67da10954..a5567ccd8803 100644
--- a/kernel/sched/stop_task.c
+++ b/kernel/sched/stop_task.c
@@ -1,4 +1,5 @@
 #include "sched.h"
+#include "walt.h"
 
 /*
  * stop-task scheduling class.
@@ -11,7 +12,8 @@
 
 #ifdef CONFIG_SMP
 static int
-select_task_rq_stop(struct task_struct *p, int cpu, int sd_flag, int flags)
+select_task_rq_stop(struct task_struct *p, int cpu, int sd_flag, int flags,
+		    int sibling_count_hint)
 {
 	return task_cpu(p); /* stop tasks as never migrate */
 }
@@ -42,12 +44,14 @@ static void
 enqueue_task_stop(struct rq *rq, struct task_struct *p, int flags)
 {
 	add_nr_running(rq, 1);
+	walt_inc_cumulative_runnable_avg(rq, p);
 }
 
 static void
 dequeue_task_stop(struct rq *rq, struct task_struct *p, int flags)
 {
 	sub_nr_running(rq, 1);
+	walt_dec_cumulative_runnable_avg(rq, p);
 }
 
 static void yield_task_stop(struct rq *rq)
diff --git a/kernel/sched/tune.c b/kernel/sched/tune.c
new file mode 100644
index 000000000000..d444fc1a4d58
--- /dev/null
+++ b/kernel/sched/tune.c
@@ -0,0 +1,955 @@
+#include <linux/cgroup.h>
+#include <linux/err.h>
+#include <linux/kernel.h>
+#include <linux/percpu.h>
+#include <linux/printk.h>
+#include <linux/rcupdate.h>
+#include <linux/slab.h>
+
+#include <trace/events/sched.h>
+
+#include "sched.h"
+#include "tune.h"
+
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+bool schedtune_initialized = false;
+#endif
+
+unsigned int sysctl_sched_cfs_boost __read_mostly;
+
+extern struct reciprocal_value schedtune_spc_rdiv;
+extern struct target_nrg schedtune_target_nrg;
+
+/* Performance Boost region (B) threshold params */
+static int perf_boost_idx;
+
+/* Performance Constraint region (C) threshold params */
+static int perf_constrain_idx;
+
+/**
+ * Performance-Energy (P-E) Space thresholds constants
+ */
+struct threshold_params {
+	int nrg_gain;
+	int cap_gain;
+};
+
+/*
+ * System specific P-E space thresholds constants
+ */
+static struct threshold_params
+threshold_gains[] = {
+	{ 0, 5 }, /*   < 10% */
+	{ 1, 5 }, /*   < 20% */
+	{ 2, 5 }, /*   < 30% */
+	{ 3, 5 }, /*   < 40% */
+	{ 4, 5 }, /*   < 50% */
+	{ 5, 4 }, /*   < 60% */
+	{ 5, 3 }, /*   < 70% */
+	{ 5, 2 }, /*   < 80% */
+	{ 5, 1 }, /*   < 90% */
+	{ 5, 0 }  /* <= 100% */
+};
+
+static int
+__schedtune_accept_deltas(int nrg_delta, int cap_delta,
+			  int perf_boost_idx, int perf_constrain_idx)
+{
+	int payoff = -INT_MAX;
+	int gain_idx = -1;
+
+	/* Performance Boost (B) region */
+	if (nrg_delta >= 0 && cap_delta > 0)
+		gain_idx = perf_boost_idx;
+	/* Performance Constraint (C) region */
+	else if (nrg_delta < 0 && cap_delta <= 0)
+		gain_idx = perf_constrain_idx;
+
+	/* Default: reject schedule candidate */
+	if (gain_idx == -1)
+		return payoff;
+
+	/*
+	 * Evaluate "Performance Boost" vs "Energy Increase"
+	 *
+	 * - Performance Boost (B) region
+	 *
+	 *   Condition: nrg_delta > 0 && cap_delta > 0
+	 *   Payoff criteria:
+	 *     cap_gain / nrg_gain  < cap_delta / nrg_delta =
+	 *     cap_gain * nrg_delta < cap_delta * nrg_gain
+	 *   Note that since both nrg_gain and nrg_delta are positive, the
+	 *   inequality does not change. Thus:
+	 *
+	 *     payoff = (cap_delta * nrg_gain) - (cap_gain * nrg_delta)
+	 *
+	 * - Performance Constraint (C) region
+	 *
+	 *   Condition: nrg_delta < 0 && cap_delta < 0
+	 *   payoff criteria:
+	 *     cap_gain / nrg_gain  > cap_delta / nrg_delta =
+	 *     cap_gain * nrg_delta < cap_delta * nrg_gain
+	 *   Note that since nrg_gain > 0 while nrg_delta < 0, the
+	 *   inequality change. Thus:
+	 *
+	 *     payoff = (cap_delta * nrg_gain) - (cap_gain * nrg_delta)
+	 *
+	 * This means that, in case of same positive defined {cap,nrg}_gain
+	 * for both the B and C regions, we can use the same payoff formula
+	 * where a positive value represents the accept condition.
+	 */
+	payoff  = cap_delta * threshold_gains[gain_idx].nrg_gain;
+	payoff -= nrg_delta * threshold_gains[gain_idx].cap_gain;
+
+	return payoff;
+}
+
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+
+/*
+ * EAS scheduler tunables for task groups.
+ */
+
+/* SchdTune tunables for a group of tasks */
+struct schedtune {
+	/* SchedTune CGroup subsystem */
+	struct cgroup_subsys_state css;
+
+	/* Boost group allocated ID */
+	int idx;
+
+	/* Boost value for tasks on that SchedTune CGroup */
+	int boost;
+
+	/* Performance Boost (B) region threshold params */
+	int perf_boost_idx;
+
+	/* Performance Constraint (C) region threshold params */
+	int perf_constrain_idx;
+
+	/* Hint to bias scheduling of tasks on that SchedTune CGroup
+	 * towards idle CPUs */
+	int prefer_idle;
+};
+
+static inline struct schedtune *css_st(struct cgroup_subsys_state *css)
+{
+	return css ? container_of(css, struct schedtune, css) : NULL;
+}
+
+static inline struct schedtune *task_schedtune(struct task_struct *tsk)
+{
+	return css_st(task_css(tsk, schedtune_cgrp_id));
+}
+
+static inline struct schedtune *parent_st(struct schedtune *st)
+{
+	return css_st(st->css.parent);
+}
+
+/*
+ * SchedTune root control group
+ * The root control group is used to defined a system-wide boosting tuning,
+ * which is applied to all tasks in the system.
+ * Task specific boost tuning could be specified by creating and
+ * configuring a child control group under the root one.
+ * By default, system-wide boosting is disabled, i.e. no boosting is applied
+ * to tasks which are not into a child control group.
+ */
+static struct schedtune
+root_schedtune = {
+	.boost	= 0,
+	.perf_boost_idx = 0,
+	.perf_constrain_idx = 0,
+	.prefer_idle = 0,
+};
+
+int
+schedtune_accept_deltas(int nrg_delta, int cap_delta,
+			struct task_struct *task)
+{
+	struct schedtune *ct;
+	int perf_boost_idx;
+	int perf_constrain_idx;
+
+	/* Optimal (O) region */
+	if (nrg_delta < 0 && cap_delta > 0) {
+		trace_sched_tune_filter(nrg_delta, cap_delta, 0, 0, 1, 0);
+		return INT_MAX;
+	}
+
+	/* Suboptimal (S) region */
+	if (nrg_delta > 0 && cap_delta < 0) {
+		trace_sched_tune_filter(nrg_delta, cap_delta, 0, 0, -1, 5);
+		return -INT_MAX;
+	}
+
+	/* Get task specific perf Boost/Constraints indexes */
+	rcu_read_lock();
+	ct = task_schedtune(task);
+	perf_boost_idx = ct->perf_boost_idx;
+	perf_constrain_idx = ct->perf_constrain_idx;
+	rcu_read_unlock();
+
+	return __schedtune_accept_deltas(nrg_delta, cap_delta,
+			perf_boost_idx, perf_constrain_idx);
+}
+
+/*
+ * Maximum number of boost groups to support
+ * When per-task boosting is used we still allow only limited number of
+ * boost groups for two main reasons:
+ * 1. on a real system we usually have only few classes of workloads which
+ *    make sense to boost with different values (e.g. background vs foreground
+ *    tasks, interactive vs low-priority tasks)
+ * 2. a limited number allows for a simpler and more memory/time efficient
+ *    implementation especially for the computation of the per-CPU boost
+ *    value
+ */
+#define BOOSTGROUPS_COUNT 5
+
+/* Array of configured boostgroups */
+static struct schedtune *allocated_group[BOOSTGROUPS_COUNT] = {
+	&root_schedtune,
+	NULL,
+};
+
+/* SchedTune boost groups
+ * Keep track of all the boost groups which impact on CPU, for example when a
+ * CPU has two RUNNABLE tasks belonging to two different boost groups and thus
+ * likely with different boost values.
+ * Since on each system we expect only a limited number of boost groups, here
+ * we use a simple array to keep track of the metrics required to compute the
+ * maximum per-CPU boosting value.
+ */
+struct boost_groups {
+	/* Maximum boost value for all RUNNABLE tasks on a CPU */
+	bool idle;
+	int boost_max;
+	struct {
+		/* The boost for tasks on that boost group */
+		int boost;
+		/* Count of RUNNABLE tasks on that boost group */
+		unsigned tasks;
+	} group[BOOSTGROUPS_COUNT];
+	/* CPU's boost group locking */
+	raw_spinlock_t lock;
+};
+
+/* Boost groups affecting each CPU in the system */
+DEFINE_PER_CPU(struct boost_groups, cpu_boost_groups);
+
+static void
+schedtune_cpu_update(int cpu)
+{
+	struct boost_groups *bg;
+	int boost_max;
+	int idx;
+
+	bg = &per_cpu(cpu_boost_groups, cpu);
+
+	/* The root boost group is always active */
+	boost_max = bg->group[0].boost;
+	for (idx = 1; idx < BOOSTGROUPS_COUNT; ++idx) {
+		/*
+		 * A boost group affects a CPU only if it has
+		 * RUNNABLE tasks on that CPU
+		 */
+		if (bg->group[idx].tasks == 0)
+			continue;
+
+		boost_max = max(boost_max, bg->group[idx].boost);
+	}
+	/* Ensures boost_max is non-negative when all cgroup boost values
+	 * are neagtive. Avoids under-accounting of cpu capacity which may cause
+	 * task stacking and frequency spikes.*/
+	boost_max = max(boost_max, 0);
+	bg->boost_max = boost_max;
+}
+
+static int
+schedtune_boostgroup_update(int idx, int boost)
+{
+	struct boost_groups *bg;
+	int cur_boost_max;
+	int old_boost;
+	int cpu;
+
+	/* Update per CPU boost groups */
+	for_each_possible_cpu(cpu) {
+		bg = &per_cpu(cpu_boost_groups, cpu);
+
+		/*
+		 * Keep track of current boost values to compute the per CPU
+		 * maximum only when it has been affected by the new value of
+		 * the updated boost group
+		 */
+		cur_boost_max = bg->boost_max;
+		old_boost = bg->group[idx].boost;
+
+		/* Update the boost value of this boost group */
+		bg->group[idx].boost = boost;
+
+		/* Check if this update increase current max */
+		if (boost > cur_boost_max && bg->group[idx].tasks) {
+			bg->boost_max = boost;
+			trace_sched_tune_boostgroup_update(cpu, 1, bg->boost_max);
+			continue;
+		}
+
+		/* Check if this update has decreased current max */
+		if (cur_boost_max == old_boost && old_boost > boost) {
+			schedtune_cpu_update(cpu);
+			trace_sched_tune_boostgroup_update(cpu, -1, bg->boost_max);
+			continue;
+		}
+
+		trace_sched_tune_boostgroup_update(cpu, 0, bg->boost_max);
+	}
+
+	return 0;
+}
+
+#define ENQUEUE_TASK  1
+#define DEQUEUE_TASK -1
+
+static inline void
+schedtune_tasks_update(struct task_struct *p, int cpu, int idx, int task_count)
+{
+	struct boost_groups *bg = &per_cpu(cpu_boost_groups, cpu);
+	int tasks = bg->group[idx].tasks + task_count;
+
+	/* Update boosted tasks count while avoiding to make it negative */
+	bg->group[idx].tasks = max(0, tasks);
+
+	trace_sched_tune_tasks_update(p, cpu, tasks, idx,
+			bg->group[idx].boost, bg->boost_max);
+
+	/* Boost group activation or deactivation on that RQ */
+	if (tasks == 1 || tasks == 0)
+		schedtune_cpu_update(cpu);
+}
+
+/*
+ * NOTE: This function must be called while holding the lock on the CPU RQ
+ */
+void schedtune_enqueue_task(struct task_struct *p, int cpu)
+{
+	struct boost_groups *bg = &per_cpu(cpu_boost_groups, cpu);
+	unsigned long irq_flags;
+	struct schedtune *st;
+	int idx;
+
+	if (!unlikely(schedtune_initialized))
+		return;
+
+	/*
+	 * When a task is marked PF_EXITING by do_exit() it's going to be
+	 * dequeued and enqueued multiple times in the exit path.
+	 * Thus we avoid any further update, since we do not want to change
+	 * CPU boosting while the task is exiting.
+	 */
+	if (p->flags & PF_EXITING)
+		return;
+
+	/*
+	 * Boost group accouting is protected by a per-cpu lock and requires
+	 * interrupt to be disabled to avoid race conditions for example on
+	 * do_exit()::cgroup_exit() and task migration.
+	 */
+	raw_spin_lock_irqsave(&bg->lock, irq_flags);
+	rcu_read_lock();
+
+	st = task_schedtune(p);
+	idx = st->idx;
+
+	schedtune_tasks_update(p, cpu, idx, ENQUEUE_TASK);
+
+	rcu_read_unlock();
+	raw_spin_unlock_irqrestore(&bg->lock, irq_flags);
+}
+
+int schedtune_can_attach(struct cgroup_taskset *tset)
+{
+	struct task_struct *task;
+	struct cgroup_subsys_state *css;
+	struct boost_groups *bg;
+	unsigned long irq_flags;
+	unsigned int cpu;
+	struct rq *rq;
+	int src_bg; /* Source boost group index */
+	int dst_bg; /* Destination boost group index */
+	int tasks;
+
+	if (!unlikely(schedtune_initialized))
+		return 0;
+
+
+	cgroup_taskset_for_each(task, css, tset) {
+
+		/*
+		 * Lock the CPU's RQ the task is enqueued to avoid race
+		 * conditions with migration code while the task is being
+		 * accounted
+		 */
+		rq = lock_rq_of(task, &irq_flags);
+
+		if (!task->on_rq) {
+			unlock_rq_of(rq, task, &irq_flags);
+			continue;
+		}
+
+		/*
+		 * Boost group accouting is protected by a per-cpu lock and requires
+		 * interrupt to be disabled to avoid race conditions on...
+		 */
+		cpu = cpu_of(rq);
+		bg = &per_cpu(cpu_boost_groups, cpu);
+		raw_spin_lock(&bg->lock);
+
+		dst_bg = css_st(css)->idx;
+		src_bg = task_schedtune(task)->idx;
+
+		/*
+		 * Current task is not changing boostgroup, which can
+		 * happen when the new hierarchy is in use.
+		 */
+		if (unlikely(dst_bg == src_bg)) {
+			raw_spin_unlock(&bg->lock);
+			unlock_rq_of(rq, task, &irq_flags);
+			continue;
+		}
+
+		/*
+		 * This is the case of a RUNNABLE task which is switching its
+		 * current boost group.
+		 */
+
+		/* Move task from src to dst boost group */
+		tasks = bg->group[src_bg].tasks - 1;
+		bg->group[src_bg].tasks = max(0, tasks);
+		bg->group[dst_bg].tasks += 1;
+
+		raw_spin_unlock(&bg->lock);
+		unlock_rq_of(rq, task, &irq_flags);
+
+		/* Update CPU boost group */
+		if (bg->group[src_bg].tasks == 0 || bg->group[dst_bg].tasks == 1)
+			schedtune_cpu_update(task_cpu(task));
+
+	}
+
+	return 0;
+}
+
+void schedtune_cancel_attach(struct cgroup_taskset *tset)
+{
+	/* This can happen only if SchedTune controller is mounted with
+	 * other hierarchies ane one of them fails. Since usually SchedTune is
+	 * mouted on its own hierarcy, for the time being we do not implement
+	 * a proper rollback mechanism */
+	WARN(1, "SchedTune cancel attach not implemented");
+}
+
+/*
+ * NOTE: This function must be called while holding the lock on the CPU RQ
+ */
+void schedtune_dequeue_task(struct task_struct *p, int cpu)
+{
+	struct boost_groups *bg = &per_cpu(cpu_boost_groups, cpu);
+	unsigned long irq_flags;
+	struct schedtune *st;
+	int idx;
+
+	if (!unlikely(schedtune_initialized))
+		return;
+
+	/*
+	 * When a task is marked PF_EXITING by do_exit() it's going to be
+	 * dequeued and enqueued multiple times in the exit path.
+	 * Thus we avoid any further update, since we do not want to change
+	 * CPU boosting while the task is exiting.
+	 * The last dequeue is already enforce by the do_exit() code path
+	 * via schedtune_exit_task().
+	 */
+	if (p->flags & PF_EXITING)
+		return;
+
+	/*
+	 * Boost group accouting is protected by a per-cpu lock and requires
+	 * interrupt to be disabled to avoid race conditions on...
+	 */
+	raw_spin_lock_irqsave(&bg->lock, irq_flags);
+	rcu_read_lock();
+
+	st = task_schedtune(p);
+	idx = st->idx;
+
+	schedtune_tasks_update(p, cpu, idx, DEQUEUE_TASK);
+
+	rcu_read_unlock();
+	raw_spin_unlock_irqrestore(&bg->lock, irq_flags);
+}
+
+void schedtune_exit_task(struct task_struct *tsk)
+{
+	struct schedtune *st;
+	unsigned long irq_flags;
+	unsigned int cpu;
+	struct rq *rq;
+	int idx;
+
+	if (!unlikely(schedtune_initialized))
+		return;
+
+	rq = lock_rq_of(tsk, &irq_flags);
+	rcu_read_lock();
+
+	cpu = cpu_of(rq);
+	st = task_schedtune(tsk);
+	idx = st->idx;
+	schedtune_tasks_update(tsk, cpu, idx, DEQUEUE_TASK);
+
+	rcu_read_unlock();
+	unlock_rq_of(rq, tsk, &irq_flags);
+}
+
+int schedtune_cpu_boost(int cpu)
+{
+	struct boost_groups *bg;
+
+	bg = &per_cpu(cpu_boost_groups, cpu);
+	return bg->boost_max;
+}
+
+int schedtune_task_boost(struct task_struct *p)
+{
+	struct schedtune *st;
+	int task_boost;
+
+	if (!unlikely(schedtune_initialized))
+		return 0;
+
+	/* Get task boost value */
+	rcu_read_lock();
+	st = task_schedtune(p);
+	task_boost = st->boost;
+	rcu_read_unlock();
+
+	return task_boost;
+}
+
+int schedtune_prefer_idle(struct task_struct *p)
+{
+	struct schedtune *st;
+	int prefer_idle;
+
+	if (!unlikely(schedtune_initialized))
+		return 0;
+
+	/* Get prefer_idle value */
+	rcu_read_lock();
+	st = task_schedtune(p);
+	prefer_idle = st->prefer_idle;
+	rcu_read_unlock();
+
+	return prefer_idle;
+}
+
+static u64
+prefer_idle_read(struct cgroup_subsys_state *css, struct cftype *cft)
+{
+	struct schedtune *st = css_st(css);
+
+	return st->prefer_idle;
+}
+
+static int
+prefer_idle_write(struct cgroup_subsys_state *css, struct cftype *cft,
+	    u64 prefer_idle)
+{
+	struct schedtune *st = css_st(css);
+	st->prefer_idle = prefer_idle;
+
+	return 0;
+}
+
+static s64
+boost_read(struct cgroup_subsys_state *css, struct cftype *cft)
+{
+	struct schedtune *st = css_st(css);
+
+	return st->boost;
+}
+
+static int
+boost_write(struct cgroup_subsys_state *css, struct cftype *cft,
+	    s64 boost)
+{
+	struct schedtune *st = css_st(css);
+	unsigned threshold_idx;
+	int boost_pct;
+
+	if (boost < -100 || boost > 100)
+		return -EINVAL;
+	boost_pct = boost;
+
+	/*
+	 * Update threshold params for Performance Boost (B)
+	 * and Performance Constraint (C) regions.
+	 * The current implementatio uses the same cuts for both
+	 * B and C regions.
+	 */
+	threshold_idx = clamp(boost_pct, 0, 99) / 10;
+	st->perf_boost_idx = threshold_idx;
+	st->perf_constrain_idx = threshold_idx;
+
+	st->boost = boost;
+	if (css == &root_schedtune.css) {
+		sysctl_sched_cfs_boost = boost;
+		perf_boost_idx  = threshold_idx;
+		perf_constrain_idx  = threshold_idx;
+	}
+
+	/* Update CPU boost */
+	schedtune_boostgroup_update(st->idx, st->boost);
+
+	trace_sched_tune_config(st->boost);
+
+	return 0;
+}
+
+static struct cftype files[] = {
+	{
+		.name = "boost",
+		.read_s64 = boost_read,
+		.write_s64 = boost_write,
+	},
+	{
+		.name = "prefer_idle",
+		.read_u64 = prefer_idle_read,
+		.write_u64 = prefer_idle_write,
+	},
+	{ }	/* terminate */
+};
+
+static int
+schedtune_boostgroup_init(struct schedtune *st)
+{
+	struct boost_groups *bg;
+	int cpu;
+
+	/* Keep track of allocated boost groups */
+	allocated_group[st->idx] = st;
+
+	/* Initialize the per CPU boost groups */
+	for_each_possible_cpu(cpu) {
+		bg = &per_cpu(cpu_boost_groups, cpu);
+		bg->group[st->idx].boost = 0;
+		bg->group[st->idx].tasks = 0;
+	}
+
+	return 0;
+}
+
+static struct cgroup_subsys_state *
+schedtune_css_alloc(struct cgroup_subsys_state *parent_css)
+{
+	struct schedtune *st;
+	int idx;
+
+	if (!parent_css)
+		return &root_schedtune.css;
+
+	/* Allow only single level hierachies */
+	if (parent_css != &root_schedtune.css) {
+		pr_err("Nested SchedTune boosting groups not allowed\n");
+		return ERR_PTR(-ENOMEM);
+	}
+
+	/* Allow only a limited number of boosting groups */
+	for (idx = 1; idx < BOOSTGROUPS_COUNT; ++idx)
+		if (!allocated_group[idx])
+			break;
+	if (idx == BOOSTGROUPS_COUNT) {
+		pr_err("Trying to create more than %d SchedTune boosting groups\n",
+		       BOOSTGROUPS_COUNT);
+		return ERR_PTR(-ENOSPC);
+	}
+
+	st = kzalloc(sizeof(*st), GFP_KERNEL);
+	if (!st)
+		goto out;
+
+	/* Initialize per CPUs boost group support */
+	st->idx = idx;
+	if (schedtune_boostgroup_init(st))
+		goto release;
+
+	return &st->css;
+
+release:
+	kfree(st);
+out:
+	return ERR_PTR(-ENOMEM);
+}
+
+static void
+schedtune_boostgroup_release(struct schedtune *st)
+{
+	/* Reset this boost group */
+	schedtune_boostgroup_update(st->idx, 0);
+
+	/* Keep track of allocated boost groups */
+	allocated_group[st->idx] = NULL;
+}
+
+static void
+schedtune_css_free(struct cgroup_subsys_state *css)
+{
+	struct schedtune *st = css_st(css);
+
+	schedtune_boostgroup_release(st);
+	kfree(st);
+}
+
+struct cgroup_subsys schedtune_cgrp_subsys = {
+	.css_alloc	= schedtune_css_alloc,
+	.css_free	= schedtune_css_free,
+	.can_attach     = schedtune_can_attach,
+	.cancel_attach  = schedtune_cancel_attach,
+	.legacy_cftypes	= files,
+	.early_init	= 1,
+};
+
+static inline void
+schedtune_init_cgroups(void)
+{
+	struct boost_groups *bg;
+	int cpu;
+
+	/* Initialize the per CPU boost groups */
+	for_each_possible_cpu(cpu) {
+		bg = &per_cpu(cpu_boost_groups, cpu);
+		memset(bg, 0, sizeof(struct boost_groups));
+		raw_spin_lock_init(&bg->lock);
+	}
+
+	pr_info("schedtune: configured to support %d boost groups\n",
+		BOOSTGROUPS_COUNT);
+
+	schedtune_initialized = true;
+}
+
+#else /* CONFIG_CGROUP_SCHEDTUNE */
+
+int
+schedtune_accept_deltas(int nrg_delta, int cap_delta,
+			struct task_struct *task)
+{
+	/* Optimal (O) region */
+	if (nrg_delta < 0 && cap_delta > 0) {
+		trace_sched_tune_filter(nrg_delta, cap_delta, 0, 0, 1, 0);
+		return INT_MAX;
+	}
+
+	/* Suboptimal (S) region */
+	if (nrg_delta > 0 && cap_delta < 0) {
+		trace_sched_tune_filter(nrg_delta, cap_delta, 0, 0, -1, 5);
+		return -INT_MAX;
+	}
+
+	return __schedtune_accept_deltas(nrg_delta, cap_delta,
+			perf_boost_idx, perf_constrain_idx);
+}
+
+#endif /* CONFIG_CGROUP_SCHEDTUNE */
+
+int
+sysctl_sched_cfs_boost_handler(struct ctl_table *table, int write,
+			       void __user *buffer, size_t *lenp,
+			       loff_t *ppos)
+{
+	int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos);
+	unsigned threshold_idx;
+	int boost_pct;
+
+	if (ret || !write)
+		return ret;
+
+	if (sysctl_sched_cfs_boost < -100 || sysctl_sched_cfs_boost > 100)
+		return -EINVAL;
+	boost_pct = sysctl_sched_cfs_boost;
+
+	/*
+	 * Update threshold params for Performance Boost (B)
+	 * and Performance Constraint (C) regions.
+	 * The current implementatio uses the same cuts for both
+	 * B and C regions.
+	 */
+	threshold_idx = clamp(boost_pct, 0, 99) / 10;
+	perf_boost_idx = threshold_idx;
+	perf_constrain_idx = threshold_idx;
+
+	return 0;
+}
+
+#ifdef CONFIG_SCHED_DEBUG
+static void
+schedtune_test_nrg(unsigned long delta_pwr)
+{
+	unsigned long test_delta_pwr;
+	unsigned long test_norm_pwr;
+	int idx;
+
+	/*
+	 * Check normalization constants using some constant system
+	 * energy values
+	 */
+	pr_info("schedtune: verify normalization constants...\n");
+	for (idx = 0; idx < 6; ++idx) {
+		test_delta_pwr = delta_pwr >> idx;
+
+		/* Normalize on max energy for target platform */
+		test_norm_pwr = reciprocal_divide(
+					test_delta_pwr << SCHED_LOAD_SHIFT,
+					schedtune_target_nrg.rdiv);
+
+		pr_info("schedtune: max_pwr/2^%d: %4lu => norm_pwr: %5lu\n",
+			idx, test_delta_pwr, test_norm_pwr);
+	}
+}
+#else
+#define schedtune_test_nrg(delta_pwr)
+#endif
+
+/*
+ * Compute the min/max power consumption of a cluster and all its CPUs
+ */
+static void
+schedtune_add_cluster_nrg(
+		struct sched_domain *sd,
+		struct sched_group *sg,
+		struct target_nrg *ste)
+{
+	struct sched_domain *sd2;
+	struct sched_group *sg2;
+
+	struct cpumask *cluster_cpus;
+	char str[32];
+
+	unsigned long min_pwr;
+	unsigned long max_pwr;
+	int cpu;
+
+	/* Get Cluster energy using EM data for the first CPU */
+	cluster_cpus = sched_group_cpus(sg);
+	snprintf(str, 32, "CLUSTER[%*pbl]",
+		 cpumask_pr_args(cluster_cpus));
+
+	min_pwr = sg->sge->idle_states[sg->sge->nr_idle_states - 1].power;
+	max_pwr = sg->sge->cap_states[sg->sge->nr_cap_states - 1].power;
+	pr_info("schedtune: %-17s min_pwr: %5lu max_pwr: %5lu\n",
+		str, min_pwr, max_pwr);
+
+	/*
+	 * Keep track of this cluster's energy in the computation of the
+	 * overall system energy
+	 */
+	ste->min_power += min_pwr;
+	ste->max_power += max_pwr;
+
+	/* Get CPU energy using EM data for each CPU in the group */
+	for_each_cpu(cpu, cluster_cpus) {
+		/* Get a SD view for the specific CPU */
+		for_each_domain(cpu, sd2) {
+			/* Get the CPU group */
+			sg2 = sd2->groups;
+			min_pwr = sg2->sge->idle_states[sg2->sge->nr_idle_states - 1].power;
+			max_pwr = sg2->sge->cap_states[sg2->sge->nr_cap_states - 1].power;
+
+			ste->min_power += min_pwr;
+			ste->max_power += max_pwr;
+
+			snprintf(str, 32, "CPU[%d]", cpu);
+			pr_info("schedtune: %-17s min_pwr: %5lu max_pwr: %5lu\n",
+				str, min_pwr, max_pwr);
+
+			/*
+			 * Assume we have EM data only at the CPU and
+			 * the upper CLUSTER level
+			 */
+			BUG_ON(!cpumask_equal(
+				sched_group_cpus(sg),
+				sched_group_cpus(sd2->parent->groups)
+				));
+			break;
+		}
+	}
+}
+
+/*
+ * Initialize the constants required to compute normalized energy.
+ * The values of these constants depends on the EM data for the specific
+ * target system and topology.
+ * Thus, this function is expected to be called by the code
+ * that bind the EM to the topology information.
+ */
+static int
+schedtune_init(void)
+{
+	struct target_nrg *ste = &schedtune_target_nrg;
+	unsigned long delta_pwr = 0;
+	struct sched_domain *sd;
+	struct sched_group *sg;
+
+	pr_info("schedtune: init normalization constants...\n");
+	ste->max_power = 0;
+	ste->min_power = 0;
+
+	rcu_read_lock();
+
+	/*
+	 * When EAS is in use, we always have a pointer to the highest SD
+	 * which provides EM data.
+	 */
+	sd = rcu_dereference(per_cpu(sd_ea, cpumask_first(cpu_online_mask)));
+	if (!sd) {
+		pr_info("schedtune: no energy model data\n");
+		goto nodata;
+	}
+
+	sg = sd->groups;
+	do {
+		schedtune_add_cluster_nrg(sd, sg, ste);
+	} while (sg = sg->next, sg != sd->groups);
+
+	rcu_read_unlock();
+
+	pr_info("schedtune: %-17s min_pwr: %5lu max_pwr: %5lu\n",
+		"SYSTEM", ste->min_power, ste->max_power);
+
+	/* Compute normalization constants */
+	delta_pwr = ste->max_power - ste->min_power;
+	ste->rdiv = reciprocal_value(delta_pwr);
+	pr_info("schedtune: using normalization constants mul: %u sh1: %u sh2: %u\n",
+		ste->rdiv.m, ste->rdiv.sh1, ste->rdiv.sh2);
+
+	schedtune_test_nrg(delta_pwr);
+
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+	schedtune_init_cgroups();
+#else
+	pr_info("schedtune: configured to support global boosting only\n");
+#endif
+
+	schedtune_spc_rdiv = reciprocal_value(100);
+
+	return 0;
+
+nodata:
+	pr_warning("schedtune: disabled!\n");
+	rcu_read_unlock();
+	return -EINVAL;
+}
+postcore_initcall(schedtune_init);
diff --git a/kernel/sched/tune.h b/kernel/sched/tune.h
new file mode 100644
index 000000000000..4f6441771e4c
--- /dev/null
+++ b/kernel/sched/tune.h
@@ -0,0 +1,55 @@
+
+#ifdef CONFIG_SCHED_TUNE
+
+#include <linux/reciprocal_div.h>
+
+/*
+ * System energy normalization constants
+ */
+struct target_nrg {
+	unsigned long min_power;
+	unsigned long max_power;
+	struct reciprocal_value rdiv;
+};
+
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+
+int schedtune_cpu_boost(int cpu);
+int schedtune_task_boost(struct task_struct *tsk);
+
+int schedtune_prefer_idle(struct task_struct *tsk);
+
+void schedtune_exit_task(struct task_struct *tsk);
+
+void schedtune_enqueue_task(struct task_struct *p, int cpu);
+void schedtune_dequeue_task(struct task_struct *p, int cpu);
+
+#else /* CONFIG_CGROUP_SCHEDTUNE */
+
+#define schedtune_cpu_boost(cpu)  get_sysctl_sched_cfs_boost()
+#define schedtune_task_boost(tsk) get_sysctl_sched_cfs_boost()
+
+#define schedtune_exit_task(task) do { } while (0)
+
+#define schedtune_enqueue_task(task, cpu) do { } while (0)
+#define schedtune_dequeue_task(task, cpu) do { } while (0)
+
+#endif /* CONFIG_CGROUP_SCHEDTUNE */
+
+int schedtune_normalize_energy(int energy);
+int schedtune_accept_deltas(int nrg_delta, int cap_delta,
+			    struct task_struct *task);
+
+#else /* CONFIG_SCHED_TUNE */
+
+#define schedtune_cpu_boost(cpu)  0
+#define schedtune_task_boost(tsk) 0
+
+#define schedtune_exit_task(task) do { } while (0)
+
+#define schedtune_enqueue_task(task, cpu) do { } while (0)
+#define schedtune_dequeue_task(task, cpu) do { } while (0)
+
+#define schedtune_accept_deltas(nrg_delta, cap_delta, task) nrg_delta
+
+#endif /* CONFIG_SCHED_TUNE */
diff --git a/kernel/sched/walt.c b/kernel/sched/walt.c
new file mode 100644
index 000000000000..8d25ffbe4fed
--- /dev/null
+++ b/kernel/sched/walt.c
@@ -0,0 +1,900 @@
+/*
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ *
+ * Window Assisted Load Tracking (WALT) implementation credits:
+ * Srivatsa Vaddagiri, Steve Muckle, Syed Rameez Mustafa, Joonwoo Park,
+ * Pavan Kumar Kondeti, Olav Haugan
+ *
+ * 2016-03-06: Integration with EAS/refactoring by Vikram Mulukutla
+ *             and Todd Kjos
+ */
+
+#include <linux/syscore_ops.h>
+#include <trace/events/sched.h>
+#include "sched.h"
+#include "walt.h"
+
+#define WINDOW_STATS_RECENT		0
+#define WINDOW_STATS_MAX		1
+#define WINDOW_STATS_MAX_RECENT_AVG	2
+#define WINDOW_STATS_AVG		3
+#define WINDOW_STATS_INVALID_POLICY	4
+
+#define EXITING_TASK_MARKER	0xdeaddead
+
+static __read_mostly unsigned int walt_ravg_hist_size = 5;
+static __read_mostly unsigned int walt_window_stats_policy =
+	WINDOW_STATS_MAX_RECENT_AVG;
+static __read_mostly unsigned int walt_account_wait_time = 1;
+static __read_mostly unsigned int walt_freq_account_wait_time = 0;
+static __read_mostly unsigned int walt_io_is_busy = 0;
+
+unsigned int sysctl_sched_walt_init_task_load_pct = 15;
+
+/* true -> use PELT based load stats, false -> use window-based load stats */
+bool __read_mostly walt_disabled = false;
+
+/*
+ * Window size (in ns). Adjust for the tick size so that the window
+ * rollover occurs just before the tick boundary.
+ */
+__read_mostly unsigned int walt_ravg_window =
+					    (20000000 / TICK_NSEC) * TICK_NSEC;
+#define MIN_SCHED_RAVG_WINDOW ((10000000 / TICK_NSEC) * TICK_NSEC)
+#define MAX_SCHED_RAVG_WINDOW ((1000000000 / TICK_NSEC) * TICK_NSEC)
+
+static unsigned int sync_cpu;
+static ktime_t ktime_last;
+static bool walt_ktime_suspended;
+
+static unsigned int task_load(struct task_struct *p)
+{
+	return p->ravg.demand;
+}
+
+static inline void fixup_cum_window_demand(struct rq *rq, s64 delta)
+{
+	rq->cum_window_demand += delta;
+	if (unlikely((s64)rq->cum_window_demand < 0))
+		rq->cum_window_demand = 0;
+}
+
+void
+walt_inc_cumulative_runnable_avg(struct rq *rq,
+				 struct task_struct *p)
+{
+	rq->cumulative_runnable_avg += p->ravg.demand;
+
+	/*
+	 * Add a task's contribution to the cumulative window demand when
+	 *
+	 * (1) task is enqueued with on_rq = 1 i.e migration,
+	 *     prio/cgroup/class change.
+	 * (2) task is waking for the first time in this window.
+	 */
+	if (p->on_rq || (p->last_sleep_ts < rq->window_start))
+		fixup_cum_window_demand(rq, p->ravg.demand);
+}
+
+void
+walt_dec_cumulative_runnable_avg(struct rq *rq,
+				 struct task_struct *p)
+{
+	rq->cumulative_runnable_avg -= p->ravg.demand;
+	BUG_ON((s64)rq->cumulative_runnable_avg < 0);
+
+	/*
+	 * on_rq will be 1 for sleeping tasks. So check if the task
+	 * is migrating or dequeuing in RUNNING state to change the
+	 * prio/cgroup/class.
+	 */
+	if (task_on_rq_migrating(p) || p->state == TASK_RUNNING)
+		fixup_cum_window_demand(rq, -(s64)p->ravg.demand);
+}
+
+static void
+fixup_cumulative_runnable_avg(struct rq *rq,
+			      struct task_struct *p, u64 new_task_load)
+{
+	s64 task_load_delta = (s64)new_task_load - task_load(p);
+
+	rq->cumulative_runnable_avg += task_load_delta;
+	if ((s64)rq->cumulative_runnable_avg < 0)
+		panic("cra less than zero: tld: %lld, task_load(p) = %u\n",
+			task_load_delta, task_load(p));
+
+	fixup_cum_window_demand(rq, task_load_delta);
+}
+
+u64 walt_ktime_clock(void)
+{
+	if (unlikely(walt_ktime_suspended))
+		return ktime_to_ns(ktime_last);
+	return ktime_get_ns();
+}
+
+static void walt_resume(void)
+{
+	walt_ktime_suspended = false;
+}
+
+static int walt_suspend(void)
+{
+	ktime_last = ktime_get();
+	walt_ktime_suspended = true;
+	return 0;
+}
+
+static struct syscore_ops walt_syscore_ops = {
+	.resume	= walt_resume,
+	.suspend = walt_suspend
+};
+
+static int __init walt_init_ops(void)
+{
+	register_syscore_ops(&walt_syscore_ops);
+	return 0;
+}
+late_initcall(walt_init_ops);
+
+void walt_inc_cfs_cumulative_runnable_avg(struct cfs_rq *cfs_rq,
+		struct task_struct *p)
+{
+	cfs_rq->cumulative_runnable_avg += p->ravg.demand;
+}
+
+void walt_dec_cfs_cumulative_runnable_avg(struct cfs_rq *cfs_rq,
+		struct task_struct *p)
+{
+	cfs_rq->cumulative_runnable_avg -= p->ravg.demand;
+}
+
+static int exiting_task(struct task_struct *p)
+{
+	if (p->flags & PF_EXITING) {
+		if (p->ravg.sum_history[0] != EXITING_TASK_MARKER) {
+			p->ravg.sum_history[0] = EXITING_TASK_MARKER;
+		}
+		return 1;
+	}
+	return 0;
+}
+
+static int __init set_walt_ravg_window(char *str)
+{
+	unsigned int adj_window;
+	bool no_walt = walt_disabled;
+
+	get_option(&str, &walt_ravg_window);
+
+	/* Adjust for CONFIG_HZ */
+	adj_window = (walt_ravg_window / TICK_NSEC) * TICK_NSEC;
+
+	/* Warn if we're a bit too far away from the expected window size */
+	WARN(adj_window < walt_ravg_window - NSEC_PER_MSEC,
+	     "tick-adjusted window size %u, original was %u\n", adj_window,
+	     walt_ravg_window);
+
+	walt_ravg_window = adj_window;
+
+	walt_disabled = walt_disabled ||
+			(walt_ravg_window < MIN_SCHED_RAVG_WINDOW ||
+			 walt_ravg_window > MAX_SCHED_RAVG_WINDOW);
+
+	WARN(!no_walt && walt_disabled,
+	     "invalid window size, disabling WALT\n");
+
+	return 0;
+}
+
+early_param("walt_ravg_window", set_walt_ravg_window);
+
+static void
+update_window_start(struct rq *rq, u64 wallclock)
+{
+	s64 delta;
+	int nr_windows;
+
+	delta = wallclock - rq->window_start;
+	/* If the MPM global timer is cleared, set delta as 0 to avoid kernel BUG happening */
+	if (delta < 0) {
+		delta = 0;
+		WARN_ONCE(1, "WALT wallclock appears to have gone backwards or reset\n");
+	}
+
+	if (delta < walt_ravg_window)
+		return;
+
+	nr_windows = div64_u64(delta, walt_ravg_window);
+	rq->window_start += (u64)nr_windows * (u64)walt_ravg_window;
+
+	rq->cum_window_demand = rq->cumulative_runnable_avg;
+}
+
+/*
+ * Translate absolute delta time accounted on a CPU
+ * to a scale where 1024 is the capacity of the most
+ * capable CPU running at FMAX
+ */
+static u64 scale_exec_time(u64 delta, struct rq *rq)
+{
+	unsigned long capcurr = capacity_curr_of(cpu_of(rq));
+
+	return (delta * capcurr) >> SCHED_CAPACITY_SHIFT;
+}
+
+static int cpu_is_waiting_on_io(struct rq *rq)
+{
+	if (!walt_io_is_busy)
+		return 0;
+
+	return atomic_read(&rq->nr_iowait);
+}
+
+void walt_account_irqtime(int cpu, struct task_struct *curr,
+				 u64 delta, u64 wallclock)
+{
+	struct rq *rq = cpu_rq(cpu);
+	unsigned long flags, nr_windows;
+	u64 cur_jiffies_ts;
+
+	raw_spin_lock_irqsave(&rq->lock, flags);
+
+	/*
+	 * cputime (wallclock) uses sched_clock so use the same here for
+	 * consistency.
+	 */
+	delta += sched_clock() - wallclock;
+	cur_jiffies_ts = get_jiffies_64();
+
+	if (is_idle_task(curr))
+		walt_update_task_ravg(curr, rq, IRQ_UPDATE, walt_ktime_clock(),
+				 delta);
+
+	nr_windows = cur_jiffies_ts - rq->irqload_ts;
+
+	if (nr_windows) {
+		if (nr_windows < 10) {
+			/* Decay CPU's irqload by 3/4 for each window. */
+			rq->avg_irqload *= (3 * nr_windows);
+			rq->avg_irqload = div64_u64(rq->avg_irqload,
+						    4 * nr_windows);
+		} else {
+			rq->avg_irqload = 0;
+		}
+		rq->avg_irqload += rq->cur_irqload;
+		rq->cur_irqload = 0;
+	}
+
+	rq->cur_irqload += delta;
+	rq->irqload_ts = cur_jiffies_ts;
+	raw_spin_unlock_irqrestore(&rq->lock, flags);
+}
+
+
+#define WALT_HIGH_IRQ_TIMEOUT 3
+
+u64 walt_irqload(int cpu) {
+	struct rq *rq = cpu_rq(cpu);
+	s64 delta;
+	delta = get_jiffies_64() - rq->irqload_ts;
+
+        /*
+	 * Current context can be preempted by irq and rq->irqload_ts can be
+	 * updated by irq context so that delta can be negative.
+	 * But this is okay and we can safely return as this means there
+	 * was recent irq occurrence.
+	 */
+
+        if (delta < WALT_HIGH_IRQ_TIMEOUT)
+		return rq->avg_irqload;
+        else
+		return 0;
+}
+
+int walt_cpu_high_irqload(int cpu) {
+	return walt_irqload(cpu) >= sysctl_sched_walt_cpu_high_irqload;
+}
+
+static int account_busy_for_cpu_time(struct rq *rq, struct task_struct *p,
+				     u64 irqtime, int event)
+{
+	if (is_idle_task(p)) {
+		/* TASK_WAKE && TASK_MIGRATE is not possible on idle task! */
+		if (event == PICK_NEXT_TASK)
+			return 0;
+
+		/* PUT_PREV_TASK, TASK_UPDATE && IRQ_UPDATE are left */
+		return irqtime || cpu_is_waiting_on_io(rq);
+	}
+
+	if (event == TASK_WAKE)
+		return 0;
+
+	if (event == PUT_PREV_TASK || event == IRQ_UPDATE ||
+					 event == TASK_UPDATE)
+		return 1;
+
+	/* Only TASK_MIGRATE && PICK_NEXT_TASK left */
+	return walt_freq_account_wait_time;
+}
+
+/*
+ * Account cpu activity in its busy time counters (rq->curr/prev_runnable_sum)
+ */
+static void update_cpu_busy_time(struct task_struct *p, struct rq *rq,
+	     int event, u64 wallclock, u64 irqtime)
+{
+	int new_window, nr_full_windows = 0;
+	int p_is_curr_task = (p == rq->curr);
+	u64 mark_start = p->ravg.mark_start;
+	u64 window_start = rq->window_start;
+	u32 window_size = walt_ravg_window;
+	u64 delta;
+
+	new_window = mark_start < window_start;
+	if (new_window) {
+		nr_full_windows = div64_u64((window_start - mark_start),
+						window_size);
+		if (p->ravg.active_windows < USHRT_MAX)
+			p->ravg.active_windows++;
+	}
+
+	/* Handle per-task window rollover. We don't care about the idle
+	 * task or exiting tasks. */
+	if (new_window && !is_idle_task(p) && !exiting_task(p)) {
+		u32 curr_window = 0;
+
+		if (!nr_full_windows)
+			curr_window = p->ravg.curr_window;
+
+		p->ravg.prev_window = curr_window;
+		p->ravg.curr_window = 0;
+	}
+
+	if (!account_busy_for_cpu_time(rq, p, irqtime, event)) {
+		/* account_busy_for_cpu_time() = 0, so no update to the
+		 * task's current window needs to be made. This could be
+		 * for example
+		 *
+		 *   - a wakeup event on a task within the current
+		 *     window (!new_window below, no action required),
+		 *   - switching to a new task from idle (PICK_NEXT_TASK)
+		 *     in a new window where irqtime is 0 and we aren't
+		 *     waiting on IO */
+
+		if (!new_window)
+			return;
+
+		/* A new window has started. The RQ demand must be rolled
+		 * over if p is the current task. */
+		if (p_is_curr_task) {
+			u64 prev_sum = 0;
+
+			/* p is either idle task or an exiting task */
+			if (!nr_full_windows) {
+				prev_sum = rq->curr_runnable_sum;
+			}
+
+			rq->prev_runnable_sum = prev_sum;
+			rq->curr_runnable_sum = 0;
+		}
+
+		return;
+	}
+
+	if (!new_window) {
+		/* account_busy_for_cpu_time() = 1 so busy time needs
+		 * to be accounted to the current window. No rollover
+		 * since we didn't start a new window. An example of this is
+		 * when a task starts execution and then sleeps within the
+		 * same window. */
+
+		if (!irqtime || !is_idle_task(p) || cpu_is_waiting_on_io(rq))
+			delta = wallclock - mark_start;
+		else
+			delta = irqtime;
+		delta = scale_exec_time(delta, rq);
+		rq->curr_runnable_sum += delta;
+		if (!is_idle_task(p) && !exiting_task(p))
+			p->ravg.curr_window += delta;
+
+		return;
+	}
+
+	if (!p_is_curr_task) {
+		/* account_busy_for_cpu_time() = 1 so busy time needs
+		 * to be accounted to the current window. A new window
+		 * has also started, but p is not the current task, so the
+		 * window is not rolled over - just split up and account
+		 * as necessary into curr and prev. The window is only
+		 * rolled over when a new window is processed for the current
+		 * task.
+		 *
+		 * Irqtime can't be accounted by a task that isn't the
+		 * currently running task. */
+
+		if (!nr_full_windows) {
+			/* A full window hasn't elapsed, account partial
+			 * contribution to previous completed window. */
+			delta = scale_exec_time(window_start - mark_start, rq);
+			if (!exiting_task(p))
+				p->ravg.prev_window += delta;
+		} else {
+			/* Since at least one full window has elapsed,
+			 * the contribution to the previous window is the
+			 * full window (window_size). */
+			delta = scale_exec_time(window_size, rq);
+			if (!exiting_task(p))
+				p->ravg.prev_window = delta;
+		}
+		rq->prev_runnable_sum += delta;
+
+		/* Account piece of busy time in the current window. */
+		delta = scale_exec_time(wallclock - window_start, rq);
+		rq->curr_runnable_sum += delta;
+		if (!exiting_task(p))
+			p->ravg.curr_window = delta;
+
+		return;
+	}
+
+	if (!irqtime || !is_idle_task(p) || cpu_is_waiting_on_io(rq)) {
+		/* account_busy_for_cpu_time() = 1 so busy time needs
+		 * to be accounted to the current window. A new window
+		 * has started and p is the current task so rollover is
+		 * needed. If any of these three above conditions are true
+		 * then this busy time can't be accounted as irqtime.
+		 *
+		 * Busy time for the idle task or exiting tasks need not
+		 * be accounted.
+		 *
+		 * An example of this would be a task that starts execution
+		 * and then sleeps once a new window has begun. */
+
+		if (!nr_full_windows) {
+			/* A full window hasn't elapsed, account partial
+			 * contribution to previous completed window. */
+			delta = scale_exec_time(window_start - mark_start, rq);
+			if (!is_idle_task(p) && !exiting_task(p))
+				p->ravg.prev_window += delta;
+
+			delta += rq->curr_runnable_sum;
+		} else {
+			/* Since at least one full window has elapsed,
+			 * the contribution to the previous window is the
+			 * full window (window_size). */
+			delta = scale_exec_time(window_size, rq);
+			if (!is_idle_task(p) && !exiting_task(p))
+				p->ravg.prev_window = delta;
+
+		}
+		/*
+		 * Rollover for normal runnable sum is done here by overwriting
+		 * the values in prev_runnable_sum and curr_runnable_sum.
+		 * Rollover for new task runnable sum has completed by previous
+		 * if-else statement.
+		 */
+		rq->prev_runnable_sum = delta;
+
+		/* Account piece of busy time in the current window. */
+		delta = scale_exec_time(wallclock - window_start, rq);
+		rq->curr_runnable_sum = delta;
+		if (!is_idle_task(p) && !exiting_task(p))
+			p->ravg.curr_window = delta;
+
+		return;
+	}
+
+	if (irqtime) {
+		/* account_busy_for_cpu_time() = 1 so busy time needs
+		 * to be accounted to the current window. A new window
+		 * has started and p is the current task so rollover is
+		 * needed. The current task must be the idle task because
+		 * irqtime is not accounted for any other task.
+		 *
+		 * Irqtime will be accounted each time we process IRQ activity
+		 * after a period of idleness, so we know the IRQ busy time
+		 * started at wallclock - irqtime. */
+
+		BUG_ON(!is_idle_task(p));
+		mark_start = wallclock - irqtime;
+
+		/* Roll window over. If IRQ busy time was just in the current
+		 * window then that is all that need be accounted. */
+		rq->prev_runnable_sum = rq->curr_runnable_sum;
+		if (mark_start > window_start) {
+			rq->curr_runnable_sum = scale_exec_time(irqtime, rq);
+			return;
+		}
+
+		/* The IRQ busy time spanned multiple windows. Process the
+		 * busy time preceding the current window start first. */
+		delta = window_start - mark_start;
+		if (delta > window_size)
+			delta = window_size;
+		delta = scale_exec_time(delta, rq);
+		rq->prev_runnable_sum += delta;
+
+		/* Process the remaining IRQ busy time in the current window. */
+		delta = wallclock - window_start;
+		rq->curr_runnable_sum = scale_exec_time(delta, rq);
+
+		return;
+	}
+
+	BUG();
+}
+
+static int account_busy_for_task_demand(struct task_struct *p, int event)
+{
+	/* No need to bother updating task demand for exiting tasks
+	 * or the idle task. */
+	if (exiting_task(p) || is_idle_task(p))
+		return 0;
+
+	/* When a task is waking up it is completing a segment of non-busy
+	 * time. Likewise, if wait time is not treated as busy time, then
+	 * when a task begins to run or is migrated, it is not running and
+	 * is completing a segment of non-busy time. */
+	if (event == TASK_WAKE || (!walt_account_wait_time &&
+			 (event == PICK_NEXT_TASK || event == TASK_MIGRATE)))
+		return 0;
+
+	return 1;
+}
+
+/*
+ * Called when new window is starting for a task, to record cpu usage over
+ * recently concluded window(s). Normally 'samples' should be 1. It can be > 1
+ * when, say, a real-time task runs without preemption for several windows at a
+ * stretch.
+ */
+static void update_history(struct rq *rq, struct task_struct *p,
+			 u32 runtime, int samples, int event)
+{
+	u32 *hist = &p->ravg.sum_history[0];
+	int ridx, widx;
+	u32 max = 0, avg, demand;
+	u64 sum = 0;
+
+	/* Ignore windows where task had no activity */
+	if (!runtime || is_idle_task(p) || exiting_task(p) || !samples)
+			goto done;
+
+	/* Push new 'runtime' value onto stack */
+	widx = walt_ravg_hist_size - 1;
+	ridx = widx - samples;
+	for (; ridx >= 0; --widx, --ridx) {
+		hist[widx] = hist[ridx];
+		sum += hist[widx];
+		if (hist[widx] > max)
+			max = hist[widx];
+	}
+
+	for (widx = 0; widx < samples && widx < walt_ravg_hist_size; widx++) {
+		hist[widx] = runtime;
+		sum += hist[widx];
+		if (hist[widx] > max)
+			max = hist[widx];
+	}
+
+	p->ravg.sum = 0;
+
+	if (walt_window_stats_policy == WINDOW_STATS_RECENT) {
+		demand = runtime;
+	} else if (walt_window_stats_policy == WINDOW_STATS_MAX) {
+		demand = max;
+	} else {
+		avg = div64_u64(sum, walt_ravg_hist_size);
+		if (walt_window_stats_policy == WINDOW_STATS_AVG)
+			demand = avg;
+		else
+			demand = max(avg, runtime);
+	}
+
+	/*
+	 * A throttled deadline sched class task gets dequeued without
+	 * changing p->on_rq. Since the dequeue decrements hmp stats
+	 * avoid decrementing it here again.
+	 *
+	 * When window is rolled over, the cumulative window demand
+	 * is reset to the cumulative runnable average (contribution from
+	 * the tasks on the runqueue). If the current task is dequeued
+	 * already, it's demand is not included in the cumulative runnable
+	 * average. So add the task demand separately to cumulative window
+	 * demand.
+	 */
+	if (!task_has_dl_policy(p) || !p->dl.dl_throttled) {
+		if (task_on_rq_queued(p))
+			fixup_cumulative_runnable_avg(rq, p, demand);
+		else if (rq->curr == p)
+			fixup_cum_window_demand(rq, demand);
+	}
+
+	p->ravg.demand = demand;
+
+done:
+	trace_walt_update_history(rq, p, runtime, samples, event);
+	return;
+}
+
+static void add_to_task_demand(struct rq *rq, struct task_struct *p,
+				u64 delta)
+{
+	delta = scale_exec_time(delta, rq);
+	p->ravg.sum += delta;
+	if (unlikely(p->ravg.sum > walt_ravg_window))
+		p->ravg.sum = walt_ravg_window;
+}
+
+/*
+ * Account cpu demand of task and/or update task's cpu demand history
+ *
+ * ms = p->ravg.mark_start;
+ * wc = wallclock
+ * ws = rq->window_start
+ *
+ * Three possibilities:
+ *
+ *	a) Task event is contained within one window.
+ *		window_start < mark_start < wallclock
+ *
+ *		ws   ms  wc
+ *		|    |   |
+ *		V    V   V
+ *		|---------------|
+ *
+ *	In this case, p->ravg.sum is updated *iff* event is appropriate
+ *	(ex: event == PUT_PREV_TASK)
+ *
+ *	b) Task event spans two windows.
+ *		mark_start < window_start < wallclock
+ *
+ *		ms   ws   wc
+ *		|    |    |
+ *		V    V    V
+ *		-----|-------------------
+ *
+ *	In this case, p->ravg.sum is updated with (ws - ms) *iff* event
+ *	is appropriate, then a new window sample is recorded followed
+ *	by p->ravg.sum being set to (wc - ws) *iff* event is appropriate.
+ *
+ *	c) Task event spans more than two windows.
+ *
+ *		ms ws_tmp			   ws  wc
+ *		|  |				   |   |
+ *		V  V				   V   V
+ *		---|-------|-------|-------|-------|------
+ *		   |				   |
+ *		   |<------ nr_full_windows ------>|
+ *
+ *	In this case, p->ravg.sum is updated with (ws_tmp - ms) first *iff*
+ *	event is appropriate, window sample of p->ravg.sum is recorded,
+ *	'nr_full_window' samples of window_size is also recorded *iff*
+ *	event is appropriate and finally p->ravg.sum is set to (wc - ws)
+ *	*iff* event is appropriate.
+ *
+ * IMPORTANT : Leave p->ravg.mark_start unchanged, as update_cpu_busy_time()
+ * depends on it!
+ */
+static void update_task_demand(struct task_struct *p, struct rq *rq,
+	     int event, u64 wallclock)
+{
+	u64 mark_start = p->ravg.mark_start;
+	u64 delta, window_start = rq->window_start;
+	int new_window, nr_full_windows;
+	u32 window_size = walt_ravg_window;
+
+	new_window = mark_start < window_start;
+	if (!account_busy_for_task_demand(p, event)) {
+		if (new_window)
+			/* If the time accounted isn't being accounted as
+			 * busy time, and a new window started, only the
+			 * previous window need be closed out with the
+			 * pre-existing demand. Multiple windows may have
+			 * elapsed, but since empty windows are dropped,
+			 * it is not necessary to account those. */
+			update_history(rq, p, p->ravg.sum, 1, event);
+		return;
+	}
+
+	if (!new_window) {
+		/* The simple case - busy time contained within the existing
+		 * window. */
+		add_to_task_demand(rq, p, wallclock - mark_start);
+		return;
+	}
+
+	/* Busy time spans at least two windows. Temporarily rewind
+	 * window_start to first window boundary after mark_start. */
+	delta = window_start - mark_start;
+	nr_full_windows = div64_u64(delta, window_size);
+	window_start -= (u64)nr_full_windows * (u64)window_size;
+
+	/* Process (window_start - mark_start) first */
+	add_to_task_demand(rq, p, window_start - mark_start);
+
+	/* Push new sample(s) into task's demand history */
+	update_history(rq, p, p->ravg.sum, 1, event);
+	if (nr_full_windows)
+		update_history(rq, p, scale_exec_time(window_size, rq),
+			       nr_full_windows, event);
+
+	/* Roll window_start back to current to process any remainder
+	 * in current window. */
+	window_start += (u64)nr_full_windows * (u64)window_size;
+
+	/* Process (wallclock - window_start) next */
+	mark_start = window_start;
+	add_to_task_demand(rq, p, wallclock - mark_start);
+}
+
+/* Reflect task activity on its demand and cpu's busy time statistics */
+void walt_update_task_ravg(struct task_struct *p, struct rq *rq,
+	     int event, u64 wallclock, u64 irqtime)
+{
+	if (walt_disabled || !rq->window_start)
+		return;
+
+	lockdep_assert_held(&rq->lock);
+
+	update_window_start(rq, wallclock);
+
+	if (!p->ravg.mark_start)
+		goto done;
+
+	update_task_demand(p, rq, event, wallclock);
+	update_cpu_busy_time(p, rq, event, wallclock, irqtime);
+
+done:
+	trace_walt_update_task_ravg(p, rq, event, wallclock, irqtime);
+
+	p->ravg.mark_start = wallclock;
+}
+
+static void reset_task_stats(struct task_struct *p)
+{
+	u32 sum = 0;
+
+	if (exiting_task(p))
+		sum = EXITING_TASK_MARKER;
+
+	memset(&p->ravg, 0, sizeof(struct ravg));
+	/* Retain EXITING_TASK marker */
+	p->ravg.sum_history[0] = sum;
+}
+
+void walt_mark_task_starting(struct task_struct *p)
+{
+	u64 wallclock;
+	struct rq *rq = task_rq(p);
+
+	if (!rq->window_start) {
+		reset_task_stats(p);
+		return;
+	}
+
+	wallclock = walt_ktime_clock();
+	p->ravg.mark_start = wallclock;
+}
+
+void walt_set_window_start(struct rq *rq)
+{
+	int cpu = cpu_of(rq);
+	struct rq *sync_rq = cpu_rq(sync_cpu);
+
+	if (likely(rq->window_start))
+		return;
+
+	if (cpu == sync_cpu) {
+		rq->window_start = 1;
+	} else {
+		raw_spin_unlock(&rq->lock);
+		double_rq_lock(rq, sync_rq);
+		rq->window_start = cpu_rq(sync_cpu)->window_start;
+		rq->curr_runnable_sum = rq->prev_runnable_sum = 0;
+		raw_spin_unlock(&sync_rq->lock);
+	}
+
+	rq->curr->ravg.mark_start = rq->window_start;
+}
+
+void walt_migrate_sync_cpu(int cpu)
+{
+	if (cpu == sync_cpu)
+		sync_cpu = smp_processor_id();
+}
+
+void walt_fixup_busy_time(struct task_struct *p, int new_cpu)
+{
+	struct rq *src_rq = task_rq(p);
+	struct rq *dest_rq = cpu_rq(new_cpu);
+	u64 wallclock;
+
+	if (!p->on_rq && p->state != TASK_WAKING)
+		return;
+
+	if (exiting_task(p)) {
+		return;
+	}
+
+	if (p->state == TASK_WAKING)
+		double_rq_lock(src_rq, dest_rq);
+
+	wallclock = walt_ktime_clock();
+
+	walt_update_task_ravg(task_rq(p)->curr, task_rq(p),
+			TASK_UPDATE, wallclock, 0);
+	walt_update_task_ravg(dest_rq->curr, dest_rq,
+			TASK_UPDATE, wallclock, 0);
+
+	walt_update_task_ravg(p, task_rq(p), TASK_MIGRATE, wallclock, 0);
+
+	/*
+	 * When a task is migrating during the wakeup, adjust
+	 * the task's contribution towards cumulative window
+	 * demand.
+	 */
+	if (p->state == TASK_WAKING &&
+	    p->last_sleep_ts >= src_rq->window_start) {
+		fixup_cum_window_demand(src_rq, -(s64)p->ravg.demand);
+		fixup_cum_window_demand(dest_rq, p->ravg.demand);
+	}
+
+	if (p->ravg.curr_window) {
+		src_rq->curr_runnable_sum -= p->ravg.curr_window;
+		dest_rq->curr_runnable_sum += p->ravg.curr_window;
+	}
+
+	if (p->ravg.prev_window) {
+		src_rq->prev_runnable_sum -= p->ravg.prev_window;
+		dest_rq->prev_runnable_sum += p->ravg.prev_window;
+	}
+
+	if ((s64)src_rq->prev_runnable_sum < 0) {
+		src_rq->prev_runnable_sum = 0;
+		WARN_ON(1);
+	}
+	if ((s64)src_rq->curr_runnable_sum < 0) {
+		src_rq->curr_runnable_sum = 0;
+		WARN_ON(1);
+	}
+
+	trace_walt_migration_update_sum(src_rq, p);
+	trace_walt_migration_update_sum(dest_rq, p);
+
+	if (p->state == TASK_WAKING)
+		double_rq_unlock(src_rq, dest_rq);
+}
+
+void walt_init_new_task_load(struct task_struct *p)
+{
+	int i;
+	u32 init_load_windows =
+			div64_u64((u64)sysctl_sched_walt_init_task_load_pct *
+                          (u64)walt_ravg_window, 100);
+	u32 init_load_pct = current->init_load_pct;
+
+	p->init_load_pct = 0;
+	memset(&p->ravg, 0, sizeof(struct ravg));
+
+	if (init_load_pct) {
+		init_load_windows = div64_u64((u64)init_load_pct *
+			  (u64)walt_ravg_window, 100);
+	}
+
+	p->ravg.demand = init_load_windows;
+	for (i = 0; i < RAVG_HIST_SIZE_MAX; ++i)
+		p->ravg.sum_history[i] = init_load_windows;
+}
diff --git a/kernel/sched/walt.h b/kernel/sched/walt.h
new file mode 100644
index 000000000000..de7edac43674
--- /dev/null
+++ b/kernel/sched/walt.h
@@ -0,0 +1,64 @@
+/*
+ * Copyright (c) 2016, The Linux Foundation. All rights reserved.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 and
+ * only version 2 as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef __WALT_H
+#define __WALT_H
+
+#ifdef CONFIG_SCHED_WALT
+
+void walt_update_task_ravg(struct task_struct *p, struct rq *rq, int event,
+		u64 wallclock, u64 irqtime);
+void walt_inc_cumulative_runnable_avg(struct rq *rq, struct task_struct *p);
+void walt_dec_cumulative_runnable_avg(struct rq *rq, struct task_struct *p);
+void walt_inc_cfs_cumulative_runnable_avg(struct cfs_rq *rq,
+		struct task_struct *p);
+void walt_dec_cfs_cumulative_runnable_avg(struct cfs_rq *rq,
+		struct task_struct *p);
+void walt_fixup_busy_time(struct task_struct *p, int new_cpu);
+void walt_init_new_task_load(struct task_struct *p);
+void walt_mark_task_starting(struct task_struct *p);
+void walt_set_window_start(struct rq *rq);
+void walt_migrate_sync_cpu(int cpu);
+void walt_init_cpu_efficiency(void);
+u64 walt_ktime_clock(void);
+void walt_account_irqtime(int cpu, struct task_struct *curr, u64 delta,
+                                  u64 wallclock);
+
+u64 walt_irqload(int cpu);
+int walt_cpu_high_irqload(int cpu);
+
+#else /* CONFIG_SCHED_WALT */
+
+static inline void walt_update_task_ravg(struct task_struct *p, struct rq *rq,
+		int event, u64 wallclock, u64 irqtime) { }
+static inline void walt_inc_cumulative_runnable_avg(struct rq *rq, struct task_struct *p) { }
+static inline void walt_dec_cumulative_runnable_avg(struct rq *rq, struct task_struct *p) { }
+static inline void walt_inc_cfs_cumulative_runnable_avg(struct cfs_rq *rq,
+		struct task_struct *p) { }
+static inline void walt_dec_cfs_cumulative_runnable_avg(struct cfs_rq *rq,
+		struct task_struct *p) { }
+static inline void walt_fixup_busy_time(struct task_struct *p, int new_cpu) { }
+static inline void walt_init_new_task_load(struct task_struct *p) { }
+static inline void walt_mark_task_starting(struct task_struct *p) { }
+static inline void walt_set_window_start(struct rq *rq) { }
+static inline void walt_migrate_sync_cpu(int cpu) { }
+static inline void walt_init_cpu_efficiency(void) { }
+static inline u64 walt_ktime_clock(void) { return 0; }
+
+#define walt_cpu_high_irqload(cpu) false
+
+#endif /* CONFIG_SCHED_WALT */
+
+extern bool walt_disabled;
+
+#endif
diff --git a/kernel/sys.c b/kernel/sys.c
index 78947de6f969..ba3ddb43dd9f 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -41,6 +41,8 @@
 #include <linux/syscore_ops.h>
 #include <linux/version.h>
 #include <linux/ctype.h>
+#include <linux/mm.h>
+#include <linux/mempolicy.h>
 
