diff options
648 files changed, 55451 insertions, 4315 deletions
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/cgroups/cgroups.txt b/Documentation/cgroups/cgroups.txt index 638bf17ff869..61dc0ec5c9a9 100644 --- a/Documentation/cgroups/cgroups.txt +++ b/Documentation/cgroups/cgroups.txt @@ -598,6 +598,15 @@ is completely unused; @cgrp->parent is still valid. (Note - can also be called for a newly-created cgroup if an error occurs after this subsystem's create() method has been called for the new cgroup). +int allow_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) +(cgroup_mutex held by caller) + +Called prior to moving a task into a cgroup; if the subsystem +returns an error, this will abort the attach operation. Used +to extend the permission checks - if all subsystems in a cgroup +return 0, the attach will be allowed to proceed, even if the +default permission check (root or same user) fails. + int can_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) (cgroup_mutex held by caller) diff --git a/Documentation/cpu-freq/governors.txt b/Documentation/cpu-freq/governors.txt index 219970ba54b7..875eecbed8c8 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/verity.txt b/Documentation/device-mapper/verity.txt index 9884681535ee..2929f6b1ccf1 100644 --- a/Documentation/device-mapper/verity.txt +++ b/Documentation/device-mapper/verity.txt @@ -10,7 +10,7 @@ Construction Parameters <version> <dev> <hash_dev> <data_block_size> <hash_block_size> <num_data_blocks> <hash_start_block> - <algorithm> <digest> <salt> + <algorithm> <digest> <salt> <mode> <version> This is the type of the on-disk hash format. @@ -62,6 +62,16 @@ Construction Parameters <salt> The hexadecimal encoding of the salt value. +<mode> + Optional. The mode of operation. + + 0 is the normal mode of operation where a corrupted block will result in an + I/O error. + + 1 is logging mode where corrupted blocks are logged and a uevent is sent to + notify user space. + + Theory of operation =================== diff --git a/Documentation/filesystems/porting b/Documentation/filesystems/porting index 4db22f6491e0..85a4a033bae7 100644 --- a/Documentation/filesystems/porting +++ b/Documentation/filesystems/porting @@ -445,3 +445,6 @@ object doesn't exist. It's remote/distributed ones that might care... [mandatory] FS_REVAL_DOT is gone; if you used to have it, add ->d_weak_revalidate() in your dentry operations instead. +-- +[mandatory] + vfs_readdir() is gone; switch to iterate_dir() instead diff --git a/Documentation/filesystems/proc.txt b/Documentation/filesystems/proc.txt index 954eab8c7fec..be35913d282b 100644 --- a/Documentation/filesystems/proc.txt +++ b/Documentation/filesystems/proc.txt @@ -369,6 +369,8 @@ is not associated with a file: [stack:1001] = the stack of the thread with tid 1001 [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. @@ -419,6 +421,7 @@ KernelPageSize: 4 kB MMUPageSize: 4 kB Locked: 374 kB VmFlags: rd ex mr mw me de +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 @@ -469,6 +472,9 @@ 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. @@ -484,6 +490,10 @@ To clear the bits for the file mapped pages associated with the process > echo 3 > /proc/PID/clear_refs Any other value written to /proc/PID/clear_refs will have no effect. +To reset the peak resident set size ("high water mark") to the process's +current value: + > echo 5 > /proc/PID/clear_refs + The /proc/pid/pagemap gives the PFN, which can be used to find the pageflags using /proc/kpageflags and number of times a page is mapped using /proc/kpagecount. For detailed explanation, see Documentation/vm/pagemap.txt. diff --git a/Documentation/filesystems/squashfs.txt b/Documentation/filesystems/squashfs.txt index 403c090aca39..e5274f84dc56 100644 --- a/Documentation/filesystems/squashfs.txt +++ b/Documentation/filesystems/squashfs.txt @@ -2,10 +2,10 @@ SQUASHFS 4.0 FILESYSTEM ======================= Squashfs is a compressed read-only filesystem for Linux. -It uses zlib/lzo/xz compression to compress files, inodes and directories. -Inodes in the system are very small and all blocks are packed to minimise -data overhead. Block sizes greater than 4K are supported up to a maximum -of 1Mbytes (default block size 128K). +It uses zlib, lz4, lzo, or xz compression to compress files, inodes and +directories. Inodes in the system are very small and all blocks are packed to +minimise data overhead. Block sizes greater than 4K are supported up to a +maximum of 1Mbytes (default block size 128K). Squashfs is intended for general read-only filesystem use, for archival use (i.e. in cases where a .tar.gz file may be used), and in constrained diff --git a/Documentation/networking/ip-sysctl.txt b/Documentation/networking/ip-sysctl.txt index a59ee432a98f..fbfc9e0d75e0 100644 --- a/Documentation/networking/ip-sysctl.txt +++ b/Documentation/networking/ip-sysctl.txt @@ -22,6 +22,15 @@ ip_no_pmtu_disc - BOOLEAN min_pmtu - INTEGER default 552 - minimum discovered Path MTU +fwmark_reflect - BOOLEAN + Controls the fwmark of kernel-generated IPv4 reply packets that are not + associated with a socket for example, TCP RSTs or ICMP echo replies). + If unset, these packets have a fwmark of zero. If set, they have the + fwmark of the packet they are replying to. Similarly affects the fwmark + used by internal routing lookups triggered by incoming packets, such as + the ones used for Path MTU Discovery. + Default: 0 + route/max_size - INTEGER Maximum number of routes allowed in the kernel. Increase this when using large numbers of interfaces and/or routes. @@ -468,6 +477,16 @@ 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 @@ -1093,6 +1112,15 @@ conf/all/forwarding - BOOLEAN proxy_ndp - BOOLEAN Do proxy ndp. +fwmark_reflect - BOOLEAN + Controls the fwmark of kernel-generated IPv6 reply packets that are not + associated with a socket for example, TCP RSTs or ICMPv6 echo replies). + If unset, these packets have a fwmark of zero. If set, they have the + fwmark of the packet they are replying to. Similarly affects the fwmark + used by internal routing lookups triggered by incoming packets, such as + the ones used for Path MTU Discovery. + Default: 0 + conf/interface/*: Change special settings per interface. @@ -1311,6 +1339,19 @@ ndisc_notify - BOOLEAN 1 - Generate unsolicited neighbour advertisements when device is brought up or hardware address changes. +optimistic_dad - BOOLEAN + Whether to perform Optimistic Duplicate Address Detection (RFC 4429). + 0: disabled (default) + 1: enabled + +use_optimistic - BOOLEAN + If enabled, do not classify optimistic addresses as deprecated during + source address selection. Preferred addresses will still be chosen + before optimistic addresses, subject to other ranking in the source + address selection algorithm. + 0: disabled (default) + 1: enabled + icmp/*: ratelimit - INTEGER Limit the maximal rates for sending ICMPv6 packets. 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/vm.txt b/Documentation/sysctl/vm.txt index dcc75a9ed919..b81fca90f7fe 100644 --- a/Documentation/sysctl/vm.txt +++ b/Documentation/sysctl/vm.txt @@ -29,6 +29,7 @@ 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 @@ -198,6 +199,21 @@ 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 is only useful when kernelcore= is specified at boot time to diff --git a/Documentation/trace/ftrace.txt b/Documentation/trace/ftrace.txt index bfe8c29b1f1d..29064c3bfcd1 100644 --- a/Documentation/trace/ftrace.txt +++ b/Documentation/trace/ftrace.txt @@ -2013,6 +2013,35 @@ 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/android/configs/README b/android/configs/README new file mode 100644 index 000000000000..8798731f8904 --- /dev/null +++ b/android/configs/README @@ -0,0 +1,15 @@ +The files in this directory are meant to be used as a base for an Android +kernel config. All devices should have the options in android-base.cfg enabled. +While not mandatory, the options in android-recommended.cfg enable advanced +Android features. + +Assuming you already have a minimalist defconfig for your device, a possible +way to enable these options would be: + + ARCH=<arch> scripts/kconfig/merge_config.sh <path_to>/<device>_defconfig android/configs/android-base.cfg android/configs/android-recommended.cfg + +This will generate a .config that can then be used to save a new defconfig or +compile a new kernel with Android features enabled. + +Because there is no tool to consistently generate these config fragments, +lets keep them alphabetically sorted instead of random. diff --git a/android/configs/android-base.cfg b/android/configs/android-base.cfg new file mode 100644 index 000000000000..1bee5d614d1b --- /dev/null +++ b/android/configs/android-base.cfg @@ -0,0 +1,152 @@ +# KEEP ALPHABETICALLY SORTED +# CONFIG_DEVKMEM is not set +# CONFIG_DEVMEM is not set +# CONFIG_INET_LRO is not set +# CONFIG_MODULES is not set +# CONFIG_OABI_COMPAT is not set +CONFIG_ANDROID=y +CONFIG_ANDROID_BINDER_IPC=y +CONFIG_ANDROID_LOW_MEMORY_KILLER=y +CONFIG_ARMV7_COMPAT=y +CONFIG_ASHMEM=y +CONFIG_AUDIT=y +CONFIG_BLK_DEV_DM=y +CONFIG_BLK_DEV_INITRD=y +CONFIG_CGROUPS=y +CONFIG_CGROUP_CPUACCT=y +CONFIG_CGROUP_DEBUG=y +CONFIG_CGROUP_FREEZER=y +CONFIG_CGROUP_SCHED=y +CONFIG_DM_CRYPT=y +CONFIG_DM_VERITY=y +CONFIG_EMBEDDED=y +CONFIG_FB=y +CONFIG_HIGH_RES_TIMERS=y +CONFIG_INET6_AH=y +CONFIG_INET6_ESP=y +CONFIG_INET6_IPCOMP=y +CONFIG_INET=y +CONFIG_INET_ESP=y +CONFIG_INET_XFRM_MODE_TUNNEL=y +CONFIG_IP6_NF_FILTER=y +CONFIG_IP6_NF_IPTABLES=y +CONFIG_IP6_NF_MANGLE=y +CONFIG_IP6_NF_RAW=y +CONFIG_IP6_NF_TARGET_REJECT=y +CONFIG_IP6_NF_TARGET_REJECT_SKERR=y +CONFIG_IPV6_MIP6=y +CONFIG_IPV6_MULTIPLE_TABLES=y +CONFIG_IPV6_OPTIMISTIC_DAD=y +CONFIG_IPV6_PRIVACY=y +CONFIG_IPV6_ROUTER_PREF=y +CONFIG_IPV6_ROUTE_INFO=y +CONFIG_IP_ADVANCED_ROUTER=y +CONFIG_IP_MULTIPLE_TABLES=y +CONFIG_IP_NF_ARPFILTER=y +CONFIG_IP_NF_ARPTABLES=y +CONFIG_IP_NF_ARP_MANGLE=y +CONFIG_IP_NF_FILTER=y +CONFIG_IP_NF_IPTABLES=y +CONFIG_IP_NF_MANGLE=y +CONFIG_IP_NF_MATCH_AH=y +CONFIG_IP_NF_MATCH_ECN=y +CONFIG_IP_NF_MATCH_TTL=y +CONFIG_IP_NF_RAW=y +CONFIG_IP_NF_SECURITY=y +CONFIG_IP_NF_TARGET_MASQUERADE=y +CONFIG_IP_NF_TARGET_NETMAP=y +CONFIG_IP_NF_TARGET_REDIRECT=y +CONFIG_IP_NF_TARGET_REJECT=y +CONFIG_IP_NF_TARGET_REJECT_SKERR=y +CONFIG_NET=y +CONFIG_NETDEVICES=y +CONFIG_NETFILTER=y +CONFIG_NETFILTER_TPROXY=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_PKTTYPE=y +CONFIG_NETFILTER_XT_MATCH_POLICY=y +CONFIG_NETFILTER_XT_MATCH_QTAGUID=y +CONFIG_NETFILTER_XT_MATCH_QUOTA2=y +CONFIG_NETFILTER_XT_MATCH_QUOTA2_LOG=y +CONFIG_NETFILTER_XT_MATCH_QUOTA=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_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_SECMARK=y +CONFIG_NETFILTER_XT_TARGET_TCPMSS=y +CONFIG_NETFILTER_XT_TARGET_TPROXY=y +CONFIG_NETFILTER_XT_TARGET_TRACE=y +CONFIG_NET_CLS_ACT=y +CONFIG_NET_CLS_U32=y +CONFIG_NET_EMATCH=y +CONFIG_NET_EMATCH_U32=y +CONFIG_NET_KEY=y +CONFIG_NET_SCHED=y +CONFIG_NET_SCH_HTB=y +CONFIG_NF_CONNTRACK=y +CONFIG_NF_CONNTRACK_AMANDA=y +CONFIG_NF_CONNTRACK_EVENTS=y +CONFIG_NF_CONNTRACK_FTP=y +CONFIG_NF_CONNTRACK_H323=y +CONFIG_NF_CONNTRACK_IPV4=y +CONFIG_NF_CONNTRACK_IPV6=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_SECMARK=y +CONFIG_NF_CONNTRACK_TFTP=y +CONFIG_NF_CT_NETLINK=y +CONFIG_NF_CT_PROTO_DCCP=y +CONFIG_NF_CT_PROTO_SCTP=y +CONFIG_NF_CT_PROTO_UDPLITE=y +CONFIG_NF_NAT=y +CONFIG_NO_HZ=y +CONFIG_PACKET=y +CONFIG_PM_AUTOSLEEP=y +CONFIG_PM_WAKELOCKS=y +CONFIG_PPP=y +CONFIG_PPPOLAC=y +CONFIG_PPPOPNS=y +CONFIG_PPP_BSDCOMP=y +CONFIG_PPP_DEFLATE=y +CONFIG_PPP_MPPE=y +CONFIG_PREEMPT=y +CONFIG_RESOURCE_COUNTERS=y +CONFIG_RTC_CLASS=y +CONFIG_RT_GROUP_SCHED=y +CONFIG_SECURITY=y +CONFIG_SECURITY_NETWORK=y +CONFIG_SECURITY_SELINUX=y +CONFIG_SND=y +CONFIG_SOUND=y +CONFIG_STAGING=y +CONFIG_SWITCH=y +CONFIG_SYNC=y +CONFIG_SYSVIPC=y +CONFIG_TUN=y +CONFIG_UNIX=y +CONFIG_USB_GADGET=y +CONFIG_USB_G_ANDROID=y +CONFIG_USB_OTG_WAKELOCK=y +CONFIG_XFRM_USER=y diff --git a/android/configs/android-recommended.cfg b/android/configs/android-recommended.cfg new file mode 100644 index 000000000000..960b9de2860d --- /dev/null +++ b/android/configs/android-recommended.cfg @@ -0,0 +1,121 @@ +# KEEP ALPHABETICALLY SORTED +# CONFIG_CORE_DUMP_DEFAULT_ELF_HEADERS is not set +# CONFIG_INPUT_MOUSE is not set +# CONFIG_LEGACY_PTYS is not set +# CONFIG_NF_CONNTRACK_SIP is not set +# CONFIG_PM_WAKELOCKS_GC is not set +# CONFIG_VT is not set +CONFIG_ANDROID_TIMED_GPIO=y +CONFIG_BACKLIGHT_LCD_SUPPORT=y +CONFIG_BLK_DEV_LOOP=y +CONFIG_BLK_DEV_RAM=y +CONFIG_BLK_DEV_RAM_SIZE=8192 +CONFIG_COMPACTION=y +CONFIG_DM_UEVENT=y +CONFIG_DRAGONRISE_FF=y +CONFIG_ENABLE_DEFAULT_TRACERS=y +CONFIG_EXT4_FS=y +CONFIG_EXT4_FS_SECURITY=y +CONFIG_FUSE_FS=y +CONFIG_GREENASIA_FF=y +CONFIG_HIDRAW=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_CYPRESS=y +CONFIG_HID_DRAGONRISE=y +CONFIG_HID_ELECOM=y +CONFIG_HID_EMS_FF=y +CONFIG_HID_EZKEY=y +CONFIG_HID_GREENASIA=y +CONFIG_HID_GYRATION=y +CONFIG_HID_HOLTEK=y +CONFIG_HID_KENSINGTON=y +CONFIG_HID_KEYTOUCH=y +CONFIG_HID_KYE=y +CONFIG_HID_LCPOWER=y +CONFIG_HID_LOGITECH=y +CONFIG_HID_LOGITECH_DJ=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_HID_PETALYNX=y +CONFIG_HID_PICOLCD=y +CONFIG_HID_PRIMAX=y +CONFIG_HID_PRODIKEYS=y +CONFIG_HID_ROCCAT=y +CONFIG_HID_SAITEK=y +CONFIG_HID_SAMSUNG=y +CONFIG_HID_SMARTJOYPLUS=y +CONFIG_HID_SONY=y +CONFIG_HID_SPEEDLINK=y +CONFIG_HID_SUNPLUS=y +CONFIG_HID_THRUSTMASTER=y +CONFIG_HID_TIVO=y +CONFIG_HID_TOPSEED=y +CONFIG_HID_TWINHAN=y +CONFIG_HID_UCLOGIC=y +CONFIG_HID_WACOM=y +CONFIG_HID_WALTOP=y +CONFIG_HID_WIIMOTE=y +CONFIG_HID_ZEROPLUS=y +CONFIG_HID_ZYDACRON=y +CONFIG_INPUT_EVDEV=y +CONFIG_INPUT_GPIO=y +CONFIG_INPUT_JOYSTICK=y +CONFIG_INPUT_KEYCHORD=y +CONFIG_INPUT_KEYRESET=y +CONFIG_INPUT_MISC=y +CONFIG_INPUT_TABLET=y +CONFIG_INPUT_UINPUT=y +CONFIG_ION=y +CONFIG_JOYSTICK_XPAD=y +CONFIG_JOYSTICK_XPAD_FF=y +CONFIG_JOYSTICK_XPAD_LEDS=y +CONFIG_KALLSYMS_ALL=y +CONFIG_KSM=y +CONFIG_LOGIG940_FF=y +CONFIG_LOGIRUMBLEPAD2_FF=y +CONFIG_LOGITECH_FF=y +CONFIG_MD=y +CONFIG_MEDIA_SUPPORT=y +CONFIG_MSDOS_FS=y +CONFIG_PANIC_TIMEOUT=5 +CONFIG_PANTHERLORD_FF=y +CONFIG_PERF_EVENTS=y +CONFIG_PM_DEBUG=y +CONFIG_PM_RUNTIME=y +CONFIG_PM_WAKELOCKS_LIMIT=0 +CONFIG_POWER_SUPPLY=y +CONFIG_PSTORE=y +CONFIG_PSTORE_CONSOLE=y +CONFIG_PSTORE_RAM=y +CONFIG_SCHEDSTATS=y +CONFIG_SMARTJOYPLUS_FF=y +CONFIG_SND=y +CONFIG_SOUND=y +CONFIG_SUSPEND_TIME=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_TABLET_USB_WACOM=y +CONFIG_TIMER_STATS=y +CONFIG_TMPFS=y +CONFIG_TMPFS_POSIX_ACL=y +CONFIG_UHID=y +CONFIG_UID_STAT=y +CONFIG_USB_ANNOUNCE_NEW_DEVICES=y +CONFIG_USB_EHCI_HCD=y +CONFIG_USB_HIDDEV=y +CONFIG_USB_USBNET=y +CONFIG_VFAT_FS=y diff --git a/arch/Kconfig b/arch/Kconfig index 00e3702ec79b..4c0a1d03ae0d 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -331,6 +331,7 @@ config HAVE_ARCH_SECCOMP_FILTER - secure_computing is called from a ptrace_event()-safe context - secure_computing return value is checked and a return value of -1 results in the system call being skipped immediately. + - seccomp syscall wired up config SECCOMP_FILTER def_bool y diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index b9e37ad6fa19..1402fcc11c2c 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -96,6 +96,7 @@ struct osf_dirent { }; struct osf_dirent_callback { + struct dir_context ctx; struct osf_dirent __user *dirent; long __user *basep; unsigned int count; @@ -146,17 +147,17 @@ SYSCALL_DEFINE4(osf_getdirentries, unsigned int, fd, { int error; struct fd arg = fdget(fd); - struct osf_dirent_callback buf; + struct osf_dirent_callback buf = { + .ctx.actor = osf_filldir, + .dirent = dirent, + .basep = basep, + .count = count + }; if (!arg.file) return -EBADF; - buf.dirent = dirent; - buf.basep = basep; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(arg.file, osf_filldir, &buf); + error = iterate_dir(arg.file, &buf.ctx); if (error >= 0) error = buf.error; if (count != buf.count) diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 9422f298f32f..1a86004f3f27 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -1958,6 +1958,15 @@ 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" @@ -1987,6 +1996,21 @@ 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 DTB to + be built by default (instead of a standalone zImage.) The image + will built in arch/arm/boot/zImage-dtb.<dtb name> + +config BUILD_ARM_APPENDED_DTB_IMAGE_NAME + string "Default dtb name" + depends on BUILD_ARM_APPENDED_DTB_IMAGE + help + name of the dtb 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 @@ -2336,6 +2360,13 @@ config NEON Say Y to include support code for NEON, the ARMv7 Advanced SIMD Extension. +config KERNEL_MODE_NEON + bool "Support for NEON in kernel mode" + default n + depends on NEON + help + Say Y to include support for NEON in kernel mode. + endmenu menu "Userspace binary formats" diff --git a/arch/arm/Kconfig.debug b/arch/arm/Kconfig.debug index 5fdb6dbc5f89..4d1793474c9c 100644 --- a/arch/arm/Kconfig.debug +++ b/arch/arm/Kconfig.debug @@ -63,6 +63,27 @@ config DEBUG_USER 8 - SIGSEGV faults 16 - SIGBUS faults +config DEBUG_RODATA + bool "Write protect kernel text section" + default n + depends on DEBUG_KERNEL && MMU + ---help--- + Mark the kernel text section as write-protected in the pagetables, + in order to catch accidental (and incorrect) writes to such const + data. This will cause the size of the kernel, plus up to 4MB, to + be mapped as pages instead of sections, which will increase TLB + pressure. + If in doubt, say "N". + +config DEBUG_RODATA_TEST + bool "Testcase for the DEBUG_RODATA feature" + depends on DEBUG_RODATA + default n + ---help--- + This option enables a testcase for the DEBUG_RODATA + feature. + If in doubt, say "N" + # These options are only for real kernel hackers who want to get their hands dirty. config DEBUG_LL bool "Kernel low-level debugging functions (read help!)" diff --git a/arch/arm/Makefile b/arch/arm/Makefile index 70bc19e2274f..9d36200374f0 100644 --- a/arch/arm/Makefile +++ b/arch/arm/Makefile @@ -265,6 +265,8 @@ libs-y := arch/arm/lib/ $(libs-y) # Default target when executing plain make ifeq ($(CONFIG_XIP_KERNEL),y) KBUILD_IMAGE := xipImage +else ifeq ($(CONFIG_BUILD_ARM_APPENDED_DTB_IMAGE),y) +KBUILD_IMAGE := zImage-dtb.$(CONFIG_BUILD_ARM_APPENDED_DTB_IMAGE_NAME) else KBUILD_IMAGE := zImage endif diff --git a/arch/arm/boot/Makefile b/arch/arm/boot/Makefile index 84aa2caf07ed..085bb96493a3 100644 --- a/arch/arm/boot/Makefile +++ b/arch/arm/boot/Makefile @@ -14,6 +14,7 @@ ifneq ($(MACHINE),) include $(srctree)/$(MACHINE)/Makefile.boot endif +include $(srctree)/arch/arm/boot/dts/Makefile # Note: the following conditions must always be true: # ZRELADDR == virt_to_phys(PAGE_OFFSET + TEXT_OFFSET) diff --git a/arch/arm/boot/compressed/head.S b/arch/arm/boot/compressed/head.S index f6e34be012ff..a8264aa9b03a 100644 --- a/arch/arm/boot/compressed/head.S +++ b/arch/arm/boot/compressed/head.S @@ -714,6 +714,8 @@ __armv7_mmu_cache_on: bic r6, r6, #1 << 31 @ 32-bit translation system bic r6, r6, #3 << 0 @ use only ttbr0 mcrne p15, 0, r3, c2, c0, 0 @ load page table pointer + mcrne p15, 0, r0, c8, c7, 0 @ flush I,D TLBs + mcr p15, 0, r0, c7, c5, 4 @ ISB mcrne p15, 0, r1, c3, c0, 0 @ load domain access control mcrne p15, 0, r6, c2, c0, 2 @ load ttb control #endif diff --git a/arch/arm/common/Kconfig b/arch/arm/common/Kconfig index 9353184d730d..ce01364a96e3 100644 --- a/arch/arm/common/Kconfig +++ b/arch/arm/common/Kconfig @@ -17,3 +17,7 @@ config SHARP_PARAM config SHARP_SCOOP bool + +config FIQ_GLUE + bool + select FIQ diff --git a/arch/arm/common/Makefile b/arch/arm/common/Makefile index 462cd580fc2d..505c479202b6 100644 --- a/arch/arm/common/Makefile +++ b/arch/arm/common/Makefile @@ -4,6 +4,7 @@ obj-y += firmware.o +obj-$(CONFIG_FIQ_GLUE) += fiq_glue.o fiq_glue_setup.o obj-$(CONFIG_ICST) += icst.o obj-$(CONFIG_SA1111) += sa1111.o obj-$(CONFIG_PCI_HOST_VIA82C505) += via82c505.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/crypto/.gitignore b/arch/arm/crypto/.gitignore new file mode 100644 index 000000000000..6231d36b3635 --- /dev/null +++ b/arch/arm/crypto/.gitignore @@ -0,0 +1 @@ +aesbs-core.S diff --git a/arch/arm/crypto/Makefile b/arch/arm/crypto/Makefile index a2c83851bc90..2cee53b27238 100644 --- a/arch/arm/crypto/Makefile +++ b/arch/arm/crypto/Makefile @@ -3,7 +3,27 @@ # obj-$(CONFIG_CRYPTO_AES_ARM) += aes-arm.o +obj-$(CONFIG_CRYPTO_AES_ARM_BS) += aes-arm-bs.o obj-$(CONFIG_CRYPTO_SHA1_ARM) += sha1-arm.o +obj-$(CONFIG_CRYPTO_SHA1_ARM_NEON) += sha1-arm-neon.o +obj-$(CONFIG_CRYPTO_SHA256_ARM) += sha256-arm.o +obj-$(CONFIG_CRYPTO_SHA512_ARM_NEON) += sha512-arm-neon.o -aes-arm-y := aes-armv4.o aes_glue.o -sha1-arm-y := sha1-armv4-large.o sha1_glue.o +aes-arm-y := aes-armv4.o aes_glue.o +aes-arm-bs-y := aesbs-core.o aesbs-glue.o +sha1-arm-y := sha1-armv4-large.o sha1_glue.o +sha1-arm-neon-y := sha1-armv7-neon.o sha1_neon_glue.o +sha256-arm-neon-$(CONFIG_KERNEL_MODE_NEON) := sha256_neon_glue.o +sha256-arm-y := sha256-core.o sha256_glue.o $(sha256-arm-neon-y) +sha512-arm-neon-y := sha512-armv7-neon.o sha512_neon_glue.o + +quiet_cmd_perl = PERL $@ + cmd_perl = $(PERL) $(<) > $(@) + +$(src)/aesbs-core.S_shipped: $(src)/bsaes-armv7.pl + $(call cmd,perl) + +$(src)/sha256-core.S_shipped: $(src)/sha256-armv4.pl + $(call cmd,perl) + +.PRECIOUS: $(obj)/aesbs-core.S $(obj)/sha256-core.S diff --git a/arch/arm/crypto/aes_glue.c b/arch/arm/crypto/aes_glue.c index e73ec2ab1316..0409b8f89782 100644 --- a/arch/arm/crypto/aes_glue.c +++ b/arch/arm/crypto/aes_glue.c @@ -6,22 +6,12 @@ #include <linux/crypto.h> #include <crypto/aes.h> -#define AES_MAXNR 14 +#include "aes_glue.h" -typedef struct { - unsigned int rd_key[4 *(AES_MAXNR + 1)]; - int rounds; -} AES_KEY; - -struct AES_CTX { - AES_KEY enc_key; - AES_KEY dec_key; -}; - -asmlinkage void AES_encrypt(const u8 *in, u8 *out, AES_KEY *ctx); -asmlinkage void AES_decrypt(const u8 *in, u8 *out, AES_KEY *ctx); -asmlinkage int private_AES_set_decrypt_key(const unsigned char *userKey, const int bits, AES_KEY *key); -asmlinkage int private_AES_set_encrypt_key(const unsigned char *userKey, const int bits, AES_KEY *key); +EXPORT_SYMBOL(AES_encrypt); +EXPORT_SYMBOL(AES_decrypt); +EXPORT_SYMBOL(private_AES_set_encrypt_key); +EXPORT_SYMBOL(private_AES_set_decrypt_key); static void aes_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) { @@ -81,7 +71,7 @@ static struct crypto_alg aes_alg = { .cipher = { .cia_min_keysize = AES_MIN_KEY_SIZE, .cia_max_keysize = AES_MAX_KEY_SIZE, - .cia_setkey = aes_set_key, + .cia_setkey = aes_set_key, .cia_encrypt = aes_encrypt, .cia_decrypt = aes_decrypt } diff --git a/arch/arm/crypto/aes_glue.h b/arch/arm/crypto/aes_glue.h new file mode 100644 index 000000000000..cca3e51eb606 --- /dev/null +++ b/arch/arm/crypto/aes_glue.h @@ -0,0 +1,19 @@ + +#define AES_MAXNR 14 + +struct AES_KEY { + unsigned int rd_key[4 * (AES_MAXNR + 1)]; + int rounds; +}; + +struct AES_CTX { + struct AES_KEY enc_key; + struct AES_KEY dec_key; +}; + +asmlinkage void AES_encrypt(const u8 *in, u8 *out, struct AES_KEY *ctx); +asmlinkage void AES_decrypt(const u8 *in, u8 *out, struct AES_KEY *ctx); +asmlinkage int private_AES_set_decrypt_key(const unsigned char *userKey, + const int bits, struct AES_KEY *key); +asmlinkage int private_AES_set_encrypt_key(const unsigned char *userKey, + const int bits, struct AES_KEY *key); diff --git a/arch/arm/crypto/aesbs-core.S_shipped b/arch/arm/crypto/aesbs-core.S_shipped new file mode 100644 index 000000000000..71e5fc7cfb18 --- /dev/null +++ b/arch/arm/crypto/aesbs-core.S_shipped @@ -0,0 +1,2544 @@ + +@ ==================================================================== +@ Written by Andy Polyakov <appro@openssl.org> for the OpenSSL +@ project. The module is, however, dual licensed under OpenSSL and +@ CRYPTOGAMS licenses depending on where you obtain it. For further +@ details see http://www.openssl.org/~appro/cryptogams/. +@ +@ Specific modes and adaptation for Linux kernel by Ard Biesheuvel +@ <ard.biesheuvel@linaro.org>. Permission to use under GPL terms is +@ granted. +@ ==================================================================== + +@ Bit-sliced AES for ARM NEON +@ +@ February 2012. +@ +@ This implementation is direct adaptation of bsaes-x86_64 module for +@ ARM NEON. Except that this module is endian-neutral [in sense that +@ it can be compiled for either endianness] by courtesy of vld1.8's +@ neutrality. Initial version doesn't implement interface to OpenSSL, +@ only low-level primitives and unsupported entry points, just enough +@ to collect performance results, which for Cortex-A8 core are: +@ +@ encrypt 19.5 cycles per byte processed with 128-bit key +@ decrypt 22.1 cycles per byte processed with 128-bit key +@ key conv. 440 cycles per 128-bit key/0.18 of 8x block +@ +@ Snapdragon S4 encrypts byte in 17.6 cycles and decrypts in 19.7, +@ which is [much] worse than anticipated (for further details see +@ http://www.openssl.org/~appro/Snapdragon-S4.html). +@ +@ Cortex-A15 manages in 14.2/16.1 cycles [when integer-only code +@ manages in 20.0 cycles]. +@ +@ When comparing to x86_64 results keep in mind that NEON unit is +@ [mostly] single-issue and thus can't [fully] benefit from +@ instruction-level parallelism. And when comparing to aes-armv4 +@ results keep in mind key schedule conversion overhead (see +@ bsaes-x86_64.pl for further details)... +@ +@ <appro@openssl.org> + +@ April-August 2013 +@ +@ Add CBC, CTR and XTS subroutines, adapt for kernel use. +@ +@ <ard.biesheuvel@linaro.org> + +#ifndef __KERNEL__ +# include "arm_arch.h" + +# define VFP_ABI_PUSH vstmdb sp!,{d8-d15} +# define VFP_ABI_POP vldmia sp!,{d8-d15} +# define VFP_ABI_FRAME 0x40 +#else +# define VFP_ABI_PUSH +# define VFP_ABI_POP +# define VFP_ABI_FRAME 0 +# define BSAES_ASM_EXTENDED_KEY +# define XTS_CHAIN_TWEAK +# define __ARM_ARCH__ 7 +#endif + +#ifdef __thumb__ +# define adrl adr +#endif + +#if __ARM_ARCH__>=7 +.text +.syntax unified @ ARMv7-capable assembler is expected to handle this +#ifdef __thumb2__ +.thumb +#else +.code 32 +#endif + +.fpu neon + +.type _bsaes_decrypt8,%function +.align 4 +_bsaes_decrypt8: + adr r6,_bsaes_decrypt8 + vldmia r4!, {q9} @ round 0 key + add r6,r6,#.LM0ISR-_bsaes_decrypt8 + + vldmia r6!, {q8} @ .LM0ISR + veor q10, q0, q9 @ xor with round0 key + veor q11, q1, q9 + vtbl.8 d0, {q10}, d16 + vtbl.8 d1, {q10}, d17 + veor q12, q2, q9 + vtbl.8 d2, {q11}, d16 + vtbl.8 d3, {q11}, d17 + veor q13, q3, q9 + vtbl.8 d4, {q12}, d16 + vtbl.8 d5, {q12}, d17 + veor q14, q4, q9 + vtbl.8 d6, {q13}, d16 + vtbl.8 d7, {q13}, d17 + veor q15, q5, q9 + vtbl.8 d8, {q14}, d16 + vtbl.8 d9, {q14}, d17 + veor q10, q6, q9 + vtbl.8 d10, {q15}, d16 + vtbl.8 d11, {q15}, d17 + veor q11, q7, q9 + vtbl.8 d12, {q10}, d16 + vtbl.8 d13, {q10}, d17 + vtbl.8 d14, {q11}, d16 + vtbl.8 d15, {q11}, d17 + vmov.i8 q8,#0x55 @ compose .LBS0 + vmov.i8 q9,#0x33 @ compose .LBS1 + vshr.u64 q10, q6, #1 + vshr.u64 q11, q4, #1 + veor q10, q10, q7 + veor q11, q11, q5 + vand q10, q10, q8 + vand q11, q11, q8 + veor q7, q7, q10 + vshl.u64 q10, q10, #1 + veor q5, q5, q11 + vshl.u64 q11, q11, #1 + veor q6, q6, q10 + veor q4, q4, q11 + vshr.u64 q10, q2, #1 + vshr.u64 q11, q0, #1 + veor q10, q10, q3 + veor q11, q11, q1 + vand q10, q10, q8 + vand q11, q11, q8 + veor q3, q3, q10 + vshl.u64 q10, q10, #1 + veor q1, q1, q11 + vshl.u64 q11, q11, #1 + veor q2, q2, q10 + veor q0, q0, q11 + vmov.i8 q8,#0x0f @ compose .LBS2 + vshr.u64 q10, q5, #2 + vshr.u64 q11, q4, #2 + veor q10, q10, q7 + veor q11, q11, q6 + vand q10, q10, q9 + vand q11, q11, q9 + veor q7, q7, q10 + vshl.u64 q10, q10, #2 + veor q6, q6, q11 + vshl.u64 q11, q11, #2 + veor q5, q5, q10 + veor q4, q4, q11 + vshr.u64 q10, q1, #2 + vshr.u64 q11, q0, #2 + veor q10, q10, q3 + veor q11, q11, q2 + vand q10, q10, q9 + vand q11, q11, q9 + veor q3, q3, q10 + vshl.u64 q10, q10, #2 + veor q2, q2, q11 + vshl.u64 q11, q11, #2 + veor q1, q1, q10 + veor q0, q0, q11 + vshr.u64 q10, q3, #4 + vshr.u64 q11, q2, #4 + veor q10, q10, q7 + veor q11, q11, q6 + vand q10, q10, q8 + vand q11, q11, q8 + veor q7, q7, q10 + vshl.u64 q10, q10, #4 + veor q6, q6, q11 + vshl.u64 q11, q11, #4 + veor q3, q3, q10 + veor q2, q2, q11 + vshr.u64 q10, q1, #4 + vshr.u64 q11, q0, #4 + veor q10, q10, q5 + veor q11, q11, q4 + vand q10, q10, q8 + vand q11, q11, q8 + veor q5, q5, q10 + vshl.u64 q10, q10, #4 + veor q4, q4, q11 + vshl.u64 q11, q11, #4 + veor q1, q1, q10 + veor q0, q0, q11 + sub r5,r5,#1 + b .Ldec_sbox +.align 4 +.Ldec_loop: + vldmia r4!, {q8-q11} + veor q8, q8, q0 + veor q9, q9, q1 + vtbl.8 d0, {q8}, d24 + vtbl.8 d1, {q8}, d25 + vldmia r4!, {q8} + veor q10, q10, q2 + vtbl.8 d2, {q9}, d24 + vtbl.8 d3, {q9}, d25 + vldmia r4!, {q9} + veor q11, q11, q3 + vtbl.8 d4, {q10}, d24 + vtbl.8 d5, {q10}, d25 + vldmia r4!, {q10} + vtbl.8 d6, {q11}, d24 + vtbl.8 d7, {q11}, d25 + vldmia r4!, {q11} + veor q8, q8, q4 + veor q9, q9, q5 + vtbl.8 d8, {q8}, d24 + vtbl.8 d9, {q8}, d25 + veor q10, q10, q6 + vtbl.8 d10, {q9}, d24 + vtbl.8 d11, {q9}, d25 + veor q11, q11, q7 + vtbl.8 d12, {q10}, d24 + vtbl.8 d13, {q10}, d25 + vtbl.8 d14, {q11}, d24 + vtbl.8 d15, {q11}, d25 +.Ldec_sbox: + veor q1, q1, q4 + veor q3, q3, q4 + + veor q4, q4, q7 + veor q1, q1, q6 + veor q2, q2, q7 + veor q6, q6, q4 + + veor q0, q0, q1 + veor q2, q2, q5 + veor q7, q7, q6 + veor q3, q3, q0 + veor q5, q5, q0 + veor q1, q1, q3 + veor q11, q3, q0 + veor q10, q7, q4 + veor q9, q1, q6 + veor q13, q4, q0 + vmov q8, q10 + veor q12, q5, q2 + + vorr q10, q10, q9 + veor q15, q11, q8 + vand q14, q11, q12 + vorr q11, q11, q12 + veor q12, q12, q9 + vand q8, q8, q9 + veor q9, q6, q2 + vand q15, q15, q12 + vand q13, q13, q9 + veor q9, q3, q7 + veor q12, q1, q5 + veor q11, q11, q13 + veor q10, q10, q13 + vand q13, q9, q12 + vorr q9, q9, q12 + veor q11, q11, q15 + veor q8, q8, q13 + veor q10, q10, q14 + veor q9, q9, q15 + veor q8, q8, q14 + vand q12, q4, q6 + veor q9, q9, q14 + vand q13, q0, q2 + vand q14, q7, q1 + vorr q15, q3, q5 + veor q11, q11, q12 + veor q9, q9, q14 + veor q8, q8, q15 + veor q10, q10, q13 + + @ Inv_GF16 0, 1, 2, 3, s0, s1, s2, s3 + + @ new smaller inversion + + vand q14, q11, q9 + vmov q12, q8 + + veor q13, q10, q14 + veor q15, q8, q14 + veor q14, q8, q14 @ q14=q15 + + vbsl q13, q9, q8 + vbsl q15, q11, q10 + veor q11, q11, q10 + + vbsl q12, q13, q14 + vbsl q8, q14, q13 + + vand q14, q12, q15 + veor q9, q9, q8 + + veor q14, q14, q11 + veor q12, q5, q2 + veor q8, q1, q6 + veor q10, q15, q14 + vand q10, q10, q5 + veor q5, q5, q1 + vand q11, q1, q15 + vand q5, q5, q14 + veor q1, q11, q10 + veor q5, q5, q11 + veor q15, q15, q13 + veor q14, q14, q9 + veor q11, q15, q14 + veor q10, q13, q9 + vand q11, q11, q12 + vand q10, q10, q2 + veor q12, q12, q8 + veor q2, q2, q6 + vand q8, q8, q15 + vand q6, q6, q13 + vand q12, q12, q14 + vand q2, q2, q9 + veor q8, q8, q12 + veor q2, q2, q6 + veor q12, q12, q11 + veor q6, q6, q10 + veor q5, q5, q12 + veor q2, q2, q12 + veor q1, q1, q8 + veor q6, q6, q8 + + veor q12, q3, q0 + veor q8, q7, q4 + veor q11, q15, q14 + veor q10, q13, q9 + vand q11, q11, q12 + vand q10, q10, q0 + veor q12, q12, q8 + veor q0, q0, q4 + vand q8, q8, q15 + vand q4, q4, q13 + vand q12, q12, q14 + vand q0, q0, q9 + veor q8, q8, q12 + veor q0, q0, q4 + veor q12, q12, q11 + veor q4, q4, q10 + veor q15, q15, q13 + veor q14, q14, q9 + veor q10, q15, q14 + vand q10, q10, q3 + veor q3, q3, q7 + vand q11, q7, q15 + vand q3, q3, q14 + veor q7, q11, q10 + veor q3, q3, q11 + veor q3, q3, q12 + veor q0, q0, q12 + veor q7, q7, q8 + veor q4, q4, q8 + veor q1, q1, q7 + veor q6, q6, q5 + + veor q4, q4, q1 + veor q2, q2, q7 + veor q5, q5, q7 + veor q4, q4, q2 + veor q7, q7, q0 + veor q4, q4, q5 + veor q3, q3, q6 + veor q6, q6, q1 + veor q3, q3, q4 + + veor q4, q4, q0 + veor q7, q7, q3 + subs r5,r5,#1 + bcc .Ldec_done + @ multiplication by 0x05-0x00-0x04-0x00 + vext.8 q8, q0, q0, #8 + vext.8 q14, q3, q3, #8 + vext.8 q15, q5, q5, #8 + veor q8, q8, q0 + vext.8 q9, q1, q1, #8 + veor q14, q14, q3 + vext.8 q10, q6, q6, #8 + veor q15, q15, q5 + vext.8 q11, q4, q4, #8 + veor q9, q9, q1 + vext.8 q12, q2, q2, #8 + veor q10, q10, q6 + vext.8 q13, q7, q7, #8 + veor q11, q11, q4 + veor q12, q12, q2 + veor q13, q13, q7 + + veor q0, q0, q14 + veor q1, q1, q14 + veor q6, q6, q8 + veor q2, q2, q10 + veor q4, q4, q9 + veor q1, q1, q15 + veor q6, q6, q15 + veor q2, q2, q14 + veor q7, q7, q11 + veor q4, q4, q14 + veor q3, q3, q12 + veor q2, q2, q15 + veor q7, q7, q15 + veor q5, q5, q13 + vext.8 q8, q0, q0, #12 @ x0 <<< 32 + vext.8 q9, q1, q1, #12 + veor q0, q0, q8 @ x0 ^ (x0 <<< 32) + vext.8 q10, q6, q6, #12 + veor q1, q1, q9 + vext.8 q11, q4, q4, #12 + veor q6, q6, q10 + vext.8 q12, q2, q2, #12 + veor q4, q4, q11 + vext.8 q13, q7, q7, #12 + veor q2, q2, q12 + vext.8 q14, q3, q3, #12 + veor q7, q7, q13 + vext.8 q15, q5, q5, #12 + veor q3, q3, q14 + + veor q9, q9, q0 + veor q5, q5, q15 + vext.8 q0, q0, q0, #8 @ (x0 ^ (x0 <<< 32)) <<< 64) + veor q10, q10, q1 + veor q8, q8, q5 + veor q9, q9, q5 + vext.8 q1, q1, q1, #8 + veor q13, q13, q2 + veor q0, q0, q8 + veor q14, q14, q7 + veor q1, q1, q9 + vext.8 q8, q2, q2, #8 + veor q12, q12, q4 + vext.8 q9, q7, q7, #8 + veor q15, q15, q3 + vext.8 q2, q4, q4, #8 + veor q11, q11, q6 + vext.8 q7, q5, q5, #8 + veor q12, q12, q5 + vext.8 q4, q3, q3, #8 + veor q11, q11, q5 + vext.8 q3, q6, q6, #8 + veor q5, q9, q13 + veor q11, q11, q2 + veor q7, q7, q15 + veor q6, q4, q14 + veor q4, q8, q12 + veor q2, q3, q10 + vmov q3, q11 + @ vmov q5, q9 + vldmia r6, {q12} @ .LISR + ite eq @ Thumb2 thing, sanity check in ARM + addeq r6,r6,#0x10 + bne .Ldec_loop + vldmia r6, {q12} @ .LISRM0 + b .Ldec_loop +.align 4 +.Ldec_done: + vmov.i8 q8,#0x55 @ compose .LBS0 + vmov.i8 q9,#0x33 @ compose .LBS1 + vshr.u64 q10, q3, #1 + vshr.u64 q11, q2, #1 + veor q10, q10, q5 + veor q11, q11, q7 + vand q10, q10, q8 + vand q11, q11, q8 + veor q5, q5, q10 + vshl.u64 q10, q10, #1 + veor q7, q7, q11 + vshl.u64 q11, q11, #1 + veor q3, q3, q10 + veor q2, q2, q11 + vshr.u64 q10, q6, #1 + vshr.u64 q11, q0, #1 + veor q10, q10, q4 + veor q11, q11, q1 + vand q10, q10, q8 + vand q11, q11, q8 + veor q4, q4, q10 + vshl.u64 q10, q10, #1 + veor q1, q1, q11 + vshl.u64 q11, q11, #1 + veor q6, q6, q10 + veor q0, q0, q11 + vmov.i8 q8,#0x0f @ compose .LBS2 + vshr.u64 q10, q7, #2 + vshr.u64 q11, q2, #2 + veor q10, q10, q5 + veor q11, q11, q3 + vand q10, q10, q9 + vand q11, q11, q9 + veor q5, q5, q10 + vshl.u64 q10, q10, #2 + veor q3, q3, q11 + vshl.u64 q11, q11, #2 + veor q7, q7, q10 + veor q2, q2, q11 + vshr.u64 q10, q1, #2 + vshr.u64 q11, q0, #2 + veor q10, q10, q4 + veor q11, q11, q6 + vand q10, q10, q9 + vand q11, q11, q9 + veor q4, q4, q10 + vshl.u64 q10, q10, #2 + veor q6, q6, q11 + vshl.u64 q11, q11, #2 + veor q1, q1, q10 + veor q0, q0, q11 + vshr.u64 q10, q4, #4 + vshr.u64 q11, q6, #4 + veor q10, q10, q5 + veor q11, q11, q3 + vand q10, q10, q8 + vand q11, q11, q8 + veor q5, q5, q10 + vshl.u64 q10, q10, #4 + veor q3, q3, q11 + vshl.u64 q11, q11, #4 + veor q4, q4, q10 + veor q6, q6, q11 + vshr.u64 q10, q1, #4 + vshr.u64 q11, q0, #4 + veor q10, q10, q7 + veor q11, q11, q2 + vand q10, q10, q8 + vand q11, q11, q8 + veor q7, q7, q10 + vshl.u64 q10, q10, #4 + veor q2, q2, q11 + vshl.u64 q11, q11, #4 + veor q1, q1, q10 + veor q0, q0, q11 + vldmia r4, {q8} @ last round key + veor q6, q6, q8 + veor q4, q4, q8 + veor q2, q2, q8 + veor q7, q7, q8 + veor q3, q3, q8 + veor q5, q5, q8 + veor q0, q0, q8 + veor q1, q1, q8 + bx lr +.size _bsaes_decrypt8,.-_bsaes_decrypt8 + +.type _bsaes_const,%object +.align 6 +_bsaes_const: +.LM0ISR: @ InvShiftRows constants + .quad 0x0a0e0206070b0f03, 0x0004080c0d010509 +.LISR: + .quad 0x0504070602010003, 0x0f0e0d0c080b0a09 +.LISRM0: + .quad 0x01040b0e0205080f, 0x0306090c00070a0d +.LM0SR: @ ShiftRows constants + .quad 0x0a0e02060f03070b, 0x0004080c05090d01 +.LSR: + .quad 0x0504070600030201, 0x0f0e0d0c0a09080b +.LSRM0: + .quad 0x0304090e00050a0f, 0x01060b0c0207080d +.LM0: + .quad 0x02060a0e03070b0f, 0x0004080c0105090d +.LREVM0SR: + .quad 0x090d01050c000408, 0x03070b0f060a0e02 +.asciz "Bit-sliced AES for NEON, CRYPTOGAMS by <appro@openssl.org>" +.align 6 +.size _bsaes_const,.-_bsaes_const + +.type _bsaes_encrypt8,%function +.align 4 +_bsaes_encrypt8: + adr r6,_bsaes_encrypt8 + vldmia r4!, {q9} @ round 0 key + sub r6,r6,#_bsaes_encrypt8-.LM0SR + + vldmia r6!, {q8} @ .LM0SR +_bsaes_encrypt8_alt: + veor q10, q0, q9 @ xor with round0 key + veor q11, q1, q9 + vtbl.8 d0, {q10}, d16 + vtbl.8 d1, {q10}, d17 + veor q12, q2, q9 + vtbl.8 d2, {q11}, d16 + vtbl.8 d3, {q11}, d17 + veor q13, q3, q9 + vtbl.8 d4, {q12}, d16 + vtbl.8 d5, {q12}, d17 + veor q14, q4, q9 + vtbl.8 d6, {q13}, d16 + vtbl.8 d7, {q13}, d17 + veor q15, q5, q9 + vtbl.8 d8, {q14}, d16 + vtbl.8 d9, {q14}, d17 + veor q10, q6, q9 + vtbl.8 d10, {q15}, d16 + vtbl.8 d11, {q15}, d17 + veor q11, q7, q9 + vtbl.8 d12, {q10}, d16 + vtbl.8 d13, {q10}, d17 + vtbl.8 d14, {q11}, d16 + vtbl.8 d15, {q11}, d17 +_bsaes_encrypt8_bitslice: + vmov.i8 q8,#0x55 @ compose .LBS0 + vmov.i8 q9,#0x33 @ compose .LBS1 + vshr.u64 q10, q6, #1 + vshr.u64 q11, q4, #1 + veor q10, q10, q7 + veor q11, q11, q5 + vand q10, q10, q8 + vand q11, q11, q8 + veor q7, q7, q10 + vshl.u64 q10, q10, #1 + veor q5, q5, q11 + vshl.u64 q11, q11, #1 + veor q6, q6, q10 + veor q4, q4, q11 + vshr.u64 q10, q2, #1 + vshr.u64 q11, q0, #1 + veor q10, q10, q3 + veor q11, q11, q1 + vand q10, q10, q8 + vand q11, q11, q8 + veor q3, q3, q10 + vshl.u64 q10, q10, #1 + veor q1, q1, q11 + vshl.u64 q11, q11, #1 + veor q2, q2, q10 + veor q0, q0, q11 + vmov.i8 q8,#0x0f @ compose .LBS2 + vshr.u64 q10, q5, #2 + vshr.u64 q11, q4, #2 + veor q10, q10, q7 + veor q11, q11, q6 + vand q10, q10, q9 + vand q11, q11, q9 + veor q7, q7, q10 + vshl.u64 q10, q10, #2 + veor q6, q6, q11 + vshl.u64 q11, q11, #2 + veor q5, q5, q10 + veor q4, q4, q11 + vshr.u64 q10, q1, #2 + vshr.u64 q11, q0, #2 + veor q10, q10, q3 + veor q11, q11, q2 + vand q10, q10, q9 + vand q11, q11, q9 + veor q3, q3, q10 + vshl.u64 q10, q10, #2 + veor q2, q2, q11 + vshl.u64 q11, q11, #2 + veor q1, q1, q10 + veor q0, q0, q11 + vshr.u64 q10, q3, #4 + vshr.u64 q11, q2, #4 + veor q10, q10, q7 + veor q11, q11, q6 + vand q10, q10, q8 + vand q11, q11, q8 + veor q7, q7, q10 + vshl.u64 q10, q10, #4 + veor q6, q6, q11 + vshl.u64 q11, q11, #4 + veor q3, q3, q10 + veor q2, q2, q11 + vshr.u64 q10, q1, #4 + vshr.u64 q11, q0, #4 + veor q10, q10, q5 + veor q11, q11, q4 + vand q10, q10, q8 + vand q11, q11, q8 + veor q5, q5, q10 + vshl.u64 q10, q10, #4 + veor q4, q4, q11 + vshl.u64 q11, q11, #4 + veor q1, q1, q10 + veor q0, q0, q11 + sub r5,r5,#1 + b .Lenc_sbox +.align 4 +.Lenc_loop: + vldmia r4!, {q8-q11} + veor q8, q8, q0 + veor q9, q9, q1 + vtbl.8 d0, {q8}, d24 + vtbl.8 d1, {q8}, d25 + vldmia r4!, {q8} + veor q10, q10, q2 + vtbl.8 d2, {q9}, d24 + vtbl.8 d3, {q9}, d25 + vldmia r4!, {q9} + veor q11, q11, q3 + vtbl.8 d4, {q10}, d24 + vtbl.8 d5, {q10}, d25 + vldmia r4!, {q10} + vtbl.8 d6, {q11}, d24 + vtbl.8 d7, {q11}, d25 + vldmia r4!, {q11} + veor q8, q8, q4 + veor q9, q9, q5 + vtbl.8 d8, {q8}, d24 + vtbl.8 d9, {q8}, d25 + veor q10, q10, q6 + vtbl.8 d10, {q9}, d24 + vtbl.8 d11, {q9}, d25 + veor q11, q11, q7 + vtbl.8 d12, {q10}, d24 + vtbl.8 d13, {q10}, d25 + vtbl.8 d14, {q11}, d24 + vtbl.8 d15, {q11}, d25 +.Lenc_sbox: + veor q2, q2, q1 + veor q5, q5, q6 + veor q3, q3, q0 + veor q6, q6, q2 + veor q5, q5, q0 + + veor q6, q6, q3 + veor q3, q3, q7 + veor q7, q7, q5 + veor q3, q3, q4 + veor q4, q4, q5 + + veor q2, q2, q7 + veor q3, q3, q1 + veor q1, q1, q5 + veor q11, q7, q4 + veor q10, q1, q2 + veor q9, q5, q3 + veor q13, q2, q4 + vmov q8, q10 + veor q12, q6, q0 + + vorr q10, q10, q9 + veor q15, q11, q8 + vand q14, q11, q12 + vorr q11, q11, q12 + veor q12, q12, q9 + vand q8, q8, q9 + veor q9, q3, q0 + vand q15, q15, q12 + vand q13, q13, q9 + veor q9, q7, q1 + veor q12, q5, q6 + veor q11, q11, q13 + veor q10, q10, q13 + vand q13, q9, q12 + vorr q9, q9, q12 + veor q11, q11, q15 + veor q8, q8, q13 + veor q10, q10, q14 + veor q9, q9, q15 + veor q8, q8, q14 + vand q12, q2, q3 + veor q9, q9, q14 + vand q13, q4, q0 + vand q14, q1, q5 + vorr q15, q7, q6 + veor q11, q11, q12 + veor q9, q9, q14 + veor q8, q8, q15 + veor q10, q10, q13 + + @ Inv_GF16 0, 1, 2, 3, s0, s1, s2, s3 + + @ new smaller inversion + + vand q14, q11, q9 + vmov q12, q8 + + veor q13, q10, q14 + veor q15, q8, q14 + veor q14, q8, q14 @ q14=q15 + + vbsl q13, q9, q8 + vbsl q15, q11, q10 + veor q11, q11, q10 + + vbsl q12, q13, q14 + vbsl q8, q14, q13 + + vand q14, q12, q15 + veor q9, q9, q8 + + veor q14, q14, q11 + veor q12, q6, q0 + veor q8, q5, q3 + veor q10, q15, q14 + vand q10, q10, q6 + veor q6, q6, q5 + vand q11, q5, q15 + vand q6, q6, q14 + veor q5, q11, q10 + veor q6, q6, q11 + veor q15, q15, q13 + veor q14, q14, q9 + veor q11, q15, q14 + veor q10, q13, q9 + vand q11, q11, q12 + vand q10, q10, q0 + veor q12, q12, q8 + veor q0, q0, q3 + vand q8, q8, q15 + vand q3, q3, q13 + vand q12, q12, q14 + vand q0, q0, q9 + veor q8, q8, q12 + veor q0, q0, q3 + veor q12, q12, q11 + veor q3, q3, q10 + veor q6, q6, q12 + veor q0, q0, q12 + veor q5, q5, q8 + veor q3, q3, q8 + + veor q12, q7, q4 + veor q8, q1, q2 + veor q11, q15, q14 + veor q10, q13, q9 + vand q11, q11, q12 + vand q10, q10, q4 + veor q12, q12, q8 + veor q4, q4, q2 + vand q8, q8, q15 + vand q2, q2, q13 + vand q12, q12, q14 + vand q4, q4, q9 + veor q8, q8, q12 + veor q4, q4, q2 + veor q12, q12, q11 + veor q2, q2, q10 + veor q15, q15, q13 + veor q14, q14, q9 + veor q10, q15, q14 + vand q10, q10, q7 + veor q7, q7, q1 + vand q11, q1, q15 + vand q7, q7, q14 + veor q1, q11, q10 + veor q7, q7, q11 + veor q7, q7, q12 + veor q4, q4, q12 + veor q1, q1, q8 + veor q2, q2, q8 + veor q7, q7, q0 + veor q1, q1, q6 + veor q6, q6, q0 + veor q4, q4, q7 + veor q0, q0, q1 + + veor q1, q1, q5 + veor q5, q5, q2 + veor q2, q2, q3 + veor q3, q3, q5 + veor q4, q4, q5 + + veor q6, q6, q3 + subs r5,r5,#1 + bcc .Lenc_done + vext.8 q8, q0, q0, #12 @ x0 <<< 32 + vext.8 q9, q1, q1, #12 + veor q0, q0, q8 @ x0 ^ (x0 <<< 32) + vext.8 q10, q4, q4, #12 + veor q1, q1, q9 + vext.8 q11, q6, q6, #12 + veor q4, q4, q10 + vext.8 q12, q3, q3, #12 + veor q6, q6, q11 + vext.8 q13, q7, q7, #12 + veor q3, q3, q12 + vext.8 q14, q2, q2, #12 + veor q7, q7, q13 + vext.8 q15, q5, q5, #12 + veor q2, q2, q14 + + veor q9, q9, q0 + veor q5, q5, q15 + vext.8 q0, q0, q0, #8 @ (x0 ^ (x0 <<< 32)) <<< 64) + veor q10, q10, q1 + veor q8, q8, q5 + veor q9, q9, q5 + vext.8 q1, q1, q1, #8 + veor q13, q13, q3 + veor q0, q0, q8 + veor q14, q14, q7 + veor q1, q1, q9 + vext.8 q8, q3, q3, #8 + veor q12, q12, q6 + vext.8 q9, q7, q7, #8 + veor q15, q15, q2 + vext.8 q3, q6, q6, #8 + veor q11, q11, q4 + vext.8 q7, q5, q5, #8 + veor q12, q12, q5 + vext.8 q6, q2, q2, #8 + veor q11, q11, q5 + vext.8 q2, q4, q4, #8 + veor q5, q9, q13 + veor q4, q8, q12 + veor q3, q3, q11 + veor q7, q7, q15 + veor q6, q6, q14 + @ vmov q4, q8 + veor q2, q2, q10 + @ vmov q5, q9 + vldmia r6, {q12} @ .LSR + ite eq @ Thumb2 thing, samity check in ARM + addeq r6,r6,#0x10 + bne .Lenc_loop + vldmia r6, {q12} @ .LSRM0 + b .Lenc_loop +.align 4 +.Lenc_done: + vmov.i8 q8,#0x55 @ compose .LBS0 + vmov.i8 q9,#0x33 @ compose .LBS1 + vshr.u64 q10, q2, #1 + vshr.u64 q11, q3, #1 + veor q10, q10, q5 + veor q11, q11, q7 + vand q10, q10, q8 + vand q11, q11, q8 + veor q5, q5, q10 + vshl.u64 q10, q10, #1 + veor q7, q7, q11 + vshl.u64 q11, q11, #1 + veor q2, q2, q10 + veor q3, q3, q11 + vshr.u64 q10, q4, #1 + vshr.u64 q11, q0, #1 + veor q10, q10, q6 + veor q11, q11, q1 + vand q10, q10, q8 + vand q11, q11, q8 + veor q6, q6, q10 + vshl.u64 q10, q10, #1 + veor q1, q1, q11 + vshl.u64 q11, q11, #1 + veor q4, q4, q10 + veor q0, q0, q11 + vmov.i8 q8,#0x0f @ compose .LBS2 + vshr.u64 q10, q7, #2 + vshr.u64 q11, q3, #2 + veor q10, q10, q5 + veor q11, q11, q2 + vand q10, q10, q9 + vand q11, q11, q9 + veor q5, q5, q10 + vshl.u64 q10, q10, #2 + veor q2, q2, q11 + vshl.u64 q11, q11, #2 + veor q7, q7, q10 + veor q3, q3, q11 + vshr.u64 q10, q1, #2 + vshr.u64 q11, q0, #2 + veor q10, q10, q6 + veor q11, q11, q4 + vand q10, q10, q9 + vand q11, q11, q9 + veor q6, q6, q10 + vshl.u64 q10, q10, #2 + veor q4, q4, q11 + vshl.u64 q11, q11, #2 + veor q1, q1, q10 + veor q0, q0, q11 + vshr.u64 q10, q6, #4 + vshr.u64 q11, q4, #4 + veor q10, q10, q5 + veor q11, q11, q2 + vand q10, q10, q8 + vand q11, q11, q8 + veor q5, q5, q10 + vshl.u64 q10, q10, #4 + veor q2, q2, q11 + vshl.u64 q11, q11, #4 + veor q6, q6, q10 + veor q4, q4, q11 + vshr.u64 q10, q1, #4 + vshr.u64 q11, q0, #4 + veor q10, q10, q7 + veor q11, q11, q3 + vand q10, q10, q8 + vand q11, q11, q8 + veor q7, q7, q10 + vshl.u64 q10, q10, #4 + veor q3, q3, q11 + vshl.u64 q11, q11, #4 + veor q1, q1, q10 + veor q0, q0, q11 + vldmia r4, {q8} @ last round key + veor q4, q4, q8 + veor q6, q6, q8 + veor q3, q3, q8 + veor q7, q7, q8 + veor q2, q2, q8 + veor q5, q5, q8 + veor q0, q0, q8 + veor q1, q1, q8 + bx lr +.size _bsaes_encrypt8,.