 #include <linux/compat.h>
 #include <linux/syscalls.h>
@@ -2072,10 +2074,158 @@ static int prctl_get_tid_address(struct task_struct *me, int __user **tid_addr)
 }
 #endif
 
+#ifdef CONFIG_MMU
+static int prctl_update_vma_anon_name(struct vm_area_struct *vma,
+		struct vm_area_struct **prev,
+		unsigned long start, unsigned long end,
+		const char __user *name_addr)
+{
+	struct mm_struct *mm = vma->vm_mm;
+	int error = 0;
+	pgoff_t pgoff;
+
+	if (name_addr == vma_get_anon_name(vma)) {
+		*prev = vma;
+		goto out;
+	}
+
+	pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
+	*prev = vma_merge(mm, *prev, start, end, vma->vm_flags, vma->anon_vma,
+				vma->vm_file, pgoff, vma_policy(vma),
+				vma->vm_userfaultfd_ctx, name_addr);
+	if (*prev) {
+		vma = *prev;
+		goto success;
+	}
+
+	*prev = vma;
+
+	if (start != vma->vm_start) {
+		error = split_vma(mm, vma, start, 1);
+		if (error)
+			goto out;
+	}
+
+	if (end != vma->vm_end) {
+		error = split_vma(mm, vma, end, 0);
+		if (error)
+			goto out;
+	}
+
+success:
+	if (!vma->vm_file)
+		vma->anon_name = name_addr;
+
+out:
+	if (error == -ENOMEM)
+		error = -EAGAIN;
+	return error;
+}
+
+static int prctl_set_vma_anon_name(unsigned long start, unsigned long end,
+			unsigned long arg)
+{
+	unsigned long tmp;
+	struct vm_area_struct *vma, *prev;
+	int unmapped_error = 0;
+	int error = -EINVAL;
+
+	/*
+	 * If the interval [start,end) covers some unmapped address
+	 * ranges, just ignore them, but return -ENOMEM at the end.
+	 * - this matches the handling in madvise.
+	 */
+	vma = find_vma_prev(current->mm, start, &prev);
+	if (vma && start > vma->vm_start)
+		prev = vma;
+
+	for (;;) {
+		/* Still start < end. */
+		error = -ENOMEM;
+		if (!vma)
+			return error;
+
+		/* Here start < (end|vma->vm_end). */
+		if (start < vma->vm_start) {
+			unmapped_error = -ENOMEM;
+			start = vma->vm_start;
+			if (start >= end)
+				return error;
+		}
+
+		/* Here vma->vm_start <= start < (end|vma->vm_end) */
+		tmp = vma->vm_end;
+		if (end < tmp)
+			tmp = end;
+
+		/* Here vma->vm_start <= start < tmp <= (end|vma->vm_end). */
+		error = prctl_update_vma_anon_name(vma, &prev, start, tmp,
+				(const char __user *)arg);
+		if (error)
+			return error;
+		start = tmp;
+		if (prev && start < prev->vm_end)
+			start = prev->vm_end;
+		error = unmapped_error;
+		if (start >= end)
+			return error;
+		if (prev)
+			vma = prev->vm_next;
+		else	/* madvise_remove dropped mmap_sem */
+			vma = find_vma(current->mm, start);
+	}
+}
+
+static int prctl_set_vma(unsigned long opt, unsigned long start,
+		unsigned long len_in, unsigned long arg)
+{
+	struct mm_struct *mm = current->mm;
+	int error;
+	unsigned long len;
+	unsigned long end;
+
+	if (start & ~PAGE_MASK)
+		return -EINVAL;
+	len = (len_in + ~PAGE_MASK) & PAGE_MASK;
+
+	/* Check to see whether len was rounded up from small -ve to zero */
+	if (len_in && !len)
+		return -EINVAL;
+
+	end = start + len;
+	if (end < start)
+		return -EINVAL;
+
+	if (end == start)
+		return 0;
+
+	down_write(&mm->mmap_sem);
+
+	switch (opt) {
+	case PR_SET_VMA_ANON_NAME:
+		error = prctl_set_vma_anon_name(start, end, arg);
+		break;
+	default:
+		error = -EINVAL;
+	}
+
+	up_write(&mm->mmap_sem);
+
+	return error;
+}
+#else /* CONFIG_MMU */
+static int prctl_set_vma(unsigned long opt, unsigned long start,
+		unsigned long len_in, unsigned long arg)
+{
+	return -EINVAL;
+}
+#endif
+
 SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
 		unsigned long, arg4, unsigned long, arg5)
 {
 	struct task_struct *me = current;
+	struct task_struct *tsk;
 	unsigned char comm[sizeof(me->comm)];
 	long error;
 
@@ -2169,7 +2319,10 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
 		error = perf_event_task_enable();
 		break;
 	case PR_GET_TIMERSLACK:
-		error = current->timer_slack_ns;
+		if (current->timer_slack_ns > ULONG_MAX)
+			error = ULONG_MAX;
+		else
+			error = current->timer_slack_ns;
 		break;
 	case PR_SET_TIMERSLACK:
 		if (arg2 <= 0)
@@ -2218,6 +2371,26 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
 	case PR_GET_TID_ADDRESS:
 		error = prctl_get_tid_address(me, (int __user **)arg2);
 		break;
+	case PR_SET_TIMERSLACK_PID:
+		if (task_pid_vnr(current) != (pid_t)arg3 &&
+				!capable(CAP_SYS_NICE))
+			return -EPERM;
+		rcu_read_lock();
+		tsk = find_task_by_vpid((pid_t)arg3);
+		if (tsk == NULL) {
+			rcu_read_unlock();
+			return -EINVAL;
+		}
+		get_task_struct(tsk);
+		rcu_read_unlock();
+		if (arg2 <= 0)
+			tsk->timer_slack_ns =
+				tsk->default_timer_slack_ns;
+		else
+			tsk->timer_slack_ns = arg2;
+		put_task_struct(tsk);
+		error = 0;
+		break;
 	case PR_SET_CHILD_SUBREAPER:
 		me->signal->is_child_subreaper = !!arg2;
 		break;
@@ -2266,6 +2439,9 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
 	case PR_GET_FP_MODE:
 		error = GET_FP_MODE(me);
 		break;
+	case PR_SET_VMA:
+		error = prctl_set_vma(arg2, arg3, arg4, arg5);
+		break;
 	default:
 		error = -EINVAL;
 		break;
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 17c59e78661b..4e2f98dd2052 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -105,6 +105,7 @@ extern char core_pattern[];
 extern unsigned int core_pipe_limit;
 #endif
 extern int pid_max;
+extern int extra_free_kbytes;
 extern int pid_max_min, pid_max_max;
 extern int percpu_pagelist_fraction;
 extern int compat_log;
@@ -305,6 +306,50 @@ static struct ctl_table kern_table[] = {
 		.extra2		= &max_sched_granularity_ns,
 	},
 	{
+		.procname	= "sched_sync_hint_enable",
+		.data		= &sysctl_sched_sync_hint_enable,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+#ifdef CONFIG_SCHED_WALT
+	{
+		.procname	= "sched_use_walt_cpu_util",
+		.data		= &sysctl_sched_use_walt_cpu_util,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
+		.procname	= "sched_use_walt_task_util",
+		.data		= &sysctl_sched_use_walt_task_util,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
+		.procname	= "sched_walt_init_task_load_pct",
+		.data		= &sysctl_sched_walt_init_task_load_pct,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
+		.procname	= "sched_walt_cpu_high_irqload",
+		.data		= &sysctl_sched_walt_cpu_high_irqload,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+#endif
+	{
+		.procname	= "sched_cstate_aware",
+		.data		= &sysctl_sched_cstate_aware,
+		.maxlen		= sizeof(unsigned int),
+		.mode		= 0644,
+		.proc_handler	= proc_dointvec,
+	},
+	{
 		.procname	= "sched_wakeup_granularity_ns",
 		.data		= &sysctl_sched_wakeup_granularity,
 		.maxlen		= sizeof(unsigned int),
@@ -435,6 +480,21 @@ static struct ctl_table kern_table[] = {
 		.extra1		= &one,
 	},
 #endif
+#ifdef CONFIG_SCHED_TUNE
+	{
+		.procname	= "sched_cfs_boost",
+		.data		= &sysctl_sched_cfs_boost,
+		.maxlen		= sizeof(sysctl_sched_cfs_boost),
+#ifdef CONFIG_CGROUP_SCHEDTUNE
+		.mode		= 0444,
+#else
+		.mode		= 0644,
+#endif
+		.proc_handler	= &sysctl_sched_cfs_boost_handler,
+		.extra1		= &zero,
+		.extra2		= &one_hundred,
+	},
+#endif
 #ifdef CONFIG_PROVE_LOCKING
 	{
 		.procname	= "prove_locking",
@@ -1396,6 +1456,14 @@ static struct ctl_table vm_table[] = {
 		.extra1		= &zero,
 	},
 	{
+		.procname	= "extra_free_kbytes",
+		.data		= &extra_free_kbytes,
+		.maxlen		= sizeof(extra_free_kbytes),
+		.mode		= 0644,
+		.proc_handler	= min_free_kbytes_sysctl_handler,
+		.extra1		= &zero,
+	},
+	{
 		.procname	= "percpu_pagelist_fraction",
 		.data		= &percpu_pagelist_fraction,
 		.maxlen		= sizeof(percpu_pagelist_fraction),
@@ -1571,6 +1639,28 @@ static struct ctl_table vm_table[] = {
 		.mode		= 0644,
 		.proc_handler	= proc_doulongvec_minmax,
 	},
+#ifdef CONFIG_HAVE_ARCH_MMAP_RND_BITS
+	{
+		.procname	= "mmap_rnd_bits",
+		.data		= &mmap_rnd_bits,
+		.maxlen		= sizeof(mmap_rnd_bits),
+		.mode		= 0600,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= (void *)&mmap_rnd_bits_min,
+		.extra2		= (void *)&mmap_rnd_bits_max,
+	},
+#endif
+#ifdef CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS
+	{
+		.procname	= "mmap_rnd_compat_bits",
+		.data		= &mmap_rnd_compat_bits,
+		.maxlen		= sizeof(mmap_rnd_compat_bits),
+		.mode		= 0600,
+		.proc_handler	= proc_dointvec_minmax,
+		.extra1		= (void *)&mmap_rnd_compat_bits_min,
+		.extra2		= (void *)&mmap_rnd_compat_bits_max,
+	},
+#endif
 	{ }
 };
 
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
index 17f7bcff1e02..405536b22c0c 100644
--- a/kernel/time/hrtimer.c
+++ b/kernel/time/hrtimer.c
@@ -984,7 +984,7 @@ static inline ktime_t hrtimer_update_lowres(struct hrtimer *timer, ktime_t tim,
  *		relative (HRTIMER_MODE_REL)
  */
 void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
-			    unsigned long delta_ns, const enum hrtimer_mode mode)
+			    u64 delta_ns, const enum hrtimer_mode mode)
 {
 	struct hrtimer_clock_base *base, *new_base;
 	unsigned long flags;
@@ -1553,7 +1553,7 @@ long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
 	struct restart_block *restart;
 	struct hrtimer_sleeper t;
 	int ret = 0;
-	unsigned long slack;
+	u64 slack;
 
 	slack = current->timer_slack_ns;
 	if (dl_task(current) || rt_task(current))
@@ -1729,7 +1729,7 @@ void __init hrtimers_init(void)
  * @clock:	timer clock, CLOCK_MONOTONIC or CLOCK_REALTIME
  */
 int __sched
-schedule_hrtimeout_range_clock(ktime_t *expires, unsigned long delta,
+schedule_hrtimeout_range_clock(ktime_t *expires, u64 delta,
 			       const enum hrtimer_mode mode, int clock)
 {
 	struct hrtimer_sleeper t;
@@ -1797,7 +1797,7 @@ schedule_hrtimeout_range_clock(ktime_t *expires, unsigned long delta,
  *
  * Returns 0 when the timer has expired otherwise -EINTR
  */
-int __sched schedule_hrtimeout_range(ktime_t *expires, unsigned long delta,
+int __sched schedule_hrtimeout_range(ktime_t *expires, u64 delta,
 				     const enum hrtimer_mode mode)
 {
 	return schedule_hrtimeout_range_clock(expires, delta, mode,
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 22c57e191a23..34f9a9c417d9 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -870,6 +870,18 @@ ktime_t tick_nohz_get_sleep_length(void)
 	return ts->sleep_length;
 }
 
+/**
+ * tick_nohz_get_idle_calls - return the current idle calls counter value
+ *
+ * Called from the schedutil frequency scaling governor in scheduler context.
+ */
+unsigned long tick_nohz_get_idle_calls(void)
+{
+	struct tick_sched *ts = this_cpu_ptr(&tick_cpu_sched);
+
+	return ts->idle_calls;
+}
+
 static void tick_nohz_account_idle_ticks(struct tick_sched *ts)
 {
 #ifndef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 6e4866834d26..738f3467d169 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -442,6 +442,35 @@ u64 ktime_get_raw_fast_ns(void)
 }
 EXPORT_SYMBOL_GPL(ktime_get_raw_fast_ns);
 
+/**
+ * ktime_get_boot_fast_ns - NMI safe and fast access to boot clock.
+ *
+ * To keep it NMI safe since we're accessing from tracing, we're not using a
+ * separate timekeeper with updates to monotonic clock and boot offset
+ * protected with seqlocks. This has the following minor side effects:
+ *
+ * (1) Its possible that a timestamp be taken after the boot offset is updated
+ * but before the timekeeper is updated. If this happens, the new boot offset
+ * is added to the old timekeeping making the clock appear to update slightly
+ * earlier:
+ *    CPU 0                                        CPU 1
+ *    timekeeping_inject_sleeptime64()
+ *    __timekeeping_inject_sleeptime(tk, delta);
+ *                                                 timestamp();
+ *    timekeeping_update(tk, TK_CLEAR_NTP...);
+ *
+ * (2) On 32-bit systems, the 64-bit boot offset (tk->offs_boot) may be
+ * partially updated.  Since the tk->offs_boot update is a rare event, this
+ * should be a rare occurrence which postprocessing should be able to handle.
+ */
+u64 notrace ktime_get_boot_fast_ns(void)
+{
+	struct timekeeper *tk = &tk_core.timekeeper;
+
+	return (ktime_get_mono_fast_ns() + ktime_to_ns(tk->offs_boot));
+}
+EXPORT_SYMBOL_GPL(ktime_get_boot_fast_ns);
+
 /* Suspend-time cycles value for halted fast timekeeper. */
 static cycle_t cycles_at_suspend;
 
diff --git a/kernel/time/timer.c b/kernel/time/timer.c
index 125407144c01..f575785c7cd8 100644
--- a/kernel/time/timer.c
+++ b/kernel/time/timer.c
@@ -1698,10 +1698,10 @@ EXPORT_SYMBOL(msleep_interruptible);
 static void __sched do_usleep_range(unsigned long min, unsigned long max)
 {
 	ktime_t kmin;
-	unsigned long delta;
+	u64 delta;
 
 	kmin = ktime_set(0, min * NSEC_PER_USEC);
-	delta = (max - min) * NSEC_PER_USEC;
+	delta = (u64)(max - min) * NSEC_PER_USEC;
 	schedule_hrtimeout_range(&kmin, delta, HRTIMER_MODE_REL);
 }
 
diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
index e45db6b0d878..006eefb6ede0 100644
--- a/kernel/trace/Kconfig
+++ b/kernel/trace/Kconfig
@@ -77,6 +77,9 @@ config EVENT_TRACING
 	select CONTEXT_SWITCH_TRACER
 	bool
 
+config GPU_TRACEPOINTS
+	bool
+
 config CONTEXT_SWITCH_TRACER
 	bool
 
@@ -162,6 +165,17 @@ config FUNCTION_GRAPH_TRACER
 	  address on the current task structure into a stack of calls.
 
 
+config PREEMPTIRQ_EVENTS
+	bool "Enable trace events for preempt and irq disable/enable"
+	select TRACE_IRQFLAGS
+	depends on DEBUG_PREEMPT || !PROVE_LOCKING
+	default n
+	help
+	  Enable tracing of disable and enable events for preemption and irqs.
+	  For tracing preempt disable/enable events, DEBUG_PREEMPT must be
+	  enabled. For tracing irq disable/enable events, PROVE_LOCKING must
+	  be disabled.
+
 config IRQSOFF_TRACER
 	bool "Interrupts-off Latency Tracer"
 	default n
diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
index 05ea5167e6bb..4b35fb97ae44 100644
--- a/kernel/trace/Makefile
+++ b/kernel/trace/Makefile
@@ -37,6 +37,7 @@ obj-$(CONFIG_TRACING) += trace_stat.o
 obj-$(CONFIG_TRACING) += trace_printk.o
 obj-$(CONFIG_CONTEXT_SWITCH_TRACER) += trace_sched_switch.o
 obj-$(CONFIG_FUNCTION_TRACER) += trace_functions.o
+obj-$(CONFIG_PREEMPTIRQ_EVENTS) += trace_irqsoff.o
 obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o
 obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o
@@ -68,6 +69,7 @@ obj-$(CONFIG_KGDB_KDB) += trace_kdb.o
 endif
 obj-$(CONFIG_PROBE_EVENTS) += trace_probe.o
 obj-$(CONFIG_UPROBE_EVENT) += trace_uprobe.o
+obj-$(CONFIG_GPU_TRACEPOINTS) += gpu-traces.o
 
 obj-$(CONFIG_TRACEPOINT_BENCHMARK) += trace_benchmark.o
 
diff --git a/kernel/trace/gpu-traces.c b/kernel/trace/gpu-traces.c
new file mode 100644
index 000000000000..a4b3f00faee3
--- /dev/null
+++ b/kernel/trace/gpu-traces.c
@@ -0,0 +1,23 @@
+/*
+ * GPU tracepoints
+ *
+ * Copyright (C) 2013 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#include <linux/module.h>
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/gpu.h>
+
+EXPORT_TRACEPOINT_SYMBOL(gpu_sched_switch);
+EXPORT_TRACEPOINT_SYMBOL(gpu_job_enqueue);
diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
index b64f35afee4e..23de746b2195 100644
--- a/kernel/trace/trace.c
+++ b/kernel/trace/trace.c
@@ -890,6 +890,7 @@ static struct {
 	{ trace_clock,			"perf",		1 },
 	{ ktime_get_mono_fast_ns,	"mono",		1 },
 	{ ktime_get_raw_fast_ns,	"mono_raw",	1 },
+	{ ktime_get_boot_fast_ns,	"boot",		1 },
 	ARCH_TRACE_CLOCKS
 };
 
@@ -1352,6 +1353,7 @@ void tracing_reset_all_online_cpus(void)
 
 #define SAVED_CMDLINES_DEFAULT 128
 #define NO_CMDLINE_MAP UINT_MAX
+static unsigned saved_tgids[SAVED_CMDLINES_DEFAULT];
 static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED;
 struct saved_cmdlines_buffer {
 	unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1];
@@ -1590,7 +1592,7 @@ static int trace_save_cmdline(struct task_struct *tsk)
 	}
 
 	set_cmdline(idx, tsk->comm);
-
+	saved_tgids[idx] = tsk->tgid;
 	arch_spin_unlock(&trace_cmdline_lock);
 
 	return 1;
@@ -1633,6 +1635,25 @@ void trace_find_cmdline(int pid, char comm[])
 	preempt_enable();
 }
 
+int trace_find_tgid(int pid)
+{
+	unsigned map;
+	int tgid;
+
+	preempt_disable();
+	arch_spin_lock(&trace_cmdline_lock);
+	map = savedcmd->map_pid_to_cmdline[pid];
+	if (map != NO_CMDLINE_MAP)
+		tgid = saved_tgids[map];
+	else
+		tgid = -1;
+
+	arch_spin_unlock(&trace_cmdline_lock);
+	preempt_enable();
+
+	return tgid;
+}
+
 void tracing_record_cmdline(struct task_struct *tsk)
 {
 	if (atomic_read(&trace_record_cmdline_disabled) || !tracing_is_on())
@@ -2583,6 +2604,13 @@ static void print_func_help_header(struct trace_buffer *buf, struct seq_file *m)
 		    "#              | |       |          |         |\n");
 }
 
+static void print_func_help_header_tgid(struct trace_buffer *buf, struct seq_file *m)
+{
+	print_event_info(buf, m);
+	seq_puts(m, "#           TASK-PID    TGID   CPU#      TIMESTAMP  FUNCTION\n");
+	seq_puts(m, "#              | |        |      |          |         |\n");
+}
+
 static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file *m)
 {
 	print_event_info(buf, m);
@@ -2595,6 +2623,18 @@ static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file
 		    "#              | |       |   ||||       |         |\n");
 }
 
+static void print_func_help_header_irq_tgid(struct trace_buffer *buf, struct seq_file *m)
+{
+	print_event_info(buf, m);
+	seq_puts(m, "#                                      _-----=> irqs-off\n");
+	seq_puts(m, "#                                     / _----=> need-resched\n");
+	seq_puts(m, "#                                    | / _---=> hardirq/softirq\n");
+	seq_puts(m, "#                                    || / _--=> preempt-depth\n");
+	seq_puts(m, "#                                    ||| /     delay\n");
+	seq_puts(m, "#           TASK-PID    TGID   CPU#  ||||    TIMESTAMP  FUNCTION\n");
+	seq_puts(m, "#              | |        |      |   ||||       |         |\n");
+}
+
 void
 print_trace_header(struct seq_file *m, struct trace_iterator *iter)
 {
@@ -2907,9 +2947,15 @@ void trace_default_header(struct seq_file *m)
 	} else {
 		if (!(trace_flags & TRACE_ITER_VERBOSE)) {
 			if (trace_flags & TRACE_ITER_IRQ_INFO)
-				print_func_help_header_irq(iter->trace_buffer, m);
+				if (trace_flags & TRACE_ITER_TGID)
+					print_func_help_header_irq_tgid(iter->trace_buffer, m);
+				else
+					print_func_help_header_irq(iter->trace_buffer, m);
 			else
-				print_func_help_header(iter->trace_buffer, m);
+				if (trace_flags & TRACE_ITER_TGID)
+					print_func_help_header_tgid(iter->trace_buffer, m);
+				else
+					print_func_help_header(iter->trace_buffer, m);
 		}
 	}
 }
@@ -4167,6 +4213,50 @@ static void trace_insert_enum_map(struct module *mod,
 }
 
 static ssize_t
+tracing_saved_tgids_read(struct file *file, char __user *ubuf,
+				size_t cnt, loff_t *ppos)
+{
+	char *file_buf;
+	char *buf;
+	int len = 0;
+	int pid;
+	int i;
+
+	file_buf = kmalloc(SAVED_CMDLINES_DEFAULT*(16+1+16), GFP_KERNEL);
+	if (!file_buf)
+		return -ENOMEM;
+
+	buf = file_buf;
+
+	for (i = 0; i < SAVED_CMDLINES_DEFAULT; i++) {
+		int tgid;
+		int r;
+
+		pid = savedcmd->map_cmdline_to_pid[i];
+		if (pid == -1 || pid == NO_CMDLINE_MAP)
+			continue;
+
+		tgid = trace_find_tgid(pid);
+		r = sprintf(buf, "%d %d\n", pid, tgid);
+		buf += r;
+		len += r;
+	}
+
+	len = simple_read_from_buffer(ubuf, cnt, ppos,
+				      file_buf, len);
+
+	kfree(file_buf);
+
+	return len;
+}
+
+static const struct file_operations tracing_saved_tgids_fops = {
+	.open	= tracing_open_generic,
+	.read	= tracing_saved_tgids_read,
+	.llseek	= generic_file_llseek,
+};
+
+static ssize_t
 tracing_set_trace_read(struct file *filp, char __user *ubuf,
 		       size_t cnt, loff_t *ppos)
 {
@@ -6799,6 +6889,9 @@ init_tracer_tracefs(struct trace_array *tr, struct dentry *d_tracer)
 	trace_create_file("trace_marker", 0220, d_tracer,
 			  tr, &tracing_mark_fops);
 
+	trace_create_file("saved_tgids", 0444, d_tracer,
+			  tr, &tracing_saved_tgids_fops);
+
 	trace_create_file("trace_clock", 0644, d_tracer, tr,
 			  &trace_clock_fops);
 
diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
index 919d9d07686f..e1265f95457f 100644
--- a/kernel/trace/trace.h
+++ b/kernel/trace/trace.h
@@ -656,6 +656,7 @@ static inline void __trace_stack(struct trace_array *tr, unsigned long flags,
 extern cycle_t ftrace_now(int cpu);
 
 extern void trace_find_cmdline(int pid, char comm[]);
+extern int trace_find_tgid(int pid);
 
 #ifdef CONFIG_DYNAMIC_FTRACE
 extern unsigned long ftrace_update_tot_cnt;
@@ -970,7 +971,8 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf,
 		FUNCTION_FLAGS					\
 		FGRAPH_FLAGS					\
 		STACK_FLAGS					\
-		BRANCH_FLAGS
+		BRANCH_FLAGS					\
+		C(TGID,			"print-tgid"),
 
 /*
  * By defining C, we can make TRACE_FLAGS a list of bit names
diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c
index 7fd6f5a26143..96c75b0e9831 100644
--- a/kernel/trace/trace_functions_graph.c
+++ b/kernel/trace/trace_functions_graph.c
@@ -64,6 +64,9 @@ struct fgraph_data {
 
 #define TRACE_GRAPH_INDENT	2
 
+/* Flag options */
+#define TRACE_GRAPH_PRINT_FLAT		0x80
+
 static unsigned int max_depth;
 
 static struct tracer_opt trace_opts[] = {
@@ -87,6 +90,8 @@ static struct tracer_opt trace_opts[] = {
 	{ TRACER_OPT(sleep-time, TRACE_GRAPH_SLEEP_TIME) },
 	/* Include time within nested functions */
 	{ TRACER_OPT(graph-time, TRACE_GRAPH_GRAPH_TIME) },
+	/* Use standard trace formatting rather than hierarchical */
+	{ TRACER_OPT(funcgraph-flat, TRACE_GRAPH_PRINT_FLAT) },
 	{ } /* Empty entry */
 };
 
@@ -1176,6 +1181,9 @@ print_graph_function_flags(struct trace_iterator *iter, u32 flags)
 	int cpu = iter->cpu;
 	int ret;
 
+	if (flags & TRACE_GRAPH_PRINT_FLAT)
+		return TRACE_TYPE_UNHANDLED;
+
 	if (data && per_cpu_ptr(data->cpu_data, cpu)->ignore) {
 		per_cpu_ptr(data->cpu_data, cpu)->ignore = 0;
 		return TRACE_TYPE_HANDLED;
@@ -1233,13 +1241,6 @@ print_graph_function(struct trace_iterator *iter)
 	return print_graph_function_flags(iter, tracer_flags.val);
 }
 
-static enum print_line_t
-print_graph_function_event(struct trace_iterator *iter, int flags,
-			   struct trace_event *event)
-{
-	return print_graph_function(iter);
-}
-
 static void print_lat_header(struct seq_file *s, u32 flags)
 {
 	static const char spaces[] = "                "	/* 16 spaces */
@@ -1308,6 +1309,11 @@ void print_graph_headers_flags(struct seq_file *s, u32 flags)
 	struct trace_iterator *iter = s->private;
 	struct trace_array *tr = iter->tr;
 
+	if (flags & TRACE_GRAPH_PRINT_FLAT) {
+		trace_default_header(s);
+		return;
+	}
+
 	if (!(tr->trace_flags & TRACE_ITER_CONTEXT_INFO))
 		return;
 
@@ -1389,19 +1395,6 @@ func_graph_set_flag(struct trace_array *tr, u32 old_flags, u32 bit, int set)
 	return 0;
 }
 
-static struct trace_event_functions graph_functions = {
-	.trace		= print_graph_function_event,
-};
-
-static struct trace_event graph_trace_entry_event = {
-	.type		= TRACE_GRAPH_ENT,
-	.funcs		= &graph_functions,
-};
-
-static struct trace_event graph_trace_ret_event = {
-	.type		= TRACE_GRAPH_RET,
-	.funcs		= &graph_functions
-};
 
 static struct tracer graph_trace __tracer_data = {
 	.name		= "function_graph",
@@ -1478,16 +1471,6 @@ static __init int init_graph_trace(void)
 {
 	max_bytes_for_cpu = snprintf(NULL, 0, "%d", nr_cpu_ids - 1);
 
-	if (!register_trace_event(&graph_trace_entry_event)) {
-		pr_warning("Warning: could not register graph trace events\n");
-		return 1;
-	}
-
-	if (!register_trace_event(&graph_trace_ret_event)) {
-		pr_warning("Warning: could not register graph trace events\n");
-		return 1;
-	}
-
 	return register_tracer(&graph_trace);
 }
 
diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
index be3222b7d72e..21b162c07e83 100644
--- a/kernel/trace/trace_irqsoff.c
+++ b/kernel/trace/trace_irqsoff.c
@@ -16,6 +16,10 @@
 
 #include "trace.h"
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/preemptirq.h>
+
+#if defined(CONFIG_IRQSOFF_TRACER) || defined(CONFIG_PREEMPT_TRACER)
 static struct trace_array		*irqsoff_trace __read_mostly;
 static int				tracer_enabled __read_mostly;
 
@@ -451,63 +455,43 @@ void time_hardirqs_off(unsigned long a0, unsigned long a1)
 #else /* !CONFIG_PROVE_LOCKING */
 
 /*
- * Stubs:
- */
-
-void trace_softirqs_on(unsigned long ip)
-{
-}
-
-void trace_softirqs_off(unsigned long ip)
-{
-}
-
-inline void print_irqtrace_events(struct task_struct *curr)
-{
-}
-
-/*
  * We are only interested in hardirq on/off events:
  */
-void trace_hardirqs_on(void)
+static inline void tracer_hardirqs_on(void)
 {
 	if (!preempt_trace() && irq_trace())
 		stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
-EXPORT_SYMBOL(trace_hardirqs_on);
 
-void trace_hardirqs_off(void)
+static inline void tracer_hardirqs_off(void)
 {
 	if (!preempt_trace() && irq_trace())
 		start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
 }
-EXPORT_SYMBOL(trace_hardirqs_off);
 
-__visible void trace_hardirqs_on_caller(unsigned long caller_addr)
+static inline void tracer_hardirqs_on_caller(unsigned long caller_addr)
 {
 	if (!preempt_trace() && irq_trace())
 		stop_critical_timing(CALLER_ADDR0, caller_addr);
 }
-EXPORT_SYMBOL(trace_hardirqs_on_caller);
 
-__visible void trace_hardirqs_off_caller(unsigned long caller_addr)
+static inline void tracer_hardirqs_off_caller(unsigned long caller_addr)
 {
 	if (!preempt_trace() && irq_trace())
 		start_critical_timing(CALLER_ADDR0, caller_addr);
 }
-EXPORT_SYMBOL(trace_hardirqs_off_caller);
 
 #endif /* CONFIG_PROVE_LOCKING */
 #endif /*  CONFIG_IRQSOFF_TRACER */
 
 #ifdef CONFIG_PREEMPT_TRACER
-void trace_preempt_on(unsigned long a0, unsigned long a1)
+static inline void tracer_preempt_on(unsigned long a0, unsigned long a1)
 {
 	if (preempt_trace() && !irq_trace())
 		stop_critical_timing(a0, a1);
 }
 
-void trace_preempt_off(unsigned long a0, unsigned long a1)
+static inline void tracer_preempt_off(unsigned long a0, unsigned long a1)
 {
 	if (preempt_trace() && !irq_trace())
 		start_critical_timing(a0, a1);
@@ -770,3 +754,100 @@ __init static int init_irqsoff_tracer(void)
 	return 0;
 }
 core_initcall(init_irqsoff_tracer);
+#endif /* IRQSOFF_TRACER || PREEMPTOFF_TRACER */
+
+#ifndef CONFIG_IRQSOFF_TRACER
+static inline void tracer_hardirqs_on(void) { }
+static inline void tracer_hardirqs_off(void) { }
+static inline void tracer_hardirqs_on_caller(unsigned long caller_addr) { }
+static inline void tracer_hardirqs_off_caller(unsigned long caller_addr) { }
+#endif
+
+#ifndef CONFIG_PREEMPT_TRACER
+static inline void tracer_preempt_on(unsigned long a0, unsigned long a1) { }
+static inline void tracer_preempt_off(unsigned long a0, unsigned long a1) { }
+#endif
+
+/* Per-cpu variable to prevent redundant calls when IRQs already off */
+static DEFINE_PER_CPU(int, tracing_irq_cpu);
+
+#if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PROVE_LOCKING)
+void trace_hardirqs_on(void)
+{
+	if (!this_cpu_read(tracing_irq_cpu))
+		return;
+
+	trace_irq_enable_rcuidle(CALLER_ADDR0, CALLER_ADDR1);
+	tracer_hardirqs_on();
+
+	this_cpu_write(tracing_irq_cpu, 0);
+}
+EXPORT_SYMBOL(trace_hardirqs_on);
+
+void trace_hardirqs_off(void)
+{
+	if (this_cpu_read(tracing_irq_cpu))
+		return;
+
+	this_cpu_write(tracing_irq_cpu, 1);
+
+	trace_irq_disable_rcuidle(CALLER_ADDR0, CALLER_ADDR1);
+	tracer_hardirqs_off();
+}
+EXPORT_SYMBOL(trace_hardirqs_off);
+
+__visible void trace_hardirqs_on_caller(unsigned long caller_addr)
+{
+	if (!this_cpu_read(tracing_irq_cpu))
+		return;
+
+	trace_irq_enable_rcuidle(CALLER_ADDR0, caller_addr);
+	tracer_hardirqs_on_caller(caller_addr);
+
+	this_cpu_write(tracing_irq_cpu, 0);
+}
+EXPORT_SYMBOL(trace_hardirqs_on_caller);
+
+__visible void trace_hardirqs_off_caller(unsigned long caller_addr)
+{
+	if (this_cpu_read(tracing_irq_cpu))
+		return;
+
+	this_cpu_write(tracing_irq_cpu, 1);
+
+	trace_irq_disable_rcuidle(CALLER_ADDR0, caller_addr);
+	tracer_hardirqs_off_caller(caller_addr);
+}
+EXPORT_SYMBOL(trace_hardirqs_off_caller);
+
+/*
+ * Stubs:
+ */
+
+void trace_softirqs_on(unsigned long ip)
+{
+}
+
+void trace_softirqs_off(unsigned long ip)
+{
+}
+
+inline void print_irqtrace_events(struct task_struct *curr)
+{
+}
+#endif
+
+#if defined(CONFIG_PREEMPT_TRACER) || \
+	(defined(CONFIG_DEBUG_PREEMPT) && defined(CONFIG_PREEMPTIRQ_EVENTS))
+void trace_preempt_on(unsigned long a0, unsigned long a1)
+{
+	trace_preempt_enable_rcuidle(a0, a1);
+	tracer_preempt_on(a0, a1);
+}
+
+void trace_preempt_off(unsigned long a0, unsigned long a1)
+{
+	trace_preempt_disable_rcuidle(a0, a1);
+	tracer_preempt_off(a0, a1);
+}
+#endif
diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c
index 282982195e09..3bc4b6de0f4d 100644
--- a/kernel/trace/trace_output.c
+++ b/kernel/trace/trace_output.c
@@ -526,11 +526,21 @@ int trace_print_context(struct trace_iterator *iter)
 	unsigned long long t;
 	unsigned long secs, usec_rem;
 	char comm[TASK_COMM_LEN];
+	int tgid;
 
 	trace_find_cmdline(entry->pid, comm);
 
-	trace_seq_printf(s, "%16s-%-5d [%03d] ",
-			       comm, entry->pid, iter->cpu);
+	trace_seq_printf(s, "%16s-%-5d ", comm, entry->pid);
+
+	if (tr->trace_flags & TRACE_ITER_TGID) {
+		tgid = trace_find_tgid(entry->pid);
+		if (tgid < 0)
+			trace_seq_puts(s, "(-----) ");
+		else
+			trace_seq_printf(s, "(%5d) ", tgid);
+	}
+
+	trace_seq_printf(s, "[%03d] ", iter->cpu);
 
 	if (tr->trace_flags & TRACE_ITER_IRQ_INFO)
 		trace_print_lat_fmt(s, entry);
@@ -845,6 +855,174 @@ static struct trace_event trace_fn_event = {
 	.funcs		= &trace_fn_funcs,
 };
 
+/* TRACE_GRAPH_ENT */
+static enum print_line_t trace_graph_ent_trace(struct trace_iterator *iter, int flags,
+					struct trace_event *event)
+{
+	struct trace_seq *s = &iter->seq;
+	struct ftrace_graph_ent_entry *field;
+
+	trace_assign_type(field, iter->ent);
+
+	trace_seq_puts(s, "graph_ent: func=");
+	if (trace_seq_has_overflowed(s))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	if (!seq_print_ip_sym(s, field->graph_ent.func, flags))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	trace_seq_puts(s, "\n");
+	if (trace_seq_has_overflowed(s))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static enum print_line_t trace_graph_ent_raw(struct trace_iterator *iter, int flags,
+				      struct trace_event *event)
+{
+	struct ftrace_graph_ent_entry *field;
+
+	trace_assign_type(field, iter->ent);
+
+	trace_seq_printf(&iter->seq, "%lx %d\n",
+			      field->graph_ent.func,
+			      field->graph_ent.depth);
+	if (trace_seq_has_overflowed(&iter->seq))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static enum print_line_t trace_graph_ent_hex(struct trace_iterator *iter, int flags,
+				      struct trace_event *event)
+{
+	struct ftrace_graph_ent_entry *field;
+	struct trace_seq *s = &iter->seq;
+
+	trace_assign_type(field, iter->ent);
+
+	SEQ_PUT_HEX_FIELD(s, field->graph_ent.func);
+	SEQ_PUT_HEX_FIELD(s, field->graph_ent.depth);
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static enum print_line_t trace_graph_ent_bin(struct trace_iterator *iter, int flags,
+				      struct trace_event *event)
+{
+	struct ftrace_graph_ent_entry *field;
+	struct trace_seq *s = &iter->seq;
+
+	trace_assign_type(field, iter->ent);
+
+	SEQ_PUT_FIELD(s, field->graph_ent.func);
+	SEQ_PUT_FIELD(s, field->graph_ent.depth);
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static struct trace_event_functions trace_graph_ent_funcs = {
+	.trace		= trace_graph_ent_trace,
+	.raw		= trace_graph_ent_raw,
+	.hex		= trace_graph_ent_hex,
+	.binary		= trace_graph_ent_bin,
+};
+
+static struct trace_event trace_graph_ent_event = {
+	.type		= TRACE_GRAPH_ENT,
+	.funcs		= &trace_graph_ent_funcs,
+};
+
+/* TRACE_GRAPH_RET */
+static enum print_line_t trace_graph_ret_trace(struct trace_iterator *iter, int flags,
+					struct trace_event *event)
+{
+	struct trace_seq *s = &iter->seq;
+	struct trace_entry *entry = iter->ent;
+	struct ftrace_graph_ret_entry *field;
+
+	trace_assign_type(field, entry);
+
+	trace_seq_puts(s, "graph_ret: func=");
+	if (trace_seq_has_overflowed(s))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	if (!seq_print_ip_sym(s, field->ret.func, flags))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	trace_seq_puts(s, "\n");
+	if (trace_seq_has_overflowed(s))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static enum print_line_t trace_graph_ret_raw(struct trace_iterator *iter, int flags,
+				      struct trace_event *event)
+{
+	struct ftrace_graph_ret_entry *field;
+
+	trace_assign_type(field, iter->ent);
+
+	trace_seq_printf(&iter->seq, "%lx %lld %lld %ld %d\n",
+			      field->ret.func,
+			      field->ret.calltime,
+			      field->ret.rettime,
+			      field->ret.overrun,
+			      field->ret.depth);
+	if (trace_seq_has_overflowed(&iter->seq))
+		return TRACE_TYPE_PARTIAL_LINE;
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static enum print_line_t trace_graph_ret_hex(struct trace_iterator *iter, int flags,
+				      struct trace_event *event)
+{
+	struct ftrace_graph_ret_entry *field;
+	struct trace_seq *s = &iter->seq;
+
+	trace_assign_type(field, iter->ent);
+
+	SEQ_PUT_HEX_FIELD(s, field->ret.func);
+	SEQ_PUT_HEX_FIELD(s, field->ret.calltime);
+	SEQ_PUT_HEX_FIELD(s, field->ret.rettime);
+	SEQ_PUT_HEX_FIELD(s, field->ret.overrun);
+	SEQ_PUT_HEX_FIELD(s, field->ret.depth);
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static enum print_line_t trace_graph_ret_bin(struct trace_iterator *iter, int flags,
+				      struct trace_event *event)
+{
+	struct ftrace_graph_ret_entry *field;
+	struct trace_seq *s = &iter->seq;
+
+	trace_assign_type(field, iter->ent);
+
+	SEQ_PUT_FIELD(s, field->ret.func);
+	SEQ_PUT_FIELD(s, field->ret.calltime);
+	SEQ_PUT_FIELD(s, field->ret.rettime);
+	SEQ_PUT_FIELD(s, field->ret.overrun);
+	SEQ_PUT_FIELD(s, field->ret.depth);
+
+	return TRACE_TYPE_HANDLED;
+}
+
+static struct trace_event_functions trace_graph_ret_funcs = {
+	.trace		= trace_graph_ret_trace,
+	.raw		= trace_graph_ret_raw,
+	.hex		= trace_graph_ret_hex,
+	.binary		= trace_graph_ret_bin,
+};
+
+static struct trace_event trace_graph_ret_event = {
+	.type		= TRACE_GRAPH_RET,
+	.funcs		= &trace_graph_ret_funcs,
+};
+
 /* TRACE_CTX an TRACE_WAKE */
 static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter,
 					     char *delim)
@@ -1222,6 +1400,8 @@ static struct trace_event trace_print_event = {
 
 static struct trace_event *events[] __initdata = {
 	&trace_fn_event,
+	&trace_graph_ent_event,
+	&trace_graph_ret_event,
 	&trace_ctx_event,
 	&trace_wake_event,
 	&trace_stack_event,
diff --git a/kernel/watchdog.c b/kernel/watchdog.c
index c1e0b5f429b6..b718530034d2 100644
--- a/kernel/watchdog.c
+++ b/kernel/watchdog.c
@@ -103,6 +103,11 @@ static DEFINE_PER_CPU(struct task_struct *, softlockup_task_ptr_saved);
 static DEFINE_PER_CPU(bool, hard_watchdog_warn);
 static DEFINE_PER_CPU(bool, watchdog_nmi_touch);
 static DEFINE_PER_CPU(unsigned long, hrtimer_interrupts_saved);
+#endif
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_OTHER_CPU
+static cpumask_t __read_mostly watchdog_cpus;
+#endif
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_NMI
 static DEFINE_PER_CPU(struct perf_event *, watchdog_ev);
 #endif
 static unsigned long soft_lockup_nmi_warn;
@@ -114,7 +119,7 @@ static unsigned long soft_lockup_nmi_warn;
 #ifdef CONFIG_HARDLOCKUP_DETECTOR
 unsigned int __read_mostly hardlockup_panic =
 			CONFIG_BOOTPARAM_HARDLOCKUP_PANIC_VALUE;
-static unsigned long hardlockup_allcpu_dumped;
+static unsigned long __maybe_unused hardlockup_allcpu_dumped;
 /*
  * We may not want to enable hard lockup detection by default in all cases,
  * for example when running the kernel as a guest on a hypervisor. In these
@@ -271,7 +276,7 @@ void touch_softlockup_watchdog_sync(void)
 	__this_cpu_write(watchdog_touch_ts, 0);
 }
 
-#ifdef CONFIG_HARDLOCKUP_DETECTOR
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_NMI
 /* watchdog detector functions */
 static bool is_hardlockup(void)
 {
@@ -285,6 +290,76 @@ static bool is_hardlockup(void)
 }
 #endif
 
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_OTHER_CPU
+static unsigned int watchdog_next_cpu(unsigned int cpu)
+{
+	cpumask_t cpus = watchdog_cpus;
+	unsigned int next_cpu;
+
+	next_cpu = cpumask_next(cpu, &cpus);
+	if (next_cpu >= nr_cpu_ids)
+		next_cpu = cpumask_first(&cpus);
+
+	if (next_cpu == cpu)
+		return nr_cpu_ids;
+
+	return next_cpu;
+}
+
+static int is_hardlockup_other_cpu(unsigned int cpu)
+{
+	unsigned long hrint = per_cpu(hrtimer_interrupts, cpu);
+
+	if (per_cpu(hrtimer_interrupts_saved, cpu) == hrint)
+		return 1;
+
+	per_cpu(hrtimer_interrupts_saved, cpu) = hrint;
+	return 0;
+}
+
+static void watchdog_check_hardlockup_other_cpu(void)
+{
+	unsigned int next_cpu;
+
+	/*
+	 * Test for hardlockups every 3 samples.  The sample period is
+	 *  watchdog_thresh * 2 / 5, so 3 samples gets us back to slightly over
+	 *  watchdog_thresh (over by 20%).
+	 */
+	if (__this_cpu_read(hrtimer_interrupts) % 3 != 0)
+		return;
+
+	/* check for a hardlockup on the next cpu */
+	next_cpu = watchdog_next_cpu(smp_processor_id());
+	if (next_cpu >= nr_cpu_ids)
+		return;
+
+	smp_rmb();
+
+	if (per_cpu(watchdog_nmi_touch, next_cpu) == true) {
+		per_cpu(watchdog_nmi_touch, next_cpu) = false;
+		return;
+	}
+
+	if (is_hardlockup_other_cpu(next_cpu)) {
+		/* only warn once */
+		if (per_cpu(hard_watchdog_warn, next_cpu) == true)
+			return;
+
+		if (hardlockup_panic)
+			panic("Watchdog detected hard LOCKUP on cpu %u", next_cpu);
+		else
+			WARN(1, "Watchdog detected hard LOCKUP on cpu %u", next_cpu);
+
+		per_cpu(hard_watchdog_warn, next_cpu) = true;
+	} else {
+		per_cpu(hard_watchdog_warn, next_cpu) = false;
+	}
+}
+#else
+static inline void watchdog_check_hardlockup_other_cpu(void) { return; }
+#endif
+
 static int is_softlockup(unsigned long touch_ts)
 {
 	unsigned long now = get_timestamp();
@@ -297,7 +372,7 @@ static int is_softlockup(unsigned long touch_ts)
 	return 0;
 }
 
-#ifdef CONFIG_HARDLOCKUP_DETECTOR
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_NMI
 
 static struct perf_event_attr wd_hw_attr = {
 	.type		= PERF_TYPE_HARDWARE,
@@ -359,7 +434,7 @@ static void watchdog_overflow_callback(struct perf_event *event,
 	__this_cpu_write(hard_watchdog_warn, false);
 	return;
 }
-#endif /* CONFIG_HARDLOCKUP_DETECTOR */
+#endif /* CONFIG_HARDLOCKUP_DETECTOR_NMI */
 
 static void watchdog_interrupt_count(void)
 {
@@ -383,6 +458,9 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer)
 	/* kick the hardlockup detector */
 	watchdog_interrupt_count();
 
+	/* test for hardlockups on the next cpu */
+	watchdog_check_hardlockup_other_cpu();
+
 	/* kick the softlockup detector */
 	wake_up_process(__this_cpu_read(softlockup_watchdog));
 
@@ -560,7 +638,7 @@ static void watchdog(unsigned int cpu)
 		watchdog_nmi_disable(cpu);
 }
 
-#ifdef CONFIG_HARDLOCKUP_DETECTOR
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_NMI
 /*
  * People like the simple clean cpu node info on boot.
  * Reduce the watchdog noise by only printing messages
@@ -659,9 +737,44 @@ static void watchdog_nmi_disable(unsigned int cpu)
 }
 
 #else
+#ifdef CONFIG_HARDLOCKUP_DETECTOR_OTHER_CPU
+static int watchdog_nmi_enable(unsigned int cpu)
+{
+	/*
+	 * The new cpu will be marked online before the first hrtimer interrupt
+	 * runs on it.  If another cpu tests for a hardlockup on the new cpu
+	 * before it has run its first hrtimer, it will get a false positive.
+	 * Touch the watchdog on the new cpu to delay the first check for at
+	 * least 3 sampling periods to guarantee one hrtimer has run on the new
+	 * cpu.
+	 */
+	per_cpu(watchdog_nmi_touch, cpu) = true;
+	smp_wmb();
+	cpumask_set_cpu(cpu, &watchdog_cpus);
+	return 0;
+}
+
+static void watchdog_nmi_disable(unsigned int cpu)
+{
+	unsigned int next_cpu = watchdog_next_cpu(cpu);
+
+	/*
+	 * Offlining this cpu will cause the cpu before this one to start
+	 * checking the one after this one.  If this cpu just finished checking
+	 * the next cpu and updating hrtimer_interrupts_saved, and then the
+	 * previous cpu checks it within one sample period, it will trigger a
+	 * false positive.  Touch the watchdog on the next cpu to prevent it.
+	 */
+	if (next_cpu < nr_cpu_ids)
+		per_cpu(watchdog_nmi_touch, next_cpu) = true;
+	smp_wmb();
+	cpumask_clear_cpu(cpu, &watchdog_cpus);
+}
+#else
 static int watchdog_nmi_enable(unsigned int cpu) { return 0; }
 static void watchdog_nmi_disable(unsigned int cpu) { return; }
-#endif /* CONFIG_HARDLOCKUP_DETECTOR */
+#endif /* CONFIG_HARDLOCKUP_DETECTOR_OTHER_CPU */
+#endif /* CONFIG_HARDLOCKUP_DETECTOR_NMI */
 
 static struct smp_hotplug_thread watchdog_threads = {
 	.store			= &softlockup_watchdog,
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index b53b375e14bd..c879d72bf9f7 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -707,15 +707,27 @@ config LOCKUP_DETECTOR
 	  The overhead should be minimal.  A periodic hrtimer runs to
 	  generate interrupts and kick the watchdog task every 4 seconds.
 	  An NMI is generated every 10 seconds or so to check for hardlockups.
+	  If NMIs are not available on the platform, every 12 seconds the
+	  hrtimer interrupt on one cpu will be used to check for hardlockups
+	  on the next cpu.
 
 	  The frequency of hrtimer and NMI events and the soft and hard lockup
 	  thresholds can be controlled through the sysctl watchdog_thresh.
 
-config HARDLOCKUP_DETECTOR
+config HARDLOCKUP_DETECTOR_NMI
 	def_bool y
 	depends on LOCKUP_DETECTOR && !HAVE_NMI_WATCHDOG
 	depends on PERF_EVENTS && HAVE_PERF_EVENTS_NMI
 
+config HARDLOCKUP_DETECTOR_OTHER_CPU
+	def_bool y
+	depends on LOCKUP_DETECTOR && SMP
+	depends on !HARDLOCKUP_DETECTOR_NMI && !HAVE_NMI_WATCHDOG
+
+config HARDLOCKUP_DETECTOR
+	def_bool y
+	depends on HARDLOCKUP_DETECTOR_NMI || HARDLOCKUP_DETECTOR_OTHER_CPU
+
 config BOOTPARAM_HARDLOCKUP_PANIC
 	bool "Panic (Reboot) On Hard Lockups"
 	depends on HARDLOCKUP_DETECTOR
@@ -855,6 +867,15 @@ config SCHED_INFO
 	bool
 	default n
 
+config PANIC_ON_RT_THROTTLING
+	bool "Panic on RT throttling"
+	help
+	  Say Y here to enable the kernel to panic when a realtime
+	  runqueue is throttled. This may be useful for detecting
+	  and debugging RT throttling issues.
+
+	  Say N if unsure.
+
 config SCHEDSTATS
 	bool "Collect scheduler statistics"
 	depends on DEBUG_KERNEL && PROC_FS
diff --git a/lib/Makefile b/lib/Makefile
index 7f1de26613d2..cb4f6aa95013 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -58,8 +58,6 @@ obj-$(CONFIG_HAS_IOMEM) += iomap_copy.o devres.o
 obj-$(CONFIG_CHECK_SIGNATURE) += check_signature.o
 obj-$(CONFIG_DEBUG_LOCKING_API_SELFTESTS) += locking-selftest.o
 
-GCOV_PROFILE_hweight.o := n
-CFLAGS_hweight.o = $(subst $(quote),,$(CONFIG_ARCH_HWEIGHT_CFLAGS))
 obj-$(CONFIG_GENERIC_HWEIGHT) += hweight.o
 
 obj-$(CONFIG_BTREE) += btree.o
diff --git a/lib/hweight.c b/lib/hweight.c
index 9a5c1f221558..43273a7d83cf 100644
--- a/lib/hweight.c
+++ b/lib/hweight.c
@@ -9,6 +9,7 @@
  * The Hamming Weight of a number is the total number of bits set in it.
  */
 
+#ifndef __HAVE_ARCH_SW_HWEIGHT
 unsigned int __sw_hweight32(unsigned int w)
 {
 #ifdef CONFIG_ARCH_HAS_FAST_MULTIPLIER
@@ -25,6 +26,7 @@ unsigned int __sw_hweight32(unsigned int w)
 #endif
 }
 EXPORT_SYMBOL(__sw_hweight32);
+#endif
 
 unsigned int __sw_hweight16(unsigned int w)
 {
@@ -43,6 +45,7 @@ unsigned int __sw_hweight8(unsigned int w)
 }
 EXPORT_SYMBOL(__sw_hweight8);
 
+#ifndef __HAVE_ARCH_SW_HWEIGHT
 unsigned long __sw_hweight64(__u64 w)
 {
 #if BITS_PER_LONG == 32
@@ -65,3 +68,4 @@ unsigned long __sw_hweight64(__u64 w)
 #endif
 }
 EXPORT_SYMBOL(__sw_hweight64);
+#endif
diff --git a/lib/libcrc32c.c b/lib/libcrc32c.c
index acf9da449f81..2c573f08adb7 100644
--- a/lib/libcrc32c.c
+++ b/lib/libcrc32c.c
@@ -42,7 +42,7 @@ static struct crypto_shash *tfm;
 u32 crc32c(u32 crc, const void *address, unsigned int length)
 {
 	SHASH_DESC_ON_STACK(shash, tfm);
-	u32 *ctx = (u32 *)shash_desc_ctx(shash);
+	u32 ret, *ctx = (u32 *)shash_desc_ctx(shash);
 	int err;
 
 	shash->tfm = tfm;
@@ -52,7 +52,9 @@ u32 crc32c(u32 crc, const void *address, unsigned int length)
 	err = crypto_shash_update(shash, address, length);
 	BUG_ON(err);
 
-	return *ctx;
+	ret = *ctx;
+	barrier_data(ctx);
+	return ret;
 }
 
 EXPORT_SYMBOL(crc32c);
diff --git a/lib/strncpy_from_user.c b/lib/strncpy_from_user.c
index 5a003a2ebd96..05efc1fa97f0 100644
--- a/lib/strncpy_from_user.c
+++ b/lib/strncpy_from_user.c
@@ -1,5 +1,6 @@
 #include <linux/compiler.h>
 #include <linux/export.h>
+#include <linux/thread_info.h>
 #include <linux/uaccess.h>
 #include <linux/kernel.h>
 #include <linux/errno.h>
@@ -109,6 +110,7 @@ long strncpy_from_user(char *dst, const char __user *src, long count)
 		unsigned long max = max_addr - src_addr;
 		long retval;
 
+		check_object_size(dst, count, false);
 		user_access_begin();
 		retval = do_strncpy_from_user(dst, src, count, max);
 		user_access_end();
diff --git a/mm/madvise.c b/mm/madvise.c
index c889fcbb530e..c154e1076303 100644
--- a/mm/madvise.c
+++ b/mm/madvise.c
@@ -104,7 +104,7 @@ static long madvise_behavior(struct vm_area_struct *vma,
 	pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
 	*prev = vma_merge(mm, *prev, start, end, new_flags, vma->anon_vma,
 			  vma->vm_file, pgoff, vma_policy(vma),
-			  vma->vm_userfaultfd_ctx);
+			  vma->vm_userfaultfd_ctx, vma_get_anon_name(vma));
 	if (*prev) {
 		vma = *prev;
 		goto success;
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index e25b93a4267d..484bedd8d811 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -5334,6 +5334,7 @@ struct cgroup_subsys memory_cgrp_subsys = {
 	.css_reset = mem_cgroup_css_reset,
 	.can_attach = mem_cgroup_can_attach,
 	.cancel_attach = mem_cgroup_cancel_attach,
+	.attach = mem_cgroup_move_task,
 	.post_attach = mem_cgroup_move_task,
 	.bind = mem_cgroup_bind,
 	.dfl_cftypes = memory_files,
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index c947014d128a..177668a9c267 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -720,7 +720,8 @@ static int mbind_range(struct mm_struct *mm, unsigned long start,
 			((vmstart - vma->vm_start) >> PAGE_SHIFT);
 		prev = vma_merge(mm, prev, vmstart, vmend, vma->vm_flags,
 				 vma->anon_vma, vma->vm_file, pgoff,
-				 new_pol, vma->vm_userfaultfd_ctx);
+				 new_pol, vma->vm_userfaultfd_ctx,
+				 vma_get_anon_name(vma));
 		if (prev) {
 			vma = prev;
 			next = vma->vm_next;
diff --git a/mm/mlock.c b/mm/mlock.c
index 9d2e773f3a95..206e86b98a03 100644
--- a/mm/mlock.c
+++ b/mm/mlock.c
@@ -513,7 +513,7 @@ static int mlock_fixup(struct vm_area_struct *vma, struct vm_area_struct **prev,
 	pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
 	*prev = vma_merge(mm, *prev, start, end, newflags, vma->anon_vma,
 			  vma->vm_file, pgoff, vma_policy(vma),
-			  vma->vm_userfaultfd_ctx);
+			  vma->vm_userfaultfd_ctx, vma_get_anon_name(vma));
 	if (*prev) {
 		vma = *prev;
 		goto success;
diff --git a/mm/mmap.c b/mm/mmap.c
index eaa460ddcaf9..19823fc3dcfa 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -58,6 +58,18 @@
 #define arch_rebalance_pgtables(addr, len)		(addr)
 #endif
 
+#ifdef CONFIG_HAVE_ARCH_MMAP_RND_BITS
+const int mmap_rnd_bits_min = CONFIG_ARCH_MMAP_RND_BITS_MIN;
+const int mmap_rnd_bits_max = CONFIG_ARCH_MMAP_RND_BITS_MAX;
+int mmap_rnd_bits __read_mostly = CONFIG_ARCH_MMAP_RND_BITS;
+#endif
+#ifdef CONFIG_HAVE_ARCH_MMAP_RND_COMPAT_BITS
+const int mmap_rnd_compat_bits_min = CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MIN;
+const int mmap_rnd_compat_bits_max = CONFIG_ARCH_MMAP_RND_COMPAT_BITS_MAX;
+int mmap_rnd_compat_bits __read_mostly = CONFIG_ARCH_MMAP_RND_COMPAT_BITS;
+#endif
+
+
 static void unmap_region(struct mm_struct *mm,
 		struct vm_area_struct *vma, struct vm_area_struct *prev,
 		unsigned long start, unsigned long end);
@@ -939,7 +951,8 @@ again:			remove_next = 1 + (end > next->vm_end);
  */
 static inline int is_mergeable_vma(struct vm_area_struct *vma,
 				struct file *file, unsigned long vm_flags,
-				struct vm_userfaultfd_ctx vm_userfaultfd_ctx)
+				struct vm_userfaultfd_ctx vm_userfaultfd_ctx,
+				const char __user *anon_name)
 {
 	/*
 	 * VM_SOFTDIRTY should not prevent from VMA merging, if we
@@ -957,6 +970,8 @@ static inline int is_mergeable_vma(struct vm_area_struct *vma,
 		return 0;
 	if (!is_mergeable_vm_userfaultfd_ctx(vma, vm_userfaultfd_ctx))
 		return 0;
+	if (vma_get_anon_name(vma) != anon_name)
+		return 0;
 	return 1;
 }
 
@@ -989,9 +1004,10 @@ static int
 can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags,
 		     struct anon_vma *anon_vma, struct file *file,
 		     pgoff_t vm_pgoff,
-		     struct vm_userfaultfd_ctx vm_userfaultfd_ctx)
+		     struct vm_userfaultfd_ctx vm_userfaultfd_ctx,
+		     const char __user *anon_name)
 {
-	if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx) &&
+	if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx, anon_name) &&
 	    is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) {
 		if (vma->vm_pgoff == vm_pgoff)
 			return 1;
@@ -1010,9 +1026,10 @@ static int
 can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags,
 		    struct anon_vma *anon_vma, struct file *file,
 		    pgoff_t vm_pgoff,
-		    struct vm_userfaultfd_ctx vm_userfaultfd_ctx)
+		    struct vm_userfaultfd_ctx vm_userfaultfd_ctx,
+		    const char __user *anon_name)
 {
-	if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx) &&
+	if (is_mergeable_vma(vma, file, vm_flags, vm_userfaultfd_ctx, anon_name) &&
 	    is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) {
 		pgoff_t vm_pglen;
 		vm_pglen = vma_pages(vma);
@@ -1023,9 +1040,9 @@ can_vma_merge_after(struct vm_area_struct *vma, unsigned long vm_flags,
 }
 
 /*
- * Given a mapping request (addr,end,vm_flags,file,pgoff), figure out
- * whether that can be merged with its predecessor or its successor.
- * Or both (it neatly fills a hole).
+ * Given a mapping request (addr,end,vm_flags,file,pgoff,anon_name),
+ * figure out whether that can be merged with its predecessor or its
+ * successor.  Or both (it neatly fills a hole).
  *
  * In most cases - when called for mmap, brk or mremap - [addr,end) is
  * certain not to be mapped by the time vma_merge is called; but when
@@ -1056,7 +1073,8 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm,
 			unsigned long end, unsigned long vm_flags,
 			struct anon_vma *anon_vma, struct file *file,
 			pgoff_t pgoff, struct mempolicy *policy,
-			struct vm_userfaultfd_ctx vm_userfaultfd_ctx)
+			struct vm_userfaultfd_ctx vm_userfaultfd_ctx,
+			const char __user *anon_name)
 {
 	pgoff_t pglen = (end - addr) >> PAGE_SHIFT;
 	struct vm_area_struct *area, *next;
@@ -1084,7 +1102,8 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm,
 			mpol_equal(vma_policy(prev), policy) &&
 			can_vma_merge_after(prev, vm_flags,
 					    anon_vma, file, pgoff,
-					    vm_userfaultfd_ctx)) {
+					    vm_userfaultfd_ctx,
+					    anon_name)) {
 		/*
 		 * OK, it can.  Can we now merge in the successor as well?
 		 */
@@ -1093,7 +1112,8 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm,
 				can_vma_merge_before(next, vm_flags,
 						     anon_vma, file,
 						     pgoff+pglen,
-						     vm_userfaultfd_ctx) &&
+						     vm_userfaultfd_ctx,
+						     anon_name) &&
 				is_mergeable_anon_vma(prev->anon_vma,
 						      next->anon_vma, NULL)) {
 							/* cases 1, 6 */
@@ -1115,7 +1135,8 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm,
 			mpol_equal(policy, vma_policy(next)) &&
 			can_vma_merge_before(next, vm_flags,
 					     anon_vma, file, pgoff+pglen,
-					     vm_userfaultfd_ctx)) {
+					     vm_userfaultfd_ctx,
+					     anon_name)) {
 		if (prev && addr < prev->vm_end)	/* case 4 */
 			err = vma_adjust(prev, prev->vm_start,
 				addr, prev->vm_pgoff, NULL);
@@ -1599,7 +1620,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
 	 * Can we just expand an old mapping?
 	 */
 	vma = vma_merge(mm, prev, addr, addr + len, vm_flags,
-			NULL, file, pgoff, NULL, NULL_VM_UFFD_CTX);
+			NULL, file, pgoff, NULL, NULL_VM_UFFD_CTX, NULL);
 	if (vma)
 		goto out;
 
@@ -2647,6 +2668,7 @@ int do_munmap(struct mm_struct *mm, unsigned long start, size_t len)
 
 	return 0;
 }
+EXPORT_SYMBOL(do_munmap);
 
 int vm_munmap(unsigned long start, size_t len)
 {
@@ -2824,7 +2846,7 @@ static unsigned long do_brk(unsigned long addr, unsigned long len)
 
 	/* Can we just expand an old private anonymous mapping? */
 	vma = vma_merge(mm, prev, addr, addr + len, flags,
-			NULL, NULL, pgoff, NULL, NULL_VM_UFFD_CTX);
+			NULL, NULL, pgoff, NULL, NULL_VM_UFFD_CTX, NULL);
 	if (vma)
 		goto out;
 
@@ -2982,7 +3004,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
 		return NULL;	/* should never get here */
 	new_vma = vma_merge(mm, prev, addr, addr + len, vma->vm_flags,
 			    vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma),
-			    vma->vm_userfaultfd_ctx);
+			    vma->vm_userfaultfd_ctx, vma_get_anon_name(vma));
 	if (new_vma) {
 		/*
 		 * Source vma may have been merged into new_vma
diff --git a/mm/mprotect.c b/mm/mprotect.c
index c0b4b2a49462..b8849a3930a0 100644
--- a/mm/mprotect.c
+++ b/mm/mprotect.c
@@ -294,7 +294,7 @@ mprotect_fixup(struct vm_area_struct *vma, struct vm_area_struct **pprev,
 	pgoff = vma->vm_pgoff + ((start - vma->vm_start) >> PAGE_SHIFT);
 	*pprev = vma_merge(mm, *pprev, start, end, newflags,
 			   vma->anon_vma, vma->vm_file, pgoff, vma_policy(vma),
-			   vma->vm_userfaultfd_ctx);
+			   vma->vm_userfaultfd_ctx, vma_get_anon_name(vma));
 	if (*pprev) {
 		vma = *pprev;
 		goto success;
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 417700241e52..02c2f11c0e63 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -231,9 +231,21 @@ compound_page_dtor * const compound_page_dtors[] = {
 #endif
 };
 
+/*
+ * Try to keep at least this much lowmem free.  Do not allow normal
+ * allocations below this point, only high priority ones. Automatically
+ * tuned according to the amount of memory in the system.
+ */
 int min_free_kbytes = 1024;
 int user_min_free_kbytes = -1;
 
+/*
+ * Extra memory for the system to try freeing. Used to temporarily
+ * free memory, to make space for new workloads. Anyone can allocate
+ * down to the min watermarks controlled by min_free_kbytes above.
+ */
+int extra_free_kbytes = 0;
+
 static unsigned long __meminitdata nr_kernel_pages;
 static unsigned long __meminitdata nr_all_pages;
 static unsigned long __meminitdata dma_reserve;
@@ -6071,6 +6083,7 @@ static void setup_per_zone_lowmem_reserve(void)
 static void __setup_per_zone_wmarks(void)
 {
 	unsigned long pages_min = min_free_kbytes >> (PAGE_SHIFT - 10);
+	unsigned long pages_low = extra_free_kbytes >> (PAGE_SHIFT - 10);
 	unsigned long lowmem_pages = 0;
 	struct zone *zone;
 	unsigned long flags;
@@ -6082,11 +6095,14 @@ static void __setup_per_zone_wmarks(void)
 	}
 
 	for_each_zone(zone) {
-		u64 tmp;
+		u64 min, low;
 
 		spin_lock_irqsave(&zone->lock, flags);
-		tmp = (u64)pages_min * zone->managed_pages;
-		do_div(tmp, lowmem_pages);
+		min = (u64)pages_min * zone->managed_pages;
+		do_div(min, lowmem_pages);
+		low = (u64)pages_low * zone->managed_pages;
+		do_div(low, vm_total_pages);
+
 		if (is_highmem(zone)) {
 			/*
 			 * __GFP_HIGH and PF_MEMALLOC allocations usually don't
@@ -6107,11 +6123,13 @@ static void __setup_per_zone_wmarks(void)
 			 * If it's a lowmem zone, reserve a number of pages
 			 * proportionate to the zone's size.
 			 */
-			zone->watermark[WMARK_MIN] = tmp;
+			zone->watermark[WMARK_MIN] = min;
 		}
 
-		zone->watermark[WMARK_LOW]  = min_wmark_pages(zone) + (tmp >> 2);
-		zone->watermark[WMARK_HIGH] = min_wmark_pages(zone) + (tmp >> 1);
+		zone->watermark[WMARK_LOW]  = min_wmark_pages(zone) +
+					low + (min >> 2);
+		zone->watermark[WMARK_HIGH] = min_wmark_pages(zone) +
+					low + (min >> 1);
 
 		__mod_zone_page_state(zone, NR_ALLOC_BATCH,
 			high_wmark_pages(zone) - low_wmark_pages(zone) -
@@ -6234,7 +6252,7 @@ core_initcall(init_per_zone_wmark_min)
 /*
  * min_free_kbytes_sysctl_handler - just a wrapper around proc_dointvec() so
  *	that we can call two helper functions whenever min_free_kbytes
- *	changes.
+ *	or extra_free_kbytes changes.
  */
 int min_free_kbytes_sysctl_handler(struct ctl_table *table, int write,
 	void __user *buffer, size_t *length, loff_t *ppos)
diff --git a/mm/shmem.c b/mm/shmem.c
index 1b11ccc0a3b7..79997e8cf807 100644
--- a/mm/shmem.c
+++ b/mm/shmem.c
@@ -3405,6 +3405,14 @@ struct file *shmem_file_setup(const char *name, loff_t size, unsigned long flags
 }
 EXPORT_SYMBOL_GPL(shmem_file_setup);
 
+void shmem_set_file(struct vm_area_struct *vma, struct file *file)
+{
+	if (vma->vm_file)
+		fput(vma->vm_file);
+	vma->vm_file = file;
+	vma->vm_ops = &shmem_vm_ops;
+}
+
 /**
  * shmem_zero_setup - setup a shared anonymous mapping
  * @vma: the vma to be mmapped is prepared by do_mmap_pgoff
@@ -3424,10 +3432,7 @@ int shmem_zero_setup(struct vm_area_struct *vma)
 	if (IS_ERR(file))
 		return PTR_ERR(file);
 
-	if (vma->vm_file)
-		fput(vma->vm_file);
-	vma->vm_file = file;
-	vma->vm_ops = &shmem_vm_ops;
+	shmem_set_file(vma, file);
 	return 0;
 }
 
diff --git a/mm/usercopy.c b/mm/usercopy.c
index c56b97b7c49c..b34996a3860b 100644
--- a/mm/usercopy.c
+++ b/mm/usercopy.c
@@ -15,6 +15,7 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/mm.h>
+#include <linux/sched.h>
 #include <linux/slab.h>
 #include <asm/sections.h>
 
@@ -124,7 +125,7 @@ static inline const char *check_kernel_text_object(const void *ptr,
 static inline const char *check_bogus_address(const void *ptr, unsigned long n)
 {
 	/* Reject if object wraps past end of memory. */
-	if (ptr + n < ptr)
+	if ((unsigned long)ptr + n < (unsigned long)ptr)
 		return "<wrapped address>";
 
 	/* Reject if NULL or ZERO-allocation. */
diff --git a/mm/util.c b/mm/util.c
index d5259b62f8d7..d7b1065644be 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -348,6 +348,7 @@ struct address_space *page_mapping(struct page *page)
 		return NULL;
 	return page->mapping;
 }
+EXPORT_SYMBOL(page_mapping);
 
 int overcommit_ratio_handler(struct ctl_table *table, int write,
 			     void __user *buffer, size_t *lenp,
diff --git a/mm/vmstat.c b/mm/vmstat.c
index c344e3609c53..44618235722c 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -460,7 +460,7 @@ static int fold_diff(int *diff)
  *
  * The function returns the number of global counters updated.
  */
-static int refresh_cpu_vm_stats(void)
+static int refresh_cpu_vm_stats(bool do_pagesets)
 {
 	struct zone *zone;
 	int i;
@@ -484,33 +484,35 @@ static int refresh_cpu_vm_stats(void)
 #endif
 			}
 		}
-		cond_resched();
 #ifdef CONFIG_NUMA
-		/*
-		 * Deal with draining the remote pageset of this
-		 * processor
-		 *
-		 * Check if there are pages remaining in this pageset
-		 * if not then there is nothing to expire.
-		 */
-		if (!__this_cpu_read(p->expire) ||
+		if (do_pagesets) {
+			cond_resched();
+			/*
+			 * Deal with draining the remote pageset of this
+			 * processor
+			 *
+			 * Check if there are pages remaining in this pageset
+			 * if not then there is nothing to expire.
+			 */
+			if (!__this_cpu_read(p->expire) ||
 			       !__this_cpu_read(p->pcp.count))
-			continue;
+				continue;
 
-		/*
-		 * We never drain zones local to this processor.
-		 */
-		if (zone_to_nid(zone) == numa_node_id()) {
-			__this_cpu_write(p->expire, 0);
-			continue;
-		}
+			/*
+			 * We never drain zones local to this processor.
+			 */
+			if (zone_to_nid(zone) == numa_node_id()) {
+				__this_cpu_write(p->expire, 0);
+				continue;
+			}
 
-		if (__this_cpu_dec_return(p->expire))
-			continue;
+			if (__this_cpu_dec_return(p->expire))
+				continue;
 
-		if (__this_cpu_read(p->pcp.count)) {
-			drain_zone_pages(zone, this_cpu_ptr(&p->pcp));
-			changes++;
+			if (__this_cpu_read(p->pcp.count)) {
+				drain_zone_pages(zone, this_cpu_ptr(&p->pcp));
+				changes++;
+			}
 		}
 #endif
 	}
@@ -1388,7 +1390,7 @@ static cpumask_var_t cpu_stat_off;
 
 static void vmstat_update(struct work_struct *w)
 {
-	if (refresh_cpu_vm_stats()) {
+	if (refresh_cpu_vm_stats(true)) {
 		/*
 		 * Counters were updated so we expect more updates
 		 * to occur in the future. Keep on running the
@@ -1420,6 +1422,23 @@ static void vmstat_update(struct work_struct *w)
 }
 
 /*
+ * Switch off vmstat processing and then fold all the remaining differentials
+ * until the diffs stay at zero. The function is used by NOHZ and can only be
+ * invoked when tick processing is not active.
+ */
+void quiet_vmstat(void)
+{
+	if (system_state != SYSTEM_RUNNING)
+		return;
+
+	do {
+		if (!cpumask_test_and_set_cpu(smp_processor_id(), cpu_stat_off))
+			cancel_delayed_work(this_cpu_ptr(&vmstat_work));
+
+	} while (refresh_cpu_vm_stats(false));
+}
+
+/*
  * Check if the diffs for a certain cpu indicate that
  * an update is needed.
  */
@@ -1451,7 +1470,7 @@ static bool need_update(int cpu)
  */
 static void vmstat_shepherd(struct work_struct *w);
 
-static DECLARE_DELAYED_WORK(shepherd, vmstat_shepherd);
+static DECLARE_DEFERRABLE_WORK(shepherd, vmstat_shepherd);
 
 static void vmstat_shepherd(struct work_struct *w)
 {
diff --git a/net/Kconfig b/net/Kconfig
index 127da94ae25e..ce9585cf343a 100644
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -86,6 +86,12 @@ source "net/netlabel/Kconfig"
 
 endif # if INET
 
+config ANDROID_PARANOID_NETWORK
+	bool "Only allow certain groups to create sockets"
+	default y
+	help
+		none
+
 config NETWORK_SECMARK
 	bool "Security Marking"
 	help
diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c
index 70306cc9d814..709ce9fb15f3 100644
--- a/net/bluetooth/af_bluetooth.c
+++ b/net/bluetooth/af_bluetooth.c
@@ -106,11 +106,40 @@ void bt_sock_unregister(int proto)
 }
 EXPORT_SYMBOL(bt_sock_unregister);
 
+#ifdef CONFIG_PARANOID_NETWORK
+static inline int current_has_bt_admin(void)
+{
+	return !current_euid();
+}
+
+static inline int current_has_bt(void)
+{
+	return current_has_bt_admin();
+}
+# else
+static inline int current_has_bt_admin(void)
+{
+	return 1;
+}
+
+static inline int current_has_bt(void)
+{
+	return 1;
+}
+#endif
+
 static int bt_sock_create(struct net *net, struct socket *sock, int proto,
 			  int kern)
 {
 	int err;
 
+	if (proto == BTPROTO_RFCOMM || proto == BTPROTO_SCO ||
+			proto == BTPROTO_L2CAP) {
+		if (!current_has_bt())
+			return -EPERM;
+	} else if (!current_has_bt_admin())
+		return -EPERM;
+
 	if (net != &init_net)
 		return -EAFNOSUPPORT;
 
diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c
index 2c8095a5d824..0346c215ff6a 100644
--- a/net/bridge/br_device.c
+++ b/net/bridge/br_device.c
@@ -48,16 +48,17 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev)
 		return NETDEV_TX_OK;
 	}
 
-	u64_stats_update_begin(&brstats->syncp);
-	brstats->tx_packets++;
-	brstats->tx_bytes += skb->len;
-	u64_stats_update_end(&brstats->syncp);
-
 	BR_INPUT_SKB_CB(skb)->brdev = dev;
 
 	skb_reset_mac_header(skb);
 	skb_pull(skb, ETH_HLEN);
 
+	u64_stats_update_begin(&brstats->syncp);
+	brstats->tx_packets++;
+	/* Exclude ETH_HLEN from byte stats for consistency with Rx chain */
+	brstats->tx_bytes += skb->len;
+	u64_stats_update_end(&brstats->syncp);
+
 	if (!br_allowed_ingress(br, br_vlan_group_rcu(br), skb, &vid))
 		goto out;
 
diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c
index 365de66436ac..cb744a352167 100644
--- a/net/core/fib_rules.c
+++ b/net/core/fib_rules.c
@@ -18,6 +18,11 @@
 #include <net/fib_rules.h>
 #include <net/ip_tunnels.h>
 
+static const struct fib_kuid_range fib_kuid_range_unset = {
+	KUIDT_INIT(0),
+	KUIDT_INIT(~0),
+};
+
 int fib_default_rule_add(struct fib_rules_ops *ops,
 			 u32 pref, u32 table, u32 flags)
 {
@@ -33,6 +38,7 @@ int fib_default_rule_add(struct fib_rules_ops *ops,
 	r->table = table;
 	r->flags = flags;
 	r->fr_net = ops->fro_net;
+	r->uid_range = fib_kuid_range_unset;
 
 	r->suppress_prefixlen = -1;
 	r->suppress_ifgroup = -1;
@@ -172,6 +178,34 @@ void fib_rules_unregister(struct fib_rules_ops *ops)
 }
 EXPORT_SYMBOL_GPL(fib_rules_unregister);
 
+static int uid_range_set(struct fib_kuid_range *range)
+{
+	return uid_valid(range->start) && uid_valid(range->end);
+}
+
+static struct fib_kuid_range nla_get_kuid_range(struct nlattr **tb)
+{
+	struct fib_rule_uid_range *in;
+	struct fib_kuid_range out;
+
+	in = (struct fib_rule_uid_range *)nla_data(tb[FRA_UID_RANGE]);
+
+	out.start = make_kuid(current_user_ns(), in->start);
+	out.end = make_kuid(current_user_ns(), in->end);
+
+	return out;
+}
+
+static int nla_put_uid_range(struct sk_buff *skb, struct fib_kuid_range *range)
+{
+	struct fib_rule_uid_range out = {
+		from_kuid_munged(current_user_ns(), range->start),
+		from_kuid_munged(current_user_ns(), range->end)
+	};
+
+	return nla_put(skb, FRA_UID_RANGE, sizeof(out), &out);
+}
+
 static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops,
 			  struct flowi *fl, int flags)
 {
@@ -189,6 +223,10 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops,
 	if (rule->tun_id && (rule->tun_id != fl->flowi_tun_key.tun_id))
 		goto out;
 
+	if (uid_lt(fl->flowi_uid, rule->uid_range.start) ||
+	    uid_gt(fl->flowi_uid, rule->uid_range.end))
+		goto out;
+
 	ret = ops->match(rule, fl, flags);
 out:
 	return (rule->flags & FIB_RULE_INVERT) ? !ret : ret;
@@ -371,6 +409,21 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh)
 	} else if (rule->action == FR_ACT_GOTO)
 		goto errout_free;
 
+	if (tb[FRA_UID_RANGE]) {
+		if (current_user_ns() != net->user_ns) {
+			err = -EPERM;
+			goto errout_free;
+		}
+
+		rule->uid_range = nla_get_kuid_range(tb);
+
+		if (!uid_range_set(&rule->uid_range) ||
+		    !uid_lte(rule->uid_range.start, rule->uid_range.end))
+			goto errout_free;
+	} else {
+		rule->uid_range = fib_kuid_range_unset;
+	}
+
 	err = ops->configure(rule, skb, frh, tb);
 	if (err < 0)
 		goto errout_free;
@@ -432,6 +485,7 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh)
 	struct fib_rules_ops *ops = NULL;
 	struct fib_rule *rule, *tmp;
 	struct nlattr *tb[FRA_MAX+1];
+	struct fib_kuid_range range;
 	int err = -EINVAL;
 
 	if (nlh->nlmsg_len < nlmsg_msg_size(sizeof(*frh)))
@@ -451,6 +505,14 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh)
 	if (err < 0)
 		goto errout;
 
+	if (tb[FRA_UID_RANGE]) {
+		range = nla_get_kuid_range(tb);
+		if (!uid_range_set(&range))
+			goto errout;
+	} else {
+		range = fib_kuid_range_unset;
+	}
+
 	list_for_each_entry(rule, &ops->rules_list, list) {
 		if (frh->action && (frh->action != rule->action))
 			continue;
@@ -483,6 +545,11 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh)
 		    (rule->tun_id != nla_get_be64(tb[FRA_TUN_ID])))
 			continue;
 
+		if (uid_range_set(&range) &&
+		    (!uid_eq(rule->uid_range.start, range.start) ||
+		     !uid_eq(rule->uid_range.end, range.end)))
+			continue;
+
 		if (!ops->compare(rule, frh, tb))
 			continue;
 
@@ -550,6 +617,7 @@ static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops,
 			 + nla_total_size(4) /* FRA_FWMARK */
 			 + nla_total_size(4) /* FRA_FWMASK */
 			 + nla_total_size(8); /* FRA_TUN_ID */
+			 + nla_total_size(sizeof(struct fib_kuid_range));
 
 	if (ops->nlmsg_payload)
 		payload += ops->nlmsg_payload(rule);
@@ -607,7 +675,9 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule,
 	    (rule->target &&
 	     nla_put_u32(skb, FRA_GOTO, rule->target)) ||
 	    (rule->tun_id &&
-	     nla_put_be64(skb, FRA_TUN_ID, rule->tun_id)))
+	     nla_put_be64(skb, FRA_TUN_ID, rule->tun_id)) ||
+	    (uid_range_set(&rule->uid_range) &&
+	     nla_put_uid_range(skb, &rule->uid_range)))
 		goto nla_put_failure;
 
 	if (rule->suppress_ifgroup != -1) {
diff --git a/net/core/sock.c b/net/core/sock.c
index cd12cb6fe366..410acdf9dcc2 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2397,8 +2397,11 @@ void sock_init_data(struct socket *sock, struct sock *sk)
 		sk->sk_type	=	sock->type;
 		sk->sk_wq	=	sock->wq;
 		sock->sk	=	sk;
-	} else
+		sk->sk_uid	=	SOCK_INODE(sock)->i_uid;
+	} else {
 		sk->sk_wq	=	NULL;
+		sk->sk_uid	=	make_kuid(sock_net(sk)->user_ns, 0);
+	}
 
 	rwlock_init(&sk->sk_callback_lock);
 	lockdep_set_class_and_name(&sk->sk_callback_lock,
diff --git a/net/core/sock_diag.c b/net/core/sock_diag.c
index 0c1d58d43f67..3963c3872c69 100644
--- a/net/core/sock_diag.c
+++ b/net/core/sock_diag.c
@@ -214,7 +214,7 @@ void sock_diag_unregister(const struct sock_diag_handler *hnld)
 }
 EXPORT_SYMBOL_GPL(sock_diag_unregister);
 
-static int __sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
+static int __sock_diag_cmd(struct sk_buff *skb, struct nlmsghdr *nlh)
 {
 	int err;
 	struct sock_diag_req *req = nlmsg_data(nlh);
@@ -234,8 +234,12 @@ static int __sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 	hndl = sock_diag_handlers[req->sdiag_family];
 	if (hndl == NULL)
 		err = -ENOENT;
-	else
+	else if (nlh->nlmsg_type == SOCK_DIAG_BY_FAMILY)
 		err = hndl->dump(skb, nlh);
+	else if (nlh->nlmsg_type == SOCK_DESTROY_BACKPORT && hndl->destroy)
+		err = hndl->destroy(skb, nlh);
+	else
+		err = -EOPNOTSUPP;
 	mutex_unlock(&sock_diag_table_mutex);
 
 	return err;
@@ -261,7 +265,8 @@ static int sock_diag_rcv_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 
 		return ret;
 	case SOCK_DIAG_BY_FAMILY:
-		return __sock_diag_rcv_msg(skb, nlh);
+	case SOCK_DESTROY_BACKPORT:
+		return __sock_diag_cmd(skb, nlh);
 	default:
 		return -EINVAL;
 	}
@@ -295,6 +300,18 @@ static int sock_diag_bind(struct net *net, int group)
 	return 0;
 }
 
+int sock_diag_destroy(struct sock *sk, int err)
+{
+	if (!ns_capable(sock_net(sk)->user_ns, CAP_NET_ADMIN))
+		return -EPERM;
+
+	if (!sk->sk_prot->diag_destroy)
+		return -EOPNOTSUPP;
+
+	return sk->sk_prot->diag_destroy(sk, err);
+}
+EXPORT_SYMBOL_GPL(sock_diag_destroy);
+
 static int __net_init diag_net_init(struct net *net)
 {
 	struct netlink_kernel_cfg cfg = {
diff --git a/net/ipv4/Kconfig b/net/ipv4/Kconfig
index 416dfa004cfb..c22920525e5d 100644
--- a/net/ipv4/Kconfig
+++ b/net/ipv4/Kconfig
@@ -436,6 +436,19 @@ config INET_UDP_DIAG
 	  Support for UDP socket monitoring interface used by the ss tool.
 	  If unsure, say Y.
 
+config INET_DIAG_DESTROY
+	bool "INET: allow privileged process to administratively close sockets"
+	depends on INET_DIAG
+	default n
+	---help---
+	  Provides a SOCK_DESTROY operation that allows privileged processes
+	  (e.g., a connection manager or a network administration tool such as
+	  ss) to close sockets opened by other processes. Closing a socket in
+	  this way interrupts any blocking read/write/connect operations on
+	  the socket and causes future socket calls to behave as if the socket
+	  had been disconnected.
+	  If unsure, say N.
+
 menuconfig TCP_CONG_ADVANCED
 	bool "TCP: advanced congestion control"
 	---help---
diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index c29809f765dc..854c4bfd6eea 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -16,6 +16,7 @@ obj-y     := route.o inetpeer.o protocol.o \
 
 obj-$(CONFIG_NET_IP_TUNNEL) += ip_tunnel.o
 obj-$(CONFIG_SYSCTL) += sysctl_net_ipv4.o
+obj-$(CONFIG_SYSFS) += sysfs_net_ipv4.o
 obj-$(CONFIG_PROC_FS) += proc.o
 obj-$(CONFIG_IP_MULTIPLE_TABLES) += fib_rules.o
 obj-$(CONFIG_IP_MROUTE) += ipmr.o
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index 967a47ff78a4..c31e5fa8f1a8 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c
@@ -89,6 +89,7 @@
 #include <linux/netfilter_ipv4.h>
 #include <linux/random.h>
 #include <linux/slab.h>
+#include <linux/netfilter/xt_qtaguid.h>
 
 #include <asm/uaccess.h>
 
@@ -121,6 +122,19 @@
 #endif
 #include <net/l3mdev.h>
 
+#ifdef CONFIG_ANDROID_PARANOID_NETWORK
+#include <linux/android_aid.h>
+
+static inline int current_has_network(void)
+{
+	return in_egroup_p(AID_INET) || capable(CAP_NET_RAW);
+}
+#else
+static inline int current_has_network(void)
+{
+	return 1;
+}
+#endif
 
 /* The inetsw table contains everything that inet_create needs to
  * build a new socket.
@@ -260,6 +274,9 @@ static int inet_create(struct net *net, struct socket *sock, int protocol,
 	if (protocol < 0 || protocol >= IPPROTO_MAX)
 		return -EINVAL;
 
+	if (!current_has_network())
+		return -EACCES;
+
 	sock->state = SS_UNCONNECTED;
 
 	/* Look for the requested type/protocol pair. */
@@ -308,8 +325,7 @@ lookup_protocol:
 	}
 
 	err = -EPERM;
-	if (sock->type == SOCK_RAW && !kern &&
-	    !ns_capable(net->user_ns, CAP_NET_RAW))
+	if (sock->type == SOCK_RAW && !kern && !capable(CAP_NET_RAW))
 		goto out_rcu_unlock;
 
 	sock->ops = answer->ops;
@@ -398,6 +414,9 @@ int inet_release(struct socket *sock)
 	if (sk) {
 		long timeout;
 
+#ifdef CONFIG_NETFILTER_XT_MATCH_QTAGUID
+		qtaguid_untag(sock, true);
+#endif
 		/* Applications forget to leave groups before exiting */
 		ip_mc_drop_socket(sk);
 
diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c
index 0cb240c749bf..ee94bd32d6dc 100644
--- a/net/ipv4/fib_frontend.c
+++ b/net/ipv4/fib_frontend.c
@@ -627,6 +627,7 @@ const struct nla_policy rtm_ipv4_policy[RTA_MAX + 1] = {
 	[RTA_FLOW]		= { .type = NLA_U32 },
 	[RTA_ENCAP_TYPE]	= { .type = NLA_U16 },
 	[RTA_ENCAP]		= { .type = NLA_NESTED },
+	[RTA_UID]		= { .type = NLA_U32 },
 };
 
 static int rtm_to_fib_config(struct net *net, struct sk_buff *skb,
diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
index 36e26977c908..ef2d4322aba7 100644
--- a/net/ipv4/icmp.c
+++ b/net/ipv4/icmp.c
@@ -425,6 +425,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb)
 	fl4.daddr = daddr;
 	fl4.saddr = saddr;
 	fl4.flowi4_mark = mark;
+	fl4.flowi4_uid = sock_net_uid(net, NULL);
 	fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos);
 	fl4.flowi4_proto = IPPROTO_ICMP;
 	fl4.flowi4_oif = l3mdev_master_ifindex(skb->dev);
@@ -473,6 +474,7 @@ static struct rtable *icmp_route_lookup(struct net *net,
 		      param->replyopts.opt.opt.faddr : iph->saddr);
 	fl4->saddr = saddr;
 	fl4->flowi4_mark = mark;
+	fl4->flowi4_uid = sock_net_uid(net, NULL);
 	fl4->flowi4_tos = RT_TOS(tos);
 	fl4->flowi4_proto = IPPROTO_ICMP;
 	fl4->fl4_icmp_type = type;
diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c
index 01acb94c4963..51512742f21f 100644
--- a/net/ipv4/inet_connection_sock.c
+++ b/net/ipv4/inet_connection_sock.c
@@ -422,7 +422,7 @@ struct dst_entry *inet_csk_route_req(const struct sock *sk,
 			   sk->sk_protocol, inet_sk_flowi_flags(sk),
 			   (opt && opt->opt.srr) ? opt->opt.faddr : ireq->ir_rmt_addr,
 			   ireq->ir_loc_addr, ireq->ir_rmt_port,
-			   htons(ireq->ir_num));
+			   htons(ireq->ir_num), sk->sk_uid);
 	security_req_classify_flow(req, flowi4_to_flowi(fl4));
 	rt = ip_route_output_flow(net, fl4, sk);
 	if (IS_ERR(rt))
@@ -458,7 +458,7 @@ struct dst_entry *inet_csk_route_child_sock(const struct sock *sk,
 			   sk->sk_protocol, inet_sk_flowi_flags(sk),
 			   (opt && opt->opt.srr) ? opt->opt.faddr : ireq->ir_rmt_addr,
 			   ireq->ir_loc_addr, ireq->ir_rmt_port,
-			   htons(ireq->ir_num));
+			   htons(ireq->ir_num), sk->sk_uid);
 	security_req_classify_flow(req, flowi4_to_flowi(fl4));
 	rt = ip_route_output_flow(net, fl4, sk);
 	if (IS_ERR(rt))
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index ab9f8a66615d..fcb83b2a61f0 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -44,6 +44,8 @@ struct inet_diag_entry {
 	u16 dport;
 	u16 family;
 	u16 userlocks;
+	u32 ifindex;
+	u32 mark;
 };
 
 static DEFINE_MUTEX(inet_diag_table_mutex);
@@ -96,6 +98,7 @@ static size_t inet_sk_attr_size(void)
 		+ nla_total_size(1) /* INET_DIAG_SHUTDOWN */
 		+ nla_total_size(1) /* INET_DIAG_TOS */
 		+ nla_total_size(1) /* INET_DIAG_TCLASS */
+		+ nla_total_size(4) /* INET_DIAG_MARK */
 		+ nla_total_size(sizeof(struct inet_diag_meminfo))
 		+ nla_total_size(sizeof(struct inet_diag_msg))
 		+ nla_total_size(SK_MEMINFO_VARS * sizeof(u32))
@@ -108,7 +111,8 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 		      struct sk_buff *skb, const struct inet_diag_req_v2 *req,
 		      struct user_namespace *user_ns,
 		      u32 portid, u32 seq, u16 nlmsg_flags,
-		      const struct nlmsghdr *unlh)
+		      const struct nlmsghdr *unlh,
+		      bool net_admin)
 {
 	const struct inet_sock *inet = inet_sk(sk);
 	const struct tcp_congestion_ops *ca_ops;
@@ -158,6 +162,9 @@ int inet_sk_diag_fill(struct sock *sk, struct inet_connection_sock *icsk,
 	}
 #endif
 
+	if (net_admin && nla_put_u32(skb, INET_DIAG_MARK, sk->sk_mark))
+		goto errout;
+
 	r->idiag_uid = from_kuid_munged(user_ns, sock_i_uid(sk));
 	r->idiag_inode = sock_i_ino(sk);
 
@@ -256,10 +263,11 @@ static int inet_csk_diag_fill(struct sock *sk,
 			      const struct inet_diag_req_v2 *req,
 			      struct user_namespace *user_ns,
 			      u32 portid, u32 seq, u16 nlmsg_flags,
-			      const struct nlmsghdr *unlh)
+			      const struct nlmsghdr *unlh,
+			      bool net_admin)
 {
-	return inet_sk_diag_fill(sk, inet_csk(sk), skb, req,
-				 user_ns, portid, seq, nlmsg_flags, unlh);
+	return inet_sk_diag_fill(sk, inet_csk(sk), skb, req, user_ns,
+				 portid, seq, nlmsg_flags, unlh, net_admin);
 }
 
 static int inet_twsk_diag_fill(struct sock *sk,
@@ -301,8 +309,9 @@ static int inet_twsk_diag_fill(struct sock *sk,
 
 static int inet_req_diag_fill(struct sock *sk, struct sk_buff *skb,
 			      u32 portid, u32 seq, u16 nlmsg_flags,
-			      const struct nlmsghdr *unlh)
+			      const struct nlmsghdr *unlh, bool net_admin)
 {
+	struct request_sock *reqsk = inet_reqsk(sk);
 	struct inet_diag_msg *r;
 	struct nlmsghdr *nlh;
 	long tmo;
@@ -316,7 +325,7 @@ static int inet_req_diag_fill(struct sock *sk, struct sk_buff *skb,
 	inet_diag_msg_common_fill(r, sk);
 	r->idiag_state = TCP_SYN_RECV;
 	r->idiag_timer = 1;
-	r->idiag_retrans = inet_reqsk(sk)->num_retrans;
+	r->idiag_retrans = reqsk->num_retrans;
 
 	BUILD_BUG_ON(offsetof(struct inet_request_sock, ir_cookie) !=
 		     offsetof(struct sock, sk_cookie));
@@ -328,6 +337,10 @@ static int inet_req_diag_fill(struct sock *sk, struct sk_buff *skb,
 	r->idiag_uid	= 0;
 	r->idiag_inode	= 0;
 
+	if (net_admin && nla_put_u32(skb, INET_DIAG_MARK,
+				     inet_rsk(reqsk)->ir_mark))
+		return -EMSGSIZE;
+
 	nlmsg_end(skb, nlh);
 	return 0;
 }
@@ -336,7 +349,7 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb,
 			const struct inet_diag_req_v2 *r,
 			struct user_namespace *user_ns,
 			u32 portid, u32 seq, u16 nlmsg_flags,
-			const struct nlmsghdr *unlh)
+			const struct nlmsghdr *unlh, bool net_admin)
 {
 	if (sk->sk_state == TCP_TIME_WAIT)
 		return inet_twsk_diag_fill(sk, skb, portid, seq,
@@ -344,46 +357,66 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb,
 
 	if (sk->sk_state == TCP_NEW_SYN_RECV)
 		return inet_req_diag_fill(sk, skb, portid, seq,
-					  nlmsg_flags, unlh);
+					  nlmsg_flags, unlh, net_admin);
 
 	return inet_csk_diag_fill(sk, skb, r, user_ns, portid, seq,
-				  nlmsg_flags, unlh);
+				  nlmsg_flags, unlh, net_admin);
 }
 
-int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo,
-			    struct sk_buff *in_skb,
-			    const struct nlmsghdr *nlh,
-			    const struct inet_diag_req_v2 *req)
+struct sock *inet_diag_find_one_icsk(struct net *net,
+				     struct inet_hashinfo *hashinfo,
+				     const struct inet_diag_req_v2 *req)
 {
-	struct net *net = sock_net(in_skb->sk);
-	struct sk_buff *rep;
 	struct sock *sk;
-	int err;
 
-	err = -EINVAL;
 	if (req->sdiag_family == AF_INET)
 		sk = inet_lookup(net, hashinfo, req->id.idiag_dst[0],
 				 req->id.idiag_dport, req->id.idiag_src[0],
 				 req->id.idiag_sport, req->id.idiag_if);
 #if IS_ENABLED(CONFIG_IPV6)
-	else if (req->sdiag_family == AF_INET6)
-		sk = inet6_lookup(net, hashinfo,
-				  (struct in6_addr *)req->id.idiag_dst,
-				  req->id.idiag_dport,
-				  (struct in6_addr *)req->id.idiag_src,
-				  req->id.idiag_sport,
-				  req->id.idiag_if);
+	else if (req->sdiag_family == AF_INET6) {
+		if (ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_dst) &&
+		    ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_src))
+			sk = inet_lookup(net, hashinfo, req->id.idiag_dst[3],
+					 req->id.idiag_dport, req->id.idiag_src[3],
+					 req->id.idiag_sport, req->id.idiag_if);
+		else
+			sk = inet6_lookup(net, hashinfo,
+					  (struct in6_addr *)req->id.idiag_dst,
+					  req->id.idiag_dport,
+					  (struct in6_addr *)req->id.idiag_src,
+					  req->id.idiag_sport,
+					  req->id.idiag_if);
+	}
 #endif
 	else
-		goto out_nosk;
+		return ERR_PTR(-EINVAL);
 
-	err = -ENOENT;
 	if (!sk)
-		goto out_nosk;
+		return ERR_PTR(-ENOENT);
 
-	err = sock_diag_check_cookie(sk, req->id.idiag_cookie);
-	if (err)
-		goto out;
+	if (sock_diag_check_cookie(sk, req->id.idiag_cookie)) {
+		sock_gen_put(sk);
+		return ERR_PTR(-ENOENT);
+	}
+
+	return sk;
+}
+EXPORT_SYMBOL_GPL(inet_diag_find_one_icsk);
+
+int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo,
+			    struct sk_buff *in_skb,
+			    const struct nlmsghdr *nlh,
+			    const struct inet_diag_req_v2 *req)
+{
+	struct net *net = sock_net(in_skb->sk);
+	struct sk_buff *rep;
+	struct sock *sk;
+	int err;
+
+	sk = inet_diag_find_one_icsk(net, hashinfo, req);
+	if (IS_ERR(sk))
+		return PTR_ERR(sk);
 
 	rep = nlmsg_new(inet_sk_attr_size(), GFP_KERNEL);
 	if (!rep) {
@@ -394,7 +427,8 @@ int inet_diag_dump_one_icsk(struct inet_hashinfo *hashinfo,
 	err = sk_diag_fill(sk, rep, req,
 			   sk_user_ns(NETLINK_CB(in_skb).sk),
 			   NETLINK_CB(in_skb).portid,
-			   nlh->nlmsg_seq, 0, nlh);
+			   nlh->nlmsg_seq, 0, nlh,
+			   netlink_net_capable(in_skb, CAP_NET_ADMIN));
 	if (err < 0) {
 		WARN_ON(err == -EMSGSIZE);
 		nlmsg_free(rep);
@@ -409,12 +443,11 @@ out:
 	if (sk)
 		sock_gen_put(sk);
 
-out_nosk:
 	return err;
 }
 EXPORT_SYMBOL_GPL(inet_diag_dump_one_icsk);
 
-static int inet_diag_get_exact(struct sk_buff *in_skb,
+static int inet_diag_cmd_exact(int cmd, struct sk_buff *in_skb,
 			       const struct nlmsghdr *nlh,
 			       const struct inet_diag_req_v2 *req)
 {
@@ -424,8 +457,12 @@ static int inet_diag_get_exact(struct sk_buff *in_skb,
 	handler = inet_diag_lock_handler(req->sdiag_protocol);
 	if (IS_ERR(handler))
 		err = PTR_ERR(handler);
-	else
+	else if (cmd == SOCK_DIAG_BY_FAMILY)
 		err = handler->dump_one(in_skb, nlh, req);
+	else if (cmd == SOCK_DESTROY_BACKPORT && handler->destroy)
+		err = handler->destroy(in_skb, req);
+	else
+		err = -EOPNOTSUPP;
 	inet_diag_unlock_handler(handler);
 
 	return err;
@@ -529,6 +566,22 @@ static int inet_diag_bc_run(const struct nlattr *_bc,
 			yes = 0;
 			break;
 		}
+		case INET_DIAG_BC_DEV_COND: {
+			u32 ifindex;
+
+			ifindex = *((const u32 *)(op + 1));
+			if (ifindex != entry->ifindex)
+				yes = 0;
+			break;
+		}
+		case INET_DIAG_BC_MARK_COND: {
+			struct inet_diag_markcond *cond;
+
+			cond = (struct inet_diag_markcond *)(op + 1);
+			if ((entry->mark & cond->mask) != cond->mark)
+				yes = 0;
+			break;
+		}
 		}
 
 		if (yes) {
@@ -571,7 +624,14 @@ int inet_diag_bc_sk(const struct nlattr *bc, struct sock *sk)
 	entry_fill_addrs(&entry, sk);
 	entry.sport = inet->inet_num;
 	entry.dport = ntohs(inet->inet_dport);
+	entry.ifindex = sk->sk_bound_dev_if;
 	entry.userlocks = sk_fullsock(sk) ? sk->sk_userlocks : 0;
+	if (sk_fullsock(sk))
+		entry.mark = sk->sk_mark;
+	else if (sk->sk_state == TCP_NEW_SYN_RECV)
+		entry.mark = inet_rsk(inet_reqsk(sk))->ir_mark;
+	else
+		entry.mark = 0;
 
 	return inet_diag_bc_run(bc, &entry);
 }
@@ -594,6 +654,17 @@ static int valid_cc(const void *bc, int len, int cc)
 	return 0;
 }
 
+/* data is u32 ifindex */
+static bool valid_devcond(const struct inet_diag_bc_op *op, int len,
+			  int *min_len)
+{
+	/* Check ifindex space. */
+	*min_len += sizeof(u32);
+	if (len < *min_len)
+		return false;
+
+	return true;
+}
 /* Validate an inet_diag_hostcond. */
 static bool valid_hostcond(const struct inet_diag_bc_op *op, int len,
 			   int *min_len)
@@ -643,10 +714,25 @@ static bool valid_port_comparison(const struct inet_diag_bc_op *op,
 	return true;
 }
 
-static int inet_diag_bc_audit(const void *bytecode, int bytecode_len)
+static bool valid_markcond(const struct inet_diag_bc_op *op, int len,
+			   int *min_len)
+{
+	*min_len += sizeof(struct inet_diag_markcond);
+	return len >= *min_len;
+}
+
+static int inet_diag_bc_audit(const struct nlattr *attr,
+			      const struct sk_buff *skb)
 {
-	const void *bc = bytecode;
-	int  len = bytecode_len;
+	bool net_admin = netlink_net_capable(skb, CAP_NET_ADMIN);
+	const void *bytecode, *bc;
+	int bytecode_len, len;
+
+	if (!attr || nla_len(attr) < sizeof(struct inet_diag_bc_op))
+		return -EINVAL;
+
+	bytecode = bc = nla_data(attr);
+	len = bytecode_len = nla_len(attr);
 
 	while (len > 0) {
 		int min_len = sizeof(struct inet_diag_bc_op);
@@ -658,6 +744,10 @@ static int inet_diag_bc_audit(const void *bytecode, int bytecode_len)
 			if (!valid_hostcond(bc, len, &min_len))
 				return -EINVAL;
 			break;
+		case INET_DIAG_BC_DEV_COND:
+			if (!valid_devcond(bc, len, &min_len))
+				return -EINVAL;
+			break;
 		case INET_DIAG_BC_S_GE:
 		case INET_DIAG_BC_S_LE:
 		case INET_DIAG_BC_D_GE:
@@ -665,6 +755,12 @@ static int inet_diag_bc_audit(const void *bytecode, int bytecode_len)
 			if (!valid_port_comparison(bc, len, &min_len))
 				return -EINVAL;
 			break;
+		case INET_DIAG_BC_MARK_COND:
+			if (!net_admin)
+				return -EPERM;
+			if (!valid_markcond(bc, len, &min_len))
+				return -EINVAL;
+			break;
 		case INET_DIAG_BC_AUTO:
 		case INET_DIAG_BC_JMP:
 		case INET_DIAG_BC_NOP:
@@ -693,7 +789,8 @@ static int inet_csk_diag_dump(struct sock *sk,
 			      struct sk_buff *skb,
 			      struct netlink_callback *cb,
 			      const struct inet_diag_req_v2 *r,
-			      const struct nlattr *bc)
+			      const struct nlattr *bc,
+			      bool net_admin)
 {
 	if (!inet_diag_bc_sk(bc, sk))
 		return 0;
@@ -701,7 +798,8 @@ static int inet_csk_diag_dump(struct sock *sk,
 	return inet_csk_diag_fill(sk, skb, r,
 				  sk_user_ns(NETLINK_CB(cb->skb).sk),
 				  NETLINK_CB(cb->skb).portid,
-				  cb->nlh->nlmsg_seq, NLM_F_MULTI, cb->nlh);
+				  cb->nlh->nlmsg_seq, NLM_F_MULTI, cb->nlh,
+				  net_admin);
 }
 
 static void twsk_build_assert(void)
@@ -737,6 +835,7 @@ void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb,
 	struct net *net = sock_net(skb->sk);
 	int i, num, s_i, s_num;
 	u32 idiag_states = r->idiag_states;
+	bool net_admin = netlink_net_capable(cb->skb, CAP_NET_ADMIN);
 
 	if (idiag_states & TCPF_SYN_RECV)
 		idiag_states |= TCPF_NEW_SYN_RECV;
@@ -778,7 +877,8 @@ void inet_diag_dump_icsk(struct inet_hashinfo *hashinfo, struct sk_buff *skb,
 				    cb->args[3] > 0)
 					goto next_listen;
 
-				if (inet_csk_diag_dump(sk, skb, cb, r, bc) < 0) {
+				if (inet_csk_diag_dump(sk, skb, cb, r,
+						       bc, net_admin) < 0) {
 					spin_unlock_bh(&ilb->lock);
 					goto done;
 				}
@@ -846,7 +946,7 @@ skip_listen_ht:
 					   sk_user_ns(NETLINK_CB(cb->skb).sk),
 					   NETLINK_CB(cb->skb).portid,
 					   cb->nlh->nlmsg_seq, NLM_F_MULTI,
-					   cb->nlh);
+					   cb->nlh, net_admin);
 			if (res < 0) {
 				spin_unlock_bh(lock);
 				goto done;
@@ -938,7 +1038,7 @@ static int inet_diag_get_exact_compat(struct sk_buff *in_skb,
 	req.idiag_states = rc->idiag_states;
 	req.id = rc->id;
 
-	return inet_diag_get_exact(in_skb, nlh, &req);
+	return inet_diag_cmd_exact(SOCK_DIAG_BY_FAMILY, in_skb, nlh, &req);
 }
 
 static int inet_diag_rcv_msg_compat(struct sk_buff *skb, struct nlmsghdr *nlh)
@@ -953,13 +1053,13 @@ static int inet_diag_rcv_msg_compat(struct sk_buff *skb, struct nlmsghdr *nlh)
 	if (nlh->nlmsg_flags & NLM_F_DUMP) {
 		if (nlmsg_attrlen(nlh, hdrlen)) {
 			struct nlattr *attr;
+			int err;
 
 			attr = nlmsg_find_attr(nlh, hdrlen,
 					       INET_DIAG_REQ_BYTECODE);
-			if (!attr ||
-			    nla_len(attr) < sizeof(struct inet_diag_bc_op) ||
-			    inet_diag_bc_audit(nla_data(attr), nla_len(attr)))
-				return -EINVAL;
+			err = inet_diag_bc_audit(attr, skb);
+			if (err)
+				return err;
 		}
 		{
 			struct netlink_dump_control c = {
@@ -972,7 +1072,7 @@ static int inet_diag_rcv_msg_compat(struct sk_buff *skb, struct nlmsghdr *nlh)
 	return inet_diag_get_exact_compat(skb, nlh);
 }
 
-static int inet_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h)
+static int inet_diag_handler_cmd(struct sk_buff *skb, struct nlmsghdr *h)
 {
 	int hdrlen = sizeof(struct inet_diag_req_v2);
 	struct net *net = sock_net(skb->sk);
@@ -980,16 +1080,17 @@ static int inet_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h)
 	if (nlmsg_len(h) < hdrlen)
 		return -EINVAL;
 
-	if (h->nlmsg_flags & NLM_F_DUMP) {
+	if (h->nlmsg_type == SOCK_DIAG_BY_FAMILY &&
+	    h->nlmsg_flags & NLM_F_DUMP) {
 		if (nlmsg_attrlen(h, hdrlen)) {
 			struct nlattr *attr;
+			int err;
 
 			attr = nlmsg_find_attr(h, hdrlen,
 					       INET_DIAG_REQ_BYTECODE);
-			if (!attr ||
-			    nla_len(attr) < sizeof(struct inet_diag_bc_op) ||
-			    inet_diag_bc_audit(nla_data(attr), nla_len(attr)))
-				return -EINVAL;
+			err = inet_diag_bc_audit(attr, skb);
+			if (err)
+				return err;
 		}
 		{
 			struct netlink_dump_control c = {
@@ -999,7 +1100,7 @@ static int inet_diag_handler_dump(struct sk_buff *skb, struct nlmsghdr *h)
 		}
 	}
 
-	return inet_diag_get_exact(skb, h, nlmsg_data(h));
+	return inet_diag_cmd_exact(h->nlmsg_type, skb, h, nlmsg_data(h));
 }
 
 static
@@ -1050,14 +1151,16 @@ int inet_diag_handler_get_info(struct sk_buff *skb, struct sock *sk)
 
 static const struct sock_diag_handler inet_diag_handler = {
 	.family = AF_INET,
-	.dump = inet_diag_handler_dump,
+	.dump = inet_diag_handler_cmd,
 	.get_info = inet_diag_handler_get_info,
+	.destroy = inet_diag_handler_cmd,
 };
 
 static const struct sock_diag_handler inet6_diag_handler = {
 	.family = AF_INET6,
-	.dump = inet_diag_handler_dump,
+	.dump = inet_diag_handler_cmd,
 	.get_info = inet_diag_handler_get_info,
+	.destroy = inet_diag_handler_cmd,
 };
 
 int inet_diag_register(const struct inet_diag_handler *h)
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 09c73dd541c5..62e41d38da78 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -1583,7 +1583,8 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb,
 			   RT_SCOPE_UNIVERSE, ip_hdr(skb)->protocol,
 			   ip_reply_arg_flowi_flags(arg),
 			   daddr, saddr,
-			   tcp_hdr(skb)->source, tcp_hdr(skb)->dest);
+			   tcp_hdr(skb)->source, tcp_hdr(skb)->dest,
+			   arg->uid);
 	security_skb_classify_flow(skb, flowi4_to_flowi(&fl4));
 	rt = ip_route_output_key(net, &fl4);
 	if (IS_ERR(rt))
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index 37a3b05d175c..a989aba861e0 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c
@@ -798,7 +798,8 @@ static int ping_v4_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 
 	flowi4_init_output(&fl4, ipc.oif, sk->sk_mark, tos,
 			   RT_SCOPE_UNIVERSE, sk->sk_protocol,
-			   inet_sk_flowi_flags(sk), faddr, saddr, 0, 0);
+			   inet_sk_flowi_flags(sk), faddr, saddr, 0, 0,
+			   sk->sk_uid);
 
 	security_sk_classify_flow(sk, flowi4_to_flowi(&fl4));
 	rt = ip_route_output_flow(net, &fl4, sk);
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 8f2cd7d09720..ca1031411aa7 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -604,7 +604,7 @@ static int raw_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 			   inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol,
 			   inet_sk_flowi_flags(sk) |
 			    (inet->hdrincl ? FLOWI_FLAG_KNOWN_NH : 0),
-			   daddr, saddr, 0, 0);
+			   daddr, saddr, 0, 0, sk->sk_uid);
 
 	if (!saddr && ipc.oif) {
 		err = l3mdev_get_saddr(net, ipc.oif, &fl4);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 0294f7c99c85..5bdc0caa7f4c 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -501,7 +501,8 @@ void __ip_select_ident(struct net *net, struct iphdr *iph, int segs)
 }
 EXPORT_SYMBOL(__ip_select_ident);
 
-static void __build_flow_key(struct flowi4 *fl4, const struct sock *sk,
+static void __build_flow_key(const struct net *net, struct flowi4 *fl4,
+			     const struct sock *sk,
 			     const struct iphdr *iph,
 			     int oif, u8 tos,
 			     u8 prot, u32 mark, int flow_flags)
@@ -517,19 +518,21 @@ static void __build_flow_key(struct flowi4 *fl4, const struct sock *sk,
 	flowi4_init_output(fl4, oif, mark, tos,
 			   RT_SCOPE_UNIVERSE, prot,
 			   flow_flags,
-			   iph->daddr, iph->saddr, 0, 0);
+			   iph->daddr, iph->saddr, 0, 0,
+			   sock_net_uid(net, sk));
 }
 
 static void build_skb_flow_key(struct flowi4 *fl4, const struct sk_buff *skb,
 			       const struct sock *sk)
 {
+	const struct net *net = dev_net(skb->dev);
 	const struct iphdr *iph = ip_hdr(skb);
 	int oif = skb->dev->ifindex;
 	u8 tos = RT_TOS(iph->tos);
 	u8 prot = iph->protocol;
 	u32 mark = skb->mark;
 
-	__build_flow_key(fl4, sk, iph, oif, tos, prot, mark, 0);
+	__build_flow_key(net, fl4, sk, iph, oif, tos, prot, mark, 0);
 }
 
 static void build_sk_flow_key(struct flowi4 *fl4, const struct sock *sk)
@@ -546,7 +549,7 @@ static void build_sk_flow_key(struct flowi4 *fl4, const struct sock *sk)
 			   RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE,
 			   inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol,
 			   inet_sk_flowi_flags(sk),
-			   daddr, inet->inet_saddr, 0, 0);
+			   daddr, inet->inet_saddr, 0, 0, sk->sk_uid);
 	rcu_read_unlock();
 }
 