-_bsaes_encrypt8 +.type _bsaes_key_convert,%function +.align 4 +_bsaes_key_convert: + adr r6,_bsaes_key_convert + vld1.8 {q7}, [r4]! @ load round 0 key + sub r6,r6,#_bsaes_key_convert-.LM0 + vld1.8 {q15}, [r4]! @ load round 1 key + + vmov.i8 q8, #0x01 @ bit masks + vmov.i8 q9, #0x02 + vmov.i8 q10, #0x04 + vmov.i8 q11, #0x08 + vmov.i8 q12, #0x10 + vmov.i8 q13, #0x20 + vldmia r6, {q14} @ .LM0 + +#ifdef __ARMEL__ + vrev32.8 q7, q7 + vrev32.8 q15, q15 +#endif + sub r5,r5,#1 + vstmia r12!, {q7} @ save round 0 key + b .Lkey_loop + +.align 4 +.Lkey_loop: + vtbl.8 d14,{q15},d28 + vtbl.8 d15,{q15},d29 + vmov.i8 q6, #0x40 + vmov.i8 q15, #0x80 + + vtst.8 q0, q7, q8 + vtst.8 q1, q7, q9 + vtst.8 q2, q7, q10 + vtst.8 q3, q7, q11 + vtst.8 q4, q7, q12 + vtst.8 q5, q7, q13 + vtst.8 q6, q7, q6 + vtst.8 q7, q7, q15 + vld1.8 {q15}, [r4]! @ load next round key + vmvn q0, q0 @ "pnot" + vmvn q1, q1 + vmvn q5, q5 + vmvn q6, q6 +#ifdef __ARMEL__ + vrev32.8 q15, q15 +#endif + subs r5,r5,#1 + vstmia r12!,{q0-q7} @ write bit-sliced round key + bne .Lkey_loop + + vmov.i8 q7,#0x63 @ compose .L63 + @ don't save last round key + bx lr +.size _bsaes_key_convert,.-_bsaes_key_convert +.extern AES_cbc_encrypt +.extern AES_decrypt + +.global bsaes_cbc_encrypt +.type bsaes_cbc_encrypt,%function +.align 5 +bsaes_cbc_encrypt: +#ifndef __KERNEL__ + cmp r2, #128 +#ifndef __thumb__ + blo AES_cbc_encrypt +#else + bhs 1f + b AES_cbc_encrypt +1: +#endif +#endif + + @ it is up to the caller to make sure we are called with enc == 0 + + mov ip, sp + stmdb sp!, {r4-r10, lr} + VFP_ABI_PUSH + ldr r8, [ip] @ IV is 1st arg on the stack + mov r2, r2, lsr#4 @ len in 16 byte blocks + sub sp, #0x10 @ scratch space to carry over the IV + mov r9, sp @ save sp + + ldr r10, [r3, #240] @ get # of rounds +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, r10, lsl#7 @ 128 bytes per inner round key + add r12, #96 @ sifze of bit-slices key schedule + + @ populate the key schedule + mov r4, r3 @ pass key + mov r5, r10 @ pass # of rounds + mov sp, r12 @ sp is sp + bl _bsaes_key_convert + vldmia sp, {q6} + vstmia r12, {q15} @ save last round key + veor q7, q7, q6 @ fix up round 0 key + vstmia sp, {q7} +#else + ldr r12, [r3, #244] + eors r12, #1 + beq 0f + + @ populate the key schedule + str r12, [r3, #244] + mov r4, r3 @ pass key + mov r5, r10 @ pass # of rounds + add r12, r3, #248 @ pass key schedule + bl _bsaes_key_convert + add r4, r3, #248 + vldmia r4, {q6} + vstmia r12, {q15} @ save last round key + veor q7, q7, q6 @ fix up round 0 key + vstmia r4, {q7} + +.align 2 +0: +#endif + + vld1.8 {q15}, [r8] @ load IV + b .Lcbc_dec_loop + +.align 4 +.Lcbc_dec_loop: + subs r2, r2, #0x8 + bmi .Lcbc_dec_loop_finish + + vld1.8 {q0-q1}, [r0]! @ load input + vld1.8 {q2-q3}, [r0]! +#ifndef BSAES_ASM_EXTENDED_KEY + mov r4, sp @ pass the key +#else + add r4, r3, #248 +#endif + vld1.8 {q4-q5}, [r0]! + mov r5, r10 + vld1.8 {q6-q7}, [r0] + sub r0, r0, #0x60 + vstmia r9, {q15} @ put aside IV + + bl _bsaes_decrypt8 + + vldmia r9, {q14} @ reload IV + vld1.8 {q8-q9}, [r0]! @ reload input + veor q0, q0, q14 @ ^= IV + vld1.8 {q10-q11}, [r0]! + veor q1, q1, q8 + veor q6, q6, q9 + vld1.8 {q12-q13}, [r0]! + veor q4, q4, q10 + veor q2, q2, q11 + vld1.8 {q14-q15}, [r0]! + veor q7, q7, q12 + vst1.8 {q0-q1}, [r1]! @ write output + veor q3, q3, q13 + vst1.8 {q6}, [r1]! + veor q5, q5, q14 + vst1.8 {q4}, [r1]! + vst1.8 {q2}, [r1]! + vst1.8 {q7}, [r1]! + vst1.8 {q3}, [r1]! + vst1.8 {q5}, [r1]! + + b .Lcbc_dec_loop + +.Lcbc_dec_loop_finish: + adds r2, r2, #8 + beq .Lcbc_dec_done + + vld1.8 {q0}, [r0]! @ load input + cmp r2, #2 + blo .Lcbc_dec_one + vld1.8 {q1}, [r0]! +#ifndef BSAES_ASM_EXTENDED_KEY + mov r4, sp @ pass the key +#else + add r4, r3, #248 +#endif + mov r5, r10 + vstmia r9, {q15} @ put aside IV + beq .Lcbc_dec_two + vld1.8 {q2}, [r0]! + cmp r2, #4 + blo .Lcbc_dec_three + vld1.8 {q3}, [r0]! + beq .Lcbc_dec_four + vld1.8 {q4}, [r0]! + cmp r2, #6 + blo .Lcbc_dec_five + vld1.8 {q5}, [r0]! + beq .Lcbc_dec_six + vld1.8 {q6}, [r0]! + sub r0, r0, #0x70 + + bl _bsaes_decrypt8 + + vldmia r9, {q14} @ reload IV + vld1.8 {q8-q9}, [r0]! @ reload input + veor q0, q0, q14 @ ^= IV + vld1.8 {q10-q11}, [r0]! + veor q1, q1, q8 + veor q6, q6, q9 + vld1.8 {q12-q13}, [r0]! + veor q4, q4, q10 + veor q2, q2, q11 + vld1.8 {q15}, [r0]! + veor q7, q7, q12 + vst1.8 {q0-q1}, [r1]! @ write output + veor q3, q3, q13 + vst1.8 {q6}, [r1]! + vst1.8 {q4}, [r1]! + vst1.8 {q2}, [r1]! + vst1.8 {q7}, [r1]! + vst1.8 {q3}, [r1]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_six: + sub r0, r0, #0x60 + bl _bsaes_decrypt8 + vldmia r9,{q14} @ reload IV + vld1.8 {q8-q9}, [r0]! @ reload input + veor q0, q0, q14 @ ^= IV + vld1.8 {q10-q11}, [r0]! + veor q1, q1, q8 + veor q6, q6, q9 + vld1.8 {q12}, [r0]! + veor q4, q4, q10 + veor q2, q2, q11 + vld1.8 {q15}, [r0]! + veor q7, q7, q12 + vst1.8 {q0-q1}, [r1]! @ write output + vst1.8 {q6}, [r1]! + vst1.8 {q4}, [r1]! + vst1.8 {q2}, [r1]! + vst1.8 {q7}, [r1]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_five: + sub r0, r0, #0x50 + bl _bsaes_decrypt8 + vldmia r9, {q14} @ reload IV + vld1.8 {q8-q9}, [r0]! @ reload input + veor q0, q0, q14 @ ^= IV + vld1.8 {q10-q11}, [r0]! + veor q1, q1, q8 + veor q6, q6, q9 + vld1.8 {q15}, [r0]! + veor q4, q4, q10 + vst1.8 {q0-q1}, [r1]! @ write output + veor q2, q2, q11 + vst1.8 {q6}, [r1]! + vst1.8 {q4}, [r1]! + vst1.8 {q2}, [r1]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_four: + sub r0, r0, #0x40 + bl _bsaes_decrypt8 + vldmia r9, {q14} @ reload IV + vld1.8 {q8-q9}, [r0]! @ reload input + veor q0, q0, q14 @ ^= IV + vld1.8 {q10}, [r0]! + veor q1, q1, q8 + veor q6, q6, q9 + vld1.8 {q15}, [r0]! + veor q4, q4, q10 + vst1.8 {q0-q1}, [r1]! @ write output + vst1.8 {q6}, [r1]! + vst1.8 {q4}, [r1]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_three: + sub r0, r0, #0x30 + bl _bsaes_decrypt8 + vldmia r9, {q14} @ reload IV + vld1.8 {q8-q9}, [r0]! @ reload input + veor q0, q0, q14 @ ^= IV + vld1.8 {q15}, [r0]! + veor q1, q1, q8 + veor q6, q6, q9 + vst1.8 {q0-q1}, [r1]! @ write output + vst1.8 {q6}, [r1]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_two: + sub r0, r0, #0x20 + bl _bsaes_decrypt8 + vldmia r9, {q14} @ reload IV + vld1.8 {q8}, [r0]! @ reload input + veor q0, q0, q14 @ ^= IV + vld1.8 {q15}, [r0]! @ reload input + veor q1, q1, q8 + vst1.8 {q0-q1}, [r1]! @ write output + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_one: + sub r0, r0, #0x10 + mov r10, r1 @ save original out pointer + mov r1, r9 @ use the iv scratch space as out buffer + mov r2, r3 + vmov q4,q15 @ just in case ensure that IV + vmov q5,q0 @ and input are preserved + bl AES_decrypt + vld1.8 {q0}, [r9,:64] @ load result + veor q0, q0, q4 @ ^= IV + vmov q15, q5 @ q5 holds input + vst1.8 {q0}, [r10] @ write output + +.Lcbc_dec_done: +#ifndef BSAES_ASM_EXTENDED_KEY + vmov.i32 q0, #0 + vmov.i32 q1, #0 +.Lcbc_dec_bzero: @ wipe key schedule [if any] + vstmia sp!, {q0-q1} + cmp sp, r9 + bne .Lcbc_dec_bzero +#endif + + mov sp, r9 + add sp, #0x10 @ add sp,r9,#0x10 is no good for thumb + vst1.8 {q15}, [r8] @ return IV + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} +.size bsaes_cbc_encrypt,.-bsaes_cbc_encrypt +.extern AES_encrypt +.global bsaes_ctr32_encrypt_blocks +.type bsaes_ctr32_encrypt_blocks,%function +.align 5 +bsaes_ctr32_encrypt_blocks: + cmp r2, #8 @ use plain AES for + blo .Lctr_enc_short @ small sizes + + mov ip, sp + stmdb sp!, {r4-r10, lr} + VFP_ABI_PUSH + ldr r8, [ip] @ ctr is 1st arg on the stack + sub sp, sp, #0x10 @ scratch space to carry over the ctr + mov r9, sp @ save sp + + ldr r10, [r3, #240] @ get # of rounds +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, r10, lsl#7 @ 128 bytes per inner round key + add r12, #96 @ size of bit-sliced key schedule + + @ populate the key schedule + mov r4, r3 @ pass key + mov r5, r10 @ pass # of rounds + mov sp, r12 @ sp is sp + bl _bsaes_key_convert + veor q7,q7,q15 @ fix up last round key + vstmia r12, {q7} @ save last round key + + vld1.8 {q0}, [r8] @ load counter + add r8, r6, #.LREVM0SR-.LM0 @ borrow r8 + vldmia sp, {q4} @ load round0 key +#else + ldr r12, [r3, #244] + eors r12, #1 + beq 0f + + @ populate the key schedule + str r12, [r3, #244] + mov r4, r3 @ pass key + mov r5, r10 @ pass # of rounds + add r12, r3, #248 @ pass key schedule + bl _bsaes_key_convert + veor q7,q7,q15 @ fix up last round key + vstmia r12, {q7} @ save last round key + +.align 2 +0: add r12, r3, #248 + vld1.8 {q0}, [r8] @ load counter + adrl r8, .LREVM0SR @ borrow r8 + vldmia r12, {q4} @ load round0 key + sub sp, #0x10 @ place for adjusted round0 key +#endif + + vmov.i32 q8,#1 @ compose 1<<96 + veor q9,q9,q9 + vrev32.8 q0,q0 + vext.8 q8,q9,q8,#4 + vrev32.8 q4,q4 + vadd.u32 q9,q8,q8 @ compose 2<<96 + vstmia sp, {q4} @ save adjusted round0 key + b .Lctr_enc_loop + +.align 4 +.Lctr_enc_loop: + vadd.u32 q10, q8, q9 @ compose 3<<96 + vadd.u32 q1, q0, q8 @ +1 + vadd.u32 q2, q0, q9 @ +2 + vadd.u32 q3, q0, q10 @ +3 + vadd.u32 q4, q1, q10 + vadd.u32 q5, q2, q10 + vadd.u32 q6, q3, q10 + vadd.u32 q7, q4, q10 + vadd.u32 q10, q5, q10 @ next counter + + @ Borrow prologue from _bsaes_encrypt8 to use the opportunity + @ to flip byte order in 32-bit counter + + vldmia sp, {q9} @ load round0 key +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x10 @ pass next round key +#else + add r4, r3, #264 +#endif + vldmia r8, {q8} @ .LREVM0SR + mov r5, r10 @ pass rounds + vstmia r9, {q10} @ save next counter + sub r6, r8, #.LREVM0SR-.LSR @ pass constants + + bl _bsaes_encrypt8_alt + + subs r2, r2, #8 + blo .Lctr_enc_loop_done + + vld1.8 {q8-q9}, [r0]! @ load input + vld1.8 {q10-q11}, [r0]! + veor q0, q8 + veor q1, q9 + vld1.8 {q12-q13}, [r0]! + veor q4, q10 + veor q6, q11 + vld1.8 {q14-q15}, [r0]! + veor q3, q12 + vst1.8 {q0-q1}, [r1]! @ write output + veor q7, q13 + veor q2, q14 + vst1.8 {q4}, [r1]! + veor q5, q15 + vst1.8 {q6}, [r1]! + vmov.i32 q8, #1 @ compose 1<<96 + vst1.8 {q3}, [r1]! + veor q9, q9, q9 + vst1.8 {q7}, [r1]! + vext.8 q8, q9, q8, #4 + vst1.8 {q2}, [r1]! + vadd.u32 q9,q8,q8 @ compose 2<<96 + vst1.8 {q5}, [r1]! + vldmia r9, {q0} @ load counter + + bne .Lctr_enc_loop + b .Lctr_enc_done + +.align 4 +.Lctr_enc_loop_done: + add r2, r2, #8 + vld1.8 {q8}, [r0]! @ load input + veor q0, q8 + vst1.8 {q0}, [r1]! @ write output + cmp r2, #2 + blo .Lctr_enc_done + vld1.8 {q9}, [r0]! + veor q1, q9 + vst1.8 {q1}, [r1]! + beq .Lctr_enc_done + vld1.8 {q10}, [r0]! + veor q4, q10 + vst1.8 {q4}, [r1]! + cmp r2, #4 + blo .Lctr_enc_done + vld1.8 {q11}, [r0]! + veor q6, q11 + vst1.8 {q6}, [r1]! + beq .Lctr_enc_done + vld1.8 {q12}, [r0]! + veor q3, q12 + vst1.8 {q3}, [r1]! + cmp r2, #6 + blo .Lctr_enc_done + vld1.8 {q13}, [r0]! + veor q7, q13 + vst1.8 {q7}, [r1]! + beq .Lctr_enc_done + vld1.8 {q14}, [r0] + veor q2, q14 + vst1.8 {q2}, [r1]! + +.Lctr_enc_done: + vmov.i32 q0, #0 + vmov.i32 q1, #0 +#ifndef BSAES_ASM_EXTENDED_KEY +.Lctr_enc_bzero: @ wipe key schedule [if any] + vstmia sp!, {q0-q1} + cmp sp, r9 + bne .Lctr_enc_bzero +#else + vstmia sp, {q0-q1} +#endif + + mov sp, r9 + add sp, #0x10 @ add sp,r9,#0x10 is no good for thumb + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} @ return + +.align 4 +.Lctr_enc_short: + ldr ip, [sp] @ ctr pointer is passed on stack + stmdb sp!, {r4-r8, lr} + + mov r4, r0 @ copy arguments + mov r5, r1 + mov r6, r2 + mov r7, r3 + ldr r8, [ip, #12] @ load counter LSW + vld1.8 {q1}, [ip] @ load whole counter value +#ifdef __ARMEL__ + rev r8, r8 +#endif + sub sp, sp, #0x10 + vst1.8 {q1}, [sp,:64] @ copy counter value + sub sp, sp, #0x10 + +.Lctr_enc_short_loop: + add r0, sp, #0x10 @ input counter value + mov r1, sp @ output on the stack + mov r2, r7 @ key + + bl AES_encrypt + + vld1.8 {q0}, [r4]! @ load input + vld1.8 {q1}, [sp,:64] @ load encrypted counter + add r8, r8, #1 +#ifdef __ARMEL__ + rev r0, r8 + str r0, [sp, #0x1c] @ next counter value +#else + str r8, [sp, #0x1c] @ next counter value +#endif + veor q0,q0,q1 + vst1.8 {q0}, [r5]! @ store output + subs r6, r6, #1 + bne .Lctr_enc_short_loop + + vmov.i32 q0, #0 + vmov.i32 q1, #0 + vstmia sp!, {q0-q1} + + ldmia sp!, {r4-r8, pc} +.size bsaes_ctr32_encrypt_blocks,.-bsaes_ctr32_encrypt_blocks +.globl bsaes_xts_encrypt +.type bsaes_xts_encrypt,%function +.align 4 +bsaes_xts_encrypt: + mov ip, sp + stmdb sp!, {r4-r10, lr} @ 0x20 + VFP_ABI_PUSH + mov r6, sp @ future r3 + + mov r7, r0 + mov r8, r1 + mov r9, r2 + mov r10, r3 + + sub r0, sp, #0x10 @ 0x10 + bic r0, #0xf @ align at 16 bytes + mov sp, r0 + +#ifdef XTS_CHAIN_TWEAK + ldr r0, [ip] @ pointer to input tweak +#else + @ generate initial tweak + ldr r0, [ip, #4] @ iv[] + mov r1, sp + ldr r2, [ip, #0] @ key2 + bl AES_encrypt + mov r0,sp @ pointer to initial tweak +#endif + + ldr r1, [r10, #240] @ get # of rounds + mov r3, r6 +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, r1, lsl#7 @ 128 bytes per inner round key + @ add r12, #96 @ size of bit-sliced key schedule + sub r12, #48 @ place for tweak[9] + + @ populate the key schedule + mov r4, r10 @ pass key + mov r5, r1 @ pass # of rounds + mov sp, r12 + add r12, #0x90 @ pass key schedule + bl _bsaes_key_convert + veor q7, q7, q15 @ fix up last round key + vstmia r12, {q7} @ save last round key +#else + ldr r12, [r10, #244] + eors r12, #1 + beq 0f + + str r12, [r10, #244] + mov r4, r10 @ pass key + mov r5, r1 @ pass # of rounds + add r12, r10, #248 @ pass key schedule + bl _bsaes_key_convert + veor q7, q7, q15 @ fix up last round key + vstmia r12, {q7} + +.align 2 +0: sub sp, #0x90 @ place for tweak[9] +#endif + + vld1.8 {q8}, [r0] @ initial tweak + adr r2, .Lxts_magic + + subs r9, #0x80 + blo .Lxts_enc_short + b .Lxts_enc_loop + +.align 4 +.Lxts_enc_loop: + vldmia r2, {q5} @ load XTS magic + vshr.s64 q6, q8, #63 + mov r0, sp + vand q6, q6, q5 + vadd.u64 q9, q8, q8 + vst1.64 {q8}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q9, #63 + veor q9, q9, q6 + vand q7, q7, q5 + vadd.u64 q10, q9, q9 + vst1.64 {q9}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q10, #63 + veor q10, q10, q7 + vand q6, q6, q5 + vld1.8 {q0}, [r7]! + vadd.u64 q11, q10, q10 + vst1.64 {q10}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q11, #63 + veor q11, q11, q6 + vand q7, q7, q5 + vld1.8 {q1}, [r7]! + veor q0, q0, q8 + vadd.u64 q12, q11, q11 + vst1.64 {q11}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q12, #63 + veor q12, q12, q7 + vand q6, q6, q5 + vld1.8 {q2}, [r7]! + veor q1, q1, q9 + vadd.u64 q13, q12, q12 + vst1.64 {q12}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q13, #63 + veor q13, q13, q6 + vand q7, q7, q5 + vld1.8 {q3}, [r7]! + veor q2, q2, q10 + vadd.u64 q14, q13, q13 + vst1.64 {q13}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q14, #63 + veor q14, q14, q7 + vand q6, q6, q5 + vld1.8 {q4}, [r7]! + veor q3, q3, q11 + vadd.u64 q15, q14, q14 + vst1.64 {q14}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q15, #63 + veor q15, q15, q6 + vand q7, q7, q5 + vld1.8 {q5}, [r7]! + veor q4, q4, q12 + vadd.u64 q8, q15, q15 + vst1.64 {q15}, [r0,:128]! + vswp d15,d14 + veor q8, q8, q7 + vst1.64 {q8}, [r0,:128] @ next round tweak + + vld1.8 {q6-q7}, [r7]! + veor q5, q5, q13 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q6, q6, q14 + mov r5, r1 @ pass rounds + veor q7, q7, q15 + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12-q13}, [r0,:128]! + veor q1, q1, q9 + veor q8, q4, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q6, q11 + vld1.64 {q14-q15}, [r0,:128]! + veor q10, q3, q12 + vst1.8 {q8-q9}, [r8]! + veor q11, q7, q13 + veor q12, q2, q14 + vst1.8 {q10-q11}, [r8]! + veor q13, q5, q15 + vst1.8 {q12-q13}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + + subs r9, #0x80 + bpl .Lxts_enc_loop + +.Lxts_enc_short: + adds r9, #0x70 + bmi .Lxts_enc_done + + vldmia r2, {q5} @ load XTS magic + vshr.s64 q7, q8, #63 + mov r0, sp + vand q7, q7, q5 + vadd.u64 q9, q8, q8 + vst1.64 {q8}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q9, #63 + veor q9, q9, q7 + vand q6, q6, q5 + vadd.u64 q10, q9, q9 + vst1.64 {q9}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q10, #63 + veor q10, q10, q6 + vand q7, q7, q5 + vld1.8 {q0}, [r7]! + subs r9, #0x10 + bmi .Lxts_enc_1 + vadd.u64 q11, q10, q10 + vst1.64 {q10}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q11, #63 + veor q11, q11, q7 + vand q6, q6, q5 + vld1.8 {q1}, [r7]! + subs r9, #0x10 + bmi .Lxts_enc_2 + veor q0, q0, q8 + vadd.u64 q12, q11, q11 + vst1.64 {q11}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q12, #63 + veor q12, q12, q6 + vand q7, q7, q5 + vld1.8 {q2}, [r7]! + subs r9, #0x10 + bmi .Lxts_enc_3 + veor q1, q1, q9 + vadd.u64 q13, q12, q12 + vst1.64 {q12}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q13, #63 + veor q13, q13, q7 + vand q6, q6, q5 + vld1.8 {q3}, [r7]! + subs r9, #0x10 + bmi .Lxts_enc_4 + veor q2, q2, q10 + vadd.u64 q14, q13, q13 + vst1.64 {q13}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q14, #63 + veor q14, q14, q6 + vand q7, q7, q5 + vld1.8 {q4}, [r7]! + subs r9, #0x10 + bmi .Lxts_enc_5 + veor q3, q3, q11 + vadd.u64 q15, q14, q14 + vst1.64 {q14}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q15, #63 + veor q15, q15, q7 + vand q6, q6, q5 + vld1.8 {q5}, [r7]! + subs r9, #0x10 + bmi .Lxts_enc_6 + veor q4, q4, q12 + sub r9, #0x10 + vst1.64 {q15}, [r0,:128] @ next round tweak + + vld1.8 {q6}, [r7]! + veor q5, q5, q13 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q6, q6, q14 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12-q13}, [r0,:128]! + veor q1, q1, q9 + veor q8, q4, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q6, q11 + vld1.64 {q14}, [r0,:128]! + veor q10, q3, q12 + vst1.8 {q8-q9}, [r8]! + veor q11, q7, q13 + veor q12, q2, q14 + vst1.8 {q10-q11}, [r8]! + vst1.8 {q12}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_6: + vst1.64 {q14}, [r0,:128] @ next round tweak + + veor q4, q4, q12 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q5, q5, q13 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12-q13}, [r0,:128]! + veor q1, q1, q9 + veor q8, q4, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q6, q11 + veor q10, q3, q12 + vst1.8 {q8-q9}, [r8]! + veor q11, q7, q13 + vst1.8 {q10-q11}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_enc_done + +@ put this in range for both ARM and Thumb mode adr instructions +.align 5 +.Lxts_magic: + .quad 1, 0x87 + +.align 5 +.Lxts_enc_5: + vst1.64 {q13}, [r0,:128] @ next round tweak + + veor q3, q3, q11 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q4, q4, q12 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12}, [r0,:128]! + veor q1, q1, q9 + veor q8, q4, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q6, q11 + veor q10, q3, q12 + vst1.8 {q8-q9}, [r8]! + vst1.8 {q10}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_4: + vst1.64 {q12}, [r0,:128] @ next round tweak + + veor q2, q2, q10 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q3, q3, q11 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + veor q1, q1, q9 + veor q8, q4, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q6, q11 + vst1.8 {q8-q9}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_3: + vst1.64 {q11}, [r0,:128] @ next round tweak + + veor q1, q1, q9 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q2, q2, q10 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10}, [r0,:128]! + veor q0, q0, q8 + veor q1, q1, q9 + veor q8, q4, q10 + vst1.8 {q0-q1}, [r8]! + vst1.8 {q8}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_2: + vst1.64 {q10}, [r0,:128] @ next round tweak + + veor q0, q0, q8 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q1, q1, q9 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + veor q0, q0, q8 + veor q1, q1, q9 + vst1.8 {q0-q1}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_1: + mov r0, sp + veor q0, q8 + mov r1, sp + vst1.8 {q0}, [sp,:128] + mov r2, r10 + mov r4, r3 @ preserve fp + + bl AES_encrypt + + vld1.8 {q0}, [sp,:128] + veor q0, q0, q8 + vst1.8 {q0}, [r8]! + mov r3, r4 + + vmov q8, q9 @ next round tweak + +.Lxts_enc_done: +#ifndef XTS_CHAIN_TWEAK + adds r9, #0x10 + beq .Lxts_enc_ret + sub r6, r8, #0x10 + +.Lxts_enc_steal: + ldrb r0, [r7], #1 + ldrb r1, [r8, #-0x10] + strb r0, [r8, #-0x10] + strb r1, [r8], #1 + + subs r9, #1 + bhi .Lxts_enc_steal + + vld1.8 {q0}, [r6] + mov r0, sp + veor q0, q0, q8 + mov r1, sp + vst1.8 {q0}, [sp,:128] + mov r2, r10 + mov r4, r3 @ preserve fp + + bl AES_encrypt + + vld1.8 {q0}, [sp,:128] + veor q0, q0, q8 + vst1.8 {q0}, [r6] + mov r3, r4 +#endif + +.Lxts_enc_ret: + bic r0, r3, #0xf + vmov.i32 q0, #0 + vmov.i32 q1, #0 +#ifdef XTS_CHAIN_TWEAK + ldr r1, [r3, #0x20+VFP_ABI_FRAME] @ chain tweak +#endif +.Lxts_enc_bzero: @ wipe key schedule [if any] + vstmia sp!, {q0-q1} + cmp sp, r0 + bne .Lxts_enc_bzero + + mov sp, r3 +#ifdef XTS_CHAIN_TWEAK + vst1.8 {q8}, [r1] +#endif + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} @ return + +.size bsaes_xts_encrypt,.-bsaes_xts_encrypt + +.globl bsaes_xts_decrypt +.type bsaes_xts_decrypt,%function +.align 4 +bsaes_xts_decrypt: + mov ip, sp + stmdb sp!, {r4-r10, lr} @ 0x20 + VFP_ABI_PUSH + mov r6, sp @ future r3 + + mov r7, r0 + mov r8, r1 + mov r9, r2 + mov r10, r3 + + sub r0, sp, #0x10 @ 0x10 + bic r0, #0xf @ align at 16 bytes + mov sp, r0 + +#ifdef XTS_CHAIN_TWEAK + ldr r0, [ip] @ pointer to input tweak +#else + @ generate initial tweak + ldr r0, [ip, #4] @ iv[] + mov r1, sp + ldr r2, [ip, #0] @ key2 + bl AES_encrypt + mov r0, sp @ pointer to initial tweak +#endif + + ldr r1, [r10, #240] @ get # of rounds + mov r3, r6 +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, r1, lsl#7 @ 128 bytes per inner round key + @ add r12, #96 @ size of bit-sliced key schedule + sub r12, #48 @ place for tweak[9] + + @ populate the key schedule + mov r4, r10 @ pass key + mov r5, r1 @ pass # of rounds + mov sp, r12 + add r12, #0x90 @ pass key schedule + bl _bsaes_key_convert + add r4, sp, #0x90 + vldmia r4, {q6} + vstmia r12, {q15} @ save last round key + veor q7, q7, q6 @ fix up round 0 key + vstmia r4, {q7} +#else + ldr r12, [r10, #244] + eors r12, #1 + beq 0f + + str r12, [r10, #244] + mov r4, r10 @ pass key + mov r5, r1 @ pass # of rounds + add r12, r10, #248 @ pass key schedule + bl _bsaes_key_convert + add r4, r10, #248 + vldmia r4, {q6} + vstmia r12, {q15} @ save last round key + veor q7, q7, q6 @ fix up round 0 key + vstmia r4, {q7} + +.align 2 +0: sub sp, #0x90 @ place for tweak[9] +#endif + vld1.8 {q8}, [r0] @ initial tweak + adr r2, .Lxts_magic + + tst r9, #0xf @ if not multiple of 16 + it ne @ Thumb2 thing, sanity check in ARM + subne r9, #0x10 @ subtract another 16 bytes + subs r9, #0x80 + + blo .Lxts_dec_short + b .Lxts_dec_loop + +.align 4 +.Lxts_dec_loop: + vldmia r2, {q5} @ load XTS magic + vshr.s64 q6, q8, #63 + mov r0, sp + vand q6, q6, q5 + vadd.u64 q9, q8, q8 + vst1.64 {q8}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q9, #63 + veor q9, q9, q6 + vand q7, q7, q5 + vadd.u64 q10, q9, q9 + vst1.64 {q9}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q10, #63 + veor q10, q10, q7 + vand q6, q6, q5 + vld1.8 {q0}, [r7]! + vadd.u64 q11, q10, q10 + vst1.64 {q10}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q11, #63 + veor q11, q11, q6 + vand q7, q7, q5 + vld1.8 {q1}, [r7]! + veor q0, q0, q8 + vadd.u64 q12, q11, q11 + vst1.64 {q11}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q12, #63 + veor q12, q12, q7 + vand q6, q6, q5 + vld1.8 {q2}, [r7]! + veor q1, q1, q9 + vadd.u64 q13, q12, q12 + vst1.64 {q12}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q13, #63 + veor q13, q13, q6 + vand q7, q7, q5 + vld1.8 {q3}, [r7]! + veor q2, q2, q10 + vadd.u64 q14, q13, q13 + vst1.64 {q13}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q14, #63 + veor q14, q14, q7 + vand q6, q6, q5 + vld1.8 {q4}, [r7]! + veor q3, q3, q11 + vadd.u64 q15, q14, q14 + vst1.64 {q14}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q15, #63 + veor q15, q15, q6 + vand q7, q7, q5 + vld1.8 {q5}, [r7]! + veor q4, q4, q12 + vadd.u64 q8, q15, q15 + vst1.64 {q15}, [r0,:128]! + vswp d15,d14 + veor q8, q8, q7 + vst1.64 {q8}, [r0,:128] @ next round tweak + + vld1.8 {q6-q7}, [r7]! + veor q5, q5, q13 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q6, q6, q14 + mov r5, r1 @ pass rounds + veor q7, q7, q15 + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12-q13}, [r0,:128]! + veor q1, q1, q9 + veor q8, q6, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q4, q11 + vld1.64 {q14-q15}, [r0,:128]! + veor q10, q2, q12 + vst1.8 {q8-q9}, [r8]! + veor q11, q7, q13 + veor q12, q3, q14 + vst1.8 {q10-q11}, [r8]! + veor q13, q5, q15 + vst1.8 {q12-q13}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + + subs r9, #0x80 + bpl .Lxts_dec_loop + +.Lxts_dec_short: + adds r9, #0x70 + bmi .Lxts_dec_done + + vldmia r2, {q5} @ load XTS magic + vshr.s64 q7, q8, #63 + mov r0, sp + vand q7, q7, q5 + vadd.u64 q9, q8, q8 + vst1.64 {q8}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q9, #63 + veor q9, q9, q7 + vand q6, q6, q5 + vadd.u64 q10, q9, q9 + vst1.64 {q9}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q10, #63 + veor q10, q10, q6 + vand q7, q7, q5 + vld1.8 {q0}, [r7]! + subs r9, #0x10 + bmi .Lxts_dec_1 + vadd.u64 q11, q10, q10 + vst1.64 {q10}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q11, #63 + veor q11, q11, q7 + vand q6, q6, q5 + vld1.8 {q1}, [r7]! + subs r9, #0x10 + bmi .Lxts_dec_2 + veor q0, q0, q8 + vadd.u64 q12, q11, q11 + vst1.64 {q11}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q12, #63 + veor q12, q12, q6 + vand q7, q7, q5 + vld1.8 {q2}, [r7]! + subs r9, #0x10 + bmi .Lxts_dec_3 + veor q1, q1, q9 + vadd.u64 q13, q12, q12 + vst1.64 {q12}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q13, #63 + veor q13, q13, q7 + vand q6, q6, q5 + vld1.8 {q3}, [r7]! + subs r9, #0x10 + bmi .Lxts_dec_4 + veor q2, q2, q10 + vadd.u64 q14, q13, q13 + vst1.64 {q13}, [r0,:128]! + vswp d13,d12 + vshr.s64 q7, q14, #63 + veor q14, q14, q6 + vand q7, q7, q5 + vld1.8 {q4}, [r7]! + subs r9, #0x10 + bmi .Lxts_dec_5 + veor q3, q3, q11 + vadd.u64 q15, q14, q14 + vst1.64 {q14}, [r0,:128]! + vswp d15,d14 + vshr.s64 q6, q15, #63 + veor q15, q15, q7 + vand q6, q6, q5 + vld1.8 {q5}, [r7]! + subs r9, #0x10 + bmi .Lxts_dec_6 + veor q4, q4, q12 + sub r9, #0x10 + vst1.64 {q15}, [r0,:128] @ next round tweak + + vld1.8 {q6}, [r7]! + veor q5, q5, q13 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q6, q6, q14 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12-q13}, [r0,:128]! + veor q1, q1, q9 + veor q8, q6, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q4, q11 + vld1.64 {q14}, [r0,:128]! + veor q10, q2, q12 + vst1.8 {q8-q9}, [r8]! + veor q11, q7, q13 + veor q12, q3, q14 + vst1.8 {q10-q11}, [r8]! + vst1.8 {q12}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_6: + vst1.64 {q14}, [r0,:128] @ next round tweak + + veor q4, q4, q12 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q5, q5, q13 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12-q13}, [r0,:128]! + veor q1, q1, q9 + veor q8, q6, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q4, q11 + veor q10, q2, q12 + vst1.8 {q8-q9}, [r8]! + veor q11, q7, q13 + vst1.8 {q10-q11}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_5: + vst1.64 {q13}, [r0,:128] @ next round tweak + + veor q3, q3, q11 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q4, q4, q12 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + vld1.64 {q12}, [r0,:128]! + veor q1, q1, q9 + veor q8, q6, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q4, q11 + veor q10, q2, q12 + vst1.8 {q8-q9}, [r8]! + vst1.8 {q10}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_4: + vst1.64 {q12}, [r0,:128] @ next round tweak + + veor q2, q2, q10 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q3, q3, q11 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10-q11}, [r0,:128]! + veor q0, q0, q8 + veor q1, q1, q9 + veor q8, q6, q10 + vst1.8 {q0-q1}, [r8]! + veor q9, q4, q11 + vst1.8 {q8-q9}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_3: + vst1.64 {q11}, [r0,:128] @ next round tweak + + veor q1, q1, q9 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q2, q2, q10 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + vld1.64 {q10}, [r0,:128]! + veor q0, q0, q8 + veor q1, q1, q9 + veor q8, q6, q10 + vst1.8 {q0-q1}, [r8]! + vst1.8 {q8}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_2: + vst1.64 {q10}, [r0,:128] @ next round tweak + + veor q0, q0, q8 +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, r10, #248 @ pass key schedule +#endif + veor q1, q1, q9 + mov r5, r1 @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {q8-q9}, [r0,:128]! + veor q0, q0, q8 + veor q1, q1, q9 + vst1.8 {q0-q1}, [r8]! + + vld1.64 {q8}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_1: + mov r0, sp + veor q0, q8 + mov r1, sp + vst1.8 {q0}, [sp,:128] + mov r2, r10 + mov r4, r3 @ preserve fp + mov r5, r2 @ preserve magic + + bl AES_decrypt + + vld1.8 {q0}, [sp,:128] + veor q0, q0, q8 + vst1.8 {q0}, [r8]! + mov r3, r4 + mov r2, r5 + + vmov q8, q9 @ next round tweak + +.Lxts_dec_done: +#ifndef XTS_CHAIN_TWEAK + adds r9, #0x10 + beq .Lxts_dec_ret + + @ calculate one round of extra tweak for the stolen ciphertext + vldmia r2, {q5} + vshr.s64 q6, q8, #63 + vand q6, q6, q5 + vadd.u64 q9, q8, q8 + vswp d13,d12 + veor q9, q9, q6 + + @ perform the final decryption with the last tweak value + vld1.8 {q0}, [r7]! + mov r0, sp + veor q0, q0, q9 + mov r1, sp + vst1.8 {q0}, [sp,:128] + mov r2, r10 + mov r4, r3 @ preserve fp + + bl AES_decrypt + + vld1.8 {q0}, [sp,:128] + veor q0, q0, q9 + vst1.8 {q0}, [r8] + + mov r6, r8 +.Lxts_dec_steal: + ldrb r1, [r8] + ldrb r0, [r7], #1 + strb r1, [r8, #0x10] + strb r0, [r8], #1 + + subs r9, #1 + bhi .Lxts_dec_steal + + vld1.8 {q0}, [r6] + mov r0, sp + veor q0, q8 + mov r1, sp + vst1.8 {q0}, [sp,:128] + mov r2, r10 + + bl AES_decrypt + + vld1.8 {q0}, [sp,:128] + veor q0, q0, q8 + vst1.8 {q0}, [r6] + mov r3, r4 +#endif + +.Lxts_dec_ret: + bic r0, r3, #0xf + vmov.i32 q0, #0 + vmov.i32 q1, #0 +#ifdef XTS_CHAIN_TWEAK + ldr r1, [r3, #0x20+VFP_ABI_FRAME] @ chain tweak +#endif +.Lxts_dec_bzero: @ wipe key schedule [if any] + vstmia sp!, {q0-q1} + cmp sp, r0 + bne .Lxts_dec_bzero + + mov sp, r3 +#ifdef XTS_CHAIN_TWEAK + vst1.8 {q8}, [r1] +#endif + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} @ return + +.size bsaes_xts_decrypt,.-bsaes_xts_decrypt +#endif diff --git a/arch/arm/crypto/aesbs-glue.c b/arch/arm/crypto/aesbs-glue.c new file mode 100644 index 000000000000..15468fbbdea3 --- /dev/null +++ b/arch/arm/crypto/aesbs-glue.c @@ -0,0 +1,434 @@ +/* + * linux/arch/arm/crypto/aesbs-glue.c - glue code for NEON bit sliced AES + * + * Copyright (C) 2013 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 + * published by the Free Software Foundation. + */ + +#include <asm/neon.h> +#include <crypto/aes.h> +#include <crypto/ablk_helper.h> +#include <crypto/algapi.h> +#include <linux/module.h> + +#include "aes_glue.h" + +#define BIT_SLICED_KEY_MAXSIZE (128 * (AES_MAXNR - 1) + 2 * AES_BLOCK_SIZE) + +struct BS_KEY { + struct AES_KEY rk; + int converted; + u8 __aligned(8) bs[BIT_SLICED_KEY_MAXSIZE]; +} __aligned(8); + +asmlinkage void bsaes_enc_key_convert(u8 out[], struct AES_KEY const *in); +asmlinkage void bsaes_dec_key_convert(u8 out[], struct AES_KEY const *in); + +asmlinkage void bsaes_cbc_encrypt(u8 const in[], u8 out[], u32 bytes, + struct BS_KEY *key, u8 iv[]); + +asmlinkage void bsaes_ctr32_encrypt_blocks(u8 const in[], u8 out[], u32 blocks, + struct BS_KEY *key, u8 const iv[]); + +asmlinkage void bsaes_xts_encrypt(u8 const in[], u8 out[], u32 bytes, + struct BS_KEY *key, u8 tweak[]); + +asmlinkage void bsaes_xts_decrypt(u8 const in[], u8 out[], u32 bytes, + struct BS_KEY *key, u8 tweak[]); + +struct aesbs_cbc_ctx { + struct AES_KEY enc; + struct BS_KEY dec; +}; + +struct aesbs_ctr_ctx { + struct BS_KEY enc; +}; + +struct aesbs_xts_ctx { + struct BS_KEY enc; + struct BS_KEY dec; + struct AES_KEY twkey; +}; + +static int aesbs_cbc_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct aesbs_cbc_ctx *ctx = crypto_tfm_ctx(tfm); + int bits = key_len * 8; + + if (private_AES_set_encrypt_key(in_key, bits, &ctx->enc)) { + tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + ctx->dec.rk = ctx->enc; + private_AES_set_decrypt_key(in_key, bits, &ctx->dec.rk); + ctx->dec.converted = 0; + return 0; +} + +static int aesbs_ctr_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct aesbs_ctr_ctx *ctx = crypto_tfm_ctx(tfm); + int bits = key_len * 8; + + if (private_AES_set_encrypt_key(in_key, bits, &ctx->enc.rk)) { + tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + ctx->enc.converted = 0; + return 0; +} + +static int aesbs_xts_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + struct aesbs_xts_ctx *ctx = crypto_tfm_ctx(tfm); + int bits = key_len * 4; + + if (private_AES_set_encrypt_key(in_key, bits, &ctx->enc.rk)) { + tfm->crt_flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + ctx->dec.rk = ctx->enc.rk; + private_AES_set_decrypt_key(in_key, bits, &ctx->dec.rk); + private_AES_set_encrypt_key(in_key + key_len / 2, bits, &ctx->twkey); + ctx->enc.converted = ctx->dec.converted = 0; + return 0; +} + +static int aesbs_cbc_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct aesbs_cbc_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + while (walk.nbytes) { + u32 blocks = walk.nbytes / AES_BLOCK_SIZE; + u8 *src = walk.src.virt.addr; + + if (walk.dst.virt.addr == walk.src.virt.addr) { + u8 *iv = walk.iv; + + do { + crypto_xor(src, iv, AES_BLOCK_SIZE); + AES_encrypt(src, src, &ctx->enc); + iv = src; + src += AES_BLOCK_SIZE; + } while (--blocks); + memcpy(walk.iv, iv, AES_BLOCK_SIZE); + } else { + u8 *dst = walk.dst.virt.addr; + + do { + crypto_xor(walk.iv, src, AES_BLOCK_SIZE); + AES_encrypt(walk.iv, dst, &ctx->enc); + memcpy(walk.iv, dst, AES_BLOCK_SIZE); + src += AES_BLOCK_SIZE; + dst += AES_BLOCK_SIZE; + } while (--blocks); + } + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); + } + return err; +} + +static int aesbs_cbc_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct aesbs_cbc_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, 8 * AES_BLOCK_SIZE); + + while ((walk.nbytes / AES_BLOCK_SIZE) >= 8) { + kernel_neon_begin(); + bsaes_cbc_encrypt(walk.src.virt.addr, walk.dst.virt.addr, + walk.nbytes, &ctx->dec, walk.iv); + kernel_neon_end(); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); + } + while (walk.nbytes) { + u32 blocks = walk.nbytes / AES_BLOCK_SIZE; + u8 *dst = walk.dst.virt.addr; + u8 *src = walk.src.virt.addr; + u8 bk[2][AES_BLOCK_SIZE]; + u8 *iv = walk.iv; + + do { + if (walk.dst.virt.addr == walk.src.virt.addr) + memcpy(bk[blocks & 1], src, AES_BLOCK_SIZE); + + AES_decrypt(src, dst, &ctx->dec.rk); + crypto_xor(dst, iv, AES_BLOCK_SIZE); + + if (walk.dst.virt.addr == walk.src.virt.addr) + iv = bk[blocks & 1]; + else + iv = src; + + dst += AES_BLOCK_SIZE; + src += AES_BLOCK_SIZE; + } while (--blocks); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); + } + return err; +} + +static void inc_be128_ctr(__be32 ctr[], u32 addend) +{ + int i; + + for (i = 3; i >= 0; i--, addend = 1) { + u32 n = be32_to_cpu(ctr[i]) + addend; + + ctr[i] = cpu_to_be32(n); + if (n >= addend) + break; + } +} + +static int aesbs_ctr_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct aesbs_ctr_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + u32 blocks; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, 8 * AES_BLOCK_SIZE); + + while ((blocks = walk.nbytes / AES_BLOCK_SIZE)) { + u32 tail = walk.nbytes % AES_BLOCK_SIZE; + __be32 *ctr = (__be32 *)walk.iv; + u32 headroom = UINT_MAX - be32_to_cpu(ctr[3]); + + /* avoid 32 bit counter overflow in the NEON code */ + if (unlikely(headroom < blocks)) { + blocks = headroom + 1; + tail = walk.nbytes - blocks * AES_BLOCK_SIZE; + } + kernel_neon_begin(); + bsaes_ctr32_encrypt_blocks(walk.src.virt.addr, + walk.dst.virt.addr, blocks, + &ctx->enc, walk.iv); + kernel_neon_end(); + inc_be128_ctr(ctr, blocks); + + nbytes -= blocks * AES_BLOCK_SIZE; + if (nbytes && nbytes == tail && nbytes <= AES_BLOCK_SIZE) + break; + + err = blkcipher_walk_done(desc, &walk, tail); + } + if (walk.nbytes) { + u8 *tdst = walk.dst.virt.addr + blocks * AES_BLOCK_SIZE; + u8 *tsrc = walk.src.virt.addr + blocks * AES_BLOCK_SIZE; + u8 ks[AES_BLOCK_SIZE]; + + AES_encrypt(walk.iv, ks, &ctx->enc.rk); + if (tdst != tsrc) + memcpy(tdst, tsrc, nbytes); + crypto_xor(tdst, ks, nbytes); + err = blkcipher_walk_done(desc, &walk, 0); + } + return err; +} + +static int aesbs_xts_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct aesbs_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, 8 * AES_BLOCK_SIZE); + + /* generate the initial tweak */ + AES_encrypt(walk.iv, walk.iv, &ctx->twkey); + + while (walk.nbytes) { + kernel_neon_begin(); + bsaes_xts_encrypt(walk.src.virt.addr, walk.dst.virt.addr, + walk.nbytes, &ctx->enc, walk.iv); + kernel_neon_end(); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); + } + return err; +} + +static int aesbs_xts_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, + struct scatterlist *src, unsigned int nbytes) +{ + struct aesbs_xts_ctx *ctx = crypto_blkcipher_ctx(desc->tfm); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt_block(desc, &walk, 8 * AES_BLOCK_SIZE); + + /* generate the initial tweak */ + AES_encrypt(walk.iv, walk.iv, &ctx->twkey); + + while (walk.nbytes) { + kernel_neon_begin(); + bsaes_xts_decrypt(walk.src.virt.addr, walk.dst.virt.addr, + walk.nbytes, &ctx->dec, walk.iv); + kernel_neon_end(); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); + } + return err; +} + +static struct crypto_alg aesbs_algs[] = { { + .cra_name = "__cbc-aes-neonbs", + .cra_driver_name = "__driver-cbc-aes-neonbs", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct aesbs_cbc_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = aesbs_cbc_set_key, + .encrypt = aesbs_cbc_encrypt, + .decrypt = aesbs_cbc_decrypt, + }, +}, { + .cra_name = "__ctr-aes-neonbs", + .cra_driver_name = "__driver-ctr-aes-neonbs", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct aesbs_ctr_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = aesbs_ctr_set_key, + .encrypt = aesbs_ctr_encrypt, + .decrypt = aesbs_ctr_encrypt, + }, +}, { + .cra_name = "__xts-aes-neonbs", + .cra_driver_name = "__driver-xts-aes-neonbs", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct aesbs_xts_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_blkcipher = { + .min_keysize = 2 * AES_MIN_KEY_SIZE, + .max_keysize = 2 * AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = aesbs_xts_set_key, + .encrypt = aesbs_xts_encrypt, + .decrypt = aesbs_xts_decrypt, + }, +}, { + .cra_name = "cbc(aes)", + .cra_driver_name = "cbc-aes-neonbs", + .cra_priority = 300, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER|CRYPTO_ALG_ASYNC, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_helper_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_ablkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = __ablk_encrypt, + .decrypt = ablk_decrypt, + } +}, { + .cra_name = "ctr(aes)", + .cra_driver_name = "ctr-aes-neonbs", + .cra_priority = 300, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER|CRYPTO_ALG_ASYNC, + .cra_blocksize = 1, + .cra_ctxsize = sizeof(struct async_helper_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_ablkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + } +}, { + .cra_name = "xts(aes)", + .cra_driver_name = "xts-aes-neonbs", + .cra_priority = 300, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER|CRYPTO_ALG_ASYNC, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_helper_ctx), + .cra_alignmask = 7, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_init = ablk_init, + .cra_exit = ablk_exit, + .cra_ablkcipher = { + .min_keysize = 2 * AES_MIN_KEY_SIZE, + .max_keysize = 2 * AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + } +} }; + +static int __init aesbs_mod_init(void) +{ + if (!cpu_has_neon()) + return -ENODEV; + + return crypto_register_algs(aesbs_algs, ARRAY_SIZE(aesbs_algs)); +} + +static void __exit aesbs_mod_exit(void) +{ + crypto_unregister_algs(aesbs_algs, ARRAY_SIZE(aesbs_algs)); +} + +module_init(aesbs_mod_init); +module_exit(aesbs_mod_exit); + +MODULE_DESCRIPTION("Bit sliced AES in CBC/CTR/XTS modes using NEON"); +MODULE_AUTHOR("Ard Biesheuvel <ard.biesheuvel@linaro.org>"); +MODULE_LICENSE("GPL"); diff --git a/arch/arm/crypto/bsaes-armv7.pl b/arch/arm/crypto/bsaes-armv7.pl new file mode 100644 index 000000000000..be068db960ee --- /dev/null +++ b/arch/arm/crypto/bsaes-armv7.pl @@ -0,0 +1,2467 @@ +#!