@@ -789,6 +792,7 @@ static void ip_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buf
 	struct rtable *rt;
 	struct flowi4 fl4;
 	const struct iphdr *iph = (const struct iphdr *) skb->data;
+	struct net *net = dev_net(skb->dev);
 	int oif = skb->dev->ifindex;
 	u8 tos = RT_TOS(iph->tos);
 	u8 prot = iph->protocol;
@@ -796,7 +800,7 @@ static void ip_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buf
 
 	rt = (struct rtable *) dst;
 
-	__build_flow_key(&fl4, sk, iph, oif, tos, prot, mark, 0);
+	__build_flow_key(net, &fl4, sk, iph, oif, tos, prot, mark, 0);
 	__ip_do_redirect(rt, skb, &fl4, true);
 }
 
@@ -1014,7 +1018,7 @@ void ipv4_update_pmtu(struct sk_buff *skb, struct net *net, u32 mtu,
 	if (!mark)
 		mark = IP4_REPLY_MARK(net, skb->mark);
 
-	__build_flow_key(&fl4, NULL, iph, oif,
+	__build_flow_key(net, &fl4, NULL, iph, oif,
 			 RT_TOS(iph->tos), protocol, mark, flow_flags);
 	rt = __ip_route_output_key(net, &fl4);
 	if (!IS_ERR(rt)) {
@@ -1030,7 +1034,7 @@ static void __ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
 	struct flowi4 fl4;
 	struct rtable *rt;
 
-	__build_flow_key(&fl4, sk, iph, 0, 0, 0, 0, 0);
+	__build_flow_key(sock_net(sk), &fl4, sk, iph, 0, 0, 0, 0, 0);
 
 	if (!fl4.flowi4_mark)
 		fl4.flowi4_mark = IP4_REPLY_MARK(sock_net(sk), skb->mark);
@@ -1049,6 +1053,7 @@ void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
 	struct rtable *rt;
 	struct dst_entry *odst = NULL;
 	bool new = false;
+	struct net *net = sock_net(sk);
 
 	bh_lock_sock(sk);
 
@@ -1062,7 +1067,7 @@ void ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu)
 		goto out;
 	}
 
-	__build_flow_key(&fl4, sk, iph, 0, 0, 0, 0, 0);
+	__build_flow_key(net, &fl4, sk, iph, 0, 0, 0, 0, 0);
 
 	rt = (struct rtable *)odst;
 	if (odst->obsolete && !odst->ops->check(odst, 0)) {
@@ -1102,7 +1107,7 @@ void ipv4_redirect(struct sk_buff *skb, struct net *net,
 	struct flowi4 fl4;
 	struct rtable *rt;
 
-	__build_flow_key(&fl4, NULL, iph, oif,
+	__build_flow_key(net, &fl4, NULL, iph, oif,
 			 RT_TOS(iph->tos), protocol, mark, flow_flags);
 	rt = __ip_route_output_key(net, &fl4);
 	if (!IS_ERR(rt)) {
@@ -1117,9 +1122,10 @@ void ipv4_sk_redirect(struct sk_buff *skb, struct sock *sk)
 	const struct iphdr *iph = (const struct iphdr *) skb->data;
 	struct flowi4 fl4;
 	struct rtable *rt;
+	struct net *net = sock_net(sk);
 
-	__build_flow_key(&fl4, sk, iph, 0, 0, 0, 0, 0);
-	rt = __ip_route_output_key(sock_net(sk), &fl4);
+	__build_flow_key(net, &fl4, sk, iph, 0, 0, 0, 0, 0);
+	rt = __ip_route_output_key(net, &fl4);
 	if (!IS_ERR(rt)) {
 		__ip_do_redirect(rt, skb, &fl4, false);
 		ip_rt_put(rt);
@@ -2495,6 +2501,11 @@ static int rt_fill_info(struct net *net,  __be32 dst, __be32 src, u32 table_id,
 	    nla_put_u32(skb, RTA_MARK, fl4->flowi4_mark))
 		goto nla_put_failure;
 
+	if (!uid_eq(fl4->flowi4_uid, INVALID_UID) &&
+	    nla_put_u32(skb, RTA_UID,
+			from_kuid_munged(current_user_ns(), fl4->flowi4_uid)))
+		goto nla_put_failure;
+
 	error = rt->dst.error;
 
 	if (rt_is_input_route(rt)) {
@@ -2547,6 +2558,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh)
 	int mark;
 	struct sk_buff *skb;
 	u32 table_id = RT_TABLE_MAIN;
+	kuid_t uid;
 
 	err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv4_policy);
 	if (err < 0)
@@ -2574,6 +2586,10 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh)
 	dst = tb[RTA_DST] ? nla_get_in_addr(tb[RTA_DST]) : 0;
 	iif = tb[RTA_IIF] ? nla_get_u32(tb[RTA_IIF]) : 0;
 	mark = tb[RTA_MARK] ? nla_get_u32(tb[RTA_MARK]) : 0;
+	if (tb[RTA_UID])
+		uid = make_kuid(current_user_ns(), nla_get_u32(tb[RTA_UID]));
+	else
+		uid = (iif ? INVALID_UID : current_uid());
 
 	memset(&fl4, 0, sizeof(fl4));
 	fl4.daddr = dst;
@@ -2581,6 +2597,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh)
 	fl4.flowi4_tos = rtm->rtm_tos;
 	fl4.flowi4_oif = tb[RTA_OIF] ? nla_get_u32(tb[RTA_OIF]) : 0;
 	fl4.flowi4_mark = mark;
+	fl4.flowi4_uid = uid;
 
 	if (netif_index_is_l3_master(net, fl4.flowi4_oif))
 		fl4.flowi4_flags = FLOWI_FLAG_L3MDEV_SRC | FLOWI_FLAG_SKIP_NH_OIF;
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index c22a74374a9c..57bbcd5b650a 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -376,7 +376,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb)
 			   RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, IPPROTO_TCP,
 			   inet_sk_flowi_flags(sk),
 			   opt->srr ? opt->faddr : ireq->ir_rmt_addr,
-			   ireq->ir_loc_addr, th->source, th->dest);
+			   ireq->ir_loc_addr, th->source, th->dest, sk->sk_uid);
 	security_req_classify_flow(req, flowi4_to_flowi(&fl4));
 	rt = ip_route_output_key(sock_net(sk), &fl4);
 	if (IS_ERR(rt)) {
diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
index 70fb352e317f..46123369144f 100644
--- a/net/ipv4/sysctl_net_ipv4.c
+++ b/net/ipv4/sysctl_net_ipv4.c
@@ -152,6 +152,21 @@ static int ipv4_ping_group_range(struct ctl_table *table, int write,
 	return ret;
 }
 
+/* Validate changes from /proc interface. */
+static int proc_tcp_default_init_rwnd(struct ctl_table *ctl, int write,
+				      void __user *buffer,
+				      size_t *lenp, loff_t *ppos)
+{
+	int old_value = *(int *)ctl->data;
+	int ret = proc_dointvec(ctl, write, buffer, lenp, ppos);
+	int new_value = *(int *)ctl->data;
+
+	if (write && ret == 0 && (new_value < 3 || new_value > 100))
+		*(int *)ctl->data = old_value;
+
+	return ret;
+}
+
 static int proc_tcp_congestion_control(struct ctl_table *ctl, int write,
 				       void __user *buffer, size_t *lenp, loff_t *ppos)
 {
@@ -761,6 +776,13 @@ static struct ctl_table ipv4_table[] = {
 		.proc_handler	= proc_dointvec_ms_jiffies,
 	},
 	{
+		.procname       = "tcp_default_init_rwnd",
+		.data           = &sysctl_tcp_default_init_rwnd,
+		.maxlen         = sizeof(int),
+		.mode           = 0644,
+		.proc_handler   = proc_tcp_default_init_rwnd
+	},
+	{
 		.procname	= "icmp_msgs_per_sec",
 		.data		= &sysctl_icmp_msgs_per_sec,
 		.maxlen		= sizeof(int),
diff --git a/net/ipv4/sysfs_net_ipv4.c b/net/ipv4/sysfs_net_ipv4.c
new file mode 100644
index 000000000000..0cbbf10026a6
--- /dev/null
+++ b/net/ipv4/sysfs_net_ipv4.c
@@ -0,0 +1,88 @@
+/*
+ * net/ipv4/sysfs_net_ipv4.c
+ *
+ * sysfs-based networking knobs (so we can, unlike with sysctl, control perms)
+ *
+ * Copyright (C) 2008 Google, Inc.
+ *
+ * Robert Love <rlove@google.com>
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+#include <linux/kobject.h>
+#include <linux/string.h>
+#include <linux/sysfs.h>
+#include <linux/init.h>
+#include <net/tcp.h>
+
+#define CREATE_IPV4_FILE(_name, _var) \
+static ssize_t _name##_show(struct kobject *kobj, \
+			    struct kobj_attribute *attr, char *buf) \
+{ \
+	return sprintf(buf, "%d\n", _var); \
+} \
+static ssize_t _name##_store(struct kobject *kobj, \
+			     struct kobj_attribute *attr, \
+			     const char *buf, size_t count) \
+{ \
+	int val, ret; \
+	ret = sscanf(buf, "%d", &val); \
+	if (ret != 1) \
+		return -EINVAL; \
+	if (val < 0) \
+		return -EINVAL; \
+	_var = val; \
+	return count; \
+} \
+static struct kobj_attribute _name##_attr = \
+	__ATTR(_name, 0644, _name##_show, _name##_store)
+
+CREATE_IPV4_FILE(tcp_wmem_min, sysctl_tcp_wmem[0]);
+CREATE_IPV4_FILE(tcp_wmem_def, sysctl_tcp_wmem[1]);
+CREATE_IPV4_FILE(tcp_wmem_max, sysctl_tcp_wmem[2]);
+
+CREATE_IPV4_FILE(tcp_rmem_min, sysctl_tcp_rmem[0]);
+CREATE_IPV4_FILE(tcp_rmem_def, sysctl_tcp_rmem[1]);
+CREATE_IPV4_FILE(tcp_rmem_max, sysctl_tcp_rmem[2]);
+
+static struct attribute *ipv4_attrs[] = {
+	&tcp_wmem_min_attr.attr,
+	&tcp_wmem_def_attr.attr,
+	&tcp_wmem_max_attr.attr,
+	&tcp_rmem_min_attr.attr,
+	&tcp_rmem_def_attr.attr,
+	&tcp_rmem_max_attr.attr,
+	NULL
+};
+
+static struct attribute_group ipv4_attr_group = {
+	.attrs = ipv4_attrs,
+};
+
+static __init int sysfs_ipv4_init(void)
+{
+	struct kobject *ipv4_kobject;
+	int ret;
+
+	ipv4_kobject = kobject_create_and_add("ipv4", kernel_kobj);
+	if (!ipv4_kobject)
+		return -ENOMEM;
+
+	ret = sysfs_create_group(ipv4_kobject, &ipv4_attr_group);
+	if (ret) {
+		kobject_put(ipv4_kobject);
+		return ret;
+	}
+
+	return 0;
+}
+
+subsys_initcall(sysfs_ipv4_init);
diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c
index 5597120c8ffd..34b5bf9e7406 100644
--- a/net/ipv4/tcp.c
+++ b/net/ipv4/tcp.c
@@ -3100,6 +3100,52 @@ void tcp_done(struct sock *sk)
 }
 EXPORT_SYMBOL_GPL(tcp_done);
 
+int tcp_abort(struct sock *sk, int err)
+{
+	if (!sk_fullsock(sk)) {
+		if (sk->sk_state == TCP_NEW_SYN_RECV) {
+			struct request_sock *req = inet_reqsk(sk);
+
+			local_bh_disable();
+			inet_csk_reqsk_queue_drop_and_put(req->rsk_listener,
+							  req);
+			local_bh_enable();
+			return 0;
+		}
+		sock_gen_put(sk);
+		return -EOPNOTSUPP;
+	}
+
+	/* Don't race with userspace socket closes such as tcp_close. */
+	lock_sock(sk);
+
+	if (sk->sk_state == TCP_LISTEN) {
+		tcp_set_state(sk, TCP_CLOSE);
+		inet_csk_listen_stop(sk);
+	}
+
+	/* Don't race with BH socket closes such as inet_csk_listen_stop. */
+	local_bh_disable();
+	bh_lock_sock(sk);
+
+	if (!sock_flag(sk, SOCK_DEAD)) {
+		sk->sk_err = err;
+		/* This barrier is coupled with smp_rmb() in tcp_poll() */
+		smp_wmb();
+		sk->sk_error_report(sk);
+		if (tcp_need_reset(sk->sk_state))
+			tcp_send_active_reset(sk, GFP_ATOMIC);
+		tcp_done(sk);
+	}
+
+	bh_unlock_sock(sk);
+	local_bh_enable();
+	release_sock(sk);
+	sock_put(sk);
+	return 0;
+}
+EXPORT_SYMBOL_GPL(tcp_abort);
+
 extern struct tcp_congestion_ops tcp_reno;
 
 static __initdata unsigned long thash_entries;
diff --git a/net/ipv4/tcp_diag.c b/net/ipv4/tcp_diag.c
index b31604086edd..4d610934fb39 100644
--- a/net/ipv4/tcp_diag.c
+++ b/net/ipv4/tcp_diag.c
@@ -10,6 +10,8 @@
  */
 
 #include <linux/module.h>
+#include <linux/net.h>
+#include <linux/sock_diag.h>
 #include <linux/inet_diag.h>
 
 #include <linux/tcp.h>
@@ -46,12 +48,29 @@ static int tcp_diag_dump_one(struct sk_buff *in_skb, const struct nlmsghdr *nlh,
 	return inet_diag_dump_one_icsk(&tcp_hashinfo, in_skb, nlh, req);
 }
 
+#ifdef CONFIG_INET_DIAG_DESTROY
+static int tcp_diag_destroy(struct sk_buff *in_skb,
+			    const struct inet_diag_req_v2 *req)
+{
+	struct net *net = sock_net(in_skb->sk);
+	struct sock *sk = inet_diag_find_one_icsk(net, &tcp_hashinfo, req);
+
+	if (IS_ERR(sk))
+		return PTR_ERR(sk);
+
+	return sock_diag_destroy(sk, ECONNABORTED);
+}
+#endif
+
 static const struct inet_diag_handler tcp_diag_handler = {
 	.dump		 = tcp_diag_dump,
 	.dump_one	 = tcp_diag_dump_one,
 	.idiag_get_info	 = tcp_diag_get_info,
 	.idiag_type	 = IPPROTO_TCP,
 	.idiag_info_size = sizeof(struct tcp_info),
+#ifdef CONFIG_INET_DIAG_DESTROY
+	.destroy	 = tcp_diag_destroy,
+#endif
 };
 
 static int __init tcp_diag_init(void)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 9e8d70160d20..f409a12d6ec1 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -102,6 +102,7 @@ int sysctl_tcp_thin_dupack __read_mostly;
 int sysctl_tcp_moderate_rcvbuf __read_mostly = 1;
 int sysctl_tcp_early_retrans __read_mostly = 3;
 int sysctl_tcp_invalid_ratelimit __read_mostly = HZ/2;
+int sysctl_tcp_default_init_rwnd __read_mostly = TCP_INIT_CWND * 2;
 
 #define FLAG_DATA		0x01 /* Incoming frame contained data.		*/
 #define FLAG_WIN_UPDATE		0x02 /* Incoming ACK was a window update.	*/
diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
index a5d790c13ef5..8ad606f0bdf6 100644
--- a/net/ipv4/tcp_ipv4.c
+++ b/net/ipv4/tcp_ipv4.c
@@ -691,6 +691,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
 		arg.bound_dev_if = sk->sk_bound_dev_if;
 
 	arg.tos = ip_hdr(skb)->tos;
+	arg.uid = sock_net_uid(net, sk && sk_fullsock(sk) ? sk : NULL);
 	ip_send_unicast_reply(*this_cpu_ptr(net->ipv4.tcp_sk),
 			      skb, &TCP_SKB_CB(skb)->header.h4.opt,
 			      ip_hdr(skb)->saddr, ip_hdr(skb)->daddr,
@@ -712,7 +713,7 @@ release_sk1:
    outside socket context is ugly, certainly. What can I do?
  */
 
-static void tcp_v4_send_ack(struct net *net,
+static void tcp_v4_send_ack(const struct sock *sk,
 			    struct sk_buff *skb, u32 seq, u32 ack,
 			    u32 win, u32 tsval, u32 tsecr, int oif,
 			    struct tcp_md5sig_key *key,
@@ -727,6 +728,7 @@ static void tcp_v4_send_ack(struct net *net,
 #endif
 			];
 	} rep;
+	struct net *net = sock_net(sk);
 	struct ip_reply_arg arg;
 
 	memset(&rep.th, 0, sizeof(struct tcphdr));
@@ -776,6 +778,7 @@ static void tcp_v4_send_ack(struct net *net,
 	if (oif)
 		arg.bound_dev_if = oif;
 	arg.tos = tos;
+	arg.uid = sock_net_uid(net, sk_fullsock(sk) ? sk : NULL);
 	ip_send_unicast_reply(*this_cpu_ptr(net->ipv4.tcp_sk),
 			      skb, &TCP_SKB_CB(skb)->header.h4.opt,
 			      ip_hdr(skb)->saddr, ip_hdr(skb)->daddr,
@@ -789,7 +792,7 @@ static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
 	struct inet_timewait_sock *tw = inet_twsk(sk);
 	struct tcp_timewait_sock *tcptw = tcp_twsk(sk);
 
-	tcp_v4_send_ack(sock_net(sk), skb,
+	tcp_v4_send_ack(sk, skb,
 			tcptw->tw_snd_nxt, tcptw->tw_rcv_nxt,
 			tcptw->tw_rcv_wnd >> tw->tw_rcv_wscale,
 			tcp_time_stamp + tcptw->tw_ts_offset,
@@ -817,7 +820,7 @@ static void tcp_v4_reqsk_send_ack(const struct sock *sk, struct sk_buff *skb,
 	 * exception of <SYN> segments, MUST be right-shifted by
 	 * Rcv.Wind.Shift bits:
 	 */
-	tcp_v4_send_ack(sock_net(sk), skb, seq,
+	tcp_v4_send_ack(sk, skb, seq,
 			tcp_rsk(req)->rcv_nxt,
 			req->rsk_rcv_wnd >> inet_rsk(req)->rcv_wscale,
 			tcp_time_stamp,
@@ -2379,6 +2382,7 @@ struct proto tcp_prot = {
 	.destroy_cgroup		= tcp_destroy_cgroup,
 	.proto_cgroup		= tcp_proto_cgroup,
 #endif
+	.diag_destroy		= tcp_abort,
 };
 EXPORT_SYMBOL(tcp_prot);
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 39c2919fe0d3..7d82c172db78 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -191,7 +191,7 @@ u32 tcp_default_init_rwnd(u32 mss)
 	 * (RFC 3517, Section 4, NextSeg() rule (2)). Further place a
 	 * limit when mss is larger than 1460.
 	 */
-	u32 init_rwnd = TCP_INIT_CWND * 2;
+	u32 init_rwnd = sysctl_tcp_default_init_rwnd;
 
 	if (mss > 1460)
 		init_rwnd = max((1460 * init_rwnd) / mss, 2U);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 301e60829c7e..9ee5087b9b5e 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c
@@ -1025,7 +1025,8 @@ int udp_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		flowi4_init_output(fl4, ipc.oif, sk->sk_mark, tos,
 				   RT_SCOPE_UNIVERSE, sk->sk_protocol,
 				   flow_flags,
-				   faddr, saddr, dport, inet->inet_sport);
+				   faddr, saddr, dport, inet->inet_sport,
+				   sk->sk_uid);
 
 		if (!saddr && ipc.oif) {
 			err = l3mdev_get_saddr(net, ipc.oif, fl4);
@@ -2264,6 +2265,20 @@ unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait)
 }
 EXPORT_SYMBOL(udp_poll);
 
+int udp_abort(struct sock *sk, int err)
+{
+	lock_sock(sk);
+
+	sk->sk_err = err;
+	sk->sk_error_report(sk);
+	udp_disconnect(sk, 0);
+
+	release_sock(sk);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(udp_abort);
+
 struct proto udp_prot = {
 	.name		   = "UDP",
 	.owner		   = THIS_MODULE,
@@ -2295,6 +2310,7 @@ struct proto udp_prot = {
 	.compat_getsockopt = compat_udp_getsockopt,
 #endif
 	.clear_sk	   = sk_prot_clear_portaddr_nulls,
+	.diag_destroy	   = udp_abort,
 };
 EXPORT_SYMBOL(udp_prot);
 
diff --git a/net/ipv4/udp_diag.c b/net/ipv4/udp_diag.c
index 6116604bf6e8..092aa60e8b92 100644
--- a/net/ipv4/udp_diag.c
+++ b/net/ipv4/udp_diag.c
@@ -20,7 +20,7 @@
 static int sk_diag_dump(struct sock *sk, struct sk_buff *skb,
 			struct netlink_callback *cb,
 			const struct inet_diag_req_v2 *req,
-			struct nlattr *bc)
+			struct nlattr *bc, bool net_admin)
 {
 	if (!inet_diag_bc_sk(bc, sk))
 		return 0;
@@ -28,7 +28,7 @@ static int sk_diag_dump(struct sock *sk, struct sk_buff *skb,
 	return inet_sk_diag_fill(sk, NULL, skb, req,
 			sk_user_ns(NETLINK_CB(cb->skb).sk),
 			NETLINK_CB(cb->skb).portid,
-			cb->nlh->nlmsg_seq, NLM_F_MULTI, cb->nlh);
+			cb->nlh->nlmsg_seq, NLM_F_MULTI, cb->nlh, net_admin);
 }
 
 static int udp_dump_one(struct udp_table *tbl, struct sk_buff *in_skb,
@@ -75,7 +75,8 @@ static int udp_dump_one(struct udp_table *tbl, struct sk_buff *in_skb,
 	err = inet_sk_diag_fill(sk, NULL, rep, req,
 			   sk_user_ns(NETLINK_CB(in_skb).sk),
 			   NETLINK_CB(in_skb).portid,
-			   nlh->nlmsg_seq, 0, nlh);
+			   nlh->nlmsg_seq, 0, nlh,
+			   netlink_net_capable(in_skb, CAP_NET_ADMIN));
 	if (err < 0) {
 		WARN_ON(err == -EMSGSIZE);
 		kfree_skb(rep);
@@ -98,6 +99,7 @@ static void udp_dump(struct udp_table *table, struct sk_buff *skb,
 {
 	int num, s_num, slot, s_slot;
 	struct net *net = sock_net(skb->sk);
+	bool net_admin = netlink_net_capable(cb->skb, CAP_NET_ADMIN);
 
 	s_slot = cb->args[0];
 	num = s_num = cb->args[1];
@@ -132,7 +134,7 @@ static void udp_dump(struct udp_table *table, struct sk_buff *skb,
 			    r->id.idiag_dport)
 				goto next;
 
-			if (sk_diag_dump(sk, skb, cb, r, bc) < 0) {
+			if (sk_diag_dump(sk, skb, cb, r, bc, net_admin) < 0) {
 				spin_unlock_bh(&hslot->lock);
 				goto done;
 			}
@@ -165,12 +167,88 @@ static void udp_diag_get_info(struct sock *sk, struct inet_diag_msg *r,
 	r->idiag_wqueue = sk_wmem_alloc_get(sk);
 }
 
+#ifdef CONFIG_INET_DIAG_DESTROY
+static int __udp_diag_destroy(struct sk_buff *in_skb,
+			      const struct inet_diag_req_v2 *req,
+			      struct udp_table *tbl)
+{
+	struct net *net = sock_net(in_skb->sk);
+	struct sock *sk;
+	int err;
+
+	rcu_read_lock();
+
+	if (req->sdiag_family == AF_INET)
+		sk = __udp4_lib_lookup(net,
+				req->id.idiag_dst[0], req->id.idiag_dport,
+				req->id.idiag_src[0], req->id.idiag_sport,
+				req->id.idiag_if, tbl);
+#if IS_ENABLED(CONFIG_IPV6)
+	else if (req->sdiag_family == AF_INET6) {
+		if (ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_dst) &&
+		    ipv6_addr_v4mapped((struct in6_addr *)req->id.idiag_src))
+			sk = __udp4_lib_lookup(net,
+					req->id.idiag_dst[3], req->id.idiag_dport,
+					req->id.idiag_src[3], req->id.idiag_sport,
+					req->id.idiag_if, tbl);
+
+		else
+			sk = __udp6_lib_lookup(net,
+					(struct in6_addr *)req->id.idiag_dst,
+					req->id.idiag_dport,
+					(struct in6_addr *)req->id.idiag_src,
+					req->id.idiag_sport,
+					req->id.idiag_if, tbl);
+	}
+#endif
+	else {
+		rcu_read_unlock();
+		return -EINVAL;
+	}
+
+	if (sk && !atomic_inc_not_zero(&sk->sk_refcnt))
+		sk = NULL;
+
+	rcu_read_unlock();
+
+	if (!sk)
+		return -ENOENT;
+
+	if (sock_diag_check_cookie(sk, req->id.idiag_cookie)) {
+		sock_put(sk);
+		return -ENOENT;
+	}
+
+	err = sock_diag_destroy(sk, ECONNABORTED);
+
+	sock_put(sk);
+
+	return err;
+}
+
+static int udp_diag_destroy(struct sk_buff *in_skb,
+			    const struct inet_diag_req_v2 *req)
+{
+	return __udp_diag_destroy(in_skb, req, &udp_table);
+}
+
+static int udplite_diag_destroy(struct sk_buff *in_skb,
+				const struct inet_diag_req_v2 *req)
+{
+	return __udp_diag_destroy(in_skb, req, &udplite_table);
+}
+
+#endif
+
 static const struct inet_diag_handler udp_diag_handler = {
 	.dump		 = udp_diag_dump,
 	.dump_one	 = udp_diag_dump_one,
 	.idiag_get_info  = udp_diag_get_info,
 	.idiag_type	 = IPPROTO_UDP,
 	.idiag_info_size = 0,
+#ifdef CONFIG_INET_DIAG_DESTROY
+	.destroy	 = udp_diag_destroy,
+#endif
 };
 
 static void udplite_diag_dump(struct sk_buff *skb, struct netlink_callback *cb,
@@ -192,6 +270,9 @@ static const struct inet_diag_handler udplite_diag_handler = {
 	.idiag_get_info  = udp_diag_get_info,
 	.idiag_type	 = IPPROTO_UDPLITE,
 	.idiag_info_size = 0,
+#ifdef CONFIG_INET_DIAG_DESTROY
+	.destroy	 = udplite_diag_destroy,
+#endif
 };
 
 static int __init udp_diag_init(void)
diff --git a/net/ipv4/xfrm4_policy.c b/net/ipv4/xfrm4_policy.c
index 7b0edb37a115..39eebc7b2831 100644
--- a/net/ipv4/xfrm4_policy.c
+++ b/net/ipv4/xfrm4_policy.c
@@ -22,14 +22,16 @@ static struct xfrm_policy_afinfo xfrm4_policy_afinfo;
 static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4,
 					    int tos, int oif,
 					    const xfrm_address_t *saddr,
-					    const xfrm_address_t *daddr)
+					    const xfrm_address_t *daddr,
+					    u32 mark)
 {
 	struct rtable *rt;
 
 	memset(fl4, 0, sizeof(*fl4));
 	fl4->daddr = daddr->a4;
 	fl4->flowi4_tos = tos;
-	fl4->flowi4_oif = oif;
+	fl4->flowi4_oif = l3mdev_master_ifindex_by_index(net, oif);
+	fl4->flowi4_mark = mark;
 	if (saddr)
 		fl4->saddr = saddr->a4;
 
@@ -44,20 +46,22 @@ static struct dst_entry *__xfrm4_dst_lookup(struct net *net, struct flowi4 *fl4,
 
 static struct dst_entry *xfrm4_dst_lookup(struct net *net, int tos, int oif,
 					  const xfrm_address_t *saddr,
-					  const xfrm_address_t *daddr)
+					  const xfrm_address_t *daddr,
+					  u32 mark)
 {
 	struct flowi4 fl4;
 
-	return __xfrm4_dst_lookup(net, &fl4, tos, oif, saddr, daddr);
+	return __xfrm4_dst_lookup(net, &fl4, tos, oif, saddr, daddr, mark);
 }
 
 static int xfrm4_get_saddr(struct net *net, int oif,
-			   xfrm_address_t *saddr, xfrm_address_t *daddr)
+			   xfrm_address_t *saddr, xfrm_address_t *daddr,
+			   u32 mark)
 {
 	struct dst_entry *dst;
 	struct flowi4 fl4;
 
-	dst = __xfrm4_dst_lookup(net, &fl4, 0, oif, NULL, daddr);
+	dst = __xfrm4_dst_lookup(net, &fl4, 0, oif, NULL, daddr, mark);
 	if (IS_ERR(dst))
 		return -EHOSTUNREACH;
 
diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c
index 92174881844d..e73b3ebe5b9a 100644
--- a/net/ipv6/addrconf.c
+++ b/net/ipv6/addrconf.c
@@ -112,6 +112,27 @@ static inline u32 cstamp_delta(unsigned long cstamp)
 	return (cstamp - INITIAL_JIFFIES) * 100UL / HZ;
 }
 
+static inline s32 rfc3315_s14_backoff_init(s32 irt)
+{
+	/* multiply 'initial retransmission time' by 0.9 .. 1.1 */
+	u64 tmp = (900000 + prandom_u32() % 200001) * (u64)irt;
+	do_div(tmp, 1000000);
+	return (s32)tmp;
+}
+
+static inline s32 rfc3315_s14_backoff_update(s32 rt, s32 mrt)
+{
+	/* multiply 'retransmission timeout' by 1.9 .. 2.1 */
+	u64 tmp = (1900000 + prandom_u32() % 200001) * (u64)rt;
+	do_div(tmp, 1000000);
+	if ((s32)tmp > mrt) {
+		/* multiply 'maximum retransmission time' by 0.9 .. 1.1 */
+		tmp = (900000 + prandom_u32() % 200001) * (u64)mrt;
+		do_div(tmp, 1000000);
+	}
+	return (s32)tmp;
+}
+
 #ifdef CONFIG_SYSCTL
 static int addrconf_sysctl_register(struct inet6_dev *idev);
 static void addrconf_sysctl_unregister(struct inet6_dev *idev);
@@ -187,6 +208,7 @@ static struct ipv6_devconf ipv6_devconf __read_mostly = {
 	.dad_transmits		= 1,
 	.rtr_solicits		= MAX_RTR_SOLICITATIONS,
 	.rtr_solicit_interval	= RTR_SOLICITATION_INTERVAL,
+	.rtr_solicit_max_interval = RTR_SOLICITATION_MAX_INTERVAL,
 	.rtr_solicit_delay	= MAX_RTR_SOLICITATION_DELAY,
 	.use_tempaddr		= 0,
 	.temp_valid_lft		= TEMP_VALID_LIFETIME,
@@ -202,9 +224,11 @@ static struct ipv6_devconf ipv6_devconf __read_mostly = {
 	.accept_ra_rtr_pref	= 1,
 	.rtr_probe_interval	= 60 * HZ,
 #ifdef CONFIG_IPV6_ROUTE_INFO
+	.accept_ra_rt_info_min_plen = 0,
 	.accept_ra_rt_info_max_plen = 0,
 #endif
 #endif
+	.accept_ra_rt_table	= 0,
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
@@ -231,6 +255,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 	.dad_transmits		= 1,
 	.rtr_solicits		= MAX_RTR_SOLICITATIONS,
 	.rtr_solicit_interval	= RTR_SOLICITATION_INTERVAL,
+	.rtr_solicit_max_interval = RTR_SOLICITATION_MAX_INTERVAL,
 	.rtr_solicit_delay	= MAX_RTR_SOLICITATION_DELAY,
 	.use_tempaddr		= 0,
 	.temp_valid_lft		= TEMP_VALID_LIFETIME,
@@ -246,9 +271,11 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = {
 	.accept_ra_rtr_pref	= 1,
 	.rtr_probe_interval	= 60 * HZ,
 #ifdef CONFIG_IPV6_ROUTE_INFO
+	.accept_ra_rt_info_min_plen = 0,
 	.accept_ra_rt_info_max_plen = 0,
 #endif
 #endif
+	.accept_ra_rt_table	= 0,
 	.proxy_ndp		= 0,
 	.accept_source_route	= 0,	/* we do not accept RH0 by default. */
 	.disable_ipv6		= 0,
@@ -2146,6 +2173,31 @@ static void  __ipv6_try_regen_rndid(struct inet6_dev *idev, struct in6_addr *tmp
 		__ipv6_regen_rndid(idev);
 }
 
+u32 addrconf_rt_table(const struct net_device *dev, u32 default_table) {
+	/* Determines into what table to put autoconf PIO/RIO/default routes
+	 * learned on this device.
+	 *
+	 * - If 0, use the same table for every device. This puts routes into
+	 *   one of RT_TABLE_{PREFIX,INFO,DFLT} depending on the type of route
+	 *   (but note that these three are currently all equal to
+	 *   RT6_TABLE_MAIN).
+	 * - If > 0, use the specified table.
+	 * - If < 0, put routes into table dev->ifindex + (-rt_table).
+	 */
+	struct inet6_dev *idev = in6_dev_get(dev);
+	u32 table;
+	int sysctl = idev->cnf.accept_ra_rt_table;
+	if (sysctl == 0) {
+		table = default_table;
+	} else if (sysctl > 0) {
+		table = (u32) sysctl;
+	} else {
+		table = (unsigned) dev->ifindex + (-sysctl);
+	}
+	in6_dev_put(idev);
+	return table;
+}
+
 /*
  *	Add prefix route.
  */
@@ -2155,7 +2207,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev,
 		      unsigned long expires, u32 flags)
 {
 	struct fib6_config cfg = {
-		.fc_table = l3mdev_fib_table(dev) ? : RT6_TABLE_PREFIX,
+		.fc_table = l3mdev_fib_table(dev) ? : addrconf_rt_table(dev, RT6_TABLE_PREFIX),
 		.fc_metric = IP6_RT_PRIO_ADDRCONF,
 		.fc_ifindex = dev->ifindex,
 		.fc_expires = expires,
@@ -2188,7 +2240,7 @@ static struct rt6_info *addrconf_get_prefix_route(const struct in6_addr *pfx,
 	struct fib6_node *fn;
 	struct rt6_info *rt = NULL;
 	struct fib6_table *table;
-	u32 tb_id = l3mdev_fib_table(dev) ? : RT6_TABLE_PREFIX;
+	u32 tb_id = l3mdev_fib_table(dev) ? : addrconf_rt_table(dev, RT6_TABLE_PREFIX);
 
 	table = fib6_get_table(dev_net(dev), tb_id);
 	if (!table)
@@ -3466,7 +3518,7 @@ static void addrconf_rs_timer(unsigned long data)
 	if (idev->if_flags & IF_RA_RCVD)
 		goto out;
 
-	if (idev->rs_probes++ < idev->cnf.rtr_solicits) {
+	if (idev->rs_probes++ < idev->cnf.rtr_solicits || idev->cnf.rtr_solicits < 0) {
 		write_unlock(&idev->lock);
 		if (!ipv6_get_lladdr(dev, &lladdr, IFA_F_TENTATIVE))
 			ndisc_send_rs(dev, &lladdr,
@@ -3475,11 +3527,13 @@ static void addrconf_rs_timer(unsigned long data)
 			goto put;
 
 		write_lock(&idev->lock);
+		idev->rs_interval = rfc3315_s14_backoff_update(
+			idev->rs_interval, idev->cnf.rtr_solicit_max_interval);
 		/* The wait after the last probe can be shorter */
 		addrconf_mod_rs_timer(idev, (idev->rs_probes ==
 					     idev->cnf.rtr_solicits) ?
 				      idev->cnf.rtr_solicit_delay :
-				      idev->cnf.rtr_solicit_interval);
+				      idev->rs_interval);
 	} else {
 		/*
 		 * Note: we do not support deprecated "all on-link"
@@ -3707,7 +3761,7 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp)
 	send_mld = ifp->scope == IFA_LINK && ipv6_lonely_lladdr(ifp);
 	send_rs = send_mld &&
 		  ipv6_accept_ra(ifp->idev) &&
-		  ifp->idev->cnf.rtr_solicits > 0 &&
+		  ifp->idev->cnf.rtr_solicits != 0 &&
 		  (dev->flags&IFF_LOOPBACK) == 0;
 	read_unlock_bh(&ifp->idev->lock);
 
@@ -3729,10 +3783,11 @@ static void addrconf_dad_completed(struct inet6_ifaddr *ifp)
 
 		write_lock_bh(&ifp->idev->lock);
 		spin_lock(&ifp->lock);
+		ifp->idev->rs_interval = rfc3315_s14_backoff_init(
+			ifp->idev->cnf.rtr_solicit_interval);
 		ifp->idev->rs_probes = 1;
 		ifp->idev->if_flags |= IF_RS_SENT;
-		addrconf_mod_rs_timer(ifp->idev,
-				      ifp->idev->cnf.rtr_solicit_interval);
+		addrconf_mod_rs_timer(ifp->idev, ifp->idev->rs_interval);
 		spin_unlock(&ifp->lock);
 		write_unlock_bh(&ifp->idev->lock);
 	}
@@ -4649,6 +4704,8 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_RTR_SOLICITS] = cnf->rtr_solicits;
 	array[DEVCONF_RTR_SOLICIT_INTERVAL] =
 		jiffies_to_msecs(cnf->rtr_solicit_interval);
+	array[DEVCONF_RTR_SOLICIT_MAX_INTERVAL] =
+		jiffies_to_msecs(cnf->rtr_solicit_max_interval);
 	array[DEVCONF_RTR_SOLICIT_DELAY] =
 		jiffies_to_msecs(cnf->rtr_solicit_delay);
 	array[DEVCONF_FORCE_MLD_VERSION] = cnf->force_mld_version;
@@ -4670,9 +4727,11 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf,
 	array[DEVCONF_RTR_PROBE_INTERVAL] =
 		jiffies_to_msecs(cnf->rtr_probe_interval);
 #ifdef CONFIG_IPV6_ROUTE_INFO
+	array[DEVCONF_ACCEPT_RA_RT_INFO_MIN_PLEN] = cnf->accept_ra_rt_info_min_plen;
 	array[DEVCONF_ACCEPT_RA_RT_INFO_MAX_PLEN] = cnf->accept_ra_rt_info_max_plen;
 #endif
 #endif
+	array[DEVCONF_ACCEPT_RA_RT_TABLE] = cnf->accept_ra_rt_table;
 	array[DEVCONF_PROXY_NDP] = cnf->proxy_ndp;
 	array[DEVCONF_ACCEPT_SOURCE_ROUTE] = cnf->accept_source_route;
 #ifdef CONFIG_IPV6_OPTIMISTIC_DAD
@@ -4856,7 +4915,7 @@ static int inet6_set_iftoken(struct inet6_dev *idev, struct in6_addr *token)
 		return -EINVAL;
 	if (!ipv6_accept_ra(idev))
 		return -EINVAL;
-	if (idev->cnf.rtr_solicits <= 0)
+	if (idev->cnf.rtr_solicits == 0)
 		return -EINVAL;
 
 	write_lock_bh(&idev->lock);
@@ -4881,8 +4940,10 @@ static int inet6_set_iftoken(struct inet6_dev *idev, struct in6_addr *token)
 
 	if (update_rs) {
 		idev->if_flags |= IF_RS_SENT;
+		idev->rs_interval = rfc3315_s14_backoff_init(
+			idev->cnf.rtr_solicit_interval);
 		idev->rs_probes = 1;
-		addrconf_mod_rs_timer(idev, idev->cnf.rtr_solicit_interval);
+		addrconf_mod_rs_timer(idev, idev->rs_interval);
 	}
 
 	/* Well, that's kinda nasty ... */
@@ -5520,6 +5581,13 @@ static struct addrconf_sysctl_table
 			.proc_handler	= proc_dointvec_jiffies,
 		},
 		{
+			.procname	= "router_solicitation_max_interval",
+			.data		= &ipv6_devconf.rtr_solicit_max_interval,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= proc_dointvec_jiffies,
+		},
+		{
 			.procname	= "router_solicitation_delay",
 			.data		= &ipv6_devconf.rtr_solicit_delay,
 			.maxlen		= sizeof(int),
@@ -5629,6 +5697,13 @@ static struct addrconf_sysctl_table
 		},
 #ifdef CONFIG_IPV6_ROUTE_INFO
 		{
+			.procname	= "accept_ra_rt_info_min_plen",
+			.data		= &ipv6_devconf.accept_ra_rt_info_min_plen,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= proc_dointvec,
+		},
+		{
 			.procname	= "accept_ra_rt_info_max_plen",
 			.data		= &ipv6_devconf.accept_ra_rt_info_max_plen,
 			.maxlen		= sizeof(int),
@@ -5638,6 +5713,13 @@ static struct addrconf_sysctl_table
 #endif
 #endif
 		{
+			.procname	= "accept_ra_rt_table",
+			.data		= &ipv6_devconf.accept_ra_rt_table,
+			.maxlen		= sizeof(int),
+			.mode		= 0644,
+			.proc_handler	= proc_dointvec,
+		},
+		{
 			.procname	= "proxy_ndp",
 			.data		= &ipv6_devconf.proxy_ndp,
 			.maxlen		= sizeof(int),
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 9f5137cd604e..1604163c2850 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c
@@ -64,6 +64,20 @@
 #include <asm/uaccess.h>
 #include <linux/mroute6.h>
 
+#ifdef CONFIG_ANDROID_PARANOID_NETWORK
+#include <linux/android_aid.h>
+
+static inline int current_has_network(void)
+{
+	return in_egroup_p(AID_INET) || capable(CAP_NET_RAW);
+}
+#else
+static inline int current_has_network(void)
+{
+	return 1;
+}
+#endif
+
 MODULE_AUTHOR("Cast of dozens");
 MODULE_DESCRIPTION("IPv6 protocol stack for Linux");
 MODULE_LICENSE("GPL");
@@ -112,6 +126,9 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol,
 	if (protocol < 0 || protocol >= IPPROTO_MAX)
 		return -EINVAL;
 
+	if (!current_has_network())
+		return -EACCES;
+
 	/* Look for the requested type/protocol pair. */
 lookup_protocol:
 	err = -ESOCKTNOSUPPORT;
@@ -158,8 +175,7 @@ lookup_protocol:
 	}
 
 	err = -EPERM;
-	if (sock->type == SOCK_RAW && !kern &&
-	    !ns_capable(net->user_ns, CAP_NET_RAW))
+	if (sock->type == SOCK_RAW && !kern && !capable(CAP_NET_RAW))
 		goto out_rcu_unlock;
 
 	sock->ops = answer->ops;
@@ -662,6 +678,7 @@ int inet6_sk_rebuild_header(struct sock *sk)
 		fl6.flowi6_mark = sk->sk_mark;
 		fl6.fl6_dport = inet->inet_dport;
 		fl6.fl6_sport = inet->inet_sport;
+		fl6.flowi6_uid = sk->sk_uid;
 		security_sk_classify_flow(sk, flowi6_to_flowi(&fl6));
 
 		rcu_read_lock();
diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c
index 0630a4d5daaa..189eb10b742d 100644
--- a/net/ipv6/ah6.c
+++ b/net/ipv6/ah6.c
@@ -662,9 +662,10 @@ static int ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		return 0;
 
 	if (type == NDISC_REDIRECT)
-		ip6_redirect(skb, net, skb->dev->ifindex, 0);
+		ip6_redirect(skb, net, skb->dev->ifindex, 0,
+			     sock_net_uid(net, NULL));
 	else
-		ip6_update_pmtu(skb, net, info, 0, 0);
+		ip6_update_pmtu(skb, net, info, 0, 0, sock_net_uid(net, NULL));
 	xfrm_state_put(x);
 
 	return 0;
diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c
index cae37bfd12ab..56528e9f3e01 100644
--- a/net/ipv6/datagram.c
+++ b/net/ipv6/datagram.c
@@ -165,6 +165,7 @@ ipv4_connected:
 	fl6.flowi6_mark = sk->sk_mark;
 	fl6.fl6_dport = inet->inet_dport;
 	fl6.fl6_sport = inet->inet_sport;
+	fl6.flowi6_uid = sk->sk_uid;
 
 	if (!fl6.flowi6_oif)
 		fl6.flowi6_oif = np->sticky_pktinfo.ipi6_ifindex;
diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c
index 111ba55fd512..cbcdd5db31f4 100644
--- a/net/ipv6/esp6.c
+++ b/net/ipv6/esp6.c
@@ -474,9 +474,10 @@ static int esp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		return 0;
 
 	if (type == NDISC_REDIRECT)
-		ip6_redirect(skb, net, skb->dev->ifindex, 0);
+		ip6_redirect(skb, net, skb->dev->ifindex, 0,
+			     sock_net_uid(net, NULL));
 	else
-		ip6_update_pmtu(skb, net, info, 0, 0);
+		ip6_update_pmtu(skb, net, info, 0, 0, sock_net_uid(net, NULL));
 	xfrm_state_put(x);
 
 	return 0;
diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c
index 9508a20fbf61..840a4388f860 100644
--- a/net/ipv6/exthdrs_core.c
+++ b/net/ipv6/exthdrs_core.c
@@ -166,15 +166,15 @@ EXPORT_SYMBOL_GPL(ipv6_find_tlv);
  * to explore inner IPv6 header, eg. ICMPv6 error messages.
  *
  * If target header is found, its offset is set in *offset and return protocol
- * number. Otherwise, return -1.
+ * number. Otherwise, return -ENOENT or -EBADMSG.
  *
  * If the first fragment doesn't contain the final protocol header or
  * NEXTHDR_NONE it is considered invalid.
  *
  * Note that non-1st fragment is special case that "the protocol number
  * of last header" is "next header" field in Fragment header. In this case,
- * *offset is meaningless and fragment offset is stored in *fragoff if fragoff
- * isn't NULL.
+ * *offset is meaningless. If fragoff is not NULL, the fragment offset is
+ * stored in *fragoff; if it is NULL, return -EINVAL.
  *
  * if flags is not NULL and it's a fragment, then the frag flag
  * IP6_FH_F_FRAG will be set. If it's an AH header, the
@@ -253,9 +253,12 @@ int ipv6_find_hdr(const struct sk_buff *skb, unsigned int *offset,
 				if (target < 0 &&
 				    ((!ipv6_ext_hdr(hp->nexthdr)) ||
 				     hp->nexthdr == NEXTHDR_NONE)) {
-					if (fragoff)
+					if (fragoff) {
 						*fragoff = _frag_off;
-					return hp->nexthdr;
+						return hp->nexthdr;
+					} else {
+						return -EINVAL;
+					}
 				}
 				if (!found)
 					return -ENOENT;
diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c
index 0a37ddc7af51..3ae2fbe07b25 100644
--- a/net/ipv6/icmp.c
+++ b/net/ipv6/icmp.c
@@ -92,13 +92,14 @@ static void icmpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	struct net *net = dev_net(skb->dev);
 
 	if (type == ICMPV6_PKT_TOOBIG)
-		ip6_update_pmtu(skb, net, info, 0, 0);
+		ip6_update_pmtu(skb, net, info, 0, 0, sock_net_uid(net, NULL));
 	else if (type == NDISC_REDIRECT)
-		ip6_redirect(skb, net, skb->dev->ifindex, 0);
+		ip6_redirect(skb, net, skb->dev->ifindex, 0,
+			     sock_net_uid(net, NULL));
 
 	if (!(type & ICMPV6_INFOMSG_MASK))
 		if (icmp6->icmp6_type == ICMPV6_ECHO_REQUEST)
-			ping_err(skb, offset, info);
+			ping_err(skb, offset, ntohl(info));
 }
 
 static int icmpv6_rcv(struct sk_buff *skb);
@@ -478,6 +479,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info)
 	fl6.flowi6_oif = iif;
 	fl6.fl6_icmp_type = type;
 	fl6.fl6_icmp_code = code;
+	fl6.flowi6_uid = sock_net_uid(net, NULL);
 	security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
 
 	sk = icmpv6_xmit_lock(net);
@@ -585,6 +587,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb)
 	fl6.flowi6_oif = l3mdev_fib_oif(skb->dev);
 	fl6.fl6_icmp_type = ICMPV6_ECHO_REPLY;
 	fl6.flowi6_mark = mark;
+	fl6.flowi6_uid = sock_net_uid(net, NULL);
 	security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
 
 	sk = icmpv6_xmit_lock(net);
diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c
index a7ca2cde2ecb..dc79ebc14189 100644
--- a/net/ipv6/inet6_connection_sock.c
+++ b/net/ipv6/inet6_connection_sock.c
@@ -86,6 +86,7 @@ struct dst_entry *inet6_csk_route_req(const struct sock *sk,
 	fl6->flowi6_mark = ireq->ir_mark;
 	fl6->fl6_dport = ireq->ir_rmt_port;
 	fl6->fl6_sport = htons(ireq->ir_num);
+	fl6->flowi6_uid = sk->sk_uid;
 	security_req_classify_flow(req, flowi6_to_flowi(fl6));
 
 	dst = ip6_dst_lookup_flow(sk, fl6, final_p);
@@ -134,6 +135,7 @@ static struct dst_entry *inet6_csk_route_socket(struct sock *sk,
 	fl6->flowi6_mark = sk->sk_mark;
 	fl6->fl6_sport = inet->inet_sport;
 	fl6->fl6_dport = inet->inet_dport;
+	fl6->flowi6_uid = sk->sk_uid;
 	security_sk_classify_flow(sk, flowi6_to_flowi(fl6));
 
 	rcu_read_lock();
diff --git a/net/ipv6/ip6_fib.c b/net/ipv6/ip6_fib.c
index c23e02a7ccb0..bf3824b59597 100644
--- a/net/ipv6/ip6_fib.c
+++ b/net/ipv6/ip6_fib.c
@@ -909,6 +909,7 @@ add:
 			fn->fn_flags |= RTN_RTINFO;
 		}
 		nsiblings = iter->rt6i_nsiblings;
+		iter->rt6i_node = NULL;
 		fib6_purge_rt(iter, fn, info->nl_net);
 		if (fn->rr_ptr == iter)
 			fn->rr_ptr = NULL;
@@ -923,6 +924,7 @@ add:
 					break;
 				if (rt6_qualify_for_ecmp(iter)) {
 					*ins = iter->dst.rt6_next;
+					iter->rt6i_node = NULL;
 					fib6_purge_rt(iter, fn, info->nl_net);
 					if (fn->rr_ptr == iter)
 						fn->rr_ptr = NULL;
diff --git a/net/ipv6/ip6_gre.c b/net/ipv6/ip6_gre.c
index c878cbf65485..9d1a54de33f2 100644
--- a/net/ipv6/ip6_gre.c
+++ b/net/ipv6/ip6_gre.c
@@ -800,6 +800,8 @@ static inline int ip6gre_xmit_ipv4(struct sk_buff *skb, struct net_device *dev)
 	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FWMARK)
 		fl6.flowi6_mark = skb->mark;
 
+	fl6.flowi6_uid = sock_net_uid(dev_net(dev), NULL);
+
 	err = ip6gre_xmit2(skb, dev, dsfield, &fl6, encap_limit, &mtu);
 	if (err != 0) {
 		/* XXX: send ICMP error even if DF is not set. */
@@ -850,6 +852,8 @@ static inline int ip6gre_xmit_ipv6(struct sk_buff *skb, struct net_device *dev)
 	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FWMARK)
 		fl6.flowi6_mark = skb->mark;
 
+	fl6.flowi6_uid = sock_net_uid(dev_net(dev), NULL);
+
 	err = ip6gre_xmit2(skb, dev, dsfield, &fl6, encap_limit, &mtu);
 	if (err != 0) {
 		if (err == -EMSGSIZE)
diff --git a/net/ipv6/ip6_tunnel.c b/net/ipv6/ip6_tunnel.c
index 97cb02dc5f02..600975c5eacf 100644
--- a/net/ipv6/ip6_tunnel.c
+++ b/net/ipv6/ip6_tunnel.c
@@ -1215,6 +1215,8 @@ ip4ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev)
 	memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6));
 	fl6.flowi6_proto = IPPROTO_IPIP;
 
+	fl6.flowi6_uid = sock_net_uid(dev_net(dev), NULL);
+
 	dsfield = ipv4_get_dsfield(iph);
 
 	if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)
@@ -1268,6 +1270,7 @@ ip6ip6_tnl_xmit(struct sk_buff *skb, struct net_device *dev)
 
 	memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6));
 	fl6.flowi6_proto = IPPROTO_IPV6;
+	fl6.flowi6_uid = sock_net_uid(dev_net(dev), NULL);
 
 	dsfield = ipv6_get_dsfield(ipv6h);
 	if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)
diff --git a/net/ipv6/ip6_vti.c b/net/ipv6/ip6_vti.c
index 7ebb14def2cb..f615f982961a 100644
--- a/net/ipv6/ip6_vti.c
+++ b/net/ipv6/ip6_vti.c
@@ -598,9 +598,10 @@ static int vti6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		return 0;
 
 	if (type == NDISC_REDIRECT)
-		ip6_redirect(skb, net, skb->dev->ifindex, 0);
+		ip6_redirect(skb, net, skb->dev->ifindex, 0,
+			     sock_net_uid(net, NULL));
 	else
-		ip6_update_pmtu(skb, net, info, 0, 0);
+		ip6_update_pmtu(skb, net, info, 0, 0, sock_net_uid(net, NULL));
 	xfrm_state_put(x);
 
 	return 0;
diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c
index 1b9316e1386a..54d165b9845a 100644
--- a/net/ipv6/ipcomp6.c
+++ b/net/ipv6/ipcomp6.c
@@ -74,9 +74,10 @@ static int ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 		return 0;
 
 	if (type == NDISC_REDIRECT)
-		ip6_redirect(skb, net, skb->dev->ifindex, 0);
+		ip6_redirect(skb, net, skb->dev->ifindex, 0,
+			     sock_net_uid(net, NULL));
 	else
-		ip6_update_pmtu(skb, net, info, 0, 0);
+		ip6_update_pmtu(skb, net, info, 0, 0, sock_net_uid(net, NULL));
 	xfrm_state_put(x);
 
 	return 0;
diff --git a/net/ipv6/ndisc.c b/net/ipv6/ndisc.c
index 84afb9a77278..3452f9037ad4 100644
--- a/net/ipv6/ndisc.c
+++ b/net/ipv6/ndisc.c
@@ -1358,6 +1358,8 @@ skip_linkparms:
 			if (ri->prefix_len == 0 &&
 			    !in6_dev->cnf.accept_ra_defrtr)
 				continue;
+			if (ri->prefix_len < in6_dev->cnf.accept_ra_rt_info_min_plen)
+				continue;
 			if (ri->prefix_len > in6_dev->cnf.accept_ra_rt_info_max_plen)
 				continue;
 			rt6_route_rcv(skb->dev, (u8 *)p, (p->nd_opt_len) << 3,
diff --git a/net/ipv6/netfilter.c b/net/ipv6/netfilter.c
index d11c46833d61..39970e212ad5 100644
--- a/net/ipv6/netfilter.c
+++ b/net/ipv6/netfilter.c
@@ -26,6 +26,7 @@ int ip6_route_me_harder(struct net *net, struct sk_buff *skb)
 	struct flowi6 fl6 = {
 		.flowi6_oif = skb->sk ? skb->sk->sk_bound_dev_if : 0,
 		.flowi6_mark = skb->mark,
+		.flowi6_uid = sock_net_uid(net, skb->sk),
 		.daddr = iph->daddr,
 		.saddr = iph->saddr,
 	};
diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c
index a830b68e63c9..4c753f4a3712 100644
--- a/net/ipv6/ping.c
+++ b/net/ipv6/ping.c
@@ -84,7 +84,7 @@ int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	struct icmp6hdr user_icmph;
 	int addr_type;
 	struct in6_addr *daddr;
-	int iif = 0;
+	int oif = 0;
 	struct flowi6 fl6;
 	int err;
 	int hlimit;
@@ -106,25 +106,30 @@ int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		if (u->sin6_family != AF_INET6) {
 			return -EAFNOSUPPORT;
 		}
-		if (sk->sk_bound_dev_if &&
-		    sk->sk_bound_dev_if != u->sin6_scope_id) {
-			return -EINVAL;
-		}
 		daddr = &(u->sin6_addr);
-		iif = u->sin6_scope_id;
+		if (__ipv6_addr_needs_scope_id(ipv6_addr_type(daddr)))
+			oif = u->sin6_scope_id;
 	} else {
 		if (sk->sk_state != TCP_ESTABLISHED)
 			return -EDESTADDRREQ;
 		daddr = &sk->sk_v6_daddr;
 	}
 
-	if (!iif)
-		iif = sk->sk_bound_dev_if;
+	if (!oif)
+		oif = sk->sk_bound_dev_if;
+
+	if (!oif)
+		oif = np->sticky_pktinfo.ipi6_ifindex;
+
+	if (!oif && ipv6_addr_is_multicast(daddr))
+		oif = np->mcast_oif;
+	else if (!oif)
+		oif = np->ucast_oif;
 
 	addr_type = ipv6_addr_type(daddr);
-	if (__ipv6_addr_needs_scope_id(addr_type) && !iif)
-		return -EINVAL;
-	if (addr_type & IPV6_ADDR_MAPPED)
+	if ((__ipv6_addr_needs_scope_id(addr_type) && !oif) ||
+	    (addr_type & IPV6_ADDR_MAPPED) ||
+	    (oif && sk->sk_bound_dev_if && oif != sk->sk_bound_dev_if))
 		return -EINVAL;
 
 	/* TODO: use ip6_datagram_send_ctl to get options from cmsg */
@@ -134,16 +139,13 @@ int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	fl6.flowi6_proto = IPPROTO_ICMPV6;
 	fl6.saddr = np->saddr;
 	fl6.daddr = *daddr;
+	fl6.flowi6_oif = oif;
 	fl6.flowi6_mark = sk->sk_mark;
+	fl6.flowi6_uid = sk->sk_uid;
 	fl6.fl6_icmp_type = user_icmph.icmp6_type;
 	fl6.fl6_icmp_code = user_icmph.icmp6_code;
 	security_sk_classify_flow(sk, flowi6_to_flowi(&fl6));
 
-	if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr))
-		fl6.flowi6_oif = np->mcast_oif;
-	else if (!fl6.flowi6_oif)
-		fl6.flowi6_oif = np->ucast_oif;
-
 	dst = ip6_sk_dst_lookup_flow(sk, &fl6,  daddr);
 	if (IS_ERR(dst))
 		return PTR_ERR(dst);
@@ -155,11 +157,6 @@ int ping_v6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 		goto dst_err_out;
 	}
 
-	if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr))
-		fl6.flowi6_oif = np->mcast_oif;
-	else if (!fl6.flowi6_oif)
-		fl6.flowi6_oif = np->ucast_oif;
-
 	pfh.icmph.type = user_icmph.icmp6_type;
 	pfh.icmph.code = user_icmph.icmp6_code;
 	pfh.icmph.checksum = 0;
diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c
index 4d52a0e2f60d..0ef8e114c8ab 100644
--- a/net/ipv6/raw.c
+++ b/net/ipv6/raw.c
@@ -774,6 +774,7 @@ static int rawv6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	memset(&fl6, 0, sizeof(fl6));
 
 	fl6.flowi6_mark = sk->sk_mark;
+	fl6.flowi6_uid = sk->sk_uid;
 
 	if (sin6) {
 		if (addr_len < SIN6_LEN_RFC2133)
diff --git a/net/ipv6/route.c b/net/ipv6/route.c
index 48917437550e..0dd92409200d 100644
--- a/net/ipv6/route.c
+++ b/net/ipv6/route.c
@@ -99,13 +99,12 @@ static void		rt6_dst_from_metrics_check(struct rt6_info *rt);
 static int rt6_score_route(struct rt6_info *rt, int oif, int strict);
 
 #ifdef CONFIG_IPV6_ROUTE_INFO
-static struct rt6_info *rt6_add_route_info(struct net *net,
+static struct rt6_info *rt6_add_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex,
-					   unsigned int pref);
-static struct rt6_info *rt6_get_route_info(struct net *net,
+					   const struct in6_addr *gwaddr, unsigned int pref);
+static struct rt6_info *rt6_get_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex);
+					   const struct in6_addr *gwaddr);
 #endif
 
 struct uncached_list {
@@ -755,7 +754,6 @@ static bool rt6_is_gw_or_nonexthop(const struct rt6_info *rt)
 int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 		  const struct in6_addr *gwaddr)
 {
-	struct net *net = dev_net(dev);
 	struct route_info *rinfo = (struct route_info *) opt;
 	struct in6_addr prefix_buf, *prefix;
 	unsigned int pref;
@@ -800,8 +798,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 	if (rinfo->prefix_len == 0)
 		rt = rt6_get_dflt_router(gwaddr, dev);
 	else
-		rt = rt6_get_route_info(net, prefix, rinfo->prefix_len,
-					gwaddr, dev->ifindex);
+		rt = rt6_get_route_info(dev, prefix, rinfo->prefix_len,	gwaddr);
 
 	if (rt && !lifetime) {
 		ip6_del_rt(rt);
@@ -809,8 +806,7 @@ int rt6_route_rcv(struct net_device *dev, u8 *opt, int len,
 	}
 
 	if (!rt && lifetime)
-		rt = rt6_add_route_info(net, prefix, rinfo->prefix_len, gwaddr, dev->ifindex,
-					pref);
+		rt = rt6_add_route_info(dev, prefix, rinfo->prefix_len, gwaddr, pref);
 	else if (rt)
 		rt->rt6i_flags = RTF_ROUTEINFO |
 				 (rt->rt6i_flags & ~RTF_PREF_MASK) | RTF_PREF(pref);
@@ -1395,7 +1391,7 @@ static void ip6_rt_update_pmtu(struct dst_entry *dst, struct sock *sk,
 }
 
 void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu,
-		     int oif, u32 mark)
+		     int oif, u32 mark, kuid_t uid)
 {
 	const struct ipv6hdr *iph = (struct ipv6hdr *) skb->data;
 	struct dst_entry *dst;
@@ -1407,6 +1403,7 @@ void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu,
 	fl6.daddr = iph->daddr;
 	fl6.saddr = iph->saddr;
 	fl6.flowlabel = ip6_flowinfo(iph);
+	fl6.flowi6_uid = uid;
 
 	dst = ip6_route_output(net, NULL, &fl6);
 	if (!dst->error)
@@ -1418,7 +1415,7 @@ EXPORT_SYMBOL_GPL(ip6_update_pmtu);
 void ip6_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, __be32 mtu)
 {
 	ip6_update_pmtu(skb, sock_net(sk), mtu,
-			sk->sk_bound_dev_if, sk->sk_mark);
+			sk->sk_bound_dev_if, sk->sk_mark, sk->sk_uid);
 }
 EXPORT_SYMBOL_GPL(ip6_sk_update_pmtu);
 
@@ -1499,7 +1496,8 @@ static struct dst_entry *ip6_route_redirect(struct net *net,
 				flags, __ip6_route_redirect);
 }
 
-void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark)
+void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark,
+		  kuid_t uid)
 {
 	const struct ipv6hdr *iph = (struct ipv6hdr *) skb->data;
 	struct dst_entry *dst;
@@ -1512,6 +1510,7 @@ void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark)
 	fl6.daddr = iph->daddr;
 	fl6.saddr = iph->saddr;
 	fl6.flowlabel = ip6_flowinfo(iph);
+	fl6.flowi6_uid = uid;
 
 	dst = ip6_route_redirect(net, &fl6, &ipv6_hdr(skb)->saddr);
 	rt6_do_redirect(dst, NULL, skb);
@@ -1533,6 +1532,7 @@ void ip6_redirect_no_header(struct sk_buff *skb, struct net *net, int oif,
 	fl6.flowi6_mark = mark;
 	fl6.daddr = msg->dest;
 	fl6.saddr = iph->daddr;
+	fl6.flowi6_uid = sock_net_uid(net, NULL);
 
 	dst = ip6_route_redirect(net, &fl6, &iph->saddr);
 	rt6_do_redirect(dst, NULL, skb);
@@ -1541,7 +1541,8 @@ void ip6_redirect_no_header(struct sk_buff *skb, struct net *net, int oif,
 
 void ip6_sk_redirect(struct sk_buff *skb, struct sock *sk)
 {
-	ip6_redirect(skb, sock_net(sk), sk->sk_bound_dev_if, sk->sk_mark);
+	ip6_redirect(skb, sock_net(sk), sk->sk_bound_dev_if, sk->sk_mark,
+		     sk->sk_uid);
 }
 EXPORT_SYMBOL_GPL(ip6_sk_redirect);
 
@@ -2263,15 +2264,16 @@ static void ip6_rt_copy_init(struct rt6_info *rt, struct rt6_info *ort)
 }
 
 #ifdef CONFIG_IPV6_ROUTE_INFO
-static struct rt6_info *rt6_get_route_info(struct net *net,
+static struct rt6_info *rt6_get_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex)
+					   const struct in6_addr *gwaddr)
 {
 	struct fib6_node *fn;
 	struct rt6_info *rt = NULL;
 	struct fib6_table *table;
 
-	table = fib6_get_table(net, RT6_TABLE_INFO);
+	table = fib6_get_table(dev_net(dev),
+			       addrconf_rt_table(dev, RT6_TABLE_INFO));
 	if (!table)
 		return NULL;
 
@@ -2281,7 +2283,7 @@ static struct rt6_info *rt6_get_route_info(struct net *net,
 		goto out;
 
 	for (rt = fn->leaf; rt; rt = rt->dst.rt6_next) {
-		if (rt->dst.dev->ifindex != ifindex)
+		if (rt->dst.dev->ifindex != dev->ifindex)
 			continue;
 		if ((rt->rt6i_flags & (RTF_ROUTEINFO|RTF_GATEWAY)) != (RTF_ROUTEINFO|RTF_GATEWAY))
 			continue;
@@ -2295,23 +2297,22 @@ out:
 	return rt;
 }
 
-static struct rt6_info *rt6_add_route_info(struct net *net,
+static struct rt6_info *rt6_add_route_info(struct net_device *dev,
 					   const struct in6_addr *prefix, int prefixlen,
-					   const struct in6_addr *gwaddr, int ifindex,
-					   unsigned int pref)
+					   const struct in6_addr *gwaddr, unsigned int pref)
 {
 	struct fib6_config cfg = {
 		.fc_metric	= IP6_RT_PRIO_USER,
-		.fc_ifindex	= ifindex,
+		.fc_ifindex	= dev->ifindex,
 		.fc_dst_len	= prefixlen,
 		.fc_flags	= RTF_GATEWAY | RTF_ADDRCONF | RTF_ROUTEINFO |
 				  RTF_UP | RTF_PREF(pref),
 		.fc_nlinfo.portid = 0,
 		.fc_nlinfo.nlh = NULL,
-		.fc_nlinfo.nl_net = net,
+		.fc_nlinfo.nl_net = dev_net(dev),
 	};
 
-	cfg.fc_table = l3mdev_fib_table_by_index(net, ifindex) ? : RT6_TABLE_INFO;
+	cfg.fc_table = l3mdev_fib_table_by_index(dev_net(dev), dev->ifindex) ? : addrconf_rt_table(dev, RT6_TABLE_INFO);
 	cfg.fc_dst = *prefix;
 	cfg.fc_gateway = *gwaddr;
 
@@ -2321,7 +2322,7 @@ static struct rt6_info *rt6_add_route_info(struct net *net,
 
 	ip6_route_add(&cfg);
 
-	return rt6_get_route_info(net, prefix, prefixlen, gwaddr, ifindex);
+	return rt6_get_route_info(dev, prefix, prefixlen, gwaddr);
 }
 #endif
 
@@ -2330,7 +2331,8 @@ struct rt6_info *rt6_get_dflt_router(const struct in6_addr *addr, struct net_dev
 	struct rt6_info *rt;
 	struct fib6_table *table;
 
-	table = fib6_get_table(dev_net(dev), RT6_TABLE_DFLT);
+	table = fib6_get_table(dev_net(dev),
+			       addrconf_rt_table(dev, RT6_TABLE_MAIN));
 	if (!table)
 		return NULL;
 
@@ -2352,7 +2354,7 @@ struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr,
 				     unsigned int pref)
 {
 	struct fib6_config cfg = {
-		.fc_table	= l3mdev_fib_table(dev) ? : RT6_TABLE_DFLT,
+		.fc_table	= l3mdev_fib_table(dev) ? : addrconf_rt_table(dev, RT6_TABLE_DFLT),
 		.fc_metric	= IP6_RT_PRIO_USER,
 		.fc_ifindex	= dev->ifindex,
 		.fc_flags	= RTF_GATEWAY | RTF_ADDRCONF | RTF_DEFAULT |
@@ -2369,28 +2371,17 @@ struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr,
 	return rt6_get_dflt_router(gwaddr, dev);
 }
 
-void rt6_purge_dflt_routers(struct net *net)
-{
-	struct rt6_info *rt;
-	struct fib6_table *table;
 
-	/* NOTE: Keep consistent with rt6_get_dflt_router */
-	table = fib6_get_table(net, RT6_TABLE_DFLT);
-	if (!table)
-		return;
+int rt6_addrconf_purge(struct rt6_info *rt, void *arg) {
+	if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF) &&
+	    (!rt->rt6i_idev || rt->rt6i_idev->cnf.accept_ra != 2))
+		return -1;
+	return 0;
+}
 
-restart:
-	read_lock_bh(&table->tb6_lock);
-	for (rt = table->tb6_root.leaf; rt; rt = rt->dst.rt6_next) {
-		if (rt->rt6i_flags & (RTF_DEFAULT | RTF_ADDRCONF) &&
-		    (!rt->rt6i_idev || rt->rt6i_idev->cnf.accept_ra != 2)) {
-			dst_hold(&rt->dst);
-			read_unlock_bh(&table->tb6_lock);
-			ip6_del_rt(rt);
-			goto restart;
-		}
-	}
-	read_unlock_bh(&table->tb6_lock);
+void rt6_purge_dflt_routers(struct net *net)
+{
+	fib6_clean_all(net, rt6_addrconf_purge, NULL);
 }
 
 static void rtmsg_to_fib6_config(struct net *net,
@@ -2715,6 +2706,7 @@ static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = {
 	[RTA_PREF]              = { .type = NLA_U8 },
 	[RTA_ENCAP_TYPE]	= { .type = NLA_U16 },
 	[RTA_ENCAP]		= { .type = NLA_NESTED },
+	[RTA_UID]		= { .type = NLA_U32 },
 };
 
 static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh,
@@ -3273,6 +3265,12 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh)
 	if (tb[RTA_MARK])
 		fl6.flowi6_mark = nla_get_u32(tb[RTA_MARK]);
 
+	if (tb[RTA_UID])
+		fl6.flowi6_uid = make_kuid(current_user_ns(),
+					   nla_get_u32(tb[RTA_UID]));
+	else
+		fl6.flowi6_uid = iif ? INVALID_UID : current_uid();
+
 	if (iif) {
 		struct net_device *dev;
 		int flags = 0;
diff --git a/net/ipv6/syncookies.c b/net/ipv6/syncookies.c
index aee87282d352..7f3667635431 100644
--- a/net/ipv6/syncookies.c
+++ b/net/ipv6/syncookies.c
@@ -229,6 +229,7 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb)
 		fl6.flowi6_mark = ireq->ir_mark;
 		fl6.fl6_dport = ireq->ir_rmt_port;
 		fl6.fl6_sport = inet_sk(sk)->inet_sport;
+		fl6.flowi6_uid = sk->sk_uid;
 		security_req_classify_flow(req, flowi6_to_flowi(&fl6));
 
 		dst = ip6_dst_lookup_flow(sk, &fl6, final_p);
diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c
index 59c908ff251a..4618f52a4abe 100644
--- a/net/ipv6/tcp_ipv6.c
+++ b/net/ipv6/tcp_ipv6.c
@@ -239,6 +239,7 @@ static int tcp_v6_connect(struct sock *sk, struct sockaddr *uaddr,
 	fl6.flowi6_mark = sk->sk_mark;
 	fl6.fl6_dport = usin->sin6_port;
 	fl6.fl6_sport = inet->inet_sport;
+	fl6.flowi6_uid = sk->sk_uid;
 
 	opt = rcu_dereference_protected(np->opt, sock_owned_by_user(sk));
 	final_p = fl6_update_dst(&fl6, opt, &final);
@@ -820,6 +821,7 @@ static void tcp_v6_send_response(const struct sock *sk, struct sk_buff *skb, u32
 	fl6.flowi6_mark = IP6_REPLY_MARK(net, skb->mark);
 	fl6.fl6_dport = t1->dest;
 	fl6.fl6_sport = t1->source;
+	fl6.flowi6_uid = sock_net_uid(net, sk && sk_fullsock(sk) ? sk : NULL);
 	security_skb_classify_flow(skb, flowi6_to_flowi(&fl6));
 
 	/* Pass a socket to ip6_dst_lookup either it is for RST
@@ -1915,6 +1917,7 @@ struct proto tcpv6_prot = {
 	.proto_cgroup		= tcp_proto_cgroup,
 #endif
 	.clear_sk		= tcp_v6_clear_sk,
+	.diag_destroy		= tcp_abort,
 };
 
 static const struct inet6_protocol tcpv6_protocol = {
diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c
index 6eb1e9293b6f..3b0e22d20e83 100644
--- a/net/ipv6/udp.c
+++ b/net/ipv6/udp.c
@@ -1249,6 +1249,7 @@ do_udp_sendmsg:
 		fl6.flowi6_oif = np->sticky_pktinfo.ipi6_ifindex;
 
 	fl6.flowi6_mark = sk->sk_mark;
+	fl6.flowi6_uid = sk->sk_uid;
 
 	if (msg->msg_controllen) {
 		opt = &opt_space;
@@ -1557,6 +1558,7 @@ struct proto udpv6_prot = {
 	.compat_getsockopt = compat_udpv6_getsockopt,
 #endif
 	.clear_sk	   = udp_v6_clear_sk,
+	.diag_destroy      = udp_abort,
 };
 
 static struct inet_protosw udpv6_protosw = {
diff --git a/net/ipv6/xfrm6_policy.c b/net/ipv6/xfrm6_policy.c
index c074771a10f7..1a8608cc104c 100644
--- a/net/ipv6/xfrm6_policy.c
+++ b/net/ipv6/xfrm6_policy.c
@@ -29,15 +29,17 @@ static struct xfrm_policy_afinfo xfrm6_policy_afinfo;
 
 static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif,
 					  const xfrm_address_t *saddr,
-					  const xfrm_address_t *daddr)
+					  const xfrm_address_t *daddr,
+					  u32 mark)
 {
 	struct flowi6 fl6;
 	struct dst_entry *dst;
 	int err;
 
 	memset(&fl6, 0, sizeof(fl6));
-	fl6.flowi6_oif = oif;
+	fl6.flowi6_oif = l3mdev_master_ifindex_by_index(net, oif);
 	fl6.flowi6_flags = FLOWI_FLAG_SKIP_NH_OIF;
+	fl6.flowi6_mark = mark;
 	memcpy(&fl6.daddr, daddr, sizeof(fl6.daddr));
 	if (saddr)
 		memcpy(&fl6.saddr, saddr, sizeof(fl6.saddr));
@@ -54,12 +56,13 @@ static struct dst_entry *xfrm6_dst_lookup(struct net *net, int tos, int oif,
 }
 
 static int xfrm6_get_saddr(struct net *net, int oif,
-			   xfrm_address_t *saddr, xfrm_address_t *daddr)
+			   xfrm_address_t *saddr, xfrm_address_t *daddr,
+			   u32 mark)
 {
 	struct dst_entry *dst;
 	struct net_device *dev;
 
-	dst = xfrm6_dst_lookup(net, 0, oif, NULL, daddr);
+	dst = xfrm6_dst_lookup(net, 0, oif, NULL, daddr, mark);
 	if (IS_ERR(dst))
 		return -EHOSTUNREACH;
 
diff --git a/net/l2tp/l2tp_ip6.c b/net/l2tp/l2tp_ip6.c
index bcdab1cba773..c8f483cd2ca9 100644
--- a/net/l2tp/l2tp_ip6.c
+++ b/net/l2tp/l2tp_ip6.c
@@ -518,6 +518,7 @@ static int l2tp_ip6_sendmsg(struct sock *sk, struct msghdr *msg, size_t len)
 	memset(&fl6, 0, sizeof(fl6));
 
 	fl6.flowi6_mark = sk->sk_mark;
+	fl6.flowi6_uid = sk->sk_uid;
 
 	if (lsa) {
 		if (addr_len < SIN6_LEN_RFC2133)
diff --git a/net/mac80211/mesh_hwmp.c b/net/mac80211/mesh_hwmp.c
index c6be0b4f4058..b6dc2d7cd650 100644
--- a/net/mac80211/mesh_hwmp.c
+++ b/net/mac80211/mesh_hwmp.c
@@ -530,7 +530,7 @@ static void hwmp_preq_frame_process(struct ieee80211_sub_if_data *sdata,
 	const u8 *target_addr, *orig_addr;
 	const u8 *da;
 	u8 target_flags, ttl, flags;
-	u32 orig_sn, target_sn, lifetime, target_metric;
+	u32 orig_sn, target_sn, lifetime, target_metric = 0;
 	bool reply = false;
 	bool forward = true;
 	bool root_is_gate;
diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig
index 4692782b5280..1959548b1161 100644
--- a/net/netfilter/Kconfig
+++ b/net/netfilter/Kconfig
@@ -1278,6 +1278,8 @@ config NETFILTER_XT_MATCH_OWNER
 	based on who created the socket: the user or group. It is also
 	possible to check whether a socket actually exists.
 
+	Conflicts with '"quota, tag, uid" match'
+
 config NETFILTER_XT_MATCH_POLICY
 	tristate 'IPsec "policy" match support'
 	depends on XFRM
@@ -1311,6 +1313,22 @@ config NETFILTER_XT_MATCH_PKTTYPE
 
 	  To compile it as a module, choose M here.  If unsure, say N.
 
+config NETFILTER_XT_MATCH_QTAGUID
+	bool '"quota, tag, owner" match and stats support'
+        depends on NETFILTER_XT_MATCH_SOCKET
+	depends on NETFILTER_XT_MATCH_OWNER=n
+	help
+	  This option replaces the `owner' match. In addition to matching
+	  on uid, it keeps stats based on a tag assigned to a socket.
+	  The full tag is comprised of a UID and an accounting tag.
+	  The tags are assignable to sockets from user space (e.g. a download
+	  manager can assign the socket to another UID for accounting).
+	  Stats and control are done via /proc/net/xt_qtaguid/.
+	  It replaces owner as it takes the same arguments, but should
+	  really be recognized by the iptables tool.
+
+	  If unsure, say `N'.
+
 config NETFILTER_XT_MATCH_QUOTA
 	tristate '"quota" match support'
 	depends on NETFILTER_ADVANCED
@@ -1321,6 +1339,29 @@ config NETFILTER_XT_MATCH_QUOTA
 	  If you want to compile it as a module, say M here and read
 	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
 
+config NETFILTER_XT_MATCH_QUOTA2
+	tristate '"quota2" match support'
+	depends on NETFILTER_ADVANCED
+	help
+	  This option adds a `quota2' match, which allows to match on a
+	  byte counter correctly and not per CPU.
+	  It allows naming the quotas.
+	  This is based on http://xtables-addons.git.sourceforge.net
+
+	  If you want to compile it as a module, say M here and read
+	  <file:Documentation/kbuild/modules.txt>.  If unsure, say `N'.
+
+config NETFILTER_XT_MATCH_QUOTA2_LOG
+	bool '"quota2" Netfilter LOG support'
+	depends on NETFILTER_XT_MATCH_QUOTA2
+	default n
+	help
+	  This option allows `quota2' to log ONCE when a quota limit
+	  is passed. It logs via NETLINK using the NETLINK_NFLOG family.
+	  It logs similarly to how ipt_ULOG would without data.
+
+	  If unsure, say `N'.
+
 config NETFILTER_XT_MATCH_RATEEST
 	tristate '"rateest" match support'
 	depends on NETFILTER_ADVANCED
diff --git a/net/netfilter/Makefile b/net/netfilter/Makefile
index 7638c36b498c..ad6a8aa63b1f 100644
--- a/net/netfilter/Makefile
+++ b/net/netfilter/Makefile
@@ -156,7 +156,9 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_CGROUP) += xt_cgroup.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_PHYSDEV) += xt_physdev.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_PKTTYPE) += xt_pkttype.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_POLICY) += xt_policy.o
+obj-$(CONFIG_NETFILTER_XT_MATCH_QTAGUID) += xt_qtaguid_print.o xt_qtaguid.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_QUOTA) += xt_quota.o
+obj-$(CONFIG_NETFILTER_XT_MATCH_QUOTA2) += xt_quota2.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_RATEEST) += xt_rateest.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_REALM) += xt_realm.o
 obj-$(CONFIG_NETFILTER_XT_MATCH_RECENT) += xt_recent.o
diff --git a/net/netfilter/xt_IDLETIMER.c b/net/netfilter/xt_IDLETIMER.c
index 29d2c31f406c..f11aa28b96ce 100644
--- a/net/netfilter/xt_IDLETIMER.c
+++ b/net/netfilter/xt_IDLETIMER.c
@@ -5,6 +5,7 @@
  * After timer expires a kevent will be sent.
  *
  * Copyright (C) 2004, 2010 Nokia Corporation
+ *
  * Written by Timo Teras <ext-timo.teras@nokia.com>
  *
  * Converted to x_tables and reworked for upstream inclusion
@@ -38,8 +39,17 @@
 #include <linux/netfilter/xt_IDLETIMER.h>
 #include <linux/kdev_t.h>
 #include <linux/kobject.h>
+#include <linux/skbuff.h>
 #include <linux/workqueue.h>
 #include <linux/sysfs.h>
+#include <linux/rtc.h>
+#include <linux/time.h>
+#include <linux/math64.h>
+#include <linux/suspend.h>
+#include <linux/notifier.h>
+#include <net/net_namespace.h>
+#include <net/sock.h>
+#include <net/inet_sock.h>
 
 struct idletimer_tg_attr {
 	struct attribute attr;
@@ -55,14 +65,110 @@ struct idletimer_tg {
 	struct kobject *kobj;
 	struct idletimer_tg_attr attr;
 
+	struct timespec delayed_timer_trigger;
+	struct timespec last_modified_timer;
+	struct timespec last_suspend_time;
+	struct notifier_block pm_nb;
+
+	int timeout;
 	unsigned int refcnt;
+	bool work_pending;
+	bool send_nl_msg;
+	bool active;
+	uid_t uid;
 };
 
 static LIST_HEAD(idletimer_tg_list);
 static DEFINE_MUTEX(list_mutex);
+static DEFINE_SPINLOCK(timestamp_lock);
 
 static struct kobject *idletimer_tg_kobj;
 
+static bool check_for_delayed_trigger(struct idletimer_tg *timer,
+		struct timespec *ts)
+{
+	bool state;
+	struct timespec temp;
+	spin_lock_bh(&timestamp_lock);
+	timer->work_pending = false;
+	if ((ts->tv_sec - timer->last_modified_timer.tv_sec) > timer->timeout ||
+			timer->delayed_timer_trigger.tv_sec != 0) {
+		state = false;
+		temp.tv_sec = timer->timeout;
+		temp.tv_nsec = 0;
+		if (timer->delayed_timer_trigger.tv_sec != 0) {
+			temp = timespec_add(timer->delayed_timer_trigger, temp);
+			ts->tv_sec = temp.tv_sec;
+			ts->tv_nsec = temp.tv_nsec;
+			timer->delayed_timer_trigger.tv_sec = 0;
+			timer->work_pending = true;
+			schedule_work(&timer->work);
+		} else {
+			temp = timespec_add(timer->last_modified_timer, temp);
+			ts->tv_sec = temp.tv_sec;
+			ts->tv_nsec = temp.tv_nsec;
+		}
+	} else {
+		state = timer->active;
+	}
+	spin_unlock_bh(&timestamp_lock);
+	return state;
+}
+
+static void notify_netlink_uevent(const char *iface, struct idletimer_tg *timer)
+{
+	char iface_msg[NLMSG_MAX_SIZE];
+	char state_msg[NLMSG_MAX_SIZE];
+	char timestamp_msg[NLMSG_MAX_SIZE];
+	char uid_msg[NLMSG_MAX_SIZE];
+	char *envp[] = { iface_msg, state_msg, timestamp_msg, uid_msg, NULL };
+	int res;
+	struct timespec ts;
+	uint64_t time_ns;
+	bool state;
+
+	res = snprintf(iface_msg, NLMSG_MAX_SIZE, "INTERFACE=%s",
+		       iface);
+	if (NLMSG_MAX_SIZE <= res) {
+		pr_err("message too long (%d)", res);
+		return;
+	}
+
+	get_monotonic_boottime(&ts);
+	state = check_for_delayed_trigger(timer, &ts);
+	res = snprintf(state_msg, NLMSG_MAX_SIZE, "STATE=%s",
+			state ? "active" : "inactive");
+
+	if (NLMSG_MAX_SIZE <= res) {
+		pr_err("message too long (%d)", res);
+		return;
+	}
+
+	if (state) {
+		res = snprintf(uid_msg, NLMSG_MAX_SIZE, "UID=%u", timer->uid);
+		if (NLMSG_MAX_SIZE <= res)
+			pr_err("message too long (%d)", res);
+	} else {
+		res = snprintf(uid_msg, NLMSG_MAX_SIZE, "UID=");
+		if (NLMSG_MAX_SIZE <= res)
+			pr_err("message too long (%d)", res);
+	}
+
+	time_ns = timespec_to_ns(&ts);
+	res = snprintf(timestamp_msg, NLMSG_MAX_SIZE, "TIME_NS=%llu", time_ns);
+	if (NLMSG_MAX_SIZE <= res) {
+		timestamp_msg[0] = '\0';
+		pr_err("message too long (%d)", res);
+	}
+
+	pr_debug("putting nlmsg: <%s> <%s> <%s> <%s>\n", iface_msg, state_msg,
+		 timestamp_msg, uid_msg);
+	kobject_uevent_env(idletimer_tg_kobj, KOBJ_CHANGE, envp);
+	return;
+
+
+}
+
 static
 struct idletimer_tg *__idletimer_tg_find_by_label(const char *label)
 {
@@ -83,6 +189,7 @@ static ssize_t idletimer_tg_show(struct kobject *kobj, struct attribute *attr,
 {
 	struct idletimer_tg *timer;
 	unsigned long expires = 0;
+	unsigned long now = jiffies;
 
 	mutex_lock(&list_mutex);
 
@@ -92,11 +199,15 @@ static ssize_t idletimer_tg_show(struct kobject *kobj, struct attribute *attr,
 
 	mutex_unlock(&list_mutex);
 
-	if (time_after(expires, jiffies))
+	if (time_after(expires, now))
 		return sprintf(buf, "%u\n",
-			       jiffies_to_msecs(expires - jiffies) / 1000);
+			       jiffies_to_msecs(expires - now) / 1000);
 
-	return sprintf(buf, "0\n");
+	if (timer->send_nl_msg)
+		return sprintf(buf, "0 %d\n",
+			jiffies_to_msecs(now - expires) / 1000);
+	else
+		return sprintf(buf, "0\n");
 }
 
 static void idletimer_tg_work(struct work_struct *work)
@@ -105,6 +216,9 @@ static void idletimer_tg_work(struct work_struct *work)
 						  work);
 
 	sysfs_notify(idletimer_tg_kobj, NULL, timer->attr.attr.name);
+
+	if (timer->send_nl_msg)
+		notify_netlink_uevent(timer->attr.attr.name, timer);
 }
 
 static void idletimer_tg_expired(unsigned long data)
@@ -112,8 +226,55 @@ static void idletimer_tg_expired(unsigned long data)
 	struct idletimer_tg *timer = (struct idletimer_tg *) data;
 
 	pr_debug("timer %s expired\n", timer->attr.attr.name);
-
+	spin_lock_bh(&timestamp_lock);
+	timer->active = false;
+	timer->work_pending = true;
 	schedule_work(&timer->work);
+	spin_unlock_bh(&timestamp_lock);
+}
+
+static int idletimer_resume(struct notifier_block *notifier,
+		unsigned long pm_event, void *unused)
+{
+	struct timespec ts;
+	unsigned long time_diff, now = jiffies;
+	struct idletimer_tg *timer = container_of(notifier,
+			struct idletimer_tg, pm_nb);
+	if (!timer)
+		return NOTIFY_DONE;
+	switch (pm_event) {
+	case PM_SUSPEND_PREPARE:
+		get_monotonic_boottime(&timer->last_suspend_time);
+		break;
+	case PM_POST_SUSPEND:
+		spin_lock_bh(&timestamp_lock);
+		if (!timer->active) {
+			spin_unlock_bh(&timestamp_lock);
+			break;
+		}
+		/* since jiffies are not updated when suspended now represents
+		 * the time it would have suspended */
+		if (time_after(timer->timer.expires, now)) {
+			get_monotonic_boottime(&ts);
+			ts = timespec_sub(ts, timer->last_suspend_time);
+			time_diff = timespec_to_jiffies(&ts);
+			if (timer->timer.expires > (time_diff + now)) {
+				mod_timer_pending(&timer->timer,
+						(timer->timer.expires - time_diff));
+			} else {
+				del_timer(&timer->timer);
+				timer->timer.expires = 0;
+				timer->active = false;
+				timer->work_pending = true;
+				schedule_work(&timer->work);
+			}
+		}
+		spin_unlock_bh(&timestamp_lock);
+		break;
+	default:
+		break;
+	}
+	return NOTIFY_DONE;
 }
 
 static int idletimer_tg_create(struct idletimer_tg_info *info)
@@ -146,6 +307,21 @@ static int idletimer_tg_create(struct idletimer_tg_info *info)
 	setup_timer(&info->timer->timer, idletimer_tg_expired,
 		    (unsigned long) info->timer);
 	info->timer->refcnt = 1;
+	info->timer->send_nl_msg = (info->send_nl_msg == 0) ? false : true;
+	info->timer->active = true;
+	info->timer->timeout = info->timeout;
+
+	info->timer->delayed_timer_trigger.tv_sec = 0;
+	info->timer->delayed_timer_trigger.tv_nsec = 0;
+	info->timer->work_pending = false;
+	info->timer->uid = 0;
+	get_monotonic_boottime(&info->timer->last_modified_timer);
+
+	info->timer->pm_nb.notifier_call = idletimer_resume;
+	ret = register_pm_notifier(&info->timer->pm_nb);
+	if (ret)
+		printk(KERN_WARNING "[%s] Failed to register pm notifier %d\n",
+				__func__, ret);
 
 	mod_timer(&info->timer->timer,
 		  msecs_to_jiffies(info->timeout * 1000) + jiffies);
@@ -162,6 +338,42 @@ out:
 	return ret;
 }
 
+static void reset_timer(const struct idletimer_tg_info *info,
+			struct sk_buff *skb)
+{
+	unsigned long now = jiffies;
+	struct idletimer_tg *timer = info->timer;
+	bool timer_prev;
+
+	spin_lock_bh(&timestamp_lock);
+	timer_prev = timer->active;
+	timer->active = true;
+	/* timer_prev is used to guard overflow problem in time_before*/
+	if (!timer_prev || time_before(timer->timer.expires, now)) {
+		pr_debug("Starting Checkentry timer (Expired, Jiffies): %lu, %lu\n",
+				timer->timer.expires, now);
+
+		/* Stores the uid resposible for waking up the radio */
+		if (skb && (skb->sk)) {
+			timer->uid = from_kuid_munged(current_user_ns(),
+					sock_i_uid(skb_to_full_sk(skb)));
+		}
+
+		/* checks if there is a pending inactive notification*/
+		if (timer->work_pending)
+			timer->delayed_timer_trigger = timer->last_modified_timer;
+		else {
+			timer->work_pending = true;
+			schedule_work(&timer->work);
+		}
+	}
+
+	get_monotonic_boottime(&timer->last_modified_timer);
+	mod_timer(&timer->timer,
+			msecs_to_jiffies(info->timeout * 1000) + now);
+	spin_unlock_bh(&timestamp_lock);
+}
+
 /*
  * The actual xt_tables plugin.
  */
@@ -169,15 +381,23 @@ static unsigned int idletimer_tg_target(struct sk_buff *skb,
 					 const struct xt_action_param *par)
 {
 	const struct idletimer_tg_info *info = par->targinfo;
+	unsigned long now = jiffies;
 
 	pr_debug("resetting timer %s, timeout period %u\n",
 		 info->label, info->timeout);
 
 	BUG_ON(!info->timer);
 
-	mod_timer(&info->timer->timer,
-		  msecs_to_jiffies(info->timeout * 1000) + jiffies);
+	info->timer->active = true;
 
+	if (time_before(info->timer->timer.expires, now)) {
+		schedule_work(&info->timer->work);
+		pr_debug("Starting timer %s (Expired, Jiffies): %lu, %lu\n",
+			 info->label, info->timer->timer.expires, now);
+	}
+
+	/* TODO: Avoid modifying timers on each packet */
+	reset_timer(info, skb);
 	return XT_CONTINUE;
 }
 
@@ -186,7 +406,7 @@ static int idletimer_tg_checkentry(const struct xt_tgchk_param *par)
 	struct idletimer_tg_info *info = par->targinfo;
 	int ret;
 
-	pr_debug("checkentry targinfo%s\n", info->label);
+	pr_debug("checkentry targinfo %s\n", info->label);
 
 	if (info->timeout == 0) {
 		pr_debug("timeout value is zero\n");
@@ -205,9 +425,7 @@ static int idletimer_tg_checkentry(const struct xt_tgchk_param *par)
 	info->timer = __idletimer_tg_find_by_label(info->label);
 	if (info->timer) {
 		info->timer->refcnt++;
-		mod_timer(&info->timer->timer,
-			  msecs_to_jiffies(info->timeout * 1000) + jiffies);
-
+		reset_timer(info, NULL);
 		pr_debug("increased refcnt of timer %s to %u\n",
 			 info->label, info->timer->refcnt);
 	} else {
@@ -220,6 +438,7 @@ static int idletimer_tg_checkentry(const struct xt_tgchk_param *par)
 	}
 
 	mutex_unlock(&list_mutex);
+
 	return 0;
 }
 
@@ -237,11 +456,13 @@ static void idletimer_tg_destroy(const struct xt_tgdtor_param *par)
 		list_del(&info->timer->entry);
 		del_timer_sync(&info->timer->timer);
 		sysfs_remove_file(idletimer_tg_kobj, &info->timer->attr.attr);
+		unregister_pm_notifier(&info->timer->pm_nb);
+		cancel_work_sync(&info->timer->work);
 		kfree(info->timer->attr.attr.name);
 		kfree(info->timer);
 	} else {
 		pr_debug("decreased refcnt of timer %s to %u\n",
-			 info->label, info->timer->refcnt);
+		info->label, info->timer->refcnt);
 	}
 
 	mutex_unlock(&list_mutex);
@@ -249,6 +470,7 @@ static void idletimer_tg_destroy(const struct xt_tgdtor_param *par)
 