/usr/bin/env perl + +# ==================================================================== +# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL +# project. The module is, however, dual licensed under OpenSSL and +# CRYPTOGAMS licenses depending on where you obtain it. For further +# details see http://www.openssl.org/~appro/cryptogams/. +# +# Specific modes and adaptation for Linux kernel by Ard Biesheuvel +# <ard.biesheuvel@linaro.org>. Permission to use under GPL terms is +# granted. +# ==================================================================== + +# Bit-sliced AES for ARM NEON +# +# February 2012. +# +# This implementation is direct adaptation of bsaes-x86_64 module for +# ARM NEON. Except that this module is endian-neutral [in sense that +# it can be compiled for either endianness] by courtesy of vld1.8's +# neutrality. Initial version doesn't implement interface to OpenSSL, +# only low-level primitives and unsupported entry points, just enough +# to collect performance results, which for Cortex-A8 core are: +# +# encrypt 19.5 cycles per byte processed with 128-bit key +# decrypt 22.1 cycles per byte processed with 128-bit key +# key conv. 440 cycles per 128-bit key/0.18 of 8x block +# +# Snapdragon S4 encrypts byte in 17.6 cycles and decrypts in 19.7, +# which is [much] worse than anticipated (for further details see +# http://www.openssl.org/~appro/Snapdragon-S4.html). +# +# Cortex-A15 manages in 14.2/16.1 cycles [when integer-only code +# manages in 20.0 cycles]. +# +# When comparing to x86_64 results keep in mind that NEON unit is +# [mostly] single-issue and thus can't [fully] benefit from +# instruction-level parallelism. And when comparing to aes-armv4 +# results keep in mind key schedule conversion overhead (see +# bsaes-x86_64.pl for further details)... +# +# <appro@openssl.org> + +# April-August 2013 +# +# Add CBC, CTR and XTS subroutines, adapt for kernel use. +# +# <ard.biesheuvel@linaro.org> + +while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {} +open STDOUT,">$output"; + +my ($inp,$out,$len,$key)=("r0","r1","r2","r3"); +my @XMM=map("q$_",(0..15)); + +{ +my ($key,$rounds,$const)=("r4","r5","r6"); + +sub Dlo() { shift=~m|q([1]?[0-9])|?"d".($1*2):""; } +sub Dhi() { shift=~m|q([1]?[0-9])|?"d".($1*2+1):""; } + +sub Sbox { +# input in lsb > [b0, b1, b2, b3, b4, b5, b6, b7] < msb +# output in lsb > [b0, b1, b4, b6, b3, b7, b2, b5] < msb +my @b=@_[0..7]; +my @t=@_[8..11]; +my @s=@_[12..15]; + &InBasisChange (@b); + &Inv_GF256 (@b[6,5,0,3,7,1,4,2],@t,@s); + &OutBasisChange (@b[7,1,4,2,6,5,0,3]); +} + +sub InBasisChange { +# input in lsb > [b0, b1, b2, b3, b4, b5, b6, b7] < msb +# output in lsb > [b6, b5, b0, b3, b7, b1, b4, b2] < msb +my @b=@_[0..7]; +$code.=<<___; + veor @b[2], @b[2], @b[1] + veor @b[5], @b[5], @b[6] + veor @b[3], @b[3], @b[0] + veor @b[6], @b[6], @b[2] + veor @b[5], @b[5], @b[0] + + veor @b[6], @b[6], @b[3] + veor @b[3], @b[3], @b[7] + veor @b[7], @b[7], @b[5] + veor @b[3], @b[3], @b[4] + veor @b[4], @b[4], @b[5] + + veor @b[2], @b[2], @b[7] + veor @b[3], @b[3], @b[1] + veor @b[1], @b[1], @b[5] +___ +} + +sub OutBasisChange { +# input in lsb > [b0, b1, b2, b3, b4, b5, b6, b7] < msb +# output in lsb > [b6, b1, b2, b4, b7, b0, b3, b5] < msb +my @b=@_[0..7]; +$code.=<<___; + veor @b[0], @b[0], @b[6] + veor @b[1], @b[1], @b[4] + veor @b[4], @b[4], @b[6] + veor @b[2], @b[2], @b[0] + veor @b[6], @b[6], @b[1] + + veor @b[1], @b[1], @b[5] + veor @b[5], @b[5], @b[3] + veor @b[3], @b[3], @b[7] + veor @b[7], @b[7], @b[5] + veor @b[2], @b[2], @b[5] + + veor @b[4], @b[4], @b[7] +___ +} + +sub InvSbox { +# input in lsb > [b0, b1, b2, b3, b4, b5, b6, b7] < msb +# output in lsb > [b0, b1, b6, b4, b2, b7, b3, b5] < msb +my @b=@_[0..7]; +my @t=@_[8..11]; +my @s=@_[12..15]; + &InvInBasisChange (@b); + &Inv_GF256 (@b[5,1,2,6,3,7,0,4],@t,@s); + &InvOutBasisChange (@b[3,7,0,4,5,1,2,6]); +} + +sub InvInBasisChange { # OutBasisChange in reverse (with twist) +my @b=@_[5,1,2,6,3,7,0,4]; +$code.=<<___ + veor @b[1], @b[1], @b[7] + veor @b[4], @b[4], @b[7] + + veor @b[7], @b[7], @b[5] + veor @b[1], @b[1], @b[3] + veor @b[2], @b[2], @b[5] + veor @b[3], @b[3], @b[7] + + veor @b[6], @b[6], @b[1] + veor @b[2], @b[2], @b[0] + veor @b[5], @b[5], @b[3] + veor @b[4], @b[4], @b[6] + veor @b[0], @b[0], @b[6] + veor @b[1], @b[1], @b[4] +___ +} + +sub InvOutBasisChange { # InBasisChange in reverse +my @b=@_[2,5,7,3,6,1,0,4]; +$code.=<<___; + veor @b[1], @b[1], @b[5] + veor @b[2], @b[2], @b[7] + + veor @b[3], @b[3], @b[1] + veor @b[4], @b[4], @b[5] + veor @b[7], @b[7], @b[5] + veor @b[3], @b[3], @b[4] + veor @b[5], @b[5], @b[0] + veor @b[3], @b[3], @b[7] + veor @b[6], @b[6], @b[2] + veor @b[2], @b[2], @b[1] + veor @b[6], @b[6], @b[3] + + veor @b[3], @b[3], @b[0] + veor @b[5], @b[5], @b[6] +___ +} + +sub Mul_GF4 { +#;************************************************************* +#;* Mul_GF4: Input x0-x1,y0-y1 Output x0-x1 Temp t0 (8) * +#;************************************************************* +my ($x0,$x1,$y0,$y1,$t0,$t1)=@_; +$code.=<<___; + veor $t0, $y0, $y1 + vand $t0, $t0, $x0 + veor $x0, $x0, $x1 + vand $t1, $x1, $y0 + vand $x0, $x0, $y1 + veor $x1, $t1, $t0 + veor $x0, $x0, $t1 +___ +} + +sub Mul_GF4_N { # not used, see next subroutine +# multiply and scale by N +my ($x0,$x1,$y0,$y1,$t0)=@_; +$code.=<<___; + veor $t0, $y0, $y1 + vand $t0, $t0, $x0 + veor $x0, $x0, $x1 + vand $x1, $x1, $y0 + vand $x0, $x0, $y1 + veor $x1, $x1, $x0 + veor $x0, $x0, $t0 +___ +} + +sub Mul_GF4_N_GF4 { +# interleaved Mul_GF4_N and Mul_GF4 +my ($x0,$x1,$y0,$y1,$t0, + $x2,$x3,$y2,$y3,$t1)=@_; +$code.=<<___; + veor $t0, $y0, $y1 + veor $t1, $y2, $y3 + vand $t0, $t0, $x0 + vand $t1, $t1, $x2 + veor $x0, $x0, $x1 + veor $x2, $x2, $x3 + vand $x1, $x1, $y0 + vand $x3, $x3, $y2 + vand $x0, $x0, $y1 + vand $x2, $x2, $y3 + veor $x1, $x1, $x0 + veor $x2, $x2, $x3 + veor $x0, $x0, $t0 + veor $x3, $x3, $t1 +___ +} +sub Mul_GF16_2 { +my @x=@_[0..7]; +my @y=@_[8..11]; +my @t=@_[12..15]; +$code.=<<___; + veor @t[0], @x[0], @x[2] + veor @t[1], @x[1], @x[3] +___ + &Mul_GF4 (@x[0], @x[1], @y[0], @y[1], @t[2..3]); +$code.=<<___; + veor @y[0], @y[0], @y[2] + veor @y[1], @y[1], @y[3] +___ + Mul_GF4_N_GF4 (@t[0], @t[1], @y[0], @y[1], @t[3], + @x[2], @x[3], @y[2], @y[3], @t[2]); +$code.=<<___; + veor @x[0], @x[0], @t[0] + veor @x[2], @x[2], @t[0] + veor @x[1], @x[1], @t[1] + veor @x[3], @x[3], @t[1] + + veor @t[0], @x[4], @x[6] + veor @t[1], @x[5], @x[7] +___ + &Mul_GF4_N_GF4 (@t[0], @t[1], @y[0], @y[1], @t[3], + @x[6], @x[7], @y[2], @y[3], @t[2]); +$code.=<<___; + veor @y[0], @y[0], @y[2] + veor @y[1], @y[1], @y[3] +___ + &Mul_GF4 (@x[4], @x[5], @y[0], @y[1], @t[2..3]); +$code.=<<___; + veor @x[4], @x[4], @t[0] + veor @x[6], @x[6], @t[0] + veor @x[5], @x[5], @t[1] + veor @x[7], @x[7], @t[1] +___ +} +sub Inv_GF256 { +#;******************************************************************** +#;* Inv_GF256: Input x0-x7 Output x0-x7 Temp t0-t3,s0-s3 (144) * +#;******************************************************************** +my @x=@_[0..7]; +my @t=@_[8..11]; +my @s=@_[12..15]; +# direct optimizations from hardware +$code.=<<___; + veor @t[3], @x[4], @x[6] + veor @t[2], @x[5], @x[7] + veor @t[1], @x[1], @x[3] + veor @s[1], @x[7], @x[6] + vmov @t[0], @t[2] + veor @s[0], @x[0], @x[2] + + vorr @t[2], @t[2], @t[1] + veor @s[3], @t[3], @t[0] + vand @s[2], @t[3], @s[0] + vorr @t[3], @t[3], @s[0] + veor @s[0], @s[0], @t[1] + vand @t[0], @t[0], @t[1] + veor @t[1], @x[3], @x[2] + vand @s[3], @s[3], @s[0] + vand @s[1], @s[1], @t[1] + veor @t[1], @x[4], @x[5] + veor @s[0], @x[1], @x[0] + veor @t[3], @t[3], @s[1] + veor @t[2], @t[2], @s[1] + vand @s[1], @t[1], @s[0] + vorr @t[1], @t[1], @s[0] + veor @t[3], @t[3], @s[3] + veor @t[0], @t[0], @s[1] + veor @t[2], @t[2], @s[2] + veor @t[1], @t[1], @s[3] + veor @t[0], @t[0], @s[2] + vand @s[0], @x[7], @x[3] + veor @t[1], @t[1], @s[2] + vand @s[1], @x[6], @x[2] + vand @s[2], @x[5], @x[1] + vorr @s[3], @x[4], @x[0] + veor @t[3], @t[3], @s[0] + veor @t[1], @t[1], @s[2] + veor @t[0], @t[0], @s[3] + veor @t[2], @t[2], @s[1] + + @ Inv_GF16 \t0, \t1, \t2, \t3, \s0, \s1, \s2, \s3 + + @ new smaller inversion + + vand @s[2], @t[3], @t[1] + vmov @s[0], @t[0] + + veor @s[1], @t[2], @s[2] + veor @s[3], @t[0], @s[2] + veor @s[2], @t[0], @s[2] @ @s[2]=@s[3] + + vbsl @s[1], @t[1], @t[0] + vbsl @s[3], @t[3], @t[2] + veor @t[3], @t[3], @t[2] + + vbsl @s[0], @s[1], @s[2] + vbsl @t[0], @s[2], @s[1] + + vand @s[2], @s[0], @s[3] + veor @t[1], @t[1], @t[0] + + veor @s[2], @s[2], @t[3] +___ +# output in s3, s2, s1, t1 + +# Mul_GF16_2 \x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \t2, \t3, \t0, \t1, \s0, \s1, \s2, \s3 + +# Mul_GF16_2 \x0, \x1, \x2, \x3, \x4, \x5, \x6, \x7, \s3, \s2, \s1, \t1, \s0, \t0, \t2, \t3 + &Mul_GF16_2(@x,@s[3,2,1],@t[1],@s[0],@t[0,2,3]); + +### output msb > [x3,x2,x1,x0,x7,x6,x5,x4] < lsb +} + +# AES linear components + +sub ShiftRows { +my @x=@_[0..7]; +my @t=@_[8..11]; +my $mask=pop; +$code.=<<___; + vldmia $key!, {@t[0]-@t[3]} + veor @t[0], @t[0], @x[0] + veor @t[1], @t[1], @x[1] + vtbl.8 `&Dlo(@x[0])`, {@t[0]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[0])`, {@t[0]}, `&Dhi($mask)` + vldmia $key!, {@t[0]} + veor @t[2], @t[2], @x[2] + vtbl.8 `&Dlo(@x[1])`, {@t[1]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[1])`, {@t[1]}, `&Dhi($mask)` + vldmia $key!, {@t[1]} + veor @t[3], @t[3], @x[3] + vtbl.8 `&Dlo(@x[2])`, {@t[2]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[2])`, {@t[2]}, `&Dhi($mask)` + vldmia $key!, {@t[2]} + vtbl.8 `&Dlo(@x[3])`, {@t[3]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[3])`, {@t[3]}, `&Dhi($mask)` + vldmia $key!, {@t[3]} + veor @t[0], @t[0], @x[4] + veor @t[1], @t[1], @x[5] + vtbl.8 `&Dlo(@x[4])`, {@t[0]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[4])`, {@t[0]}, `&Dhi($mask)` + veor @t[2], @t[2], @x[6] + vtbl.8 `&Dlo(@x[5])`, {@t[1]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[5])`, {@t[1]}, `&Dhi($mask)` + veor @t[3], @t[3], @x[7] + vtbl.8 `&Dlo(@x[6])`, {@t[2]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[6])`, {@t[2]}, `&Dhi($mask)` + vtbl.8 `&Dlo(@x[7])`, {@t[3]}, `&Dlo($mask)` + vtbl.8 `&Dhi(@x[7])`, {@t[3]}, `&Dhi($mask)` +___ +} + +sub MixColumns { +# modified to emit output in order suitable for feeding back to aesenc[last] +my @x=@_[0..7]; +my @t=@_[8..15]; +my $inv=@_[16]; # optional +$code.=<<___; + vext.8 @t[0], @x[0], @x[0], #12 @ x0 <<< 32 + vext.8 @t[1], @x[1], @x[1], #12 + veor @x[0], @x[0], @t[0] @ x0 ^ (x0 <<< 32) + vext.8 @t[2], @x[2], @x[2], #12 + veor @x[1], @x[1], @t[1] + vext.8 @t[3], @x[3], @x[3], #12 + veor @x[2], @x[2], @t[2] + vext.8 @t[4], @x[4], @x[4], #12 + veor @x[3], @x[3], @t[3] + vext.8 @t[5], @x[5], @x[5], #12 + veor @x[4], @x[4], @t[4] + vext.8 @t[6], @x[6], @x[6], #12 + veor @x[5], @x[5], @t[5] + vext.8 @t[7], @x[7], @x[7], #12 + veor @x[6], @x[6], @t[6] + + veor @t[1], @t[1], @x[0] + veor @x[7], @x[7], @t[7] + vext.8 @x[0], @x[0], @x[0], #8 @ (x0 ^ (x0 <<< 32)) <<< 64) + veor @t[2], @t[2], @x[1] + veor @t[0], @t[0], @x[7] + veor @t[1], @t[1], @x[7] + vext.8 @x[1], @x[1], @x[1], #8 + veor @t[5], @t[5], @x[4] + veor @x[0], @x[0], @t[0] + veor @t[6], @t[6], @x[5] + veor @x[1], @x[1], @t[1] + vext.8 @t[0], @x[4], @x[4], #8 + veor @t[4], @t[4], @x[3] + vext.8 @t[1], @x[5], @x[5], #8 + veor @t[7], @t[7], @x[6] + vext.8 @x[4], @x[3], @x[3], #8 + veor @t[3], @t[3], @x[2] + vext.8 @x[5], @x[7], @x[7], #8 + veor @t[4], @t[4], @x[7] + vext.8 @x[3], @x[6], @x[6], #8 + veor @t[3], @t[3], @x[7] + vext.8 @x[6], @x[2], @x[2], #8 + veor @x[7], @t[1], @t[5] +___ +$code.=<<___ if (!$inv); + veor @x[2], @t[0], @t[4] + veor @x[4], @x[4], @t[3] + veor @x[5], @x[5], @t[7] + veor @x[3], @x[3], @t[6] + @ vmov @x[2], @t[0] + veor @x[6], @x[6], @t[2] + @ vmov @x[7], @t[1] +___ +$code.=<<___ if ($inv); + veor @t[3], @t[3], @x[4] + veor @x[5], @x[5], @t[7] + veor @x[2], @x[3], @t[6] + veor @x[3], @t[0], @t[4] + veor @x[4], @x[6], @t[2] + vmov @x[6], @t[3] + @ vmov @x[7], @t[1] +___ +} + +sub InvMixColumns_orig { +my @x=@_[0..7]; +my @t=@_[8..15]; + +$code.=<<___; + @ multiplication by 0x0e + vext.8 @t[7], @x[7], @x[7], #12 + vmov @t[2], @x[2] + veor @x[2], @x[2], @x[5] @ 2 5 + veor @x[7], @x[7], @x[5] @ 7 5 + vext.8 @t[0], @x[0], @x[0], #12 + vmov @t[5], @x[5] + veor @x[5], @x[5], @x[0] @ 5 0 [1] + veor @x[0], @x[0], @x[1] @ 0 1 + vext.8 @t[1], @x[1], @x[1], #12 + veor @x[1], @x[1], @x[2] @ 1 25 + veor @x[0], @x[0], @x[6] @ 01 6 [2] + vext.8 @t[3], @x[3], @x[3], #12 + veor @x[1], @x[1], @x[3] @ 125 3 [4] + veor @x[2], @x[2], @x[0] @ 25 016 [3] + veor @x[3], @x[3], @x[7] @ 3 75 + veor @x[7], @x[7], @x[6] @ 75 6 [0] + vext.8 @t[6], @x[6], @x[6], #12 + vmov @t[4], @x[4] + veor @x[6], @x[6], @x[4] @ 6 4 + veor @x[4], @x[4], @x[3] @ 4 375 [6] + veor @x[3], @x[3], @x[7] @ 375 756=36 + veor @x[6], @x[6], @t[5] @ 64 5 [7] + veor @x[3], @x[3], @t[2] @ 36 2 + vext.8 @t[5], @t[5], @t[5], #12 + veor @x[3], @x[3], @t[4] @ 362 4 [5] +___ + my @y = @x[7,5,0,2,1,3,4,6]; +$code.=<<___; + @ multiplication by 0x0b + veor @y[1], @y[1], @y[0] + veor @y[0], @y[0], @t[0] + vext.8 @t[2], @t[2], @t[2], #12 + veor @y[1], @y[1], @t[1] + veor @y[0], @y[0], @t[5] + vext.8 @t[4], @t[4], @t[4], #12 + veor @y[1], @y[1], @t[6] + veor @y[0], @y[0], @t[7] + veor @t[7], @t[7], @t[6] @ clobber t[7] + + veor @y[3], @y[3], @t[0] + veor @y[1], @y[1], @y[0] + vext.8 @t[0], @t[0], @t[0], #12 + veor @y[2], @y[2], @t[1] + veor @y[4], @y[4], @t[1] + vext.8 @t[1], @t[1], @t[1], #12 + veor @y[2], @y[2], @t[2] + veor @y[3], @y[3], @t[2] + veor @y[5], @y[5], @t[2] + veor @y[2], @y[2], @t[7] + vext.8 @t[2], @t[2], @t[2], #12 + veor @y[3], @y[3], @t[3] + veor @y[6], @y[6], @t[3] + veor @y[4], @y[4], @t[3] + veor @y[7], @y[7], @t[4] + vext.8 @t[3], @t[3], @t[3], #12 + veor @y[5], @y[5], @t[4] + veor @y[7], @y[7], @t[7] + veor @t[7], @t[7], @t[5] @ clobber t[7] even more + veor @y[3], @y[3], @t[5] + veor @y[4], @y[4], @t[4] + + veor @y[5], @y[5], @t[7] + vext.8 @t[4], @t[4], @t[4], #12 + veor @y[6], @y[6], @t[7] + veor @y[4], @y[4], @t[7] + + veor @t[7], @t[7], @t[5] + vext.8 @t[5], @t[5], @t[5], #12 + + @ multiplication by 0x0d + veor @y[4], @y[4], @y[7] + veor @t[7], @t[7], @t[6] @ restore t[7] + veor @y[7], @y[7], @t[4] + vext.8 @t[6], @t[6], @t[6], #12 + veor @y[2], @y[2], @t[0] + veor @y[7], @y[7], @t[5] + vext.8 @t[7], @t[7], @t[7], #12 + veor @y[2], @y[2], @t[2] + + veor @y[3], @y[3], @y[1] + veor @y[1], @y[1], @t[1] + veor @y[0], @y[0], @t[0] + veor @y[3], @y[3], @t[0] + veor @y[1], @y[1], @t[5] + veor @y[0], @y[0], @t[5] + vext.8 @t[0], @t[0], @t[0], #12 + veor @y[1], @y[1], @t[7] + veor @y[0], @y[0], @t[6] + veor @y[3], @y[3], @y[1] + veor @y[4], @y[4], @t[1] + vext.8 @t[1], @t[1], @t[1], #12 + + veor @y[7], @y[7], @t[7] + veor @y[4], @y[4], @t[2] + veor @y[5], @y[5], @t[2] + veor @y[2], @y[2], @t[6] + veor @t[6], @t[6], @t[3] @ clobber t[6] + vext.8 @t[2], @t[2], @t[2], #12 + veor @y[4], @y[4], @y[7] + veor @y[3], @y[3], @t[6] + + veor @y[6], @y[6], @t[6] + veor @y[5], @y[5], @t[5] + vext.8 @t[5], @t[5], @t[5], #12 + veor @y[6], @y[6], @t[4] + vext.8 @t[4], @t[4], @t[4], #12 + veor @y[5], @y[5], @t[6] + veor @y[6], @y[6], @t[7] + vext.8 @t[7], @t[7], @t[7], #12 + veor @t[6], @t[6], @t[3] @ restore t[6] + vext.8 @t[3], @t[3], @t[3], #12 + + @ multiplication by 0x09 + veor @y[4], @y[4], @y[1] + veor @t[1], @t[1], @y[1] @ t[1]=y[1] + veor @t[0], @t[0], @t[5] @ clobber t[0] + vext.8 @t[6], @t[6], @t[6], #12 + veor @t[1], @t[1], @t[5] + veor @y[3], @y[3], @t[0] + veor @t[0], @t[0], @y[0] @ t[0]=y[0] + veor @t[1], @t[1], @t[6] + veor @t[6], @t[6], @t[7] @ clobber t[6] + veor @y[4], @y[4], @t[1] + veor @y[7], @y[7], @t[4] + veor @y[6], @y[6], @t[3] + veor @y[5], @y[5], @t[2] + veor @t[4], @t[4], @y[4] @ t[4]=y[4] + veor @t[3], @t[3], @y[3] @ t[3]=y[3] + veor @t[5], @t[5], @y[5] @ t[5]=y[5] + veor @t[2], @t[2], @y[2] @ t[2]=y[2] + veor @t[3], @t[3], @t[7] + veor @XMM[5], @t[5], @t[6] + veor @XMM[6], @t[6], @y[6] @ t[6]=y[6] + veor @XMM[2], @t[2], @t[6] + veor @XMM[7], @t[7], @y[7] @ t[7]=y[7] + + vmov @XMM[0], @t[0] + vmov @XMM[1], @t[1] + @ vmov @XMM[2], @t[2] + vmov @XMM[3], @t[3] + vmov @XMM[4], @t[4] + @ vmov @XMM[5], @t[5] + @ vmov @XMM[6], @t[6] + @ vmov @XMM[7], @t[7] +___ +} + +sub InvMixColumns { +my @x=@_[0..7]; +my @t=@_[8..15]; + +# Thanks to Jussi Kivilinna for providing pointer to +# +# | 0e 0b 0d 09 | | 02 03 01 01 | | 05 00 04 00 | +# | 09 0e 0b 0d | = | 01 02 03 01 | x | 00 05 00 04 | +# | 0d 09 0e 0b | | 01 01 02 03 | | 04 00 05 00 | +# | 0b 0d 09 0e | | 03 01 01 02 | | 00 04 00 05 | + +$code.=<<___; + @ multiplication by 0x05-0x00-0x04-0x00 + vext.8 @t[0], @x[0], @x[0], #8 + vext.8 @t[6], @x[6], @x[6], #8 + vext.8 @t[7], @x[7], @x[7], #8 + veor @t[0], @t[0], @x[0] + vext.8 @t[1], @x[1], @x[1], #8 + veor @t[6], @t[6], @x[6] + vext.8 @t[2], @x[2], @x[2], #8 + veor @t[7], @t[7], @x[7] + vext.8 @t[3], @x[3], @x[3], #8 + veor @t[1], @t[1], @x[1] + vext.8 @t[4], @x[4], @x[4], #8 + veor @t[2], @t[2], @x[2] + vext.8 @t[5], @x[5], @x[5], #8 + veor @t[3], @t[3], @x[3] + veor @t[4], @t[4], @x[4] + veor @t[5], @t[5], @x[5] + + veor @x[0], @x[0], @t[6] + veor @x[1], @x[1], @t[6] + veor @x[2], @x[2], @t[0] + veor @x[4], @x[4], @t[2] + veor @x[3], @x[3], @t[1] + veor @x[1], @x[1], @t[7] + veor @x[2], @x[2], @t[7] + veor @x[4], @x[4], @t[6] + veor @x[5], @x[5], @t[3] + veor @x[3], @x[3], @t[6] + veor @x[6], @x[6], @t[4] + veor @x[4], @x[4], @t[7] + veor @x[5], @x[5], @t[7] + veor @x[7], @x[7], @t[5] +___ + &MixColumns (@x,@t,1); # flipped 2<->3 and 4<->6 +} + +sub swapmove { +my ($a,$b,$n,$mask,$t)=@_; +$code.=<<___; + vshr.u64 $t, $b, #$n + veor $t, $t, $a + vand $t, $t, $mask + veor $a, $a, $t + vshl.u64 $t, $t, #$n + veor $b, $b, $t +___ +} +sub swapmove2x { +my ($a0,$b0,$a1,$b1,$n,$mask,$t0,$t1)=@_; +$code.=<<___; + vshr.u64 $t0, $b0, #$n + vshr.u64 $t1, $b1, #$n + veor $t0, $t0, $a0 + veor $t1, $t1, $a1 + vand $t0, $t0, $mask + vand $t1, $t1, $mask + veor $a0, $a0, $t0 + vshl.u64 $t0, $t0, #$n + veor $a1, $a1, $t1 + vshl.u64 $t1, $t1, #$n + veor $b0, $b0, $t0 + veor $b1, $b1, $t1 +___ +} + +sub bitslice { +my @x=reverse(@_[0..7]); +my ($t0,$t1,$t2,$t3)=@_[8..11]; +$code.=<<___; + vmov.i8 $t0,#0x55 @ compose .LBS0 + vmov.i8 $t1,#0x33 @ compose .LBS1 +___ + &swapmove2x(@x[0,1,2,3],1,$t0,$t2,$t3); + &swapmove2x(@x[4,5,6,7],1,$t0,$t2,$t3); +$code.=<<___; + vmov.i8 $t0,#0x0f @ compose .LBS2 +___ + &swapmove2x(@x[0,2,1,3],2,$t1,$t2,$t3); + &swapmove2x(@x[4,6,5,7],2,$t1,$t2,$t3); + + &swapmove2x(@x[0,4,1,5],4,$t0,$t2,$t3); + &swapmove2x(@x[2,6,3,7],4,$t0,$t2,$t3); +} + +$code.=<<___; +#ifndef __KERNEL__ +# include "arm_arch.h" + +# define VFP_ABI_PUSH vstmdb sp!,{d8-d15} +# define VFP_ABI_POP vldmia sp!,{d8-d15} +# define VFP_ABI_FRAME 0x40 +#else +# define VFP_ABI_PUSH +# define VFP_ABI_POP +# define VFP_ABI_FRAME 0 +# define BSAES_ASM_EXTENDED_KEY +# define XTS_CHAIN_TWEAK +# define __ARM_ARCH__ 7 +#endif + +#ifdef __thumb__ +# define adrl adr +#endif + +#if __ARM_ARCH__>=7 +.text +.syntax unified @ ARMv7-capable assembler is expected to handle this +#ifdef __thumb2__ +.thumb +#else +.code 32 +#endif + +.fpu neon + +.type _bsaes_decrypt8,%function +.align 4 +_bsaes_decrypt8: + adr $const,_bsaes_decrypt8 + vldmia $key!, {@XMM[9]} @ round 0 key + add $const,$const,#.LM0ISR-_bsaes_decrypt8 + + vldmia $const!, {@XMM[8]} @ .LM0ISR + veor @XMM[10], @XMM[0], @XMM[9] @ xor with round0 key + veor @XMM[11], @XMM[1], @XMM[9] + vtbl.8 `&Dlo(@XMM[0])`, {@XMM[10]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[0])`, {@XMM[10]}, `&Dhi(@XMM[8])` + veor @XMM[12], @XMM[2], @XMM[9] + vtbl.8 `&Dlo(@XMM[1])`, {@XMM[11]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[1])`, {@XMM[11]}, `&Dhi(@XMM[8])` + veor @XMM[13], @XMM[3], @XMM[9] + vtbl.8 `&Dlo(@XMM[2])`, {@XMM[12]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[2])`, {@XMM[12]}, `&Dhi(@XMM[8])` + veor @XMM[14], @XMM[4], @XMM[9] + vtbl.8 `&Dlo(@XMM[3])`, {@XMM[13]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[3])`, {@XMM[13]}, `&Dhi(@XMM[8])` + veor @XMM[15], @XMM[5], @XMM[9] + vtbl.8 `&Dlo(@XMM[4])`, {@XMM[14]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[4])`, {@XMM[14]}, `&Dhi(@XMM[8])` + veor @XMM[10], @XMM[6], @XMM[9] + vtbl.8 `&Dlo(@XMM[5])`, {@XMM[15]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[5])`, {@XMM[15]}, `&Dhi(@XMM[8])` + veor @XMM[11], @XMM[7], @XMM[9] + vtbl.8 `&Dlo(@XMM[6])`, {@XMM[10]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[6])`, {@XMM[10]}, `&Dhi(@XMM[8])` + vtbl.8 `&Dlo(@XMM[7])`, {@XMM[11]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[7])`, {@XMM[11]}, `&Dhi(@XMM[8])` +___ + &bitslice (@XMM[0..7, 8..11]); +$code.=<<___; + sub $rounds,$rounds,#1 + b .Ldec_sbox +.align 4 +.Ldec_loop: +___ + &ShiftRows (@XMM[0..7, 8..12]); +$code.=".Ldec_sbox:\n"; + &InvSbox (@XMM[0..7, 8..15]); +$code.=<<___; + subs $rounds,$rounds,#1 + bcc .Ldec_done +___ + &InvMixColumns (@XMM[0,1,6,4,2,7,3,5, 8..15]); +$code.=<<___; + vldmia $const, {@XMM[12]} @ .LISR + ite eq @ Thumb2 thing, sanity check in ARM + addeq $const,$const,#0x10 + bne .Ldec_loop + vldmia $const, {@XMM[12]} @ .LISRM0 + b .Ldec_loop +.align 4 +.Ldec_done: +___ + &bitslice (@XMM[0,1,6,4,2,7,3,5, 8..11]); +$code.=<<___; + vldmia $key, {@XMM[8]} @ last round key + veor @XMM[6], @XMM[6], @XMM[8] + veor @XMM[4], @XMM[4], @XMM[8] + veor @XMM[2], @XMM[2], @XMM[8] + veor @XMM[7], @XMM[7], @XMM[8] + veor @XMM[3], @XMM[3], @XMM[8] + veor @XMM[5], @XMM[5], @XMM[8] + veor @XMM[0], @XMM[0], @XMM[8] + veor @XMM[1], @XMM[1], @XMM[8] + bx lr +.size _bsaes_decrypt8,.-_bsaes_decrypt8 + +.type _bsaes_const,%object +.align 6 +_bsaes_const: +.LM0ISR: @ InvShiftRows constants + .quad 0x0a0e0206070b0f03, 0x0004080c0d010509 +.LISR: + .quad 0x0504070602010003, 0x0f0e0d0c080b0a09 +.LISRM0: + .quad 0x01040b0e0205080f, 0x0306090c00070a0d +.LM0SR: @ ShiftRows constants + .quad 0x0a0e02060f03070b, 0x0004080c05090d01 +.LSR: + .quad 0x0504070600030201, 0x0f0e0d0c0a09080b +.LSRM0: + .quad 0x0304090e00050a0f, 0x01060b0c0207080d +.LM0: + .quad 0x02060a0e03070b0f, 0x0004080c0105090d +.LREVM0SR: + .quad 0x090d01050c000408, 0x03070b0f060a0e02 +.asciz "Bit-sliced AES for NEON, CRYPTOGAMS by <appro\@openssl.org>" +.align 6 +.size _bsaes_const,.-_bsaes_const + +.type _bsaes_encrypt8,%function +.align 4 +_bsaes_encrypt8: + adr $const,_bsaes_encrypt8 + vldmia $key!, {@XMM[9]} @ round 0 key + sub $const,$const,#_bsaes_encrypt8-.LM0SR + + vldmia $const!, {@XMM[8]} @ .LM0SR +_bsaes_encrypt8_alt: + veor @XMM[10], @XMM[0], @XMM[9] @ xor with round0 key + veor @XMM[11], @XMM[1], @XMM[9] + vtbl.8 `&Dlo(@XMM[0])`, {@XMM[10]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[0])`, {@XMM[10]}, `&Dhi(@XMM[8])` + veor @XMM[12], @XMM[2], @XMM[9] + vtbl.8 `&Dlo(@XMM[1])`, {@XMM[11]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[1])`, {@XMM[11]}, `&Dhi(@XMM[8])` + veor @XMM[13], @XMM[3], @XMM[9] + vtbl.8 `&Dlo(@XMM[2])`, {@XMM[12]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[2])`, {@XMM[12]}, `&Dhi(@XMM[8])` + veor @XMM[14], @XMM[4], @XMM[9] + vtbl.8 `&Dlo(@XMM[3])`, {@XMM[13]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[3])`, {@XMM[13]}, `&Dhi(@XMM[8])` + veor @XMM[15], @XMM[5], @XMM[9] + vtbl.8 `&Dlo(@XMM[4])`, {@XMM[14]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[4])`, {@XMM[14]}, `&Dhi(@XMM[8])` + veor @XMM[10], @XMM[6], @XMM[9] + vtbl.8 `&Dlo(@XMM[5])`, {@XMM[15]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[5])`, {@XMM[15]}, `&Dhi(@XMM[8])` + veor @XMM[11], @XMM[7], @XMM[9] + vtbl.8 `&Dlo(@XMM[6])`, {@XMM[10]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[6])`, {@XMM[10]}, `&Dhi(@XMM[8])` + vtbl.8 `&Dlo(@XMM[7])`, {@XMM[11]}, `&Dlo(@XMM[8])` + vtbl.8 `&Dhi(@XMM[7])`, {@XMM[11]}, `&Dhi(@XMM[8])` +_bsaes_encrypt8_bitslice: +___ + &bitslice (@XMM[0..7, 8..11]); +$code.=<<___; + sub $rounds,$rounds,#1 + b .Lenc_sbox +.align 4 +.Lenc_loop: +___ + &ShiftRows (@XMM[0..7, 8..12]); +$code.=".Lenc_sbox:\n"; + &Sbox (@XMM[0..7, 8..15]); +$code.=<<___; + subs $rounds,$rounds,#1 + bcc .Lenc_done +___ + &MixColumns (@XMM[0,1,4,6,3,7,2,5, 8..15]); +$code.=<<___; + vldmia $const, {@XMM[12]} @ .LSR + ite eq @ Thumb2 thing, samity check in ARM + addeq $const,$const,#0x10 + bne .Lenc_loop + vldmia $const, {@XMM[12]} @ .LSRM0 + b .Lenc_loop +.align 4 +.Lenc_done: +___ + # output in lsb > [t0, t1, t4, t6, t3, t7, t2, t5] < msb + &bitslice (@XMM[0,1,4,6,3,7,2,5, 8..11]); +$code.=<<___; + vldmia $key, {@XMM[8]} @ last round key + veor @XMM[4], @XMM[4], @XMM[8] + veor @XMM[6], @XMM[6], @XMM[8] + veor @XMM[3], @XMM[3], @XMM[8] + veor @XMM[7], @XMM[7], @XMM[8] + veor @XMM[2], @XMM[2], @XMM[8] + veor @XMM[5], @XMM[5], @XMM[8] + veor @XMM[0], @XMM[0], @XMM[8] + veor @XMM[1], @XMM[1], @XMM[8] + bx lr +.size _bsaes_encrypt8,.-_bsaes_encrypt8 +___ +} +{ +my ($out,$inp,$rounds,$const)=("r12","r4","r5","r6"); + +sub bitslice_key { +my @x=reverse(@_[0..7]); +my ($bs0,$bs1,$bs2,$t2,$t3)=@_[8..12]; + + &swapmove (@x[0,1],1,$bs0,$t2,$t3); +$code.=<<___; + @ &swapmove(@x[2,3],1,$t0,$t2,$t3); + vmov @x[2], @x[0] + vmov @x[3], @x[1] +___ + #&swapmove2x(@x[4,5,6,7],1,$t0,$t2,$t3); + + &swapmove2x (@x[0,2,1,3],2,$bs1,$t2,$t3); +$code.=<<___; + @ &swapmove2x(@x[4,6,5,7],2,$t1,$t2,$t3); + vmov @x[4], @x[0] + vmov @x[6], @x[2] + vmov @x[5], @x[1] + vmov @x[7], @x[3] +___ + &swapmove2x (@x[0,4,1,5],4,$bs2,$t2,$t3); + &swapmove2x (@x[2,6,3,7],4,$bs2,$t2,$t3); +} + +$code.=<<___; +.type _bsaes_key_convert,%function +.align 4 +_bsaes_key_convert: + adr $const,_bsaes_key_convert + vld1.8 {@XMM[7]}, [$inp]! @ load round 0 key + sub $const,$const,#_bsaes_key_convert-.LM0 + vld1.8 {@XMM[15]}, [$inp]! @ load round 1 key + + vmov.i8 @XMM[8], #0x01 @ bit masks + vmov.i8 @XMM[9], #0x02 + vmov.i8 @XMM[10], #0x04 + vmov.i8 @XMM[11], #0x08 + vmov.i8 @XMM[12], #0x10 + vmov.i8 @XMM[13], #0x20 + vldmia $const, {@XMM[14]} @ .LM0 + +#ifdef __ARMEL__ + vrev32.8 @XMM[7], @XMM[7] + vrev32.8 @XMM[15], @XMM[15] +#endif + sub $rounds,$rounds,#1 + vstmia $out!, {@XMM[7]} @ save round 0 key + b .Lkey_loop + +.align 4 +.Lkey_loop: + vtbl.8 `&Dlo(@XMM[7])`,{@XMM[15]},`&Dlo(@XMM[14])` + vtbl.8 `&Dhi(@XMM[7])`,{@XMM[15]},`&Dhi(@XMM[14])` + vmov.i8 @XMM[6], #0x40 + vmov.i8 @XMM[15], #0x80 + + vtst.8 @XMM[0], @XMM[7], @XMM[8] + vtst.8 @XMM[1], @XMM[7], @XMM[9] + vtst.8 @XMM[2], @XMM[7], @XMM[10] + vtst.8 @XMM[3], @XMM[7], @XMM[11] + vtst.8 @XMM[4], @XMM[7], @XMM[12] + vtst.8 @XMM[5], @XMM[7], @XMM[13] + vtst.8 @XMM[6], @XMM[7], @XMM[6] + vtst.8 @XMM[7], @XMM[7], @XMM[15] + vld1.8 {@XMM[15]}, [$inp]! @ load next round key + vmvn @XMM[0], @XMM[0] @ "pnot" + vmvn @XMM[1], @XMM[1] + vmvn @XMM[5], @XMM[5] + vmvn @XMM[6], @XMM[6] +#ifdef __ARMEL__ + vrev32.8 @XMM[15], @XMM[15] +#endif + subs $rounds,$rounds,#1 + vstmia $out!,{@XMM[0]-@XMM[7]} @ write bit-sliced round key + bne .Lkey_loop + + vmov.i8 @XMM[7],#0x63 @ compose .L63 + @ don't save last round key + bx lr +.size _bsaes_key_convert,.-_bsaes_key_convert +___ +} + +if (0) { # following four functions are unsupported interface + # used for benchmarking... +$code.=<<___; +.globl bsaes_enc_key_convert +.type bsaes_enc_key_convert,%function +.align 4 +bsaes_enc_key_convert: + stmdb sp!,{r4-r6,lr} + vstmdb sp!,{d8-d15} @ ABI specification says so + + ldr r5,[$inp,#240] @ pass rounds + mov r4,$inp @ pass key + mov r12,$out @ pass key schedule + bl _bsaes_key_convert + veor @XMM[7],@XMM[7],@XMM[15] @ fix up last round key + vstmia r12, {@XMM[7]} @ save last round key + + vldmia sp!,{d8-d15} + ldmia sp!,{r4-r6,pc} +.size bsaes_enc_key_convert,.-bsaes_enc_key_convert + +.globl bsaes_encrypt_128 +.type bsaes_encrypt_128,%function +.align 4 +bsaes_encrypt_128: + stmdb sp!,{r4-r6,lr} + vstmdb sp!,{d8-d15} @ ABI specification says so +.Lenc128_loop: + vld1.8 {@XMM[0]-@XMM[1]}, [$inp]! @ load input + vld1.8 {@XMM[2]-@XMM[3]}, [$inp]! + mov r4,$key @ pass the key + vld1.8 {@XMM[4]-@XMM[5]}, [$inp]! + mov r5,#10 @ pass rounds + vld1.8 {@XMM[6]-@XMM[7]}, [$inp]! + + bl _bsaes_encrypt8 + + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + vst1.8 {@XMM[4]}, [$out]! + vst1.8 {@XMM[6]}, [$out]! + vst1.8 {@XMM[3]}, [$out]! + vst1.8 {@XMM[7]}, [$out]! + vst1.8 {@XMM[2]}, [$out]! + subs $len,$len,#0x80 + vst1.8 {@XMM[5]}, [$out]! + bhi .Lenc128_loop + + vldmia sp!,{d8-d15} + ldmia sp!,{r4-r6,pc} +.size bsaes_encrypt_128,.-bsaes_encrypt_128 + +.globl bsaes_dec_key_convert +.type bsaes_dec_key_convert,%function +.align 4 +bsaes_dec_key_convert: + stmdb sp!,{r4-r6,lr} + vstmdb sp!,{d8-d15} @ ABI specification says so + + ldr r5,[$inp,#240] @ pass rounds + mov r4,$inp @ pass key + mov r12,$out @ pass key schedule + bl _bsaes_key_convert + vldmia $out, {@XMM[6]} + vstmia r12, {@XMM[15]} @ save last round key + veor @XMM[7], @XMM[7], @XMM[6] @ fix up round 0 key + vstmia $out, {@XMM[7]} + + vldmia sp!,{d8-d15} + ldmia sp!,{r4-r6,pc} +.size bsaes_dec_key_convert,.-bsaes_dec_key_convert + +.globl bsaes_decrypt_128 +.type bsaes_decrypt_128,%function +.align 4 +bsaes_decrypt_128: + stmdb sp!,{r4-r6,lr} + vstmdb sp!,{d8-d15} @ ABI specification says so +.Ldec128_loop: + vld1.8 {@XMM[0]-@XMM[1]}, [$inp]! @ load input + vld1.8 {@XMM[2]-@XMM[3]}, [$inp]! + mov r4,$key @ pass the key + vld1.8 {@XMM[4]-@XMM[5]}, [$inp]! + mov r5,#10 @ pass rounds + vld1.8 {@XMM[6]-@XMM[7]}, [$inp]! + + bl _bsaes_decrypt8 + + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + vst1.8 {@XMM[6]}, [$out]! + vst1.8 {@XMM[4]}, [$out]! + vst1.8 {@XMM[2]}, [$out]! + vst1.8 {@XMM[7]}, [$out]! + vst1.8 {@XMM[3]}, [$out]! + subs $len,$len,#0x80 + vst1.8 {@XMM[5]}, [$out]! + bhi .Ldec128_loop + + vldmia sp!,{d8-d15} + ldmia sp!,{r4-r6,pc} +.size bsaes_decrypt_128,.-bsaes_decrypt_128 +___ +} +{ +my ($inp,$out,$len,$key, $ivp,$fp,$rounds)=map("r$_",(0..3,8..10)); +my ($keysched)=("sp"); + +$code.=<<___; +.extern AES_cbc_encrypt +.extern AES_decrypt + +.global bsaes_cbc_encrypt +.type bsaes_cbc_encrypt,%function +.align 5 +bsaes_cbc_encrypt: +#ifndef __KERNEL__ + cmp $len, #128 +#ifndef __thumb__ + blo AES_cbc_encrypt +#else + bhs 1f + b AES_cbc_encrypt +1: +#endif +#endif + + @ it is up to the caller to make sure we are called with enc == 0 + + mov ip, sp + stmdb sp!, {r4-r10, lr} + VFP_ABI_PUSH + ldr $ivp, [ip] @ IV is 1st arg on the stack + mov $len, $len, lsr#4 @ len in 16 byte blocks + sub sp, #0x10 @ scratch space to carry over the IV + mov $fp, sp @ save sp + + ldr $rounds, [$key, #240] @ get # of rounds +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, $rounds, lsl#7 @ 128 bytes per inner round key + add r12, #`128-32` @ sifze of bit-slices key schedule + + @ populate the key schedule + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + mov sp, r12 @ sp is $keysched + bl _bsaes_key_convert + vldmia $keysched, {@XMM[6]} + vstmia r12, {@XMM[15]} @ save last round key + veor @XMM[7], @XMM[7], @XMM[6] @ fix up round 0 key + vstmia $keysched, {@XMM[7]} +#else + ldr r12, [$key, #244] + eors r12, #1 + beq 0f + + @ populate the key schedule + str r12, [$key, #244] + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + add r12, $key, #248 @ pass key schedule + bl _bsaes_key_convert + add r4, $key, #248 + vldmia r4, {@XMM[6]} + vstmia r12, {@XMM[15]} @ save last round key + veor @XMM[7], @XMM[7], @XMM[6] @ fix up round 0 key + vstmia r4, {@XMM[7]} + +.align 2 +0: +#endif + + vld1.8 {@XMM[15]}, [$ivp] @ load IV + b .Lcbc_dec_loop + +.align 4 +.Lcbc_dec_loop: + subs $len, $len, #0x8 + bmi .Lcbc_dec_loop_finish + + vld1.8 {@XMM[0]-@XMM[1]}, [$inp]! @ load input + vld1.8 {@XMM[2]-@XMM[3]}, [$inp]! +#ifndef BSAES_ASM_EXTENDED_KEY + mov r4, $keysched @ pass the key +#else + add r4, $key, #248 +#endif + vld1.8 {@XMM[4]-@XMM[5]}, [$inp]! + mov r5, $rounds + vld1.8 {@XMM[6]-@XMM[7]}, [$inp] + sub $inp, $inp, #0x60 + vstmia $fp, {@XMM[15]} @ put aside IV + + bl _bsaes_decrypt8 + + vldmia $fp, {@XMM[14]} @ reload IV + vld1.8 {@XMM[8]-@XMM[9]}, [$inp]! @ reload input + veor @XMM[0], @XMM[0], @XMM[14] @ ^= IV + vld1.8 {@XMM[10]-@XMM[11]}, [$inp]! + veor @XMM[1], @XMM[1], @XMM[8] + veor @XMM[6], @XMM[6], @XMM[9] + vld1.8 {@XMM[12]-@XMM[13]}, [$inp]! + veor @XMM[4], @XMM[4], @XMM[10] + veor @XMM[2], @XMM[2], @XMM[11] + vld1.8 {@XMM[14]-@XMM[15]}, [$inp]! + veor @XMM[7], @XMM[7], @XMM[12] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + veor @XMM[3], @XMM[3], @XMM[13] + vst1.8 {@XMM[6]}, [$out]! + veor @XMM[5], @XMM[5], @XMM[14] + vst1.8 {@XMM[4]}, [$out]! + vst1.8 {@XMM[2]}, [$out]! + vst1.8 {@XMM[7]}, [$out]! + vst1.8 {@XMM[3]}, [$out]! + vst1.8 {@XMM[5]}, [$out]! + + b .Lcbc_dec_loop + +.Lcbc_dec_loop_finish: + adds $len, $len, #8 + beq .Lcbc_dec_done + + vld1.8 {@XMM[0]}, [$inp]! @ load input + cmp $len, #2 + blo .Lcbc_dec_one + vld1.8 {@XMM[1]}, [$inp]! +#ifndef BSAES_ASM_EXTENDED_KEY + mov r4, $keysched @ pass the key +#else + add r4, $key, #248 +#endif + mov r5, $rounds + vstmia $fp, {@XMM[15]} @ put aside IV + beq .Lcbc_dec_two + vld1.8 {@XMM[2]}, [$inp]! + cmp $len, #4 + blo .Lcbc_dec_three + vld1.8 {@XMM[3]}, [$inp]! + beq .Lcbc_dec_four + vld1.8 {@XMM[4]}, [$inp]! + cmp $len, #6 + blo .Lcbc_dec_five + vld1.8 {@XMM[5]}, [$inp]! + beq .Lcbc_dec_six + vld1.8 {@XMM[6]}, [$inp]! + sub $inp, $inp, #0x70 + + bl _bsaes_decrypt8 + + vldmia $fp, {@XMM[14]} @ reload IV + vld1.8 {@XMM[8]-@XMM[9]}, [$inp]! @ reload input + veor @XMM[0], @XMM[0], @XMM[14] @ ^= IV + vld1.8 {@XMM[10]-@XMM[11]}, [$inp]! + veor @XMM[1], @XMM[1], @XMM[8] + veor @XMM[6], @XMM[6], @XMM[9] + vld1.8 {@XMM[12]-@XMM[13]}, [$inp]! + veor @XMM[4], @XMM[4], @XMM[10] + veor @XMM[2], @XMM[2], @XMM[11] + vld1.8 {@XMM[15]}, [$inp]! + veor @XMM[7], @XMM[7], @XMM[12] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + veor @XMM[3], @XMM[3], @XMM[13] + vst1.8 {@XMM[6]}, [$out]! + vst1.8 {@XMM[4]}, [$out]! + vst1.8 {@XMM[2]}, [$out]! + vst1.8 {@XMM[7]}, [$out]! + vst1.8 {@XMM[3]}, [$out]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_six: + sub $inp, $inp, #0x60 + bl _bsaes_decrypt8 + vldmia $fp,{@XMM[14]} @ reload IV + vld1.8 {@XMM[8]-@XMM[9]}, [$inp]! @ reload input + veor @XMM[0], @XMM[0], @XMM[14] @ ^= IV + vld1.8 {@XMM[10]-@XMM[11]}, [$inp]! + veor @XMM[1], @XMM[1], @XMM[8] + veor @XMM[6], @XMM[6], @XMM[9] + vld1.8 {@XMM[12]}, [$inp]! + veor @XMM[4], @XMM[4], @XMM[10] + veor @XMM[2], @XMM[2], @XMM[11] + vld1.8 {@XMM[15]}, [$inp]! + veor @XMM[7], @XMM[7], @XMM[12] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + vst1.8 {@XMM[6]}, [$out]! + vst1.8 {@XMM[4]}, [$out]! + vst1.8 {@XMM[2]}, [$out]! + vst1.8 {@XMM[7]}, [$out]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_five: + sub $inp, $inp, #0x50 + bl _bsaes_decrypt8 + vldmia $fp, {@XMM[14]} @ reload IV + vld1.8 {@XMM[8]-@XMM[9]}, [$inp]! @ reload input + veor @XMM[0], @XMM[0], @XMM[14] @ ^= IV + vld1.8 {@XMM[10]-@XMM[11]}, [$inp]! + veor @XMM[1], @XMM[1], @XMM[8] + veor @XMM[6], @XMM[6], @XMM[9] + vld1.8 {@XMM[15]}, [$inp]! + veor @XMM[4], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + veor @XMM[2], @XMM[2], @XMM[11] + vst1.8 {@XMM[6]}, [$out]! + vst1.8 {@XMM[4]}, [$out]! + vst1.8 {@XMM[2]}, [$out]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_four: + sub $inp, $inp, #0x40 + bl _bsaes_decrypt8 + vldmia $fp, {@XMM[14]} @ reload IV + vld1.8 {@XMM[8]-@XMM[9]}, [$inp]! @ reload input + veor @XMM[0], @XMM[0], @XMM[14] @ ^= IV + vld1.8 {@XMM[10]}, [$inp]! + veor @XMM[1], @XMM[1], @XMM[8] + veor @XMM[6], @XMM[6], @XMM[9] + vld1.8 {@XMM[15]}, [$inp]! + veor @XMM[4], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + vst1.8 {@XMM[6]}, [$out]! + vst1.8 {@XMM[4]}, [$out]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_three: + sub $inp, $inp, #0x30 + bl _bsaes_decrypt8 + vldmia $fp, {@XMM[14]} @ reload IV + vld1.8 {@XMM[8]-@XMM[9]}, [$inp]! @ reload input + veor @XMM[0], @XMM[0], @XMM[14] @ ^= IV + vld1.8 {@XMM[15]}, [$inp]! + veor @XMM[1], @XMM[1], @XMM[8] + veor @XMM[6], @XMM[6], @XMM[9] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + vst1.8 {@XMM[6]}, [$out]! + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_two: + sub $inp, $inp, #0x20 + bl _bsaes_decrypt8 + vldmia $fp, {@XMM[14]} @ reload IV + vld1.8 {@XMM[8]}, [$inp]! @ reload input + veor @XMM[0], @XMM[0], @XMM[14] @ ^= IV + vld1.8 {@XMM[15]}, [$inp]! @ reload input + veor @XMM[1], @XMM[1], @XMM[8] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + b .Lcbc_dec_done +.align 4 +.Lcbc_dec_one: + sub $inp, $inp, #0x10 + mov $rounds, $out @ save original out pointer + mov $out, $fp @ use the iv scratch space as out buffer + mov r2, $key + vmov @XMM[4],@XMM[15] @ just in case ensure that IV + vmov @XMM[5],@XMM[0] @ and input are preserved + bl AES_decrypt + vld1.8 {@XMM[0]}, [$fp,:64] @ load result + veor @XMM[0], @XMM[0], @XMM[4] @ ^= IV + vmov @XMM[15], @XMM[5] @ @XMM[5] holds input + vst1.8 {@XMM[0]}, [$rounds] @ write output + +.Lcbc_dec_done: +#ifndef BSAES_ASM_EXTENDED_KEY + vmov.i32 q0, #0 + vmov.i32 q1, #0 +.Lcbc_dec_bzero: @ wipe key schedule [if any] + vstmia $keysched!, {q0-q1} + cmp $keysched, $fp + bne .Lcbc_dec_bzero +#endif + + mov sp, $fp + add sp, #0x10 @ add sp,$fp,#0x10 is no good for thumb + vst1.8 {@XMM[15]}, [$ivp] @ return IV + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} +.size bsaes_cbc_encrypt,.-bsaes_cbc_encrypt +___ +} +{ +my ($inp,$out,$len,$key, $ctr,$fp,$rounds)=(map("r$_",(0..3,8..10))); +my $const = "r6"; # shared with _bsaes_encrypt8_alt +my $keysched = "sp"; + +$code.=<<___; +.extern AES_encrypt +.global bsaes_ctr32_encrypt_blocks +.type bsaes_ctr32_encrypt_blocks,%function +.align 5 +bsaes_ctr32_encrypt_blocks: + cmp $len, #8 @ use plain AES for + blo .Lctr_enc_short @ small sizes + + mov ip, sp + stmdb sp!, {r4-r10, lr} + VFP_ABI_PUSH + ldr $ctr, [ip] @ ctr is 1st arg on the stack + sub sp, sp, #0x10 @ scratch space to carry over the ctr + mov $fp, sp @ save sp + + ldr $rounds, [$key, #240] @ get # of rounds +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, $rounds, lsl#7 @ 128 bytes per inner round key + add r12, #`128-32` @ size of bit-sliced key schedule + + @ populate the key schedule + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + mov sp, r12 @ sp is $keysched + bl _bsaes_key_convert + veor @XMM[7],@XMM[7],@XMM[15] @ fix up last round key + vstmia r12, {@XMM[7]} @ save last round key + + vld1.8 {@XMM[0]}, [$ctr] @ load counter + add $ctr, $const, #.LREVM0SR-.LM0 @ borrow $ctr + vldmia $keysched, {@XMM[4]} @ load round0 key +#else + ldr r12, [$key, #244] + eors r12, #1 + beq 0f + + @ populate the key schedule + str r12, [$key, #244] + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + add r12, $key, #248 @ pass key schedule + bl _bsaes_key_convert + veor @XMM[7],@XMM[7],@XMM[15] @ fix up last round key + vstmia r12, {@XMM[7]} @ save last round key + +.align 2 +0: add r12, $key, #248 + vld1.8 {@XMM[0]}, [$ctr] @ load counter + adrl $ctr, .LREVM0SR @ borrow $ctr + vldmia r12, {@XMM[4]} @ load round0 key + sub sp, #0x10 @ place for adjusted round0 key +#endif + + vmov.i32 @XMM[8],#1 @ compose 1<<96 + veor @XMM[9],@XMM[9],@XMM[9] + vrev32.8 @XMM[0],@XMM[0] + vext.8 @XMM[8],@XMM[9],@XMM[8],#4 + vrev32.8 @XMM[4],@XMM[4] + vadd.u32 @XMM[9],@XMM[8],@XMM[8] @ compose 2<<96 + vstmia $keysched, {@XMM[4]} @ save adjusted round0 key + b .Lctr_enc_loop + +.align 4 +.Lctr_enc_loop: + vadd.u32 @XMM[10], @XMM[8], @XMM[9] @ compose 3<<96 + vadd.u32 @XMM[1], @XMM[0], @XMM[8] @ +1 + vadd.u32 @XMM[2], @XMM[0], @XMM[9] @ +2 + vadd.u32 @XMM[3], @XMM[0], @XMM[10] @ +3 + vadd.u32 @XMM[4], @XMM[1], @XMM[10] + vadd.u32 @XMM[5], @XMM[2], @XMM[10] + vadd.u32 @XMM[6], @XMM[3], @XMM[10] + vadd.u32 @XMM[7], @XMM[4], @XMM[10] + vadd.u32 @XMM[10], @XMM[5], @XMM[10] @ next counter + + @ Borrow prologue from _bsaes_encrypt8 to use the opportunity + @ to flip byte order in 32-bit counter + + vldmia $keysched, {@XMM[9]} @ load round0 key +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, $keysched, #0x10 @ pass next round key +#else + add r4, $key, #`248+16` +#endif + vldmia $ctr, {@XMM[8]} @ .LREVM0SR + mov r5, $rounds @ pass rounds + vstmia $fp, {@XMM[10]} @ save next counter + sub $const, $ctr, #.LREVM0SR-.LSR @ pass constants + + bl _bsaes_encrypt8_alt + + subs $len, $len, #8 + blo .Lctr_enc_loop_done + + vld1.8 {@XMM[8]-@XMM[9]}, [$inp]! @ load input + vld1.8 {@XMM[10]-@XMM[11]}, [$inp]! + veor @XMM[0], @XMM[8] + veor @XMM[1], @XMM[9] + vld1.8 {@XMM[12]-@XMM[13]}, [$inp]! + veor @XMM[4], @XMM[10] + veor @XMM[6], @XMM[11] + vld1.8 {@XMM[14]-@XMM[15]}, [$inp]! + veor @XMM[3], @XMM[12] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! @ write output + veor @XMM[7], @XMM[13] + veor @XMM[2], @XMM[14] + vst1.8 {@XMM[4]}, [$out]! + veor @XMM[5], @XMM[15] + vst1.8 {@XMM[6]}, [$out]! + vmov.i32 @XMM[8], #1 @ compose 1<<96 + vst1.8 {@XMM[3]}, [$out]! + veor @XMM[9], @XMM[9], @XMM[9] + vst1.8 {@XMM[7]}, [$out]! + vext.8 @XMM[8], @XMM[9], @XMM[8], #4 + vst1.8 {@XMM[2]}, [$out]! + vadd.u32 @XMM[9],@XMM[8],@XMM[8] @ compose 2<<96 + vst1.8 {@XMM[5]}, [$out]! + vldmia $fp, {@XMM[0]} @ load counter + + bne .Lctr_enc_loop + b .Lctr_enc_done + +.align 4 +.Lctr_enc_loop_done: + add $len, $len, #8 + vld1.8 {@XMM[8]}, [$inp]! @ load input + veor @XMM[0], @XMM[8] + vst1.8 {@XMM[0]}, [$out]! @ write output + cmp $len, #2 + blo .Lctr_enc_done + vld1.8 {@XMM[9]}, [$inp]! + veor @XMM[1], @XMM[9] + vst1.8 {@XMM[1]}, [$out]! + beq .Lctr_enc_done + vld1.8 {@XMM[10]}, [$inp]! + veor @XMM[4], @XMM[10] + vst1.8 {@XMM[4]}, [$out]! + cmp $len, #4 + blo .Lctr_enc_done + vld1.8 {@XMM[11]}, [$inp]! + veor @XMM[6], @XMM[11] + vst1.8 {@XMM[6]}, [$out]! + beq .Lctr_enc_done + vld1.8 {@XMM[12]}, [$inp]! + veor @XMM[3], @XMM[12] + vst1.8 {@XMM[3]}, [$out]! + cmp $len, #6 + blo .Lctr_enc_done + vld1.8 {@XMM[13]}, [$inp]! + veor @XMM[7], @XMM[13] + vst1.8 {@XMM[7]}, [$out]! + beq .Lctr_enc_done + vld1.8 {@XMM[14]}, [$inp] + veor @XMM[2], @XMM[14] + vst1.8 {@XMM[2]}, [$out]! + +.Lctr_enc_done: + vmov.i32 q0, #0 + vmov.i32 q1, #0 +#ifndef BSAES_ASM_EXTENDED_KEY +.Lctr_enc_bzero: @ wipe key schedule [if any] + vstmia $keysched!, {q0-q1} + cmp $keysched, $fp + bne .Lctr_enc_bzero +#else + vstmia $keysched, {q0-q1} +#endif + + mov sp, $fp + add sp, #0x10 @ add sp,$fp,#0x10 is no good for thumb + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} @ return + +.align 4 +.Lctr_enc_short: + ldr ip, [sp] @ ctr pointer is passed on stack + stmdb sp!, {r4-r8, lr} + + mov r4, $inp @ copy arguments + mov r5, $out + mov r6, $len + mov r7, $key + ldr r8, [ip, #12] @ load counter LSW + vld1.8 {@XMM[1]}, [ip] @ load whole counter value +#ifdef __ARMEL__ + rev r8, r8 +#endif + sub sp, sp, #0x10 + vst1.8 {@XMM[1]}, [sp,:64] @ copy counter value + sub sp, sp, #0x10 + +.Lctr_enc_short_loop: + add r0, sp, #0x10 @ input counter value + mov r1, sp @ output on the stack + mov r2, r7 @ key + + bl AES_encrypt + + vld1.8 {@XMM[0]}, [r4]! @ load input + vld1.8 {@XMM[1]}, [sp,:64] @ load encrypted counter + add r8, r8, #1 +#ifdef __ARMEL__ + rev r0, r8 + str r0, [sp, #0x1c] @ next counter value +#else + str r8, [sp, #0x1c] @ next counter value +#endif + veor @XMM[0],@XMM[0],@XMM[1] + vst1.8 {@XMM[0]}, [r5]! @ store output + subs r6, r6, #1 + bne .Lctr_enc_short_loop + + vmov.i32 q0, #0 + vmov.i32 q1, #0 + vstmia sp!, {q0-q1} + + ldmia sp!, {r4-r8, pc} +.size bsaes_ctr32_encrypt_blocks,.-bsaes_ctr32_encrypt_blocks +___ +} +{ +###################################################################### +# void bsaes_xts_[en|de]crypt(const char *inp,char *out,size_t len, +# const AES_KEY *key1, const AES_KEY *key2, +# const unsigned char iv[16]); +# +my ($inp,$out,$len,$key,$rounds,$magic,$fp)=(map("r$_",(7..10,1..3))); +my $const="r6"; # returned by _bsaes_key_convert +my $twmask=@XMM[5]; +my @T=@XMM[6..7]; + +$code.=<<___; +.globl bsaes_xts_encrypt +.type bsaes_xts_encrypt,%function +.align 4 +bsaes_xts_encrypt: + mov ip, sp + stmdb sp!, {r4-r10, lr} @ 0x20 + VFP_ABI_PUSH + mov r6, sp @ future $fp + + mov $inp, r0 + mov $out, r1 + mov $len, r2 + mov $key, r3 + + sub r0, sp, #0x10 @ 0x10 + bic r0, #0xf @ align at 16 bytes + mov sp, r0 + +#ifdef XTS_CHAIN_TWEAK + ldr r0, [ip] @ pointer to input tweak +#else + @ generate initial tweak + ldr r0, [ip, #4] @ iv[] + mov r1, sp + ldr r2, [ip, #0] @ key2 + bl AES_encrypt + mov r0,sp @ pointer to initial tweak +#endif + + ldr $rounds, [$key, #240] @ get # of rounds + mov $fp, r6 +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, $rounds, lsl#7 @ 128 bytes per inner round key + @ add r12, #`128-32` @ size of bit-sliced key schedule + sub r12, #`32+16` @ place for tweak[9] + + @ populate the key schedule + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + mov sp, r12 + add r12, #0x90 @ pass key schedule + bl _bsaes_key_convert + veor @XMM[7], @XMM[7], @XMM[15] @ fix up last round key + vstmia r12, {@XMM[7]} @ save last round key +#else + ldr r12, [$key, #244] + eors r12, #1 + beq 0f + + str r12, [$key, #244] + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + add r12, $key, #248 @ pass key schedule + bl _bsaes_key_convert + veor @XMM[7], @XMM[7], @XMM[15] @ fix up last round key + vstmia r12, {@XMM[7]} + +.align 2 +0: sub sp, #0x90 @ place for tweak[9] +#endif + + vld1.8 {@XMM[8]}, [r0] @ initial tweak + adr $magic, .Lxts_magic + + subs $len, #0x80 + blo .Lxts_enc_short + b .Lxts_enc_loop + +.align 4 +.Lxts_enc_loop: + vldmia $magic, {$twmask} @ load XTS magic + vshr.s64 @T[0], @XMM[8], #63 + mov r0, sp + vand @T[0], @T[0], $twmask +___ +for($i=9;$i<16;$i++) { +$code.=<<___; + vadd.u64 @XMM[$i], @XMM[$i-1], @XMM[$i-1] + vst1.64 {@XMM[$i-1]}, [r0,:128]! + vswp `&Dhi("@T[0]")`,`&Dlo("@T[0]")` + vshr.s64 @T[1], @XMM[$i], #63 + veor @XMM[$i], @XMM[$i], @T[0] + vand @T[1], @T[1], $twmask +___ + @T=reverse(@T); + +$code.=<<___ if ($i>=10); + vld1.8 {@XMM[$i-10]}, [$inp]! +___ +$code.=<<___ if ($i>=11); + veor @XMM[$i-11], @XMM[$i-11], @XMM[$i-3] +___ +} +$code.=<<___; + vadd.u64 @XMM[8], @XMM[15], @XMM[15] + vst1.64 {@XMM[15]}, [r0,:128]! + vswp `&Dhi("@T[0]")`,`&Dlo("@T[0]")` + veor @XMM[8], @XMM[8], @T[0] + vst1.64 {@XMM[8]}, [r0,:128] @ next round tweak + + vld1.8 {@XMM[6]-@XMM[7]}, [$inp]! + veor @XMM[5], @XMM[5], @XMM[13] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[6], @XMM[6], @XMM[14] + mov r5, $rounds @ pass rounds + veor @XMM[7], @XMM[7], @XMM[15] + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]-@XMM[13]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[6], @XMM[11] + vld1.64 {@XMM[14]-@XMM[15]}, [r0,:128]! + veor @XMM[10], @XMM[3], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + veor @XMM[11], @XMM[7], @XMM[13] + veor @XMM[12], @XMM[2], @XMM[14] + vst1.8 {@XMM[10]-@XMM[11]}, [$out]! + veor @XMM[13], @XMM[5], @XMM[15] + vst1.8 {@XMM[12]-@XMM[13]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + + subs $len, #0x80 + bpl .Lxts_enc_loop + +.Lxts_enc_short: + adds $len, #0x70 + bmi .Lxts_enc_done + + vldmia $magic, {$twmask} @ load XTS magic + vshr.s64 @T[0], @XMM[8], #63 + mov r0, sp + vand @T[0], @T[0], $twmask +___ +for($i=9;$i<16;$i++) { +$code.=<<___; + vadd.u64 @XMM[$i], @XMM[$i-1], @XMM[$i-1] + vst1.64 {@XMM[$i-1]}, [r0,:128]! + vswp `&Dhi("@T[0]")`,`&Dlo("@T[0]")` + vshr.s64 @T[1], @XMM[$i], #63 + veor @XMM[$i], @XMM[$i], @T[0] + vand @T[1], @T[1], $twmask +___ + @T=reverse(@T); + +$code.=<<___ if ($i>=10); + vld1.8 {@XMM[$i-10]}, [$inp]! + subs $len, #0x10 + bmi .Lxts_enc_`$i-9` +___ +$code.=<<___ if ($i>=11); + veor @XMM[$i-11], @XMM[$i-11], @XMM[$i-3] +___ +} +$code.=<<___; + sub $len, #0x10 + vst1.64 {@XMM[15]}, [r0,:128] @ next round tweak + + vld1.8 {@XMM[6]}, [$inp]! + veor @XMM[5], @XMM[5], @XMM[13] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[6], @XMM[6], @XMM[14] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]-@XMM[13]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[6], @XMM[11] + vld1.64 {@XMM[14]}, [r0,:128]! + veor @XMM[10], @XMM[3], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + veor @XMM[11], @XMM[7], @XMM[13] + veor @XMM[12], @XMM[2], @XMM[14] + vst1.8 {@XMM[10]-@XMM[11]}, [$out]! + vst1.8 {@XMM[12]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_6: + vst1.64 {@XMM[14]}, [r0,:128] @ next round tweak + + veor @XMM[4], @XMM[4], @XMM[12] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[5], @XMM[5], @XMM[13] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]-@XMM[13]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[6], @XMM[11] + veor @XMM[10], @XMM[3], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + veor @XMM[11], @XMM[7], @XMM[13] + vst1.8 {@XMM[10]-@XMM[11]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_enc_done + +@ put this in range for both ARM and Thumb mode adr instructions +.align 5 +.Lxts_magic: + .quad 1, 0x87 + +.align 5 +.Lxts_enc_5: + vst1.64 {@XMM[13]}, [r0,:128] @ next round tweak + + veor @XMM[3], @XMM[3], @XMM[11] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[4], @XMM[4], @XMM[12] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[6], @XMM[11] + veor @XMM[10], @XMM[3], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + vst1.8 {@XMM[10]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_4: + vst1.64 {@XMM[12]}, [r0,:128] @ next round tweak + + veor @XMM[2], @XMM[2], @XMM[10] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[3], @XMM[3], @XMM[11] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[6], @XMM[11] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_3: + vst1.64 {@XMM[11]}, [r0,:128] @ next round tweak + + veor @XMM[1], @XMM[1], @XMM[9] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[2], @XMM[2], @XMM[10] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {@XMM[8]-@XMM[9]}, [r0,:128]! + vld1.64 {@XMM[10]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[4], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + vst1.8 {@XMM[8]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_2: + vst1.64 {@XMM[10]}, [r0,:128] @ next round tweak + + veor @XMM[0], @XMM[0], @XMM[8] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[1], @XMM[1], @XMM[9] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_encrypt8 + + vld1.64 {@XMM[8]-@XMM[9]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + veor @XMM[1], @XMM[1], @XMM[ 9] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_enc_done +.align 4 +.Lxts_enc_1: + mov r0, sp + veor @XMM[0], @XMM[8] + mov r1, sp + vst1.8 {@XMM[0]}, [sp,:128] + mov r2, $key + mov r4, $fp @ preserve fp + + bl AES_encrypt + + vld1.8 {@XMM[0]}, [sp,:128] + veor @XMM[0], @XMM[0], @XMM[8] + vst1.8 {@XMM[0]}, [$out]! + mov $fp, r4 + + vmov @XMM[8], @XMM[9] @ next round tweak + +.Lxts_enc_done: +#ifndef XTS_CHAIN_TWEAK + adds $len, #0x10 + beq .Lxts_enc_ret + sub r6, $out, #0x10 + +.Lxts_enc_steal: + ldrb r0, [$inp], #1 + ldrb r1, [$out, #-0x10] + strb r0, [$out, #-0x10] + strb r1, [$out], #1 + + subs $len, #1 + bhi .Lxts_enc_steal + + vld1.8 {@XMM[0]}, [r6] + mov r0, sp + veor @XMM[0], @XMM[0], @XMM[8] + mov r1, sp + vst1.8 {@XMM[0]}, [sp,:128] + mov r2, $key + mov r4, $fp @ preserve fp + + bl AES_encrypt + + vld1.8 {@XMM[0]}, [sp,:128] + veor @XMM[0], @XMM[0], @XMM[8] + vst1.8 {@XMM[0]}, [r6] + mov $fp, r4 +#endif + +.Lxts_enc_ret: + bic r0, $fp, #0xf + vmov.i32 q0, #0 + vmov.i32 q1, #0 +#ifdef XTS_CHAIN_TWEAK + ldr r1, [$fp, #0x20+VFP_ABI_FRAME] @ chain tweak +#endif +.Lxts_enc_bzero: @ wipe key schedule [if any] + vstmia sp!, {q0-q1} + cmp sp, r0 + bne .Lxts_enc_bzero + + mov sp, $fp +#ifdef XTS_CHAIN_TWEAK + vst1.8 {@XMM[8]}, [r1] +#endif + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} @ return + +.size bsaes_xts_encrypt,.-bsaes_xts_encrypt + +.globl bsaes_xts_decrypt +.type bsaes_xts_decrypt,%function +.align 4 +bsaes_xts_decrypt: + mov ip, sp + stmdb sp!, {r4-r10, lr} @ 0x20 + VFP_ABI_PUSH + mov r6, sp @ future $fp + + mov $inp, r0 + mov $out, r1 + mov $len, r2 + mov $key, r3 + + sub r0, sp, #0x10 @ 0x10 + bic r0, #0xf @ align at 16 bytes + mov sp, r0 + +#ifdef XTS_CHAIN_TWEAK + ldr r0, [ip] @ pointer to input tweak +#else + @ generate initial tweak + ldr r0, [ip, #4] @ iv[] + mov r1, sp + ldr r2, [ip, #0] @ key2 + bl AES_encrypt + mov r0, sp @ pointer to initial tweak +#endif + + ldr $rounds, [$key, #240] @ get # of rounds + mov $fp, r6 +#ifndef BSAES_ASM_EXTENDED_KEY + @ allocate the key schedule on the stack + sub r12, sp, $rounds, lsl#7 @ 128 bytes per inner round key + @ add r12, #`128-32` @ size of bit-sliced key schedule + sub r12, #`32+16` @ place for tweak[9] + + @ populate the key schedule + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + mov sp, r12 + add r12, #0x90 @ pass key schedule + bl _bsaes_key_convert + add r4, sp, #0x90 + vldmia r4, {@XMM[6]} + vstmia r12, {@XMM[15]} @ save last round key + veor @XMM[7], @XMM[7], @XMM[6] @ fix up round 0 key + vstmia r4, {@XMM[7]} +#else + ldr r12, [$key, #244] + eors r12, #1 + beq 0f + + str r12, [$key, #244] + mov r4, $key @ pass key + mov r5, $rounds @ pass # of rounds + add r12, $key, #248 @ pass key schedule + bl _bsaes_key_convert + add r4, $key, #248 + vldmia r4, {@XMM[6]} + vstmia r12, {@XMM[15]} @ save last round key + veor @XMM[7], @XMM[7], @XMM[6] @ fix up round 0 key + vstmia r4, {@XMM[7]} + +.align 2 +0: sub sp, #0x90 @ place for tweak[9] +#endif + vld1.8 {@XMM[8]}, [r0] @ initial tweak + adr $magic, .Lxts_magic + + tst $len, #0xf @ if not multiple of 16 + it ne @ Thumb2 thing, sanity check in ARM + subne $len, #0x10 @ subtract another 16 bytes + subs $len, #0x80 + + blo .Lxts_dec_short + b .Lxts_dec_loop + +.align 4 +.Lxts_dec_loop: + vldmia $magic, {$twmask} @ load XTS magic + vshr.s64 @T[0], @XMM[8], #63 + mov r0, sp + vand @T[0], @T[0], $twmask +___ +for($i=9;$i<16;$i++) { +$code.=<<___; + vadd.u64 @XMM[$i], @XMM[$i-1], @XMM[$i-1] + vst1.64 {@XMM[$i-1]}, [r0,:128]! + vswp `&Dhi("@T[0]")`,`&Dlo("@T[0]")` + vshr.s64 @T[1], @XMM[$i], #63 + veor @XMM[$i], @XMM[$i], @T[0] + vand @T[1], @T[1], $twmask +___ + @T=reverse(@T); + +$code.=<<___ if ($i>=10); + vld1.8 {@XMM[$i-10]}, [$inp]! +___ +$code.=<<___ if ($i>=11); + veor @XMM[$i-11], @XMM[$i-11], @XMM[$i-3] +___ +} +$code.=<<___; + vadd.u64 @XMM[8], @XMM[15], @XMM[15] + vst1.64 {@XMM[15]}, [r0,:128]! + vswp `&Dhi("@T[0]")`,`&Dlo("@T[0]")` + veor @XMM[8], @XMM[8], @T[0] + vst1.64 {@XMM[8]}, [r0,:128] @ next round tweak + + vld1.8 {@XMM[6]-@XMM[7]}, [$inp]! + veor @XMM[5], @XMM[5], @XMM[13] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[6], @XMM[6], @XMM[14] + mov r5, $rounds @ pass rounds + veor @XMM[7], @XMM[7], @XMM[15] + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]-@XMM[13]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[6], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[4], @XMM[11] + vld1.64 {@XMM[14]-@XMM[15]}, [r0,:128]! + veor @XMM[10], @XMM[2], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + veor @XMM[11], @XMM[7], @XMM[13] + veor @XMM[12], @XMM[3], @XMM[14] + vst1.8 {@XMM[10]-@XMM[11]}, [$out]! + veor @XMM[13], @XMM[5], @XMM[15] + vst1.8 {@XMM[12]-@XMM[13]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + + subs $len, #0x80 + bpl .Lxts_dec_loop + +.Lxts_dec_short: + adds $len, #0x70 + bmi .Lxts_dec_done + + vldmia $magic, {$twmask} @ load XTS magic + vshr.s64 @T[0], @XMM[8], #63 + mov r0, sp + vand @T[0], @T[0], $twmask +___ +for($i=9;$i<16;$i++) { +$code.=<<___; + vadd.u64 @XMM[$i], @XMM[$i-1], @XMM[$i-1] + vst1.64 {@XMM[$i-1]}, [r0,:128]! + vswp `&Dhi("@T[0]")`,`&Dlo("@T[0]")` + vshr.s64 @T[1], @XMM[$i], #63 + veor @XMM[$i], @XMM[$i], @T[0] + vand @T[1], @T[1], $twmask +___ + @T=reverse(@T); + +$code.=<<___ if ($i>=10); + vld1.8 {@XMM[$i-10]}, [$inp]! + subs $len, #0x10 + bmi .Lxts_dec_`$i-9` +___ +$code.=<<___ if ($i>=11); + veor @XMM[$i-11], @XMM[$i-11], @XMM[$i-3] +___ +} +$code.=<<___; + sub $len, #0x10 + vst1.64 {@XMM[15]}, [r0,:128] @ next round tweak + + vld1.8 {@XMM[6]}, [$inp]! + veor @XMM[5], @XMM[5], @XMM[13] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[6], @XMM[6], @XMM[14] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]-@XMM[13]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[6], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[4], @XMM[11] + vld1.64 {@XMM[14]}, [r0,:128]! + veor @XMM[10], @XMM[2], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + veor @XMM[11], @XMM[7], @XMM[13] + veor @XMM[12], @XMM[3], @XMM[14] + vst1.8 {@XMM[10]-@XMM[11]}, [$out]! + vst1.8 {@XMM[12]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_6: + vst1.64 {@XMM[14]}, [r0,:128] @ next round tweak + + veor @XMM[4], @XMM[4], @XMM[12] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[5], @XMM[5], @XMM[13] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]-@XMM[13]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[6], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[4], @XMM[11] + veor @XMM[10], @XMM[2], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + veor @XMM[11], @XMM[7], @XMM[13] + vst1.8 {@XMM[10]-@XMM[11]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_5: + vst1.64 {@XMM[13]}, [r0,:128] @ next round tweak + + veor @XMM[3], @XMM[3], @XMM[11] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[4], @XMM[4], @XMM[12] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + vld1.64 {@XMM[12]}, [r0,:128]! + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[6], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[4], @XMM[11] + veor @XMM[10], @XMM[2], @XMM[12] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + vst1.8 {@XMM[10]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_4: + vst1.64 {@XMM[12]}, [r0,:128] @ next round tweak + + veor @XMM[2], @XMM[2], @XMM[10] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[3], @XMM[3], @XMM[11] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {@XMM[ 8]-@XMM[ 9]}, [r0,:128]! + vld1.64 {@XMM[10]-@XMM[11]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[6], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + veor @XMM[9], @XMM[4], @XMM[11] + vst1.8 {@XMM[8]-@XMM[9]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_3: + vst1.64 {@XMM[11]}, [r0,:128] @ next round tweak + + veor @XMM[1], @XMM[1], @XMM[9] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[2], @XMM[2], @XMM[10] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {@XMM[8]-@XMM[9]}, [r0,:128]! + vld1.64 {@XMM[10]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + veor @XMM[1], @XMM[1], @XMM[ 9] + veor @XMM[8], @XMM[6], @XMM[10] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + vst1.8 {@XMM[8]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_2: + vst1.64 {@XMM[10]}, [r0,:128] @ next round tweak + + veor @XMM[0], @XMM[0], @XMM[8] +#ifndef BSAES_ASM_EXTENDED_KEY + add r4, sp, #0x90 @ pass key schedule +#else + add r4, $key, #248 @ pass key schedule +#endif + veor @XMM[1], @XMM[1], @XMM[9] + mov r5, $rounds @ pass rounds + mov r0, sp + + bl _bsaes_decrypt8 + + vld1.64 {@XMM[8]-@XMM[9]}, [r0,:128]! + veor @XMM[0], @XMM[0], @XMM[ 8] + veor @XMM[1], @XMM[1], @XMM[ 9] + vst1.8 {@XMM[0]-@XMM[1]}, [$out]! + + vld1.64 {@XMM[8]}, [r0,:128] @ next round tweak + b .Lxts_dec_done +.align 4 +.Lxts_dec_1: + mov r0, sp + veor @XMM[0], @XMM[8] + mov r1, sp + vst1.8 {@XMM[0]}, [sp,:128] + mov r2, $key + mov r4, $fp @ preserve fp + mov r5, $magic @ preserve magic + + bl AES_decrypt + + vld1.8 {@XMM[0]}, [sp,:128] + veor @XMM[0], @XMM[0], @XMM[8] + vst1.8 {@XMM[0]}, [$out]! + mov $fp, r4 + mov $magic, r5 + + vmov @XMM[8], @XMM[9] @ next round tweak + +.Lxts_dec_done: +#ifndef XTS_CHAIN_TWEAK + adds $len, #0x10 + beq .Lxts_dec_ret + + @ calculate one round of extra tweak for the stolen ciphertext + vldmia $magic, {$twmask} + vshr.s64 @XMM[6], @XMM[8], #63 + vand @XMM[6], @XMM[6], $twmask + vadd.u64 @XMM[9], @XMM[8], @XMM[8] + vswp `&Dhi("@XMM[6]")`,`&Dlo("@XMM[6]")` + veor @XMM[9], @XMM[9], @XMM[6] + + @ perform the final decryption with the last tweak value + vld1.8 {@XMM[0]}, [$inp]! + mov r0, sp + veor @XMM[0], @XMM[0], @XMM[9] + mov r1, sp + vst1.8 {@XMM[0]}, [sp,:128] + mov r2, $key + mov r4, $fp @ preserve fp + + bl AES_decrypt + + vld1.8 {@XMM[0]}, [sp,:128] + veor @XMM[0], @XMM[0], @XMM[9] + vst1.8 {@XMM[0]}, [$out] + + mov r6, $out +.Lxts_dec_steal: + ldrb r1, [$out] + ldrb r0, [$inp], #1 + strb r1, [$out, #0x10] + strb r0, [$out], #1 + + subs $len, #1 + bhi .Lxts_dec_steal + + vld1.8 {@XMM[0]}, [r6] + mov r0, sp + veor @XMM[0], @XMM[8] + mov r1, sp + vst1.8 {@XMM[0]}, [sp,:128] + mov r2, $key + + bl AES_decrypt + + vld1.8 {@XMM[0]}, [sp,:128] + veor @XMM[0], @XMM[0], @XMM[8] + vst1.8 {@XMM[0]}, [r6] + mov $fp, r4 +#endif + +.Lxts_dec_ret: + bic r0, $fp, #0xf + vmov.i32 q0, #0 + vmov.i32 q1, #0 +#ifdef XTS_CHAIN_TWEAK + ldr r1, [$fp, #0x20+VFP_ABI_FRAME] @ chain tweak +#endif +.Lxts_dec_bzero: @ wipe key schedule [if any] + vstmia sp!, {q0-q1} + cmp sp, r0 + bne .Lxts_dec_bzero + + mov sp, $fp +#ifdef XTS_CHAIN_TWEAK + vst1.8 {@XMM[8]}, [r1] +#endif + VFP_ABI_POP + ldmia sp!, {r4-r10, pc} @ return + +.size bsaes_xts_decrypt,.-bsaes_xts_decrypt +___ +} +$code.=<<___; +#endif +___ + +$code =~ s/\`([^\`]*)\`/eval($1)/gem; + +open SELF,$0; +while(<SELF>) { + next if (/^#!/); + last if (!s/^#/@/ and !/^$/); + print; +} +close SELF; + +print $code; + +close STDOUT; diff --git a/arch/arm/crypto/sha1-armv7-neon.S b/arch/arm/crypto/sha1-armv7-neon.S new file mode 100644 index 000000000000..50013c0e2864 --- /dev/null +++ b/arch/arm/crypto/sha1-armv7-neon.S @@ -0,0 +1,634 @@ +/* sha1-armv7-neon.S - ARM/NEON accelerated SHA-1 transform function + * + * Copyright © 2013-2014 Jussi Kivilinna <jussi.kivilinna@iki.fi> + * + * 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 <linux/linkage.h> + + +.syntax unified +.code 32 +.fpu neon + +.text + + +/* Context structure */ + +#define state_h0 0 +#define state_h1 4 +#define state_h2 8 +#define state_h3 12 +#define state_h4 16 + + +/* Constants */ + +#define K1 0x5A827999 +#define K2 0x6ED9EBA1 +#define K3 0x8F1BBCDC +#define K4 0xCA62C1D6 +.align 4 +.LK_VEC: +.LK1: .long K1, K1, K1, K1 +.LK2: .long K2, K2, K2, K2 +.LK3: .long K3, K3, K3, K3 +.LK4: .long K4, K4, K4, K4 + + +/* Register macros */ + +#define RSTATE r0 +#define RDATA r1 +#define RNBLKS r2 +#define ROLDSTACK r3 +#define RWK lr + +#define _a r4 +#define _b r5 +#define _c r6 +#define _d r7 +#define _e r8 + +#define RT0 r9 +#define RT1 r10 +#define RT2 r11 +#define RT3 r12 + +#define W0 q0 +#define W1 q1 +#define W2 q2 +#define W3 q3 +#define W4 q4 +#define W5 q5 +#define W6 q6 +#define W7 q7 + +#define tmp0 q8 +#define tmp1 q9 +#define tmp2 q10 +#define tmp3 q11 + +#define qK1 q12 +#define qK2 q13 +#define qK3 q14 +#define qK4 q15 + + +/* Round function macros. */ + +#define WK_offs(i) (((i) & 15) * 4) + +#define _R_F1(a,b,c,d,e,i,pre1,pre2,pre3,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + ldr RT3, [sp, WK_offs(i)]; \ + pre1(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + bic RT0, d, b; \ + add e, e, a, ror #(32 - 5); \ + and RT1, c, b; \ + pre2(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + add RT0, RT0, RT3; \ + add e, e, RT1; \ + ror b, #(32 - 30); \ + pre3(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + add e, e, RT0; + +#define _R_F2(a,b,c,d,e,i,pre1,pre2,pre3,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + ldr RT3, [sp, WK_offs(i)]; \ + pre1(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + eor RT0, d, b; \ + add e, e, a, ror #(32 - 5); \ + eor RT0, RT0, c; \ + pre2(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + add e, e, RT3; \ + ror b, #(32 - 30); \ + pre3(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + add e, e, RT0; \ + +#define _R_F3(a,b,c,d,e,i,pre1,pre2,pre3,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + ldr RT3, [sp, WK_offs(i)]; \ + pre1(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + eor RT0, b, c; \ + and RT1, b, c; \ + add e, e, a, ror #(32 - 5); \ + pre2(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + and RT0, RT0, d; \ + add RT1, RT1, RT3; \ + add e, e, RT0; \ + ror b, #(32 - 30); \ + pre3(i16,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28); \ + add e, e, RT1; + +#define _R_F4(a,b,c,d,e,i,pre1,pre2,pre3,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + _R_F2(a,b,c,d,e,i,pre1,pre2,pre3,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) + +#define _R(a,b,c,d,e,f,i,pre1,pre2,pre3,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + _R_##f(a,b,c,d,e,i,pre1,pre2,pre3,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) + +#define R(a,b,c,d,e,f,i) \ + _R_##f(a,b,c,d,e,i,dummy,dummy,dummy,i16,\ + W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) + +#define dummy(...) + + +/* Input expansion macros. */ + +/********* Precalc macros for rounds 0-15 *************************************/ + +#define W_PRECALC_00_15() \ + add RWK, sp, #(WK_offs(0)); \ + \ + vld1.32 {tmp0, tmp1}, [RDATA]!; \ + vrev32.8 W0, tmp0; /* big => little */ \ + vld1.32 {tmp2, tmp3}, [RDATA]!; \ + vadd.u32 tmp0, W0, curK; \ + vrev32.8 W7, tmp1; /* big => little */ \ + vrev32.8 W6, tmp2; /* big => little */ \ + vadd.u32 tmp1, W7, curK; \ + vrev32.8 W5, tmp3; /* big => little */ \ + vadd.u32 tmp2, W6, curK; \ + vst1.32 {tmp0, tmp1}, [RWK]!; \ + vadd.u32 tmp3, W5, curK; \ + vst1.32 {tmp2, tmp3}, [RWK]; \ + +#define WPRECALC_00_15_0(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vld1.32 {tmp0, tmp1}, [RDATA]!; \ + +#define WPRECALC_00_15_1(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + add RWK, sp, #(WK_offs(0)); \ + +#define WPRECALC_00_15_2(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vrev32.8 W0, tmp0; /* big => little */ \ + +#define WPRECALC_00_15_3(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vld1.32 {tmp2, tmp3}, [RDATA]!; \ + +#define WPRECALC_00_15_4(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vadd.u32 tmp0, W0, curK; \ + +#define WPRECALC_00_15_5(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vrev32.8 W7, tmp1; /* big => little */ \ + +#define WPRECALC_00_15_6(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vrev32.8 W6, tmp2; /* big => little */ \ + +#define WPRECALC_00_15_7(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vadd.u32 tmp1, W7, curK; \ + +#define WPRECALC_00_15_8(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vrev32.8 W5, tmp3; /* big => little */ \ + +#define WPRECALC_00_15_9(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vadd.u32 tmp2, W6, curK; \ + +#define WPRECALC_00_15_10(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vst1.32 {tmp0, tmp1}, [RWK]!; \ + +#define WPRECALC_00_15_11(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vadd.u32 tmp3, W5, curK; \ + +#define WPRECALC_00_15_12(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vst1.32 {tmp2, tmp3}, [RWK]; \ + + +/********* Precalc macros for rounds 16-31 ************************************/ + +#define WPRECALC_16_31_0(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor tmp0, tmp0; \ + vext.8 W, W_m16, W_m12, #8; \ + +#define WPRECALC_16_31_1(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + add RWK, sp, #(WK_offs(i)); \ + vext.8 tmp0, W_m04, tmp0, #4; \ + +#define WPRECALC_16_31_2(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor tmp0, tmp0, W_m16; \ + veor.32 W, W, W_m08; \ + +#define WPRECALC_16_31_3(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor tmp1, tmp1; \ + veor W, W, tmp0; \ + +#define WPRECALC_16_31_4(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vshl.u32 tmp0, W, #1; \ + +#define WPRECALC_16_31_5(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vext.8 tmp1, tmp1, W, #(16-12); \ + vshr.u32 W, W, #31; \ + +#define WPRECALC_16_31_6(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vorr tmp0, tmp0, W; \ + vshr.u32 W, tmp1, #30; \ + +#define WPRECALC_16_31_7(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vshl.u32 tmp1, tmp1, #2; \ + +#define WPRECALC_16_31_8(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor tmp0, tmp0, W; \ + +#define WPRECALC_16_31_9(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor W, tmp0, tmp1; \ + +#define WPRECALC_16_31_10(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vadd.u32 tmp0, W, curK; \ + +#define WPRECALC_16_31_11(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vst1.32 {tmp0}, [RWK]; + + +/********* Precalc macros for rounds 32-79 ************************************/ + +#define WPRECALC_32_79_0(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor W, W_m28; \ + +#define WPRECALC_32_79_1(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vext.8 tmp0, W_m08, W_m04, #8; \ + +#define WPRECALC_32_79_2(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor W, W_m16; \ + +#define WPRECALC_32_79_3(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + veor W, tmp0; \ + +#define WPRECALC_32_79_4(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + add RWK, sp, #(WK_offs(i&~3)); \ + +#define WPRECALC_32_79_5(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vshl.u32 tmp1, W, #2; \ + +#define WPRECALC_32_79_6(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vshr.u32 tmp0, W, #30; \ + +#define WPRECALC_32_79_7(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vorr W, tmp0, tmp1; \ + +#define WPRECALC_32_79_8(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vadd.u32 tmp0, W, curK; \ + +#define WPRECALC_32_79_9(i,W,W_m04,W_m08,W_m12,W_m16,W_m20,W_m24,W_m28) \ + vst1.32 {tmp0}, [RWK]; + + +/* + * Transform nblks*64 bytes (nblks*16 32-bit words) at DATA. + * + * unsigned int + * sha1_transform_neon (void *ctx, const unsigned char *data, + * unsigned int nblks) + */ +.align 3 +ENTRY(sha1_transform_neon) + /* input: + * r0: ctx, CTX + * r1: data (64*nblks bytes) + * r2: nblks + */ + + cmp RNBLKS, #0; + beq .Ldo_nothing; + + push {r4-r12, lr}; + /*vpush {q4-q7};*/ + + adr RT3, .LK_VEC; + + mov ROLDSTACK, sp; + + /* Align stack. */ + sub RT0, sp, #(16*4); + and RT0, #(~(16-1)); + mov sp, RT0; + + vld1.32 {qK1-qK2}, [RT3]!; /* Load K1,K2 */ + + /* Get the values of the chaining variables. */ + ldm RSTATE, {_a-_e}; + + vld1.32 {qK3-qK4}, [RT3]; /* Load K3,K4 */ + +#undef curK +#define curK qK1 + /* Precalc 0-15. */ + W_PRECALC_00_15(); + +.Loop: + /* Transform 0-15 + Precalc 16-31. */ + _R( _a, _b, _c, _d, _e, F1, 0, + WPRECALC_16_31_0, WPRECALC_16_31_1, WPRECALC_16_31_2, 16, + W4, W5, W6, W7, W0, _, _, _ ); + _R( _e, _a, _b, _c, _d, F1, 1, + WPRECALC_16_31_3, WPRECALC_16_31_4, WPRECALC_16_31_5, 16, + W4, W5, W6, W7, W0, _, _, _ ); + _R( _d, _e, _a, _b, _c, F1, 2, + WPRECALC_16_31_6, WPRECALC_16_31_7, WPRECALC_16_31_8, 16, + W4, W5, W6, W7, W0, _, _, _ ); + _R( _c, _d, _e, _a, _b, F1, 3, + WPRECALC_16_31_9, WPRECALC_16_31_10,WPRECALC_16_31_11,16, + W4, W5, W6, W7, W0, _, _, _ ); + +#undef curK +#define curK qK2 + _R( _b, _c, _d, _e, _a, F1, 4, + WPRECALC_16_31_0, WPRECALC_16_31_1, WPRECALC_16_31_2, 20, + W3, W4, W5, W6, W7, _, _, _ ); + _R( _a, _b, _c, _d, _e, F1, 5, + WPRECALC_16_31_3, WPRECALC_16_31_4, WPRECALC_16_31_5, 20, + W3, W4, W5, W6, W7, _, _, _ ); + _R( _e, _a, _b, _c, _d, F1, 6, + WPRECALC_16_31_6, WPRECALC_16_31_7, WPRECALC_16_31_8, 20, + W3, W4, W5, W6, W7, _, _, _ ); + _R( _d, _e, _a, _b, _c, F1, 7, + WPRECALC_16_31_9, WPRECALC_16_31_10,WPRECALC_16_31_11,20, + W3, W4, W5, W6, W7, _, _, _ ); + + _R( _c, _d, _e, _a, _b, F1, 8, + WPRECALC_16_31_0, WPRECALC_16_31_1, WPRECALC_16_31_2, 24, + W2, W3, W4, W5, W6, _, _, _ ); + _R( _b, _c, _d, _e, _a, F1, 9, + WPRECALC_16_31_3, WPRECALC_16_31_4, WPRECALC_16_31_5, 24, + W2, W3, W4, W5, W6, _, _, _ ); + _R( _a, _b, _c, _d, _e, F1, 10, + WPRECALC_16_31_6, WPRECALC_16_31_7, WPRECALC_16_31_8, 24, + W2, W3, W4, W5, W6, _, _, _ ); + _R( _e, _a, _b, _c, _d, F1, 11, + WPRECALC_16_31_9, WPRECALC_16_31_10,WPRECALC_16_31_11,24, + W2, W3, W4, W5, W6, _, _, _ ); + + _R( _d, _e, _a, _b, _c, F1, 12, + WPRECALC_16_31_0, WPRECALC_16_31_1, WPRECALC_16_31_2, 28, + W1, W2, W3, W4, W5, _, _, _ ); + _R( _c, _d, _e, _a, _b, F1, 13, + WPRECALC_16_31_3, WPRECALC_16_31_4, WPRECALC_16_31_5, 28, + W1, W2, W3, W4, W5, _, _, _ ); + _R( _b, _c, _d, _e, _a, F1, 14, + WPRECALC_16_31_6, WPRECALC_16_31_7, WPRECALC_16_31_8, 28, + W1, W2, W3, W4, W5, _, _, _ ); + _R( _a, _b, _c, _d, _e, F1, 15, + WPRECALC_16_31_9, WPRECALC_16_31_10,WPRECALC_16_31_11,28, + W1, W2, W3, W4, W5, _, _, _ ); + + /* Transform 16-63 + Precalc 32-79. */ + _R( _e, _a, _b, _c, _d, F1, 16, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 32, + W0, W1, W2, W3, W4, W5, W6, W7); + _R( _d, _e, _a, _b, _c, F1, 17, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 32, + W0, W1, W2, W3, W4, W5, W6, W7); + _R( _c, _d, _e, _a, _b, F1, 18, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 32, + W0, W1, W2, W3, W4, W5, W6, W7); + _R( _b, _c, _d, _e, _a, F1, 19, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 32, + W0, W1, W2, W3, W4, W5, W6, W7); + + _R( _a, _b, _c, _d, _e, F2, 20, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 36, + W7, W0, W1, W2, W3, W4, W5, W6); + _R( _e, _a, _b, _c, _d, F2, 21, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 36, + W7, W0, W1, W2, W3, W4, W5, W6); + _R( _d, _e, _a, _b, _c, F2, 22, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 36, + W7, W0, W1, W2, W3, W4, W5, W6); + _R( _c, _d, _e, _a, _b, F2, 23, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 36, + W7, W0, W1, W2, W3, W4, W5, W6); + +#undef curK +#define curK qK3 + _R( _b, _c, _d, _e, _a, F2, 24, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 40, + W6, W7, W0, W1, W2, W3, W4, W5); + _R( _a, _b, _c, _d, _e, F2, 25, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 40, + W6, W7, W0, W1, W2, W3, W4, W5); + _R( _e, _a, _b, _c, _d, F2, 26, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 40, + W6, W7, W0, W1, W2, W3, W4, W5); + _R( _d, _e, _a, _b, _c, F2, 27, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 40, + W6, W7, W0, W1, W2, W3, W4, W5); + + _R( _c, _d, _e, _a, _b, F2, 28, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 44, + W5, W6, W7, W0, W1, W2, W3, W4); + _R( _b, _c, _d, _e, _a, F2, 29, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 44, + W5, W6, W7, W0, W1, W2, W3, W4); + _R( _a, _b, _c, _d, _e, F2, 30, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 44, + W5, W6, W7, W0, W1, W2, W3, W4); + _R( _e, _a, _b, _c, _d, F2, 31, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 44, + W5, W6, W7, W0, W1, W2, W3, W4); + + _R( _d, _e, _a, _b, _c, F2, 32, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 48, + W4, W5, W6, W7, W0, W1, W2, W3); + _R( _c, _d, _e, _a, _b, F2, 33, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 48, + W4, W5, W6, W7, W0, W1, W2, W3); + _R( _b, _c, _d, _e, _a, F2, 34, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 48, + W4, W5, W6, W7, W0, W1, W2, W3); + _R( _a, _b, _c, _d, _e, F2, 35, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 48, + W4, W5, W6, W7, W0, W1, W2, W3); + + _R( _e, _a, _b, _c, _d, F2, 36, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 52, + W3, W4, W5, W6, W7, W0, W1, W2); + _R( _d, _e, _a, _b, _c, F2, 37, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 52, + W3, W4, W5, W6, W7, W0, W1, W2); + _R( _c, _d, _e, _a, _b, F2, 38, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 52, + W3, W4, W5, W6, W7, W0, W1, W2); + _R( _b, _c, _d, _e, _a, F2, 39, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 52, + W3, W4, W5, W6, W7, W0, W1, W2); + + _R( _a, _b, _c, _d, _e, F3, 40, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 56, + W2, W3, W4, W5, W6, W7, W0, W1); + _R( _e, _a, _b, _c, _d, F3, 41, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 56, + W2, W3, W4, W5, W6, W7, W0, W1); + _R( _d, _e, _a, _b, _c, F3, 42, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 56, + W2, W3, W4, W5, W6, W7, W0, W1); + _R( _c, _d, _e, _a, _b, F3, 43, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 56, + W2, W3, W4, W5, W6, W7, W0, W1); + +#undef curK +#define curK qK4 + _R( _b, _c, _d, _e, _a, F3, 44, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 60, + W1, W2, W3, W4, W5, W6, W7, W0); + _R( _a, _b, _c, _d, _e, F3, 45, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 60, + W1, W2, W3, W4, W5, W6, W7, W0); + _R( _e, _a, _b, _c, _d, F3, 46, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 60, + W1, W2, W3, W4, W5, W6, W7, W0); + _R( _d, _e, _a, _b, _c, F3, 47, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 60, + W1, W2, W3, W4, W5, W6, W7, W0); + + _R( _c, _d, _e, _a, _b, F3, 48, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 64, + W0, W1, W2, W3, W4, W5, W6, W7); + _R( _b, _c, _d, _e, _a, F3, 49, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 64, + W0, W1, W2, W3, W4, W5, W6, W7); + _R( _a, _b, _c, _d, _e, F3, 50, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 64, + W0, W1, W2, W3, W4, W5, W6, W7); + _R( _e, _a, _b, _c, _d, F3, 51, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 64, + W0, W1, W2, W3, W4, W5, W6, W7); + + _R( _d, _e, _a, _b, _c, F3, 52, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 68, + W7, W0, W1, W2, W3, W4, W5, W6); + _R( _c, _d, _e, _a, _b, F3, 53, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 68, + W7, W0, W1, W2, W3, W4, W5, W6); + _R( _b, _c, _d, _e, _a, F3, 54, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 68, + W7, W0, W1, W2, W3, W4, W5, W6); + _R( _a, _b, _c, _d, _e, F3, 55, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 68, + W7, W0, W1, W2, W3, W4, W5, W6); + + _R( _e, _a, _b, _c, _d, F3, 56, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 72, + W6, W7, W0, W1, W2, W3, W4, W5); + _R( _d, _e, _a, _b, _c, F3, 57, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 72, + W6, W7, W0, W1, W2, W3, W4, W5); + _R( _c, _d, _e, _a, _b, F3, 58, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 72, + W6, W7, W0, W1, W2, W3, W4, W5); + _R( _b, _c, _d, _e, _a, F3, 59, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 72, + W6, W7, W0, W1, W2, W3, W4, W5); + + subs RNBLKS, #1; + + _R( _a, _b, _c, _d, _e, F4, 60, + WPRECALC_32_79_0, WPRECALC_32_79_1, WPRECALC_32_79_2, 76, + W5, W6, W7, W0, W1, W2, W3, W4); + _R( _e, _a, _b, _c, _d, F4, 61, + WPRECALC_32_79_3, WPRECALC_32_79_4, WPRECALC_32_79_5, 76, + W5, W6, W7, W0, W1, W2, W3, W4); + _R( _d, _e, _a, _b, _c, F4, 62, + WPRECALC_32_79_6, dummy, WPRECALC_32_79_7, 76, + W5, W6, W7, W0, W1, W2, W3, W4); + _R( _c, _d, _e, _a, _b, F4, 63, + WPRECALC_32_79_8, dummy, WPRECALC_32_79_9, 76, + W5, W6, W7, W0, W1, W2, W3, W4); + + beq .Lend; + + /* Transform 64-79 + Precalc 0-15 of next block. */ +#undef curK +#define curK qK1 + _R( _b, _c, _d, _e, _a, F4, 64, + WPRECALC_00_15_0, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _a, _b, _c, _d, _e, F4, 65, + WPRECALC_00_15_1, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _e, _a, _b, _c, _d, F4, 66, + WPRECALC_00_15_2, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _d, _e, _a, _b, _c, F4, 67, + WPRECALC_00_15_3, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + + _R( _c, _d, _e, _a, _b, F4, 68, + dummy, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _b, _c, _d, _e, _a, F4, 69, + dummy, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _a, _b, _c, _d, _e, F4, 70, + WPRECALC_00_15_4, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _e, _a, _b, _c, _d, F4, 71, + WPRECALC_00_15_5, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + + _R( _d, _e, _a, _b, _c, F4, 72, + dummy, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _c, _d, _e, _a, _b, F4, 73, + dummy, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _b, _c, _d, _e, _a, F4, 74, + WPRECALC_00_15_6, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _a, _b, _c, _d, _e, F4, 75, + WPRECALC_00_15_7, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + + _R( _e, _a, _b, _c, _d, F4, 76, + WPRECALC_00_15_8, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _d, _e, _a, _b, _c, F4, 77, + WPRECALC_00_15_9, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _c, _d, _e, _a, _b, F4, 78, + WPRECALC_00_15_10, dummy, dummy, _, _, _, _, _, _, _, _, _ ); + _R( _b, _c, _d, _e, _a, F4, 79, + WPRECALC_00_15_11, dummy, WPRECALC_00_15_12, _, _, _, _, _, _, _, _, _ ); + + /* Update the chaining variables. */ + ldm RSTATE, {RT0-RT3}; + add _a, RT0; + ldr RT0, [RSTATE, #state_h4]; + add _b, RT1; + add _c, RT2; + add _d, RT3; + add _e, RT0; + stm RSTATE, {_a-_e}; + + b .Loop; + +.Lend: + /* Transform 64-79 */ + R( _b, _c, _d, _e, _a, F4, 64 ); + R( _a, _b, _c, _d, _e, F4, 65 ); + R( _e, _a, _b, _c, _d, F4, 66 ); + R( _d, _e, _a, _b, _c, F4, 67 ); + R( _c, _d, _e, _a, _b, F4, 68 ); + R( _b, _c, _d, _e, _a, F4, 69 ); + R( _a, _b, _c, _d, _e, F4, 70 ); + R( _e, _a, _b, _c, _d, F4, 71 ); + R( _d, _e, _a, _b, _c, F4, 72 ); + R( _c, _d, _e, _a, _b, F4, 73 ); + R( _b, _c, _d, _e, _a, F4, 74 ); + R( _a, _b, _c, _d, _e, F4, 75 ); + R( _e, _a, _b, _c, _d, F4, 76 ); + R( _d, _e, _a, _b, _c, F4, 77 ); + R( _c, _d, _e, _a, _b, F4, 78 ); + R( _b, _c, _d, _e, _a, F4, 79 ); + + mov sp, ROLDSTACK; + + /* Update the chaining variables. */ + ldm RSTATE, {RT0-RT3}; + add _a, RT0; + ldr RT0, [RSTATE, #state_h4]; + add _b, RT1; + add _c, RT2; + add _d, RT3; + /*vpop {q4-q7};*/ + add _e, RT0; + stm RSTATE, {_a-_e}; + + pop {r4-r12, pc}; + +.Ldo_nothing: + bx lr +ENDPROC(sha1_transform_neon) diff --git a/arch/arm/crypto/sha1_glue.c b/arch/arm/crypto/sha1_glue.c index ace4cd67464c..e31b0440c613 100644 --- a/arch/arm/crypto/sha1_glue.c +++ b/arch/arm/crypto/sha1_glue.c @@ -23,32 +23,27 @@ #include <linux/types.h> #include <crypto/sha.h> #include <asm/byteorder.h> +#include <asm/crypto/sha1.h> -struct SHA1_CTX { - uint32_t h0,h1,h2,h3,h4; - u64 count; - u8 data[SHA1_BLOCK_SIZE]; -}; -asmlinkage void sha1_block_data_order(struct SHA1_CTX *digest, +asmlinkage void sha1_block_data_order(u32 *digest, const unsigned char *data, unsigned int rounds); static int sha1_init(struct shash_desc *desc) { - struct SHA1_CTX *sctx = shash_desc_ctx(desc); - memset(sctx, 0, sizeof(*sctx)); - sctx->h0 = SHA1_H0; - sctx->h1 = SHA1_H1; - sctx->h2 = SHA1_H2; - sctx->h3 = SHA1_H3; - sctx->h4 = SHA1_H4; + struct sha1_state *sctx = shash_desc_ctx(desc); + + *sctx = (struct sha1_state){ + .state = { SHA1_H0, SHA1_H1, SHA1_H2, SHA1_H3, SHA1_H4 }, + }; + return 0; } -static int __sha1_update(struct SHA1_CTX *sctx, const u8 *data, - unsigned int len, unsigned int partial) +static int __sha1_update(struct sha1_state *sctx, const u8 *data, + unsigned int len, unsigned int partial) { unsigned int done = 0; @@ -56,43 +51,44 @@ static int __sha1_update(struct SHA1_CTX *sctx, const u8 *data, if (partial) { done = SHA1_BLOCK_SIZE - partial; - memcpy(sctx->data + partial, data, done); - sha1_block_data_order(sctx, sctx->data, 1); + memcpy(sctx->buffer + partial, data, done); + sha1_block_data_order(sctx->state, sctx->buffer, 1); } if (len - done >= SHA1_BLOCK_SIZE) { const unsigned int rounds = (len - done) / SHA1_BLOCK_SIZE; - sha1_block_data_order(sctx, data + done, rounds); + sha1_block_data_order(sctx->state, data + done, rounds); done += rounds * SHA1_BLOCK_SIZE; } - memcpy(sctx->data, data + done, len - done); + memcpy(sctx->buffer, data + done, len - done); return 0; } -static int sha1_update(struct shash_desc *desc, const u8 *data, - unsigned int len) +int sha1_update_arm(struct shash_desc *desc, const u8 *data, + unsigned int len) { - struct SHA1_CTX *sctx = shash_desc_ctx(desc); + struct sha1_state *sctx = shash_desc_ctx(desc); unsigned int partial = sctx->count % SHA1_BLOCK_SIZE; int res; /* Handle the fast case right here */ if (partial + len < SHA1_BLOCK_SIZE) { sctx->count += len; - memcpy(sctx->data + partial, data, len); + memcpy(sctx->buffer + partial, data, len); return 0; } res = __sha1_update(sctx, data, len, partial); return res; } +EXPORT_SYMBOL_GPL(sha1_update_arm); /* Add padding and return the message digest. */ static int sha1_final(struct shash_desc *desc, u8 *out) { - struct SHA1_CTX *sctx = shash_desc_ctx(desc); + struct sha1_state *sctx = shash_desc_ctx(desc); unsigned int i, index, padlen; __be32 *dst = (__be32 *)out; __be64 bits; @@ -106,7 +102,7 @@ static int sha1_final(struct shash_desc *desc, u8 *out) /* We need to fill a whole block for __sha1_update() */ if (padlen <= 56) { sctx->count += padlen; - memcpy(sctx->data + index, padding, padlen); + memcpy(sctx->buffer + index, padding, padlen); } else { __sha1_update(sctx, padding, padlen, index); } @@ -114,7 +110,7 @@ static int sha1_final(struct shash_desc *desc, u8 *out) /* Store state in digest */ for (i = 0; i < 5; i++) - dst[i] = cpu_to_be32(((u32 *)sctx)[i]); + dst[i] = cpu_to_be32(sctx->state[i]); /* Wipe context */ memset(sctx, 0, sizeof(*sctx)); @@ -124,7 +120,7 @@ static int sha1_final(struct shash_desc *desc, u8 *out) static int sha1_export(struct shash_desc *desc, void *out) { - struct SHA1_CTX *sctx = shash_desc_ctx(desc); + struct sha1_state *sctx = shash_desc_ctx(desc); memcpy(out, sctx, sizeof(*sctx)); return 0; } @@ -132,7 +128,7 @@ static int sha1_export(struct shash_desc *desc, void *out) static int sha1_import(struct shash_desc *desc, const void *in) { - struct SHA1_CTX *sctx = shash_desc_ctx(desc); + struct sha1_state *sctx = shash_desc_ctx(desc); memcpy(sctx, in, sizeof(*sctx)); return 0; } @@ -141,12 +137,12 @@ static int sha1_import(struct shash_desc *desc, const void *in) static struct shash_alg alg = { .digestsize = SHA1_DIGEST_SIZE, .init = sha1_init, - .update = sha1_update, + .update = sha1_update_arm, .final = sha1_final, .export = sha1_export, .import = sha1_import, - .descsize = sizeof(struct SHA1_CTX), - .statesize = sizeof(struct SHA1_CTX), + .descsize = sizeof(struct sha1_state), + .statesize = sizeof(struct sha1_state), .base = { .cra_name = "sha1", .cra_driver_name= "sha1-asm", diff --git a/arch/arm/crypto/sha1_neon_glue.c b/arch/arm/crypto/sha1_neon_glue.c new file mode 100644 index 000000000000..6f1b411b1d55 --- /dev/null +++ b/arch/arm/crypto/sha1_neon_glue.c @@ -0,0 +1,197 @@ +/* + * Glue code for the SHA1 Secure Hash Algorithm assembler implementation using + * ARM NEON instructions. + * + * Copyright © 2014 Jussi Kivilinna <jussi.kivilinna@iki.fi> + * + * This file is based on sha1_generic.c and sha1_ssse3_glue.c: + * Copyright (c) Alan Smithee. + * Copyright (c) Andrew McDonald <andrew@mcdonald.org.uk> + * Copyright (c) Jean-Francois Dive <jef@linuxbe.org> + * Copyright (c) Mathias Krause <minipli@googlemail.com> + * Copyright (c) Chandramouli Narayanan <mouli@linux.intel.