 static struct xt_target idletimer_tg __read_mostly = {
 	.name		= "IDLETIMER",
+	.revision	= 1,
 	.family		= NFPROTO_UNSPEC,
 	.target		= idletimer_tg_target,
 	.targetsize     = sizeof(struct idletimer_tg_info),
@@ -314,3 +536,4 @@ MODULE_DESCRIPTION("Xtables: idle time monitor");
 MODULE_LICENSE("GPL v2");
 MODULE_ALIAS("ipt_IDLETIMER");
 MODULE_ALIAS("ip6t_IDLETIMER");
+MODULE_ALIAS("arpt_IDLETIMER");
diff --git a/net/netfilter/xt_qtaguid.c b/net/netfilter/xt_qtaguid.c
new file mode 100644
index 000000000000..6f5cdc5b505f
--- /dev/null
+++ b/net/netfilter/xt_qtaguid.c
@@ -0,0 +1,3018 @@
+/*
+ * Kernel iptables module to track stats for packets based on user tags.
+ *
+ * (C) 2011 Google, Inc
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * There are run-time debug flags enabled via the debug_mask module param, or
+ * via the DEFAULT_DEBUG_MASK. See xt_qtaguid_internal.h.
+ */
+#define DEBUG
+
+#include <linux/file.h>
+#include <linux/inetdevice.h>
+#include <linux/module.h>
+#include <linux/miscdevice.h>
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_qtaguid.h>
+#include <linux/ratelimit.h>
+#include <linux/seq_file.h>
+#include <linux/skbuff.h>
+#include <linux/workqueue.h>
+#include <net/addrconf.h>
+#include <net/sock.h>
+#include <net/tcp.h>
+#include <net/udp.h>
+
+#if defined(CONFIG_IP6_NF_IPTABLES) || defined(CONFIG_IP6_NF_IPTABLES_MODULE)
+#include <linux/netfilter_ipv6/ip6_tables.h>
+#endif
+
+#include <linux/netfilter/xt_socket.h>
+#include "xt_qtaguid_internal.h"
+#include "xt_qtaguid_print.h"
+#include "../../fs/proc/internal.h"
+
+/*
+ * We only use the xt_socket funcs within a similar context to avoid unexpected
+ * return values.
+ */
+#define XT_SOCKET_SUPPORTED_HOOKS \
+	((1 << NF_INET_PRE_ROUTING) | (1 << NF_INET_LOCAL_IN))
+
+
+static const char *module_procdirname = "xt_qtaguid";
+static struct proc_dir_entry *xt_qtaguid_procdir;
+
+static unsigned int proc_iface_perms = S_IRUGO;
+module_param_named(iface_perms, proc_iface_perms, uint, S_IRUGO | S_IWUSR);
+
+static struct proc_dir_entry *xt_qtaguid_stats_file;
+static unsigned int proc_stats_perms = S_IRUGO;
+module_param_named(stats_perms, proc_stats_perms, uint, S_IRUGO | S_IWUSR);
+
+static struct proc_dir_entry *xt_qtaguid_ctrl_file;
+
+/* Everybody can write. But proc_ctrl_write_limited is true by default which
+ * limits what can be controlled. See the can_*() functions.
+ */
+static unsigned int proc_ctrl_perms = S_IRUGO | S_IWUGO;
+module_param_named(ctrl_perms, proc_ctrl_perms, uint, S_IRUGO | S_IWUSR);
+
+/* Limited by default, so the gid of the ctrl and stats proc entries
+ * will limit what can be done. See the can_*() functions.
+ */
+static bool proc_stats_readall_limited = true;
+static bool proc_ctrl_write_limited = true;
+
+module_param_named(stats_readall_limited, proc_stats_readall_limited, bool,
+		   S_IRUGO | S_IWUSR);
+module_param_named(ctrl_write_limited, proc_ctrl_write_limited, bool,
+		   S_IRUGO | S_IWUSR);
+
+/*
+ * Limit the number of active tags (via socket tags) for a given UID.
+ * Multiple processes could share the UID.
+ */
+static int max_sock_tags = DEFAULT_MAX_SOCK_TAGS;
+module_param(max_sock_tags, int, S_IRUGO | S_IWUSR);
+
+/*
+ * After the kernel has initiallized this module, it is still possible
+ * to make it passive.
+ * Setting passive to Y:
+ *  - the iface stats handling will not act on notifications.
+ *  - iptables matches will never match.
+ *  - ctrl commands silently succeed.
+ *  - stats are always empty.
+ * This is mostly usefull when a bug is suspected.
+ */
+static bool module_passive;
+module_param_named(passive, module_passive, bool, S_IRUGO | S_IWUSR);
+
+/*
+ * Control how qtaguid data is tracked per proc/uid.
+ * Setting tag_tracking_passive to Y:
+ *  - don't create proc specific structs to track tags
+ *  - don't check that active tag stats exceed some limits.
+ *  - don't clean up socket tags on process exits.
+ * This is mostly usefull when a bug is suspected.
+ */
+static bool qtu_proc_handling_passive;
+module_param_named(tag_tracking_passive, qtu_proc_handling_passive, bool,
+		   S_IRUGO | S_IWUSR);
+
+#define QTU_DEV_NAME "xt_qtaguid"
+
+uint qtaguid_debug_mask = DEFAULT_DEBUG_MASK;
+module_param_named(debug_mask, qtaguid_debug_mask, uint, S_IRUGO | S_IWUSR);
+
+/*---------------------------------------------------------------------------*/
+static const char *iface_stat_procdirname = "iface_stat";
+static struct proc_dir_entry *iface_stat_procdir;
+/*
+ * The iface_stat_all* will go away once userspace gets use to the new fields
+ * that have a format line.
+ */
+static const char *iface_stat_all_procfilename = "iface_stat_all";
+static struct proc_dir_entry *iface_stat_all_procfile;
+static const char *iface_stat_fmt_procfilename = "iface_stat_fmt";
+static struct proc_dir_entry *iface_stat_fmt_procfile;
+
+
+static LIST_HEAD(iface_stat_list);
+static DEFINE_SPINLOCK(iface_stat_list_lock);
+
+static struct rb_root sock_tag_tree = RB_ROOT;
+static DEFINE_SPINLOCK(sock_tag_list_lock);
+
+static struct rb_root tag_counter_set_tree = RB_ROOT;
+static DEFINE_SPINLOCK(tag_counter_set_list_lock);
+
+static struct rb_root uid_tag_data_tree = RB_ROOT;
+static DEFINE_SPINLOCK(uid_tag_data_tree_lock);
+
+static struct rb_root proc_qtu_data_tree = RB_ROOT;
+/* No proc_qtu_data_tree_lock; use uid_tag_data_tree_lock */
+
+static struct qtaguid_event_counts qtu_events;
+/*----------------------------------------------*/
+static bool can_manipulate_uids(void)
+{
+	/* root pwnd */
+	return in_egroup_p(xt_qtaguid_ctrl_file->gid)
+		|| unlikely(!from_kuid(&init_user_ns, current_fsuid())) || unlikely(!proc_ctrl_write_limited)
+		|| unlikely(uid_eq(current_fsuid(), xt_qtaguid_ctrl_file->uid));
+}
+
+static bool can_impersonate_uid(kuid_t uid)
+{
+	return uid_eq(uid, current_fsuid()) || can_manipulate_uids();
+}
+
+static bool can_read_other_uid_stats(kuid_t uid)
+{
+	/* root pwnd */
+	return in_egroup_p(xt_qtaguid_stats_file->gid)
+		|| unlikely(!from_kuid(&init_user_ns, current_fsuid())) || uid_eq(uid, current_fsuid())
+		|| unlikely(!proc_stats_readall_limited)
+		|| unlikely(uid_eq(current_fsuid(), xt_qtaguid_ctrl_file->uid));
+}
+
+static inline void dc_add_byte_packets(struct data_counters *counters, int set,
+				  enum ifs_tx_rx direction,
+				  enum ifs_proto ifs_proto,
+				  int bytes,
+				  int packets)
+{
+	counters->bpc[set][direction][ifs_proto].bytes += bytes;
+	counters->bpc[set][direction][ifs_proto].packets += packets;
+}
+
+static struct tag_node *tag_node_tree_search(struct rb_root *root, tag_t tag)
+{
+	struct rb_node *node = root->rb_node;
+
+	while (node) {
+		struct tag_node *data = rb_entry(node, struct tag_node, node);
+		int result;
+		RB_DEBUG("qtaguid: tag_node_tree_search(0x%llx): "
+			 " node=%p data=%p\n", tag, node, data);
+		result = tag_compare(tag, data->tag);
+		RB_DEBUG("qtaguid: tag_node_tree_search(0x%llx): "
+			 " data.tag=0x%llx (uid=%u) res=%d\n",
+			 tag, data->tag, get_uid_from_tag(data->tag), result);
+		if (result < 0)
+			node = node->rb_left;
+		else if (result > 0)
+			node = node->rb_right;
+		else
+			return data;
+	}
+	return NULL;
+}
+
+static void tag_node_tree_insert(struct tag_node *data, struct rb_root *root)
+{
+	struct rb_node **new = &(root->rb_node), *parent = NULL;
+
+	/* Figure out where to put new node */
+	while (*new) {
+		struct tag_node *this = rb_entry(*new, struct tag_node,
+						 node);
+		int result = tag_compare(data->tag, this->tag);
+		RB_DEBUG("qtaguid: %s(): tag=0x%llx"
+			 " (uid=%u)\n", __func__,
+			 this->tag,
+			 get_uid_from_tag(this->tag));
+		parent = *new;
+		if (result < 0)
+			new = &((*new)->rb_left);
+		else if (result > 0)
+			new = &((*new)->rb_right);
+		else
+			BUG();
+	}
+
+	/* Add new node and rebalance tree. */
+	rb_link_node(&data->node, parent, new);
+	rb_insert_color(&data->node, root);
+}
+
+static void tag_stat_tree_insert(struct tag_stat *data, struct rb_root *root)
+{
+	tag_node_tree_insert(&data->tn, root);
+}
+
+static struct tag_stat *tag_stat_tree_search(struct rb_root *root, tag_t tag)
+{
+	struct tag_node *node = tag_node_tree_search(root, tag);
+	if (!node)
+		return NULL;
+	return rb_entry(&node->node, struct tag_stat, tn.node);
+}
+
+static void tag_counter_set_tree_insert(struct tag_counter_set *data,
+					struct rb_root *root)
+{
+	tag_node_tree_insert(&data->tn, root);
+}
+
+static struct tag_counter_set *tag_counter_set_tree_search(struct rb_root *root,
+							   tag_t tag)
+{
+	struct tag_node *node = tag_node_tree_search(root, tag);
+	if (!node)
+		return NULL;
+	return rb_entry(&node->node, struct tag_counter_set, tn.node);
+
+}
+
+static void tag_ref_tree_insert(struct tag_ref *data, struct rb_root *root)
+{
+	tag_node_tree_insert(&data->tn, root);
+}
+
+static struct tag_ref *tag_ref_tree_search(struct rb_root *root, tag_t tag)
+{
+	struct tag_node *node = tag_node_tree_search(root, tag);
+	if (!node)
+		return NULL;
+	return rb_entry(&node->node, struct tag_ref, tn.node);
+}
+
+static struct sock_tag *sock_tag_tree_search(struct rb_root *root,
+					     const struct sock *sk)
+{
+	struct rb_node *node = root->rb_node;
+
+	while (node) {
+		struct sock_tag *data = rb_entry(node, struct sock_tag,
+						 sock_node);
+		if (sk < data->sk)
+			node = node->rb_left;
+		else if (sk > data->sk)
+			node = node->rb_right;
+		else
+			return data;
+	}
+	return NULL;
+}
+
+static void sock_tag_tree_insert(struct sock_tag *data, struct rb_root *root)
+{
+	struct rb_node **new = &(root->rb_node), *parent = NULL;
+
+	/* Figure out where to put new node */
+	while (*new) {
+		struct sock_tag *this = rb_entry(*new, struct sock_tag,
+						 sock_node);
+		parent = *new;
+		if (data->sk < this->sk)
+			new = &((*new)->rb_left);
+		else if (data->sk > this->sk)
+			new = &((*new)->rb_right);
+		else
+			BUG();
+	}
+
+	/* Add new node and rebalance tree. */
+	rb_link_node(&data->sock_node, parent, new);
+	rb_insert_color(&data->sock_node, root);
+}
+
+static void sock_tag_tree_erase(struct rb_root *st_to_free_tree)
+{
+	struct rb_node *node;
+	struct sock_tag *st_entry;
+
+	node = rb_first(st_to_free_tree);
+	while (node) {
+		st_entry = rb_entry(node, struct sock_tag, sock_node);
+		node = rb_next(node);
+		CT_DEBUG("qtaguid: %s(): "
+			 "erase st: sk=%p tag=0x%llx (uid=%u)\n", __func__,
+			 st_entry->sk,
+			 st_entry->tag,
+			 get_uid_from_tag(st_entry->tag));
+		rb_erase(&st_entry->sock_node, st_to_free_tree);
+		sock_put(st_entry->sk);
+		kfree(st_entry);
+	}
+}
+
+static struct proc_qtu_data *proc_qtu_data_tree_search(struct rb_root *root,
+						       const pid_t pid)
+{
+	struct rb_node *node = root->rb_node;
+
+	while (node) {
+		struct proc_qtu_data *data = rb_entry(node,
+						      struct proc_qtu_data,
+						      node);
+		if (pid < data->pid)
+			node = node->rb_left;
+		else if (pid > data->pid)
+			node = node->rb_right;
+		else
+			return data;
+	}
+	return NULL;
+}
+
+static void proc_qtu_data_tree_insert(struct proc_qtu_data *data,
+				      struct rb_root *root)
+{
+	struct rb_node **new = &(root->rb_node), *parent = NULL;
+
+	/* Figure out where to put new node */
+	while (*new) {
+		struct proc_qtu_data *this = rb_entry(*new,
+						      struct proc_qtu_data,
+						      node);
+		parent = *new;
+		if (data->pid < this->pid)
+			new = &((*new)->rb_left);
+		else if (data->pid > this->pid)
+			new = &((*new)->rb_right);
+		else
+			BUG();
+	}
+
+	/* Add new node and rebalance tree. */
+	rb_link_node(&data->node, parent, new);
+	rb_insert_color(&data->node, root);
+}
+
+static void uid_tag_data_tree_insert(struct uid_tag_data *data,
+				     struct rb_root *root)
+{
+	struct rb_node **new = &(root->rb_node), *parent = NULL;
+
+	/* Figure out where to put new node */
+	while (*new) {
+		struct uid_tag_data *this = rb_entry(*new,
+						     struct uid_tag_data,
+						     node);
+		parent = *new;
+		if (data->uid < this->uid)
+			new = &((*new)->rb_left);
+		else if (data->uid > this->uid)
+			new = &((*new)->rb_right);
+		else
+			BUG();
+	}
+
+	/* Add new node and rebalance tree. */
+	rb_link_node(&data->node, parent, new);
+	rb_insert_color(&data->node, root);
+}
+
+static struct uid_tag_data *uid_tag_data_tree_search(struct rb_root *root,
+						     uid_t uid)
+{
+	struct rb_node *node = root->rb_node;
+
+	while (node) {
+		struct uid_tag_data *data = rb_entry(node,
+						     struct uid_tag_data,
+						     node);
+		if (uid < data->uid)
+			node = node->rb_left;
+		else if (uid > data->uid)
+			node = node->rb_right;
+		else
+			return data;
+	}
+	return NULL;
+}
+
+/*
+ * Allocates a new uid_tag_data struct if needed.
+ * Returns a pointer to the found or allocated uid_tag_data.
+ * Returns a PTR_ERR on failures, and lock is not held.
+ * If found is not NULL:
+ *   sets *found to true if not allocated.
+ *   sets *found to false if allocated.
+ */
+struct uid_tag_data *get_uid_data(uid_t uid, bool *found_res)
+{
+	struct uid_tag_data *utd_entry;
+
+	/* Look for top level uid_tag_data for the UID */
+	utd_entry = uid_tag_data_tree_search(&uid_tag_data_tree, uid);
+	DR_DEBUG("qtaguid: get_uid_data(%u) utd=%p\n", uid, utd_entry);
+
+	if (found_res)
+		*found_res = utd_entry;
+	if (utd_entry)
+		return utd_entry;
+
+	utd_entry = kzalloc(sizeof(*utd_entry), GFP_ATOMIC);
+	if (!utd_entry) {
+		pr_err("qtaguid: get_uid_data(%u): "
+		       "tag data alloc failed\n", uid);
+		return ERR_PTR(-ENOMEM);
+	}
+
+	utd_entry->uid = uid;
+	utd_entry->tag_ref_tree = RB_ROOT;
+	uid_tag_data_tree_insert(utd_entry, &uid_tag_data_tree);
+	DR_DEBUG("qtaguid: get_uid_data(%u) new utd=%p\n", uid, utd_entry);
+	return utd_entry;
+}
+
+/* Never returns NULL. Either PTR_ERR or a valid ptr. */
+static struct tag_ref *new_tag_ref(tag_t new_tag,
+				   struct uid_tag_data *utd_entry)
+{
+	struct tag_ref *tr_entry;
+	int res;
+
+	if (utd_entry->num_active_tags + 1 > max_sock_tags) {
+		pr_info("qtaguid: new_tag_ref(0x%llx): "
+			"tag ref alloc quota exceeded. max=%d\n",
+			new_tag, max_sock_tags);
+		res = -EMFILE;
+		goto err_res;
+
+	}
+
+	tr_entry = kzalloc(sizeof(*tr_entry), GFP_ATOMIC);
+	if (!tr_entry) {
+		pr_err("qtaguid: new_tag_ref(0x%llx): "
+		       "tag ref alloc failed\n",
+		       new_tag);
+		res = -ENOMEM;
+		goto err_res;
+	}
+	tr_entry->tn.tag = new_tag;
+	/* tr_entry->num_sock_tags  handled by caller */
+	utd_entry->num_active_tags++;
+	tag_ref_tree_insert(tr_entry, &utd_entry->tag_ref_tree);
+	DR_DEBUG("qtaguid: new_tag_ref(0x%llx): "
+		 " inserted new tag ref %p\n",
+		 new_tag, tr_entry);
+	return tr_entry;
+
+err_res:
+	return ERR_PTR(res);
+}
+
+static struct tag_ref *lookup_tag_ref(tag_t full_tag,
+				      struct uid_tag_data **utd_res)
+{
+	struct uid_tag_data *utd_entry;
+	struct tag_ref *tr_entry;
+	bool found_utd;
+	uid_t uid = get_uid_from_tag(full_tag);
+
+	DR_DEBUG("qtaguid: lookup_tag_ref(tag=0x%llx (uid=%u))\n",
+		 full_tag, uid);
+
+	utd_entry = get_uid_data(uid, &found_utd);
+	if (IS_ERR_OR_NULL(utd_entry)) {
+		if (utd_res)
+			*utd_res = utd_entry;
+		return NULL;
+	}
+
+	tr_entry = tag_ref_tree_search(&utd_entry->tag_ref_tree, full_tag);
+	if (utd_res)
+		*utd_res = utd_entry;
+	DR_DEBUG("qtaguid: lookup_tag_ref(0x%llx) utd_entry=%p tr_entry=%p\n",
+		 full_tag, utd_entry, tr_entry);
+	return tr_entry;
+}
+
+/* Never returns NULL. Either PTR_ERR or a valid ptr. */
+static struct tag_ref *get_tag_ref(tag_t full_tag,
+				   struct uid_tag_data **utd_res)
+{
+	struct uid_tag_data *utd_entry;
+	struct tag_ref *tr_entry;
+
+	DR_DEBUG("qtaguid: get_tag_ref(0x%llx)\n",
+		 full_tag);
+	spin_lock_bh(&uid_tag_data_tree_lock);
+	tr_entry = lookup_tag_ref(full_tag, &utd_entry);
+	BUG_ON(IS_ERR_OR_NULL(utd_entry));
+	if (!tr_entry)
+		tr_entry = new_tag_ref(full_tag, utd_entry);
+
+	spin_unlock_bh(&uid_tag_data_tree_lock);
+	if (utd_res)
+		*utd_res = utd_entry;
+	DR_DEBUG("qtaguid: get_tag_ref(0x%llx) utd=%p tr=%p\n",
+		 full_tag, utd_entry, tr_entry);
+	return tr_entry;
+}
+
+/* Checks and maybe frees the UID Tag Data entry */
+static void put_utd_entry(struct uid_tag_data *utd_entry)
+{
+	/* Are we done with the UID tag data entry? */
+	if (RB_EMPTY_ROOT(&utd_entry->tag_ref_tree) &&
+		!utd_entry->num_pqd) {
+		DR_DEBUG("qtaguid: %s(): "
+			 "erase utd_entry=%p uid=%u "
+			 "by pid=%u tgid=%u uid=%u\n", __func__,
+			 utd_entry, utd_entry->uid,
+			 current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+		BUG_ON(utd_entry->num_active_tags);
+		rb_erase(&utd_entry->node, &uid_tag_data_tree);
+		kfree(utd_entry);
+	} else {
+		DR_DEBUG("qtaguid: %s(): "
+			 "utd_entry=%p still has %d tags %d proc_qtu_data\n",
+			 __func__, utd_entry, utd_entry->num_active_tags,
+			 utd_entry->num_pqd);
+		BUG_ON(!(utd_entry->num_active_tags ||
+			 utd_entry->num_pqd));
+	}
+}
+
+/*
+ * If no sock_tags are using this tag_ref,
+ * decrements refcount of utd_entry, removes tr_entry
+ * from utd_entry->tag_ref_tree and frees.
+ */
+static void free_tag_ref_from_utd_entry(struct tag_ref *tr_entry,
+					struct uid_tag_data *utd_entry)
+{
+	DR_DEBUG("qtaguid: %s(): %p tag=0x%llx (uid=%u)\n", __func__,
+		 tr_entry, tr_entry->tn.tag,
+		 get_uid_from_tag(tr_entry->tn.tag));
+	if (!tr_entry->num_sock_tags) {
+		BUG_ON(!utd_entry->num_active_tags);
+		utd_entry->num_active_tags--;
+		rb_erase(&tr_entry->tn.node, &utd_entry->tag_ref_tree);
+		DR_DEBUG("qtaguid: %s(): erased %p\n", __func__, tr_entry);
+		kfree(tr_entry);
+	}
+}
+
+static void put_tag_ref_tree(tag_t full_tag, struct uid_tag_data *utd_entry)
+{
+	struct rb_node *node;
+	struct tag_ref *tr_entry;
+	tag_t acct_tag;
+
+	DR_DEBUG("qtaguid: %s(tag=0x%llx (uid=%u))\n", __func__,
+		 full_tag, get_uid_from_tag(full_tag));
+	acct_tag = get_atag_from_tag(full_tag);
+	node = rb_first(&utd_entry->tag_ref_tree);
+	while (node) {
+		tr_entry = rb_entry(node, struct tag_ref, tn.node);
+		node = rb_next(node);
+		if (!acct_tag || tr_entry->tn.tag == full_tag)
+			free_tag_ref_from_utd_entry(tr_entry, utd_entry);
+	}
+}
+
+static ssize_t read_proc_u64(struct file *file, char __user *buf,
+			 size_t size, loff_t *ppos)
+{
+	uint64_t *valuep = PDE_DATA(file_inode(file));
+	char tmp[24];
+	size_t tmp_size;
+
+	tmp_size = scnprintf(tmp, sizeof(tmp), "%llu\n", *valuep);
+	return simple_read_from_buffer(buf, size, ppos, tmp, tmp_size);
+}
+
+static ssize_t read_proc_bool(struct file *file, char __user *buf,
+			  size_t size, loff_t *ppos)
+{
+	bool *valuep = PDE_DATA(file_inode(file));
+	char tmp[24];
+	size_t tmp_size;
+
+	tmp_size = scnprintf(tmp, sizeof(tmp), "%u\n", *valuep);
+	return simple_read_from_buffer(buf, size, ppos, tmp, tmp_size);
+}
+
+static int get_active_counter_set(tag_t tag)
+{
+	int active_set = 0;
+	struct tag_counter_set *tcs;
+
+	MT_DEBUG("qtaguid: get_active_counter_set(tag=0x%llx)"
+		 " (uid=%u)\n",
+		 tag, get_uid_from_tag(tag));
+	/* For now we only handle UID tags for active sets */
+	tag = get_utag_from_tag(tag);
+	spin_lock_bh(&tag_counter_set_list_lock);
+	tcs = tag_counter_set_tree_search(&tag_counter_set_tree, tag);
+	if (tcs)
+		active_set = tcs->active_set;
+	spin_unlock_bh(&tag_counter_set_list_lock);
+	return active_set;
+}
+
+/*
+ * Find the entry for tracking the specified interface.
+ * Caller must hold iface_stat_list_lock
+ */
+static struct iface_stat *get_iface_entry(const char *ifname)
+{
+	struct iface_stat *iface_entry;
+
+	/* Find the entry for tracking the specified tag within the interface */
+	if (ifname == NULL) {
+		pr_info("qtaguid: iface_stat: get() NULL device name\n");
+		return NULL;
+	}
+
+	/* Iterate over interfaces */
+	list_for_each_entry(iface_entry, &iface_stat_list, list) {
+		if (!strcmp(ifname, iface_entry->ifname))
+			goto done;
+	}
+	iface_entry = NULL;
+done:
+	return iface_entry;
+}
+
+/* This is for fmt2 only */
+static void pp_iface_stat_header(struct seq_file *m)
+{
+	seq_puts(m,
+		 "ifname "
+		 "total_skb_rx_bytes total_skb_rx_packets "
+		 "total_skb_tx_bytes total_skb_tx_packets "
+		 "rx_tcp_bytes rx_tcp_packets "
+		 "rx_udp_bytes rx_udp_packets "
+		 "rx_other_bytes rx_other_packets "
+		 "tx_tcp_bytes tx_tcp_packets "
+		 "tx_udp_bytes tx_udp_packets "
+		 "tx_other_bytes tx_other_packets\n"
+	);
+}
+
+static void pp_iface_stat_line(struct seq_file *m,
+			       struct iface_stat *iface_entry)
+{
+	struct data_counters *cnts;
+	int cnt_set = 0;   /* We only use one set for the device */
+	cnts = &iface_entry->totals_via_skb;
+	seq_printf(m, "%s %llu %llu %llu %llu %llu %llu %llu %llu "
+		   "%llu %llu %llu %llu %llu %llu %llu %llu\n",
+		   iface_entry->ifname,
+		   dc_sum_bytes(cnts, cnt_set, IFS_RX),
+		   dc_sum_packets(cnts, cnt_set, IFS_RX),
+		   dc_sum_bytes(cnts, cnt_set, IFS_TX),
+		   dc_sum_packets(cnts, cnt_set, IFS_TX),
+		   cnts->bpc[cnt_set][IFS_RX][IFS_TCP].bytes,
+		   cnts->bpc[cnt_set][IFS_RX][IFS_TCP].packets,
+		   cnts->bpc[cnt_set][IFS_RX][IFS_UDP].bytes,
+		   cnts->bpc[cnt_set][IFS_RX][IFS_UDP].packets,
+		   cnts->bpc[cnt_set][IFS_RX][IFS_PROTO_OTHER].bytes,
+		   cnts->bpc[cnt_set][IFS_RX][IFS_PROTO_OTHER].packets,
+		   cnts->bpc[cnt_set][IFS_TX][IFS_TCP].bytes,
+		   cnts->bpc[cnt_set][IFS_TX][IFS_TCP].packets,
+		   cnts->bpc[cnt_set][IFS_TX][IFS_UDP].bytes,
+		   cnts->bpc[cnt_set][IFS_TX][IFS_UDP].packets,
+		   cnts->bpc[cnt_set][IFS_TX][IFS_PROTO_OTHER].bytes,
+		   cnts->bpc[cnt_set][IFS_TX][IFS_PROTO_OTHER].packets);
+}
+
+struct proc_iface_stat_fmt_info {
+	int fmt;
+};
+
+static void *iface_stat_fmt_proc_start(struct seq_file *m, loff_t *pos)
+{
+	struct proc_iface_stat_fmt_info *p = m->private;
+	loff_t n = *pos;
+
+	/*
+	 * This lock will prevent iface_stat_update() from changing active,
+	 * and in turn prevent an interface from unregistering itself.
+	 */
+	spin_lock_bh(&iface_stat_list_lock);
+
+	if (unlikely(module_passive))
+		return NULL;
+
+	if (!n && p->fmt == 2)
+		pp_iface_stat_header(m);
+
+	return seq_list_start(&iface_stat_list, n);
+}
+
+static void *iface_stat_fmt_proc_next(struct seq_file *m, void *p, loff_t *pos)
+{
+	return seq_list_next(p, &iface_stat_list, pos);
+}
+
+static void iface_stat_fmt_proc_stop(struct seq_file *m, void *p)
+{
+	spin_unlock_bh(&iface_stat_list_lock);
+}
+
+static int iface_stat_fmt_proc_show(struct seq_file *m, void *v)
+{
+	struct proc_iface_stat_fmt_info *p = m->private;
+	struct iface_stat *iface_entry;
+	struct rtnl_link_stats64 dev_stats, *stats;
+	struct rtnl_link_stats64 no_dev_stats = {0};
+
+
+	CT_DEBUG("qtaguid:proc iface_stat_fmt pid=%u tgid=%u uid=%u\n",
+		 current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+
+	iface_entry = list_entry(v, struct iface_stat, list);
+
+	if (iface_entry->active) {
+		stats = dev_get_stats(iface_entry->net_dev,
+				      &dev_stats);
+	} else {
+		stats = &no_dev_stats;
+	}
+	/*
+	 * If the meaning of the data changes, then update the fmtX
+	 * string.
+	 */
+	if (p->fmt == 1) {
+		seq_printf(m, "%s %d %llu %llu %llu %llu %llu %llu %llu %llu\n",
+			   iface_entry->ifname,
+			   iface_entry->active,
+			   iface_entry->totals_via_dev[IFS_RX].bytes,
+			   iface_entry->totals_via_dev[IFS_RX].packets,
+			   iface_entry->totals_via_dev[IFS_TX].bytes,
+			   iface_entry->totals_via_dev[IFS_TX].packets,
+			   stats->rx_bytes, stats->rx_packets,
+			   stats->tx_bytes, stats->tx_packets
+			   );
+	} else {
+		pp_iface_stat_line(m, iface_entry);
+	}
+	return 0;
+}
+
+static const struct file_operations read_u64_fops = {
+	.read		= read_proc_u64,
+	.llseek		= default_llseek,
+};
+
+static const struct file_operations read_bool_fops = {
+	.read		= read_proc_bool,
+	.llseek		= default_llseek,
+};
+
+static void iface_create_proc_worker(struct work_struct *work)
+{
+	struct proc_dir_entry *proc_entry;
+	struct iface_stat_work *isw = container_of(work, struct iface_stat_work,
+						   iface_work);
+	struct iface_stat *new_iface  = isw->iface_entry;
+
+	/* iface_entries are not deleted, so safe to manipulate. */
+	proc_entry = proc_mkdir(new_iface->ifname, iface_stat_procdir);
+	if (IS_ERR_OR_NULL(proc_entry)) {
+		pr_err("qtaguid: iface_stat: create_proc(): alloc failed.\n");
+		kfree(isw);
+		return;
+	}
+
+	new_iface->proc_ptr = proc_entry;
+
+	proc_create_data("tx_bytes", proc_iface_perms, proc_entry,
+			 &read_u64_fops,
+			 &new_iface->totals_via_dev[IFS_TX].bytes);
+	proc_create_data("rx_bytes", proc_iface_perms, proc_entry,
+			 &read_u64_fops,
+			 &new_iface->totals_via_dev[IFS_RX].bytes);
+	proc_create_data("tx_packets", proc_iface_perms, proc_entry,
+			 &read_u64_fops,
+			 &new_iface->totals_via_dev[IFS_TX].packets);
+	proc_create_data("rx_packets", proc_iface_perms, proc_entry,
+			 &read_u64_fops,
+			 &new_iface->totals_via_dev[IFS_RX].packets);
+	proc_create_data("active", proc_iface_perms, proc_entry,
+			 &read_bool_fops, &new_iface->active);
+
+	IF_DEBUG("qtaguid: iface_stat: create_proc(): done "
+		 "entry=%p dev=%s\n", new_iface, new_iface->ifname);
+	kfree(isw);
+}
+
+/*
+ * Will set the entry's active state, and
+ * update the net_dev accordingly also.
+ */
+static void _iface_stat_set_active(struct iface_stat *entry,
+				   struct net_device *net_dev,
+				   bool activate)
+{
+	if (activate) {
+		entry->net_dev = net_dev;
+		entry->active = true;
+		IF_DEBUG("qtaguid: %s(%s): "
+			 "enable tracking. rfcnt=%d\n", __func__,
+			 entry->ifname,
+			 __this_cpu_read(*net_dev->pcpu_refcnt));
+	} else {
+		entry->active = false;
+		entry->net_dev = NULL;
+		IF_DEBUG("qtaguid: %s(%s): "
+			 "disable tracking. rfcnt=%d\n", __func__,
+			 entry->ifname,
+			 __this_cpu_read(*net_dev->pcpu_refcnt));
+
+	}
+}
+
+/* Caller must hold iface_stat_list_lock */
+static struct iface_stat *iface_alloc(struct net_device *net_dev)
+{
+	struct iface_stat *new_iface;
+	struct iface_stat_work *isw;
+
+	new_iface = kzalloc(sizeof(*new_iface), GFP_ATOMIC);
+	if (new_iface == NULL) {
+		pr_err("qtaguid: iface_stat: create(%s): "
+		       "iface_stat alloc failed\n", net_dev->name);
+		return NULL;
+	}
+	new_iface->ifname = kstrdup(net_dev->name, GFP_ATOMIC);
+	if (new_iface->ifname == NULL) {
+		pr_err("qtaguid: iface_stat: create(%s): "
+		       "ifname alloc failed\n", net_dev->name);
+		kfree(new_iface);
+		return NULL;
+	}
+	spin_lock_init(&new_iface->tag_stat_list_lock);
+	new_iface->tag_stat_tree = RB_ROOT;
+	_iface_stat_set_active(new_iface, net_dev, true);
+
+	/*
+	 * ipv6 notifier chains are atomic :(
+	 * No create_proc_read_entry() for you!
+	 */
+	isw = kmalloc(sizeof(*isw), GFP_ATOMIC);
+	if (!isw) {
+		pr_err("qtaguid: iface_stat: create(%s): "
+		       "work alloc failed\n", new_iface->ifname);
+		_iface_stat_set_active(new_iface, net_dev, false);
+		kfree(new_iface->ifname);
+		kfree(new_iface);
+		return NULL;
+	}
+	isw->iface_entry = new_iface;
+	INIT_WORK(&isw->iface_work, iface_create_proc_worker);
+	schedule_work(&isw->iface_work);
+	list_add(&new_iface->list, &iface_stat_list);
+	return new_iface;
+}
+
+static void iface_check_stats_reset_and_adjust(struct net_device *net_dev,
+					       struct iface_stat *iface)
+{
+	struct rtnl_link_stats64 dev_stats, *stats;
+	bool stats_rewound;
+
+	stats = dev_get_stats(net_dev, &dev_stats);
+	/* No empty packets */
+	stats_rewound =
+		(stats->rx_bytes < iface->last_known[IFS_RX].bytes)
+		|| (stats->tx_bytes < iface->last_known[IFS_TX].bytes);
+
+	IF_DEBUG("qtaguid: %s(%s): iface=%p netdev=%p "
+		 "bytes rx/tx=%llu/%llu "
+		 "active=%d last_known=%d "
+		 "stats_rewound=%d\n", __func__,
+		 net_dev ? net_dev->name : "?",
+		 iface, net_dev,
+		 stats->rx_bytes, stats->tx_bytes,
+		 iface->active, iface->last_known_valid, stats_rewound);
+
+	if (iface->active && iface->last_known_valid && stats_rewound) {
+		pr_warn_once("qtaguid: iface_stat: %s(%s): "
+			     "iface reset its stats unexpectedly\n", __func__,
+			     net_dev->name);
+
+		iface->totals_via_dev[IFS_TX].bytes +=
+			iface->last_known[IFS_TX].bytes;
+		iface->totals_via_dev[IFS_TX].packets +=
+			iface->last_known[IFS_TX].packets;
+		iface->totals_via_dev[IFS_RX].bytes +=
+			iface->last_known[IFS_RX].bytes;
+		iface->totals_via_dev[IFS_RX].packets +=
+			iface->last_known[IFS_RX].packets;
+		iface->last_known_valid = false;
+		IF_DEBUG("qtaguid: %s(%s): iface=%p "
+			 "used last known bytes rx/tx=%llu/%llu\n", __func__,
+			 iface->ifname, iface, iface->last_known[IFS_RX].bytes,
+			 iface->last_known[IFS_TX].bytes);
+	}
+}
+
+/*
+ * Create a new entry for tracking the specified interface.
+ * Do nothing if the entry already exists.
+ * Called when an interface is configured with a valid IP address.
+ */
+static void iface_stat_create(struct net_device *net_dev,
+			      struct in_ifaddr *ifa)
+{
+	struct in_device *in_dev = NULL;
+	const char *ifname;
+	struct iface_stat *entry;
+	__be32 ipaddr = 0;
+	struct iface_stat *new_iface;
+
+	IF_DEBUG("qtaguid: iface_stat: create(%s): ifa=%p netdev=%p\n",
+		 net_dev ? net_dev->name : "?",
+		 ifa, net_dev);
+	if (!net_dev) {
+		pr_err("qtaguid: iface_stat: create(): no net dev\n");
+		return;
+	}
+
+	ifname = net_dev->name;
+	if (!ifa) {
+		in_dev = in_dev_get(net_dev);
+		if (!in_dev) {
+			pr_err("qtaguid: iface_stat: create(%s): no inet dev\n",
+			       ifname);
+			return;
+		}
+		IF_DEBUG("qtaguid: iface_stat: create(%s): in_dev=%p\n",
+			 ifname, in_dev);
+		for (ifa = in_dev->ifa_list; ifa; ifa = ifa->ifa_next) {
+			IF_DEBUG("qtaguid: iface_stat: create(%s): "
+				 "ifa=%p ifa_label=%s\n",
+				 ifname, ifa,
+				 ifa->ifa_label ? ifa->ifa_label : "(null)");
+			if (ifa->ifa_label && !strcmp(ifname, ifa->ifa_label))
+				break;
+		}
+	}
+
+	if (!ifa) {
+		IF_DEBUG("qtaguid: iface_stat: create(%s): no matching IP\n",
+			 ifname);
+		goto done_put;
+	}
+	ipaddr = ifa->ifa_local;
+
+	spin_lock_bh(&iface_stat_list_lock);
+	entry = get_iface_entry(ifname);
+	if (entry != NULL) {
+		IF_DEBUG("qtaguid: iface_stat: create(%s): entry=%p\n",
+			 ifname, entry);
+		iface_check_stats_reset_and_adjust(net_dev, entry);
+		_iface_stat_set_active(entry, net_dev, true);
+		IF_DEBUG("qtaguid: %s(%s): "
+			 "tracking now %d on ip=%pI4\n", __func__,
+			 entry->ifname, true, &ipaddr);
+		goto done_unlock_put;
+	}
+
+	new_iface = iface_alloc(net_dev);
+	IF_DEBUG("qtaguid: iface_stat: create(%s): done "
+		 "entry=%p ip=%pI4\n", ifname, new_iface, &ipaddr);
+done_unlock_put:
+	spin_unlock_bh(&iface_stat_list_lock);
+done_put:
+	if (in_dev)
+		in_dev_put(in_dev);
+}
+
+static void iface_stat_create_ipv6(struct net_device *net_dev,
+				   struct inet6_ifaddr *ifa)
+{
+	struct in_device *in_dev;
+	const char *ifname;
+	struct iface_stat *entry;
+	struct iface_stat *new_iface;
+	int addr_type;
+
+	IF_DEBUG("qtaguid: iface_stat: create6(): ifa=%p netdev=%p->name=%s\n",
+		 ifa, net_dev, net_dev ? net_dev->name : "");
+	if (!net_dev) {
+		pr_err("qtaguid: iface_stat: create6(): no net dev!\n");
+		return;
+	}
+	ifname = net_dev->name;
+
+	in_dev = in_dev_get(net_dev);
+	if (!in_dev) {
+		pr_err("qtaguid: iface_stat: create6(%s): no inet dev\n",
+		       ifname);
+		return;
+	}
+
+	IF_DEBUG("qtaguid: iface_stat: create6(%s): in_dev=%p\n",
+		 ifname, in_dev);
+
+	if (!ifa) {
+		IF_DEBUG("qtaguid: iface_stat: create6(%s): no matching IP\n",
+			 ifname);
+		goto done_put;
+	}
+	addr_type = ipv6_addr_type(&ifa->addr);
+
+	spin_lock_bh(&iface_stat_list_lock);
+	entry = get_iface_entry(ifname);
+	if (entry != NULL) {
+		IF_DEBUG("qtaguid: %s(%s): entry=%p\n", __func__,
+			 ifname, entry);
+		iface_check_stats_reset_and_adjust(net_dev, entry);
+		_iface_stat_set_active(entry, net_dev, true);
+		IF_DEBUG("qtaguid: %s(%s): "
+			 "tracking now %d on ip=%pI6c\n", __func__,
+			 entry->ifname, true, &ifa->addr);
+		goto done_unlock_put;
+	}
+
+	new_iface = iface_alloc(net_dev);
+	IF_DEBUG("qtaguid: iface_stat: create6(%s): done "
+		 "entry=%p ip=%pI6c\n", ifname, new_iface, &ifa->addr);
+
+done_unlock_put:
+	spin_unlock_bh(&iface_stat_list_lock);
+done_put:
+	in_dev_put(in_dev);
+}
+
+static struct sock_tag *get_sock_stat_nl(const struct sock *sk)
+{
+	MT_DEBUG("qtaguid: get_sock_stat_nl(sk=%p)\n", sk);
+	return sock_tag_tree_search(&sock_tag_tree, sk);
+}
+
+static struct sock_tag *get_sock_stat(const struct sock *sk)
+{
+	struct sock_tag *sock_tag_entry;
+	MT_DEBUG("qtaguid: get_sock_stat(sk=%p)\n", sk);
+	if (!sk)
+		return NULL;
+	spin_lock_bh(&sock_tag_list_lock);
+	sock_tag_entry = get_sock_stat_nl(sk);
+	spin_unlock_bh(&sock_tag_list_lock);
+	return sock_tag_entry;
+}
+
+static int ipx_proto(const struct sk_buff *skb,
+		     struct xt_action_param *par)
+{
+	int thoff = 0, tproto;
+
+	switch (par->family) {
+	case NFPROTO_IPV6:
+		tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL);
+		if (tproto < 0)
+			MT_DEBUG("%s(): transport header not found in ipv6"
+				 " skb=%p\n", __func__, skb);
+		break;
+	case NFPROTO_IPV4:
+		tproto = ip_hdr(skb)->protocol;
+		break;
+	default:
+		tproto = IPPROTO_RAW;
+	}
+	return tproto;
+}
+
+static void
+data_counters_update(struct data_counters *dc, int set,
+		     enum ifs_tx_rx direction, int proto, int bytes)
+{
+	switch (proto) {
+	case IPPROTO_TCP:
+		dc_add_byte_packets(dc, set, direction, IFS_TCP, bytes, 1);
+		break;
+	case IPPROTO_UDP:
+		dc_add_byte_packets(dc, set, direction, IFS_UDP, bytes, 1);
+		break;
+	case IPPROTO_IP:
+	default:
+		dc_add_byte_packets(dc, set, direction, IFS_PROTO_OTHER, bytes,
+				    1);
+		break;
+	}
+}
+
+/*
+ * Update stats for the specified interface. Do nothing if the entry
+ * does not exist (when a device was never configured with an IP address).
+ * Called when an device is being unregistered.
+ */
+static void iface_stat_update(struct net_device *net_dev, bool stash_only)
+{
+	struct rtnl_link_stats64 dev_stats, *stats;
+	struct iface_stat *entry;
+
+	stats = dev_get_stats(net_dev, &dev_stats);
+	spin_lock_bh(&iface_stat_list_lock);
+	entry = get_iface_entry(net_dev->name);
+	if (entry == NULL) {
+		IF_DEBUG("qtaguid: iface_stat: update(%s): not tracked\n",
+			 net_dev->name);
+		spin_unlock_bh(&iface_stat_list_lock);
+		return;
+	}
+
+	IF_DEBUG("qtaguid: %s(%s): entry=%p\n", __func__,
+		 net_dev->name, entry);
+	if (!entry->active) {
+		IF_DEBUG("qtaguid: %s(%s): already disabled\n", __func__,
+			 net_dev->name);
+		spin_unlock_bh(&iface_stat_list_lock);
+		return;
+	}
+
+	if (stash_only) {
+		entry->last_known[IFS_TX].bytes = stats->tx_bytes;
+		entry->last_known[IFS_TX].packets = stats->tx_packets;
+		entry->last_known[IFS_RX].bytes = stats->rx_bytes;
+		entry->last_known[IFS_RX].packets = stats->rx_packets;
+		entry->last_known_valid = true;
+		IF_DEBUG("qtaguid: %s(%s): "
+			 "dev stats stashed rx/tx=%llu/%llu\n", __func__,
+			 net_dev->name, stats->rx_bytes, stats->tx_bytes);
+		spin_unlock_bh(&iface_stat_list_lock);
+		return;
+	}
+	entry->totals_via_dev[IFS_TX].bytes += stats->tx_bytes;
+	entry->totals_via_dev[IFS_TX].packets += stats->tx_packets;
+	entry->totals_via_dev[IFS_RX].bytes += stats->rx_bytes;
+	entry->totals_via_dev[IFS_RX].packets += stats->rx_packets;
+	/* We don't need the last_known[] anymore */
+	entry->last_known_valid = false;
+	_iface_stat_set_active(entry, net_dev, false);
+	IF_DEBUG("qtaguid: %s(%s): "
+		 "disable tracking. rx/tx=%llu/%llu\n", __func__,
+		 net_dev->name, stats->rx_bytes, stats->tx_bytes);
+	spin_unlock_bh(&iface_stat_list_lock);
+}
+
+/* Guarantied to return a net_device that has a name */
+static void get_dev_and_dir(const struct sk_buff *skb,
+			    struct xt_action_param *par,
+			    enum ifs_tx_rx *direction,
+			    const struct net_device **el_dev)
+{
+	BUG_ON(!direction || !el_dev);
+
+	if (par->in) {
+		*el_dev = par->in;
+		*direction = IFS_RX;
+	} else if (par->out) {
+		*el_dev = par->out;
+		*direction = IFS_TX;
+	} else {
+		pr_err("qtaguid[%d]: %s(): no par->in/out?!!\n",
+		       par->hooknum, __func__);
+		BUG();
+	}
+	if (unlikely(!(*el_dev)->name)) {
+		pr_err("qtaguid[%d]: %s(): no dev->name?!!\n",
+		       par->hooknum, __func__);
+		BUG();
+	}
+	if (skb->dev && *el_dev != skb->dev) {
+		MT_DEBUG("qtaguid[%d]: skb->dev=%p %s vs par->%s=%p %s\n",
+			 par->hooknum, skb->dev, skb->dev->name,
+			 *direction == IFS_RX ? "in" : "out",  *el_dev,
+			 (*el_dev)->name);
+	}
+}
+
+/*
+ * Update stats for the specified interface from the skb.
+ * Do nothing if the entry
+ * does not exist (when a device was never configured with an IP address).
+ * Called on each sk.
+ */
+static void iface_stat_update_from_skb(const struct sk_buff *skb,
+				       struct xt_action_param *par)
+{
+	struct iface_stat *entry;
+	const struct net_device *el_dev;
+	enum ifs_tx_rx direction;
+	int bytes = skb->len;
+	int proto;
+
+	get_dev_and_dir(skb, par, &direction, &el_dev);
+	proto = ipx_proto(skb, par);
+	MT_DEBUG("qtaguid[%d]: iface_stat: %s(%s): "
+		 "type=%d fam=%d proto=%d dir=%d\n",
+		 par->hooknum, __func__, el_dev->name, el_dev->type,
+		 par->family, proto, direction);
+
+	spin_lock_bh(&iface_stat_list_lock);
+	entry = get_iface_entry(el_dev->name);
+	if (entry == NULL) {
+		IF_DEBUG("qtaguid[%d]: iface_stat: %s(%s): not tracked\n",
+			 par->hooknum, __func__, el_dev->name);
+		spin_unlock_bh(&iface_stat_list_lock);
+		return;
+	}
+
+	IF_DEBUG("qtaguid[%d]: %s(%s): entry=%p\n", par->hooknum,  __func__,
+		 el_dev->name, entry);
+
+	data_counters_update(&entry->totals_via_skb, 0, direction, proto,
+			     bytes);
+	spin_unlock_bh(&iface_stat_list_lock);
+}
+
+static void tag_stat_update(struct tag_stat *tag_entry,
+			enum ifs_tx_rx direction, int proto, int bytes)
+{
+	int active_set;
+	active_set = get_active_counter_set(tag_entry->tn.tag);
+	MT_DEBUG("qtaguid: tag_stat_update(tag=0x%llx (uid=%u) set=%d "
+		 "dir=%d proto=%d bytes=%d)\n",
+		 tag_entry->tn.tag, get_uid_from_tag(tag_entry->tn.tag),
+		 active_set, direction, proto, bytes);
+	data_counters_update(&tag_entry->counters, active_set, direction,
+			     proto, bytes);
+	if (tag_entry->parent_counters)
+		data_counters_update(tag_entry->parent_counters, active_set,
+				     direction, proto, bytes);
+}
+
+/*
+ * Create a new entry for tracking the specified {acct_tag,uid_tag} within
+ * the interface.
+ * iface_entry->tag_stat_list_lock should be held.
+ */
+static struct tag_stat *create_if_tag_stat(struct iface_stat *iface_entry,
+					   tag_t tag)
+{
+	struct tag_stat *new_tag_stat_entry = NULL;
+	IF_DEBUG("qtaguid: iface_stat: %s(): ife=%p tag=0x%llx"
+		 " (uid=%u)\n", __func__,
+		 iface_entry, tag, get_uid_from_tag(tag));
+	new_tag_stat_entry = kzalloc(sizeof(*new_tag_stat_entry), GFP_ATOMIC);
+	if (!new_tag_stat_entry) {
+		pr_err("qtaguid: iface_stat: tag stat alloc failed\n");
+		goto done;
+	}
+	new_tag_stat_entry->tn.tag = tag;
+	tag_stat_tree_insert(new_tag_stat_entry, &iface_entry->tag_stat_tree);
+done:
+	return new_tag_stat_entry;
+}
+
+static void if_tag_stat_update(const char *ifname, uid_t uid,
+			       const struct sock *sk, enum ifs_tx_rx direction,
+			       int proto, int bytes)
+{
+	struct tag_stat *tag_stat_entry;
+	tag_t tag, acct_tag;
+	tag_t uid_tag;
+	struct data_counters *uid_tag_counters;
+	struct sock_tag *sock_tag_entry;
+	struct iface_stat *iface_entry;
+	struct tag_stat *new_tag_stat = NULL;
+	MT_DEBUG("qtaguid: if_tag_stat_update(ifname=%s "
+		"uid=%u sk=%p dir=%d proto=%d bytes=%d)\n",
+		 ifname, uid, sk, direction, proto, bytes);
+
+	spin_lock_bh(&iface_stat_list_lock);
+	iface_entry = get_iface_entry(ifname);
+	if (!iface_entry) {
+		pr_err_ratelimited("qtaguid: tag_stat: stat_update() "
+				   "%s not found\n", ifname);
+		spin_unlock_bh(&iface_stat_list_lock);
+		return;
+	}
+	/* It is ok to process data when an iface_entry is inactive */
+
+	MT_DEBUG("qtaguid: tag_stat: stat_update() dev=%s entry=%p\n",
+		 ifname, iface_entry);
+
+	/*
+	 * Look for a tagged sock.
+	 * It will have an acct_uid.
+	 */
+	sock_tag_entry = get_sock_stat(sk);
+	if (sock_tag_entry) {
+		tag = sock_tag_entry->tag;
+		acct_tag = get_atag_from_tag(tag);
+		uid_tag = get_utag_from_tag(tag);
+	} else {
+		acct_tag = make_atag_from_value(0);
+		tag = combine_atag_with_uid(acct_tag, uid);
+		uid_tag = make_tag_from_uid(uid);
+	}
+	MT_DEBUG("qtaguid: tag_stat: stat_update(): "
+		 " looking for tag=0x%llx (uid=%u) in ife=%p\n",
+		 tag, get_uid_from_tag(tag), iface_entry);
+	/* Loop over tag list under this interface for {acct_tag,uid_tag} */
+	spin_lock_bh(&iface_entry->tag_stat_list_lock);
+
+	tag_stat_entry = tag_stat_tree_search(&iface_entry->tag_stat_tree,
+					      tag);
+	if (tag_stat_entry) {
+		/*
+		 * Updating the {acct_tag, uid_tag} entry handles both stats:
+		 * {0, uid_tag} will also get updated.
+		 */
+		tag_stat_update(tag_stat_entry, direction, proto, bytes);
+		goto unlock;
+	}
+
+	/* Loop over tag list under this interface for {0,uid_tag} */
+	tag_stat_entry = tag_stat_tree_search(&iface_entry->tag_stat_tree,
+					      uid_tag);
+	if (!tag_stat_entry) {
+		/* Here: the base uid_tag did not exist */
+		/*
+		 * No parent counters. So
+		 *  - No {0, uid_tag} stats and no {acc_tag, uid_tag} stats.
+		 */
+		new_tag_stat = create_if_tag_stat(iface_entry, uid_tag);
+		if (!new_tag_stat)
+			goto unlock;
+		uid_tag_counters = &new_tag_stat->counters;
+	} else {
+		uid_tag_counters = &tag_stat_entry->counters;
+	}
+
+	if (acct_tag) {
+		/* Create the child {acct_tag, uid_tag} and hook up parent. */
+		new_tag_stat = create_if_tag_stat(iface_entry, tag);
+		if (!new_tag_stat)
+			goto unlock;
+		new_tag_stat->parent_counters = uid_tag_counters;
+	} else {
+		/*
+		 * For new_tag_stat to be still NULL here would require:
+		 *  {0, uid_tag} exists
+		 *  and {acct_tag, uid_tag} doesn't exist
+		 *  AND acct_tag == 0.
+		 * Impossible. This reassures us that new_tag_stat
+		 * below will always be assigned.
+		 */
+		BUG_ON(!new_tag_stat);
+	}
+	tag_stat_update(new_tag_stat, direction, proto, bytes);
+unlock:
+	spin_unlock_bh(&iface_entry->tag_stat_list_lock);
+	spin_unlock_bh(&iface_stat_list_lock);
+}
+
+static int iface_netdev_event_handler(struct notifier_block *nb,
+				      unsigned long event, void *ptr) {
+	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
+
+	if (unlikely(module_passive))
+		return NOTIFY_DONE;
+
+	IF_DEBUG("qtaguid: iface_stat: netdev_event(): "
+		 "ev=0x%lx/%s netdev=%p->name=%s\n",
+		 event, netdev_evt_str(event), dev, dev ? dev->name : "");
+
+	switch (event) {
+	case NETDEV_UP:
+		iface_stat_create(dev, NULL);
+		atomic64_inc(&qtu_events.iface_events);
+		break;
+	case NETDEV_DOWN:
+	case NETDEV_UNREGISTER:
+		iface_stat_update(dev, event == NETDEV_DOWN);
+		atomic64_inc(&qtu_events.iface_events);
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+static int iface_inet6addr_event_handler(struct notifier_block *nb,
+					 unsigned long event, void *ptr)
+{
+	struct inet6_ifaddr *ifa = ptr;
+	struct net_device *dev;
+
+	if (unlikely(module_passive))
+		return NOTIFY_DONE;
+
+	IF_DEBUG("qtaguid: iface_stat: inet6addr_event(): "
+		 "ev=0x%lx/%s ifa=%p\n",
+		 event, netdev_evt_str(event), ifa);
+
+	switch (event) {
+	case NETDEV_UP:
+		BUG_ON(!ifa || !ifa->idev);
+		dev = (struct net_device *)ifa->idev->dev;
+		iface_stat_create_ipv6(dev, ifa);
+		atomic64_inc(&qtu_events.iface_events);
+		break;
+	case NETDEV_DOWN:
+	case NETDEV_UNREGISTER:
+		BUG_ON(!ifa || !ifa->idev);
+		dev = (struct net_device *)ifa->idev->dev;
+		iface_stat_update(dev, event == NETDEV_DOWN);
+		atomic64_inc(&qtu_events.iface_events);
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+static int iface_inetaddr_event_handler(struct notifier_block *nb,
+					unsigned long event, void *ptr)
+{
+	struct in_ifaddr *ifa = ptr;
+	struct net_device *dev;
+
+	if (unlikely(module_passive))
+		return NOTIFY_DONE;
+
+	IF_DEBUG("qtaguid: iface_stat: inetaddr_event(): "
+		 "ev=0x%lx/%s ifa=%p\n",
+		 event, netdev_evt_str(event), ifa);
+
+	switch (event) {
+	case NETDEV_UP:
+		BUG_ON(!ifa || !ifa->ifa_dev);
+		dev = ifa->ifa_dev->dev;
+		iface_stat_create(dev, ifa);
+		atomic64_inc(&qtu_events.iface_events);
+		break;
+	case NETDEV_DOWN:
+	case NETDEV_UNREGISTER:
+		BUG_ON(!ifa || !ifa->ifa_dev);
+		dev = ifa->ifa_dev->dev;
+		iface_stat_update(dev, event == NETDEV_DOWN);
+		atomic64_inc(&qtu_events.iface_events);
+		break;
+	}
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block iface_netdev_notifier_blk = {
+	.notifier_call = iface_netdev_event_handler,
+};
+
+static struct notifier_block iface_inetaddr_notifier_blk = {
+	.notifier_call = iface_inetaddr_event_handler,
+};
+
+static struct notifier_block iface_inet6addr_notifier_blk = {
+	.notifier_call = iface_inet6addr_event_handler,
+};
+
+static const struct seq_operations iface_stat_fmt_proc_seq_ops = {
+	.start	= iface_stat_fmt_proc_start,
+	.next	= iface_stat_fmt_proc_next,
+	.stop	= iface_stat_fmt_proc_stop,
+	.show	= iface_stat_fmt_proc_show,
+};
+
+static int proc_iface_stat_fmt_open(struct inode *inode, struct file *file)
+{
+	struct proc_iface_stat_fmt_info *s;
+
+	s = __seq_open_private(file, &iface_stat_fmt_proc_seq_ops,
+			sizeof(struct proc_iface_stat_fmt_info));
+	if (!s)
+		return -ENOMEM;
+
+	s->fmt = (uintptr_t)PDE_DATA(inode);
+	return 0;
+}
+
+static const struct file_operations proc_iface_stat_fmt_fops = {
+	.open		= proc_iface_stat_fmt_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release_private,
+};
+
+static int __init iface_stat_init(struct proc_dir_entry *parent_procdir)
+{
+	int err;
+
+	iface_stat_procdir = proc_mkdir(iface_stat_procdirname, parent_procdir);
+	if (!iface_stat_procdir) {
+		pr_err("qtaguid: iface_stat: init failed to create proc entry\n");
+		err = -1;
+		goto err;
+	}
+
+	iface_stat_all_procfile = proc_create_data(iface_stat_all_procfilename,
+						   proc_iface_perms,
+						   parent_procdir,
+						   &proc_iface_stat_fmt_fops,
+						   (void *)1 /* fmt1 */);
+	if (!iface_stat_all_procfile) {
+		pr_err("qtaguid: iface_stat: init "
+		       " failed to create stat_old proc entry\n");
+		err = -1;
+		goto err_zap_entry;
+	}
+
+	iface_stat_fmt_procfile = proc_create_data(iface_stat_fmt_procfilename,
+						   proc_iface_perms,
+						   parent_procdir,
+						   &proc_iface_stat_fmt_fops,
+						   (void *)2 /* fmt2 */);
+	if (!iface_stat_fmt_procfile) {
+		pr_err("qtaguid: iface_stat: init "
+		       " failed to create stat_all proc entry\n");
+		err = -1;
+		goto err_zap_all_stats_entry;
+	}
+
+
+	err = register_netdevice_notifier(&iface_netdev_notifier_blk);
+	if (err) {
+		pr_err("qtaguid: iface_stat: init "
+		       "failed to register dev event handler\n");
+		goto err_zap_all_stats_entries;
+	}
+	err = register_inetaddr_notifier(&iface_inetaddr_notifier_blk);
+	if (err) {
+		pr_err("qtaguid: iface_stat: init "
+		       "failed to register ipv4 dev event handler\n");
+		goto err_unreg_nd;
+	}
+
+	err = register_inet6addr_notifier(&iface_inet6addr_notifier_blk);
+	if (err) {
+		pr_err("qtaguid: iface_stat: init "
+		       "failed to register ipv6 dev event handler\n");
+		goto err_unreg_ip4_addr;
+	}
+	return 0;
+
+err_unreg_ip4_addr:
+	unregister_inetaddr_notifier(&iface_inetaddr_notifier_blk);
+err_unreg_nd:
+	unregister_netdevice_notifier(&iface_netdev_notifier_blk);
+err_zap_all_stats_entries:
+	remove_proc_entry(iface_stat_fmt_procfilename, parent_procdir);
+err_zap_all_stats_entry:
+	remove_proc_entry(iface_stat_all_procfilename, parent_procdir);
+err_zap_entry:
+	remove_proc_entry(iface_stat_procdirname, parent_procdir);
+err:
+	return err;
+}
+
+static struct sock *qtaguid_find_sk(const struct sk_buff *skb,
+				    struct xt_action_param *par)
+{
+	struct sock *sk;
+	unsigned int hook_mask = (1 << par->hooknum);
+
+	MT_DEBUG("qtaguid[%d]: find_sk(skb=%p) family=%d\n",
+		 par->hooknum, skb, par->family);
+
+	/*
+	 * Let's not abuse the the xt_socket_get*_sk(), or else it will
+	 * return garbage SKs.
+	 */
+	if (!(hook_mask & XT_SOCKET_SUPPORTED_HOOKS))
+		return NULL;
+
+	switch (par->family) {
+	case NFPROTO_IPV6:
+		sk = xt_socket_lookup_slow_v6(dev_net(skb->dev), skb, par->in);
+		break;
+	case NFPROTO_IPV4:
+		sk = xt_socket_lookup_slow_v4(dev_net(skb->dev), skb, par->in);
+		break;
+	default:
+		return NULL;
+	}
+
+	if (sk) {
+		MT_DEBUG("qtaguid[%d]: %p->sk_proto=%u->sk_state=%d\n",
+			 par->hooknum, sk, sk->sk_protocol, sk->sk_state);
+		/*
+		 * When in TCP_TIME_WAIT the sk is not a "struct sock" but
+		 * "struct inet_timewait_sock" which is missing fields.
+		 */
+		if (!sk_fullsock(sk) || sk->sk_state  == TCP_TIME_WAIT) {
+			sock_gen_put(sk);
+			sk = NULL;
+		}
+	}
+	return sk;
+}
+
+static void account_for_uid(const struct sk_buff *skb,
+			    const struct sock *alternate_sk, uid_t uid,
+			    struct xt_action_param *par)
+{
+	const struct net_device *el_dev;
+	enum ifs_tx_rx direction;
+	int proto;
+
+	get_dev_and_dir(skb, par, &direction, &el_dev);
+	proto = ipx_proto(skb, par);
+	MT_DEBUG("qtaguid[%d]: dev name=%s type=%d fam=%d proto=%d dir=%d\n",
+		 par->hooknum, el_dev->name, el_dev->type,
+		 par->family, proto, direction);
+
+	if_tag_stat_update(el_dev->name, uid,
+			   skb->sk ? skb->sk : alternate_sk,
+			   direction,
+			   proto, skb->len);
+}
+
+static bool qtaguid_mt(const struct sk_buff *skb, struct xt_action_param *par)
+{
+	const struct xt_qtaguid_match_info *info = par->matchinfo;
+	const struct file *filp;
+	bool got_sock = false;
+	struct sock *sk;
+	kuid_t sock_uid;
+	bool res;
+	bool set_sk_callback_lock = false;
+	/*
+	 * TODO: unhack how to force just accounting.
+	 * For now we only do tag stats when the uid-owner is not requested
+	 */
+	bool do_tag_stat = !(info->match & XT_QTAGUID_UID);
+
+	if (unlikely(module_passive))
+		return (info->match ^ info->invert) == 0;
+
+	MT_DEBUG("qtaguid[%d]: entered skb=%p par->in=%p/out=%p fam=%d\n",
+		 par->hooknum, skb, par->in, par->out, par->family);
+
+	atomic64_inc(&qtu_events.match_calls);
+	if (skb == NULL) {
+		res = (info->match ^ info->invert) == 0;
+		goto ret_res;
+	}
+
+	switch (par->hooknum) {
+	case NF_INET_PRE_ROUTING:
+	case NF_INET_POST_ROUTING:
+		atomic64_inc(&qtu_events.match_calls_prepost);
+		iface_stat_update_from_skb(skb, par);
+		/*
+		 * We are done in pre/post. The skb will get processed
+		 * further alter.
+		 */
+		res = (info->match ^ info->invert);
+		goto ret_res;
+		break;
+	/* default: Fall through and do UID releated work */
+	}
+
+	sk = skb_to_full_sk(skb);
+	/*
+	 * When in TCP_TIME_WAIT the sk is not a "struct sock" but
+	 * "struct inet_timewait_sock" which is missing fields.
+	 * So we ignore it.
+	 */
+	if (sk && sk->sk_state == TCP_TIME_WAIT)
+		sk = NULL;
+	if (sk == NULL) {
+		/*
+		 * A missing sk->sk_socket happens when packets are in-flight
+		 * and the matching socket is already closed and gone.
+		 */
+		sk = qtaguid_find_sk(skb, par);
+		/*
+		 * If we got the socket from the find_sk(), we will need to put
+		 * it back, as nf_tproxy_get_sock_v4() got it.
+		 */
+		got_sock = sk;
+		if (sk)
+			atomic64_inc(&qtu_events.match_found_sk_in_ct);
+		else
+			atomic64_inc(&qtu_events.match_found_no_sk_in_ct);
+	} else {
+		atomic64_inc(&qtu_events.match_found_sk);
+	}
+	MT_DEBUG("qtaguid[%d]: sk=%p got_sock=%d fam=%d proto=%d\n",
+		 par->hooknum, sk, got_sock, par->family, ipx_proto(skb, par));
+
+	if (!sk) {
+		/*
+		 * Here, the qtaguid_find_sk() using connection tracking
+		 * couldn't find the owner, so for now we just count them
+		 * against the system.
+		 */
+		if (do_tag_stat)
+			account_for_uid(skb, sk, 0, par);
+		MT_DEBUG("qtaguid[%d]: leaving (sk=NULL)\n", par->hooknum);
+		res = (info->match ^ info->invert) == 0;
+		atomic64_inc(&qtu_events.match_no_sk);
+		goto put_sock_ret_res;
+	} else if (info->match & info->invert & XT_QTAGUID_SOCKET) {
+		res = false;
+		goto put_sock_ret_res;
+	}
+	sock_uid = sk->sk_uid;
+	if (do_tag_stat)
+		account_for_uid(skb, sk, from_kuid(&init_user_ns, sock_uid),
+				par);
+
+	/*
+	 * The following two tests fail the match when:
+	 *    id not in range AND no inverted condition requested
+	 * or id     in range AND    inverted condition requested
+	 * Thus (!a && b) || (a && !b) == a ^ b
+	 */
+	if (info->match & XT_QTAGUID_UID) {
+		kuid_t uid_min = make_kuid(&init_user_ns, info->uid_min);
+		kuid_t uid_max = make_kuid(&init_user_ns, info->uid_max);
+
+		if ((uid_gte(sock_uid, uid_min) &&
+		     uid_lte(sock_uid, uid_max)) ^
+		    !(info->invert & XT_QTAGUID_UID)) {
+			MT_DEBUG("qtaguid[%d]: leaving uid not matching\n",
+				 par->hooknum);
+			res = false;
+			goto put_sock_ret_res;
+		}
+	}
+	if (info->match & XT_QTAGUID_GID) {
+		kgid_t gid_min = make_kgid(&init_user_ns, info->gid_min);
+		kgid_t gid_max = make_kgid(&init_user_ns, info->gid_max);
+		set_sk_callback_lock = true;
+		read_lock_bh(&sk->sk_callback_lock);
+		MT_DEBUG("qtaguid[%d]: sk=%p->sk_socket=%p->file=%p\n",
+			 par->hooknum, sk, sk->sk_socket,
+			 sk->sk_socket ? sk->sk_socket->file : (void *)-1LL);
+		filp = sk->sk_socket ? sk->sk_socket->file : NULL;
+		if (!filp) {
+			res = ((info->match ^ info->invert) &
+			       XT_QTAGUID_GID) == 0;
+			atomic64_inc(&qtu_events.match_no_sk_gid);
+			goto put_sock_ret_res;
+		}
+		MT_DEBUG("qtaguid[%d]: filp...uid=%u\n",
+			 par->hooknum, filp ?
+			 from_kuid(&init_user_ns, filp->f_cred->fsuid) : -1);
+		if ((gid_gte(filp->f_cred->fsgid, gid_min) &&
+				gid_lte(filp->f_cred->fsgid, gid_max)) ^
+			!(info->invert & XT_QTAGUID_GID)) {
+			MT_DEBUG("qtaguid[%d]: leaving gid not matching\n",
+				par->hooknum);
+			res = false;
+			goto put_sock_ret_res;
+		}
+	}
+	MT_DEBUG("qtaguid[%d]: leaving matched\n", par->hooknum);
+	res = true;
+
+put_sock_ret_res:
+	if (got_sock)
+		sock_gen_put(sk);
+	if (set_sk_callback_lock)
+		read_unlock_bh(&sk->sk_callback_lock);
+ret_res:
+	MT_DEBUG("qtaguid[%d]: left %d\n", par->hooknum, res);
+	return res;
+}
+
+#ifdef DDEBUG
+/*
+ * This function is not in xt_qtaguid_print.c because of locks visibility.
+ * The lock of sock_tag_list must be aquired before calling this function
+ */
+static void prdebug_full_state_locked(int indent_level, const char *fmt, ...)
+{
+	va_list args;
+	char *fmt_buff;
+	char *buff;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	fmt_buff = kasprintf(GFP_ATOMIC,
+			     "qtaguid: %s(): %s {\n", __func__, fmt);
+	BUG_ON(!fmt_buff);
+	va_start(args, fmt);
+	buff = kvasprintf(GFP_ATOMIC,
+			  fmt_buff, args);
+	BUG_ON(!buff);
+	pr_debug("%s", buff);
+	kfree(fmt_buff);
+	kfree(buff);
+	va_end(args);
+
+	prdebug_sock_tag_tree(indent_level, &sock_tag_tree);
+
+	spin_lock_bh(&uid_tag_data_tree_lock);
+	prdebug_uid_tag_data_tree(indent_level, &uid_tag_data_tree);
+	prdebug_proc_qtu_data_tree(indent_level, &proc_qtu_data_tree);
+	spin_unlock_bh(&uid_tag_data_tree_lock);
+
+	spin_lock_bh(&iface_stat_list_lock);
+	prdebug_iface_stat_list(indent_level, &iface_stat_list);
+	spin_unlock_bh(&iface_stat_list_lock);
+
+	pr_debug("qtaguid: %s(): }\n", __func__);
+}
+#else
+static void prdebug_full_state_locked(int indent_level, const char *fmt, ...) {}
+#endif
+
+struct proc_ctrl_print_info {
+	struct sock *sk; /* socket found by reading to sk_pos */
+	loff_t sk_pos;
+};
+
+static void *qtaguid_ctrl_proc_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	struct proc_ctrl_print_info *pcpi = m->private;
+	struct sock_tag *sock_tag_entry = v;
+	struct rb_node *node;
+
+	(*pos)++;
+
+	if (!v || v  == SEQ_START_TOKEN)
+		return NULL;
+
+	node = rb_next(&sock_tag_entry->sock_node);
+	if (!node) {
+		pcpi->sk = NULL;
+		sock_tag_entry = SEQ_START_TOKEN;
+	} else {
+		sock_tag_entry = rb_entry(node, struct sock_tag, sock_node);
+		pcpi->sk = sock_tag_entry->sk;
+	}
+	pcpi->sk_pos = *pos;
+	return sock_tag_entry;
+}
+
+static void *qtaguid_ctrl_proc_start(struct seq_file *m, loff_t *pos)
+{
+	struct proc_ctrl_print_info *pcpi = m->private;
+	struct sock_tag *sock_tag_entry;
+	struct rb_node *node;
+
+	spin_lock_bh(&sock_tag_list_lock);
+
+	if (unlikely(module_passive))
+		return NULL;
+
+	if (*pos == 0) {
+		pcpi->sk_pos = 0;
+		node = rb_first(&sock_tag_tree);
+		if (!node) {
+			pcpi->sk = NULL;
+			return SEQ_START_TOKEN;
+		}
+		sock_tag_entry = rb_entry(node, struct sock_tag, sock_node);
+		pcpi->sk = sock_tag_entry->sk;
+	} else {
+		sock_tag_entry = (pcpi->sk ? get_sock_stat_nl(pcpi->sk) :
+						NULL) ?: SEQ_START_TOKEN;
+		if (*pos != pcpi->sk_pos) {
+			/* seq_read skipped a next call */
+			*pos = pcpi->sk_pos;
+			return qtaguid_ctrl_proc_next(m, sock_tag_entry, pos);
+		}
+	}
+	return sock_tag_entry;
+}
+
+static void qtaguid_ctrl_proc_stop(struct seq_file *m, void *v)
+{
+	spin_unlock_bh(&sock_tag_list_lock);
+}
+
+/*
+ * Procfs reader to get all active socket tags using style "1)" as described in
+ * fs/proc/generic.c
+ */
+static int qtaguid_ctrl_proc_show(struct seq_file *m, void *v)
+{
+	struct sock_tag *sock_tag_entry = v;
+	uid_t uid;
+
+	CT_DEBUG("qtaguid: proc ctrl pid=%u tgid=%u uid=%u\n",
+		 current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+
+	if (sock_tag_entry != SEQ_START_TOKEN) {
+		int sk_ref_count;
+		uid = get_uid_from_tag(sock_tag_entry->tag);
+		CT_DEBUG("qtaguid: proc_read(): sk=%p tag=0x%llx (uid=%u) "
+			 "pid=%u\n",
+			 sock_tag_entry->sk,
+			 sock_tag_entry->tag,
+			 uid,
+			 sock_tag_entry->pid
+			);
+		sk_ref_count = atomic_read(
+			&sock_tag_entry->sk->sk_refcnt);
+		seq_printf(m, "sock=%pK tag=0x%llx (uid=%u) pid=%u "
+			   "f_count=%d\n",
+			   sock_tag_entry->sk,
+			   sock_tag_entry->tag, uid,
+			   sock_tag_entry->pid, sk_ref_count);
+	} else {
+		seq_printf(m, "events: sockets_tagged=%llu "
+			   "sockets_untagged=%llu "
+			   "counter_set_changes=%llu "
+			   "delete_cmds=%llu "
+			   "iface_events=%llu "
+			   "match_calls=%llu "
+			   "match_calls_prepost=%llu "
+			   "match_found_sk=%llu "
+			   "match_found_sk_in_ct=%llu "
+			   "match_found_no_sk_in_ct=%llu "
+			   "match_no_sk=%llu "
+			   "match_no_sk_gid=%llu\n",
+			   (u64)atomic64_read(&qtu_events.sockets_tagged),
+			   (u64)atomic64_read(&qtu_events.sockets_untagged),
+			   (u64)atomic64_read(&qtu_events.counter_set_changes),
+			   (u64)atomic64_read(&qtu_events.delete_cmds),
+			   (u64)atomic64_read(&qtu_events.iface_events),
+			   (u64)atomic64_read(&qtu_events.match_calls),
+			   (u64)atomic64_read(&qtu_events.match_calls_prepost),
+			   (u64)atomic64_read(&qtu_events.match_found_sk),
+			   (u64)atomic64_read(&qtu_events.match_found_sk_in_ct),
+			   (u64)atomic64_read(&qtu_events.match_found_no_sk_in_ct),
+			   (u64)atomic64_read(&qtu_events.match_no_sk),
+			   (u64)atomic64_read(&qtu_events.match_no_sk_gid));
+
+		/* Count the following as part of the last item_index. No need
+		 * to lock the sock_tag_list here since it is already locked when
+		 * starting the seq_file operation
+		 */
+		prdebug_full_state_locked(0, "proc ctrl");
+	}
+
+	return 0;
+}
+
+/*
+ * Delete socket tags, and stat tags associated with a given
+ * accouting tag and uid.
+ */
+static int ctrl_cmd_delete(const char *input)
+{
+	char cmd;
+	int uid_int;
+	kuid_t uid;
+	uid_t entry_uid;
+	tag_t acct_tag;
+	tag_t tag;
+	int res, argc;
+	struct iface_stat *iface_entry;
+	struct rb_node *node;
+	struct sock_tag *st_entry;
+	struct rb_root st_to_free_tree = RB_ROOT;
+	struct tag_stat *ts_entry;
+	struct tag_counter_set *tcs_entry;
+	struct tag_ref *tr_entry;
+	struct uid_tag_data *utd_entry;
+
+	argc = sscanf(input, "%c %llu %u", &cmd, &acct_tag, &uid_int);
+	uid = make_kuid(&init_user_ns, uid_int);
+	CT_DEBUG("qtaguid: ctrl_delete(%s): argc=%d cmd=%c "
+		 "user_tag=0x%llx uid=%u\n", input, argc, cmd,
+		 acct_tag, uid_int);
+	if (argc < 2) {
+		res = -EINVAL;
+		goto err;
+	}
+	if (!valid_atag(acct_tag)) {
+		pr_info("qtaguid: ctrl_delete(%s): invalid tag\n", input);
+		res = -EINVAL;
+		goto err;
+	}
+	if (argc < 3) {
+		uid = current_fsuid();
+		uid_int = from_kuid(&init_user_ns, uid);
+	} else if (!can_impersonate_uid(uid)) {
+		pr_info("qtaguid: ctrl_delete(%s): "
+			"insufficient priv from pid=%u tgid=%u uid=%u\n",
+			input, current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+		res = -EPERM;
+		goto err;
+	}
+
+	tag = combine_atag_with_uid(acct_tag, uid_int);
+	CT_DEBUG("qtaguid: ctrl_delete(%s): "
+		 "looking for tag=0x%llx (uid=%u)\n",
+		 input, tag, uid_int);
+
+	/* Delete socket tags */
+	spin_lock_bh(&sock_tag_list_lock);
+	node = rb_first(&sock_tag_tree);
+	while (node) {
+		st_entry = rb_entry(node, struct sock_tag, sock_node);
+		entry_uid = get_uid_from_tag(st_entry->tag);
+		node = rb_next(node);
+		if (entry_uid != uid_int)
+			continue;
+
+		CT_DEBUG("qtaguid: ctrl_delete(%s): st tag=0x%llx (uid=%u)\n",
+			 input, st_entry->tag, entry_uid);
+
+		if (!acct_tag || st_entry->tag == tag) {
+			rb_erase(&st_entry->sock_node, &sock_tag_tree);
+			/* Can't sockfd_put() within spinlock, do it later. */
+			sock_tag_tree_insert(st_entry, &st_to_free_tree);
+			tr_entry = lookup_tag_ref(st_entry->tag, NULL);
+			BUG_ON(tr_entry->num_sock_tags <= 0);
+			tr_entry->num_sock_tags--;
+			/*
+			 * TODO: remove if, and start failing.
+			 * This is a hack to work around the fact that in some
+			 * places we have "if (IS_ERR_OR_NULL(pqd_entry))"
+			 * and are trying to work around apps
+			 * that didn't open the /dev/xt_qtaguid.
+			 */
+			if (st_entry->list.next && st_entry->list.prev)
+				list_del(&st_entry->list);
+		}
+	}
+	spin_unlock_bh(&sock_tag_list_lock);
+
+	sock_tag_tree_erase(&st_to_free_tree);
+
+	/* Delete tag counter-sets */
+	spin_lock_bh(&tag_counter_set_list_lock);
+	/* Counter sets are only on the uid tag, not full tag */
+	tcs_entry = tag_counter_set_tree_search(&tag_counter_set_tree, tag);
+	if (tcs_entry) {
+		CT_DEBUG("qtaguid: ctrl_delete(%s): "
+			 "erase tcs: tag=0x%llx (uid=%u) set=%d\n",
+			 input,
+			 tcs_entry->tn.tag,
+			 get_uid_from_tag(tcs_entry->tn.tag),
+			 tcs_entry->active_set);
+		rb_erase(&tcs_entry->tn.node, &tag_counter_set_tree);
+		kfree(tcs_entry);
+	}
+	spin_unlock_bh(&tag_counter_set_list_lock);
+
+	/*
+	 * If acct_tag is 0, then all entries belonging to uid are
+	 * erased.
+	 */
+	spin_lock_bh(&iface_stat_list_lock);
+	list_for_each_entry(iface_entry, &iface_stat_list, list) {
+		spin_lock_bh(&iface_entry->tag_stat_list_lock);
+		node = rb_first(&iface_entry->tag_stat_tree);
+		while (node) {
+			ts_entry = rb_entry(node, struct tag_stat, tn.node);
+			entry_uid = get_uid_from_tag(ts_entry->tn.tag);
+			node = rb_next(node);
+
+			CT_DEBUG("qtaguid: ctrl_delete(%s): "
+				 "ts tag=0x%llx (uid=%u)\n",
+				 input, ts_entry->tn.tag, entry_uid);
+
+			if (entry_uid != uid_int)
+				continue;
+			if (!acct_tag || ts_entry->tn.tag == tag) {
+				CT_DEBUG("qtaguid: ctrl_delete(%s): "
+					 "erase ts: %s 0x%llx %u\n",
+					 input, iface_entry->ifname,
+					 get_atag_from_tag(ts_entry->tn.tag),
+					 entry_uid);
+				rb_erase(&ts_entry->tn.node,
+					 &iface_entry->tag_stat_tree);
+				kfree(ts_entry);
+			}
+		}
+		spin_unlock_bh(&iface_entry->tag_stat_list_lock);
+	}
+	spin_unlock_bh(&iface_stat_list_lock);
+
+	/* Cleanup the uid_tag_data */
+	spin_lock_bh(&uid_tag_data_tree_lock);
+	node = rb_first(&uid_tag_data_tree);
+	while (node) {
+		utd_entry = rb_entry(node, struct uid_tag_data, node);
+		entry_uid = utd_entry->uid;
+		node = rb_next(node);
+
+		CT_DEBUG("qtaguid: ctrl_delete(%s): "
+			 "utd uid=%u\n",
+			 input, entry_uid);
+
+		if (entry_uid != uid_int)
+			continue;
+		/*
+		 * Go over the tag_refs, and those that don't have
+		 * sock_tags using them are freed.
+		 */
+		put_tag_ref_tree(tag, utd_entry);
+		put_utd_entry(utd_entry);
+	}
+	spin_unlock_bh(&uid_tag_data_tree_lock);
+
+	atomic64_inc(&qtu_events.delete_cmds);
+	res = 0;
+
+err:
+	return res;
+}
+
+static int ctrl_cmd_counter_set(const char *input)
+{
+	char cmd;
+	uid_t uid = 0;
+	tag_t tag;
+	int res, argc;
+	struct tag_counter_set *tcs;
+	int counter_set;
+
+	argc = sscanf(input, "%c %d %u", &cmd, &counter_set, &uid);
+	CT_DEBUG("qtaguid: ctrl_counterset(%s): argc=%d cmd=%c "
+		 "set=%d uid=%u\n", input, argc, cmd,
+		 counter_set, uid);
+	if (argc != 3) {
+		res = -EINVAL;
+		goto err;
+	}
+	if (counter_set < 0 || counter_set >= IFS_MAX_COUNTER_SETS) {
+		pr_info("qtaguid: ctrl_counterset(%s): invalid counter_set range\n",
+			input);
+		res = -EINVAL;
+		goto err;
+	}
+	if (!can_manipulate_uids()) {
+		pr_info("qtaguid: ctrl_counterset(%s): "