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 <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <crypto/sha.h> +#include <asm/byteorder.h> +#include <asm/neon.h> +#include <asm/simd.h> +#include <asm/crypto/sha1.h> + + +asmlinkage void sha1_transform_neon(void *state_h, const char *data, + unsigned int rounds); + + +static int sha1_neon_init(struct shash_desc *desc) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + + *sctx = (struct sha1_state){ + .state = { SHA1_H0, SHA1_H1, SHA1_H2, SHA1_H3, SHA1_H4 }, + }; + + return 0; +} + +static int __sha1_neon_update(struct shash_desc *desc, const u8 *data, + unsigned int len, unsigned int partial) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + unsigned int done = 0; + + sctx->count += len; + + if (partial) { + done = SHA1_BLOCK_SIZE - partial; + memcpy(sctx->buffer + partial, data, done); + sha1_transform_neon(sctx->state, sctx->buffer, 1); + } + + if (len - done >= SHA1_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / SHA1_BLOCK_SIZE; + + sha1_transform_neon(sctx->state, data + done, rounds); + done += rounds * SHA1_BLOCK_SIZE; + } + + memcpy(sctx->buffer, data + done, len - done); + + return 0; +} + +static int sha1_neon_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->count % SHA1_BLOCK_SIZE; + int res; + + /* Handle the fast case right here */ + if (partial + len < SHA1_BLOCK_SIZE) { + sctx->count += len; + memcpy(sctx->buffer + partial, data, len); + + return 0; + } + + if (!may_use_simd()) { + res = sha1_update_arm(desc, data, len); + } else { + kernel_neon_begin(); + res = __sha1_neon_update(desc, data, len, partial); + kernel_neon_end(); + } + + return res; +} + + +/* Add padding and return the message digest. */ +static int sha1_neon_final(struct shash_desc *desc, u8 *out) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + __be32 *dst = (__be32 *)out; + __be64 bits; + static const u8 padding[SHA1_BLOCK_SIZE] = { 0x80, }; + + bits = cpu_to_be64(sctx->count << 3); + + /* Pad out to 56 mod 64 and append length */ + index = sctx->count % SHA1_BLOCK_SIZE; + padlen = (index < 56) ? (56 - index) : ((SHA1_BLOCK_SIZE+56) - index); + if (!may_use_simd()) { + sha1_update_arm(desc, padding, padlen); + sha1_update_arm(desc, (const u8 *)&bits, sizeof(bits)); + } else { + kernel_neon_begin(); + /* We need to fill a whole block for __sha1_neon_update() */ + if (padlen <= 56) { + sctx->count += padlen; + memcpy(sctx->buffer + index, padding, padlen); + } else { + __sha1_neon_update(desc, padding, padlen, index); + } + __sha1_neon_update(desc, (const u8 *)&bits, sizeof(bits), 56); + kernel_neon_end(); + } + + /* Store state in digest */ + for (i = 0; i < 5; i++) + dst[i] = cpu_to_be32(sctx->state[i]); + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int sha1_neon_export(struct shash_desc *desc, void *out) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + + memcpy(out, sctx, sizeof(*sctx)); + + return 0; +} + +static int sha1_neon_import(struct shash_desc *desc, const void *in) +{ + struct sha1_state *sctx = shash_desc_ctx(desc); + + memcpy(sctx, in, sizeof(*sctx)); + + return 0; +} + +static struct shash_alg alg = { + .digestsize = SHA1_DIGEST_SIZE, + .init = sha1_neon_init, + .update = sha1_neon_update, + .final = sha1_neon_final, + .export = sha1_neon_export, + .import = sha1_neon_import, + .descsize = sizeof(struct sha1_state), + .statesize = sizeof(struct sha1_state), + .base = { + .cra_name = "sha1", + .cra_driver_name = "sha1-neon", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA1_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}; + +static int __init sha1_neon_mod_init(void) +{ + if (!cpu_has_neon()) + return -ENODEV; + + return crypto_register_shash(&alg); +} + +static void __exit sha1_neon_mod_fini(void) +{ + crypto_unregister_shash(&alg); +} + +module_init(sha1_neon_mod_init); +module_exit(sha1_neon_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA1 Secure Hash Algorithm, NEON accelerated"); +MODULE_ALIAS("sha1"); diff --git a/arch/arm/crypto/sha256-armv4.pl b/arch/arm/crypto/sha256-armv4.pl new file mode 100644 index 000000000000..fac0533ea633 --- /dev/null +++ b/arch/arm/crypto/sha256-armv4.pl @@ -0,0 +1,716 @@ +#!/usr/bin/env perl + +# ==================================================================== +# Written by Andy Polyakov <appro@openssl.org> for the OpenSSL +# project. The module is, however, dual licensed under OpenSSL and +# CRYPTOGAMS licenses depending on where you obtain it. For further +# details see http://www.openssl.org/~appro/cryptogams/. +# +# Permission to use under GPL terms is granted. +# ==================================================================== + +# SHA256 block procedure for ARMv4. May 2007. + +# Performance is ~2x better than gcc 3.4 generated code and in "abso- +# lute" terms is ~2250 cycles per 64-byte block or ~35 cycles per +# byte [on single-issue Xscale PXA250 core]. + +# July 2010. +# +# Rescheduling for dual-issue pipeline resulted in 22% improvement on +# Cortex A8 core and ~20 cycles per processed byte. + +# February 2011. +# +# Profiler-assisted and platform-specific optimization resulted in 16% +# improvement on Cortex A8 core and ~15.4 cycles per processed byte. + +# September 2013. +# +# Add NEON implementation. On Cortex A8 it was measured to process one +# byte in 12.5 cycles or 23% faster than integer-only code. Snapdragon +# S4 does it in 12.5 cycles too, but it's 50% faster than integer-only +# code (meaning that latter performs sub-optimally, nothing was done +# about it). + +# May 2014. +# +# Add ARMv8 code path performing at 2.0 cpb on Apple A7. + +while (($output=shift) && ($output!~/^\w[\w\-]*\.\w+$/)) {} +open STDOUT,">$output"; + +$ctx="r0"; $t0="r0"; +$inp="r1"; $t4="r1"; +$len="r2"; $t1="r2"; +$T1="r3"; $t3="r3"; +$A="r4"; +$B="r5"; +$C="r6"; +$D="r7"; +$E="r8"; +$F="r9"; +$G="r10"; +$H="r11"; +@V=($A,$B,$C,$D,$E,$F,$G,$H); +$t2="r12"; +$Ktbl="r14"; + +@Sigma0=( 2,13,22); +@Sigma1=( 6,11,25); +@sigma0=( 7,18, 3); +@sigma1=(17,19,10); + +sub BODY_00_15 { +my ($i,$a,$b,$c,$d,$e,$f,$g,$h) = @_; + +$code.=<<___ if ($i<16); +#if __ARM_ARCH__>=7 + @ ldr $t1,[$inp],#4 @ $i +# if $i==15 + str $inp,[sp,#17*4] @ make room for $t4 +# endif + eor $t0,$e,$e,ror#`$Sigma1[1]-$Sigma1[0]` + add $a,$a,$t2 @ h+=Maj(a,b,c) from the past + eor $t0,$t0,$e,ror#`$Sigma1[2]-$Sigma1[0]` @ Sigma1(e) +# ifndef __ARMEB__ + rev $t1,$t1 +# endif +#else + @ ldrb $t1,[$inp,#3] @ $i + add $a,$a,$t2 @ h+=Maj(a,b,c) from the past + ldrb $t2,[$inp,#2] + ldrb $t0,[$inp,#1] + orr $t1,$t1,$t2,lsl#8 + ldrb $t2,[$inp],#4 + orr $t1,$t1,$t0,lsl#16 +# if $i==15 + str $inp,[sp,#17*4] @ make room for $t4 +# endif + eor $t0,$e,$e,ror#`$Sigma1[1]-$Sigma1[0]` + orr $t1,$t1,$t2,lsl#24 + eor $t0,$t0,$e,ror#`$Sigma1[2]-$Sigma1[0]` @ Sigma1(e) +#endif +___ +$code.=<<___; + ldr $t2,[$Ktbl],#4 @ *K256++ + add $h,$h,$t1 @ h+=X[i] + str $t1,[sp,#`$i%16`*4] + eor $t1,$f,$g + add $h,$h,$t0,ror#$Sigma1[0] @ h+=Sigma1(e) + and $t1,$t1,$e + add $h,$h,$t2 @ h+=K256[i] + eor $t1,$t1,$g @ Ch(e,f,g) + eor $t0,$a,$a,ror#`$Sigma0[1]-$Sigma0[0]` + add $h,$h,$t1 @ h+=Ch(e,f,g) +#if $i==31 + and $t2,$t2,#0xff + cmp $t2,#0xf2 @ done? +#endif +#if $i<15 +# if __ARM_ARCH__>=7 + ldr $t1,[$inp],#4 @ prefetch +# else + ldrb $t1,[$inp,#3] +# endif + eor $t2,$a,$b @ a^b, b^c in next round +#else + ldr $t1,[sp,#`($i+2)%16`*4] @ from future BODY_16_xx + eor $t2,$a,$b @ a^b, b^c in next round + ldr $t4,[sp,#`($i+15)%16`*4] @ from future BODY_16_xx +#endif + eor $t0,$t0,$a,ror#`$Sigma0[2]-$Sigma0[0]` @ Sigma0(a) + and $t3,$t3,$t2 @ (b^c)&=(a^b) + add $d,$d,$h @ d+=h + eor $t3,$t3,$b @ Maj(a,b,c) + add $h,$h,$t0,ror#$Sigma0[0] @ h+=Sigma0(a) + @ add $h,$h,$t3 @ h+=Maj(a,b,c) +___ + ($t2,$t3)=($t3,$t2); +} + +sub BODY_16_XX { +my ($i,$a,$b,$c,$d,$e,$f,$g,$h) = @_; + +$code.=<<___; + @ ldr $t1,[sp,#`($i+1)%16`*4] @ $i + @ ldr $t4,[sp,#`($i+14)%16`*4] + mov $t0,$t1,ror#$sigma0[0] + add $a,$a,$t2 @ h+=Maj(a,b,c) from the past + mov $t2,$t4,ror#$sigma1[0] + eor $t0,$t0,$t1,ror#$sigma0[1] + eor $t2,$t2,$t4,ror#$sigma1[1] + eor $t0,$t0,$t1,lsr#$sigma0[2] @ sigma0(X[i+1]) + ldr $t1,[sp,#`($i+0)%16`*4] + eor $t2,$t2,$t4,lsr#$sigma1[2] @ sigma1(X[i+14]) + ldr $t4,[sp,#`($i+9)%16`*4] + + add $t2,$t2,$t0 + eor $t0,$e,$e,ror#`$Sigma1[1]-$Sigma1[0]` @ from BODY_00_15 + add $t1,$t1,$t2 + eor $t0,$t0,$e,ror#`$Sigma1[2]-$Sigma1[0]` @ Sigma1(e) + add $t1,$t1,$t4 @ X[i] +___ + &BODY_00_15(@_); +} + +$code=<<___; +#ifndef __KERNEL__ +# include "arm_arch.h" +#else +# define __ARM_ARCH__ __LINUX_ARM_ARCH__ +# define __ARM_MAX_ARCH__ 7 +#endif + +.text +#if __ARM_ARCH__<7 +.code 32 +#else +.syntax unified +# ifdef __thumb2__ +# define adrl adr +.thumb +# else +.code 32 +# endif +#endif + +.type K256,%object +.align 5 +K256: +.word 0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5 +.word 0x3956c25b,0x59f111f1,0x923f82a4,0xab1c5ed5 +.word 0xd807aa98,0x12835b01,0x243185be,0x550c7dc3 +.word 0x72be5d74,0x80deb1fe,0x9bdc06a7,0xc19bf174 +.word 0xe49b69c1,0xefbe4786,0x0fc19dc6,0x240ca1cc +.word 0x2de92c6f,0x4a7484aa,0x5cb0a9dc,0x76f988da +.word 0x983e5152,0xa831c66d,0xb00327c8,0xbf597fc7 +.word 0xc6e00bf3,0xd5a79147,0x06ca6351,0x14292967 +.word 0x27b70a85,0x2e1b2138,0x4d2c6dfc,0x53380d13 +.word 0x650a7354,0x766a0abb,0x81c2c92e,0x92722c85 +.word 0xa2bfe8a1,0xa81a664b,0xc24b8b70,0xc76c51a3 +.word 0xd192e819,0xd6990624,0xf40e3585,0x106aa070 +.word 0x19a4c116,0x1e376c08,0x2748774c,0x34b0bcb5 +.word 0x391c0cb3,0x4ed8aa4a,0x5b9cca4f,0x682e6ff3 +.word 0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208 +.word 0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2 +.size K256,.-K256 +.word 0 @ terminator +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) +.LOPENSSL_armcap: +.word OPENSSL_armcap_P-sha256_block_data_order +#endif +.align 5 + +.global sha256_block_data_order +.type sha256_block_data_order,%function +sha256_block_data_order: +#if __ARM_ARCH__<7 + sub r3,pc,#8 @ sha256_block_data_order +#else + adr r3,sha256_block_data_order +#endif +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) + ldr r12,.LOPENSSL_armcap + ldr r12,[r3,r12] @ OPENSSL_armcap_P + tst r12,#ARMV8_SHA256 + bne .LARMv8 + tst r12,#ARMV7_NEON + bne .LNEON +#endif + add $len,$inp,$len,lsl#6 @ len to point at the end of inp + stmdb sp!,{$ctx,$inp,$len,r4-r11,lr} + ldmia $ctx,{$A,$B,$C,$D,$E,$F,$G,$H} + sub $Ktbl,r3,#256+32 @ K256 + sub sp,sp,#16*4 @ alloca(X[16]) +.Loop: +# if __ARM_ARCH__>=7 + ldr $t1,[$inp],#4 +# else + ldrb $t1,[$inp,#3] +# endif + eor $t3,$B,$C @ magic + eor $t2,$t2,$t2 +___ +for($i=0;$i<16;$i++) { &BODY_00_15($i,@V); unshift(@V,pop(@V)); } +$code.=".Lrounds_16_xx:\n"; +for (;$i<32;$i++) { &BODY_16_XX($i,@V); unshift(@V,pop(@V)); } +$code.=<<___; +#if __ARM_ARCH__>=7 + ite eq @ Thumb2 thing, sanity check in ARM +#endif + ldreq $t3,[sp,#16*4] @ pull ctx + bne .Lrounds_16_xx + + add $A,$A,$t2 @ h+=Maj(a,b,c) from the past + ldr $t0,[$t3,#0] + ldr $t1,[$t3,#4] + ldr $t2,[$t3,#8] + add $A,$A,$t0 + ldr $t0,[$t3,#12] + add $B,$B,$t1 + ldr $t1,[$t3,#16] + add $C,$C,$t2 + ldr $t2,[$t3,#20] + add $D,$D,$t0 + ldr $t0,[$t3,#24] + add $E,$E,$t1 + ldr $t1,[$t3,#28] + add $F,$F,$t2 + ldr $inp,[sp,#17*4] @ pull inp + ldr $t2,[sp,#18*4] @ pull inp+len + add $G,$G,$t0 + add $H,$H,$t1 + stmia $t3,{$A,$B,$C,$D,$E,$F,$G,$H} + cmp $inp,$t2 + sub $Ktbl,$Ktbl,#256 @ rewind Ktbl + bne .Loop + + add sp,sp,#`16+3`*4 @ destroy frame +#if __ARM_ARCH__>=5 + ldmia sp!,{r4-r11,pc} +#else + ldmia sp!,{r4-r11,lr} + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + bx lr @ interoperable with Thumb ISA:-) +#endif +.size sha256_block_data_order,.-sha256_block_data_order +___ +###################################################################### +# NEON stuff +# +{{{ +my @X=map("q$_",(0..3)); +my ($T0,$T1,$T2,$T3,$T4,$T5)=("q8","q9","q10","q11","d24","d25"); +my $Xfer=$t4; +my $j=0; + +sub Dlo() { shift=~m|q([1]?[0-9])|?"d".($1*2):""; } +sub Dhi() { shift=~m|q([1]?[0-9])|?"d".($1*2+1):""; } + +sub AUTOLOAD() # thunk [simplified] x86-style perlasm +{ my $opcode = $AUTOLOAD; $opcode =~ s/.*:://; $opcode =~ s/_/\./; + my $arg = pop; + $arg = "#$arg" if ($arg*1 eq $arg); + $code .= "\t$opcode\t".join(',',@_,$arg)."\n"; +} + +sub Xupdate() +{ use integer; + my $body = shift; + my @insns = (&$body,&$body,&$body,&$body); + my ($a,$b,$c,$d,$e,$f,$g,$h); + + &vext_8 ($T0,@X[0],@X[1],4); # X[1..4] + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + &vext_8 ($T1,@X[2],@X[3],4); # X[9..12] + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T2,$T0,$sigma0[0]); + eval(shift(@insns)); + eval(shift(@insns)); + &vadd_i32 (@X[0],@X[0],$T1); # X[0..3] += X[9..12] + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T1,$T0,$sigma0[2]); + eval(shift(@insns)); + eval(shift(@insns)); + &vsli_32 ($T2,$T0,32-$sigma0[0]); + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T3,$T0,$sigma0[1]); + eval(shift(@insns)); + eval(shift(@insns)); + &veor ($T1,$T1,$T2); + eval(shift(@insns)); + eval(shift(@insns)); + &vsli_32 ($T3,$T0,32-$sigma0[1]); + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T4,&Dhi(@X[3]),$sigma1[0]); + eval(shift(@insns)); + eval(shift(@insns)); + &veor ($T1,$T1,$T3); # sigma0(X[1..4]) + eval(shift(@insns)); + eval(shift(@insns)); + &vsli_32 ($T4,&Dhi(@X[3]),32-$sigma1[0]); + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T5,&Dhi(@X[3]),$sigma1[2]); + eval(shift(@insns)); + eval(shift(@insns)); + &vadd_i32 (@X[0],@X[0],$T1); # X[0..3] += sigma0(X[1..4]) + eval(shift(@insns)); + eval(shift(@insns)); + &veor ($T5,$T5,$T4); + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T4,&Dhi(@X[3]),$sigma1[1]); + eval(shift(@insns)); + eval(shift(@insns)); + &vsli_32 ($T4,&Dhi(@X[3]),32-$sigma1[1]); + eval(shift(@insns)); + eval(shift(@insns)); + &veor ($T5,$T5,$T4); # sigma1(X[14..15]) + eval(shift(@insns)); + eval(shift(@insns)); + &vadd_i32 (&Dlo(@X[0]),&Dlo(@X[0]),$T5);# X[0..1] += sigma1(X[14..15]) + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T4,&Dlo(@X[0]),$sigma1[0]); + eval(shift(@insns)); + eval(shift(@insns)); + &vsli_32 ($T4,&Dlo(@X[0]),32-$sigma1[0]); + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T5,&Dlo(@X[0]),$sigma1[2]); + eval(shift(@insns)); + eval(shift(@insns)); + &veor ($T5,$T5,$T4); + eval(shift(@insns)); + eval(shift(@insns)); + &vshr_u32 ($T4,&Dlo(@X[0]),$sigma1[1]); + eval(shift(@insns)); + eval(shift(@insns)); + &vld1_32 ("{$T0}","[$Ktbl,:128]!"); + eval(shift(@insns)); + eval(shift(@insns)); + &vsli_32 ($T4,&Dlo(@X[0]),32-$sigma1[1]); + eval(shift(@insns)); + eval(shift(@insns)); + &veor ($T5,$T5,$T4); # sigma1(X[16..17]) + eval(shift(@insns)); + eval(shift(@insns)); + &vadd_i32 (&Dhi(@X[0]),&Dhi(@X[0]),$T5);# X[2..3] += sigma1(X[16..17]) + eval(shift(@insns)); + eval(shift(@insns)); + &vadd_i32 ($T0,$T0,@X[0]); + while($#insns>=2) { eval(shift(@insns)); } + &vst1_32 ("{$T0}","[$Xfer,:128]!"); + eval(shift(@insns)); + eval(shift(@insns)); + + push(@X,shift(@X)); # "rotate" X[] +} + +sub Xpreload() +{ use integer; + my $body = shift; + my @insns = (&$body,&$body,&$body,&$body); + my ($a,$b,$c,$d,$e,$f,$g,$h); + + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + &vld1_32 ("{$T0}","[$Ktbl,:128]!"); + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + &vrev32_8 (@X[0],@X[0]); + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + eval(shift(@insns)); + &vadd_i32 ($T0,$T0,@X[0]); + foreach (@insns) { eval; } # remaining instructions + &vst1_32 ("{$T0}","[$Xfer,:128]!"); + + push(@X,shift(@X)); # "rotate" X[] +} + +sub body_00_15 () { + ( + '($a,$b,$c,$d,$e,$f,$g,$h)=@V;'. + '&add ($h,$h,$t1)', # h+=X[i]+K[i] + '&eor ($t1,$f,$g)', + '&eor ($t0,$e,$e,"ror#".($Sigma1[1]-$Sigma1[0]))', + '&add ($a,$a,$t2)', # h+=Maj(a,b,c) from the past + '&and ($t1,$t1,$e)', + '&eor ($t2,$t0,$e,"ror#".($Sigma1[2]-$Sigma1[0]))', # Sigma1(e) + '&eor ($t0,$a,$a,"ror#".($Sigma0[1]-$Sigma0[0]))', + '&eor ($t1,$t1,$g)', # Ch(e,f,g) + '&add ($h,$h,$t2,"ror#$Sigma1[0]")', # h+=Sigma1(e) + '&eor ($t2,$a,$b)', # a^b, b^c in next round + '&eor ($t0,$t0,$a,"ror#".($Sigma0[2]-$Sigma0[0]))', # Sigma0(a) + '&add ($h,$h,$t1)', # h+=Ch(e,f,g) + '&ldr ($t1,sprintf "[sp,#%d]",4*(($j+1)&15)) if (($j&15)!=15);'. + '&ldr ($t1,"[$Ktbl]") if ($j==15);'. + '&ldr ($t1,"[sp,#64]") if ($j==31)', + '&and ($t3,$t3,$t2)', # (b^c)&=(a^b) + '&add ($d,$d,$h)', # d+=h + '&add ($h,$h,$t0,"ror#$Sigma0[0]");'. # h+=Sigma0(a) + '&eor ($t3,$t3,$b)', # Maj(a,b,c) + '$j++; unshift(@V,pop(@V)); ($t2,$t3)=($t3,$t2);' + ) +} + +$code.=<<___; +#if __ARM_MAX_ARCH__>=7 +.arch armv7-a +.fpu neon + +.global sha256_block_data_order_neon +.type sha256_block_data_order_neon,%function +.align 4 +sha256_block_data_order_neon: +.LNEON: + stmdb sp!,{r4-r12,lr} + + sub $H,sp,#16*4+16 + adrl $Ktbl,K256 + bic $H,$H,#15 @ align for 128-bit stores + mov $t2,sp + mov sp,$H @ alloca + add $len,$inp,$len,lsl#6 @ len to point at the end of inp + + vld1.8 {@X[0]},[$inp]! + vld1.8 {@X[1]},[$inp]! + vld1.8 {@X[2]},[$inp]! + vld1.8 {@X[3]},[$inp]! + vld1.32 {$T0},[$Ktbl,:128]! + vld1.32 {$T1},[$Ktbl,:128]! + vld1.32 {$T2},[$Ktbl,:128]! + vld1.32 {$T3},[$Ktbl,:128]! + vrev32.8 @X[0],@X[0] @ yes, even on + str $ctx,[sp,#64] + vrev32.8 @X[1],@X[1] @ big-endian + str $inp,[sp,#68] + mov $Xfer,sp + vrev32.8 @X[2],@X[2] + str $len,[sp,#72] + vrev32.8 @X[3],@X[3] + str $t2,[sp,#76] @ save original sp + vadd.i32 $T0,$T0,@X[0] + vadd.i32 $T1,$T1,@X[1] + vst1.32 {$T0},[$Xfer,:128]! + vadd.i32 $T2,$T2,@X[2] + vst1.32 {$T1},[$Xfer,:128]! + vadd.i32 $T3,$T3,@X[3] + vst1.32 {$T2},[$Xfer,:128]! + vst1.32 {$T3},[$Xfer,:128]! + + ldmia $ctx,{$A-$H} + sub $Xfer,$Xfer,#64 + ldr $t1,[sp,#0] + eor $t2,$t2,$t2 + eor $t3,$B,$C + b .L_00_48 + +.align 4 +.L_00_48: +___ + &Xupdate(\&body_00_15); + &Xupdate(\&body_00_15); + &Xupdate(\&body_00_15); + &Xupdate(\&body_00_15); +$code.=<<___; + teq $t1,#0 @ check for K256 terminator + ldr $t1,[sp,#0] + sub $Xfer,$Xfer,#64 + bne .L_00_48 + + ldr $inp,[sp,#68] + ldr $t0,[sp,#72] + sub $Ktbl,$Ktbl,#256 @ rewind $Ktbl + teq $inp,$t0 + it eq + subeq $inp,$inp,#64 @ avoid SEGV + vld1.8 {@X[0]},[$inp]! @ load next input block + vld1.8 {@X[1]},[$inp]! + vld1.8 {@X[2]},[$inp]! + vld1.8 {@X[3]},[$inp]! + it ne + strne $inp,[sp,#68] + mov $Xfer,sp +___ + &Xpreload(\&body_00_15); + &Xpreload(\&body_00_15); + &Xpreload(\&body_00_15); + &Xpreload(\&body_00_15); +$code.=<<___; + ldr $t0,[$t1,#0] + add $A,$A,$t2 @ h+=Maj(a,b,c) from the past + ldr $t2,[$t1,#4] + ldr $t3,[$t1,#8] + ldr $t4,[$t1,#12] + add $A,$A,$t0 @ accumulate + ldr $t0,[$t1,#16] + add $B,$B,$t2 + ldr $t2,[$t1,#20] + add $C,$C,$t3 + ldr $t3,[$t1,#24] + add $D,$D,$t4 + ldr $t4,[$t1,#28] + add $E,$E,$t0 + str $A,[$t1],#4 + add $F,$F,$t2 + str $B,[$t1],#4 + add $G,$G,$t3 + str $C,[$t1],#4 + add $H,$H,$t4 + str $D,[$t1],#4 + stmia $t1,{$E-$H} + + ittte ne + movne $Xfer,sp + ldrne $t1,[sp,#0] + eorne $t2,$t2,$t2 + ldreq sp,[sp,#76] @ restore original sp + itt ne + eorne $t3,$B,$C + bne .L_00_48 + + ldmia sp!,{r4-r12,pc} +.size sha256_block_data_order_neon,.-sha256_block_data_order_neon +#endif +___ +}}} +###################################################################### +# ARMv8 stuff +# +{{{ +my ($ABCD,$EFGH,$abcd)=map("q$_",(0..2)); +my @MSG=map("q$_",(8..11)); +my ($W0,$W1,$ABCD_SAVE,$EFGH_SAVE)=map("q$_",(12..15)); +my $Ktbl="r3"; + +$code.=<<___; +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) + +# ifdef __thumb2__ +# define INST(a,b,c,d) .byte c,d|0xc,a,b +# else +# define INST(a,b,c,d) .byte a,b,c,d +# endif + +.type sha256_block_data_order_armv8,%function +.align 5 +sha256_block_data_order_armv8: +.LARMv8: + vld1.32 {$ABCD,$EFGH},[$ctx] +# ifdef __thumb2__ + adr $Ktbl,.LARMv8 + sub $Ktbl,$Ktbl,#.LARMv8-K256 +# else + adrl $Ktbl,K256 +# endif + add $len,$inp,$len,lsl#6 @ len to point at the end of inp + +.Loop_v8: + vld1.8 {@MSG[0]-@MSG[1]},[$inp]! + vld1.8 {@MSG[2]-@MSG[3]},[$inp]! + vld1.32 {$W0},[$Ktbl]! + vrev32.8 @MSG[0],@MSG[0] + vrev32.8 @MSG[1],@MSG[1] + vrev32.8 @MSG[2],@MSG[2] + vrev32.8 @MSG[3],@MSG[3] + vmov $ABCD_SAVE,$ABCD @ offload + vmov $EFGH_SAVE,$EFGH + teq $inp,$len +___ +for($i=0;$i<12;$i++) { +$code.=<<___; + vld1.32 {$W1},[$Ktbl]! + vadd.i32 $W0,$W0,@MSG[0] + sha256su0 @MSG[0],@MSG[1] + vmov $abcd,$ABCD + sha256h $ABCD,$EFGH,$W0 + sha256h2 $EFGH,$abcd,$W0 + sha256su1 @MSG[0],@MSG[2],@MSG[3] +___ + ($W0,$W1)=($W1,$W0); push(@MSG,shift(@MSG)); +} +$code.=<<___; + vld1.32 {$W1},[$Ktbl]! + vadd.i32 $W0,$W0,@MSG[0] + vmov $abcd,$ABCD + sha256h $ABCD,$EFGH,$W0 + sha256h2 $EFGH,$abcd,$W0 + + vld1.32 {$W0},[$Ktbl]! + vadd.i32 $W1,$W1,@MSG[1] + vmov $abcd,$ABCD + sha256h $ABCD,$EFGH,$W1 + sha256h2 $EFGH,$abcd,$W1 + + vld1.32 {$W1},[$Ktbl] + vadd.i32 $W0,$W0,@MSG[2] + sub $Ktbl,$Ktbl,#256-16 @ rewind + vmov $abcd,$ABCD + sha256h $ABCD,$EFGH,$W0 + sha256h2 $EFGH,$abcd,$W0 + + vadd.i32 $W1,$W1,@MSG[3] + vmov $abcd,$ABCD + sha256h $ABCD,$EFGH,$W1 + sha256h2 $EFGH,$abcd,$W1 + + vadd.i32 $ABCD,$ABCD,$ABCD_SAVE + vadd.i32 $EFGH,$EFGH,$EFGH_SAVE + it ne + bne .Loop_v8 + + vst1.32 {$ABCD,$EFGH},[$ctx] + + ret @ bx lr +.size sha256_block_data_order_armv8,.-sha256_block_data_order_armv8 +#endif +___ +}}} +$code.=<<___; +.asciz "SHA256 block transform for ARMv4/NEON/ARMv8, CRYPTOGAMS by <appro\@openssl.org>" +.align 2 +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) +.comm OPENSSL_armcap_P,4,4 +#endif +___ + +open SELF,$0; +while(<SELF>) { + next if (/^#!/); + last if (!s/^#/@/ and !/^$/); + print; +} +close SELF; + +{ my %opcode = ( + "sha256h" => 0xf3000c40, "sha256h2" => 0xf3100c40, + "sha256su0" => 0xf3ba03c0, "sha256su1" => 0xf3200c40 ); + + sub unsha256 { + my ($mnemonic,$arg)=@_; + + if ($arg =~ m/q([0-9]+)(?:,\s*q([0-9]+))?,\s*q([0-9]+)/o) { + my $word = $opcode{$mnemonic}|(($1&7)<<13)|(($1&8)<<19) + |(($2&7)<<17)|(($2&8)<<4) + |(($3&7)<<1) |(($3&8)<<2); + # since ARMv7 instructions are always encoded little-endian. + # correct solution is to use .inst directive, but older + # assemblers don't implement it:-( + sprintf "INST(0x%02x,0x%02x,0x%02x,0x%02x)\t@ %s %s", + $word&0xff,($word>>8)&0xff, + ($word>>16)&0xff,($word>>24)&0xff, + $mnemonic,$arg; + } + } +} + +foreach (split($/,$code)) { + + s/\`([^\`]*)\`/eval $1/geo; + + s/\b(sha256\w+)\s+(q.*)/unsha256($1,$2)/geo; + + s/\bret\b/bx lr/go or + s/\bbx\s+lr\b/.word\t0xe12fff1e/go; # make it possible to compile with -march=armv4 + + print $_,"\n"; +} + +close STDOUT; # enforce flush diff --git a/arch/arm/crypto/sha256-core.S_shipped b/arch/arm/crypto/sha256-core.S_shipped new file mode 100644 index 000000000000..555a1a8eec90 --- /dev/null +++ b/arch/arm/crypto/sha256-core.S_shipped @@ -0,0 +1,2808 @@ + +@ ==================================================================== +@ Written by Andy Polyakov <appro@openssl.org> for the OpenSSL +@ project. The module is, however, dual licensed under OpenSSL and +@ CRYPTOGAMS licenses depending on where you obtain it. For further +@ details see http://www.openssl.org/~appro/cryptogams/. +@ +@ Permission to use under GPL terms is granted. +@ ==================================================================== + +@ SHA256 block procedure for ARMv4. May 2007. + +@ Performance is ~2x better than gcc 3.4 generated code and in "abso- +@ lute" terms is ~2250 cycles per 64-byte block or ~35 cycles per +@ byte [on single-issue Xscale PXA250 core]. + +@ July 2010. +@ +@ Rescheduling for dual-issue pipeline resulted in 22% improvement on +@ Cortex A8 core and ~20 cycles per processed byte. + +@ February 2011. +@ +@ Profiler-assisted and platform-specific optimization resulted in 16% +@ improvement on Cortex A8 core and ~15.4 cycles per processed byte. + +@ September 2013. +@ +@ Add NEON implementation. On Cortex A8 it was measured to process one +@ byte in 12.5 cycles or 23% faster than integer-only code. Snapdragon +@ S4 does it in 12.5 cycles too, but it's 50% faster than integer-only +@ code (meaning that latter performs sub-optimally, nothing was done +@ about it). + +@ May 2014. +@ +@ Add ARMv8 code path performing at 2.0 cpb on Apple A7. + +#ifndef __KERNEL__ +# include "arm_arch.h" +#else +# define __ARM_ARCH__ __LINUX_ARM_ARCH__ +# define __ARM_MAX_ARCH__ 7 +#endif + +.text +#if __ARM_ARCH__<7 +.code 32 +#else +.syntax unified +# ifdef __thumb2__ +# define adrl adr +.thumb +# else +.code 32 +# endif +#endif + +.type K256,%object +.align 5 +K256: +.word 0x428a2f98,0x71374491,0xb5c0fbcf,0xe9b5dba5 +.word 0x3956c25b,0x59f111f1,0x923f82a4,0xab1c5ed5 +.word 0xd807aa98,0x12835b01,0x243185be,0x550c7dc3 +.word 0x72be5d74,0x80deb1fe,0x9bdc06a7,0xc19bf174 +.word 0xe49b69c1,0xefbe4786,0x0fc19dc6,0x240ca1cc +.word 0x2de92c6f,0x4a7484aa,0x5cb0a9dc,0x76f988da +.word 0x983e5152,0xa831c66d,0xb00327c8,0xbf597fc7 +.word 0xc6e00bf3,0xd5a79147,0x06ca6351,0x14292967 +.word 0x27b70a85,0x2e1b2138,0x4d2c6dfc,0x53380d13 +.word 0x650a7354,0x766a0abb,0x81c2c92e,0x92722c85 +.word 0xa2bfe8a1,0xa81a664b,0xc24b8b70,0xc76c51a3 +.word 0xd192e819,0xd6990624,0xf40e3585,0x106aa070 +.word 0x19a4c116,0x1e376c08,0x2748774c,0x34b0bcb5 +.word 0x391c0cb3,0x4ed8aa4a,0x5b9cca4f,0x682e6ff3 +.word 0x748f82ee,0x78a5636f,0x84c87814,0x8cc70208 +.word 0x90befffa,0xa4506ceb,0xbef9a3f7,0xc67178f2 +.size K256,.-K256 +.word 0 @ terminator +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) +.LOPENSSL_armcap: +.word OPENSSL_armcap_P-sha256_block_data_order +#endif +.align 5 + +.global sha256_block_data_order +.type sha256_block_data_order,%function +sha256_block_data_order: +#if __ARM_ARCH__<7 + sub r3,pc,#8 @ sha256_block_data_order +#else + adr r3,sha256_block_data_order +#endif +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) + ldr r12,.LOPENSSL_armcap + ldr r12,[r3,r12] @ OPENSSL_armcap_P + tst r12,#ARMV8_SHA256 + bne .LARMv8 + tst r12,#ARMV7_NEON + bne .LNEON +#endif + add r2,r1,r2,lsl#6 @ len to point at the end of inp + stmdb sp!,{r0,r1,r2,r4-r11,lr} + ldmia r0,{r4,r5,r6,r7,r8,r9,r10,r11} + sub r14,r3,#256+32 @ K256 + sub sp,sp,#16*4 @ alloca(X[16]) +.Loop: +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 +# else + ldrb r2,[r1,#3] +# endif + eor r3,r5,r6 @ magic + eor r12,r12,r12 +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 0 +# if 0==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r8,r8,ror#5 + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r8,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 0 + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 0==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r8,r8,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r8,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r11,r11,r2 @ h+=X[i] + str r2,[sp,#0*4] + eor r2,r9,r10 + add r11,r11,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r8 + add r11,r11,r12 @ h+=K256[i] + eor r2,r2,r10 @ Ch(e,f,g) + eor r0,r4,r4,ror#11 + add r11,r11,r2 @ h+=Ch(e,f,g) +#if 0==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 0<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r4,r5 @ a^b, b^c in next round +#else + ldr r2,[sp,#2*4] @ from future BODY_16_xx + eor r12,r4,r5 @ a^b, b^c in next round + ldr r1,[sp,#15*4] @ from future BODY_16_xx +#endif + eor r0,r0,r4,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r7,r7,r11 @ d+=h + eor r3,r3,r5 @ Maj(a,b,c) + add r11,r11,r0,ror#2 @ h+=Sigma0(a) + @ add r11,r11,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 1 +# if 1==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r7,r7,ror#5 + add r11,r11,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r7,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 1 + add r11,r11,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 1==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r7,r7,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r7,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r10,r10,r2 @ h+=X[i] + str r2,[sp,#1*4] + eor r2,r8,r9 + add r10,r10,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r7 + add r10,r10,r3 @ h+=K256[i] + eor r2,r2,r9 @ Ch(e,f,g) + eor r0,r11,r11,ror#11 + add r10,r10,r2 @ h+=Ch(e,f,g) +#if 1==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 1<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r11,r4 @ a^b, b^c in next round +#else + ldr r2,[sp,#3*4] @ from future BODY_16_xx + eor r3,r11,r4 @ a^b, b^c in next round + ldr r1,[sp,#0*4] @ from future BODY_16_xx +#endif + eor r0,r0,r11,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r6,r6,r10 @ d+=h + eor r12,r12,r4 @ Maj(a,b,c) + add r10,r10,r0,ror#2 @ h+=Sigma0(a) + @ add r10,r10,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 2 +# if 2==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r6,r6,ror#5 + add r10,r10,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r6,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 2 + add r10,r10,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 2==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r6,r6,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r6,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r9,r9,r2 @ h+=X[i] + str r2,[sp,#2*4] + eor r2,r7,r8 + add r9,r9,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r6 + add r9,r9,r12 @ h+=K256[i] + eor r2,r2,r8 @ Ch(e,f,g) + eor r0,r10,r10,ror#11 + add r9,r9,r2 @ h+=Ch(e,f,g) +#if 2==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 2<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r10,r11 @ a^b, b^c in next round +#else + ldr r2,[sp,#4*4] @ from future BODY_16_xx + eor r12,r10,r11 @ a^b, b^c in next round + ldr r1,[sp,#1*4] @ from future BODY_16_xx +#endif + eor r0,r0,r10,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r5,r5,r9 @ d+=h + eor r3,r3,r11 @ Maj(a,b,c) + add r9,r9,r0,ror#2 @ h+=Sigma0(a) + @ add r9,r9,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 3 +# if 3==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r5,r5,ror#5 + add r9,r9,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r5,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 3 + add r9,r9,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 3==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r5,r5,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r5,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r8,r8,r2 @ h+=X[i] + str r2,[sp,#3*4] + eor r2,r6,r7 + add r8,r8,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r5 + add r8,r8,r3 @ h+=K256[i] + eor r2,r2,r7 @ Ch(e,f,g) + eor r0,r9,r9,ror#11 + add r8,r8,r2 @ h+=Ch(e,f,g) +#if 3==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 3<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r9,r10 @ a^b, b^c in next round +#else + ldr r2,[sp,#5*4] @ from future BODY_16_xx + eor r3,r9,r10 @ a^b, b^c in next round + ldr r1,[sp,#2*4] @ from future BODY_16_xx +#endif + eor r0,r0,r9,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r4,r4,r8 @ d+=h + eor r12,r12,r10 @ Maj(a,b,c) + add r8,r8,r0,ror#2 @ h+=Sigma0(a) + @ add r8,r8,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 4 +# if 4==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r4,r4,ror#5 + add r8,r8,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r4,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 4 + add r8,r8,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 4==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r4,r4,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r4,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r7,r7,r2 @ h+=X[i] + str r2,[sp,#4*4] + eor r2,r5,r6 + add r7,r7,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r4 + add r7,r7,r12 @ h+=K256[i] + eor r2,r2,r6 @ Ch(e,f,g) + eor r0,r8,r8,ror#11 + add r7,r7,r2 @ h+=Ch(e,f,g) +#if 4==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 4<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r8,r9 @ a^b, b^c in next round +#else + ldr r2,[sp,#6*4] @ from future BODY_16_xx + eor r12,r8,r9 @ a^b, b^c in next round + ldr r1,[sp,#3*4] @ from future BODY_16_xx +#endif + eor r0,r0,r8,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r11,r11,r7 @ d+=h + eor r3,r3,r9 @ Maj(a,b,c) + add r7,r7,r0,ror#2 @ h+=Sigma0(a) + @ add r7,r7,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 5 +# if 5==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r11,r11,ror#5 + add r7,r7,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r11,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 5 + add r7,r7,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 5==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r11,r11,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r11,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r6,r6,r2 @ h+=X[i] + str r2,[sp,#5*4] + eor r2,r4,r5 + add r6,r6,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r11 + add r6,r6,r3 @ h+=K256[i] + eor r2,r2,r5 @ Ch(e,f,g) + eor r0,r7,r7,ror#11 + add r6,r6,r2 @ h+=Ch(e,f,g) +#if 5==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 5<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r7,r8 @ a^b, b^c in next round +#else + ldr r2,[sp,#7*4] @ from future BODY_16_xx + eor r3,r7,r8 @ a^b, b^c in next round + ldr r1,[sp,#4*4] @ from future BODY_16_xx +#endif + eor r0,r0,r7,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r10,r10,r6 @ d+=h + eor r12,r12,r8 @ Maj(a,b,c) + add r6,r6,r0,ror#2 @ h+=Sigma0(a) + @ add r6,r6,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 6 +# if 6==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r10,r10,ror#5 + add r6,r6,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r10,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 6 + add r6,r6,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 6==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r10,r10,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r10,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r5,r5,r2 @ h+=X[i] + str r2,[sp,#6*4] + eor r2,r11,r4 + add r5,r5,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r10 + add r5,r5,r12 @ h+=K256[i] + eor r2,r2,r4 @ Ch(e,f,g) + eor r0,r6,r6,ror#11 + add r5,r5,r2 @ h+=Ch(e,f,g) +#if 6==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 6<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r6,r7 @ a^b, b^c in next round +#else + ldr r2,[sp,#8*4] @ from future BODY_16_xx + eor r12,r6,r7 @ a^b, b^c in next round + ldr r1,[sp,#5*4] @ from future BODY_16_xx +#endif + eor r0,r0,r6,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r9,r9,r5 @ d+=h + eor r3,r3,r7 @ Maj(a,b,c) + add r5,r5,r0,ror#2 @ h+=Sigma0(a) + @ add r5,r5,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 7 +# if 7==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r9,r9,ror#5 + add r5,r5,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r9,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 7 + add r5,r5,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 7==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r9,r9,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r9,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r4,r4,r2 @ h+=X[i] + str r2,[sp,#7*4] + eor r2,r10,r11 + add r4,r4,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r9 + add r4,r4,r3 @ h+=K256[i] + eor r2,r2,r11 @ Ch(e,f,g) + eor r0,r5,r5,ror#11 + add r4,r4,r2 @ h+=Ch(e,f,g) +#if 7==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 7<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r5,r6 @ a^b, b^c in next round +#else + ldr r2,[sp,#9*4] @ from future BODY_16_xx + eor r3,r5,r6 @ a^b, b^c in next round + ldr r1,[sp,#6*4] @ from future BODY_16_xx +#endif + eor r0,r0,r5,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r8,r8,r4 @ d+=h + eor r12,r12,r6 @ Maj(a,b,c) + add r4,r4,r0,ror#2 @ h+=Sigma0(a) + @ add r4,r4,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 8 +# if 8==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r8,r8,ror#5 + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r8,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 8 + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 8==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r8,r8,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r8,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r11,r11,r2 @ h+=X[i] + str r2,[sp,#8*4] + eor r2,r9,r10 + add r11,r11,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r8 + add r11,r11,r12 @ h+=K256[i] + eor r2,r2,r10 @ Ch(e,f,g) + eor r0,r4,r4,ror#11 + add r11,r11,r2 @ h+=Ch(e,f,g) +#if 8==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 8<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r4,r5 @ a^b, b^c in next round +#else + ldr r2,[sp,#10*4] @ from future BODY_16_xx + eor r12,r4,r5 @ a^b, b^c in next round + ldr r1,[sp,#7*4] @ from future BODY_16_xx +#endif + eor r0,r0,r4,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r7,r7,r11 @ d+=h + eor r3,r3,r5 @ Maj(a,b,c) + add r11,r11,r0,ror#2 @ h+=Sigma0(a) + @ add r11,r11,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 9 +# if 9==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r7,r7,ror#5 + add r11,r11,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r7,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 9 + add r11,r11,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 9==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r7,r7,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r7,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r10,r10,r2 @ h+=X[i] + str r2,[sp,#9*4] + eor r2,r8,r9 + add r10,r10,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r7 + add r10,r10,r3 @ h+=K256[i] + eor r2,r2,r9 @ Ch(e,f,g) + eor r0,r11,r11,ror#11 + add r10,r10,r2 @ h+=Ch(e,f,g) +#if 9==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 9<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r11,r4 @ a^b, b^c in next round +#else + ldr r2,[sp,#11*4] @ from future BODY_16_xx + eor r3,r11,r4 @ a^b, b^c in next round + ldr r1,[sp,#8*4] @ from future BODY_16_xx +#endif + eor r0,r0,r11,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r6,r6,r10 @ d+=h + eor r12,r12,r4 @ Maj(a,b,c) + add r10,r10,r0,ror#2 @ h+=Sigma0(a) + @ add r10,r10,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 10 +# if 10==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r6,r6,ror#5 + add r10,r10,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r6,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 10 + add r10,r10,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 10==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r6,r6,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r6,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r9,r9,r2 @ h+=X[i] + str r2,[sp,#10*4] + eor r2,r7,r8 + add r9,r9,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r6 + add r9,r9,r12 @ h+=K256[i] + eor r2,r2,r8 @ Ch(e,f,g) + eor r0,r10,r10,ror#11 + add r9,r9,r2 @ h+=Ch(e,f,g) +#if 10==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 10<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r10,r11 @ a^b, b^c in next round +#else + ldr r2,[sp,#12*4] @ from future BODY_16_xx + eor r12,r10,r11 @ a^b, b^c in next round + ldr r1,[sp,#9*4] @ from future BODY_16_xx +#endif + eor r0,r0,r10,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r5,r5,r9 @ d+=h + eor r3,r3,r11 @ Maj(a,b,c) + add r9,r9,r0,ror#2 @ h+=Sigma0(a) + @ add r9,r9,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 11 +# if 11==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r5,r5,ror#5 + add r9,r9,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r5,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 11 + add r9,r9,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 11==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r5,r5,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r5,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r8,r8,r2 @ h+=X[i] + str r2,[sp,#11*4] + eor r2,r6,r7 + add r8,r8,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r5 + add r8,r8,r3 @ h+=K256[i] + eor r2,r2,r7 @ Ch(e,f,g) + eor r0,r9,r9,ror#11 + add r8,r8,r2 @ h+=Ch(e,f,g) +#if 11==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 11<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r9,r10 @ a^b, b^c in next round +#else + ldr r2,[sp,#13*4] @ from future BODY_16_xx + eor r3,r9,r10 @ a^b, b^c in next round + ldr r1,[sp,#10*4] @ from future BODY_16_xx +#endif + eor r0,r0,r9,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r4,r4,r8 @ d+=h + eor r12,r12,r10 @ Maj(a,b,c) + add r8,r8,r0,ror#2 @ h+=Sigma0(a) + @ add r8,r8,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 12 +# if 12==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r4,r4,ror#5 + add r8,r8,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r4,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 12 + add r8,r8,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 12==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r4,r4,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r4,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r7,r7,r2 @ h+=X[i] + str r2,[sp,#12*4] + eor r2,r5,r6 + add r7,r7,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r4 + add r7,r7,r12 @ h+=K256[i] + eor r2,r2,r6 @ Ch(e,f,g) + eor r0,r8,r8,ror#11 + add r7,r7,r2 @ h+=Ch(e,f,g) +#if 12==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 12<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r8,r9 @ a^b, b^c in next round +#else + ldr r2,[sp,#14*4] @ from future BODY_16_xx + eor r12,r8,r9 @ a^b, b^c in next round + ldr r1,[sp,#11*4] @ from future BODY_16_xx +#endif + eor r0,r0,r8,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r11,r11,r7 @ d+=h + eor r3,r3,r9 @ Maj(a,b,c) + add r7,r7,r0,ror#2 @ h+=Sigma0(a) + @ add r7,r7,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 13 +# if 13==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r11,r11,ror#5 + add r7,r7,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r11,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 13 + add r7,r7,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 13==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r11,r11,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r11,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r6,r6,r2 @ h+=X[i] + str r2,[sp,#13*4] + eor r2,r4,r5 + add r6,r6,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r11 + add r6,r6,r3 @ h+=K256[i] + eor r2,r2,r5 @ Ch(e,f,g) + eor r0,r7,r7,ror#11 + add r6,r6,r2 @ h+=Ch(e,f,g) +#if 13==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 13<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r7,r8 @ a^b, b^c in next round +#else + ldr r2,[sp,#15*4] @ from future BODY_16_xx + eor r3,r7,r8 @ a^b, b^c in next round + ldr r1,[sp,#12*4] @ from future BODY_16_xx +#endif + eor r0,r0,r7,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r10,r10,r6 @ d+=h + eor r12,r12,r8 @ Maj(a,b,c) + add r6,r6,r0,ror#2 @ h+=Sigma0(a) + @ add r6,r6,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 14 +# if 14==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r10,r10,ror#5 + add r6,r6,r12 @ h+=Maj(a,b,c) from the past + eor r0,r0,r10,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 14 + add r6,r6,r12 @ h+=Maj(a,b,c) from the past + ldrb r12,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r12,lsl#8 + ldrb r12,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 14==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r10,r10,ror#5 + orr r2,r2,r12,lsl#24 + eor r0,r0,r10,ror#19 @ Sigma1(e) +#endif + ldr r12,[r14],#4 @ *K256++ + add r5,r5,r2 @ h+=X[i] + str r2,[sp,#14*4] + eor r2,r11,r4 + add r5,r5,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r10 + add r5,r5,r12 @ h+=K256[i] + eor r2,r2,r4 @ Ch(e,f,g) + eor r0,r6,r6,ror#11 + add r5,r5,r2 @ h+=Ch(e,f,g) +#if 14==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 14<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r6,r7 @ a^b, b^c in next round +#else + ldr r2,[sp,#0*4] @ from future BODY_16_xx + eor r12,r6,r7 @ a^b, b^c in next round + ldr r1,[sp,#13*4] @ from future BODY_16_xx +#endif + eor r0,r0,r6,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r9,r9,r5 @ d+=h + eor r3,r3,r7 @ Maj(a,b,c) + add r5,r5,r0,ror#2 @ h+=Sigma0(a) + @ add r5,r5,r3 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + @ ldr r2,[r1],#4 @ 15 +# if 15==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r9,r9,ror#5 + add r5,r5,r3 @ h+=Maj(a,b,c) from the past + eor r0,r0,r9,ror#19 @ Sigma1(e) +# ifndef __ARMEB__ + rev r2,r2 +# endif +#else + @ ldrb r2,[r1,#3] @ 15 + add r5,r5,r3 @ h+=Maj(a,b,c) from the past + ldrb r3,[r1,#2] + ldrb r0,[r1,#1] + orr r2,r2,r3,lsl#8 + ldrb r3,[r1],#4 + orr r2,r2,r0,lsl#16 +# if 15==15 + str r1,[sp,#17*4] @ make room for r1 +# endif + eor r0,r9,r9,ror#5 + orr r2,r2,r3,lsl#24 + eor r0,r0,r9,ror#19 @ Sigma1(e) +#endif + ldr r3,[r14],#4 @ *K256++ + add r4,r4,r2 @ h+=X[i] + str r2,[sp,#15*4] + eor r2,r10,r11 + add r4,r4,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r9 + add r4,r4,r3 @ h+=K256[i] + eor r2,r2,r11 @ Ch(e,f,g) + eor r0,r5,r5,ror#11 + add r4,r4,r2 @ h+=Ch(e,f,g) +#if 15==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 15<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r5,r6 @ a^b, b^c in next round +#else + ldr r2,[sp,#1*4] @ from future BODY_16_xx + eor r3,r5,r6 @ a^b, b^c in next round + ldr r1,[sp,#14*4] @ from future BODY_16_xx +#endif + eor r0,r0,r5,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r8,r8,r4 @ d+=h + eor r12,r12,r6 @ Maj(a,b,c) + add r4,r4,r0,ror#2 @ h+=Sigma0(a) + @ add r4,r4,r12 @ h+=Maj(a,b,c) +.Lrounds_16_xx: + @ ldr r2,[sp,#1*4] @ 16 + @ ldr r1,[sp,#14*4] + mov r0,r2,ror#7 + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#0*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#9*4] + + add r12,r12,r0 + eor r0,r8,r8,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r8,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r11,r11,r2 @ h+=X[i] + str r2,[sp,#0*4] + eor r2,r9,r10 + add r11,r11,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r8 + add r11,r11,r12 @ h+=K256[i] + eor r2,r2,r10 @ Ch(e,f,g) + eor r0,r4,r4,ror#11 + add r11,r11,r2 @ h+=Ch(e,f,g) +#if 16==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 16<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r4,r5 @ a^b, b^c in next round +#else + ldr r2,[sp,#2*4] @ from future BODY_16_xx + eor r12,r4,r5 @ a^b, b^c in next round + ldr r1,[sp,#15*4] @ from future BODY_16_xx +#endif + eor r0,r0,r4,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r7,r7,r11 @ d+=h + eor r3,r3,r5 @ Maj(a,b,c) + add r11,r11,r0,ror#2 @ h+=Sigma0(a) + @ add r11,r11,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#2*4] @ 17 + @ ldr r1,[sp,#15*4] + mov r0,r2,ror#7 + add r11,r11,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#1*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#10*4] + + add r3,r3,r0 + eor r0,r7,r7,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r7,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r10,r10,r2 @ h+=X[i] + str r2,[sp,#1*4] + eor r2,r8,r9 + add r10,r10,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r7 + add r10,r10,r3 @ h+=K256[i] + eor r2,r2,r9 @ Ch(e,f,g) + eor r0,r11,r11,ror#11 + add r10,r10,r2 @ h+=Ch(e,f,g) +#if 17==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 17<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r11,r4 @ a^b, b^c in next round +#else + ldr r2,[sp,#3*4] @ from future BODY_16_xx + eor r3,r11,r4 @ a^b, b^c in next round + ldr r1,[sp,#0*4] @ from future BODY_16_xx +#endif + eor r0,r0,r11,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r6,r6,r10 @ d+=h + eor r12,r12,r4 @ Maj(a,b,c) + add r10,r10,r0,ror#2 @ h+=Sigma0(a) + @ add r10,r10,r12 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#3*4] @ 18 + @ ldr r1,[sp,#0*4] + mov r0,r2,ror#7 + add r10,r10,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#2*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#11*4] + + add r12,r12,r0 + eor r0,r6,r6,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r6,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r9,r9,r2 @ h+=X[i] + str r2,[sp,#2*4] + eor r2,r7,r8 + add r9,r9,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r6 + add r9,r9,r12 @ h+=K256[i] + eor r2,r2,r8 @ Ch(e,f,g) + eor r0,r10,r10,ror#11 + add r9,r9,r2 @ h+=Ch(e,f,g) +#if 18==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 18<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r10,r11 @ a^b, b^c in next round +#else + ldr r2,[sp,#4*4] @ from future BODY_16_xx + eor r12,r10,r11 @ a^b, b^c in next round + ldr r1,[sp,#1*4] @ from future BODY_16_xx +#endif + eor r0,r0,r10,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r5,r5,r9 @ d+=h + eor r3,r3,r11 @ Maj(a,b,c) + add r9,r9,r0,ror#2 @ h+=Sigma0(a) + @ add r9,r9,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#4*4] @ 19 + @ ldr r1,[sp,#1*4] + mov r0,r2,ror#7 + add r9,r9,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#3*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#12*4] + + add r3,r3,r0 + eor r0,r5,r5,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r5,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r8,r8,r2 @ h+=X[i] + str r2,[sp,#3*4] + eor r2,r6,r7 + add r8,r8,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r5 + add r8,r8,r3 @ h+=K256[i] + eor r2,r2,r7 @ Ch(e,f,g) + eor r0,r9,r9,ror#11 + add r8,r8,r2 @ h+=Ch(e,f,g) +#if 19==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 19<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r9,r10 @ a^b, b^c in next round +#else + ldr r2,[sp,#5*4] @ from future BODY_16_xx + eor r3,r9,r10 @ a^b, b^c in next round + ldr r1,[sp,#2*4] @ from future BODY_16_xx +#endif + eor r0,r0,r9,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r4,r4,r8 @ d+=h + eor r12,r12,r10 @ Maj(a,b,c) + add r8,r8,r0,ror#2 @ h+=Sigma0(a) + @ add r8,r8,r12 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#5*4] @ 20 + @ ldr r1,[sp,#2*4] + mov r0,r2,ror#7 + add r8,r8,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#4*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#13*4] + + add r12,r12,r0 + eor r0,r4,r4,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r4,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r7,r7,r2 @ h+=X[i] + str r2,[sp,#4*4] + eor r2,r5,r6 + add r7,r7,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r4 + add r7,r7,r12 @ h+=K256[i] + eor r2,r2,r6 @ Ch(e,f,g) + eor r0,r8,r8,ror#11 + add r7,r7,r2 @ h+=Ch(e,f,g) +#if 20==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 20<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r8,r9 @ a^b, b^c in next round +#else + ldr r2,[sp,#6*4] @ from future BODY_16_xx + eor r12,r8,r9 @ a^b, b^c in next round + ldr r1,[sp,#3*4] @ from future BODY_16_xx +#endif + eor r0,r0,r8,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r11,r11,r7 @ d+=h + eor r3,r3,r9 @ Maj(a,b,c) + add r7,r7,r0,ror#2 @ h+=Sigma0(a) + @ add r7,r7,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#6*4] @ 21 + @ ldr r1,[sp,#3*4] + mov r0,r2,ror#7 + add r7,r7,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#5*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#14*4] + + add r3,r3,r0 + eor r0,r11,r11,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r11,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r6,r6,r2 @ h+=X[i] + str r2,[sp,#5*4] + eor r2,r4,r5 + add r6,r6,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r11 + add r6,r6,r3 @ h+=K256[i] + eor r2,r2,r5 @ Ch(e,f,g) + eor r0,r7,r7,ror#11 + add r6,r6,r2 @ h+=Ch(e,f,g) +#if 21==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 21<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r7,r8 @ a^b, b^c in next round +#else + ldr r2,[sp,#7*4] @ from future BODY_16_xx + eor r3,r7,r8 @ a^b, b^c in next round + ldr r1,[sp,#4*4] @ from future BODY_16_xx +#endif + eor r0,r0,r7,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r10,r10,r6 @ d+=h + eor r12,r12,r8 @ Maj(a,b,c) + add r6,r6,r0,ror#2 @ h+=Sigma0(a) + @ add r6,r6,r12 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#7*4] @ 22 + @ ldr r1,[sp,#4*4] + mov r0,r2,ror#7 + add r6,r6,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#6*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#15*4] + + add r12,r12,r0 + eor r0,r10,r10,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r10,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r5,r5,r2 @ h+=X[i] + str r2,[sp,#6*4] + eor r2,r11,r4 + add r5,r5,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r10 + add r5,r5,r12 @ h+=K256[i] + eor r2,r2,r4 @ Ch(e,f,g) + eor r0,r6,r6,ror#11 + add r5,r5,r2 @ h+=Ch(e,f,g) +#if 22==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 22<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r6,r7 @ a^b, b^c in next round +#else + ldr r2,[sp,#8*4] @ from future BODY_16_xx + eor r12,r6,r7 @ a^b, b^c in next round + ldr r1,[sp,#5*4] @ from future BODY_16_xx +#endif + eor r0,r0,r6,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r9,r9,r5 @ d+=h + eor r3,r3,r7 @ Maj(a,b,c) + add r5,r5,r0,ror#2 @ h+=Sigma0(a) + @ add r5,r5,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#8*4] @ 23 + @ ldr r1,[sp,#5*4] + mov r0,r2,ror#7 + add r5,r5,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#7*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#0*4] + + add r3,r3,r0 + eor r0,r9,r9,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r9,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r4,r4,r2 @ h+=X[i] + str r2,[sp,#7*4] + eor r2,r10,r11 + add r4,r4,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r9 + add r4,r4,r3 @ h+=K256[i] + eor r2,r2,r11 @ Ch(e,f,g) + eor r0,r5,r5,ror#11 + add r4,r4,r2 @ h+=Ch(e,f,g) +#if 23==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 23<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r5,r6 @ a^b, b^c in next round +#else + ldr r2,[sp,#9*4] @ from future BODY_16_xx + eor r3,r5,r6 @ a^b, b^c in next round + ldr r1,[sp,#6*4] @ from future BODY_16_xx +#endif + eor r0,r0,r5,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r8,r8,r4 @ d+=h + eor r12,r12,r6 @ Maj(a,b,c) + add r4,r4,r0,ror#2 @ h+=Sigma0(a) + @ add r4,r4,r12 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#9*4] @ 24 + @ ldr r1,[sp,#6*4] + mov r0,r2,ror#7 + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#8*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#1*4] + + add r12,r12,r0 + eor r0,r8,r8,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r8,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r11,r11,r2 @ h+=X[i] + str r2,[sp,#8*4] + eor r2,r9,r10 + add r11,r11,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r8 + add r11,r11,r12 @ h+=K256[i] + eor r2,r2,r10 @ Ch(e,f,g) + eor r0,r4,r4,ror#11 + add r11,r11,r2 @ h+=Ch(e,f,g) +#if 24==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 24<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r4,r5 @ a^b, b^c in next round +#else + ldr r2,[sp,#10*4] @ from future BODY_16_xx + eor r12,r4,r5 @ a^b, b^c in next round + ldr r1,[sp,#7*4] @ from future BODY_16_xx +#endif + eor r0,r0,r4,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r7,r7,r11 @ d+=h + eor r3,r3,r5 @ Maj(a,b,c) + add r11,r11,r0,ror#2 @ h+=Sigma0(a) + @ add r11,r11,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#10*4] @ 25 + @ ldr r1,[sp,#7*4] + mov r0,r2,ror#7 + add r11,r11,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#9*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#2*4] + + add r3,r3,r0 + eor r0,r7,r7,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r7,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r10,r10,r2 @ h+=X[i] + str r2,[sp,#9*4] + eor r2,r8,r9 + add r10,r10,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r7 + add r10,r10,r3 @ h+=K256[i] + eor r2,r2,r9 @ Ch(e,f,g) + eor r0,r11,r11,ror#11 + add r10,r10,r2 @ h+=Ch(e,f,g) +#if 25==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 25<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r11,r4 @ a^b, b^c in next round +#else + ldr r2,[sp,#11*4] @ from future BODY_16_xx + eor r3,r11,r4 @ a^b, b^c in next round + ldr r1,[sp,#8*4] @ from future BODY_16_xx +#endif + eor r0,r0,r11,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r6,r6,r10 @ d+=h + eor r12,r12,r4 @ Maj(a,b,c) + add r10,r10,r0,ror#2 @ h+=Sigma0(a) + @ add r10,r10,r12 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#11*4] @ 26 + @ ldr r1,[sp,#8*4] + mov r0,r2,ror#7 + add r10,r10,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#10*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#3*4] + + add r12,r12,r0 + eor r0,r6,r6,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r6,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r9,r9,r2 @ h+=X[i] + str r2,[sp,#10*4] + eor r2,r7,r8 + add r9,r9,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r6 + add r9,r9,r12 @ h+=K256[i] + eor r2,r2,r8 @ Ch(e,f,g) + eor r0,r10,r10,ror#11 + add r9,r9,r2 @ h+=Ch(e,f,g) +#if 26==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 26<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r10,r11 @ a^b, b^c in next round +#else + ldr r2,[sp,#12*4] @ from future BODY_16_xx + eor r12,r10,r11 @ a^b, b^c in next round + ldr r1,[sp,#9*4] @ from future BODY_16_xx +#endif + eor r0,r0,r10,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r5,r5,r9 @ d+=h + eor r3,r3,r11 @ Maj(a,b,c) + add r9,r9,r0,ror#2 @ h+=Sigma0(a) + @ add r9,r9,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#12*4] @ 27 + @ ldr r1,[sp,#9*4] + mov r0,r2,ror#7 + add r9,r9,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#11*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#4*4] + + add r3,r3,r0 + eor r0,r5,r5,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r5,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r8,r8,r2 @ h+=X[i] + str r2,[sp,#11*4] + eor r2,r6,r7 + add r8,r8,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r5 + add r8,r8,r3 @ h+=K256[i] + eor r2,r2,r7 @ Ch(e,f,g) + eor r0,r9,r9,ror#11 + add r8,r8,r2 @ h+=Ch(e,f,g) +#if 27==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 27<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r9,r10 @ a^b, b^c in next round +#else + ldr r2,[sp,#13*4] @ from future BODY_16_xx + eor r3,r9,r10 @ a^b, b^c in next round + ldr r1,[sp,#10*4] @ from future BODY_16_xx +#endif + eor r0,r0,r9,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r4,r4,r8 @ d+=h + eor r12,r12,r10 @ Maj(a,b,c) + add r8,r8,r0,ror#2 @ h+=Sigma0(a) + @ add r8,r8,r12 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#13*4] @ 28 + @ ldr r1,[sp,#10*4] + mov r0,r2,ror#7 + add r8,r8,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#12*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#5*4] + + add r12,r12,r0 + eor r0,r4,r4,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r4,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r7,r7,r2 @ h+=X[i] + str r2,[sp,#12*4] + eor r2,r5,r6 + add r7,r7,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r4 + add r7,r7,r12 @ h+=K256[i] + eor r2,r2,r6 @ Ch(e,f,g) + eor r0,r8,r8,ror#11 + add r7,r7,r2 @ h+=Ch(e,f,g) +#if 28==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 28<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r8,r9 @ a^b, b^c in next round +#else + ldr r2,[sp,#14*4] @ from future BODY_16_xx + eor r12,r8,r9 @ a^b, b^c in next round + ldr r1,[sp,#11*4] @ from future BODY_16_xx +#endif + eor r0,r0,r8,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r11,r11,r7 @ d+=h + eor r3,r3,r9 @ Maj(a,b,c) + add r7,r7,r0,ror#2 @ h+=Sigma0(a) + @ add r7,r7,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#14*4] @ 29 + @ ldr r1,[sp,#11*4] + mov r0,r2,ror#7 + add r7,r7,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#13*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#6*4] + + add r3,r3,r0 + eor r0,r11,r11,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r11,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r6,r6,r2 @ h+=X[i] + str r2,[sp,#13*4] + eor r2,r4,r5 + add r6,r6,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r11 + add r6,r6,r3 @ h+=K256[i] + eor r2,r2,r5 @ Ch(e,f,g) + eor r0,r7,r7,ror#11 + add r6,r6,r2 @ h+=Ch(e,f,g) +#if 29==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 29<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r7,r8 @ a^b, b^c in next round +#else + ldr r2,[sp,#15*4] @ from future BODY_16_xx + eor r3,r7,r8 @ a^b, b^c in next round + ldr r1,[sp,#12*4] @ from future BODY_16_xx +#endif + eor r0,r0,r7,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r10,r10,r6 @ d+=h + eor r12,r12,r8 @ Maj(a,b,c) + add r6,r6,r0,ror#2 @ h+=Sigma0(a) + @ add r6,r6,r12 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#15*4] @ 30 + @ ldr r1,[sp,#12*4] + mov r0,r2,ror#7 + add r6,r6,r12 @ h+=Maj(a,b,c) from the past + mov r12,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r12,r12,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#14*4] + eor r12,r12,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#7*4] + + add r12,r12,r0 + eor r0,r10,r10,ror#5 @ from BODY_00_15 + add r2,r2,r12 + eor r0,r0,r10,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r12,[r14],#4 @ *K256++ + add r5,r5,r2 @ h+=X[i] + str r2,[sp,#14*4] + eor r2,r11,r4 + add r5,r5,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r10 + add r5,r5,r12 @ h+=K256[i] + eor r2,r2,r4 @ Ch(e,f,g) + eor r0,r6,r6,ror#11 + add r5,r5,r2 @ h+=Ch(e,f,g) +#if 30==31 + and r12,r12,#0xff + cmp r12,#0xf2 @ done? +#endif +#if 30<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r12,r6,r7 @ a^b, b^c in next round +#else + ldr r2,[sp,#0*4] @ from future BODY_16_xx + eor r12,r6,r7 @ a^b, b^c in next round + ldr r1,[sp,#13*4] @ from future BODY_16_xx +#endif + eor r0,r0,r6,ror#20 @ Sigma0(a) + and r3,r3,r12 @ (b^c)&=(a^b) + add r9,r9,r5 @ d+=h + eor r3,r3,r7 @ Maj(a,b,c) + add r5,r5,r0,ror#2 @ h+=Sigma0(a) + @ add r5,r5,r3 @ h+=Maj(a,b,c) + @ ldr r2,[sp,#0*4] @ 31 + @ ldr r1,[sp,#13*4] + mov r0,r2,ror#7 + add r5,r5,r3 @ h+=Maj(a,b,c) from the past + mov r3,r1,ror#17 + eor r0,r0,r2,ror#18 + eor r3,r3,r1,ror#19 + eor r0,r0,r2,lsr#3 @ sigma0(X[i+1]) + ldr r2,[sp,#15*4] + eor r3,r3,r1,lsr#10 @ sigma1(X[i+14]) + ldr r1,[sp,#8*4] + + add r3,r3,r0 + eor r0,r9,r9,ror#5 @ from BODY_00_15 + add r2,r2,r3 + eor r0,r0,r9,ror#19 @ Sigma1(e) + add r2,r2,r1 @ X[i] + ldr r3,[r14],#4 @ *K256++ + add r4,r4,r2 @ h+=X[i] + str r2,[sp,#15*4] + eor r2,r10,r11 + add r4,r4,r0,ror#6 @ h+=Sigma1(e) + and r2,r2,r9 + add r4,r4,r3 @ h+=K256[i] + eor r2,r2,r11 @ Ch(e,f,g) + eor r0,r5,r5,ror#11 + add r4,r4,r2 @ h+=Ch(e,f,g) +#if 31==31 + and r3,r3,#0xff + cmp r3,#0xf2 @ done? +#endif +#if 31<15 +# if __ARM_ARCH__>=7 + ldr r2,[r1],#4 @ prefetch +# else + ldrb r2,[r1,#3] +# endif + eor r3,r5,r6 @ a^b, b^c in next round +#else + ldr r2,[sp,#1*4] @ from future BODY_16_xx + eor r3,r5,r6 @ a^b, b^c in next round + ldr r1,[sp,#14*4] @ from future BODY_16_xx +#endif + eor r0,r0,r5,ror#20 @ Sigma0(a) + and r12,r12,r3 @ (b^c)&=(a^b) + add r8,r8,r4 @ d+=h + eor r12,r12,r6 @ Maj(a,b,c) + add r4,r4,r0,ror#2 @ h+=Sigma0(a) + @ add r4,r4,r12 @ h+=Maj(a,b,c) +#if __ARM_ARCH__>=7 + ite eq @ Thumb2 thing, sanity check in ARM +#endif + ldreq r3,[sp,#16*4] @ pull ctx + bne .Lrounds_16_xx + + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + ldr r0,[r3,#0] + ldr r2,[r3,#4] + ldr r12,[r3,#8] + add r4,r4,r0 + ldr r0,[r3,#12] + add r5,r5,r2 + ldr r2,[r3,#16] + add r6,r6,r12 + ldr r12,[r3,#20] + add r7,r7,r0 + ldr r0,[r3,#24] + add r8,r8,r2 + ldr r2,[r3,#28] + add r9,r9,r12 + ldr r1,[sp,#17*4] @ pull inp + ldr r12,[sp,#18*4] @ pull inp+len + add r10,r10,r0 + add r11,r11,r2 + stmia r3,{r4,r5,r6,r7,r8,r9,r10,r11} + cmp r1,r12 + sub r14,r14,#256 @ rewind Ktbl + bne .Loop + + add sp,sp,#19*4 @ destroy frame +#if __ARM_ARCH__>=5 + ldmia sp!,{r4-r11,pc} +#else + ldmia sp!,{r4-r11,lr} + tst lr,#1 + moveq pc,lr @ be binary compatible with V4, yet + .word 0xe12fff1e @ interoperable with Thumb ISA:-) +#endif +.size sha256_block_data_order,.-sha256_block_data_order +#if __ARM_MAX_ARCH__>=7 +.arch armv7-a +.fpu neon + +.global sha256_block_data_order_neon +.type sha256_block_data_order_neon,%function +.align 4 +sha256_block_data_order_neon: +.LNEON: + stmdb sp!,{r4-r12,lr} + + sub r11,sp,#16*4+16 + adrl r14,K256 + bic r11,r11,#15 @ align for 128-bit stores + mov r12,sp + mov sp,r11 @ alloca + add r2,r1,r2,lsl#6 @ len to point at the end of inp + + vld1.8 {q0},[r1]! + vld1.8 {q1},[r1]! + vld1.8 {q2},[r1]! + vld1.8 {q3},[r1]! + vld1.32 {q8},[r14,:128]! + vld1.32 {q9},[r14,:128]! + vld1.32 {q10},[r14,:128]! + vld1.32 {q11},[r14,:128]! + vrev32.8 q0,q0 @ yes, even on + str r0,[sp,#64] + vrev32.8 q1,q1 @ big-endian + str r1,[sp,#68] + mov r1,sp + vrev32.8 q2,q2 + str r2,[sp,#72] + vrev32.8 q3,q3 + str r12,[sp,#76] @ save original sp + vadd.i32 q8,q8,q0 + vadd.i32 q9,q9,q1 + vst1.32 {q8},[r1,:128]! + vadd.i32 q10,q10,q2 + vst1.32 {q9},[r1,:128]! + vadd.i32 q11,q11,q3 + vst1.32 {q10},[r1,:128]! + vst1.32 {q11},[r1,:128]! + + ldmia r0,{r4-r11} + sub r1,r1,#64 + ldr r2,[sp,#0] + eor r12,r12,r12 + eor r3,r5,r6 + b .L_00_48 + +.align 4 +.L_00_48: + vext.8 q8,q0,q1,#4 + add r11,r11,r2 + eor r2,r9,r10 + eor r0,r8,r8,ror#5 + vext.8 q9,q2,q3,#4 + add r4,r4,r12 + and r2,r2,r8 + eor r12,r0,r8,ror#19 + vshr.u32 q10,q8,#7 + eor r0,r4,r4,ror#11 + eor r2,r2,r10 + vadd.i32 q0,q0,q9 + add r11,r11,r12,ror#6 + eor r12,r4,r5 + vshr.u32 q9,q8,#3 + eor r0,r0,r4,ror#20 + add r11,r11,r2 + vsli.32 q10,q8,#25 + ldr r2,[sp,#4] + and r3,r3,r12 + vshr.u32 q11,q8,#18 + add r7,r7,r11 + add r11,r11,r0,ror#2 + eor r3,r3,r5 + veor q9,q9,q10 + add r10,r10,r2 + vsli.32 q11,q8,#14 + eor r2,r8,r9 + eor r0,r7,r7,ror#5 + vshr.u32 d24,d7,#17 + add r11,r11,r3 + and r2,r2,r7 + veor q9,q9,q11 + eor r3,r0,r7,ror#19 + eor r0,r11,r11,ror#11 + vsli.32 d24,d7,#15 + eor r2,r2,r9 + add r10,r10,r3,ror#6 + vshr.u32 d25,d7,#10 + eor r3,r11,r4 + eor r0,r0,r11,ror#20 + vadd.i32 q0,q0,q9 + add r10,r10,r2 + ldr r2,[sp,#8] + veor d25,d25,d24 + and r12,r12,r3 + add r6,r6,r10 + vshr.u32 d24,d7,#19 + add r10,r10,r0,ror#2 + eor r12,r12,r4 + vsli.32 d24,d7,#13 + add r9,r9,r2 + eor r2,r7,r8 + veor d25,d25,d24 + eor r0,r6,r6,ror#5 + add r10,r10,r12 + vadd.i32 d0,d0,d25 + and r2,r2,r6 + eor r12,r0,r6,ror#19 + vshr.u32 d24,d0,#17 + eor r0,r10,r10,ror#11 + eor r2,r2,r8 + vsli.32 d24,d0,#15 + add r9,r9,r12,ror#6 + eor r12,r10,r11 + vshr.u32 d25,d0,#10 + eor r0,r0,r10,ror#20 + add r9,r9,r2 + veor d25,d25,d24 + ldr r2,[sp,#12] + and r3,r3,r12 + vshr.u32 d24,d0,#19 + add r5,r5,r9 + add r9,r9,r0,ror#2 + eor r3,r3,r11 + vld1.32 {q8},[r14,:128]! + add r8,r8,r2 + vsli.32 d24,d0,#13 + eor r2,r6,r7 + eor r0,r5,r5,ror#5 + veor d25,d25,d24 + add r9,r9,r3 + and r2,r2,r5 + vadd.i32 d1,d1,d25 + eor r3,r0,r5,ror#19 + eor r0,r9,r9,ror#11 + vadd.i32 q8,q8,q0 + eor r2,r2,r7 + add r8,r8,r3,ror#6 + eor r3,r9,r10 + eor r0,r0,r9,ror#20 + add r8,r8,r2 + ldr r2,[sp,#16] + and r12,r12,r3 + add r4,r4,r8 + vst1.32 {q8},[r1,:128]! + add r8,r8,r0,ror#2 + eor r12,r12,r10 + vext.8 q8,q1,q2,#4 + add r7,r7,r2 + eor r2,r5,r6 + eor r0,r4,r4,ror#5 + vext.8 q9,q3,q0,#4 + add r8,r8,r12 + and r2,r2,r4 + eor r12,r0,r4,ror#19 + vshr.u32 q10,q8,#7 + eor r0,r8,r8,ror#11 + eor r2,r2,r6 + vadd.i32 q1,q1,q9 + add r7,r7,r12,ror#6 + eor r12,r8,r9 + vshr.u32 q9,q8,#3 + eor r0,r0,r8,ror#20 + add r7,r7,r2 + vsli.32 q10,q8,#25 + ldr r2,[sp,#20] + and r3,r3,r12 + vshr.u32 q11,q8,#18 + add r11,r11,r7 + add r7,r7,r0,ror#2 + eor r3,r3,r9 + veor q9,q9,q10 + add r6,r6,r2 + vsli.32 q11,q8,#14 + eor r2,r4,r5 + eor r0,r11,r11,ror#5 + vshr.u32 d24,d1,#17 + add r7,r7,r3 + and r2,r2,r11 + veor q9,q9,q11 + eor r3,r0,r11,ror#19 + eor r0,r7,r7,ror#11 + vsli.32 d24,d1,#15 + eor r2,r2,r5 + add r6,r6,r3,ror#6 + vshr.u32 d25,d1,#10 + eor r3,r7,r8 + eor r0,r0,r7,ror#20 + vadd.i32 q1,q1,q9 + add r6,r6,r2 + ldr r2,[sp,#24] + veor d25,d25,d24 + and r12,r12,r3 + add r10,r10,r6 + vshr.u32 d24,d1,#19 + add r6,r6,r0,ror#2 + eor r12,r12,r8 + vsli.32 d24,d1,#13 + add r5,r5,r2 + eor r2,r11,r4 + veor d25,d25,d24 + eor r0,r10,r10,ror#5 + add r6,r6,r12 + vadd.i32 d2,d2,d25 + and r2,r2,r10 + eor r12,r0,r10,ror#19 + vshr.u32 d24,d2,#17 + eor r0,r6,r6,ror#11 + eor r2,r2,r4 + vsli.32 d24,d2,#15 + add r5,r5,r12,ror#6 + eor r12,r6,r7 + vshr.u32 d25,d2,#10 + eor r0,r0,r6,ror#20 + add r5,r5,r2 + veor d25,d25,d24 + ldr r2,[sp,#28] + and r3,r3,r12 + vshr.u32 d24,d2,#19 + add r9,r9,r5 + add r5,r5,r0,ror#2 + eor r3,r3,r7 + vld1.32 {q8},[r14,:128]! + add r4,r4,r2 + vsli.32 d24,d2,#13 + eor r2,r10,r11 + eor r0,r9,r9,ror#5 + veor d25,d25,d24 + add r5,r5,r3 + and r2,r2,r9 + vadd.i32 d3,d3,d25 + eor r3,r0,r9,ror#19 + eor r0,r5,r5,ror#11 + vadd.i32 q8,q8,q1 + eor r2,r2,r11 + add r4,r4,r3,ror#6 + eor r3,r5,r6 + eor r0,r0,r5,ror#20 + add r4,r4,r2 + ldr r2,[sp,#32] + and r12,r12,r3 + add r8,r8,r4 + vst1.32 {q8},[r1,:128]! + add r4,r4,r0,ror#2 + eor r12,r12,r6 + vext.8 q8,q2,q3,#4 + add r11,r11,r2 + eor r2,r9,r10 + eor r0,r8,r8,ror#5 + vext.8 q9,q0,q1,#4 + add r4,r4,r12 + and r2,r2,r8 + eor r12,r0,r8,ror#19 + vshr.u32 q10,q8,#7 + eor r0,r4,r4,ror#11 + eor r2,r2,r10 + vadd.i32 q2,q2,q9 + add r11,r11,r12,ror#6 + eor r12,r4,r5 + vshr.u32 q9,q8,#3 + eor r0,r0,r4,ror#20 + add r11,r11,r2 + vsli.32 q10,q8,#25 + ldr r2,[sp,#36] + and r3,r3,r12 + vshr.u32 q11,q8,#18 + add r7,r7,r11 + add r11,r11,r0,ror#2 + eor r3,r3,r5 + veor q9,q9,q10 + add r10,r10,r2 + vsli.32 q11,q8,#14 + eor r2,r8,r9 + eor r0,r7,r7,ror#5 + vshr.u32 d24,d3,#17 + add r11,r11,r3 + and r2,r2,r7 + veor q9,q9,q11 + eor r3,r0,r7,ror#19 + eor r0,r11,r11,ror#11 + vsli.32 d24,d3,#15 + eor r2,r2,r9 + add r10,r10,r3,ror#6 + vshr.u32 d25,d3,#10 + eor r3,r11,r4 + eor r0,r0,r11,ror#20 + vadd.i32 q2,q2,q9 + add r10,r10,r2 + ldr r2,[sp,#40] + veor d25,d25,d24 + and r12,r12,r3 + add r6,r6,r10 + vshr.u32 d24,d3,#19 + add r10,r10,r0,ror#2 + eor r12,r12,r4 + vsli.32 d24,d3,#13 + add r9,r9,r2 + eor r2,r7,r8 + veor d25,d25,d24 + eor r0,r6,r6,ror#5 + add r10,r10,r12 + vadd.i32 d4,d4,d25 + and r2,r2,r6 + eor r12,r0,r6,ror#19 + vshr.u32 d24,d4,#17 + eor r0,r10,r10,ror#11 + eor r2,r2,r8 + vsli.32 d24,d4,#15 + add r9,r9,r12,ror#6 + eor r12,r10,r11 + vshr.u32 d25,d4,#10 + eor r0,r0,r10,ror#20 + add r9,r9,r2 + veor d25,d25,d24 + ldr r2,[sp,#44] + and r3,r3,r12 + vshr.u32 d24,d4,#19 + add r5,r5,r9 + add r9,r9,r0,ror#2 + eor r3,r3,r11 + vld1.32 {q8},[r14,:128]! + add r8,r8,r2 + vsli.32 d24,d4,#13 + eor r2,r6,r7 + eor r0,r5,r5,ror#5 + veor d25,d25,d24 + add r9,r9,r3 + and r2,r2,r5 + vadd.i32 d5,d5,d25 + eor r3,r0,r5,ror#19 + eor r0,r9,r9,ror#11 + vadd.i32 q8,q8,q2 + eor r2,r2,r7 + add r8,r8,r3,ror#6 + eor r3,r9,r10 + eor r0,r0,r9,ror#20 + add r8,r8,r2 + ldr r2,[sp,#48] + and r12,r12,r3 + add r4,r4,r8 + vst1.32 {q8},[r1,:128]! + add r8,r8,r0,ror#2 + eor r12,r12,r10 + vext.8 q8,q3,q0,#4 + add r7,r7,r2 + eor r2,r5,r6 + eor r0,r4,r4,ror#5 + vext.8 q9,q1,q2,#4 + add r8,r8,r12 + and r2,r2,r4 + eor r12,r0,r4,ror#19 + vshr.u32 q10,q8,#7 + eor r0,r8,r8,ror#11 + eor r2,r2,r6 + vadd.i32 q3,q3,q9 + add r7,r7,r12,ror#6 + eor r12,r8,r9 + vshr.u32 q9,q8,#3 + eor r0,r0,r8,ror#20 + add r7,r7,r2 + vsli.32 q10,q8,#25 + ldr r2,[sp,#52] + and r3,r3,r12 + vshr.u32 q11,q8,#18 + add r11,r11,r7 + add r7,r7,r0,ror#2 + eor r3,r3,r9 + veor q9,q9,q10 + add r6,r6,r2 + vsli.32 q11,q8,#14 + eor r2,r4,r5 + eor r0,r11,r11,ror#5 + vshr.u32 d24,d5,#17 + add r7,r7,r3 + and r2,r2,r11 + veor q9,q9,q11 + eor r3,r0,r11,ror#19 + eor r0,r7,r7,ror#11 + vsli.32 d24,d5,#15 + eor r2,r2,r5 + add r6,r6,r3,ror#6 + vshr.u32 d25,d5,#10 + eor r3,r7,r8 + eor r0,r0,r7,ror#20 + vadd.i32 q3,q3,q9 + add r6,r6,r2 + ldr r2,[sp,#56] + veor d25,d25,d24 + and r12,r12,r3 + add r10,r10,r6 + vshr.u32 d24,d5,#19 + add r6,r6,r0,ror#2 + eor r12,r12,r8 + vsli.32 d24,d5,#13 + add r5,r5,r2 + eor r2,r11,r4 + veor d25,d25,d24 + eor r0,r10,r10,ror#5 + add r6,r6,r12 + vadd.i32 d6,d6,d25 + and r2,r2,r10 + eor r12,r0,r10,ror#19 + vshr.u32 d24,d6,#17 + eor r0,r6,r6,ror#11 + eor r2,r2,r4 + vsli.32 d24,d6,#15 + add r5,r5,r12,ror#6 + eor r12,r6,r7 + vshr.u32 d25,d6,#10 + eor r0,r0,r6,ror#20 + add r5,r5,r2 + veor d25,d25,d24 + ldr r2,[sp,#60] + and r3,r3,r12 + vshr.u32 d24,d6,#19 + add r9,r9,r5 + add r5,r5,r0,ror#2 + eor r3,r3,r7 + vld1.32 {q8},[r14,:128]! + add r4,r4,r2 + vsli.32 d24,d6,#13 + eor r2,r10,r11 + eor r0,r9,r9,ror#5 + veor d25,d25,d24 + add r5,r5,r3 + and r2,r2,r9 + vadd.i32 d7,d7,d25 + eor r3,r0,r9,ror#19 + eor r0,r5,r5,ror#11 + vadd.i32 q8,q8,q3 + eor r2,r2,r11 + add r4,r4,r3,ror#6 + eor r3,r5,r6 + eor r0,r0,r5,ror#20 + add r4,r4,r2 + ldr r2,[r14] + and r12,r12,r3 + add r8,r8,r4 + vst1.32 {q8},[r1,:128]! + add r4,r4,r0,ror#2 + eor r12,r12,r6 + teq r2,#0 @ check for K256 terminator + ldr r2,[sp,#0] + sub r1,r1,#64 + bne .L_00_48 + + ldr r1,[sp,#68] + ldr r0,[sp,#72] + sub r14,r14,#256 @ rewind r14 + teq r1,r0 + it eq + subeq r1,r1,#64 @ avoid SEGV + vld1.8 {q0},[r1]! @ load next input block + vld1.8 {q1},[r1]! + vld1.8 {q2},[r1]! + vld1.8 {q3},[r1]! + it ne + strne r1,[sp,#68] + mov r1,sp + add r11,r11,r2 + eor r2,r9,r10 + eor r0,r8,r8,ror#5 + add r4,r4,r12 + vld1.32 {q8},[r14,:128]! + and r2,r2,r8 + eor r12,r0,r8,ror#19 + eor r0,r4,r4,ror#11 + eor r2,r2,r10 + vrev32.8 q0,q0 + add r11,r11,r12,ror#6 + eor r12,r4,r5 + eor r0,r0,r4,ror#20 + add r11,r11,r2 + vadd.i32 q8,q8,q0 + ldr r2,[sp,#4] + and r3,r3,r12 + add r7,r7,r11 + add r11,r11,r0,ror#2 + eor r3,r3,r5 + add r10,r10,r2 + eor r2,r8,r9 + eor r0,r7,r7,ror#5 + add r11,r11,r3 + and r2,r2,r7 + eor r3,r0,r7,ror#19 + eor r0,r11,r11,ror#11 + eor r2,r2,r9 + add r10,r10,r3,ror#6 + eor r3,r11,r4 + eor r0,r0,r11,ror#20 + add r10,r10,r2 + ldr r2,[sp,#8] + and r12,r12,r3 + add r6,r6,r10 + add r10,r10,r0,ror#2 + eor r12,r12,r4 + add r9,r9,r2 + eor r2,r7,r8 + eor r0,r6,r6,ror#5 + add r10,r10,r12 + and r2,r2,r6 + eor r12,r0,r6,ror#19 + eor r0,r10,r10,ror#11 + eor r2,r2,r8 + add r9,r9,r12,ror#6 + eor r12,r10,r11 + eor r0,r0,r10,ror#20 + add r9,r9,r2 + ldr r2,[sp,#12] + and r3,r3,r12 + add r5,r5,r9 + add r9,r9,r0,ror#2 + eor r3,r3,r11 + add r8,r8,r2 + eor r2,r6,r7 + eor r0,r5,r5,ror#5 + add r9,r9,r3 + and r2,r2,r5 + eor r3,r0,r5,ror#19 + eor r0,r9,r9,ror#11 + eor r2,r2,r7 + add r8,r8,r3,ror#6 + eor r3,r9,r10 + eor r0,r0,r9,ror#20 + add r8,r8,r2 + ldr r2,[sp,#16] + and r12,r12,r3 + add r4,r4,r8 + add r8,r8,r0,ror#2 + eor r12,r12,r10 + vst1.32 {q8},[r1,:128]! + add r7,r7,r2 + eor r2,r5,r6 + eor r0,r4,r4,ror#5 + add r8,r8,r12 + vld1.32 {q8},[r14,:128]! + and r2,r2,r4 + eor r12,r0,r4,ror#19 + eor r0,r8,r8,ror#11 + eor r2,r2,r6 + vrev32.8 q1,q1 + add r7,r7,r12,ror#6 + eor r12,r8,r9 + eor r0,r0,r8,ror#20 + add r7,r7,r2 + vadd.i32 q8,q8,q1 + ldr r2,[sp,#20] + and r3,r3,r12 + add r11,r11,r7 + add r7,r7,r0,ror#2 + eor r3,r3,r9 + add r6,r6,r2 + eor r2,r4,r5 + eor r0,r11,r11,ror#5 + add r7,r7,r3 + and r2,r2,r11 + eor r3,r0,r11,ror#19 + eor r0,r7,r7,ror#11 + eor r2,r2,r5 + add r6,r6,r3,ror#6 + eor r3,r7,r8 + eor r0,r0,r7,ror#20 + add r6,r6,r2 + ldr r2,[sp,#24] + and r12,r12,r3 + add r10,r10,r6 + add r6,r6,r0,ror#2 + eor r12,r12,r8 + add r5,r5,r2 + eor r2,r11,r4 + eor r0,r10,r10,ror#5 + add r6,r6,r12 + and r2,r2,r10 + eor r12,r0,r10,ror#19 + eor r0,r6,r6,ror#11 + eor r2,r2,r4 + add r5,r5,r12,ror#6 + eor r12,r6,r7 + eor r0,r0,r6,ror#20 + add r5,r5,r2 + ldr r2,[sp,#28] + and r3,r3,r12 + add r9,r9,r5 + add r5,r5,r0,ror#2 + eor r3,r3,r7 + add r4,r4,r2 + eor r2,r10,r11 + eor r0,r9,r9,ror#5 + add r5,r5,r3 + and r2,r2,r9 + eor r3,r0,r9,ror#19 + eor r0,r5,r5,ror#11 + eor r2,r2,r11 + add r4,r4,r3,ror#6 + eor r3,r5,r6 + eor r0,r0,r5,ror#20 + add r4,r4,r2 + ldr r2,[sp,#32] + and r12,r12,r3 + add r8,r8,r4 + add r4,r4,r0,ror#2 + eor r12,r12,r6 + vst1.32 {q8},[r1,:128]! + add r11,r11,r2 + eor r2,r9,r10 + eor r0,r8,r8,ror#5 + add r4,r4,r12 + vld1.32 {q8},[r14,:128]! + and r2,r2,r8 + eor r12,r0,r8,ror#19 + eor r0,r4,r4,ror#11 + eor r2,r2,r10 + vrev32.8 q2,q2 + add r11,r11,r12,ror#6 + eor r12,r4,r5 + eor r0,r0,r4,ror#20 + add r11,r11,r2 + vadd.i32 q8,q8,q2 + ldr r2,[sp,#36] + and r3,r3,r12 + add r7,r7,r11 + add r11,r11,r0,ror#2 + eor r3,r3,r5 + add r10,r10,r2 + eor r2,r8,r9 + eor r0,r7,r7,ror#5 + add r11,r11,r3 + and r2,r2,r7 + eor r3,r0,r7,ror#19 + eor r0,r11,r11,ror#11 + eor r2,r2,r9 + add r10,r10,r3,ror#6 + eor r3,r11,r4 + eor r0,r0,r11,ror#20 + add r10,r10,r2 + ldr r2,[sp,#40] + and r12,r12,r3 + add r6,r6,r10 + add r10,r10,r0,ror#2 + eor r12,r12,r4 + add r9,r9,r2 + eor r2,r7,r8 + eor r0,r6,r6,ror#5 + add r10,r10,r12 + and r2,r2,r6 + eor r12,r0,r6,ror#19 + eor r0,r10,r10,ror#11 + eor r2,r2,r8 + add r9,r9,r12,ror#6 + eor r12,r10,r11 + eor r0,r0,r10,ror#20 + add r9,r9,r2 + ldr r2,[sp,#44] + and r3,r3,r12 + add r5,r5,r9 + add r9,r9,r0,ror#2 + eor r3,r3,r11 + add r8,r8,r2 + eor r2,r6,r7 + eor r0,r5,r5,ror#5 + add r9,r9,r3 + and r2,r2,r5 + eor r3,r0,r5,ror#19 + eor r0,r9,r9,ror#11 + eor r2,r2,r7 + add r8,r8,r3,ror#6 + eor r3,r9,r10 + eor r0,r0,r9,ror#20 + add r8,r8,r2 + ldr r2,[sp,#48] + and r12,r12,r3 + add r4,r4,r8 + add r8,r8,r0,ror#2 + eor r12,r12,r10 + vst1.32 {q8},[r1,:128]! + add r7,r7,r2 + eor r2,r5,r6 + eor r0,r4,r4,ror#5 + add r8,r8,r12 + vld1.32 {q8},[r14,:128]! + and r2,r2,r4 + eor r12,r0,r4,ror#19 + eor r0,r8,r8,ror#11 + eor r2,r2,r6 + vrev32.8 q3,q3 + add r7,r7,r12,ror#6 + eor r12,r8,r9 + eor r0,r0,r8,ror#20 + add r7,r7,r2 + vadd.i32 q8,q8,q3 + ldr r2,[sp,#52] + and r3,r3,r12 + add r11,r11,r7 + add r7,r7,r0,ror#2 + eor r3,r3,r9 + add r6,r6,r2 + eor r2,r4,r5 + eor r0,r11,r11,ror#5 + add r7,r7,r3 + and r2,r2,r11 + eor r3,r0,r11,ror#19 + eor r0,r7,r7,ror#11 + eor r2,r2,r5 + add r6,r6,r3,ror#6 + eor r3,r7,r8 + eor r0,r0,r7,ror#20 + add r6,r6,r2 + ldr r2,[sp,#56] + and r12,r12,r3 + add r10,r10,r6 + add r6,r6,r0,ror#2 + eor r12,r12,r8 + add r5,r5,r2 + eor r2,r11,r4 + eor r0,r10,r10,ror#5 + add r6,r6,r12 + and r2,r2,r10 + eor r12,r0,r10,ror#19 + eor r0,r6,r6,ror#11 + eor r2,r2,r4 + add r5,r5,r12,ror#6 + eor r12,r6,r7 + eor r0,r0,r6,ror#20 + add r5,r5,r2 + ldr r2,[sp,#60] + and r3,r3,r12 + add r9,r9,r5 + add r5,r5,r0,ror#2 + eor r3,r3,r7 + add r4,r4,r2 + eor r2,r10,r11 + eor r0,r9,r9,ror#5 + add r5,r5,r3 + and r2,r2,r9 + eor r3,r0,r9,ror#19 + eor r0,r5,r5,ror#11 + eor r2,r2,r11 + add r4,r4,r3,ror#6 + eor r3,r5,r6 + eor r0,r0,r5,ror#20 + add r4,r4,r2 + ldr r2,[sp,#64] + and r12,r12,r3 + add r8,r8,r4 + add r4,r4,r0,ror#2 + eor r12,r12,r6 + vst1.32 {q8},[r1,:128]! + ldr r0,[r2,#0] + add r4,r4,r12 @ h+=Maj(a,b,c) from the past + ldr r12,[r2,#4] + ldr r3,[r2,#8] + ldr r1,[r2,#12] + add r4,r4,r0 @ accumulate + ldr r0,[r2,#16] + add r5,r5,r12 + ldr r12,[r2,#20] + add r6,r6,r3 + ldr r3,[r2,#24] + add r7,r7,r1 + ldr r1,[r2,#28] + add r8,r8,r0 + str r4,[r2],#4 + add r9,r9,r12 + str r5,[r2],#4 + add r10,r10,r3 + str r6,[r2],#4 + add r11,r11,r1 + str r7,[r2],#4 + stmia r2,{r8-r11} + + ittte ne + movne r1,sp + ldrne r2,[sp,#0] + eorne r12,r12,r12 + ldreq sp,[sp,#76] @ restore original sp + itt ne + eorne r3,r5,r6 + bne .L_00_48 + + ldmia sp!,{r4-r12,pc} +.size sha256_block_data_order_neon,.-sha256_block_data_order_neon +#endif +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) + +# ifdef __thumb2__ +# define INST(a,b,c,d) .byte c,d|0xc,a,b +# else +# define INST(a,b,c,d) .byte a,b,c,d +# endif + +.type sha256_block_data_order_armv8,%function +.align 5 +sha256_block_data_order_armv8: +.LARMv8: + vld1.32 {q0,q1},[r0] +# ifdef __thumb2__ + adr r3,.LARMv8 + sub r3,r3,#.LARMv8-K256 +# else + adrl r3,K256 +# endif + add r2,r1,r2,lsl#6 @ len to point at the end of inp + +.Loop_v8: + vld1.8 {q8-q9},[r1]! + vld1.8 {q10-q11},[r1]! + vld1.32 {q12},[r3]! + vrev32.8 q8,q8 + vrev32.8 q9,q9 + vrev32.8 q10,q10 + vrev32.8 q11,q11 + vmov q14,q0 @ offload + vmov q15,q1 + teq r1,r2 + vld1.32 {q13},[r3]! + vadd.i32 q12,q12,q8 + INST(0xe2,0x03,0xfa,0xf3) @ sha256su0 q8,q9 + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + INST(0xe6,0x0c,0x64,0xf3) @ sha256su1 q8,q10,q11 + vld1.32 {q12},[r3]! + vadd.i32 q13,q13,q9 + INST(0xe4,0x23,0xfa,0xf3) @ sha256su0 q9,q10 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + INST(0xe0,0x2c,0x66,0xf3) @ sha256su1 q9,q11,q8 + vld1.32 {q13},[r3]! + vadd.i32 q12,q12,q10 + INST(0xe6,0x43,0xfa,0xf3) @ sha256su0 q10,q11 + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + INST(0xe2,0x4c,0x60,0xf3) @ sha256su1 q10,q8,q9 + vld1.32 {q12},[r3]! + vadd.i32 q13,q13,q11 + INST(0xe0,0x63,0xfa,0xf3) @ sha256su0 q11,q8 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + INST(0xe4,0x6c,0x62,0xf3) @ sha256su1 q11,q9,q10 + vld1.32 {q13},[r3]! + vadd.i32 q12,q12,q8 + INST(0xe2,0x03,0xfa,0xf3) @ sha256su0 q8,q9 + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + INST(0xe6,0x0c,0x64,0xf3) @ sha256su1 q8,q10,q11 + vld1.32 {q12},[r3]! + vadd.i32 q13,q13,q9 + INST(0xe4,0x23,0xfa,0xf3) @ sha256su0 q9,q10 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + INST(0xe0,0x2c,0x66,0xf3) @ sha256su1 q9,q11,q8 + vld1.32 {q13},[r3]! + vadd.i32 q12,q12,q10 + INST(0xe6,0x43,0xfa,0xf3) @ sha256su0 q10,q11 + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + INST(0xe2,0x4c,0x60,0xf3) @ sha256su1 q10,q8,q9 + vld1.32 {q12},[r3]! + vadd.i32 q13,q13,q11 + INST(0xe0,0x63,0xfa,0xf3) @ sha256su0 q11,q8 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + INST(0xe4,0x6c,0x62,0xf3) @ sha256su1 q11,q9,q10 + vld1.32 {q13},[r3]! + vadd.i32 q12,q12,q8 + INST(0xe2,0x03,0xfa,0xf3) @ sha256su0 q8,q9 + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + INST(0xe6,0x0c,0x64,0xf3) @ sha256su1 q8,q10,q11 + vld1.32 {q12},[r3]! + vadd.i32 q13,q13,q9 + INST(0xe4,0x23,0xfa,0xf3) @ sha256su0 q9,q10 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + INST(0xe0,0x2c,0x66,0xf3) @ sha256su1 q9,q11,q8 + vld1.32 {q13},[r3]! + vadd.i32 q12,q12,q10 + INST(0xe6,0x43,0xfa,0xf3) @ sha256su0 q10,q11 + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + INST(0xe2,0x4c,0x60,0xf3) @ sha256su1 q10,q8,q9 + vld1.32 {q12},[r3]! + vadd.i32 q13,q13,q11 + INST(0xe0,0x63,0xfa,0xf3) @ sha256su0 q11,q8 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + INST(0xe4,0x6c,0x62,0xf3) @ sha256su1 q11,q9,q10 + vld1.32 {q13},[r3]! + vadd.i32 q12,q12,q8 + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + + vld1.32 {q12},[r3]! + vadd.i32 q13,q13,q9 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + + vld1.32 {q13},[r3] + vadd.i32 q12,q12,q10 + sub r3,r3,#256-16 @ rewind + vmov q2,q0 + INST(0x68,0x0c,0x02,0xf3) @ sha256h q0,q1,q12 + INST(0x68,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q12 + + vadd.i32 q13,q13,q11 + vmov q2,q0 + INST(0x6a,0x0c,0x02,0xf3) @ sha256h q0,q1,q13 + INST(0x6a,0x2c,0x14,0xf3) @ sha256h2 q1,q2,q13 + + vadd.i32 q0,q0,q14 + vadd.i32 q1,q1,q15 + it ne + bne .Loop_v8 + + vst1.32 {q0,q1},[r0] + + bx lr @ bx lr +.size sha256_block_data_order_armv8,.-sha256_block_data_order_armv8 +#endif +.asciz "SHA256 block transform for ARMv4/NEON/ARMv8, CRYPTOGAMS by <appro@openssl.org>" +.align 2 +#if __ARM_MAX_ARCH__>=7 && !defined(__KERNEL__) +.comm OPENSSL_armcap_P,4,4 +#endif diff --git a/arch/arm/crypto/sha256_glue.c b/arch/arm/crypto/sha256_glue.c new file mode 100644 index 000000000000..bb03482e53e2 --- /dev/null +++ b/arch/arm/crypto/sha256_glue.c @@ -0,0 +1,246 @@ +/* + * Glue code for the SHA256 Secure Hash Algorithm assembly implementation + * using optimized ARM assembler and NEON instructions. + * + * Copyright © 2015 Google Inc. + * + * This file is based on sha256_ssse3_glue.c: + * Copyright (C) 2013 Intel Corporation + * Author: Tim Chen <tim.c.chen@linux.intel.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 <crypto/internal/hash.h> +#include <linux/crypto.