+			"insufficient priv from pid=%u tgid=%u uid=%u\n",
+			input, current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+		res = -EPERM;
+		goto err;
+	}
+
+	tag = make_tag_from_uid(uid);
+	spin_lock_bh(&tag_counter_set_list_lock);
+	tcs = tag_counter_set_tree_search(&tag_counter_set_tree, tag);
+	if (!tcs) {
+		tcs = kzalloc(sizeof(*tcs), GFP_ATOMIC);
+		if (!tcs) {
+			spin_unlock_bh(&tag_counter_set_list_lock);
+			pr_err("qtaguid: ctrl_counterset(%s): "
+			       "failed to alloc counter set\n",
+			       input);
+			res = -ENOMEM;
+			goto err;
+		}
+		tcs->tn.tag = tag;
+		tag_counter_set_tree_insert(tcs, &tag_counter_set_tree);
+		CT_DEBUG("qtaguid: ctrl_counterset(%s): added tcs tag=0x%llx "
+			 "(uid=%u) set=%d\n",
+			 input, tag, get_uid_from_tag(tag), counter_set);
+	}
+	tcs->active_set = counter_set;
+	spin_unlock_bh(&tag_counter_set_list_lock);
+	atomic64_inc(&qtu_events.counter_set_changes);
+	res = 0;
+
+err:
+	return res;
+}
+
+static int ctrl_cmd_tag(const char *input)
+{
+	char cmd;
+	int sock_fd = 0;
+	kuid_t uid;
+	unsigned int uid_int = 0;
+	tag_t acct_tag = make_atag_from_value(0);
+	tag_t full_tag;
+	struct socket *el_socket;
+	int res, argc;
+	struct sock_tag *sock_tag_entry;
+	struct tag_ref *tag_ref_entry;
+	struct uid_tag_data *uid_tag_data_entry;
+	struct proc_qtu_data *pqd_entry;
+
+	/* Unassigned args will get defaulted later. */
+	argc = sscanf(input, "%c %d %llu %u", &cmd, &sock_fd, &acct_tag, &uid_int);
+	uid = make_kuid(&init_user_ns, uid_int);
+	CT_DEBUG("qtaguid: ctrl_tag(%s): argc=%d cmd=%c sock_fd=%d "
+		 "acct_tag=0x%llx uid=%u\n", input, argc, cmd, sock_fd,
+		 acct_tag, uid_int);
+	if (argc < 2) {
+		res = -EINVAL;
+		goto err;
+	}
+	el_socket = sockfd_lookup(sock_fd, &res);  /* This locks the file */
+	if (!el_socket) {
+		pr_info("qtaguid: ctrl_tag(%s): failed to lookup"
+			" sock_fd=%d err=%d pid=%u tgid=%u uid=%u\n",
+			input, sock_fd, res, current->pid, current->tgid,
+			from_kuid(&init_user_ns, current_fsuid()));
+		goto err;
+	}
+	CT_DEBUG("qtaguid: ctrl_tag(%s): socket->...->sk_refcnt=%d ->sk=%p\n",
+		 input, atomic_read(&el_socket->sk->sk_refcnt),
+		 el_socket->sk);
+	if (argc < 3) {
+		acct_tag = make_atag_from_value(0);
+	} else if (!valid_atag(acct_tag)) {
+		pr_info("qtaguid: ctrl_tag(%s): invalid tag\n", input);
+		res = -EINVAL;
+		goto err_put;
+	}
+	CT_DEBUG("qtaguid: ctrl_tag(%s): "
+		 "pid=%u tgid=%u uid=%u euid=%u fsuid=%u "
+		 "ctrl.gid=%u in_group()=%d in_egroup()=%d\n",
+		 input, current->pid, current->tgid,
+		 from_kuid(&init_user_ns, current_uid()),
+		 from_kuid(&init_user_ns, current_euid()),
+		 from_kuid(&init_user_ns, current_fsuid()),
+		 from_kgid(&init_user_ns, xt_qtaguid_ctrl_file->gid),
+		 in_group_p(xt_qtaguid_ctrl_file->gid),
+		 in_egroup_p(xt_qtaguid_ctrl_file->gid));
+	if (argc < 4) {
+		uid = current_fsuid();
+		uid_int = from_kuid(&init_user_ns, uid);
+	} else if (!can_impersonate_uid(uid)) {
+		pr_info("qtaguid: ctrl_tag(%s): "
+			"insufficient priv from pid=%u tgid=%u uid=%u\n",
+			input, current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+		res = -EPERM;
+		goto err_put;
+	}
+	full_tag = combine_atag_with_uid(acct_tag, uid_int);
+
+	spin_lock_bh(&sock_tag_list_lock);
+	sock_tag_entry = get_sock_stat_nl(el_socket->sk);
+	tag_ref_entry = get_tag_ref(full_tag, &uid_tag_data_entry);
+	if (IS_ERR(tag_ref_entry)) {
+		res = PTR_ERR(tag_ref_entry);
+		spin_unlock_bh(&sock_tag_list_lock);
+		goto err_put;
+	}
+	tag_ref_entry->num_sock_tags++;
+	if (sock_tag_entry) {
+		struct tag_ref *prev_tag_ref_entry;
+
+		CT_DEBUG("qtaguid: ctrl_tag(%s): retag for sk=%p "
+			 "st@%p ...->sk_refcnt=%d\n",
+			 input, el_socket->sk, sock_tag_entry,
+			 atomic_read(&el_socket->sk->sk_refcnt));
+		prev_tag_ref_entry = lookup_tag_ref(sock_tag_entry->tag,
+						    &uid_tag_data_entry);
+		BUG_ON(IS_ERR_OR_NULL(prev_tag_ref_entry));
+		BUG_ON(prev_tag_ref_entry->num_sock_tags <= 0);
+		prev_tag_ref_entry->num_sock_tags--;
+		sock_tag_entry->tag = full_tag;
+	} else {
+		CT_DEBUG("qtaguid: ctrl_tag(%s): newtag for sk=%p\n",
+			 input, el_socket->sk);
+		sock_tag_entry = kzalloc(sizeof(*sock_tag_entry),
+					 GFP_ATOMIC);
+		if (!sock_tag_entry) {
+			pr_err("qtaguid: ctrl_tag(%s): "
+			       "socket tag alloc failed\n",
+			       input);
+			spin_unlock_bh(&sock_tag_list_lock);
+			res = -ENOMEM;
+			goto err_tag_unref_put;
+		}
+		/*
+		 * Hold the sk refcount here to make sure the sk pointer cannot
+		 * be freed and reused
+		 */
+		sock_hold(el_socket->sk);
+		sock_tag_entry->sk = el_socket->sk;
+		sock_tag_entry->pid = current->tgid;
+		sock_tag_entry->tag = combine_atag_with_uid(acct_tag, uid_int);
+		spin_lock_bh(&uid_tag_data_tree_lock);
+		pqd_entry = proc_qtu_data_tree_search(
+			&proc_qtu_data_tree, current->tgid);
+		/*
+		 * TODO: remove if, and start failing.
+		 * At first, we want to catch user-space code that is not
+		 * opening the /dev/xt_qtaguid.
+		 */
+		if (IS_ERR_OR_NULL(pqd_entry))
+			pr_warn_once(
+				"qtaguid: %s(): "
+				"User space forgot to open /dev/xt_qtaguid? "
+				"pid=%u tgid=%u uid=%u\n", __func__,
+				current->pid, current->tgid,
+				from_kuid(&init_user_ns, current_fsuid()));
+		else
+			list_add(&sock_tag_entry->list,
+				 &pqd_entry->sock_tag_list);
+		spin_unlock_bh(&uid_tag_data_tree_lock);
+
+		sock_tag_tree_insert(sock_tag_entry, &sock_tag_tree);
+		atomic64_inc(&qtu_events.sockets_tagged);
+	}
+	spin_unlock_bh(&sock_tag_list_lock);
+	/* We keep the ref to the sk until it is untagged */
+	CT_DEBUG("qtaguid: ctrl_tag(%s): done st@%p ...->sk_refcnt=%d\n",
+		 input, sock_tag_entry,
+		 atomic_read(&el_socket->sk->sk_refcnt));
+	sockfd_put(el_socket);
+	return 0;
+
+err_tag_unref_put:
+	BUG_ON(tag_ref_entry->num_sock_tags <= 0);
+	tag_ref_entry->num_sock_tags--;
+	free_tag_ref_from_utd_entry(tag_ref_entry, uid_tag_data_entry);
+err_put:
+	CT_DEBUG("qtaguid: ctrl_tag(%s): done. ...->sk_refcnt=%d\n",
+		 input, atomic_read(&el_socket->sk->sk_refcnt) - 1);
+	/* Release the sock_fd that was grabbed by sockfd_lookup(). */
+	sockfd_put(el_socket);
+	return res;
+
+err:
+	CT_DEBUG("qtaguid: ctrl_tag(%s): done.\n", input);
+	return res;
+}
+
+static int ctrl_cmd_untag(const char *input)
+{
+	char cmd;
+	int sock_fd = 0;
+	struct socket *el_socket;
+	int res, argc;
+
+	argc = sscanf(input, "%c %d", &cmd, &sock_fd);
+	CT_DEBUG("qtaguid: ctrl_untag(%s): argc=%d cmd=%c sock_fd=%d\n",
+		 input, argc, cmd, sock_fd);
+	if (argc < 2) {
+		res = -EINVAL;
+		return res;
+	}
+	el_socket = sockfd_lookup(sock_fd, &res);  /* This locks the file */
+	if (!el_socket) {
+		pr_info("qtaguid: ctrl_untag(%s): failed to lookup"
+			" sock_fd=%d err=%d pid=%u tgid=%u uid=%u\n",
+			input, sock_fd, res, current->pid, current->tgid,
+			from_kuid(&init_user_ns, current_fsuid()));
+		return res;
+	}
+	CT_DEBUG("qtaguid: ctrl_untag(%s): socket->...->f_count=%ld ->sk=%p\n",
+		 input, atomic_long_read(&el_socket->file->f_count),
+		 el_socket->sk);
+	res = qtaguid_untag(el_socket, false);
+	sockfd_put(el_socket);
+	return res;
+}
+
+int qtaguid_untag(struct socket *el_socket, bool kernel)
+{
+	int res;
+	pid_t pid;
+	struct sock_tag *sock_tag_entry;
+	struct tag_ref *tag_ref_entry;
+	struct uid_tag_data *utd_entry;
+	struct proc_qtu_data *pqd_entry;
+
+	spin_lock_bh(&sock_tag_list_lock);
+	sock_tag_entry = get_sock_stat_nl(el_socket->sk);
+	if (!sock_tag_entry) {
+		spin_unlock_bh(&sock_tag_list_lock);
+		res = -EINVAL;
+		return res;
+	}
+	/*
+	 * The socket already belongs to the current process
+	 * so it can do whatever it wants to it.
+	 */
+	rb_erase(&sock_tag_entry->sock_node, &sock_tag_tree);
+
+	tag_ref_entry = lookup_tag_ref(sock_tag_entry->tag, &utd_entry);
+	BUG_ON(!tag_ref_entry);
+	BUG_ON(tag_ref_entry->num_sock_tags <= 0);
+	spin_lock_bh(&uid_tag_data_tree_lock);
+	if (kernel)
+		pid = sock_tag_entry->pid;
+	else
+		pid = current->tgid;
+	pqd_entry = proc_qtu_data_tree_search(
+		&proc_qtu_data_tree, pid);
+	/*
+	 * TODO: remove if, and start failing.
+	 * At first, we want to catch user-space code that is not
+	 * opening the /dev/xt_qtaguid.
+	 */
+	if (IS_ERR_OR_NULL(pqd_entry) || !sock_tag_entry->list.next) {
+		pr_warn_once("qtaguid: %s(): "
+			     "User space forgot to open /dev/xt_qtaguid? "
+			     "pid=%u tgid=%u sk_pid=%u, uid=%u\n", __func__,
+			     current->pid, current->tgid, sock_tag_entry->pid,
+			     from_kuid(&init_user_ns, current_fsuid()));
+	} else {
+		list_del(&sock_tag_entry->list);
+	}
+	spin_unlock_bh(&uid_tag_data_tree_lock);
+	/*
+	 * We don't free tag_ref from the utd_entry here,
+	 * only during a cmd_delete().
+	 */
+	tag_ref_entry->num_sock_tags--;
+	spin_unlock_bh(&sock_tag_list_lock);
+	/*
+	 * Release the sock_fd that was grabbed at tag time.
+	 */
+	sock_put(sock_tag_entry->sk);
+	CT_DEBUG("qtaguid: done. st@%p ...->sk_refcnt=%d\n",
+		 sock_tag_entry,
+		 atomic_read(&el_socket->sk->sk_refcnt));
+
+	kfree(sock_tag_entry);
+	atomic64_inc(&qtu_events.sockets_untagged);
+
+	return 0;
+}
+
+static ssize_t qtaguid_ctrl_parse(const char *input, size_t count)
+{
+	char cmd;
+	ssize_t res;
+
+	CT_DEBUG("qtaguid: ctrl(%s): pid=%u tgid=%u uid=%u\n",
+		 input, current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+
+	cmd = input[0];
+	/* Collect params for commands */
+	switch (cmd) {
+	case 'd':
+		res = ctrl_cmd_delete(input);
+		break;
+
+	case 's':
+		res = ctrl_cmd_counter_set(input);
+		break;
+
+	case 't':
+		res = ctrl_cmd_tag(input);
+		break;
+
+	case 'u':
+		res = ctrl_cmd_untag(input);
+		break;
+
+	default:
+		res = -EINVAL;
+		goto err;
+	}
+	if (!res)
+		res = count;
+err:
+	CT_DEBUG("qtaguid: ctrl(%s): res=%zd\n", input, res);
+	return res;
+}
+
+#define MAX_QTAGUID_CTRL_INPUT_LEN 255
+static ssize_t qtaguid_ctrl_proc_write(struct file *file, const char __user *buffer,
+				   size_t count, loff_t *offp)
+{
+	char input_buf[MAX_QTAGUID_CTRL_INPUT_LEN];
+
+	if (unlikely(module_passive))
+		return count;
+
+	if (count >= MAX_QTAGUID_CTRL_INPUT_LEN)
+		return -EINVAL;
+
+	if (copy_from_user(input_buf, buffer, count))
+		return -EFAULT;
+
+	input_buf[count] = '\0';
+	return qtaguid_ctrl_parse(input_buf, count);
+}
+
+struct proc_print_info {
+	struct iface_stat *iface_entry;
+	int item_index;
+	tag_t tag; /* tag found by reading to tag_pos */
+	off_t tag_pos;
+	int tag_item_index;
+};
+
+static void pp_stats_header(struct seq_file *m)
+{
+	seq_puts(m,
+		 "idx iface acct_tag_hex uid_tag_int cnt_set "
+		 "rx_bytes rx_packets "
+		 "tx_bytes tx_packets "
+		 "rx_tcp_bytes rx_tcp_packets "
+		 "rx_udp_bytes rx_udp_packets "
+		 "rx_other_bytes rx_other_packets "
+		 "tx_tcp_bytes tx_tcp_packets "
+		 "tx_udp_bytes tx_udp_packets "
+		 "tx_other_bytes tx_other_packets\n");
+}
+
+static int pp_stats_line(struct seq_file *m, struct tag_stat *ts_entry,
+			 int cnt_set)
+{
+	struct data_counters *cnts;
+	tag_t tag = ts_entry->tn.tag;
+	uid_t stat_uid = get_uid_from_tag(tag);
+	struct proc_print_info *ppi = m->private;
+	/* Detailed tags are not available to everybody */
+	if (!can_read_other_uid_stats(make_kuid(&init_user_ns,stat_uid))) {
+		CT_DEBUG("qtaguid: stats line: "
+			 "%s 0x%llx %u: insufficient priv "
+			 "from pid=%u tgid=%u uid=%u stats.gid=%u\n",
+			 ppi->iface_entry->ifname,
+			 get_atag_from_tag(tag), stat_uid,
+			 current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()),
+			 from_kgid(&init_user_ns,xt_qtaguid_stats_file->gid));
+		return 0;
+	}
+	ppi->item_index++;
+	cnts = &ts_entry->counters;
+	seq_printf(m, "%d %s 0x%llx %u %u "
+		"%llu %llu "
+		"%llu %llu "
+		"%llu %llu "
+		"%llu %llu "
+		"%llu %llu "
+		"%llu %llu "
+		"%llu %llu "
+		"%llu %llu\n",
+		ppi->item_index,
+		ppi->iface_entry->ifname,
+		get_atag_from_tag(tag),
+		stat_uid,
+		cnt_set,
+		dc_sum_bytes(cnts, cnt_set, IFS_RX),
+		dc_sum_packets(cnts, cnt_set, IFS_RX),
+		dc_sum_bytes(cnts, cnt_set, IFS_TX),
+		dc_sum_packets(cnts, cnt_set, IFS_TX),
+		cnts->bpc[cnt_set][IFS_RX][IFS_TCP].bytes,
+		cnts->bpc[cnt_set][IFS_RX][IFS_TCP].packets,
+		cnts->bpc[cnt_set][IFS_RX][IFS_UDP].bytes,
+		cnts->bpc[cnt_set][IFS_RX][IFS_UDP].packets,
+		cnts->bpc[cnt_set][IFS_RX][IFS_PROTO_OTHER].bytes,
+		cnts->bpc[cnt_set][IFS_RX][IFS_PROTO_OTHER].packets,
+		cnts->bpc[cnt_set][IFS_TX][IFS_TCP].bytes,
+		cnts->bpc[cnt_set][IFS_TX][IFS_TCP].packets,
+		cnts->bpc[cnt_set][IFS_TX][IFS_UDP].bytes,
+		cnts->bpc[cnt_set][IFS_TX][IFS_UDP].packets,
+		cnts->bpc[cnt_set][IFS_TX][IFS_PROTO_OTHER].bytes,
+		cnts->bpc[cnt_set][IFS_TX][IFS_PROTO_OTHER].packets);
+	return seq_has_overflowed(m) ? -ENOSPC : 1;
+}
+
+static bool pp_sets(struct seq_file *m, struct tag_stat *ts_entry)
+{
+	int ret;
+	int counter_set;
+	for (counter_set = 0; counter_set < IFS_MAX_COUNTER_SETS;
+	     counter_set++) {
+		ret = pp_stats_line(m, ts_entry, counter_set);
+		if (ret < 0)
+			return false;
+	}
+	return true;
+}
+
+static int qtaguid_stats_proc_iface_stat_ptr_valid(struct iface_stat *ptr)
+{
+	struct iface_stat *iface_entry;
+
+	if (!ptr)
+		return false;
+
+	list_for_each_entry(iface_entry, &iface_stat_list, list)
+		if (iface_entry == ptr)
+			return true;
+	return false;
+}
+
+static void qtaguid_stats_proc_next_iface_entry(struct proc_print_info *ppi)
+{
+	spin_unlock_bh(&ppi->iface_entry->tag_stat_list_lock);
+	list_for_each_entry_continue(ppi->iface_entry, &iface_stat_list, list) {
+		spin_lock_bh(&ppi->iface_entry->tag_stat_list_lock);
+		return;
+	}
+	ppi->iface_entry = NULL;
+}
+
+static void *qtaguid_stats_proc_next(struct seq_file *m, void *v, loff_t *pos)
+{
+	struct proc_print_info *ppi = m->private;
+	struct tag_stat *ts_entry;
+	struct rb_node *node;
+
+	if (!v) {
+		pr_err("qtaguid: %s(): unexpected v: NULL\n", __func__);
+		return NULL;
+	}
+
+	(*pos)++;
+
+	if (!ppi->iface_entry || unlikely(module_passive))
+		return NULL;
+
+	if (v == SEQ_START_TOKEN)
+		node = rb_first(&ppi->iface_entry->tag_stat_tree);
+	else
+		node = rb_next(&((struct tag_stat *)v)->tn.node);
+
+	while (!node) {
+		qtaguid_stats_proc_next_iface_entry(ppi);
+		if (!ppi->iface_entry)
+			return NULL;
+		node = rb_first(&ppi->iface_entry->tag_stat_tree);
+	}
+
+	ts_entry = rb_entry(node, struct tag_stat, tn.node);
+	ppi->tag = ts_entry->tn.tag;
+	ppi->tag_pos = *pos;
+	ppi->tag_item_index = ppi->item_index;
+	return ts_entry;
+}
+
+static void *qtaguid_stats_proc_start(struct seq_file *m, loff_t *pos)
+{
+	struct proc_print_info *ppi = m->private;
+	struct tag_stat *ts_entry = NULL;
+
+	spin_lock_bh(&iface_stat_list_lock);
+
+	if (*pos == 0) {
+		ppi->item_index = 1;
+		ppi->tag_pos = 0;
+		if (list_empty(&iface_stat_list)) {
+			ppi->iface_entry = NULL;
+		} else {
+			ppi->iface_entry = list_first_entry(&iface_stat_list,
+							    struct iface_stat,
+							    list);
+			spin_lock_bh(&ppi->iface_entry->tag_stat_list_lock);
+		}
+		return SEQ_START_TOKEN;
+	}
+	if (!qtaguid_stats_proc_iface_stat_ptr_valid(ppi->iface_entry)) {
+		if (ppi->iface_entry) {
+			pr_err("qtaguid: %s(): iface_entry %p not found\n",
+			       __func__, ppi->iface_entry);
+			ppi->iface_entry = NULL;
+		}
+		return NULL;
+	}
+
+	spin_lock_bh(&ppi->iface_entry->tag_stat_list_lock);
+
+	if (!ppi->tag_pos) {
+		/* seq_read skipped first next call */
+		ts_entry = SEQ_START_TOKEN;
+	} else {
+		ts_entry = tag_stat_tree_search(
+				&ppi->iface_entry->tag_stat_tree, ppi->tag);
+		if (!ts_entry) {
+			pr_info("qtaguid: %s(): tag_stat.tag 0x%llx not found. Abort.\n",
+				__func__, ppi->tag);
+			return NULL;
+		}
+	}
+
+	if (*pos == ppi->tag_pos) { /* normal resume */
+		ppi->item_index = ppi->tag_item_index;
+	} else {
+		/* seq_read skipped a next call */
+		*pos = ppi->tag_pos;
+		ts_entry = qtaguid_stats_proc_next(m, ts_entry, pos);
+	}
+
+	return ts_entry;
+}
+
+static void qtaguid_stats_proc_stop(struct seq_file *m, void *v)
+{
+	struct proc_print_info *ppi = m->private;
+	if (ppi->iface_entry)
+		spin_unlock_bh(&ppi->iface_entry->tag_stat_list_lock);
+	spin_unlock_bh(&iface_stat_list_lock);
+}
+
+/*
+ * Procfs reader to get all tag stats using style "1)" as described in
+ * fs/proc/generic.c
+ * Groups all protocols tx/rx bytes.
+ */
+static int qtaguid_stats_proc_show(struct seq_file *m, void *v)
+{
+	struct tag_stat *ts_entry = v;
+
+	if (v == SEQ_START_TOKEN)
+		pp_stats_header(m);
+	else
+		pp_sets(m, ts_entry);
+
+	return 0;
+}
+
+/*------------------------------------------*/
+static int qtudev_open(struct inode *inode, struct file *file)
+{
+	struct uid_tag_data *utd_entry;
+	struct proc_qtu_data  *pqd_entry;
+	struct proc_qtu_data  *new_pqd_entry;
+	int res;
+	bool utd_entry_found;
+
+	if (unlikely(qtu_proc_handling_passive))
+		return 0;
+
+	DR_DEBUG("qtaguid: qtudev_open(): pid=%u tgid=%u uid=%u\n",
+		 current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+
+	spin_lock_bh(&uid_tag_data_tree_lock);
+
+	/* Look for existing uid data, or alloc one. */
+	utd_entry = get_uid_data(from_kuid(&init_user_ns, current_fsuid()), &utd_entry_found);
+	if (IS_ERR_OR_NULL(utd_entry)) {
+		res = PTR_ERR(utd_entry);
+		goto err_unlock;
+	}
+
+	/* Look for existing PID based proc_data */
+	pqd_entry = proc_qtu_data_tree_search(&proc_qtu_data_tree,
+					      current->tgid);
+	if (pqd_entry) {
+		pr_err("qtaguid: qtudev_open(): %u/%u %u "
+		       "%s already opened\n",
+		       current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()),
+		       QTU_DEV_NAME);
+		res = -EBUSY;
+		goto err_unlock_free_utd;
+	}
+
+	new_pqd_entry = kzalloc(sizeof(*new_pqd_entry), GFP_ATOMIC);
+	if (!new_pqd_entry) {
+		pr_err("qtaguid: qtudev_open(): %u/%u %u: "
+		       "proc data alloc failed\n",
+		       current->pid, current->tgid, from_kuid(&init_user_ns, current_fsuid()));
+		res = -ENOMEM;
+		goto err_unlock_free_utd;
+	}
+	new_pqd_entry->pid = current->tgid;
+	INIT_LIST_HEAD(&new_pqd_entry->sock_tag_list);
+	new_pqd_entry->parent_tag_data = utd_entry;
+	utd_entry->num_pqd++;
+
+	proc_qtu_data_tree_insert(new_pqd_entry,
+				  &proc_qtu_data_tree);
+
+	spin_unlock_bh(&uid_tag_data_tree_lock);
+	DR_DEBUG("qtaguid: tracking data for uid=%u in pqd=%p\n",
+		 from_kuid(&init_user_ns, current_fsuid()), new_pqd_entry);
+	file->private_data = new_pqd_entry;
+	return 0;
+
+err_unlock_free_utd:
+	if (!utd_entry_found) {
+		rb_erase(&utd_entry->node, &uid_tag_data_tree);
+		kfree(utd_entry);
+	}
+err_unlock:
+	spin_unlock_bh(&uid_tag_data_tree_lock);
+	return res;
+}
+
+static int qtudev_release(struct inode *inode, struct file *file)
+{
+	struct proc_qtu_data  *pqd_entry = file->private_data;
+	struct uid_tag_data  *utd_entry = pqd_entry->parent_tag_data;
+	struct sock_tag *st_entry;
+	struct rb_root st_to_free_tree = RB_ROOT;
+	struct list_head *entry, *next;
+	struct tag_ref *tr;
+
+	if (unlikely(qtu_proc_handling_passive))
+		return 0;
+
+	/*
+	 * Do not trust the current->pid, it might just be a kworker cleaning
+	 * up after a dead proc.
+	 */
+	DR_DEBUG("qtaguid: qtudev_release(): "
+		 "pid=%u tgid=%u uid=%u "
+		 "pqd_entry=%p->pid=%u utd_entry=%p->active_tags=%d\n",
+		 current->pid, current->tgid, pqd_entry->parent_tag_data->uid,
+		 pqd_entry, pqd_entry->pid, utd_entry,
+		 utd_entry->num_active_tags);
+
+	spin_lock_bh(&sock_tag_list_lock);
+	spin_lock_bh(&uid_tag_data_tree_lock);
+
+	list_for_each_safe(entry, next, &pqd_entry->sock_tag_list) {
+		st_entry = list_entry(entry, struct sock_tag, list);
+		DR_DEBUG("qtaguid: %s(): "
+			 "erase sock_tag=%p->sk=%p pid=%u tgid=%u uid=%u\n",
+			 __func__,
+			 st_entry, st_entry->sk,
+			 current->pid, current->tgid,
+			 pqd_entry->parent_tag_data->uid);
+
+		utd_entry = uid_tag_data_tree_search(
+			&uid_tag_data_tree,
+			get_uid_from_tag(st_entry->tag));
+		BUG_ON(IS_ERR_OR_NULL(utd_entry));
+		DR_DEBUG("qtaguid: %s(): "
+			 "looking for tag=0x%llx in utd_entry=%p\n", __func__,
+			 st_entry->tag, utd_entry);
+		tr = tag_ref_tree_search(&utd_entry->tag_ref_tree,
+					 st_entry->tag);
+		BUG_ON(!tr);
+		BUG_ON(tr->num_sock_tags <= 0);
+		tr->num_sock_tags--;
+		free_tag_ref_from_utd_entry(tr, utd_entry);
+
+		rb_erase(&st_entry->sock_node, &sock_tag_tree);
+		list_del(&st_entry->list);
+		/* Can't sockfd_put() within spinlock, do it later. */
+		sock_tag_tree_insert(st_entry, &st_to_free_tree);
+
+		/*
+		 * Try to free the utd_entry if no other proc_qtu_data is
+		 * using it (num_pqd is 0) and it doesn't have active tags
+		 * (num_active_tags is 0).
+		 */
+		put_utd_entry(utd_entry);
+	}
+
+	rb_erase(&pqd_entry->node, &proc_qtu_data_tree);
+	BUG_ON(pqd_entry->parent_tag_data->num_pqd < 1);
+	pqd_entry->parent_tag_data->num_pqd--;
+	put_utd_entry(pqd_entry->parent_tag_data);
+	kfree(pqd_entry);
+	file->private_data = NULL;
+
+	spin_unlock_bh(&uid_tag_data_tree_lock);
+	spin_unlock_bh(&sock_tag_list_lock);
+
+
+	sock_tag_tree_erase(&st_to_free_tree);
+
+	spin_lock_bh(&sock_tag_list_lock);
+	prdebug_full_state_locked(0, "%s(): pid=%u tgid=%u", __func__,
+			   current->pid, current->tgid);
+	spin_unlock_bh(&sock_tag_list_lock);
+	return 0;
+}
+
+/*------------------------------------------*/
+static const struct file_operations qtudev_fops = {
+	.owner = THIS_MODULE,
+	.open = qtudev_open,
+	.release = qtudev_release,
+};
+
+static struct miscdevice qtu_device = {
+	.minor = MISC_DYNAMIC_MINOR,
+	.name = QTU_DEV_NAME,
+	.fops = &qtudev_fops,
+	/* How sad it doesn't allow for defaults: .mode = S_IRUGO | S_IWUSR */
+};
+
+static const struct seq_operations proc_qtaguid_ctrl_seqops = {
+	.start = qtaguid_ctrl_proc_start,
+	.next = qtaguid_ctrl_proc_next,
+	.stop = qtaguid_ctrl_proc_stop,
+	.show = qtaguid_ctrl_proc_show,
+};
+
+static int proc_qtaguid_ctrl_open(struct inode *inode, struct file *file)
+{
+	return seq_open_private(file, &proc_qtaguid_ctrl_seqops,
+				sizeof(struct proc_ctrl_print_info));
+}
+
+static const struct file_operations proc_qtaguid_ctrl_fops = {
+	.open		= proc_qtaguid_ctrl_open,
+	.read		= seq_read,
+	.write		= qtaguid_ctrl_proc_write,
+	.llseek		= seq_lseek,
+	.release	= seq_release_private,
+};
+
+static const struct seq_operations proc_qtaguid_stats_seqops = {
+	.start = qtaguid_stats_proc_start,
+	.next = qtaguid_stats_proc_next,
+	.stop = qtaguid_stats_proc_stop,
+	.show = qtaguid_stats_proc_show,
+};
+
+static int proc_qtaguid_stats_open(struct inode *inode, struct file *file)
+{
+	return seq_open_private(file, &proc_qtaguid_stats_seqops,
+				sizeof(struct proc_print_info));
+}
+
+static const struct file_operations proc_qtaguid_stats_fops = {
+	.open		= proc_qtaguid_stats_open,
+	.read		= seq_read,
+	.llseek		= seq_lseek,
+	.release	= seq_release_private,
+};
+
+/*------------------------------------------*/
+static int __init qtaguid_proc_register(struct proc_dir_entry **res_procdir)
+{
+	int ret;
+	*res_procdir = proc_mkdir(module_procdirname, init_net.proc_net);
+	if (!*res_procdir) {
+		pr_err("qtaguid: failed to create proc/.../xt_qtaguid\n");
+		ret = -ENOMEM;
+		goto no_dir;
+	}
+
+	xt_qtaguid_ctrl_file = proc_create_data("ctrl", proc_ctrl_perms,
+						*res_procdir,
+						&proc_qtaguid_ctrl_fops,
+						NULL);
+	if (!xt_qtaguid_ctrl_file) {
+		pr_err("qtaguid: failed to create xt_qtaguid/ctrl "
+			" file\n");
+		ret = -ENOMEM;
+		goto no_ctrl_entry;
+	}
+
+	xt_qtaguid_stats_file = proc_create_data("stats", proc_stats_perms,
+						 *res_procdir,
+						 &proc_qtaguid_stats_fops,
+						 NULL);
+	if (!xt_qtaguid_stats_file) {
+		pr_err("qtaguid: failed to create xt_qtaguid/stats "
+			"file\n");
+		ret = -ENOMEM;
+		goto no_stats_entry;
+	}
+	/*
+	 * TODO: add support counter hacking
+	 * xt_qtaguid_stats_file->write_proc = qtaguid_stats_proc_write;
+	 */
+	return 0;
+
+no_stats_entry:
+	remove_proc_entry("ctrl", *res_procdir);
+no_ctrl_entry:
+	remove_proc_entry("xt_qtaguid", NULL);
+no_dir:
+	return ret;
+}
+
+static struct xt_match qtaguid_mt_reg __read_mostly = {
+	/*
+	 * This module masquerades as the "owner" module so that iptables
+	 * tools can deal with it.
+	 */
+	.name       = "owner",
+	.revision   = 1,
+	.family     = NFPROTO_UNSPEC,
+	.match      = qtaguid_mt,
+	.matchsize  = sizeof(struct xt_qtaguid_match_info),
+	.me         = THIS_MODULE,
+};
+
+static int __init qtaguid_mt_init(void)
+{
+	if (qtaguid_proc_register(&xt_qtaguid_procdir)
+	    || iface_stat_init(xt_qtaguid_procdir)
+	    || xt_register_match(&qtaguid_mt_reg)
+	    || misc_register(&qtu_device))
+		return -1;
+	return 0;
+}
+
+/*
+ * TODO: allow unloading of the module.
+ * For now stats are permanent.
+ * Kconfig forces'y/n' and never an 'm'.
+ */
+
+module_init(qtaguid_mt_init);
+MODULE_AUTHOR("jpa <jpa@google.com>");
+MODULE_DESCRIPTION("Xtables: socket owner+tag matching and associated stats");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("ipt_owner");
+MODULE_ALIAS("ip6t_owner");
+MODULE_ALIAS("ipt_qtaguid");
+MODULE_ALIAS("ip6t_qtaguid");
diff --git a/net/netfilter/xt_qtaguid_internal.h b/net/netfilter/xt_qtaguid_internal.h
new file mode 100644
index 000000000000..c7052707a6a4
--- /dev/null
+++ b/net/netfilter/xt_qtaguid_internal.h
@@ -0,0 +1,350 @@
+/*
+ * Kernel iptables module to track stats for packets based on user tags.
+ *
+ * (C) 2011 Google, Inc
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __XT_QTAGUID_INTERNAL_H__
+#define __XT_QTAGUID_INTERNAL_H__
+
+#include <linux/types.h>
+#include <linux/rbtree.h>
+#include <linux/spinlock_types.h>
+#include <linux/workqueue.h>
+
+/* Iface handling */
+#define IDEBUG_MASK (1<<0)
+/* Iptable Matching. Per packet. */
+#define MDEBUG_MASK (1<<1)
+/* Red-black tree handling. Per packet. */
+#define RDEBUG_MASK (1<<2)
+/* procfs ctrl/stats handling */
+#define CDEBUG_MASK (1<<3)
+/* dev and resource tracking */
+#define DDEBUG_MASK (1<<4)
+
+/* E.g (IDEBUG_MASK | CDEBUG_MASK | DDEBUG_MASK) */
+#define DEFAULT_DEBUG_MASK 0
+
+/*
+ * (Un)Define these *DEBUG to compile out/in the pr_debug calls.
+ * All undef: text size ~ 0x3030; all def: ~ 0x4404.
+ */
+#define IDEBUG
+#define MDEBUG
+#define RDEBUG
+#define CDEBUG
+#define DDEBUG
+
+#define MSK_DEBUG(mask, ...) do {                           \
+		if (unlikely(qtaguid_debug_mask & (mask)))  \
+			pr_debug(__VA_ARGS__);              \
+	} while (0)
+#ifdef IDEBUG
+#define IF_DEBUG(...) MSK_DEBUG(IDEBUG_MASK, __VA_ARGS__)
+#else
+#define IF_DEBUG(...) no_printk(__VA_ARGS__)
+#endif
+#ifdef MDEBUG
+#define MT_DEBUG(...) MSK_DEBUG(MDEBUG_MASK, __VA_ARGS__)
+#else
+#define MT_DEBUG(...) no_printk(__VA_ARGS__)
+#endif
+#ifdef RDEBUG
+#define RB_DEBUG(...) MSK_DEBUG(RDEBUG_MASK, __VA_ARGS__)
+#else
+#define RB_DEBUG(...) no_printk(__VA_ARGS__)
+#endif
+#ifdef CDEBUG
+#define CT_DEBUG(...) MSK_DEBUG(CDEBUG_MASK, __VA_ARGS__)
+#else
+#define CT_DEBUG(...) no_printk(__VA_ARGS__)
+#endif
+#ifdef DDEBUG
+#define DR_DEBUG(...) MSK_DEBUG(DDEBUG_MASK, __VA_ARGS__)
+#else
+#define DR_DEBUG(...) no_printk(__VA_ARGS__)
+#endif
+
+extern uint qtaguid_debug_mask;
+
+/*---------------------------------------------------------------------------*/
+/*
+ * Tags:
+ *
+ * They represent what the data usage counters will be tracked against.
+ * By default a tag is just based on the UID.
+ * The UID is used as the base for policing, and can not be ignored.
+ * So a tag will always at least represent a UID (uid_tag).
+ *
+ * A tag can be augmented with an "accounting tag" which is associated
+ * with a UID.
+ * User space can set the acct_tag portion of the tag which is then used
+ * with sockets: all data belonging to that socket will be counted against the
+ * tag. The policing is then based on the tag's uid_tag portion,
+ * and stats are collected for the acct_tag portion separately.
+ *
+ * There could be
+ * a:  {acct_tag=1, uid_tag=10003}
+ * b:  {acct_tag=2, uid_tag=10003}
+ * c:  {acct_tag=3, uid_tag=10003}
+ * d:  {acct_tag=0, uid_tag=10003}
+ * a, b, and c represent tags associated with specific sockets.
+ * d is for the totals for that uid, including all untagged traffic.
+ * Typically d is used with policing/quota rules.
+ *
+ * We want tag_t big enough to distinguish uid_t and acct_tag.
+ * It might become a struct if needed.
+ * Nothing should be using it as an int.
+ */
+typedef uint64_t tag_t;  /* Only used via accessors */
+
+#define TAG_UID_MASK 0xFFFFFFFFULL
+#define TAG_ACCT_MASK (~0xFFFFFFFFULL)
+
+static inline int tag_compare(tag_t t1, tag_t t2)
+{
+	return t1 < t2 ? -1 : t1 == t2 ? 0 : 1;
+}
+
+static inline tag_t combine_atag_with_uid(tag_t acct_tag, uid_t uid)
+{
+	return acct_tag | uid;
+}
+static inline tag_t make_tag_from_uid(uid_t uid)
+{
+	return uid;
+}
+static inline uid_t get_uid_from_tag(tag_t tag)
+{
+	return tag & TAG_UID_MASK;
+}
+static inline tag_t get_utag_from_tag(tag_t tag)
+{
+	return tag & TAG_UID_MASK;
+}
+static inline tag_t get_atag_from_tag(tag_t tag)
+{
+	return tag & TAG_ACCT_MASK;
+}
+
+static inline bool valid_atag(tag_t tag)
+{
+	return !(tag & TAG_UID_MASK);
+}
+static inline tag_t make_atag_from_value(uint32_t value)
+{
+	return (uint64_t)value << 32;
+}
+/*---------------------------------------------------------------------------*/
+
+/*
+ * Maximum number of socket tags that a UID is allowed to have active.
+ * Multiple processes belonging to the same UID contribute towards this limit.
+ * Special UIDs that can impersonate a UID also contribute (e.g. download
+ * manager, ...)
+ */
+#define DEFAULT_MAX_SOCK_TAGS 1024
+
+/*
+ * For now we only track 2 sets of counters.
+ * The default set is 0.
+ * Userspace can activate another set for a given uid being tracked.
+ */
+#define IFS_MAX_COUNTER_SETS 2
+
+enum ifs_tx_rx {
+	IFS_TX,
+	IFS_RX,
+	IFS_MAX_DIRECTIONS
+};
+
+/* For now, TCP, UDP, the rest */
+enum ifs_proto {
+	IFS_TCP,
+	IFS_UDP,
+	IFS_PROTO_OTHER,
+	IFS_MAX_PROTOS
+};
+
+struct byte_packet_counters {
+	uint64_t bytes;
+	uint64_t packets;
+};
+
+struct data_counters {
+	struct byte_packet_counters bpc[IFS_MAX_COUNTER_SETS][IFS_MAX_DIRECTIONS][IFS_MAX_PROTOS];
+};
+
+static inline uint64_t dc_sum_bytes(struct data_counters *counters,
+				    int set,
+				    enum ifs_tx_rx direction)
+{
+	return counters->bpc[set][direction][IFS_TCP].bytes
+		+ counters->bpc[set][direction][IFS_UDP].bytes
+		+ counters->bpc[set][direction][IFS_PROTO_OTHER].bytes;
+}
+
+static inline uint64_t dc_sum_packets(struct data_counters *counters,
+				      int set,
+				      enum ifs_tx_rx direction)
+{
+	return counters->bpc[set][direction][IFS_TCP].packets
+		+ counters->bpc[set][direction][IFS_UDP].packets
+		+ counters->bpc[set][direction][IFS_PROTO_OTHER].packets;
+}
+
+
+/* Generic X based nodes used as a base for rb_tree ops */
+struct tag_node {
+	struct rb_node node;
+	tag_t tag;
+};
+
+struct tag_stat {
+	struct tag_node tn;
+	struct data_counters counters;
+	/*
+	 * If this tag is acct_tag based, we need to count against the
+	 * matching parent uid_tag.
+	 */
+	struct data_counters *parent_counters;
+};
+
+struct iface_stat {
+	struct list_head list;  /* in iface_stat_list */
+	char *ifname;
+	bool active;
+	/* net_dev is only valid for active iface_stat */
+	struct net_device *net_dev;
+
+	struct byte_packet_counters totals_via_dev[IFS_MAX_DIRECTIONS];
+	struct data_counters totals_via_skb;
+	/*
+	 * We keep the last_known, because some devices reset their counters
+	 * just before NETDEV_UP, while some will reset just before
+	 * NETDEV_REGISTER (which is more normal).
+	 * So now, if the device didn't do a NETDEV_UNREGISTER and we see
+	 * its current dev stats smaller that what was previously known, we
+	 * assume an UNREGISTER and just use the last_known.
+	 */
+	struct byte_packet_counters last_known[IFS_MAX_DIRECTIONS];
+	/* last_known is usable when last_known_valid is true */
+	bool last_known_valid;
+
+	struct proc_dir_entry *proc_ptr;
+
+	struct rb_root tag_stat_tree;
+	spinlock_t tag_stat_list_lock;
+};
+
+/* This is needed to create proc_dir_entries from atomic context. */
+struct iface_stat_work {
+	struct work_struct iface_work;
+	struct iface_stat *iface_entry;
+};
+
+/*
+ * Track tag that this socket is transferring data for, and not necessarily
+ * the uid that owns the socket.
+ * This is the tag against which tag_stat.counters will be billed.
+ * These structs need to be looked up by sock and pid.
+ */
+struct sock_tag {
+	struct rb_node sock_node;
+	struct sock *sk;  /* Only used as a number, never dereferenced */
+	/* Used to associate with a given pid */
+	struct list_head list;   /* in proc_qtu_data.sock_tag_list */
+	pid_t pid;
+
+	tag_t tag;
+};
+
+struct qtaguid_event_counts {
+	/* Various successful events */
+	atomic64_t sockets_tagged;
+	atomic64_t sockets_untagged;
+	atomic64_t counter_set_changes;
+	atomic64_t delete_cmds;
+	atomic64_t iface_events;  /* Number of NETDEV_* events handled */
+
+	atomic64_t match_calls;   /* Number of times iptables called mt */
+	/* Number of times iptables called mt from pre or post routing hooks */
+	atomic64_t match_calls_prepost;
+	/*
+	 * match_found_sk_*: numbers related to the netfilter matching
+	 * function finding a sock for the sk_buff.
+	 * Total skbs processed is sum(match_found*).
+	 */
+	atomic64_t match_found_sk;   /* An sk was already in the sk_buff. */
+	/* The connection tracker had or didn't have the sk. */
+	atomic64_t match_found_sk_in_ct;
+	atomic64_t match_found_no_sk_in_ct;
+	/*
+	 * No sk could be found. No apparent owner. Could happen with
+	 * unsolicited traffic.
+	 */
+	atomic64_t match_no_sk;
+	/*
+	 * The file ptr in the sk_socket wasn't there and we couldn't get GID.
+	 * This might happen for traffic while the socket is being closed.
+	 */
+	atomic64_t match_no_sk_gid;
+};
+
+/* Track the set active_set for the given tag. */
+struct tag_counter_set {
+	struct tag_node tn;
+	int active_set;
+};
+
+/*----------------------------------------------*/
+/*
+ * The qtu uid data is used to track resources that are created directly or
+ * indirectly by processes (uid tracked).
+ * It is shared by the processes with the same uid.
+ * Some of the resource will be counted to prevent further rogue allocations,
+ * some will need freeing once the owner process (uid) exits.
+ */
+struct uid_tag_data {
+	struct rb_node node;
+	uid_t uid;
+
+	/*
+	 * For the uid, how many accounting tags have been set.
+	 */
+	int num_active_tags;
+	/* Track the number of proc_qtu_data that reference it */
+	int num_pqd;
+	struct rb_root tag_ref_tree;
+	/* No tag_node_tree_lock; use uid_tag_data_tree_lock */
+};
+
+struct tag_ref {
+	struct tag_node tn;
+
+	/*
+	 * This tracks the number of active sockets that have a tag on them
+	 * which matches this tag_ref.tn.tag.
+	 * A tag ref can live on after the sockets are untagged.
+	 * A tag ref can only be removed during a tag delete command.
+	 */
+	int num_sock_tags;
+};
+
+struct proc_qtu_data {
+	struct rb_node node;
+	pid_t pid;
+
+	struct uid_tag_data *parent_tag_data;
+
+	/* Tracks the sock_tags that need freeing upon this proc's death */
+	struct list_head sock_tag_list;
+	/* No spinlock_t sock_tag_list_lock; use the global one. */
+};
+
+/*----------------------------------------------*/
+#endif  /* ifndef __XT_QTAGUID_INTERNAL_H__ */
diff --git a/net/netfilter/xt_qtaguid_print.c b/net/netfilter/xt_qtaguid_print.c
new file mode 100644
index 000000000000..2a7190d285e6
--- /dev/null
+++ b/net/netfilter/xt_qtaguid_print.c
@@ -0,0 +1,566 @@
+/*
+ * Pretty printing Support for iptables xt_qtaguid module.
+ *
+ * (C) 2011 Google, Inc
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+/*
+ * Most of the functions in this file just waste time if DEBUG is not defined.
+ * The matching xt_qtaguid_print.h will static inline empty funcs if the needed
+ * debug flags ore not defined.
+ * Those funcs that fail to allocate memory will panic as there is no need to
+ * hobble allong just pretending to do the requested work.
+ */
+
+#define DEBUG
+
+#include <linux/fs.h>
+#include <linux/gfp.h>
+#include <linux/net.h>
+#include <linux/rbtree.h>
+#include <linux/slab.h>
+#include <linux/spinlock_types.h>
+#include <net/sock.h>
+
+#include "xt_qtaguid_internal.h"
+#include "xt_qtaguid_print.h"
+
+#ifdef DDEBUG
+
+static void _bug_on_err_or_null(void *ptr)
+{
+	if (IS_ERR_OR_NULL(ptr)) {
+		pr_err("qtaguid: kmalloc failed\n");
+		BUG();
+	}
+}
+
+char *pp_tag_t(tag_t *tag)
+{
+	char *res;
+
+	if (!tag)
+		res = kasprintf(GFP_ATOMIC, "tag_t@null{}");
+	else
+		res = kasprintf(GFP_ATOMIC,
+				"tag_t@%p{tag=0x%llx, uid=%u}",
+				tag, *tag, get_uid_from_tag(*tag));
+	_bug_on_err_or_null(res);
+	return res;
+}
+
+char *pp_data_counters(struct data_counters *dc, bool showValues)
+{
+	char *res;
+
+	if (!dc)
+		res = kasprintf(GFP_ATOMIC, "data_counters@null{}");
+	else if (showValues)
+		res = kasprintf(
+			GFP_ATOMIC, "data_counters@%p{"
+			"set0{"
+			"rx{"
+			"tcp{b=%llu, p=%llu}, "
+			"udp{b=%llu, p=%llu},"
+			"other{b=%llu, p=%llu}}, "
+			"tx{"
+			"tcp{b=%llu, p=%llu}, "
+			"udp{b=%llu, p=%llu},"
+			"other{b=%llu, p=%llu}}}, "
+			"set1{"
+			"rx{"
+			"tcp{b=%llu, p=%llu}, "
+			"udp{b=%llu, p=%llu},"
+			"other{b=%llu, p=%llu}}, "
+			"tx{"
+			"tcp{b=%llu, p=%llu}, "
+			"udp{b=%llu, p=%llu},"
+			"other{b=%llu, p=%llu}}}}",
+			dc,
+			dc->bpc[0][IFS_RX][IFS_TCP].bytes,
+			dc->bpc[0][IFS_RX][IFS_TCP].packets,
+			dc->bpc[0][IFS_RX][IFS_UDP].bytes,
+			dc->bpc[0][IFS_RX][IFS_UDP].packets,
+			dc->bpc[0][IFS_RX][IFS_PROTO_OTHER].bytes,
+			dc->bpc[0][IFS_RX][IFS_PROTO_OTHER].packets,
+			dc->bpc[0][IFS_TX][IFS_TCP].bytes,
+			dc->bpc[0][IFS_TX][IFS_TCP].packets,
+			dc->bpc[0][IFS_TX][IFS_UDP].bytes,
+			dc->bpc[0][IFS_TX][IFS_UDP].packets,
+			dc->bpc[0][IFS_TX][IFS_PROTO_OTHER].bytes,
+			dc->bpc[0][IFS_TX][IFS_PROTO_OTHER].packets,
+			dc->bpc[1][IFS_RX][IFS_TCP].bytes,
+			dc->bpc[1][IFS_RX][IFS_TCP].packets,
+			dc->bpc[1][IFS_RX][IFS_UDP].bytes,
+			dc->bpc[1][IFS_RX][IFS_UDP].packets,
+			dc->bpc[1][IFS_RX][IFS_PROTO_OTHER].bytes,
+			dc->bpc[1][IFS_RX][IFS_PROTO_OTHER].packets,
+			dc->bpc[1][IFS_TX][IFS_TCP].bytes,
+			dc->bpc[1][IFS_TX][IFS_TCP].packets,
+			dc->bpc[1][IFS_TX][IFS_UDP].bytes,
+			dc->bpc[1][IFS_TX][IFS_UDP].packets,
+			dc->bpc[1][IFS_TX][IFS_PROTO_OTHER].bytes,
+			dc->bpc[1][IFS_TX][IFS_PROTO_OTHER].packets);
+	else
+		res = kasprintf(GFP_ATOMIC, "data_counters@%p{...}", dc);
+	_bug_on_err_or_null(res);
+	return res;
+}
+
+char *pp_tag_node(struct tag_node *tn)
+{
+	char *tag_str;
+	char *res;
+
+	if (!tn) {
+		res = kasprintf(GFP_ATOMIC, "tag_node@null{}");
+		_bug_on_err_or_null(res);
+		return res;
+	}
+	tag_str = pp_tag_t(&tn->tag);
+	res = kasprintf(GFP_ATOMIC,
+			"tag_node@%p{tag=%s}",
+			tn, tag_str);
+	_bug_on_err_or_null(res);
+	kfree(tag_str);
+	return res;
+}
+
+char *pp_tag_ref(struct tag_ref *tr)
+{
+	char *tn_str;
+	char *res;
+
+	if (!tr) {
+		res = kasprintf(GFP_ATOMIC, "tag_ref@null{}");
+		_bug_on_err_or_null(res);
+		return res;
+	}
+	tn_str = pp_tag_node(&tr->tn);
+	res = kasprintf(GFP_ATOMIC,
+			"tag_ref@%p{%s, num_sock_tags=%d}",
+			tr, tn_str, tr->num_sock_tags);
+	_bug_on_err_or_null(res);
+	kfree(tn_str);
+	return res;
+}
+
+char *pp_tag_stat(struct tag_stat *ts)
+{
+	char *tn_str;
+	char *counters_str;
+	char *parent_counters_str;
+	char *res;
+
+	if (!ts) {
+		res = kasprintf(GFP_ATOMIC, "tag_stat@null{}");
+		_bug_on_err_or_null(res);
+		return res;
+	}
+	tn_str = pp_tag_node(&ts->tn);
+	counters_str = pp_data_counters(&ts->counters, true);
+	parent_counters_str = pp_data_counters(ts->parent_counters, false);
+	res = kasprintf(GFP_ATOMIC,
+			"tag_stat@%p{%s, counters=%s, parent_counters=%s}",
+			ts, tn_str, counters_str, parent_counters_str);
+	_bug_on_err_or_null(res);
+	kfree(tn_str);
+	kfree(counters_str);
+	kfree(parent_counters_str);
+	return res;
+}
+
+char *pp_iface_stat(struct iface_stat *is)
+{
+	char *res;
+	if (!is) {
+		res = kasprintf(GFP_ATOMIC, "iface_stat@null{}");
+	} else {
+		struct data_counters *cnts = &is->totals_via_skb;
+		res = kasprintf(GFP_ATOMIC, "iface_stat@%p{"
+				"list=list_head{...}, "
+				"ifname=%s, "
+				"total_dev={rx={bytes=%llu, "
+				"packets=%llu}, "
+				"tx={bytes=%llu, "
+				"packets=%llu}}, "
+				"total_skb={rx={bytes=%llu, "
+				"packets=%llu}, "
+				"tx={bytes=%llu, "
+				"packets=%llu}}, "
+				"last_known_valid=%d, "
+				"last_known={rx={bytes=%llu, "
+				"packets=%llu}, "
+				"tx={bytes=%llu, "
+				"packets=%llu}}, "
+				"active=%d, "
+				"net_dev=%p, "
+				"proc_ptr=%p, "
+				"tag_stat_tree=rb_root{...}}",
+				is,
+				is->ifname,
+				is->totals_via_dev[IFS_RX].bytes,
+				is->totals_via_dev[IFS_RX].packets,
+				is->totals_via_dev[IFS_TX].bytes,
+				is->totals_via_dev[IFS_TX].packets,
+				dc_sum_bytes(cnts, 0, IFS_RX),
+				dc_sum_packets(cnts, 0, IFS_RX),
+				dc_sum_bytes(cnts, 0, IFS_TX),
+				dc_sum_packets(cnts, 0, IFS_TX),
+				is->last_known_valid,
+				is->last_known[IFS_RX].bytes,
+				is->last_known[IFS_RX].packets,
+				is->last_known[IFS_TX].bytes,
+				is->last_known[IFS_TX].packets,
+				is->active,
+				is->net_dev,
+				is->proc_ptr);
+	}
+	_bug_on_err_or_null(res);
+	return res;
+}
+
+char *pp_sock_tag(struct sock_tag *st)
+{
+	char *tag_str;
+	char *res;
+
+	if (!st) {
+		res = kasprintf(GFP_ATOMIC, "sock_tag@null{}");
+		_bug_on_err_or_null(res);
+		return res;
+	}
+	tag_str = pp_tag_t(&st->tag);
+	res = kasprintf(GFP_ATOMIC, "sock_tag@%p{"
+			"sock_node=rb_node{...}, "
+			"sk=%p (f_count=%d), list=list_head{...}, "
+			"pid=%u, tag=%s}",
+			st, st->sk, atomic_read(
+				&st->sk->sk_refcnt),
+			st->pid, tag_str);
+	_bug_on_err_or_null(res);
+	kfree(tag_str);
+	return res;
+}
+
+char *pp_uid_tag_data(struct uid_tag_data *utd)
+{
+	char *res;
+
+	if (!utd)
+		res = kasprintf(GFP_ATOMIC, "uid_tag_data@null{}");
+	else
+		res = kasprintf(GFP_ATOMIC, "uid_tag_data@%p{"
+				"uid=%u, num_active_acct_tags=%d, "
+				"num_pqd=%d, "
+				"tag_node_tree=rb_root{...}, "
+				"proc_qtu_data_tree=rb_root{...}}",
+				utd, utd->uid,
+				utd->num_active_tags, utd->num_pqd);
+	_bug_on_err_or_null(res);
+	return res;
+}
+
+char *pp_proc_qtu_data(struct proc_qtu_data *pqd)
+{
+	char *parent_tag_data_str;
+	char *res;
+
+	if (!pqd) {
+		res = kasprintf(GFP_ATOMIC, "proc_qtu_data@null{}");
+		_bug_on_err_or_null(res);
+		return res;
+	}
+	parent_tag_data_str = pp_uid_tag_data(pqd->parent_tag_data);
+	res = kasprintf(GFP_ATOMIC, "proc_qtu_data@%p{"
+			"node=rb_node{...}, pid=%u, "
+			"parent_tag_data=%s, "
+			"sock_tag_list=list_head{...}}",
+			pqd, pqd->pid, parent_tag_data_str
+		);
+	_bug_on_err_or_null(res);
+	kfree(parent_tag_data_str);
+	return res;
+}
+
+/*------------------------------------------*/
+void prdebug_sock_tag_tree(int indent_level,
+			   struct rb_root *sock_tag_tree)
+{
+	struct rb_node *node;
+	struct sock_tag *sock_tag_entry;
+	char *str;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	if (RB_EMPTY_ROOT(sock_tag_tree)) {
+		str = "sock_tag_tree=rb_root{}";
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		return;
+	}
+
+	str = "sock_tag_tree=rb_root{";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+	indent_level++;
+	for (node = rb_first(sock_tag_tree);
+	     node;
+	     node = rb_next(node)) {
+		sock_tag_entry = rb_entry(node, struct sock_tag, sock_node);
+		str = pp_sock_tag(sock_tag_entry);
+		pr_debug("%*d: %s,\n", indent_level*2, indent_level, str);
+		kfree(str);
+	}
+	indent_level--;
+	str = "}";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+}
+
+void prdebug_sock_tag_list(int indent_level,
+			   struct list_head *sock_tag_list)
+{
+	struct sock_tag *sock_tag_entry;
+	char *str;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	if (list_empty(sock_tag_list)) {
+		str = "sock_tag_list=list_head{}";
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		return;
+	}
+
+	str = "sock_tag_list=list_head{";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+	indent_level++;
+	list_for_each_entry(sock_tag_entry, sock_tag_list, list) {
+		str = pp_sock_tag(sock_tag_entry);
+		pr_debug("%*d: %s,\n", indent_level*2, indent_level, str);
+		kfree(str);
+	}
+	indent_level--;
+	str = "}";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+}
+
+void prdebug_proc_qtu_data_tree(int indent_level,
+				struct rb_root *proc_qtu_data_tree)
+{
+	char *str;
+	struct rb_node *node;
+	struct proc_qtu_data *proc_qtu_data_entry;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	if (RB_EMPTY_ROOT(proc_qtu_data_tree)) {
+		str = "proc_qtu_data_tree=rb_root{}";
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		return;
+	}
+
+	str = "proc_qtu_data_tree=rb_root{";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+	indent_level++;
+	for (node = rb_first(proc_qtu_data_tree);
+	     node;
+	     node = rb_next(node)) {
+		proc_qtu_data_entry = rb_entry(node,
+					       struct proc_qtu_data,
+					       node);
+		str = pp_proc_qtu_data(proc_qtu_data_entry);
+		pr_debug("%*d: %s,\n", indent_level*2, indent_level,
+			 str);
+		kfree(str);
+		indent_level++;
+		prdebug_sock_tag_list(indent_level,
+				      &proc_qtu_data_entry->sock_tag_list);
+		indent_level--;
+
+	}
+	indent_level--;
+	str = "}";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+}
+
+void prdebug_tag_ref_tree(int indent_level, struct rb_root *tag_ref_tree)
+{
+	char *str;
+	struct rb_node *node;
+	struct tag_ref *tag_ref_entry;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	if (RB_EMPTY_ROOT(tag_ref_tree)) {
+		str = "tag_ref_tree{}";
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		return;
+	}
+
+	str = "tag_ref_tree{";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+	indent_level++;
+	for (node = rb_first(tag_ref_tree);
+	     node;
+	     node = rb_next(node)) {
+		tag_ref_entry = rb_entry(node,
+					 struct tag_ref,
+					 tn.node);
+		str = pp_tag_ref(tag_ref_entry);
+		pr_debug("%*d: %s,\n", indent_level*2, indent_level,
+			 str);
+		kfree(str);
+	}
+	indent_level--;
+	str = "}";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+}
+
+void prdebug_uid_tag_data_tree(int indent_level,
+			       struct rb_root *uid_tag_data_tree)
+{
+	char *str;
+	struct rb_node *node;
+	struct uid_tag_data *uid_tag_data_entry;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	if (RB_EMPTY_ROOT(uid_tag_data_tree)) {
+		str = "uid_tag_data_tree=rb_root{}";
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		return;
+	}
+
+	str = "uid_tag_data_tree=rb_root{";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+	indent_level++;
+	for (node = rb_first(uid_tag_data_tree);
+	     node;
+	     node = rb_next(node)) {
+		uid_tag_data_entry = rb_entry(node, struct uid_tag_data,
+					      node);
+		str = pp_uid_tag_data(uid_tag_data_entry);
+		pr_debug("%*d: %s,\n", indent_level*2, indent_level, str);
+		kfree(str);
+		if (!RB_EMPTY_ROOT(&uid_tag_data_entry->tag_ref_tree)) {
+			indent_level++;
+			prdebug_tag_ref_tree(indent_level,
+					     &uid_tag_data_entry->tag_ref_tree);
+			indent_level--;
+		}
+	}
+	indent_level--;
+	str = "}";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+}
+
+void prdebug_tag_stat_tree(int indent_level,
+				  struct rb_root *tag_stat_tree)
+{
+	char *str;
+	struct rb_node *node;
+	struct tag_stat *ts_entry;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	if (RB_EMPTY_ROOT(tag_stat_tree)) {
+		str = "tag_stat_tree{}";
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		return;
+	}
+
+	str = "tag_stat_tree{";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+	indent_level++;
+	for (node = rb_first(tag_stat_tree);
+	     node;
+	     node = rb_next(node)) {
+		ts_entry = rb_entry(node, struct tag_stat, tn.node);
+		str = pp_tag_stat(ts_entry);
+		pr_debug("%*d: %s\n", indent_level*2, indent_level,
+			 str);
+		kfree(str);
+	}
+	indent_level--;
+	str = "}";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+}
+
+void prdebug_iface_stat_list(int indent_level,
+			     struct list_head *iface_stat_list)
+{
+	char *str;
+	struct iface_stat *iface_entry;
+
+	if (!unlikely(qtaguid_debug_mask & DDEBUG_MASK))
+		return;
+
+	if (list_empty(iface_stat_list)) {
+		str = "iface_stat_list=list_head{}";
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		return;
+	}
+
+	str = "iface_stat_list=list_head{";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+	indent_level++;
+	list_for_each_entry(iface_entry, iface_stat_list, list) {
+		str = pp_iface_stat(iface_entry);
+		pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+		kfree(str);
+
+		spin_lock_bh(&iface_entry->tag_stat_list_lock);
+		if (!RB_EMPTY_ROOT(&iface_entry->tag_stat_tree)) {
+			indent_level++;
+			prdebug_tag_stat_tree(indent_level,
+					      &iface_entry->tag_stat_tree);
+			indent_level--;
+		}
+		spin_unlock_bh(&iface_entry->tag_stat_list_lock);
+	}
+	indent_level--;
+	str = "}";
+	pr_debug("%*d: %s\n", indent_level*2, indent_level, str);
+}
+
+#endif  /* ifdef DDEBUG */
+/*------------------------------------------*/
+static const char * const netdev_event_strings[] = {
+	"netdev_unknown",
+	"NETDEV_UP",
+	"NETDEV_DOWN",
+	"NETDEV_REBOOT",
+	"NETDEV_CHANGE",
+	"NETDEV_REGISTER",
+	"NETDEV_UNREGISTER",
+	"NETDEV_CHANGEMTU",
+	"NETDEV_CHANGEADDR",
+	"NETDEV_GOING_DOWN",
+	"NETDEV_CHANGENAME",
+	"NETDEV_FEAT_CHANGE",
+	"NETDEV_BONDING_FAILOVER",
+	"NETDEV_PRE_UP",
+	"NETDEV_PRE_TYPE_CHANGE",
+	"NETDEV_POST_TYPE_CHANGE",
+	"NETDEV_POST_INIT",
+	"NETDEV_UNREGISTER_BATCH",
+	"NETDEV_RELEASE",
+	"NETDEV_NOTIFY_PEERS",
+	"NETDEV_JOIN",
+};
+
+const char *netdev_evt_str(int netdev_event)
+{
+	if (netdev_event < 0
+	    || netdev_event >= ARRAY_SIZE(netdev_event_strings))
+		return "bad event num";
+	return netdev_event_strings[netdev_event];
+}
diff --git a/net/netfilter/xt_qtaguid_print.h b/net/netfilter/xt_qtaguid_print.h
new file mode 100644
index 000000000000..b63871a0be5a
--- /dev/null
+++ b/net/netfilter/xt_qtaguid_print.h
@@ -0,0 +1,120 @@
+/*
+ * Pretty printing Support for iptables xt_qtaguid module.
+ *
+ * (C) 2011 Google, Inc
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+#ifndef __XT_QTAGUID_PRINT_H__
+#define __XT_QTAGUID_PRINT_H__
+
+#include "xt_qtaguid_internal.h"
+
+#ifdef DDEBUG
+
+char *pp_tag_t(tag_t *tag);
+char *pp_data_counters(struct data_counters *dc, bool showValues);
+char *pp_tag_node(struct tag_node *tn);
+char *pp_tag_ref(struct tag_ref *tr);
+char *pp_tag_stat(struct tag_stat *ts);
+char *pp_iface_stat(struct iface_stat *is);
+char *pp_sock_tag(struct sock_tag *st);
+char *pp_uid_tag_data(struct uid_tag_data *qtd);
+char *pp_proc_qtu_data(struct proc_qtu_data *pqd);
+
+/*------------------------------------------*/
+void prdebug_sock_tag_list(int indent_level,
+			   struct list_head *sock_tag_list);
+void prdebug_sock_tag_tree(int indent_level,
+			   struct rb_root *sock_tag_tree);
+void prdebug_proc_qtu_data_tree(int indent_level,
+				struct rb_root *proc_qtu_data_tree);
+void prdebug_tag_ref_tree(int indent_level, struct rb_root *tag_ref_tree);
+void prdebug_uid_tag_data_tree(int indent_level,
+			       struct rb_root *uid_tag_data_tree);
+void prdebug_tag_stat_tree(int indent_level,
+			   struct rb_root *tag_stat_tree);
+void prdebug_iface_stat_list(int indent_level,
+			     struct list_head *iface_stat_list);
+
+#else
+
+/*------------------------------------------*/
+static inline char *pp_tag_t(tag_t *tag)
+{
+	return NULL;
+}
+static inline char *pp_data_counters(struct data_counters *dc, bool showValues)
+{
+	return NULL;
+}
+static inline char *pp_tag_node(struct tag_node *tn)
+{
+	return NULL;
+}
+static inline char *pp_tag_ref(struct tag_ref *tr)
+{
+	return NULL;
+}
+static inline char *pp_tag_stat(struct tag_stat *ts)
+{
+	return NULL;
+}
+static inline char *pp_iface_stat(struct iface_stat *is)
+{
+	return NULL;
+}
+static inline char *pp_sock_tag(struct sock_tag *st)
+{
+	return NULL;
+}
+static inline char *pp_uid_tag_data(struct uid_tag_data *qtd)
+{
+	return NULL;
+}
+static inline char *pp_proc_qtu_data(struct proc_qtu_data *pqd)
+{
+	return NULL;
+}
+
+/*------------------------------------------*/
+static inline
+void prdebug_sock_tag_list(int indent_level,
+			   struct list_head *sock_tag_list)
+{
+}
+static inline
+void prdebug_sock_tag_tree(int indent_level,
+			   struct rb_root *sock_tag_tree)
+{
+}
+static inline
+void prdebug_proc_qtu_data_tree(int indent_level,
+				struct rb_root *proc_qtu_data_tree)
+{
+}
+static inline
+void prdebug_tag_ref_tree(int indent_level, struct rb_root *tag_ref_tree)
+{
+}
+static inline
+void prdebug_uid_tag_data_tree(int indent_level,
+			       struct rb_root *uid_tag_data_tree)
+{
+}
+static inline
+void prdebug_tag_stat_tree(int indent_level,
+			   struct rb_root *tag_stat_tree)
+{
+}
+static inline
+void prdebug_iface_stat_list(int indent_level,
+			     struct list_head *iface_stat_list)
+{
+}
+#endif
+/*------------------------------------------*/
+const char *netdev_evt_str(int netdev_event);
+#endif  /* ifndef __XT_QTAGUID_PRINT_H__ */
diff --git a/net/netfilter/xt_quota2.c b/net/netfilter/xt_quota2.c
new file mode 100644
index 000000000000..834594aa0085
--- /dev/null
+++ b/net/netfilter/xt_quota2.c
@@ -0,0 +1,401 @@
+/*
+ * xt_quota2 - enhanced xt_quota that can count upwards and in packets
+ * as a minimal accounting match.
+ * by Jan Engelhardt <jengelh@medozas.de>, 2008
+ *
+ * Originally based on xt_quota.c:
+ * 	netfilter module to enforce network quotas
+ * 	Sam Johnston <samj@samj.net>
+ *
+ *	This program is free software; you can redistribute it and/or modify
+ *	it under the terms of the GNU General Public License; either
+ *	version 2 of the License, as published by the Free Software Foundation.
+ */
+#include <linux/list.h>
+#include <linux/module.h>
+#include <linux/proc_fs.h>
+#include <linux/skbuff.h>
+#include <linux/spinlock.h>
+#include <asm/atomic.h>
+#include <net/netlink.h>
+
+#include <linux/netfilter/x_tables.h>
+#include <linux/netfilter/xt_quota2.h>
+
+#ifdef CONFIG_NETFILTER_XT_MATCH_QUOTA2_LOG
+/* For compatibility, these definitions are copied from the
+ * deprecated header file <linux/netfilter_ipv4/ipt_ULOG.h> */
+#define ULOG_MAC_LEN	80
+#define ULOG_PREFIX_LEN	32
+
+/* Format of the ULOG packets passed through netlink */
+typedef struct ulog_packet_msg {
+	unsigned long mark;
+	long timestamp_sec;
+	long timestamp_usec;
+	unsigned int hook;
+	char indev_name[IFNAMSIZ];
+	char outdev_name[IFNAMSIZ];
+	size_t data_len;
+	char prefix[ULOG_PREFIX_LEN];
+	unsigned char mac_len;
+	unsigned char mac[ULOG_MAC_LEN];
+	unsigned char payload[0];
+} ulog_packet_msg_t;
+#endif
+
+/**
+ * @lock:	lock to protect quota writers from each other
+ */
+struct xt_quota_counter {
+	u_int64_t quota;
+	spinlock_t lock;
+	struct list_head list;
+	atomic_t ref;
+	char name[sizeof(((struct xt_quota_mtinfo2 *)NULL)->name)];
+	struct proc_dir_entry *procfs_entry;
+};
+
+#ifdef CONFIG_NETFILTER_XT_MATCH_QUOTA2_LOG
+/* Harald's favorite number +1 :D From ipt_ULOG.C */
+static int qlog_nl_event = 112;
+module_param_named(event_num, qlog_nl_event, uint, S_IRUGO | S_IWUSR);
+MODULE_PARM_DESC(event_num,
+		 "Event number for NETLINK_NFLOG message. 0 disables log."
+		 "111 is what ipt_ULOG uses.");
+static struct sock *nflognl;
+#endif
+
+static LIST_HEAD(counter_list);
+static DEFINE_SPINLOCK(counter_list_lock);
+
+static struct proc_dir_entry *proc_xt_quota;
+static unsigned int quota_list_perms = S_IRUGO | S_IWUSR;
+static kuid_t quota_list_uid = KUIDT_INIT(0);
+static kgid_t quota_list_gid = KGIDT_INIT(0);
+module_param_named(perms, quota_list_perms, uint, S_IRUGO | S_IWUSR);
+
+#ifdef CONFIG_NETFILTER_XT_MATCH_QUOTA2_LOG
+static void quota2_log(unsigned int hooknum,
+		       const struct sk_buff *skb,
+		       const struct net_device *in,
+		       const struct net_device *out,
+		       const char *prefix)
+{
+	ulog_packet_msg_t *pm;
+	struct sk_buff *log_skb;
+	size_t size;
+	struct nlmsghdr *nlh;
+
+	if (!qlog_nl_event)
+		return;
+
+	size = NLMSG_SPACE(sizeof(*pm));
+	size = max(size, (size_t)NLMSG_GOODSIZE);
+	log_skb = alloc_skb(size, GFP_ATOMIC);
+	if (!log_skb) {
+		pr_err("xt_quota2: cannot alloc skb for logging\n");
+		return;
+	}
+
+	nlh = nlmsg_put(log_skb, /*pid*/0, /*seq*/0, qlog_nl_event,
+			sizeof(*pm), 0);
+	if (!nlh) {
+		pr_err("xt_quota2: nlmsg_put failed\n");
+		kfree_skb(log_skb);
+		return;
+	}
+	pm = nlmsg_data(nlh);
+	if (skb->tstamp.tv64 == 0)
+		__net_timestamp((struct sk_buff *)skb);
+	pm->data_len = 0;
+	pm->hook = hooknum;
+	if (prefix != NULL)
+		strlcpy(pm->prefix, prefix, sizeof(pm->prefix));
+	else
+		*(pm->prefix) = '\0';
+	if (in)
+		strlcpy(pm->indev_name, in->name, sizeof(pm->indev_name));
+	else
+		pm->indev_name[0] = '\0';
+
+	if (out)
+		strlcpy(pm->outdev_name, out->name, sizeof(pm->outdev_name));
+	else
+		pm->outdev_name[0] = '\0';
+
+	NETLINK_CB(log_skb).dst_group = 1;
+	pr_debug("throwing 1 packets to netlink group 1\n");
+	netlink_broadcast(nflognl, log_skb, 0, 1, GFP_ATOMIC);
+}
+#else
+static void quota2_log(unsigned int hooknum,
+		       const struct sk_buff *skb,
+		       const struct net_device *in,
+		       const struct net_device *out,
+		       const char *prefix)
+{
+}
+#endif  /* if+else CONFIG_NETFILTER_XT_MATCH_QUOTA2_LOG */
+
+static ssize_t quota_proc_read(struct file *file, char __user *buf,
+			   size_t size, loff_t *ppos)
+{
+	struct xt_quota_counter *e = PDE_DATA(file_inode(file));
+	char tmp[24];
+	size_t tmp_size;
+
+	spin_lock_bh(&e->lock);
+	tmp_size = scnprintf(tmp, sizeof(tmp), "%llu\n", e->quota);
+	spin_unlock_bh(&e->lock);
+	return simple_read_from_buffer(buf, size, ppos, tmp, tmp_size);
+}
+
+static ssize_t quota_proc_write(struct file *file, const char __user *input,
+                            size_t size, loff_t *ppos)
+{
+	struct xt_quota_counter *e = PDE_DATA(file_inode(file));
+	char buf[sizeof("18446744073709551616")];
+
+	if (size > sizeof(buf))
+		size = sizeof(buf);
+	if (copy_from_user(buf, input, size) != 0)
+		return -EFAULT;
+	buf[sizeof(buf)-1] = '\0';
+
+	spin_lock_bh(&e->lock);
+	e->quota = simple_strtoull(buf, NULL, 0);
+	spin_unlock_bh(&e->lock);
+	return size;
+}
+
+static const struct file_operations q2_counter_fops = {
+	.read		= quota_proc_read,
+	.write		= quota_proc_write,
+	.llseek		= default_llseek,
+};
+
+static struct xt_quota_counter *
+q2_new_counter(const struct xt_quota_mtinfo2 *q, bool anon)
+{
+	struct xt_quota_counter *e;
+	unsigned int size;
+
+	/* Do not need all the procfs things for anonymous counters. */
+	size = anon ? offsetof(typeof(*e), list) : sizeof(*e);
+	e = kmalloc(size, GFP_KERNEL);
+	if (e == NULL)
+		return NULL;
+
+	e->quota = q->quota;
+	spin_lock_init(&e->lock);
+	if (!anon) {
+		INIT_LIST_HEAD(&e->list);
+		atomic_set(&e->ref, 1);
+		strlcpy(e->name, q->name, sizeof(e->name));
+	}
+	return e;
+}
+
+/**
+ * q2_get_counter - get ref to counter or create new
+ * @name:	name of counter
+ */
+static struct xt_quota_counter *
+q2_get_counter(const struct xt_quota_mtinfo2 *q)
+{
+	struct proc_dir_entry *p;
+	struct xt_quota_counter *e = NULL;
+	struct xt_quota_counter *new_e;
+
+	if (*q->name == '\0')
+		return q2_new_counter(q, true);
+
+	/* No need to hold a lock while getting a new counter */
+	new_e = q2_new_counter(q, false);
+	if (new_e == NULL)
+		goto out;
+
+	spin_lock_bh(&counter_list_lock);
+	list_for_each_entry(e, &counter_list, list)
+		if (strcmp(e->name, q->name) == 0) {
+			atomic_inc(&e->ref);
+			spin_unlock_bh(&counter_list_lock);
+			kfree(new_e);
+			pr_debug("xt_quota2: old counter name=%s", e->name);
+			return e;
+		}
+	e = new_e;
+	pr_debug("xt_quota2: new_counter name=%s", e->name);
+	list_add_tail(&e->list, &counter_list);
+	/* The entry having a refcount of 1 is not directly destructible.
+	 * This func has not yet returned the new entry, thus iptables
+	 * has not references for destroying this entry.
+	 * For another rule to try to destroy it, it would 1st need for this
+	 * func* to be re-invoked, acquire a new ref for the same named quota.
+	 * Nobody will access the e->procfs_entry either.
+	 * So release the lock. */
+	spin_unlock_bh(&counter_list_lock);
+
+	/* create_proc_entry() is not spin_lock happy */
+	p = e->procfs_entry = proc_create_data(e->name, quota_list_perms,
+	                      proc_xt_quota, &q2_counter_fops, e);
+
+	if (IS_ERR_OR_NULL(p)) {
+		spin_lock_bh(&counter_list_lock);
+		list_del(&e->list);
+		spin_unlock_bh(&counter_list_lock);
+		goto out;
+	}
+	proc_set_user(p, quota_list_uid, quota_list_gid);
+	return e;
+
+ out:
+	kfree(e);
+	return NULL;
+}
+
+static int quota_mt2_check(const struct xt_mtchk_param *par)
+{
+	struct xt_quota_mtinfo2 *q = par->matchinfo;
+
+	pr_debug("xt_quota2: check() flags=0x%04x", q->flags);
+
+	if (q->flags & ~XT_QUOTA_MASK)
+		return -EINVAL;
+
+	q->name[sizeof(q->name)-1] = '\0';
+	if (*q->name == '.' || strchr(q->name, '/') != NULL) {
+		printk(KERN_ERR "xt_quota.3: illegal name\n");
+		return -EINVAL;
+	}
+
+	q->master = q2_get_counter(q);
+	if (q->master == NULL) {
+		printk(KERN_ERR "xt_quota.3: memory alloc failure\n");
+		return -ENOMEM;
+	}
+
+	return 0;
+}
+
+static void quota_mt2_destroy(const struct xt_mtdtor_param *par)
+{
+	struct xt_quota_mtinfo2 *q = par->matchinfo;
+	struct xt_quota_counter *e = q->master;
+
+	if (*q->name == '\0') {
+		kfree(e);
+		return;
+	}
+
+	spin_lock_bh(&counter_list_lock);
+	if (!atomic_dec_and_test(&e->ref)) {
+		spin_unlock_bh(&counter_list_lock);
+		return;
+	}
+
+	list_del(&e->list);
+	remove_proc_entry(e->name, proc_xt_quota);
+	spin_unlock_bh(&counter_list_lock);
+	kfree(e);
+}
+
+static bool
+quota_mt2(const struct sk_buff *skb, struct xt_action_param *par)
+{
+	struct xt_quota_mtinfo2 *q = (void *)par->matchinfo;
+	struct xt_quota_counter *e = q->master;
+	bool ret = q->flags & XT_QUOTA_INVERT;
+
+	spin_lock_bh(&e->lock);
+	if (q->flags & XT_QUOTA_GROW) {
+		/*
+		 * While no_change is pointless in "grow" mode, we will
+		 * implement it here simply to have a consistent behavior.
+		 */
+		if (!(q->flags & XT_QUOTA_NO_CHANGE)) {
+			e->quota += (q->flags & XT_QUOTA_PACKET) ? 1 : skb->len;
+		}
+		ret = true;
+	} else {
+		if (e->quota >= skb->len) {
+			if (!(q->flags & XT_QUOTA_NO_CHANGE))
+				e->quota -= (q->flags & XT_QUOTA_PACKET) ? 1 : skb->len;
+			ret = !ret;
+		} else {
+			/* We are transitioning, log that fact. */
+			if (e->quota) {
+				quota2_log(par->hooknum,
+					   skb,
+					   par->in,
+					   par->out,
+					   q->name);
+			}
+			/* we do not allow even small packets from now on */
+			e->quota = 0;
+		}
+	}
+	spin_unlock_bh(&e->lock);
+	return ret;
+}
+
+static struct xt_match quota_mt2_reg[] __read_mostly = {
+	{
+		.name       = "quota2",
+		.revision   = 3,
+		.family     = NFPROTO_IPV4,
+		.checkentry = quota_mt2_check,
+		.match      = quota_mt2,
+		.destroy    = quota_mt2_destroy,
+		.matchsize  = sizeof(struct xt_quota_mtinfo2),
+		.me         = THIS_MODULE,
+	},
+	{
+		.name       = "quota2",
+		.revision   = 3,
+		.family     = NFPROTO_IPV6,
+		.checkentry = quota_mt2_check,
+		.match      = quota_mt2,
+		.destroy    = quota_mt2_destroy,
+		.matchsize  = sizeof(struct xt_quota_mtinfo2),
+		.me         = THIS_MODULE,
+	},
+};
+
+static int __init quota_mt2_init(void)
+{
+	int ret;
+	pr_debug("xt_quota2: init()");
+
+#ifdef CONFIG_NETFILTER_XT_MATCH_QUOTA2_LOG
+	nflognl = netlink_kernel_create(&init_net, NETLINK_NFLOG, NULL);
+	if (!nflognl)
+		return -ENOMEM;
+#endif
+
+	proc_xt_quota = proc_mkdir("xt_quota", init_net.proc_net);
+	if (proc_xt_quota == NULL)
+		return -EACCES;
+
+	ret = xt_register_matches(quota_mt2_reg, ARRAY_SIZE(quota_mt2_reg));
+	if (ret < 0)
+		remove_proc_entry("xt_quota", init_net.proc_net);
+	pr_debug("xt_quota2: init() %d", ret);
+	return ret;
+}
+
+static void __exit quota_mt2_exit(void)
+{
+	xt_unregister_matches(quota_mt2_reg, ARRAY_SIZE(quota_mt2_reg));
+	remove_proc_entry("xt_quota", init_net.proc_net);
+}
+
+module_init(quota_mt2_init);
+module_exit(quota_mt2_exit);
+MODULE_DESCRIPTION("Xtables: countdown quota match; up counter");
+MODULE_AUTHOR("Sam Johnston <samj@samj.net>");
+MODULE_AUTHOR("Jan Engelhardt <jengelh@medozas.de>");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("ipt_quota2");
+MODULE_ALIAS("ip6t_quota2");
diff --git a/net/netfilter/xt_socket.c b/net/netfilter/xt_socket.c
index 2ec08f04b816..8a2a489b2cd3 100644
--- a/net/netfilter/xt_socket.c
+++ b/net/netfilter/xt_socket.c
@@ -143,11 +143,12 @@ static bool xt_socket_sk_is_transparent(struct sock *sk)
 	}
 }
 