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <linux/string.h> +#include <crypto/sha.h> +#include <asm/byteorder.h> +#include <asm/simd.h> +#include <asm/neon.h> +#include "sha256_glue.h" + +asmlinkage void sha256_block_data_order(u32 *digest, const void *data, + unsigned int num_blks); + + +int sha256_init(struct shash_desc *desc) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + + sctx->state[0] = SHA256_H0; + sctx->state[1] = SHA256_H1; + sctx->state[2] = SHA256_H2; + sctx->state[3] = SHA256_H3; + sctx->state[4] = SHA256_H4; + sctx->state[5] = SHA256_H5; + sctx->state[6] = SHA256_H6; + sctx->state[7] = SHA256_H7; + sctx->count = 0; + + return 0; +} + +int sha224_init(struct shash_desc *desc) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + + sctx->state[0] = SHA224_H0; + sctx->state[1] = SHA224_H1; + sctx->state[2] = SHA224_H2; + sctx->state[3] = SHA224_H3; + sctx->state[4] = SHA224_H4; + sctx->state[5] = SHA224_H5; + sctx->state[6] = SHA224_H6; + sctx->state[7] = SHA224_H7; + sctx->count = 0; + + return 0; +} + +int __sha256_update(struct shash_desc *desc, const u8 *data, unsigned int len, + unsigned int partial) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int done = 0; + + sctx->count += len; + + if (partial) { + done = SHA256_BLOCK_SIZE - partial; + memcpy(sctx->buf + partial, data, done); + sha256_block_data_order(sctx->state, sctx->buf, 1); + } + + if (len - done >= SHA256_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / SHA256_BLOCK_SIZE; + + sha256_block_data_order(sctx->state, data + done, rounds); + done += rounds * SHA256_BLOCK_SIZE; + } + + memcpy(sctx->buf, data + done, len - done); + + return 0; +} + +int sha256_update(struct shash_desc *desc, const u8 *data, unsigned int len) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->count % SHA256_BLOCK_SIZE; + + /* Handle the fast case right here */ + if (partial + len < SHA256_BLOCK_SIZE) { + sctx->count += len; + memcpy(sctx->buf + partial, data, len); + + return 0; + } + + return __sha256_update(desc, data, len, partial); +} + +/* Add padding and return the message digest. */ +static int sha256_final(struct shash_desc *desc, u8 *out) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + __be32 *dst = (__be32 *)out; + __be64 bits; + static const u8 padding[SHA256_BLOCK_SIZE] = { 0x80, }; + + /* save number of bits */ + bits = cpu_to_be64(sctx->count << 3); + + /* Pad out to 56 mod 64 and append length */ + index = sctx->count % SHA256_BLOCK_SIZE; + padlen = (index < 56) ? (56 - index) : ((SHA256_BLOCK_SIZE+56)-index); + + /* We need to fill a whole block for __sha256_update */ + if (padlen <= 56) { + sctx->count += padlen; + memcpy(sctx->buf + index, padding, padlen); + } else { + __sha256_update(desc, padding, padlen, index); + } + __sha256_update(desc, (const u8 *)&bits, sizeof(bits), 56); + + /* Store state in digest */ + for (i = 0; i < 8; i++) + dst[i] = cpu_to_be32(sctx->state[i]); + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int sha224_final(struct shash_desc *desc, u8 *out) +{ + u8 D[SHA256_DIGEST_SIZE]; + + sha256_final(desc, D); + + memcpy(out, D, SHA224_DIGEST_SIZE); + memset(D, 0, SHA256_DIGEST_SIZE); + + return 0; +} + +int sha256_export(struct shash_desc *desc, void *out) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + + memcpy(out, sctx, sizeof(*sctx)); + + return 0; +} + +int sha256_import(struct shash_desc *desc, const void *in) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + + memcpy(sctx, in, sizeof(*sctx)); + + return 0; +} + +static struct shash_alg algs[] = { { + .digestsize = SHA256_DIGEST_SIZE, + .init = sha256_init, + .update = sha256_update, + .final = sha256_final, + .export = sha256_export, + .import = sha256_import, + .descsize = sizeof(struct sha256_state), + .statesize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha256", + .cra_driver_name = "sha256-asm", + .cra_priority = 150, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA256_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}, { + .digestsize = SHA224_DIGEST_SIZE, + .init = sha224_init, + .update = sha256_update, + .final = sha224_final, + .export = sha256_export, + .import = sha256_import, + .descsize = sizeof(struct sha256_state), + .statesize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha224", + .cra_driver_name = "sha224-asm", + .cra_priority = 150, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA224_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; + +static int __init sha256_mod_init(void) +{ + int res = crypto_register_shashes(algs, ARRAY_SIZE(algs)); + + if (res < 0) + return res; + + if (IS_ENABLED(CONFIG_KERNEL_MODE_NEON) && cpu_has_neon()) { + res = crypto_register_shashes(sha256_neon_algs, + ARRAY_SIZE(sha256_neon_algs)); + + if (res < 0) + crypto_unregister_shashes(algs, ARRAY_SIZE(algs)); + } + + return res; +} + +static void __exit sha256_mod_fini(void) +{ + crypto_unregister_shashes(algs, ARRAY_SIZE(algs)); + + if (IS_ENABLED(CONFIG_KERNEL_MODE_NEON) && cpu_has_neon()) + crypto_unregister_shashes(sha256_neon_algs, + ARRAY_SIZE(sha256_neon_algs)); +} + +module_init(sha256_mod_init); +module_exit(sha256_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA256 Secure Hash Algorithm (ARM), including NEON"); + +MODULE_ALIAS_CRYPTO("sha256"); diff --git a/arch/arm/crypto/sha256_glue.h b/arch/arm/crypto/sha256_glue.h new file mode 100644 index 000000000000..0312f4ffe8cc --- /dev/null +++ b/arch/arm/crypto/sha256_glue.h @@ -0,0 +1,23 @@ +#ifndef _CRYPTO_SHA256_GLUE_H +#define _CRYPTO_SHA256_GLUE_H + +#include <linux/crypto.h> +#include <crypto/sha.h> + +extern struct shash_alg sha256_neon_algs[2]; + +extern int sha256_init(struct shash_desc *desc); + +extern int sha224_init(struct shash_desc *desc); + +extern int __sha256_update(struct shash_desc *desc, const u8 *data, + unsigned int len, unsigned int partial); + +extern int sha256_update(struct shash_desc *desc, const u8 *data, + unsigned int len); + +extern int sha256_export(struct shash_desc *desc, void *out); + +extern int sha256_import(struct shash_desc *desc, const void *in); + +#endif /* _CRYPTO_SHA256_GLUE_H */ diff --git a/arch/arm/crypto/sha256_neon_glue.c b/arch/arm/crypto/sha256_neon_glue.c new file mode 100644 index 000000000000..3ff0a7f1d092 --- /dev/null +++ b/arch/arm/crypto/sha256_neon_glue.c @@ -0,0 +1,172 @@ +/* + * Glue code for the SHA256 Secure Hash Algorithm assembly implementation + * using NEON instructions. + * + * Copyright © 2015 Google Inc. + * + * This file is based on sha512_neon_glue.c: + * Copyright © 2014 Jussi Kivilinna <jussi.kivilinna@iki.fi> + * + * 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 <crypto/internal/hash.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <linux/string.h> +#include <crypto/sha.h> +#include <asm/byteorder.h> +#include <asm/simd.h> +#include <asm/neon.h> +#include "sha256_glue.h" + +asmlinkage void sha256_block_data_order_neon(u32 *digest, const void *data, + unsigned int num_blks); + + +static int __sha256_neon_update(struct shash_desc *desc, const u8 *data, + unsigned int len, unsigned int partial) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int done = 0; + + sctx->count += len; + + if (partial) { + done = SHA256_BLOCK_SIZE - partial; + memcpy(sctx->buf + partial, data, done); + sha256_block_data_order_neon(sctx->state, sctx->buf, 1); + } + + if (len - done >= SHA256_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / SHA256_BLOCK_SIZE; + + sha256_block_data_order_neon(sctx->state, data + done, rounds); + done += rounds * SHA256_BLOCK_SIZE; + } + + memcpy(sctx->buf, data + done, len - done); + + return 0; +} + +static int sha256_neon_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->count % SHA256_BLOCK_SIZE; + int res; + + /* Handle the fast case right here */ + if (partial + len < SHA256_BLOCK_SIZE) { + sctx->count += len; + memcpy(sctx->buf + partial, data, len); + + return 0; + } + + if (!may_use_simd()) { + res = __sha256_update(desc, data, len, partial); + } else { + kernel_neon_begin(); + res = __sha256_neon_update(desc, data, len, partial); + kernel_neon_end(); + } + + return res; +} + +/* Add padding and return the message digest. */ +static int sha256_neon_final(struct shash_desc *desc, u8 *out) +{ + struct sha256_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + __be32 *dst = (__be32 *)out; + __be64 bits; + static const u8 padding[SHA256_BLOCK_SIZE] = { 0x80, }; + + /* save number of bits */ + bits = cpu_to_be64(sctx->count << 3); + + /* Pad out to 56 mod 64 and append length */ + index = sctx->count % SHA256_BLOCK_SIZE; + padlen = (index < 56) ? (56 - index) : ((SHA256_BLOCK_SIZE+56)-index); + + if (!may_use_simd()) { + sha256_update(desc, padding, padlen); + sha256_update(desc, (const u8 *)&bits, sizeof(bits)); + } else { + kernel_neon_begin(); + /* We need to fill a whole block for __sha256_neon_update() */ + if (padlen <= 56) { + sctx->count += padlen; + memcpy(sctx->buf + index, padding, padlen); + } else { + __sha256_neon_update(desc, padding, padlen, index); + } + __sha256_neon_update(desc, (const u8 *)&bits, + sizeof(bits), 56); + kernel_neon_end(); + } + + /* Store state in digest */ + for (i = 0; i < 8; i++) + dst[i] = cpu_to_be32(sctx->state[i]); + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int sha224_neon_final(struct shash_desc *desc, u8 *out) +{ + u8 D[SHA256_DIGEST_SIZE]; + + sha256_neon_final(desc, D); + + memcpy(out, D, SHA224_DIGEST_SIZE); + memset(D, 0, SHA256_DIGEST_SIZE); + + return 0; +} + +struct shash_alg sha256_neon_algs[] = { { + .digestsize = SHA256_DIGEST_SIZE, + .init = sha256_init, + .update = sha256_neon_update, + .final = sha256_neon_final, + .export = sha256_export, + .import = sha256_import, + .descsize = sizeof(struct sha256_state), + .statesize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha256", + .cra_driver_name = "sha256-neon", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA256_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}, { + .digestsize = SHA224_DIGEST_SIZE, + .init = sha224_init, + .update = sha256_neon_update, + .final = sha224_neon_final, + .export = sha256_export, + .import = sha256_import, + .descsize = sizeof(struct sha256_state), + .statesize = sizeof(struct sha256_state), + .base = { + .cra_name = "sha224", + .cra_driver_name = "sha224-neon", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA224_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; diff --git a/arch/arm/crypto/sha512-armv7-neon.S b/arch/arm/crypto/sha512-armv7-neon.S new file mode 100644 index 000000000000..fe99472e507c --- /dev/null +++ b/arch/arm/crypto/sha512-armv7-neon.S @@ -0,0 +1,455 @@ +/* sha512-armv7-neon.S - ARM/NEON assembly implementation of SHA-512 transform + * + * Copyright © 2013-2014 Jussi Kivilinna <jussi.kivilinna@iki.fi> + * + * 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 <linux/linkage.h> + + +.syntax unified +.code 32 +.fpu neon + +.text + +/* structure of SHA512_CONTEXT */ +#define hd_a 0 +#define hd_b ((hd_a) + 8) +#define hd_c ((hd_b) + 8) +#define hd_d ((hd_c) + 8) +#define hd_e ((hd_d) + 8) +#define hd_f ((hd_e) + 8) +#define hd_g ((hd_f) + 8) + +/* register macros */ +#define RK %r2 + +#define RA d0 +#define RB d1 +#define RC d2 +#define RD d3 +#define RE d4 +#define RF d5 +#define RG d6 +#define RH d7 + +#define RT0 d8 +#define RT1 d9 +#define RT2 d10 +#define RT3 d11 +#define RT4 d12 +#define RT5 d13 +#define RT6 d14 +#define RT7 d15 + +#define RT01q q4 +#define RT23q q5 +#define RT45q q6 +#define RT67q q7 + +#define RW0 d16 +#define RW1 d17 +#define RW2 d18 +#define RW3 d19 +#define RW4 d20 +#define RW5 d21 +#define RW6 d22 +#define RW7 d23 +#define RW8 d24 +#define RW9 d25 +#define RW10 d26 +#define RW11 d27 +#define RW12 d28 +#define RW13 d29 +#define RW14 d30 +#define RW15 d31 + +#define RW01q q8 +#define RW23q q9 +#define RW45q q10 +#define RW67q q11 +#define RW89q q12 +#define RW1011q q13 +#define RW1213q q14 +#define RW1415q q15 + +/*********************************************************************** + * ARM assembly implementation of sha512 transform + ***********************************************************************/ +#define rounds2_0_63(ra, rb, rc, rd, re, rf, rg, rh, rw0, rw1, rw01q, rw2, \ + rw23q, rw1415q, rw9, rw10, interleave_op, arg1) \ + /* t1 = h + Sum1 (e) + Ch (e, f, g) + k[t] + w[t]; */ \ + vshr.u64 RT2, re, #14; \ + vshl.u64 RT3, re, #64 - 14; \ + interleave_op(arg1); \ + vshr.u64 RT4, re, #18; \ + vshl.u64 RT5, re, #64 - 18; \ + vld1.64 {RT0}, [RK]!; \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, re, #41; \ + vshl.u64 RT5, re, #64 - 41; \ + vadd.u64 RT0, RT0, rw0; \ + veor.64 RT23q, RT23q, RT45q; \ + vmov.64 RT7, re; \ + veor.64 RT1, RT2, RT3; \ + vbsl.64 RT7, rf, rg; \ + \ + vadd.u64 RT1, RT1, rh; \ + vshr.u64 RT2, ra, #28; \ + vshl.u64 RT3, ra, #64 - 28; \ + vadd.u64 RT1, RT1, RT0; \ + vshr.u64 RT4, ra, #34; \ + vshl.u64 RT5, ra, #64 - 34; \ + vadd.u64 RT1, RT1, RT7; \ + \ + /* h = Sum0 (a) + Maj (a, b, c); */ \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, ra, #39; \ + vshl.u64 RT5, ra, #64 - 39; \ + veor.64 RT0, ra, rb; \ + veor.64 RT23q, RT23q, RT45q; \ + vbsl.64 RT0, rc, rb; \ + vadd.u64 rd, rd, RT1; /* d+=t1; */ \ + veor.64 rh, RT2, RT3; \ + \ + /* t1 = g + Sum1 (d) + Ch (d, e, f) + k[t] + w[t]; */ \ + vshr.u64 RT2, rd, #14; \ + vshl.u64 RT3, rd, #64 - 14; \ + vadd.u64 rh, rh, RT0; \ + vshr.u64 RT4, rd, #18; \ + vshl.u64 RT5, rd, #64 - 18; \ + vadd.u64 rh, rh, RT1; /* h+=t1; */ \ + vld1.64 {RT0}, [RK]!; \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, rd, #41; \ + vshl.u64 RT5, rd, #64 - 41; \ + vadd.u64 RT0, RT0, rw1; \ + veor.64 RT23q, RT23q, RT45q; \ + vmov.64 RT7, rd; \ + veor.64 RT1, RT2, RT3; \ + vbsl.64 RT7, re, rf; \ + \ + vadd.u64 RT1, RT1, rg; \ + vshr.u64 RT2, rh, #28; \ + vshl.u64 RT3, rh, #64 - 28; \ + vadd.u64 RT1, RT1, RT0; \ + vshr.u64 RT4, rh, #34; \ + vshl.u64 RT5, rh, #64 - 34; \ + vadd.u64 RT1, RT1, RT7; \ + \ + /* g = Sum0 (h) + Maj (h, a, b); */ \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, rh, #39; \ + vshl.u64 RT5, rh, #64 - 39; \ + veor.64 RT0, rh, ra; \ + veor.64 RT23q, RT23q, RT45q; \ + vbsl.64 RT0, rb, ra; \ + vadd.u64 rc, rc, RT1; /* c+=t1; */ \ + veor.64 rg, RT2, RT3; \ + \ + /* w[0] += S1 (w[14]) + w[9] + S0 (w[1]); */ \ + /* w[1] += S1 (w[15]) + w[10] + S0 (w[2]); */ \ + \ + /**** S0(w[1:2]) */ \ + \ + /* w[0:1] += w[9:10] */ \ + /* RT23q = rw1:rw2 */ \ + vext.u64 RT23q, rw01q, rw23q, #1; \ + vadd.u64 rw0, rw9; \ + vadd.u64 rg, rg, RT0; \ + vadd.u64 rw1, rw10;\ + vadd.u64 rg, rg, RT1; /* g+=t1; */ \ + \ + vshr.u64 RT45q, RT23q, #1; \ + vshl.u64 RT67q, RT23q, #64 - 1; \ + vshr.u64 RT01q, RT23q, #8; \ + veor.u64 RT45q, RT45q, RT67q; \ + vshl.u64 RT67q, RT23q, #64 - 8; \ + veor.u64 RT45q, RT45q, RT01q; \ + vshr.u64 RT01q, RT23q, #7; \ + veor.u64 RT45q, RT45q, RT67q; \ + \ + /**** S1(w[14:15]) */ \ + vshr.u64 RT23q, rw1415q, #6; \ + veor.u64 RT01q, RT01q, RT45q; \ + vshr.u64 RT45q, rw1415q, #19; \ + vshl.u64 RT67q, rw1415q, #64 - 19; \ + veor.u64 RT23q, RT23q, RT45q; \ + vshr.u64 RT45q, rw1415q, #61; \ + veor.u64 RT23q, RT23q, RT67q; \ + vshl.u64 RT67q, rw1415q, #64 - 61; \ + veor.u64 RT23q, RT23q, RT45q; \ + vadd.u64 rw01q, RT01q; /* w[0:1] += S(w[1:2]) */ \ + veor.u64 RT01q, RT23q, RT67q; +#define vadd_RT01q(rw01q) \ + /* w[0:1] += S(w[14:15]) */ \ + vadd.u64 rw01q, RT01q; + +#define dummy(_) /*_*/ + +#define rounds2_64_79(ra, rb, rc, rd, re, rf, rg, rh, rw0, rw1, \ + interleave_op1, arg1, interleave_op2, arg2) \ + /* t1 = h + Sum1 (e) + Ch (e, f, g) + k[t] + w[t]; */ \ + vshr.u64 RT2, re, #14; \ + vshl.u64 RT3, re, #64 - 14; \ + interleave_op1(arg1); \ + vshr.u64 RT4, re, #18; \ + vshl.u64 RT5, re, #64 - 18; \ + interleave_op2(arg2); \ + vld1.64 {RT0}, [RK]!; \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, re, #41; \ + vshl.u64 RT5, re, #64 - 41; \ + vadd.u64 RT0, RT0, rw0; \ + veor.64 RT23q, RT23q, RT45q; \ + vmov.64 RT7, re; \ + veor.64 RT1, RT2, RT3; \ + vbsl.64 RT7, rf, rg; \ + \ + vadd.u64 RT1, RT1, rh; \ + vshr.u64 RT2, ra, #28; \ + vshl.u64 RT3, ra, #64 - 28; \ + vadd.u64 RT1, RT1, RT0; \ + vshr.u64 RT4, ra, #34; \ + vshl.u64 RT5, ra, #64 - 34; \ + vadd.u64 RT1, RT1, RT7; \ + \ + /* h = Sum0 (a) + Maj (a, b, c); */ \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, ra, #39; \ + vshl.u64 RT5, ra, #64 - 39; \ + veor.64 RT0, ra, rb; \ + veor.64 RT23q, RT23q, RT45q; \ + vbsl.64 RT0, rc, rb; \ + vadd.u64 rd, rd, RT1; /* d+=t1; */ \ + veor.64 rh, RT2, RT3; \ + \ + /* t1 = g + Sum1 (d) + Ch (d, e, f) + k[t] + w[t]; */ \ + vshr.u64 RT2, rd, #14; \ + vshl.u64 RT3, rd, #64 - 14; \ + vadd.u64 rh, rh, RT0; \ + vshr.u64 RT4, rd, #18; \ + vshl.u64 RT5, rd, #64 - 18; \ + vadd.u64 rh, rh, RT1; /* h+=t1; */ \ + vld1.64 {RT0}, [RK]!; \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, rd, #41; \ + vshl.u64 RT5, rd, #64 - 41; \ + vadd.u64 RT0, RT0, rw1; \ + veor.64 RT23q, RT23q, RT45q; \ + vmov.64 RT7, rd; \ + veor.64 RT1, RT2, RT3; \ + vbsl.64 RT7, re, rf; \ + \ + vadd.u64 RT1, RT1, rg; \ + vshr.u64 RT2, rh, #28; \ + vshl.u64 RT3, rh, #64 - 28; \ + vadd.u64 RT1, RT1, RT0; \ + vshr.u64 RT4, rh, #34; \ + vshl.u64 RT5, rh, #64 - 34; \ + vadd.u64 RT1, RT1, RT7; \ + \ + /* g = Sum0 (h) + Maj (h, a, b); */ \ + veor.64 RT23q, RT23q, RT45q; \ + vshr.u64 RT4, rh, #39; \ + vshl.u64 RT5, rh, #64 - 39; \ + veor.64 RT0, rh, ra; \ + veor.64 RT23q, RT23q, RT45q; \ + vbsl.64 RT0, rb, ra; \ + vadd.u64 rc, rc, RT1; /* c+=t1; */ \ + veor.64 rg, RT2, RT3; +#define vadd_rg_RT0(rg) \ + vadd.u64 rg, rg, RT0; +#define vadd_rg_RT1(rg) \ + vadd.u64 rg, rg, RT1; /* g+=t1; */ + +.align 3 +ENTRY(sha512_transform_neon) + /* Input: + * %r0: SHA512_CONTEXT + * %r1: data + * %r2: u64 k[] constants + * %r3: nblks + */ + push {%lr}; + + mov %lr, #0; + + /* Load context to d0-d7 */ + vld1.64 {RA-RD}, [%r0]!; + vld1.64 {RE-RH}, [%r0]; + sub %r0, #(4*8); + + /* Load input to w[16], d16-d31 */ + /* NOTE: Assumes that on ARMv7 unaligned accesses are always allowed. */ + vld1.64 {RW0-RW3}, [%r1]!; + vld1.64 {RW4-RW7}, [%r1]!; + vld1.64 {RW8-RW11}, [%r1]!; + vld1.64 {RW12-RW15}, [%r1]!; +#ifdef __ARMEL__ + /* byteswap */ + vrev64.8 RW01q, RW01q; + vrev64.8 RW23q, RW23q; + vrev64.8 RW45q, RW45q; + vrev64.8 RW67q, RW67q; + vrev64.8 RW89q, RW89q; + vrev64.8 RW1011q, RW1011q; + vrev64.8 RW1213q, RW1213q; + vrev64.8 RW1415q, RW1415q; +#endif + + /* EABI says that d8-d15 must be preserved by callee. */ + /*vpush {RT0-RT7};*/ + +.Loop: + rounds2_0_63(RA, RB, RC, RD, RE, RF, RG, RH, RW0, RW1, RW01q, RW2, + RW23q, RW1415q, RW9, RW10, dummy, _); + b .Lenter_rounds; + +.Loop_rounds: + rounds2_0_63(RA, RB, RC, RD, RE, RF, RG, RH, RW0, RW1, RW01q, RW2, + RW23q, RW1415q, RW9, RW10, vadd_RT01q, RW1415q); +.Lenter_rounds: + rounds2_0_63(RG, RH, RA, RB, RC, RD, RE, RF, RW2, RW3, RW23q, RW4, + RW45q, RW01q, RW11, RW12, vadd_RT01q, RW01q); + rounds2_0_63(RE, RF, RG, RH, RA, RB, RC, RD, RW4, RW5, RW45q, RW6, + RW67q, RW23q, RW13, RW14, vadd_RT01q, RW23q); + rounds2_0_63(RC, RD, RE, RF, RG, RH, RA, RB, RW6, RW7, RW67q, RW8, + RW89q, RW45q, RW15, RW0, vadd_RT01q, RW45q); + rounds2_0_63(RA, RB, RC, RD, RE, RF, RG, RH, RW8, RW9, RW89q, RW10, + RW1011q, RW67q, RW1, RW2, vadd_RT01q, RW67q); + rounds2_0_63(RG, RH, RA, RB, RC, RD, RE, RF, RW10, RW11, RW1011q, RW12, + RW1213q, RW89q, RW3, RW4, vadd_RT01q, RW89q); + add %lr, #16; + rounds2_0_63(RE, RF, RG, RH, RA, RB, RC, RD, RW12, RW13, RW1213q, RW14, + RW1415q, RW1011q, RW5, RW6, vadd_RT01q, RW1011q); + cmp %lr, #64; + rounds2_0_63(RC, RD, RE, RF, RG, RH, RA, RB, RW14, RW15, RW1415q, RW0, + RW01q, RW1213q, RW7, RW8, vadd_RT01q, RW1213q); + bne .Loop_rounds; + + subs %r3, #1; + + rounds2_64_79(RA, RB, RC, RD, RE, RF, RG, RH, RW0, RW1, + vadd_RT01q, RW1415q, dummy, _); + rounds2_64_79(RG, RH, RA, RB, RC, RD, RE, RF, RW2, RW3, + vadd_rg_RT0, RG, vadd_rg_RT1, RG); + beq .Lhandle_tail; + vld1.64 {RW0-RW3}, [%r1]!; + rounds2_64_79(RE, RF, RG, RH, RA, RB, RC, RD, RW4, RW5, + vadd_rg_RT0, RE, vadd_rg_RT1, RE); + rounds2_64_79(RC, RD, RE, RF, RG, RH, RA, RB, RW6, RW7, + vadd_rg_RT0, RC, vadd_rg_RT1, RC); +#ifdef __ARMEL__ + vrev64.8 RW01q, RW01q; + vrev64.8 RW23q, RW23q; +#endif + vld1.64 {RW4-RW7}, [%r1]!; + rounds2_64_79(RA, RB, RC, RD, RE, RF, RG, RH, RW8, RW9, + vadd_rg_RT0, RA, vadd_rg_RT1, RA); + rounds2_64_79(RG, RH, RA, RB, RC, RD, RE, RF, RW10, RW11, + vadd_rg_RT0, RG, vadd_rg_RT1, RG); +#ifdef __ARMEL__ + vrev64.8 RW45q, RW45q; + vrev64.8 RW67q, RW67q; +#endif + vld1.64 {RW8-RW11}, [%r1]!; + rounds2_64_79(RE, RF, RG, RH, RA, RB, RC, RD, RW12, RW13, + vadd_rg_RT0, RE, vadd_rg_RT1, RE); + rounds2_64_79(RC, RD, RE, RF, RG, RH, RA, RB, RW14, RW15, + vadd_rg_RT0, RC, vadd_rg_RT1, RC); +#ifdef __ARMEL__ + vrev64.8 RW89q, RW89q; + vrev64.8 RW1011q, RW1011q; +#endif + vld1.64 {RW12-RW15}, [%r1]!; + vadd_rg_RT0(RA); + vadd_rg_RT1(RA); + + /* Load context */ + vld1.64 {RT0-RT3}, [%r0]!; + vld1.64 {RT4-RT7}, [%r0]; + sub %r0, #(4*8); + +#ifdef __ARMEL__ + vrev64.8 RW1213q, RW1213q; + vrev64.8 RW1415q, RW1415q; +#endif + + vadd.u64 RA, RT0; + vadd.u64 RB, RT1; + vadd.u64 RC, RT2; + vadd.u64 RD, RT3; + vadd.u64 RE, RT4; + vadd.u64 RF, RT5; + vadd.u64 RG, RT6; + vadd.u64 RH, RT7; + + /* Store the first half of context */ + vst1.64 {RA-RD}, [%r0]!; + sub RK, $(8*80); + vst1.64 {RE-RH}, [%r0]; /* Store the last half of context */ + mov %lr, #0; + sub %r0, #(4*8); + + b .Loop; + +.Lhandle_tail: + rounds2_64_79(RE, RF, RG, RH, RA, RB, RC, RD, RW4, RW5, + vadd_rg_RT0, RE, vadd_rg_RT1, RE); + rounds2_64_79(RC, RD, RE, RF, RG, RH, RA, RB, RW6, RW7, + vadd_rg_RT0, RC, vadd_rg_RT1, RC); + rounds2_64_79(RA, RB, RC, RD, RE, RF, RG, RH, RW8, RW9, + vadd_rg_RT0, RA, vadd_rg_RT1, RA); + rounds2_64_79(RG, RH, RA, RB, RC, RD, RE, RF, RW10, RW11, + vadd_rg_RT0, RG, vadd_rg_RT1, RG); + rounds2_64_79(RE, RF, RG, RH, RA, RB, RC, RD, RW12, RW13, + vadd_rg_RT0, RE, vadd_rg_RT1, RE); + rounds2_64_79(RC, RD, RE, RF, RG, RH, RA, RB, RW14, RW15, + vadd_rg_RT0, RC, vadd_rg_RT1, RC); + + /* Load context to d16-d23 */ + vld1.64 {RW0-RW3}, [%r0]!; + vadd_rg_RT0(RA); + vld1.64 {RW4-RW7}, [%r0]; + vadd_rg_RT1(RA); + sub %r0, #(4*8); + + vadd.u64 RA, RW0; + vadd.u64 RB, RW1; + vadd.u64 RC, RW2; + vadd.u64 RD, RW3; + vadd.u64 RE, RW4; + vadd.u64 RF, RW5; + vadd.u64 RG, RW6; + vadd.u64 RH, RW7; + + /* Store the first half of context */ + vst1.64 {RA-RD}, [%r0]!; + + /* Clear used registers */ + /* d16-d31 */ + veor.u64 RW01q, RW01q; + veor.u64 RW23q, RW23q; + veor.u64 RW45q, RW45q; + veor.u64 RW67q, RW67q; + vst1.64 {RE-RH}, [%r0]; /* Store the last half of context */ + veor.u64 RW89q, RW89q; + veor.u64 RW1011q, RW1011q; + veor.u64 RW1213q, RW1213q; + veor.u64 RW1415q, RW1415q; + /* d8-d15 */ + /*vpop {RT0-RT7};*/ + /* d0-d7 (q0-q3) */ + veor.u64 %q0, %q0; + veor.u64 %q1, %q1; + veor.u64 %q2, %q2; + veor.u64 %q3, %q3; + + pop {%pc}; +ENDPROC(sha512_transform_neon) diff --git a/arch/arm/crypto/sha512_neon_glue.c b/arch/arm/crypto/sha512_neon_glue.c new file mode 100644 index 000000000000..0d2758ff5e12 --- /dev/null +++ b/arch/arm/crypto/sha512_neon_glue.c @@ -0,0 +1,305 @@ +/* + * Glue code for the SHA512 Secure Hash Algorithm assembly implementation + * using NEON instructions. + * + * Copyright © 2014 Jussi Kivilinna <jussi.kivilinna@iki.fi> + * + * This file is based on sha512_ssse3_glue.c: + * Copyright (C) 2013 Intel Corporation + * Author: Tim Chen <tim.c.chen@linux.intel.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 <crypto/internal/hash.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/mm.h> +#include <linux/cryptohash.h> +#include <linux/types.h> +#include <linux/string.h> +#include <crypto/sha.h> +#include <asm/byteorder.h> +#include <asm/simd.h> +#include <asm/neon.h> + + +static const u64 sha512_k[] = { + 0x428a2f98d728ae22ULL, 0x7137449123ef65cdULL, + 0xb5c0fbcfec4d3b2fULL, 0xe9b5dba58189dbbcULL, + 0x3956c25bf348b538ULL, 0x59f111f1b605d019ULL, + 0x923f82a4af194f9bULL, 0xab1c5ed5da6d8118ULL, + 0xd807aa98a3030242ULL, 0x12835b0145706fbeULL, + 0x243185be4ee4b28cULL, 0x550c7dc3d5ffb4e2ULL, + 0x72be5d74f27b896fULL, 0x80deb1fe3b1696b1ULL, + 0x9bdc06a725c71235ULL, 0xc19bf174cf692694ULL, + 0xe49b69c19ef14ad2ULL, 0xefbe4786384f25e3ULL, + 0x0fc19dc68b8cd5b5ULL, 0x240ca1cc77ac9c65ULL, + 0x2de92c6f592b0275ULL, 0x4a7484aa6ea6e483ULL, + 0x5cb0a9dcbd41fbd4ULL, 0x76f988da831153b5ULL, + 0x983e5152ee66dfabULL, 0xa831c66d2db43210ULL, + 0xb00327c898fb213fULL, 0xbf597fc7beef0ee4ULL, + 0xc6e00bf33da88fc2ULL, 0xd5a79147930aa725ULL, + 0x06ca6351e003826fULL, 0x142929670a0e6e70ULL, + 0x27b70a8546d22ffcULL, 0x2e1b21385c26c926ULL, + 0x4d2c6dfc5ac42aedULL, 0x53380d139d95b3dfULL, + 0x650a73548baf63deULL, 0x766a0abb3c77b2a8ULL, + 0x81c2c92e47edaee6ULL, 0x92722c851482353bULL, + 0xa2bfe8a14cf10364ULL, 0xa81a664bbc423001ULL, + 0xc24b8b70d0f89791ULL, 0xc76c51a30654be30ULL, + 0xd192e819d6ef5218ULL, 0xd69906245565a910ULL, + 0xf40e35855771202aULL, 0x106aa07032bbd1b8ULL, + 0x19a4c116b8d2d0c8ULL, 0x1e376c085141ab53ULL, + 0x2748774cdf8eeb99ULL, 0x34b0bcb5e19b48a8ULL, + 0x391c0cb3c5c95a63ULL, 0x4ed8aa4ae3418acbULL, + 0x5b9cca4f7763e373ULL, 0x682e6ff3d6b2b8a3ULL, + 0x748f82ee5defb2fcULL, 0x78a5636f43172f60ULL, + 0x84c87814a1f0ab72ULL, 0x8cc702081a6439ecULL, + 0x90befffa23631e28ULL, 0xa4506cebde82bde9ULL, + 0xbef9a3f7b2c67915ULL, 0xc67178f2e372532bULL, + 0xca273eceea26619cULL, 0xd186b8c721c0c207ULL, + 0xeada7dd6cde0eb1eULL, 0xf57d4f7fee6ed178ULL, + 0x06f067aa72176fbaULL, 0x0a637dc5a2c898a6ULL, + 0x113f9804bef90daeULL, 0x1b710b35131c471bULL, + 0x28db77f523047d84ULL, 0x32caab7b40c72493ULL, + 0x3c9ebe0a15c9bebcULL, 0x431d67c49c100d4cULL, + 0x4cc5d4becb3e42b6ULL, 0x597f299cfc657e2aULL, + 0x5fcb6fab3ad6faecULL, 0x6c44198c4a475817ULL +}; + + +asmlinkage void sha512_transform_neon(u64 *digest, const void *data, + const u64 k[], unsigned int num_blks); + + +static int sha512_neon_init(struct shash_desc *desc) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + + sctx->state[0] = SHA512_H0; + sctx->state[1] = SHA512_H1; + sctx->state[2] = SHA512_H2; + sctx->state[3] = SHA512_H3; + sctx->state[4] = SHA512_H4; + sctx->state[5] = SHA512_H5; + sctx->state[6] = SHA512_H6; + sctx->state[7] = SHA512_H7; + sctx->count[0] = sctx->count[1] = 0; + + return 0; +} + +static int __sha512_neon_update(struct shash_desc *desc, const u8 *data, + unsigned int len, unsigned int partial) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + unsigned int done = 0; + + sctx->count[0] += len; + if (sctx->count[0] < len) + sctx->count[1]++; + + if (partial) { + done = SHA512_BLOCK_SIZE - partial; + memcpy(sctx->buf + partial, data, done); + sha512_transform_neon(sctx->state, sctx->buf, sha512_k, 1); + } + + if (len - done >= SHA512_BLOCK_SIZE) { + const unsigned int rounds = (len - done) / SHA512_BLOCK_SIZE; + + sha512_transform_neon(sctx->state, data + done, sha512_k, + rounds); + + done += rounds * SHA512_BLOCK_SIZE; + } + + memcpy(sctx->buf, data + done, len - done); + + return 0; +} + +static int sha512_neon_update(struct shash_desc *desc, const u8 *data, + unsigned int len) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + unsigned int partial = sctx->count[0] % SHA512_BLOCK_SIZE; + int res; + + /* Handle the fast case right here */ + if (partial + len < SHA512_BLOCK_SIZE) { + sctx->count[0] += len; + if (sctx->count[0] < len) + sctx->count[1]++; + memcpy(sctx->buf + partial, data, len); + + return 0; + } + + if (!may_use_simd()) { + res = crypto_sha512_update(desc, data, len); + } else { + kernel_neon_begin(); + res = __sha512_neon_update(desc, data, len, partial); + kernel_neon_end(); + } + + return res; +} + + +/* Add padding and return the message digest. */ +static int sha512_neon_final(struct shash_desc *desc, u8 *out) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + unsigned int i, index, padlen; + __be64 *dst = (__be64 *)out; + __be64 bits[2]; + static const u8 padding[SHA512_BLOCK_SIZE] = { 0x80, }; + + /* save number of bits */ + bits[1] = cpu_to_be64(sctx->count[0] << 3); + bits[0] = cpu_to_be64(sctx->count[1] << 3 | sctx->count[0] >> 61); + + /* Pad out to 112 mod 128 and append length */ + index = sctx->count[0] & 0x7f; + padlen = (index < 112) ? (112 - index) : ((128+112) - index); + + if (!may_use_simd()) { + crypto_sha512_update(desc, padding, padlen); + crypto_sha512_update(desc, (const u8 *)&bits, sizeof(bits)); + } else { + kernel_neon_begin(); + /* We need to fill a whole block for __sha512_neon_update() */ + if (padlen <= 112) { + sctx->count[0] += padlen; + if (sctx->count[0] < padlen) + sctx->count[1]++; + memcpy(sctx->buf + index, padding, padlen); + } else { + __sha512_neon_update(desc, padding, padlen, index); + } + __sha512_neon_update(desc, (const u8 *)&bits, + sizeof(bits), 112); + kernel_neon_end(); + } + + /* Store state in digest */ + for (i = 0; i < 8; i++) + dst[i] = cpu_to_be64(sctx->state[i]); + + /* Wipe context */ + memset(sctx, 0, sizeof(*sctx)); + + return 0; +} + +static int sha512_neon_export(struct shash_desc *desc, void *out) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + + memcpy(out, sctx, sizeof(*sctx)); + + return 0; +} + +static int sha512_neon_import(struct shash_desc *desc, const void *in) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + + memcpy(sctx, in, sizeof(*sctx)); + + return 0; +} + +static int sha384_neon_init(struct shash_desc *desc) +{ + struct sha512_state *sctx = shash_desc_ctx(desc); + + sctx->state[0] = SHA384_H0; + sctx->state[1] = SHA384_H1; + sctx->state[2] = SHA384_H2; + sctx->state[3] = SHA384_H3; + sctx->state[4] = SHA384_H4; + sctx->state[5] = SHA384_H5; + sctx->state[6] = SHA384_H6; + sctx->state[7] = SHA384_H7; + + sctx->count[0] = sctx->count[1] = 0; + + return 0; +} + +static int sha384_neon_final(struct shash_desc *desc, u8 *hash) +{ + u8 D[SHA512_DIGEST_SIZE]; + + sha512_neon_final(desc, D); + + memcpy(hash, D, SHA384_DIGEST_SIZE); + memset(D, 0, SHA512_DIGEST_SIZE); + + return 0; +} + +static struct shash_alg algs[] = { { + .digestsize = SHA512_DIGEST_SIZE, + .init = sha512_neon_init, + .update = sha512_neon_update, + .final = sha512_neon_final, + .export = sha512_neon_export, + .import = sha512_neon_import, + .descsize = sizeof(struct sha512_state), + .statesize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha512", + .cra_driver_name = "sha512-neon", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA512_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +}, { + .digestsize = SHA384_DIGEST_SIZE, + .init = sha384_neon_init, + .update = sha512_neon_update, + .final = sha384_neon_final, + .export = sha512_neon_export, + .import = sha512_neon_import, + .descsize = sizeof(struct sha512_state), + .statesize = sizeof(struct sha512_state), + .base = { + .cra_name = "sha384", + .cra_driver_name = "sha384-neon", + .cra_priority = 250, + .cra_flags = CRYPTO_ALG_TYPE_SHASH, + .cra_blocksize = SHA384_BLOCK_SIZE, + .cra_module = THIS_MODULE, + } +} }; + +static int __init sha512_neon_mod_init(void) +{ + if (!cpu_has_neon()) + return -ENODEV; + + return crypto_register_shashes(algs, ARRAY_SIZE(algs)); +} + +static void __exit sha512_neon_mod_fini(void) +{ + crypto_unregister_shashes(algs, ARRAY_SIZE(algs)); +} + +module_init(sha512_neon_mod_init); +module_exit(sha512_neon_mod_fini); + +MODULE_LICENSE("GPL"); +MODULE_DESCRIPTION("SHA512 Secure Hash Algorithm, NEON accelerated"); + +MODULE_ALIAS("sha512"); +MODULE_ALIAS("sha384"); diff --git a/arch/arm/include/asm/Kbuild b/arch/arm/include/asm/Kbuild index d3db39860b9c..6577b8aeb711 100644 --- a/arch/arm/include/asm/Kbuild +++ b/arch/arm/include/asm/Kbuild @@ -24,6 +24,7 @@ generic-y += sembuf.h generic-y += serial.h generic-y += shmbuf.h generic-y += siginfo.h +generic-y += simd.h generic-y += sizes.h generic-y += socket.h generic-y += sockios.h diff --git a/arch/arm/include/asm/barrier.h b/arch/arm/include/asm/barrier.h index 60f15e274e6d..2f59f7443396 100644 --- a/arch/arm/include/asm/barrier.h +++ b/arch/arm/include/asm/barrier.h @@ -59,6 +59,21 @@ #define smp_wmb() dmb(ishst) #endif +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ___p1; \ +}) + #define read_barrier_depends() do { } while(0) #define smp_read_barrier_depends() do { } while(0) diff --git a/arch/arm/include/asm/cacheflush.h b/arch/arm/include/asm/cacheflush.h index 2059f019bef4..455e6637c881 100644 --- a/arch/arm/include/asm/cacheflush.h +++ b/arch/arm/include/asm/cacheflush.h @@ -16,6 +16,7 @@ #include <asm/shmparam.h> #include <asm/cachetype.h> #include <asm/outercache.h> +#include <asm/rodata.h> #define CACHE_COLOUR(vaddr) ((vaddr & (SHMLBA - 1)) >> PAGE_SHIFT) diff --git a/arch/arm/include/asm/crypto/sha1.h b/arch/arm/include/asm/crypto/sha1.h new file mode 100644 index 000000000000..75e6a417416b --- /dev/null +++ b/arch/arm/include/asm/crypto/sha1.h @@ -0,0 +1,10 @@ +#ifndef ASM_ARM_CRYPTO_SHA1_H +#define ASM_ARM_CRYPTO_SHA1_H + +#include <linux/crypto.h> +#include <crypto/sha.h> + +extern int sha1_update_arm(struct shash_desc *desc, const u8 *data, + unsigned int len); + +#endif 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/hardirq.h b/arch/arm/include/asm/hardirq.h index 3d7351c844aa..fe3ea776dc34 100644 --- a/arch/arm/include/asm/hardirq.h +++ b/arch/arm/include/asm/hardirq.h @@ -5,7 +5,7 @@ #include <linux/threads.h> #include <asm/irq.h> -#define NR_IPI 7 +#define NR_IPI 8 typedef struct { unsigned int __softirq_pending; diff --git a/arch/arm/include/asm/hardware/cache-l2x0.h b/arch/arm/include/asm/hardware/cache-l2x0.h index 3b2c40b5bfa2..0ca0f5a7c84b 100644 --- a/arch/arm/include/asm/hardware/cache-l2x0.h +++ b/arch/arm/include/asm/hardware/cache-l2x0.h @@ -66,6 +66,7 @@ #define L2X0_STNDBY_MODE_EN (1 << 0) /* Registers shifts and masks */ +#define L2X0_CACHE_ID_REV_MASK (0x3f) #define L2X0_CACHE_ID_PART_MASK (0xf << 6) #define L2X0_CACHE_ID_PART_L210 (1 << 6) #define L2X0_CACHE_ID_PART_L310 (3 << 6) @@ -106,6 +107,8 @@ #define L2X0_WAY_SIZE_SHIFT 3 +#define REV_PL310_R2P0 4 + #ifndef __ASSEMBLY__ extern void __init l2x0_init(void __iomem *base, u32 aux_val, u32 aux_mask); #if defined(CONFIG_CACHE_L2X0) && defined(CONFIG_OF) diff --git a/arch/arm/include/asm/irq.h b/arch/arm/include/asm/irq.h index 53c15dec7af6..809203a4b71b 100644 --- a/arch/arm/include/asm/irq.h +++ b/arch/arm/include/asm/irq.h @@ -35,6 +35,9 @@ extern void (*handle_arch_irq)(struct pt_regs *); extern void set_handle_irq(void (*handle_irq)(struct pt_regs *)); #endif +void arch_trigger_all_cpu_backtrace(void); +#define arch_trigger_all_cpu_backtrace arch_trigger_all_cpu_backtrace + #endif #endif diff --git a/arch/arm/include/asm/mach/mmc.h b/arch/arm/include/asm/mach/mmc.h new file mode 100644 index 000000000000..bca864ac945f --- /dev/null +++ b/arch/arm/include/asm/mach/mmc.h @@ -0,0 +1,28 @@ +/* + * arch/arm/include/asm/mach/mmc.h + */ +#ifndef ASMARM_MACH_MMC_H +#define ASMARM_MACH_MMC_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 mmc_platform_data { + unsigned int ocr_mask; /* available voltages */ + int built_in; /* built-in device flag */ + int card_present; /* card detect state */ + u32 (*translate_vdd)(struct device *, unsigned int); + unsigned int (*status)(struct device *); + struct embedded_sdio_data *embedded_sdio; + int (*register_status_notify)(void (*callback)(int card_present, void *dev_id), void *dev_id); +}; + +#endif diff --git a/arch/arm/include/asm/neon.h b/arch/arm/include/asm/neon.h new file mode 100644 index 000000000000..8f730fe70093 --- /dev/null +++ b/arch/arm/include/asm/neon.h @@ -0,0 +1,36 @@ +/* + * linux/arch/arm/include/asm/neon.h + * + * Copyright (C) 2013 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 + * published by the Free Software Foundation. + */ + +#include <asm/hwcap.h> + +#define cpu_has_neon() (!!(elf_hwcap & HWCAP_NEON)) + +#ifdef __ARM_NEON__ + +/* + * If you are affected by the BUILD_BUG below, it probably means that you are + * using NEON code /and/ calling the kernel_neon_begin() function from the same + * compilation unit. To prevent issues that may arise from GCC reordering or + * generating(1) NEON instructions outside of these begin/end functions, the + * only supported way of using NEON code in the kernel is by isolating it in a + * separate compilation unit, and calling it from another unit from inside a + * kernel_neon_begin/kernel_neon_end pair. + * + * (1) Current GCC (4.7) might generate NEON instructions at O3 level if + * -mpfu=neon is set. + */ + +#define kernel_neon_begin() \ + BUILD_BUG_ON_MSG(1, "kernel_neon_begin() called from NEON code") + +#else +void kernel_neon_begin(void); +#endif +void kernel_neon_end(void); diff --git a/arch/arm/include/asm/rodata.h b/arch/arm/include/asm/rodata.h new file mode 100644 index 000000000000..8c8add87bbc5 --- /dev/null +++ b/arch/arm/include/asm/rodata.h @@ -0,0 +1,32 @@ +/* + * arch/arm/include/asm/rodata.h + * + * Copyright (C) 2011 Google, Inc. + * + * Author: Colin Cross <ccross@android.