-static struct sock *xt_socket_lookup_slow_v4(struct net *net,
+struct sock *xt_socket_lookup_slow_v4(struct net *net,
 					     const struct sk_buff *skb,
 					     const struct net_device *indev)
 {
 	const struct iphdr *iph = ip_hdr(skb);
+	struct sock *sk = skb->sk;
 	__be32 uninitialized_var(daddr), uninitialized_var(saddr);
 	__be16 uninitialized_var(dport), uninitialized_var(sport);
 	u8 uninitialized_var(protocol);
@@ -198,9 +199,16 @@ static struct sock *xt_socket_lookup_slow_v4(struct net *net,
 	}
 #endif
 
-	return xt_socket_get_sock_v4(net, protocol, saddr, daddr,
-				     sport, dport, indev);
+	if (sk)
+		atomic_inc(&sk->sk_refcnt);
+	else
+		sk = xt_socket_get_sock_v4(dev_net(skb->dev), protocol,
+					   saddr, daddr, sport, dport,
+					   indev);
+
+	return sk;
 }
+EXPORT_SYMBOL(xt_socket_lookup_slow_v4);
 
 static bool
 socket_match(const struct sk_buff *skb, struct xt_action_param *par,
@@ -232,8 +240,7 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par,
 		    transparent)
 			pskb->mark = sk->sk_mark;
 
-		if (sk != skb->sk)
-			sock_gen_put(sk);
+		sock_gen_put(sk);
 
 		if (wildcard || !transparent)
 			sk = NULL;
@@ -336,10 +343,11 @@ xt_socket_get_sock_v6(struct net *net, const u8 protocol,
 	return NULL;
 }
 
-static struct sock *xt_socket_lookup_slow_v6(struct net *net,
+struct sock *xt_socket_lookup_slow_v6(struct net *net,
 					     const struct sk_buff *skb,
 					     const struct net_device *indev)
 {
+	struct sock *sk = skb->sk;
 	__be16 uninitialized_var(dport), uninitialized_var(sport);
 	const struct in6_addr *daddr = NULL, *saddr = NULL;
 	struct ipv6hdr *iph = ipv6_hdr(skb);
@@ -373,9 +381,16 @@ static struct sock *xt_socket_lookup_slow_v6(struct net *net,
 		return NULL;
 	}
 
-	return xt_socket_get_sock_v6(net, tproto, saddr, daddr,
-				     sport, dport, indev);
+	if (sk)
+		atomic_inc(&sk->sk_refcnt);
+	else
+		sk = xt_socket_get_sock_v6(dev_net(skb->dev), tproto,
+					   saddr, daddr, sport, dport,
+					   indev);
+
+	return sk;
 }
+EXPORT_SYMBOL(xt_socket_lookup_slow_v6);
 
 static bool
 socket_mt6_v1_v2_v3(const struct sk_buff *skb, struct xt_action_param *par)
diff --git a/net/nfc/hci/core.c b/net/nfc/hci/core.c
index 2b0f0ac498d2..5a58f9f38095 100644
--- a/net/nfc/hci/core.c
+++ b/net/nfc/hci/core.c
@@ -209,6 +209,11 @@ void nfc_hci_cmd_received(struct nfc_hci_dev *hdev, u8 pipe, u8 cmd,
 		}
 		create_info = (struct hci_create_pipe_resp *)skb->data;
 
+		if (create_info->pipe >= NFC_HCI_MAX_PIPES) {
+			status = NFC_HCI_ANY_E_NOK;
+			goto exit;
+		}
+
 		/* Save the new created pipe and bind with local gate,
 		 * the description for skb->data[3] is destination gate id
 		 * but since we received this cmd from host controller, we
@@ -232,6 +237,11 @@ void nfc_hci_cmd_received(struct nfc_hci_dev *hdev, u8 pipe, u8 cmd,
 		}
 		delete_info = (struct hci_delete_pipe_noti *)skb->data;
 
+		if (delete_info->pipe >= NFC_HCI_MAX_PIPES) {
+			status = NFC_HCI_ANY_E_NOK;
+			goto exit;
+		}
+
 		hdev->pipes[delete_info->pipe].gate = NFC_HCI_INVALID_GATE;
 		hdev->pipes[delete_info->pipe].dest_host = NFC_HCI_INVALID_HOST;
 		break;
diff --git a/net/rfkill/Kconfig b/net/rfkill/Kconfig
index 598d374f6a35..44c3be9d1021 100644
--- a/net/rfkill/Kconfig
+++ b/net/rfkill/Kconfig
@@ -10,6 +10,11 @@ menuconfig RFKILL
 	  To compile this driver as a module, choose M here: the
 	  module will be called rfkill.
 
+config RFKILL_PM
+	bool "Power off on suspend"
+	depends on RFKILL && PM
+	default y
+
 # LED trigger support
 config RFKILL_LEDS
 	bool
diff --git a/net/rfkill/core.c b/net/rfkill/core.c
index cf5b69ab1829..71e1f0def5a5 100644
--- a/net/rfkill/core.c
+++ b/net/rfkill/core.c
@@ -802,8 +802,7 @@ void rfkill_resume_polling(struct rfkill *rfkill)
 }
 EXPORT_SYMBOL(rfkill_resume_polling);
 
-#ifdef CONFIG_PM_SLEEP
-static int rfkill_suspend(struct device *dev)
+static __maybe_unused int rfkill_suspend(struct device *dev)
 {
 	struct rfkill *rfkill = to_rfkill(dev);
 
@@ -812,7 +811,7 @@ static int rfkill_suspend(struct device *dev)
 	return 0;
 }
 
-static int rfkill_resume(struct device *dev)
+static __maybe_unused int rfkill_resume(struct device *dev)
 {
 	struct rfkill *rfkill = to_rfkill(dev);
 	bool cur;
@@ -828,17 +827,13 @@ static int rfkill_resume(struct device *dev)
 }
 
 static SIMPLE_DEV_PM_OPS(rfkill_pm_ops, rfkill_suspend, rfkill_resume);
-#define RFKILL_PM_OPS (&rfkill_pm_ops)
-#else
-#define RFKILL_PM_OPS NULL
-#endif
 
 static struct class rfkill_class = {
 	.name		= "rfkill",
 	.dev_release	= rfkill_release,
 	.dev_groups	= rfkill_dev_groups,
 	.dev_uevent	= rfkill_dev_uevent,
-	.pm		= RFKILL_PM_OPS,
+	.pm		= IS_ENABLED(CONFIG_RFKILL_PM) ? &rfkill_pm_ops : NULL,
 };
 
 bool rfkill_blocked(struct rfkill *rfkill)
diff --git a/net/socket.c b/net/socket.c
index fbfa9d2492cf..18aff3d804ec 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -520,9 +520,23 @@ static ssize_t sockfs_listxattr(struct dentry *dentry, char *buffer,
 	return used;
 }
 
+static int sockfs_setattr(struct dentry *dentry, struct iattr *iattr)
+{
+	int err = simple_setattr(dentry, iattr);
+
+	if (!err && (iattr->ia_valid & ATTR_UID)) {
+		struct socket *sock = SOCKET_I(d_inode(dentry));
+
+		sock->sk->sk_uid = iattr->ia_uid;
+	}
+
+	return err;
+}
+
 static const struct inode_operations sockfs_inode_ops = {
 	.getxattr = sockfs_getxattr,
 	.listxattr = sockfs_listxattr,
+	.setattr = sockfs_setattr,
 };
 
 /**
diff --git a/net/wireless/scan.c b/net/wireless/scan.c
index 8dde12a11725..6e7b86ca2abd 100644
--- a/net/wireless/scan.c
+++ b/net/wireless/scan.c
@@ -69,7 +69,7 @@ module_param(bss_entries_limit, int, 0644);
 MODULE_PARM_DESC(bss_entries_limit,
                  "limit to number of scan BSS entries (per wiphy, default 1000)");
 
-#define IEEE80211_SCAN_RESULT_EXPIRE	(30 * HZ)
+#define IEEE80211_SCAN_RESULT_EXPIRE	(7 * HZ)
 
 static void bss_free(struct cfg80211_internal_bss *bss)
 {
diff --git a/net/xfrm/xfrm_algo.c b/net/xfrm/xfrm_algo.c
index f07224d8b88f..9adfdd711b31 100644
--- a/net/xfrm/xfrm_algo.c
+++ b/net/xfrm/xfrm_algo.c
@@ -239,7 +239,7 @@ static struct xfrm_algo_desc aalg_list[] = {
 
 	.uinfo = {
 		.auth = {
-			.icv_truncbits = 96,
+			.icv_truncbits = 128,
 			.icv_fullbits = 256,
 		}
 	},
diff --git a/net/xfrm/xfrm_output.c b/net/xfrm/xfrm_output.c
index ff4a91fcab9f..16e828f2540f 100644
--- a/net/xfrm/xfrm_output.c
+++ b/net/xfrm/xfrm_output.c
@@ -66,6 +66,9 @@ static int xfrm_output_one(struct sk_buff *skb, int err)
 			goto error_nolock;
 		}
 
+		if (x->props.output_mark)
+			skb->mark = x->props.output_mark;
+
 		err = x->outer_mode->output(x, skb);
 		if (err) {
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMOUTSTATEMODEERROR);
diff --git a/net/xfrm/xfrm_policy.c b/net/xfrm/xfrm_policy.c
index 0e01250f2072..4096f699ba00 100644
--- a/net/xfrm/xfrm_policy.c
+++ b/net/xfrm/xfrm_policy.c
@@ -119,7 +119,7 @@ static inline struct dst_entry *__xfrm_dst_lookup(struct net *net,
 						  int tos, int oif,
 						  const xfrm_address_t *saddr,
 						  const xfrm_address_t *daddr,
-						  int family)
+						  int family, u32 mark)
 {
 	struct xfrm_policy_afinfo *afinfo;
 	struct dst_entry *dst;
@@ -128,7 +128,7 @@ static inline struct dst_entry *__xfrm_dst_lookup(struct net *net,
 	if (unlikely(afinfo == NULL))
 		return ERR_PTR(-EAFNOSUPPORT);
 
-	dst = afinfo->dst_lookup(net, tos, oif, saddr, daddr);
+	dst = afinfo->dst_lookup(net, tos, oif, saddr, daddr, mark);
 
 	xfrm_policy_put_afinfo(afinfo);
 
@@ -139,7 +139,7 @@ static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x,
 						int tos, int oif,
 						xfrm_address_t *prev_saddr,
 						xfrm_address_t *prev_daddr,
-						int family)
+						int family, u32 mark)
 {
 	struct net *net = xs_net(x);
 	xfrm_address_t *saddr = &x->props.saddr;
@@ -155,7 +155,7 @@ static inline struct dst_entry *xfrm_dst_lookup(struct xfrm_state *x,
 		daddr = x->coaddr;
 	}
 
-	dst = __xfrm_dst_lookup(net, tos, oif, saddr, daddr, family);
+	dst = __xfrm_dst_lookup(net, tos, oif, saddr, daddr, family, mark);
 
 	if (!IS_ERR(dst)) {
 		if (prev_saddr != saddr)
@@ -1395,14 +1395,14 @@ int __xfrm_sk_clone_policy(struct sock *sk, const struct sock *osk)
 
 static int
 xfrm_get_saddr(struct net *net, int oif, xfrm_address_t *local,
-	       xfrm_address_t *remote, unsigned short family)
+	       xfrm_address_t *remote, unsigned short family, u32 mark)
 {
 	int err;
 	struct xfrm_policy_afinfo *afinfo = xfrm_policy_get_afinfo(family);
 
 	if (unlikely(afinfo == NULL))
 		return -EINVAL;
-	err = afinfo->get_saddr(net, oif, local, remote);
+	err = afinfo->get_saddr(net, oif, local, remote, mark);
 	xfrm_policy_put_afinfo(afinfo);
 	return err;
 }
@@ -1433,7 +1433,7 @@ xfrm_tmpl_resolve_one(struct xfrm_policy *policy, const struct flowi *fl,
 			if (xfrm_addr_any(local, tmpl->encap_family)) {
 				error = xfrm_get_saddr(net, fl->flowi_oif,
 						       &tmp, remote,
-						       tmpl->encap_family);
+						       tmpl->encap_family, 0);
 				if (error)
 					goto fail;
 				local = &tmp;
@@ -1712,7 +1712,8 @@ static struct dst_entry *xfrm_bundle_create(struct xfrm_policy *policy,
 		if (xfrm[i]->props.mode != XFRM_MODE_TRANSPORT) {
 			family = xfrm[i]->props.family;
 			dst = xfrm_dst_lookup(xfrm[i], tos, fl->flowi_oif,
-					      &saddr, &daddr, family);
+					      &saddr, &daddr, family,
+					      xfrm[i]->props.output_mark);
 			err = PTR_ERR(dst);
 			if (IS_ERR(dst))
 				goto put_states;
diff --git a/net/xfrm/xfrm_state.c b/net/xfrm/xfrm_state.c
index 9895a8c56d8c..c0aa1b18fe8f 100644
--- a/net/xfrm/xfrm_state.c
+++ b/net/xfrm/xfrm_state.c
@@ -1869,6 +1869,7 @@ int xfrm_user_policy(struct sock *sk, int optname, u8 __user *optval, int optlen
 	if (err >= 0) {
 		xfrm_sk_policy_insert(sk, err, pol);
 		xfrm_pol_put(pol);
+		__sk_dst_reset(sk);
 		err = 0;
 	}
 
diff --git a/net/xfrm/xfrm_user.c b/net/xfrm/xfrm_user.c
index 7a5a64e70b4d..68010a01ea36 100644
--- a/net/xfrm/xfrm_user.c
+++ b/net/xfrm/xfrm_user.c
@@ -584,6 +584,9 @@ static struct xfrm_state *xfrm_state_construct(struct net *net,
 
 	xfrm_mark_get(attrs, &x->mark);
 
+	if (attrs[XFRMA_OUTPUT_MARK])
+		x->props.output_mark = nla_get_u32(attrs[XFRMA_OUTPUT_MARK]);
+
 	err = __xfrm_init_state(x, false);
 	if (err)
 		goto error;
@@ -867,6 +870,11 @@ static int copy_to_user_state_extra(struct xfrm_state *x,
 		goto out;
 	if (x->security)
 		ret = copy_sec_ctx(x->security, skb);
+	if (x->props.output_mark) {
+		ret = nla_put_u32(skb, XFRMA_OUTPUT_MARK, x->props.output_mark);
+		if (ret)
+			goto out;
+	}
 out:
 	return ret;
 }
@@ -1691,6 +1699,10 @@ static struct sk_buff *xfrm_policy_netlink(struct sk_buff *in_skb,
 	struct sk_buff *skb;
 	int err;
 
+	err = verify_policy_dir(dir);
+	if (err)
+		return ERR_PTR(err);
+
 	skb = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
 	if (!skb)
 		return ERR_PTR(-ENOMEM);
@@ -2216,6 +2228,10 @@ static int xfrm_do_migrate(struct sk_buff *skb, struct nlmsghdr *nlh,
 	int n = 0;
 	struct net *net = sock_net(skb->sk);
 
+	err = verify_policy_dir(pi->dir);
+	if (err)
+		return err;
+
 	if (attrs[XFRMA_MIGRATE] == NULL)
 		return -EINVAL;
 
@@ -2331,6 +2347,11 @@ static int xfrm_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type,
 {
 	struct net *net = &init_net;
 	struct sk_buff *skb;
+	int err;
+
+	err = verify_policy_dir(dir);
+	if (err)
+		return err;
 
 	skb = nlmsg_new(xfrm_migrate_msgsize(num_migrate, !!k), GFP_ATOMIC);
 	if (skb == NULL)
@@ -2406,6 +2427,7 @@ static const struct nla_policy xfrma_policy[XFRMA_MAX+1] = {
 	[XFRMA_SA_EXTRA_FLAGS]	= { .type = NLA_U32 },
 	[XFRMA_PROTO]		= { .type = NLA_U8 },
 	[XFRMA_ADDRESS_FILTER]	= { .len = sizeof(struct xfrm_address_filter) },
+	[XFRMA_OUTPUT_MARK]	= { .len = NLA_U32 },
 };
 
 static const struct nla_policy xfrma_spd_policy[XFRMA_SPD_MAX+1] = {
@@ -2622,6 +2644,8 @@ static inline size_t xfrm_sa_len(struct xfrm_state *x)
 		l += nla_total_size(sizeof(*x->coaddr));
 	if (x->props.extra_flags)
 		l += nla_total_size(sizeof(x->props.extra_flags));
+	if (x->props.output_mark)
+		l += nla_total_size(sizeof(x->props.output_mark));
 
 	/* Must count x->lastused as it may become non-zero behind our back. */
 	l += nla_total_size(sizeof(u64));
@@ -2985,6 +3009,11 @@ out_free_skb:
 
 static int xfrm_send_policy_notify(struct xfrm_policy *xp, int dir, const struct km_event *c)
 {
+	int err;
+
+	err = verify_policy_dir(dir);
+	if (err)
+		return err;
 
 	switch (c->event) {
 	case XFRM_MSG_NEWPOLICY:
diff --git a/scripts/Kbuild.include b/scripts/Kbuild.include
index 1db6d73c8dd2..8a1bb64f1dcd 100644
--- a/scripts/Kbuild.include
+++ b/scripts/Kbuild.include
@@ -107,16 +107,30 @@ as-option = $(call try-run,\
 as-instr = $(call try-run,\
 	printf "%b\n" "$(1)" | $(CC) $(KBUILD_AFLAGS) -c -x assembler -o "$$TMP" -,$(2),$(3))
 
+# __cc-option
+# Usage: MY_CFLAGS += $(call __cc-option,$(CC),$(MY_CFLAGS),-march=winchip-c6,-march=i586)
+__cc-option = $(call try-run,\
+	$(1) -Werror $(2) $(3) -c -x c /dev/null -o "$$TMP",$(3),$(4))
+
+# Do not attempt to build with gcc plugins during cc-option tests.
+# (And this uses delayed resolution so the flags will be up to date.)
+CC_OPTION_CFLAGS = $(filter-out $(GCC_PLUGINS_CFLAGS),$(KBUILD_CFLAGS))
+
 # cc-option
 # Usage: cflags-y += $(call cc-option,-march=winchip-c6,-march=i586)
 
-cc-option = $(call try-run,\
-	$(CC) $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) $(1) -c -x c /dev/null -o "$$TMP",$(1),$(2))
+cc-option = $(call __cc-option, $(CC),\
+	$(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS),$(1),$(2))
+
+# hostcc-option
+# Usage: cflags-y += $(call hostcc-option,-march=winchip-c6,-march=i586)
+hostcc-option = $(call __cc-option, $(HOSTCC),\
+	$(HOSTCFLAGS) $(HOST_EXTRACFLAGS),$(1),$(2))
 
 # cc-option-yn
 # Usage: flag := $(call cc-option-yn,-march=winchip-c6)
 cc-option-yn = $(call try-run,\
-	$(CC) $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) $(1) -c -x c /dev/null -o "$$TMP",y,n)
+	$(CC) -Werror $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) $(1) -c -x c /dev/null -o "$$TMP",y,n)
 
 # cc-option-align
 # Prefix align with either -falign or -malign
@@ -126,7 +140,7 @@ cc-option-align = $(subst -functions=0,,\
 # cc-disable-warning
 # Usage: cflags-y += $(call cc-disable-warning,unused-but-set-variable)
 cc-disable-warning = $(call try-run,\
-	$(CC) $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) -W$(strip $(1)) -c -x c /dev/null -o "$$TMP",-Wno-$(strip $(1)))
+	$(CC) -Werror $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS) -W$(strip $(1)) -c -x c /dev/null -o "$$TMP",-Wno-$(strip $(1)))
 
 # cc-name
 # Expands to either gcc or clang
diff --git a/scripts/Makefile.build b/scripts/Makefile.build
index 01df30af4d4a..411281c36898 100644
--- a/scripts/Makefile.build
+++ b/scripts/Makefile.build
@@ -174,6 +174,14 @@ cmd_cc_symtypes_c =                                                         \
 $(obj)/%.symtypes : $(src)/%.c FORCE
 	$(call cmd,cc_symtypes_c)
 
+# LLVM assembly
+# Generate .ll files from .c
+quiet_cmd_cc_ll_c = CC $(quiet_modtag)  $@
+      cmd_cc_ll_c = $(CC) $(c_flags) -emit-llvm -S -o $@ $<
+
+$(obj)/%.ll: $(src)/%.c FORCE
+	$(call if_changed_dep,cc_ll_c)
+
 # C (.c) files
 # The C file is compiled and updated dependency information is generated.
 # (See cmd_cc_o_c + relevant part of rule_cc_o_c)
diff --git a/scripts/Makefile.clean b/scripts/Makefile.clean
index 55c96cb8070f..a3245aea56e1 100644
--- a/scripts/Makefile.clean
+++ b/scripts/Makefile.clean
@@ -11,7 +11,7 @@ include scripts/Kbuild.include
 
 # The filename Kbuild has precedence over Makefile
 kbuild-dir := $(if $(filter /%,$(src)),$(src),$(srctree)/$(src))
-include $(if $(wildcard $(kbuild-dir)/Kbuild), $(kbuild-dir)/Kbuild, $(kbuild-dir)/Makefile)
+-include $(if $(wildcard $(kbuild-dir)/Kbuild), $(kbuild-dir)/Kbuild, $(kbuild-dir)/Makefile)
 
 # Figure out what we need to build from the various variables
 # ==========================================================================
diff --git a/scripts/Makefile.extrawarn b/scripts/Makefile.extrawarn
index da3386a9d244..abe5f47b1ab0 100644
--- a/scripts/Makefile.extrawarn
+++ b/scripts/Makefile.extrawarn
@@ -61,7 +61,6 @@ ifeq ($(cc-name),clang)
 KBUILD_CFLAGS += $(call cc-disable-warning, initializer-overrides)
 KBUILD_CFLAGS += $(call cc-disable-warning, unused-value)
 KBUILD_CFLAGS += $(call cc-disable-warning, format)
-KBUILD_CFLAGS += $(call cc-disable-warning, unknown-warning-option)
 KBUILD_CFLAGS += $(call cc-disable-warning, sign-compare)
 KBUILD_CFLAGS += $(call cc-disable-warning, format-zero-length)
 KBUILD_CFLAGS += $(call cc-disable-warning, uninitialized)
diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib
index 79e86613712f..c3b3c94e2446 100644
--- a/scripts/Makefile.lib
+++ b/scripts/Makefile.lib
@@ -294,6 +294,12 @@ $(obj)/%.dtb: $(src)/%.dts FORCE
 
 dtc-tmp = $(subst $(comma),_,$(dot-target).dts.tmp)
 
+# cat
+# ---------------------------------------------------------------------------
+# Concatentate multiple files together
+quiet_cmd_cat = CAT     $@
+cmd_cat = (cat $(filter-out FORCE,$^) > $@) || (rm -f $@; false)
+
 # Bzip2
 # ---------------------------------------------------------------------------
 
@@ -388,3 +394,34 @@ quiet_cmd_xzmisc = XZMISC  $@
 cmd_xzmisc = (cat $(filter-out FORCE,$^) | \
 	xz --check=crc32 --lzma2=dict=1MiB) > $@ || \
 	(rm -f $@ ; false)
+
+# ASM offsets
+# ---------------------------------------------------------------------------
+
+# Default sed regexp - multiline due to syntax constraints
+#
+# Use [:space:] because LLVM's integrated assembler inserts <tab> around
+# the .ascii directive whereas GCC keeps the <space> as-is.
+define sed-offsets
+	's:^[[:space:]]*\.ascii[[:space:]]*"\(.*\)".*:\1:; \
+	/^->/{s:->#\(.*\):/* \1 */:; \
+	s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; \
+	s:->::; p;}'
+endef
+
+# Use filechk to avoid rebuilds when a header changes, but the resulting file
+# does not
+define filechk_offsets
+	(set -e; \
+	 echo "#ifndef $2"; \
+	 echo "#define $2"; \
+	 echo "/*"; \
+	 echo " * DO NOT MODIFY."; \
+	 echo " *"; \
+	 echo " * This file was generated by Kbuild"; \
+	 echo " */"; \
+	 echo ""; \
+	 sed -ne $(sed-offsets); \
+	 echo ""; \
+	 echo "#endif" )
+endef
diff --git a/scripts/Makefile.modinst b/scripts/Makefile.modinst
index 07650eeaaf06..6f4c3f5a7ae3 100644
--- a/scripts/Makefile.modinst
+++ b/scripts/Makefile.modinst
@@ -29,7 +29,7 @@ quiet_cmd_modules_install = INSTALL $@
 INSTALL_MOD_DIR ?= extra
 ext-mod-dir = $(INSTALL_MOD_DIR)$(subst $(patsubst %/,%,$(KBUILD_EXTMOD)),,$(@D))
 
-modinst_dir = $(if $(KBUILD_EXTMOD),$(ext-mod-dir),kernel/$(@D))
+modinst_dir ?= $(if $(KBUILD_EXTMOD),$(ext-mod-dir),kernel/$(@D))
 
 $(modules):
 	$(call cmd,modules_install,$(MODLIB)/$(modinst_dir))
diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index 6ac6550d751c..fabdf0e5c28c 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -2348,6 +2348,7 @@ sub process {
 
 # Check for git id commit length and improperly formed commit descriptions
 		if ($in_commit_log && !$commit_log_possible_stack_dump &&
+		    $line !~ /^This reverts commit [0-9a-f]{7,40}/ &&
 		    ($line =~ /\bcommit\s+[0-9a-f]{5,}\b/i ||
 		     ($line =~ /\b[0-9a-f]{12,40}\b/i &&
 		      $line !~ /[\<\[][0-9a-f]{12,40}[\>\]]/i &&
diff --git a/scripts/mod/Makefile b/scripts/mod/Makefile
index c11212ff3510..e0cb2e4a3b15 100644
--- a/scripts/mod/Makefile
+++ b/scripts/mod/Makefile
@@ -5,32 +5,8 @@ modpost-objs	:= modpost.o file2alias.o sumversion.o
 
 devicetable-offsets-file := devicetable-offsets.h
 
-define sed-y
-	"/^->/{s:->#\(.*\):/* \1 */:; \
-	s:^->\([^ ]*\) [\$$#]*\([-0-9]*\) \(.*\):#define \1 \2 /* \3 */:; \
-	s:^->\([^ ]*\) [\$$#]*\([^ ]*\) \(.*\):#define \1 \2 /* \3 */:; \
-	s:->::; p;}"
-endef
-
-quiet_cmd_offsets = GEN     $@
-define cmd_offsets
-	(set -e; \
-	 echo "#ifndef __DEVICETABLE_OFFSETS_H__"; \
-	 echo "#define __DEVICETABLE_OFFSETS_H__"; \
-	 echo "/*"; \
-	 echo " * DO NOT MODIFY."; \
-	 echo " *"; \
-	 echo " * This file was generated by Kbuild"; \
-	 echo " *"; \
-	 echo " */"; \
-	 echo ""; \
-	 sed -ne $(sed-y) $<; \
-	 echo ""; \
-	 echo "#endif" ) > $@
-endef
-
-$(obj)/$(devicetable-offsets-file): $(obj)/devicetable-offsets.s
-	$(call if_changed,offsets)
+$(obj)/$(devicetable-offsets-file): $(obj)/devicetable-offsets.s FORCE
+	$(call filechk,offsets,__DEVICETABLE_OFFSETS_H__)
 
 targets += $(devicetable-offsets-file) devicetable-offsets.s
 
diff --git a/security/Kconfig b/security/Kconfig
index ddb3e8a8d9bd..3aa60791f84d 100644
--- a/security/Kconfig
+++ b/security/Kconfig
@@ -18,6 +18,15 @@ config SECURITY_DMESG_RESTRICT
 
 	  If you are unsure how to answer this question, answer N.
 
+config SECURITY_PERF_EVENTS_RESTRICT
+	bool "Restrict unprivileged use of performance events"
+	depends on PERF_EVENTS
+	help
+	  If you say Y here, the kernel.perf_event_paranoid sysctl
+	  will be set to 3 by default, and no unprivileged use of the
+	  perf_event_open syscall will be permitted unless it is
+	  changed.
+
 config SECURITY
 	bool "Enable different security models"
 	depends on SYSFS
@@ -136,6 +145,7 @@ config HAVE_ARCH_HARDENED_USERCOPY
 config HARDENED_USERCOPY
 	bool "Harden memory copies between kernel and userspace"
 	depends on HAVE_ARCH_HARDENED_USERCOPY
+	depends on HAVE_HARDENED_USERCOPY_ALLOCATOR
 	select BUG
 	help
 	  This option checks for obviously wrong memory regions when
diff --git a/security/commoncap.c b/security/commoncap.c
index 48071ed7c445..7fa251aea32f 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
@@ -31,6 +31,10 @@
 #include <linux/binfmts.h>
 #include <linux/personality.h>
 
+#ifdef CONFIG_ANDROID_PARANOID_NETWORK
+#include <linux/android_aid.h>
+#endif
+
 /*
  * If a non-root user executes a setuid-root binary in
  * !secure(SECURE_NOROOT) mode, then we raise capabilities.
@@ -73,6 +77,13 @@ int cap_capable(const struct cred *cred, struct user_namespace *targ_ns,
 {
 	struct user_namespace *ns = targ_ns;
 
+#ifdef CONFIG_ANDROID_PARANOID_NETWORK
+	if (cap == CAP_NET_RAW && in_egroup_p(AID_NET_RAW))
+		return 0;
+	if (cap == CAP_NET_ADMIN && in_egroup_p(AID_NET_ADMIN))
+		return 0;
+#endif
+
 	/* See if cred has the capability in the target user namespace
 	 * by examining the target user namespace and all of the target
 	 * user namespace's parents.
diff --git a/security/inode.c b/security/inode.c
index 16622aef9bde..0f1a041bf6cb 100644
--- a/security/inode.c
+++ b/security/inode.c
@@ -100,7 +100,7 @@ struct dentry *securityfs_create_file(const char *name, umode_t mode,
 	dir = d_inode(parent);
 
 	mutex_lock(&dir->i_mutex);
-	dentry = lookup_one_len(name, parent, strlen(name));
+	dentry = lookup_one_len2(name, mount, parent, strlen(name));
 	if (IS_ERR(dentry))
 		goto out;
 
diff --git a/security/lsm_audit.c b/security/lsm_audit.c
index cccbf3068cdc..45d927ab807d 100644
--- a/security/lsm_audit.c
+++ b/security/lsm_audit.c
@@ -220,7 +220,7 @@ static void dump_common_audit_data(struct audit_buffer *ab,
 	 */
 	BUILD_BUG_ON(sizeof(a->u) > sizeof(void *)*2);
 
-	audit_log_format(ab, " pid=%d comm=", task_pid_nr(current));
+	audit_log_format(ab, " pid=%d comm=", task_tgid_nr(current));
 	audit_log_untrustedstring(ab, memcpy(comm, current->comm, sizeof(comm)));
 
 	switch (a->type) {
@@ -294,7 +294,7 @@ static void dump_common_audit_data(struct audit_buffer *ab,
 	case LSM_AUDIT_DATA_TASK: {
 		struct task_struct *tsk = a->u.tsk;
 		if (tsk) {
-			pid_t pid = task_pid_nr(tsk);
+			pid_t pid = task_tgid_nr(tsk);
 			if (pid) {
 				char comm[sizeof(tsk->comm)];
 				audit_log_format(ab, " opid=%d ocomm=", pid);
diff --git a/security/security.c b/security/security.c
index 46f405ce6b0f..ae05ab153c5a 100644
--- a/security/security.c
+++ b/security/security.c
@@ -498,6 +498,7 @@ int security_path_chown(struct path *path, kuid_t uid, kgid_t gid)
 		return 0;
 	return call_int_hook(path_chown, 0, path, uid, gid);
 }
+EXPORT_SYMBOL(security_path_chown);
 
 int security_path_chroot(struct path *path)
 {
diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c
index 4b56c3b6c25f..34427384605d 100644
--- a/security/selinux/hooks.c
+++ b/security/selinux/hooks.c
@@ -420,6 +420,7 @@ static int selinux_is_sblabel_mnt(struct super_block *sb)
 		!strcmp(sb->s_type->name, "sysfs") ||
 		!strcmp(sb->s_type->name, "pstore") ||
 		!strcmp(sb->s_type->name, "debugfs") ||
+		!strcmp(sb->s_type->name, "tracefs") ||
 		!strcmp(sb->s_type->name, "rootfs");
 }
 
@@ -739,6 +740,7 @@ static int selinux_set_mnt_opts(struct super_block *sb,
 		sbsec->flags |= SE_SBPROC | SE_SBGENFS;
 
 	if (!strcmp(sb->s_type->name, "debugfs") ||
+	    !strcmp(sb->s_type->name, "tracefs") ||
 	    !strcmp(sb->s_type->name, "sysfs") ||
 	    !strcmp(sb->s_type->name, "pstore"))
 		sbsec->flags |= SE_SBGENFS;
@@ -3653,6 +3655,38 @@ static int selinux_kernel_module_request(char *kmod_name)
 			    SYSTEM__MODULE_REQUEST, &ad);
 }
 
+static int selinux_kernel_module_from_file(struct file *file)
+{
+	struct common_audit_data ad;
+	struct inode_security_struct *isec;
+	struct file_security_struct *fsec;
+	struct inode *inode;
+	u32 sid = current_sid();
+	int rc;
+
+	/* init_module */
+	if (file == NULL)
+		return avc_has_perm(sid, sid, SECCLASS_SYSTEM,
+					SYSTEM__MODULE_LOAD, NULL);
+
+	/* finit_module */
+	ad.type = LSM_AUDIT_DATA_PATH;
+	ad.u.path = file->f_path;
+
+	inode = file_inode(file);
+	isec = inode->i_security;
+	fsec = file->f_security;
+
+	if (sid != fsec->sid) {
+		rc = avc_has_perm(sid, fsec->sid, SECCLASS_FD, FD__USE, &ad);
+		if (rc)
+			return rc;
+	}
+
+	return avc_has_perm(sid, isec->sid, SECCLASS_SYSTEM,
+				SYSTEM__MODULE_LOAD, &ad);
+}
+
 static int selinux_task_setpgid(struct task_struct *p, pid_t pgid)
 {
 	return current_has_perm(p, PROCESS__SETPGID);
@@ -5943,6 +5977,7 @@ static struct security_hook_list selinux_hooks[] = {
 	LSM_HOOK_INIT(kernel_act_as, selinux_kernel_act_as),
 	LSM_HOOK_INIT(kernel_create_files_as, selinux_kernel_create_files_as),
 	LSM_HOOK_INIT(kernel_module_request, selinux_kernel_module_request),
+	LSM_HOOK_INIT(kernel_module_from_file, selinux_kernel_module_from_file),
 	LSM_HOOK_INIT(task_setpgid, selinux_task_setpgid),
 	LSM_HOOK_INIT(task_getpgid, selinux_task_getpgid),
 	LSM_HOOK_INIT(task_getsid, selinux_task_getsid),
diff --git a/security/selinux/include/classmap.h b/security/selinux/include/classmap.h
index 5a4eef59aeff..b393d29ae857 100644
--- a/security/selinux/include/classmap.h
+++ b/security/selinux/include/classmap.h
@@ -32,7 +32,7 @@ struct security_class_mapping secclass_map[] = {
 	    "setsockcreate", NULL } },
 	{ "system",
 	  { "ipc_info", "syslog_read", "syslog_mod",
-	    "syslog_console", "module_request", NULL } },
+	    "syslog_console", "module_request", "module_load", NULL } },
 	{ "capability",
 	  { "chown", "dac_override", "dac_read_search",
 	    "fowner", "fsetid", "kill", "setgid", "setuid", "setpcap",
diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c
index 3628d3a868b6..31345376aa9b 100644
--- a/security/selinux/ss/avtab.c
+++ b/security/selinux/ss/avtab.c
@@ -374,6 +374,32 @@ void avtab_hash_eval(struct avtab *h, char *tag)
 	       chain2_len_sum);
 }
 
+/*
+ * extended permissions compatibility. Make ToT Android kernels compatible
+ * with Android M releases
+ */
+#define AVTAB_OPTYPE_ALLOWED	0x1000
+#define AVTAB_OPTYPE_AUDITALLOW	0x2000
+#define AVTAB_OPTYPE_DONTAUDIT	0x4000
+#define AVTAB_OPTYPE		(AVTAB_OPTYPE_ALLOWED | \
+				AVTAB_OPTYPE_AUDITALLOW | \
+				AVTAB_OPTYPE_DONTAUDIT)
+#define AVTAB_XPERMS_OPTYPE	4
+
+#define avtab_xperms_to_optype(x) (x << AVTAB_XPERMS_OPTYPE)
+#define avtab_optype_to_xperms(x) (x >> AVTAB_XPERMS_OPTYPE)
+
+static unsigned int avtab_android_m_compat;
+
+static void avtab_android_m_compat_set(void)
+{
+	if (!avtab_android_m_compat) {
+		pr_info("SELinux:  Android master kernel running Android"
+				" M policy in compatibility mode.\n");
+		avtab_android_m_compat = 1;
+	}
+}
+
 static uint16_t spec_order[] = {
 	AVTAB_ALLOWED,
 	AVTAB_AUDITDENY,
@@ -398,6 +424,7 @@ int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol,
 	struct avtab_datum datum;
 	struct avtab_extended_perms xperms;
 	__le32 buf32[ARRAY_SIZE(xperms.perms.p)];
+	unsigned int android_m_compat_optype = 0;
 	int i, rc;
 	unsigned set;
 
@@ -488,6 +515,13 @@ int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol,
 	key.target_class = le16_to_cpu(buf16[items++]);
 	key.specified = le16_to_cpu(buf16[items++]);
 
+	if ((key.specified & AVTAB_OPTYPE) &&
+			(vers == POLICYDB_VERSION_XPERMS_IOCTL)) {
+		key.specified = avtab_optype_to_xperms(key.specified);
+		android_m_compat_optype = 1;
+		avtab_android_m_compat_set();
+	}
+
 	if (!policydb_type_isvalid(pol, key.source_type) ||
 	    !policydb_type_isvalid(pol, key.target_type) ||
 	    !policydb_class_isvalid(pol, key.target_class)) {
@@ -518,10 +552,22 @@ int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol,
 			printk(KERN_ERR "SELinux: avtab: truncated entry\n");
 			return rc;
 		}
-		rc = next_entry(&xperms.driver, fp, sizeof(u8));
-		if (rc) {
-			printk(KERN_ERR "SELinux: avtab: truncated entry\n");
-			return rc;
+		if (avtab_android_m_compat ||
+			    ((xperms.specified != AVTAB_XPERMS_IOCTLFUNCTION) &&
+			    (xperms.specified != AVTAB_XPERMS_IOCTLDRIVER) &&
+			    (vers == POLICYDB_VERSION_XPERMS_IOCTL))) {
+			xperms.driver = xperms.specified;
+			if (android_m_compat_optype)
+				xperms.specified = AVTAB_XPERMS_IOCTLDRIVER;
+			else
+				xperms.specified = AVTAB_XPERMS_IOCTLFUNCTION;
+			avtab_android_m_compat_set();
+		} else {
+			rc = next_entry(&xperms.driver, fp, sizeof(u8));
+			if (rc) {
+				printk(KERN_ERR "SELinux: avtab: truncated entry\n");
+				return rc;
+			}
 		}
 		rc = next_entry(buf32, fp, sizeof(u32)*ARRAY_SIZE(xperms.perms.p));
 		if (rc) {
@@ -607,15 +653,22 @@ int avtab_write_item(struct policydb *p, struct avtab_node *cur, void *fp)
 	buf16[0] = cpu_to_le16(cur->key.source_type);
 	buf16[1] = cpu_to_le16(cur->key.target_type);
 	buf16[2] = cpu_to_le16(cur->key.target_class);
-	buf16[3] = cpu_to_le16(cur->key.specified);
+	if (avtab_android_m_compat && (cur->key.specified & AVTAB_XPERMS) &&
+		    (cur->datum.u.xperms->specified == AVTAB_XPERMS_IOCTLDRIVER))
+		buf16[3] = cpu_to_le16(avtab_xperms_to_optype(cur->key.specified));
+	else
+		buf16[3] = cpu_to_le16(cur->key.specified);
 	rc = put_entry(buf16, sizeof(u16), 4, fp);
 	if (rc)
 		return rc;
 
 	if (cur->key.specified & AVTAB_XPERMS) {
-		rc = put_entry(&cur->datum.u.xperms->specified, sizeof(u8), 1, fp);
-		if (rc)
-			return rc;
+		if (avtab_android_m_compat == 0) {
+			rc = put_entry(&cur->datum.u.xperms->specified,
+					sizeof(u8), 1, fp);
+			if (rc)
+				return rc;
+		}
 		rc = put_entry(&cur->datum.u.xperms->driver, sizeof(u8), 1, fp);
 		if (rc)
 			return rc;
diff --git a/tools/perf/util/evsel.c b/tools/perf/util/evsel.c
index 39a8bd842d0d..6e0a16c7176a 100644
--- a/tools/perf/util/evsel.c
+++ b/tools/perf/util/evsel.c
@@ -2356,12 +2356,15 @@ int perf_evsel__open_strerror(struct perf_evsel *evsel, struct target *target,
 	case EPERM:
 	case EACCES:
 		return scnprintf(msg, size,
-		 "You may not have permission to collect %sstats.\n"
-		 "Consider tweaking /proc/sys/kernel/perf_event_paranoid:\n"
-		 " -1 - Not paranoid at all\n"
-		 "  0 - Disallow raw tracepoint access for unpriv\n"
-		 "  1 - Disallow cpu events for unpriv\n"
-		 "  2 - Disallow kernel profiling for unpriv",
+		 "You may not have permission to collect %sstats.\n\n"
+		 "Consider tweaking /proc/sys/kernel/perf_event_paranoid,\n"
+		 "which controls use of the performance events system by\n"
+		 "unprivileged users (without CAP_SYS_ADMIN).\n\n"
+		 "The default value is 1:\n\n"
+		 "  -1: Allow use of (almost) all events by all users\n"
+		 ">= 0: Disallow raw tracepoint access by users without CAP_IOC_LOCK\n"
+		 ">= 1: Disallow CPU event access by users without CAP_SYS_ADMIN\n"
+		 ">= 2: Disallow kernel profiling by users without CAP_SYS_ADMIN",
 				 target->system_wide ? "system-wide " : "");
 	case ENOENT:
 		return scnprintf(msg, size, "The %s event is not supported.",