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. + */ +#ifndef _ASMARM_RODATA_H +#define _ASMARM_RODATA_H + +#ifndef __ASSEMBLY__ + +#ifdef CONFIG_DEBUG_RODATA + +int set_memory_rw(unsigned long virt, int numpages); +int set_memory_ro(unsigned long virt, int numpages); + +void mark_rodata_ro(void); +void set_kernel_text_rw(void); +void set_kernel_text_ro(void); +#else +static inline void set_kernel_text_rw(void) { } +static inline void set_kernel_text_ro(void) { } +#endif + +#endif + +#endif diff --git a/arch/arm/include/asm/smp.h b/arch/arm/include/asm/smp.h index 610ccf33f5e7..67a18a5ed9fa 100644 --- a/arch/arm/include/asm/smp.h +++ b/arch/arm/include/asm/smp.h @@ -82,6 +82,7 @@ extern void arch_send_call_function_ipi_mask(const struct cpumask *mask); extern void arch_send_wakeup_ipi_mask(const struct cpumask *mask); extern int register_ipi_completion(struct completion *completion, int cpu); +extern void smp_send_all_cpu_backtrace(void); struct smp_operations { #ifdef CONFIG_SMP diff --git a/arch/arm/include/asm/syscall.h b/arch/arm/include/asm/syscall.h index 73ddd7239b33..ed805f1d3785 100644 --- a/arch/arm/include/asm/syscall.h +++ b/arch/arm/include/asm/syscall.h @@ -103,8 +103,7 @@ static inline void syscall_set_arguments(struct task_struct *task, memcpy(®s->ARM_r0 + i, args, n * sizeof(args[0])); } -static inline int syscall_get_arch(struct task_struct *task, - struct pt_regs *regs) +static inline int syscall_get_arch(void) { /* ARM tasks don't change audit architectures on the fly. */ return AUDIT_ARCH_ARM; diff --git a/arch/arm/include/asm/unistd.h b/arch/arm/include/asm/unistd.h index cbd61977c996..43876245fc57 100644 --- a/arch/arm/include/asm/unistd.h +++ b/arch/arm/include/asm/unistd.h @@ -15,7 +15,7 @@ #include <uapi/asm/unistd.h> -#define __NR_syscalls (380) +#define __NR_syscalls (384) #define __ARM_NR_cmpxchg (__ARM_NR_BASE+0x00fff0) #define __ARCH_WANT_STAT64 diff --git a/arch/arm/include/uapi/asm/unistd.h b/arch/arm/include/uapi/asm/unistd.h index af33b44990ed..17407c92c0da 100644 --- a/arch/arm/include/uapi/asm/unistd.h +++ b/arch/arm/include/uapi/asm/unistd.h @@ -406,6 +406,12 @@ #define __NR_process_vm_writev (__NR_SYSCALL_BASE+377) #define __NR_kcmp (__NR_SYSCALL_BASE+378) #define __NR_finit_module (__NR_SYSCALL_BASE+379) +/* Reserve for later +#define __NR_sched_setattr (__NR_SYSCALL_BASE+380) +#define __NR_sched_getattr (__NR_SYSCALL_BASE+381) +#define __NR_renameat2 (__NR_SYSCALL_BASE+382) +*/ +#define __NR_seccomp (__NR_SYSCALL_BASE+383) /* * This may need to be greater than __NR_last_syscall+1 in order to diff --git a/arch/arm/kernel/calls.S b/arch/arm/kernel/calls.S index c6ca7e376773..725f844926ea 100644 --- a/arch/arm/kernel/calls.S +++ b/arch/arm/kernel/calls.S @@ -389,6 +389,11 @@ CALL(sys_process_vm_writev) CALL(sys_kcmp) CALL(sys_finit_module) +/* 380 */ CALL(sys_ni_syscall) /* reserved sys_sched_setattr */ + CALL(sys_ni_syscall) /* reserved sys_sched_getattr */ + CALL(sys_ni_syscall) /* reserved sys_renameat2 */ + CALL(sys_seccomp) + #ifndef syscalls_counted .equ syscalls_padding, ((NR_syscalls + 3) & ~3) - NR_syscalls #define syscalls_counted diff --git a/arch/arm/kernel/ftrace.c b/arch/arm/kernel/ftrace.c index 34e56647dcee..6a740a93f4bb 100644 --- a/arch/arm/kernel/ftrace.c +++ b/arch/arm/kernel/ftrace.c @@ -13,6 +13,7 @@ */ #include <linux/ftrace.h> +#include <linux/module.h> #include <linux/uaccess.h> #include <asm/cacheflush.h> @@ -63,6 +64,20 @@ static unsigned long adjust_address(struct dyn_ftrace *rec, unsigned long addr) } #endif +int ftrace_arch_code_modify_prepare(void) +{ + set_kernel_text_rw(); + set_all_modules_text_rw(); + return 0; +} + +int ftrace_arch_code_modify_post_process(void) +{ + set_all_modules_text_ro(); + set_kernel_text_ro(); + return 0; +} + static unsigned long ftrace_call_replace(unsigned long pc, unsigned long addr) { return arm_gen_branch_link(pc, addr); diff --git a/arch/arm/kernel/kgdb.c b/arch/arm/kernel/kgdb.c index 778c2f7024ff..b321c8fbb87d 100644 --- a/arch/arm/kernel/kgdb.c +++ b/arch/arm/kernel/kgdb.c @@ -144,6 +144,8 @@ 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; @@ -151,6 +153,8 @@ 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/process.c b/arch/arm/kernel/process.c index ac4c2e5e17e4..978002e5b406 100644 --- a/arch/arm/kernel/process.c +++ b/arch/arm/kernel/process.c @@ -32,6 +32,7 @@ #include <linux/hw_breakpoint.h> #include <linux/cpuidle.h> #include <linux/leds.h> +#include <linux/console.h> #include <asm/cacheflush.h> #include <asm/idmap.h> @@ -57,9 +58,46 @@ static const char *isa_modes[] = { "ARM" , "Thumb" , "Jazelle", "ThumbEE" }; +#ifdef CONFIG_SMP +void arch_trigger_all_cpu_backtrace(void) +{ + smp_send_all_cpu_backtrace(); +} +#else +void arch_trigger_all_cpu_backtrace(void) +{ + dump_stack(); +} +#endif + extern void call_with_stack(void (*fn)(void *), void *arg, void *sp); typedef void (*phys_reset_t)(unsigned long); +#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 + /* * A temporary stack to use for CPU reset. This is static so that we * don't clobber it with the identity mapping. When running with this @@ -147,6 +185,7 @@ 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(); @@ -156,6 +195,7 @@ void arch_cpu_idle_enter(void) void arch_cpu_idle_exit(void) { ledtrig_cpu(CPU_LED_IDLE_END); + idle_notifier_call_chain(IDLE_END); } #ifdef CONFIG_HOTPLUG_CPU @@ -195,6 +235,16 @@ __setup("reboot=", reboot_setup); */ void machine_shutdown(void) { +#ifdef CONFIG_SMP + /* + * Disable preemption so we're guaranteed to + * run to power off or reboot and prevent + * the possibility of switching to another + * thread that might wind up blocking on + * one of the stopped CPUs. + */ + preempt_disable(); +#endif disable_nonboot_cpus(); } @@ -240,6 +290,10 @@ void machine_restart(char *cmd) { smp_send_stop(); + /* Flush the console to make sure all the relevant messages make it + * out to the console drivers */ + arm_machine_flush_console(); + arm_pm_restart(reboot_mode, cmd); /* Give a grace period for failure to restart of 1s */ @@ -251,6 +305,77 @@ 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; + + 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) { unsigned long flags; @@ -307,6 +432,8 @@ 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/ptrace.c b/arch/arm/kernel/ptrace.c index 03deeffd9f6d..394424b25254 100644 --- a/arch/arm/kernel/ptrace.c +++ b/arch/arm/kernel/ptrace.c @@ -916,7 +916,7 @@ enum ptrace_syscall_dir { PTRACE_SYSCALL_EXIT, }; -static int tracehook_report_syscall(struct pt_regs *regs, +static void tracehook_report_syscall(struct pt_regs *regs, enum ptrace_syscall_dir dir) { unsigned long ip; @@ -934,7 +934,6 @@ static int tracehook_report_syscall(struct pt_regs *regs, current_thread_info()->syscall = -1; regs->ARM_ip = ip; - return current_thread_info()->syscall; } asmlinkage int syscall_trace_enter(struct pt_regs *regs, int scno) @@ -946,7 +945,9 @@ asmlinkage int syscall_trace_enter(struct pt_regs *regs, int scno) return -1; if (test_thread_flag(TIF_SYSCALL_TRACE)) - scno = tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER); + tracehook_report_syscall(regs, PTRACE_SYSCALL_ENTER); + + scno = current_thread_info()->syscall; if (test_thread_flag(TIF_SYSCALL_TRACEPOINT)) trace_sys_enter(regs, scno); diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index d6e3d1ca4ddf..f2724e475b96 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -70,6 +70,7 @@ enum ipi_msg_type { IPI_CALL_FUNC_SINGLE, IPI_CPU_STOP, IPI_COMPLETION, + IPI_CPU_BACKTRACE, }; static DECLARE_COMPLETION(cpu_running); @@ -468,6 +469,7 @@ static const char *ipi_types[NR_IPI] = { S(IPI_CALL_FUNC_SINGLE, "Single function call interrupts"), S(IPI_CPU_STOP, "CPU stop interrupts"), S(IPI_COMPLETION, "completion interrupts"), + S(IPI_CPU_BACKTRACE, "CPU backtrace"), }; void show_ipi_list(struct seq_file *p, int prec) @@ -606,6 +608,58 @@ static void ipi_complete(unsigned int cpu) complete(per_cpu(cpu_completion, cpu)); } +static cpumask_t backtrace_mask; +static DEFINE_RAW_SPINLOCK(backtrace_lock); + +/* "in progress" flag of arch_trigger_all_cpu_backtrace */ +static unsigned long backtrace_flag; + +void smp_send_all_cpu_backtrace(void) +{ + unsigned int this_cpu = smp_processor_id(); + int i; + + if (test_and_set_bit(0, &backtrace_flag)) + /* + * If there is already a trigger_all_cpu_backtrace() in progress + * (backtrace_flag == 1), don't output double cpu dump infos. + */ + return; + + cpumask_copy(&backtrace_mask, cpu_online_mask); + cpu_clear(this_cpu, backtrace_mask); + + pr_info("Backtrace for cpu %d (current):\n", this_cpu); + dump_stack(); + + pr_info("\nsending IPI to all other CPUs:\n"); + smp_cross_call(&backtrace_mask, IPI_CPU_BACKTRACE); + + /* Wait for up to 10 seconds for all other CPUs to do the backtrace */ + for (i = 0; i < 10 * 1000; i++) { + if (cpumask_empty(&backtrace_mask)) + break; + mdelay(1); + } + + clear_bit(0, &backtrace_flag); + smp_mb__after_clear_bit(); +} + +/* + * ipi_cpu_backtrace - handle IPI from smp_send_all_cpu_backtrace() + */ +static void ipi_cpu_backtrace(unsigned int cpu, struct pt_regs *regs) +{ + if (cpu_isset(cpu, backtrace_mask)) { + raw_spin_lock(&backtrace_lock); + pr_warning("IPI backtrace for cpu %d\n", cpu); + show_regs(regs); + raw_spin_unlock(&backtrace_lock); + cpu_clear(cpu, backtrace_mask); + } +} + /* * Main handler for inter-processor interrupts */ @@ -663,6 +717,10 @@ void handle_IPI(int ipinr, struct pt_regs *regs) irq_exit(); break; + case IPI_CPU_BACKTRACE: + ipi_cpu_backtrace(cpu, regs); + break; + default: printk(KERN_CRIT "CPU%u: Unknown IPI message 0x%x\n", cpu, ipinr); diff --git a/arch/arm/mm/Makefile b/arch/arm/mm/Makefile index 224a9cc09877..2d861a210ac5 100644 --- a/arch/arm/mm/Makefile +++ b/arch/arm/mm/Makefile @@ -7,6 +7,7 @@ obj-y := dma-mapping.o extable.o fault.o init.o \ obj-$(CONFIG_MMU) += fault-armv.o flush.o idmap.o ioremap.o \ mmap.o pgd.o mmu.o +obj-$(CONFIG_DEBUG_RODATA) += rodata.o ifneq ($(CONFIG_MMU),y) obj-y += nommu.o diff --git a/arch/arm/mm/cache-l2x0.c b/arch/arm/mm/cache-l2x0.c index c465faca51b0..90a130f98acf 100644 --- a/arch/arm/mm/cache-l2x0.c +++ b/arch/arm/mm/cache-l2x0.c @@ -33,6 +33,9 @@ static void __iomem *l2x0_base; static DEFINE_RAW_SPINLOCK(l2x0_lock); static u32 l2x0_way_mask; /* Bitmask of active ways */ static u32 l2x0_size; +static u32 l2x0_cache_id; +static unsigned int l2x0_sets; +static unsigned int l2x0_ways; static unsigned long sync_reg_offset = L2X0_CACHE_SYNC; /* Aurora don't have the cache ID register available, so we have to @@ -49,6 +52,13 @@ struct l2x0_of_data { static bool of_init = false; +static inline bool is_pl310_rev(int rev) +{ + return (l2x0_cache_id & + (L2X0_CACHE_ID_PART_MASK | L2X0_CACHE_ID_REV_MASK)) == + (L2X0_CACHE_ID_PART_L310 | rev); +} + static inline void cache_wait_way(void __iomem *reg, unsigned long mask) { /* wait for cache operation by line or way to complete */ @@ -137,6 +147,23 @@ static void l2x0_cache_sync(void) raw_spin_unlock_irqrestore(&l2x0_lock, flags); } +#ifdef CONFIG_PL310_ERRATA_727915 +static void l2x0_for_each_set_way(void __iomem *reg) +{ + int set; + int way; + unsigned long flags; + + for (way = 0; way < l2x0_ways; way++) { + raw_spin_lock_irqsave(&l2x0_lock, flags); + for (set = 0; set < l2x0_sets; set++) + writel_relaxed((way << 28) | (set << 5), reg); + cache_sync(); + raw_spin_unlock_irqrestore(&l2x0_lock, flags); + } +} +#endif + static void __l2x0_flush_all(void) { debug_writel(0x03); @@ -150,6 +177,13 @@ static void l2x0_flush_all(void) { unsigned long flags; +#ifdef CONFIG_PL310_ERRATA_727915 + if (is_pl310_rev(REV_PL310_R2P0)) { + l2x0_for_each_set_way(l2x0_base + L2X0_CLEAN_INV_LINE_IDX); + return; + } +#endif + /* clean all ways */ raw_spin_lock_irqsave(&l2x0_lock, flags); __l2x0_flush_all(); @@ -160,11 +194,20 @@ static void l2x0_clean_all(void) { unsigned long flags; +#ifdef CONFIG_PL310_ERRATA_727915 + if (is_pl310_rev(REV_PL310_R2P0)) { + l2x0_for_each_set_way(l2x0_base + L2X0_CLEAN_LINE_IDX); + return; + } +#endif + /* clean all ways */ raw_spin_lock_irqsave(&l2x0_lock, flags); + debug_writel(0x03); writel_relaxed(l2x0_way_mask, l2x0_base + L2X0_CLEAN_WAY); cache_wait_way(l2x0_base + L2X0_CLEAN_WAY, l2x0_way_mask); cache_sync(); + debug_writel(0x00); raw_spin_unlock_irqrestore(&l2x0_lock, flags); } @@ -323,65 +366,64 @@ static void l2x0_unlock(u32 cache_id) void __init l2x0_init(void __iomem *base, u32 aux_val, u32 aux_mask) { u32 aux; - u32 cache_id; u32 way_size = 0; - int ways; int way_size_shift = L2X0_WAY_SIZE_SHIFT; const char *type; l2x0_base = base; if (cache_id_part_number_from_dt) - cache_id = cache_id_part_number_from_dt; + l2x0_cache_id = cache_id_part_number_from_dt; else - cache_id = readl_relaxed(l2x0_base + L2X0_CACHE_ID); + l2x0_cache_id = readl_relaxed(l2x0_base + L2X0_CACHE_ID); aux = readl_relaxed(l2x0_base + L2X0_AUX_CTRL); aux &= aux_mask; aux |= aux_val; /* Determine the number of ways */ - switch (cache_id & L2X0_CACHE_ID_PART_MASK) { + switch (l2x0_cache_id & L2X0_CACHE_ID_PART_MASK) { case L2X0_CACHE_ID_PART_L310: if (aux & (1 << 16)) - ways = 16; + l2x0_ways = 16; else - ways = 8; + l2x0_ways = 8; type = "L310"; #ifdef CONFIG_PL310_ERRATA_753970 /* Unmapped register. */ sync_reg_offset = L2X0_DUMMY_REG; #endif - if ((cache_id & L2X0_CACHE_ID_RTL_MASK) <= L2X0_CACHE_ID_RTL_R3P0) + if ((l2x0_cache_id & L2X0_CACHE_ID_RTL_MASK) <= L2X0_CACHE_ID_RTL_R3P0) outer_cache.set_debug = pl310_set_debug; break; case L2X0_CACHE_ID_PART_L210: - ways = (aux >> 13) & 0xf; + l2x0_ways = (aux >> 13) & 0xf; type = "L210"; break; case AURORA_CACHE_ID: sync_reg_offset = AURORA_SYNC_REG; - ways = (aux >> 13) & 0xf; - ways = 2 << ((ways + 1) >> 2); + l2x0_ways = (aux >> 13) & 0xf; + l2x0_ways = 2 << ((l2x0_ways + 1) >> 2); way_size_shift = AURORA_WAY_SIZE_SHIFT; type = "Aurora"; break; default: /* Assume unknown chips have 8 ways */ - ways = 8; + l2x0_ways = 8; type = "L2x0 series"; break; } - l2x0_way_mask = (1 << ways) - 1; + l2x0_way_mask = (1 << l2x0_ways) - 1; /* * L2 cache Size = Way size * Number of ways */ way_size = (aux & L2X0_AUX_CTRL_WAY_SIZE_MASK) >> 17; - way_size = 1 << (way_size + way_size_shift); + way_size = SZ_1K << (way_size + way_size_shift); - l2x0_size = ways * way_size * SZ_1K; + l2x0_size = l2x0_ways * way_size; + l2x0_sets = way_size / CACHE_LINE_SIZE; /* * Check if l2x0 controller is already enabled. @@ -390,7 +432,7 @@ void __init l2x0_init(void __iomem *base, u32 aux_val, u32 aux_mask) */ if (!(readl_relaxed(l2x0_base + L2X0_CTRL) & L2X0_CTRL_EN)) { /* Make sure that I&D is not locked down when starting */ - l2x0_unlock(cache_id); + l2x0_unlock(l2x0_cache_id); /* l2x0 controller is disabled */ writel_relaxed(aux, l2x0_base + L2X0_AUX_CTRL); @@ -419,7 +461,7 @@ void __init l2x0_init(void __iomem *base, u32 aux_val, u32 aux_mask) printk(KERN_INFO "%s cache controller enabled\n", type); printk(KERN_INFO "l2x0: %d ways, CACHE_ID 0x%08x, AUX_CTRL 0x%08x, Cache size: %d B\n", - ways, cache_id, aux, l2x0_size); + l2x0_ways, l2x0_cache_id, aux, l2x0_size); } #ifdef CONFIG_OF diff --git a/arch/arm/mm/cache-v6.S b/arch/arm/mm/cache-v6.S index d8fd4d4bd3d4..7a3d3d8d98d7 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 mov pc, 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 54fcddafec15..9820ad4b80c0 100644 --- a/arch/arm/mm/fault.c +++ b/arch/arm/mm/fault.c @@ -274,10 +274,10 @@ 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 (in_atomic() || !mm) + if (in_atomic() || irqs_disabled() || !mm) goto no_context; if (user_mode(regs)) diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c index 4c7d5cddef35..a4e5cd5c3067 100644 --- a/arch/arm/mm/mmu.c +++ b/arch/arm/mm/mmu.c @@ -605,11 +605,25 @@ static void __init *early_alloc(unsigned long sz) return early_alloc_aligned(sz, sz); } -static pte_t * __init early_pte_alloc(pmd_t *pmd, unsigned long addr, unsigned long prot) +static pte_t * __init early_pte_alloc(pmd_t *pmd) +{ + if (pmd_none(*pmd) || pmd_bad(*pmd)) + return early_alloc(PTE_HWTABLE_OFF + PTE_HWTABLE_SIZE); + return pmd_page_vaddr(*pmd); +} + +static void __init early_pte_install(pmd_t *pmd, pte_t *pte, unsigned long prot) +{ + __pmd_populate(pmd, __pa(pte), prot); + BUG_ON(pmd_bad(*pmd)); +} + +static pte_t * __init early_pte_alloc_and_install(pmd_t *pmd, + unsigned long addr, unsigned long prot) { if (pmd_none(*pmd)) { - pte_t *pte = early_alloc(PTE_HWTABLE_OFF + PTE_HWTABLE_SIZE); - __pmd_populate(pmd, __pa(pte), prot); + pte_t *pte = early_pte_alloc(pmd); + early_pte_install(pmd, pte, prot); } BUG_ON(pmd_bad(*pmd)); return pte_offset_kernel(pmd, addr); @@ -619,11 +633,17 @@ static void __init alloc_init_pte(pmd_t *pmd, unsigned long addr, unsigned long end, unsigned long pfn, const struct mem_type *type) { - pte_t *pte = early_pte_alloc(pmd, addr, type->prot_l1); + pte_t *start_pte = early_pte_alloc(pmd); + pte_t *pte = start_pte + pte_index(addr); + + /* If replacing a section mapping, the whole section must be replaced */ + BUG_ON(!pmd_none(*pmd) && pmd_bad(*pmd) && ((addr | end) & ~PMD_MASK)); + do { set_pte_ext(pte, pfn_pte(pfn, __pgprot(type->prot_pte)), 0); pfn++; } while (pte++, addr += PAGE_SIZE, addr != end); + early_pte_install(pmd, start_pte, type->prot_l1); } static void __init __map_init_section(pmd_t *pmd, unsigned long addr, @@ -655,7 +675,8 @@ static void __init __map_init_section(pmd_t *pmd, unsigned long addr, static void __init alloc_init_pmd(pud_t *pud, unsigned long addr, unsigned long end, phys_addr_t phys, - const struct mem_type *type) + const struct mem_type *type, + bool force_pages) { pmd_t *pmd = pmd_offset(pud, addr); unsigned long next; @@ -672,7 +693,8 @@ static void __init alloc_init_pmd(pud_t *pud, unsigned long addr, * aligned to a section boundary. */ if (type->prot_sect && - ((addr | next | phys) & ~SECTION_MASK) == 0) { + ((addr | next | phys) & ~SECTION_MASK) == 0 && + !force_pages) { __map_init_section(pmd, addr, next, phys, type); } else { alloc_init_pte(pmd, addr, next, @@ -686,14 +708,15 @@ static void __init alloc_init_pmd(pud_t *pud, unsigned long addr, static void __init alloc_init_pud(pgd_t *pgd, unsigned long addr, unsigned long end, phys_addr_t phys, - const struct mem_type *type) + const struct mem_type *type, + bool force_pages) { pud_t *pud = pud_offset(pgd, addr); unsigned long next; do { next = pud_addr_end(addr, end); - alloc_init_pmd(pud, addr, next, phys, type); + alloc_init_pmd(pud, addr, next, phys, type, force_pages); phys += next - addr; } while (pud++, addr = next, addr != end); } @@ -767,7 +790,7 @@ static void __init create_36bit_mapping(struct map_desc *md, * offsets, and we take full advantage of sections and * supersections. */ -static void __init create_mapping(struct map_desc *md) +static void __init create_mapping(struct map_desc *md, bool force_pages) { unsigned long addr, length, end; phys_addr_t phys; @@ -817,7 +840,7 @@ static void __init create_mapping(struct map_desc *md) do { unsigned long next = pgd_addr_end(addr, end); - alloc_init_pud(pgd, addr, next, phys, type); + alloc_init_pud(pgd, addr, next, phys, type, force_pages); phys += next - addr; addr = next; @@ -839,7 +862,7 @@ void __init iotable_init(struct map_desc *io_desc, int nr) svm = early_alloc_aligned(sizeof(*svm) * nr, __alignof__(*svm)); for (md = io_desc; nr; md++, nr--) { - create_mapping(md); + create_mapping(md, false); vm = &svm->vm; vm->addr = (void *)(md->virtual & PAGE_MASK); @@ -960,7 +983,7 @@ void __init debug_ll_io_init(void) map.virtual &= PAGE_MASK; map.length = PAGE_SIZE; map.type = MT_DEVICE; - create_mapping(&map); + create_mapping(&map, false); } #endif @@ -1005,6 +1028,28 @@ void __init sanity_check_meminfo(void) struct membank *bank = &meminfo.bank[j]; *bank = meminfo.bank[i]; +#ifdef CONFIG_SPARSEMEM + if (pfn_to_section_nr(bank_pfn_start(bank)) != + pfn_to_section_nr(bank_pfn_end(bank) - 1)) { + phys_addr_t sz; + unsigned long start_pfn = bank_pfn_start(bank); + unsigned long end_pfn = SECTION_ALIGN_UP(start_pfn + 1); + sz = ((phys_addr_t)(end_pfn - start_pfn) << PAGE_SHIFT); + + if (meminfo.nr_banks >= NR_BANKS) { + pr_crit("NR_BANKS too low, ignoring %lld bytes of memory\n", + (unsigned long long)(bank->size - sz)); + } else { + memmove(bank + 1, bank, + (meminfo.nr_banks - i) * sizeof(*bank)); + meminfo.nr_banks++; + bank[1].size -= sz; + bank[1].start = __pfn_to_phys(end_pfn); + } + bank->size = sz; + } +#endif + if (bank->start > ULONG_MAX) highmem = 1; @@ -1202,7 +1247,7 @@ static void __init devicemaps_init(struct machine_desc *mdesc) map.virtual = MODULES_VADDR; map.length = ((unsigned long)_etext - map.virtual + ~SECTION_MASK) & SECTION_MASK; map.type = MT_ROM; - create_mapping(&map); + create_mapping(&map, false); #endif /* @@ -1213,14 +1258,14 @@ static void __init devicemaps_init(struct machine_desc *mdesc) map.virtual = FLUSH_BASE; map.length = SZ_1M; map.type = MT_CACHECLEAN; - create_mapping(&map); + create_mapping(&map, false); #endif #ifdef FLUSH_BASE_MINICACHE map.pfn = __phys_to_pfn(FLUSH_BASE_PHYS + SZ_1M); map.virtual = FLUSH_BASE_MINICACHE; map.length = SZ_1M; map.type = MT_MINICLEAN; - create_mapping(&map); + create_mapping(&map, false); #endif /* @@ -1236,13 +1281,13 @@ static void __init devicemaps_init(struct machine_desc *mdesc) #else map.type = MT_LOW_VECTORS; #endif - create_mapping(&map); + create_mapping(&map, false); if (!vectors_high()) { map.virtual = 0; map.length = PAGE_SIZE * 2; map.type = MT_LOW_VECTORS; - create_mapping(&map); + create_mapping(&map, false); } /* Now create a kernel read-only mapping */ @@ -1250,7 +1295,7 @@ static void __init devicemaps_init(struct machine_desc *mdesc) map.virtual = 0xffff0000 + PAGE_SIZE; map.length = PAGE_SIZE; map.type = MT_LOW_VECTORS; - create_mapping(&map); + create_mapping(&map, false); /* * Ask the machine support to map in the statically mapped devices. @@ -1275,20 +1320,23 @@ static void __init devicemaps_init(struct machine_desc *mdesc) static void __init kmap_init(void) { #ifdef CONFIG_HIGHMEM - pkmap_page_table = early_pte_alloc(pmd_off_k(PKMAP_BASE), + pkmap_page_table = early_pte_alloc_and_install(pmd_off_k(PKMAP_BASE), PKMAP_BASE, _PAGE_KERNEL_TABLE); #endif } + static void __init map_lowmem(void) { struct memblock_region *reg; + phys_addr_t start; + phys_addr_t end; + struct map_desc map; /* Map all the lowmem memory banks. */ for_each_memblock(memory, reg) { - phys_addr_t start = reg->base; - phys_addr_t end = start + reg->size; - struct map_desc map; + start = reg->base; + end = start + reg->size; if (end > arm_lowmem_limit) end = arm_lowmem_limit; @@ -1300,8 +1348,20 @@ static void __init map_lowmem(void) map.length = end - start; map.type = MT_MEMORY; - create_mapping(&map); + create_mapping(&map, false); } + +#ifdef CONFIG_DEBUG_RODATA + start = __pa(_stext) & PMD_MASK; + end = ALIGN(__pa(__end_rodata), PMD_SIZE); + + map.pfn = __phys_to_pfn(start); + map.virtual = __phys_to_virt(start); + map.length = end - start; + map.type = MT_MEMORY; + + create_mapping(&map, true); +#endif } /* diff --git a/arch/arm/mm/rodata.c b/arch/arm/mm/rodata.c new file mode 100644 index 000000000000..9a8eb841c428 --- /dev/null +++ b/arch/arm/mm/rodata.c @@ -0,0 +1,159 @@ +/* + * linux/arch/arm/mm/rodata.c + * + * Copyright (C) 2011 Google, Inc. + * + * Author: Colin Cross <ccross@android.com> + * + * Based on x86 implementation in arch/x86/mm/init_32.c + * + * 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/kernel.h> +#include <linux/mm.h> +#include <linux/module.h> + +#include <asm/cache.h> +#include <asm/pgtable.h> +#include <asm/rodata.h> +#include <asm/sections.h> +#include <asm/tlbflush.h> + +#include "mm.h" + +static int kernel_set_to_readonly __read_mostly; + +#ifdef CONFIG_DEBUG_RODATA_TEST +static const int rodata_test_data = 0xC3; + +static noinline void rodata_test(void) +{ + int result; + + pr_info("%s: attempting to write to read-only section:\n", __func__); + + if (*(volatile int *)&rodata_test_data != 0xC3) { + pr_err("read only data changed before test\n"); + return; + } + + /* + * Attempt to to write to rodata_test_data, trapping the expected + * data abort. If the trap executed, result will be 1. If it didn't, + * result will be 0xFF. + */ + asm volatile( + "0: str %[zero], [%[rodata_test_data]]\n" + " mov %[result], #0xFF\n" + " b 2f\n" + "1: mov %[result], #1\n" + "2:\n" + + /* Exception fixup - if store at label 0 faults, jumps to 1 */ + ".pushsection __ex_table, \"a\"\n" + " .long 0b, 1b\n" + ".popsection\n" + + : [result] "=r" (result) + : [rodata_test_data] "r" (&rodata_test_data), [zero] "r" (0) + : "memory" + ); + + if (result == 1) + pr_info("write to read-only section trapped, success\n"); + else + pr_err("write to read-only section NOT trapped, test failed\n"); + + if (*(volatile int *)&rodata_test_data != 0xC3) + pr_err("read only data changed during write\n"); +} +#else +static inline void rodata_test(void) { } +#endif + +static int set_page_attributes(unsigned long virt, int numpages, + pte_t (*f)(pte_t)) +{ + pmd_t *pmd; + pte_t *pte; + unsigned long start = virt; + unsigned long end = virt + (numpages << PAGE_SHIFT); + unsigned long pmd_end; + + while (virt < end) { + pmd = pmd_off_k(virt); + pmd_end = min(ALIGN(virt + 1, PMD_SIZE), end); + + if ((pmd_val(*pmd) & PMD_TYPE_MASK) != PMD_TYPE_TABLE) { + pr_err("%s: pmd %p=%08lx for %08lx not page table\n", + __func__, pmd, pmd_val(*pmd), virt); + virt = pmd_end; + continue; + } + + while (virt < pmd_end) { + pte = pte_offset_kernel(pmd, virt); + set_pte_ext(pte, f(*pte), 0); + virt += PAGE_SIZE; + } + } + + flush_tlb_kernel_range(start, end); + + return 0; +} + +int set_memory_ro(unsigned long virt, int numpages) +{ + return set_page_attributes(virt, numpages, pte_wrprotect); +} +EXPORT_SYMBOL(set_memory_ro); + +int set_memory_rw(unsigned long virt, int numpages) +{ + return set_page_attributes(virt, numpages, pte_mkwrite); +} +EXPORT_SYMBOL(set_memory_rw); + +void set_kernel_text_rw(void) +{ + unsigned long start = PAGE_ALIGN((unsigned long)_text); + unsigned long size = PAGE_ALIGN((unsigned long)__end_rodata) - start; + + if (!kernel_set_to_readonly) + return; + + pr_debug("Set kernel text: %lx - %lx to read-write\n", + start, start + size); + + set_memory_rw(start, size >> PAGE_SHIFT); +} + +void set_kernel_text_ro(void) +{ + unsigned long start = PAGE_ALIGN((unsigned long)_text); + unsigned long size = PAGE_ALIGN((unsigned long)__end_rodata) - start; + + if (!kernel_set_to_readonly) + return; + + pr_info_once("Write protecting the kernel text section %lx - %lx\n", + start, start + size); + + pr_debug("Set kernel text: %lx - %lx to read only\n", + start, start + size); + + set_memory_ro(start, size >> PAGE_SHIFT); +} + +void mark_rodata_ro(void) +{ + kernel_set_to_readonly = 1; + + set_kernel_text_ro(); + + rodata_test(); +} diff --git a/arch/arm/vfp/vfpmodule.c b/arch/arm/vfp/vfpmodule.c index 5dfbb0b8e7f4..452fb3ad68aa 100644 --- a/arch/arm/vfp/vfpmodule.c +++ b/arch/arm/vfp/vfpmodule.c @@ -20,6 +20,7 @@ #include <linux/init.h> #include <linux/uaccess.h> #include <linux/user.h> +#include <linux/export.h> #include <asm/cp15.h> #include <asm/cputype.h> @@ -648,6 +649,52 @@ static int vfp_hotplug(struct notifier_block *b, unsigned long action, return NOTIFY_OK; } +#ifdef CONFIG_KERNEL_MODE_NEON + +/* + * Kernel-side NEON support functions + */ +void kernel_neon_begin(void) +{ + struct thread_info *thread = current_thread_info(); + unsigned int cpu; + u32 fpexc; + + /* + * Kernel mode NEON is only allowed outside of interrupt context + * with preemption disabled. This will make sure that the kernel + * mode NEON register contents never need to be preserved. + */ + BUG_ON(in_interrupt()); + cpu = get_cpu(); + + fpexc = fmrx(FPEXC) | FPEXC_EN; + fmxr(FPEXC, fpexc); + + /* + * Save the userland NEON/VFP state. Under UP, + * the owner could be a task other than 'current' + */ + if (vfp_state_in_hw(cpu, thread)) + vfp_save_state(&thread->vfpstate, fpexc); +#ifndef CONFIG_SMP + else if (vfp_current_hw_state[cpu] != NULL) + vfp_save_state(vfp_current_hw_state[cpu], fpexc); +#endif + vfp_current_hw_state[cpu] = NULL; +} +EXPORT_SYMBOL(kernel_neon_begin); + +void kernel_neon_end(void) +{ + /* Disable the NEON/VFP unit. */ + fmxr(FPEXC, fmrx(FPEXC) & ~FPEXC_EN); + put_cpu(); +} +EXPORT_SYMBOL(kernel_neon_end); + +#endif /* CONFIG_KERNEL_MODE_NEON */ + /* * VFP support code initialisation. */ @@ -731,4 +778,4 @@ static int __init vfp_init(void) return 0; } -late_initcall(vfp_init); +core_initcall(vfp_init); diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 9742106cb51c..58ed097828b6 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -15,6 +15,7 @@ config ARM64 select ARM_GIC select ARM_GIC_V3 select BUILDTIME_EXTABLE_SORT + select AUDIT_ARCH_COMPAT_GENERIC select CLONE_BACKWARDS select COMMON_CLK select CPU_PM if (SUSPEND || CPU_IDLE) @@ -34,6 +35,8 @@ config ARM64 select HARDIRQS_SW_RESEND select HAVE_ARCH_JUMP_LABEL select HAVE_ARCH_KGDB + select HAVE_ARCH_AUDITSYSCALL + select HAVE_ARCH_SECCOMP_FILTER select HAVE_ARCH_TRACEHOOK select HAVE_C_RECORDMCOUNT select HAVE_CC_STACKPROTECTOR @@ -336,6 +339,31 @@ config HOTPLUG_CPU Say Y here to experiment with turning CPUs off and on. CPUs can be controlled through /sys/devices/system/cpu. +config SWP_EMULATE + bool "Emulate SWP/SWPB instructions" + help + ARMv6 architecture deprecates use of the SWP/SWPB instructions. ARMv8 + oblosetes the use of SWP/SWPB instructions. ARMv7 multiprocessing + extensions introduce the ability to disable these instructions, + triggering an undefined instruction exception when executed. Say Y + here to enable software emulation of these instructions for userspace + (not kernel) using LDREX/STREX. Also creates /proc/cpu/swp_emulation + for statistics. + + In some older versions of glibc [<=2.8] SWP is used during futex + trylock() operations with the assumption that the code will not + be preempted. This invalid assumption may be more likely to fail + with SWP emulation enabled, leading to deadlock of the user + application. + + NOTE: when accessing uncached shared regions, LDREX/STREX rely + on an external transaction monitoring block called a global + monitor to maintain update atomicity. If your system does not + implement a global monitor, this option can cause programs that + perform SWP operations to uncached memory to deadlock. + + If unsure, say Y. + source kernel/Kconfig.preempt config HZ @@ -378,6 +406,27 @@ config ARCH_WANT_HUGE_PMD_SHARE config HAVE_ARCH_TRANSPARENT_HUGEPAGE def_bool y +config ARMV7_COMPAT + bool "Kernel support for ARMv7 applications" + depends on COMPAT + select SWP_EMULATE + help + This option enables features that allow that ran on an ARMv7 or older + processor to continue functioning. + + If you want to execute ARMv7 applications, say Y + +config ARMV7_COMPAT_CPUINFO + bool "Report backwards compatible cpu features in /proc/cpuinfo" + depends on ARMV7_COMPAT + default y + help + This option makes /proc/cpuinfo list CPU features that an ARMv7 or + earlier kernel would report, but are not optional on an ARMv8 or later + processor. + + If you want to execute ARMv7 applications, say Y + source "mm/Kconfig" config FORCE_MAX_ZONEORDER @@ -385,6 +434,19 @@ config FORCE_MAX_ZONEORDER default "14" if (ARM64_64K_PAGES && TRANSPARENT_HUGEPAGE) default "11" +config SECCOMP + bool "Enable seccomp to safely compute untrusted bytecode" + ---help--- + This kernel feature is useful for number crunching applications + that may need to compute untrusted bytecode during their + execution. By using pipes or other transports made available to + the process as file descriptors supporting the read/write + syscalls, it's possible to isolate those applications in + their own address space using seccomp. Once seccomp is + enabled via prctl(PR_SET_SECCOMP), it cannot be disabled + and the task is only allowed to execute a few safe syscalls + defined by each seccomp mode. + endmenu menu "Boot options" @@ -397,6 +459,23 @@ 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 @@ -404,6 +483,22 @@ 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 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 + +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. config EFI bool "UEFI runtime support" diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile index 3e7882c4e034..9e64836cb256 100644 --- a/arch/arm64/Makefile +++ b/arch/arm64/Makefile @@ -20,6 +20,7 @@ LIBGCC := $(shell $(CC) $(KBUILD_CFLAGS) -print-libgcc-file-name) KBUILD_DEFCONFIG := defconfig KBUILD_CFLAGS += -mgeneral-regs-only +KBUILD_CFLAGS += -fno-pic ifeq ($(CONFIG_CPU_BIG_ENDIAN), y) KBUILD_CPPFLAGS += -mbig-endian AS += -EB @@ -53,7 +54,12 @@ libs-y := arch/arm64/lib/ $(libs-y) libs-y += $(LIBGCC) # Default target when executing plain make +ifeq ($(CONFIG_BUILD_ARM64_APPENDED_DTB_IMAGE),y) +KBUILD_IMAGE := Image.gz-dtb +else KBUILD_IMAGE := Image.gz +endif + KBUILD_DTBS := dtbs all: $(KBUILD_IMAGE) $(KBUILD_DTBS) @@ -72,6 +78,9 @@ zinstall install: vmlinux dtbs: scripts $(Q)$(MAKE) $(build)=$(boot)/dts dtbs +Image.gz-dtb: vmlinux scripts dtbs + $(Q)$(MAKE) $(build)=$(boot) $(boot)/$@ + PHONY += vdso_install vdso_install: $(Q)$(MAKE) $(build)=arch/arm64/kernel/vdso $@ diff --git a/arch/arm64/boot/.gitignore b/arch/arm64/boot/.gitignore index 8dab0bb6ae66..eb3551131b1e 100644 --- a/arch/arm64/boot/.gitignore +++ b/arch/arm64/boot/.gitignore @@ -1,2 +1,3 @@ Image Image.gz +Image.gz-dtb diff --git a/arch/arm64/boot/Makefile b/arch/arm64/boot/Makefile index 5a0e3ab854a5..df519849fa00 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.gz: $(obj)/Image FORCE $(call if_changed,gzip) +$(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 ef388176116d..661015fd7748 100644 --- a/arch/arm64/boot/dts/Makefile +++ b/arch/arm64/boot/dts/Makefile @@ -4,8 +4,15 @@ dtb-$(CONFIG_ARCH_VEXPRESS) += juno.dtb dtb-$(CONFIG_ARCH_XGENE) += apm-mustang.dtb targets += dtbs -targets += $(dtb-y) -dtbs: $(addprefix $(obj)/, $(dtb-y)) +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) -clean-files := *.dtb +dtbs: $(addprefix $(obj)/, $(DTB_LIST)) + +clean-files := dts/*.dtb *.dtb diff --git a/arch/arm64/crypto/Makefile b/arch/arm64/crypto/Makefile index 2070a56ecc46..a3f935fde975 100644 --- a/arch/arm64/crypto/Makefile +++ b/arch/arm64/crypto/Makefile @@ -35,4 +35,4 @@ AFLAGS_aes-neon.o := -DINTERLEAVE=4 CFLAGS_aes-glue-ce.o := -DUSE_V8_CRYPTO_EXTENSIONS $(obj)/aes-glue-%.o: $(src)/aes-glue.c FORCE - $(call if_changed_dep,cc_o_c) + $(call if_changed_rule,cc_o_c) diff --git a/arch/arm64/crypto/aes-glue.c b/arch/arm64/crypto/aes-glue.c index 60f2f4c12256..79cd911ef88c 100644 --- a/arch/arm64/crypto/aes-glue.c +++ b/arch/arm64/crypto/aes-glue.c @@ -106,7 +106,7 @@ static int ecb_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, for (first = 1; (blocks = (walk.nbytes / AES_BLOCK_SIZE)); first = 0) { aes_ecb_encrypt(walk.dst.virt.addr, walk.src.virt.addr, (u8 *)ctx->key_enc, rounds, blocks, first); - err = blkcipher_walk_done(desc, &walk, 0); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); } kernel_neon_end(); return err; @@ -128,7 +128,7 @@ static int ecb_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, for (first = 1; (blocks = (walk.nbytes / AES_BLOCK_SIZE)); first = 0) { aes_ecb_decrypt(walk.dst.virt.addr, walk.src.virt.addr, (u8 *)ctx->key_dec, rounds, blocks, first); - err = blkcipher_walk_done(desc, &walk, 0); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); } kernel_neon_end(); return err; @@ -151,7 +151,7 @@ static int cbc_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, aes_cbc_encrypt(walk.dst.virt.addr, walk.src.virt.addr, (u8 *)ctx->key_enc, rounds, blocks, walk.iv, first); - err = blkcipher_walk_done(desc, &walk, 0); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); } kernel_neon_end(); return err; @@ -174,7 +174,7 @@ static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, aes_cbc_decrypt(walk.dst.virt.addr, walk.src.virt.addr, (u8 *)ctx->key_dec, rounds, blocks, walk.iv, first); - err = blkcipher_walk_done(desc, &walk, 0); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); } kernel_neon_end(); return err; @@ -243,7 +243,7 @@ static int xts_encrypt(struct blkcipher_desc *desc, struct scatterlist *dst, aes_xts_encrypt(walk.dst.virt.addr, walk.src.virt.addr, (u8 *)ctx->key1.key_enc, rounds, blocks, (u8 *)ctx->key2.key_enc, walk.iv, first); - err = blkcipher_walk_done(desc, &walk, 0); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); } kernel_neon_end(); @@ -267,7 +267,7 @@ static int xts_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst, aes_xts_decrypt(walk.dst.virt.addr, walk.src.virt.addr, (u8 *)ctx->key1.key_dec, rounds, blocks, (u8 *)ctx->key2.key_enc, walk.iv, first); - err = blkcipher_walk_done(desc, &walk, 0); + err = blkcipher_walk_done(desc, &walk, walk.nbytes % AES_BLOCK_SIZE); } kernel_neon_end(); diff --git a/arch/arm64/crypto/ghash-ce-core.S b/arch/arm64/crypto/ghash-ce-core.S index b9e6eaf41c9b..607928ad1fdf 100644 --- a/arch/arm64/crypto/ghash-ce-core.S +++ b/arch/arm64/crypto/ghash-ce-core.S @@ -19,13 +19,15 @@ #include <linux/linkage.h> #include <asm/assembler.h> - DATA .req v0 - SHASH .req v1 - IN1 .req v2 + SHASH .req v0 + SHASH2 .req v1 T1 .req v2 T2 .req v3 - T3 .req v4 - VZR .req v5 + MASK .req v4 + XL .req v5 + XM .req v6 + XH .req v7 + IN1 .req v7 .text .arch armv8-a+crypto @@ -35,61 +37,51 @@ * struct ghash_key const *k, const char *head) */ ENTRY(pmull_ghash_update) - ld1 {DATA.16b}, [x1] ld1 {SHASH.16b}, [x3] - eor VZR.16b, VZR.16b, VZR.16b + ld1 {XL.16b}, [x1] + movi MASK.16b, #0xe1 + ext SHASH2.16b, SHASH.16b, SHASH.16b, #8 + shl MASK.2d, MASK.2d, #57 + eor SHASH2.16b, SHASH2.16b, SHASH.16b /* do the head block first, if supplied */ cbz x4, 0f - ld1 {IN1.2d}, [x4] + ld1 {T1.2d}, [x4] b 1f -0: ld1 {IN1.2d}, [x2], #16 +0: ld1 {T1.2d}, [x2], #16 sub w0, w0, #1 -1: ext IN1.16b, IN1.16b, IN1.16b, #8 -CPU_LE( rev64 IN1.16b, IN1.16b ) - eor DATA.16b, DATA.16b, IN1.16b - /* multiply DATA by SHASH in GF(2^128) */ - ext T2.16b, DATA.16b, DATA.16b, #8 - ext T3.16b, SHASH.16b, SHASH.16b, #8 - eor T2.16b, T2.16b, DATA.16b - eor T3.16b, T3.16b, SHASH.16b +1: /* multiply XL by SHASH in GF(2^128) */ +CPU_LE( rev64 T1.16b, T1.16b ) - pmull2 T1.1q, SHASH.2d, DATA.2d // a1 * b1 - pmull DATA.1q, SHASH.1d, DATA.1d // a0 * b0 - pmull T2.1q, T2.1d, T3.1d // (a1 + a0)(b1 + b0) - eor T2.16b, T2.16b, T1.16b // (a0 * b1) + (a1 * b0) - eor T2.16b, T2.16b, DATA.16b + ext T2.16b, XL.16b, XL.16b, #8 + ext IN1.16b, T1.16b, T1.16b, #8 + eor T1.16b, T1.16b, T2.16b + eor XL.16b, XL.16b, IN1.16b - ext T3.16b, VZR.16b, T2.16b, #8 - ext T2.16b, T2.16b, VZR.16b, #8 - eor DATA.16b, DATA.16b, T3.16b - eor T1.16b, T1.16b, T2.16b // <T1:DATA> is result of - // carry-less multiplication + pmull2 XH.1q, SHASH.2d, XL.2d // a1 * b1 + eor T1.16b, T1.16b, XL.16b + pmull XL.1q, SHASH.1d, XL.1d // a0 * b0 + pmull XM.1q, SHASH2.1d, T1.1d // (a1 + a0)(b1 + b0) - /* first phase of the reduction */ - shl T3.2d, DATA.2d, #1 - eor T3.16b, T3.16b, DATA.16b - shl T3.2d, T3.2d, #5 - eor T3.16b, T3.16b, DATA.16b - shl T3.2d, T3.2d, #57 - ext T2.16b, VZR.16b, T3.16b, #8 - ext T3.16b, T3.16b, VZR.16b, #8 - eor DATA.16b, DATA.16b, T2.16b - eor T1.16b, T1.16b, T3.16b + 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 + pmull T2.1q, XL.1d, MASK.1d - /* second phase of the reduction */ - ushr T2.2d, DATA.2d, #5 - eor T2.16b, T2.16b, DATA.16b - ushr T2.2d, T2.2d, #1 - eor T2.16b, T2.16b, DATA.16b - ushr T2.2d, T2.2d, #1 - eor T1.16b, T1.16b, T2.16b - eor DATA.16b, DATA.16b, T1.16b + mov XH.d[0], XM.d[1] + mov XM.d[1], XL.d[0] + + eor XL.16b, XM.16b, T2.16b + ext T2.16b, XL.16b, XL.16b, #8 + pmull XL.1q, XL.1d, MASK.1d + eor T2.16b, T2.16b, XH.16b + eor XL.16b, XL.16b, T2.16b cbnz w0, 0b - st1 {DATA.16b}, [x1] + st1 {XL.16b}, [x1] ret ENDPROC(pmull_ghash_update) diff --git a/arch/arm64/crypto/ghash-ce-glue.c b/arch/arm64/crypto/ghash-ce-glue.c index b92baf3f68c7..833ec1e3f3e9 100644 --- a/arch/arm64/crypto/ghash-ce-glue.c +++ b/arch/arm64/crypto/ghash-ce-glue.c @@ -67,11 +67,12 @@ static int ghash_update(struct shash_desc *desc, const u8 *src, blocks = len / GHASH_BLOCK_SIZE; len %= GHASH_BLOCK_SIZE; - kernel_neon_begin_partial(6); + kernel_neon_begin_partial(8); pmull_ghash_update(blocks, ctx->digest, src, key, partial ? ctx->buf : NULL); kernel_neon_end(); src += blocks * GHASH_BLOCK_SIZE; + partial = 0; } if (len) memcpy(ctx->buf + partial, src, len); @@ -88,7 +89,7 @@ static int ghash_final(struct shash_desc *desc, u8 *dst) memset(ctx->buf + partial, 0, GHASH_BLOCK_SIZE - partial); - kernel_neon_begin_partial(6); + kernel_neon_begin_partial(8); pmull_ghash_update(1, ctx->digest, ctx->buf, key, NULL); kernel_neon_end(); } diff --git a/arch/arm64/include/asm/cmpxchg.h b/arch/arm64/include/asm/cmpxchg.h index a84d4c8acbbe..ddb9d7830558 100644 --- a/arch/arm64/include/asm/cmpxchg.h +++ b/arch/arm64/include/asm/cmpxchg.h @@ -168,8 +168,8 @@ static inline unsigned long __cmpxchg_mb(volatile void *ptr, unsigned long old, ({ \ __typeof__(*(ptr)) __ret; \ __ret = (__typeof__(*(ptr))) \ - __cmpxchg_mb((ptr), (unsigned long)(o), (unsigned long)(n), \ - sizeof(*(ptr))); \ + __cmpxchg_mb((ptr), (unsigned long)(o), (unsigned long)(n), \ + sizeof(*(ptr))); \ __ret; \ }) @@ -177,8 +177,8 @@ static inline unsigned long __cmpxchg_mb(volatile void *ptr, unsigned long old, ({ \ __typeof__(*(ptr)) __ret; \ __ret = (__typeof__(*(ptr))) \ - __cmpxchg((ptr), (unsigned long)(o), \ - (unsigned long)(n), sizeof(*(ptr))); \ + __cmpxchg((ptr), (unsigned long)(o), \ + (unsigned long)(n), sizeof(*(ptr))); \ __ret; \ }) diff --git a/arch/arm64/include/asm/compat.h b/arch/arm64/include/asm/compat.h index 56de5aadede2..e94e8dde78b4 100644 --- a/arch/arm64/include/asm/compat.h +++ b/arch/arm64/include/asm/compat.h @@ -205,6 +205,13 @@ typedef struct compat_siginfo { compat_long_t _band; /* POLL_IN, POLL_OUT, POLL_MSG */ int _fd; } _sigpoll; + + /* SIGSYS */ + struct { + compat_uptr_t _call_addr; /* calling user insn */ + int _syscall; /* triggering system call number */ + unsigned int _arch; /* AUDIT_ARCH_* of syscall */ + } _sigsys; } _sifields; } compat_siginfo_t; diff --git a/arch/arm64/include/asm/fpsimd.h b/arch/arm64/include/asm/fpsimd.h index c43b4ac13008..50f559f574fe 100644 --- a/arch/arm64/include/asm/fpsimd.h +++ b/arch/arm64/include/asm/fpsimd.h @@ -37,8 +37,21 @@ struct fpsimd_state { u32 fpcr; }; }; + /* the id of the last cpu to have restored this state */ + unsigned int cpu; }; +/* + * Struct for stacking the bottom 'n' FP/SIMD registers. + */ +struct fpsimd_partial_state { + u32 fpsr; + u32 fpcr; + u32 num_regs; + __uint128_t vregs[32]; +}; + + #if defined(__KERNEL__) && defined(CONFIG_COMPAT) /* Masks for extracting the FPSR and FPCR from the FPSCR */ #define VFP_FPSCR_STAT_MASK 0xf800009f @@ -58,6 +71,16 @@ extern void fpsimd_load_state(struct fpsimd_state *state); extern void fpsimd_thread_switch(struct task_struct *next); extern void fpsimd_flush_thread(void); +extern void fpsimd_preserve_current_state(void); +extern void fpsimd_restore_current_state(void); +extern void fpsimd_update_current_state(struct fpsimd_state *state); + +extern void fpsimd_flush_task_state(struct task_struct *target); + +extern void fpsimd_save_partial_state(struct fpsimd_partial_state *state, + u32 num_regs); +extern void fpsimd_load_partial_state(struct fpsimd_partial_state *state); + #endif #endif diff --git a/arch/arm64/include/asm/fpsimdmacros.h b/arch/arm64/include/asm/fpsimdmacros.h index bbec599c96bd..768414d55e64 100644 --- a/arch/arm64/include/asm/fpsimdmacros.h +++ b/arch/arm64/include/asm/fpsimdmacros.h @@ -62,3 +62,38 @@ ldr w\tmpnr, [\state, #16 * 2 + 4] msr fpcr, x\tmpnr .endm + +.altmacro +.macro fpsimd_save_partial state, numnr, tmpnr1, tmpnr2 + mrs x\tmpnr1, fpsr + str w\numnr, [\state, #8] + mrs x\tmpnr2, fpcr + stp w\tmpnr1, w\tmpnr2, [\state] + adr x\tmpnr1, 0f + add \state, \state, x\numnr, lsl #4 + sub x\tmpnr1, x\tmpnr1, x\numnr, lsl #1 + br x\tmpnr1 + .irp qa, 30, 28, 26, 24, 22, 20, 18, 16, 14, 12, 10, 8, 6, 4, 2, 0 + .irp qb, %(qa + 1) + stp q\qa, q\qb, [\state, # -16 * \qa - 16] + .endr + .endr +0: +.endm + +.macro fpsimd_restore_partial state, tmpnr1, tmpnr2 + ldp w\tmpnr1, w\tmpnr2, [\state] + msr fpsr, x\tmpnr1 + msr fpcr, x\tmpnr2 + adr x\tmpnr1, 0f + ldr w\tmpnr2, [\state, #8] + add \state, \state, x\tmpnr2, lsl #4 + sub x\tmpnr1, x\tmpnr1, x\tmpnr2, lsl #1 + br x\tmpnr1 + .irp qa, 30, 28, 26, 24, 22, 20, 18, 16, 14, 12, 10, 8, 6, 4, 2, 0 + .irp qb, %(qa + 1) + ldp q\qa, q\qb, [\state, # -16 * \qa - 16] + .endr + .endr +0: +.endm diff --git a/arch/arm64/include/asm/neon.h b/arch/arm64/include/asm/neon.h new file mode 100644 index 000000000000..13ce4cc18e26 --- /dev/null +++ b/arch/arm64/include/asm/neon.h @@ -0,0 +1,18 @@ +/* + * linux/arch/arm64/include/asm/neon.h + * + * Copyright (C) 2013 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 + * published by the Free Software Foundation. + */ + +#include <linux/types.h> + +#define cpu_has_neon() (1) + +#define kernel_neon_begin() kernel_neon_begin_partial(32) + +void kernel_neon_begin_partial(u32 num_regs); +void kernel_neon_end(void); diff --git a/arch/arm64/include/asm/opcodes.h b/arch/arm64/include/asm/opcodes.h new file mode 100644 index 000000000000..fd189a522aee --- /dev/null +++ b/arch/arm64/include/asm/opcodes.h @@ -0,0 +1,231 @@ +/* + * Copied from arch/arm/include/asm/opcodes.h + * + * 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 __ASM_ARM_OPCODES_H +#define __ASM_ARM_OPCODES_H + +#ifndef __ASSEMBLY__ +#include <linux/linkage.h> +extern asmlinkage unsigned int arm_check_condition(u32 opcode, u64 psr); +#endif + +#define ARM_OPCODE_CONDTEST_FAIL 0 +#define ARM_OPCODE_CONDTEST_PASS 1 +#define ARM_OPCODE_CONDTEST_UNCOND 2 + + +/* + * Assembler opcode byteswap helpers. + * These are only intended for use by this header: don't use them directly, + * because they will be suboptimal in most cases. + */ +#define ___asm_opcode_swab32(x) ( \ + (((x) << 24) & 0xFF000000) \ + | (((x) << 8) & 0x00FF0000) \ + | (((x) >> 8) & 0x0000FF00) \ + | (((x) >> 24) & 0x000000FF) \ +) +#define ___asm_opcode_swab16(x) ( \ + (((x) << 8) & 0xFF00) \ + | (((x) >> 8) & 0x00FF) \ +) +#define ___asm_opcode_swahb32(x) ( \ + (((x) << 8) & 0xFF00FF00) \ + | (((x) >> 8) & 0x00FF00FF) \ +) +#define ___asm_opcode_swahw32(x) ( \ + (((x) << 16) & 0xFFFF0000) \ + | (((x) >> 16) & 0x0000FFFF) \ +) +#define ___asm_opcode_identity32(x) ((x) & 0xFFFFFFFF) +#define ___asm_opcode_identity16(x) ((x) & 0xFFFF) + + +/* + * Opcode byteswap helpers + * + * These macros help with converting instructions between a canonical integer + * format and in-memory representation, in an endianness-agnostic manner. + * + * __mem_to_opcode_*() convert from in-memory representation to canonical form. + * __opcode_to_mem_*() convert from canonical form to in-memory representation. + * + * + * Canonical instruction representation: + * + * ARM: 0xKKLLMMNN + * Thumb 16-bit: 0x0000KKLL, where KK < 0xE8 + * Thumb 32-bit: 0xKKLLMMNN, where KK >= 0xE8 + * + * There is no way to distinguish an ARM instruction in canonical representation + * from a Thumb instruction (just as these cannot be distinguished in memory). + * Where this distinction is important, it needs to be tracked separately. + * + * Note that values in the range 0x0000E800..0xE7FFFFFF intentionally do not + * represent any valid Thumb-2 instruction. For this range, + * __opcode_is_thumb32() and __opcode_is_thumb16() will both be false. + * + * The ___asm variants are intended only for use by this header, in situations + * involving inline assembler. For .S files, the normal __opcode_*() macros + * should do the right thing. + */ +#ifdef __ASSEMBLY__ + +#define ___opcode_swab32(x) ___asm_opcode_swab32(x) +#define ___opcode_swab16(x) ___asm_opcode_swab16(x) +#define ___opcode_swahb32(x) ___asm_opcode_swahb32(x) +#define ___opcode_swahw32(x) ___asm_opcode_swahw32(x) +#define ___opcode_identity32(x) ___asm_opcode_identity32(x) +#define ___opcode_identity16(x) ___asm_opcode_identity16(x) + +#else /* ! __ASSEMBLY__ */ + +#include <linux/types.h> +#include <linux/swab.h> + +#define ___opcode_swab32(x) swab32(x) +#define ___opcode_swab16(x) swab16(x) +#define ___opcode_swahb32(x) swahb32(x) +#define ___opcode_swahw32(x) swahw32(x) +#define ___opcode_identity32(x) ((u32)(x)) +#define ___opcode_identity16(x) ((u16)(x)) + +#endif /* ! __ASSEMBLY__ */ + + +#ifdef CONFIG_CPU_ENDIAN_BE8 + +#define __opcode_to_mem_arm(x) ___opcode_swab32(x) +#define __opcode_to_mem_thumb16(x) ___opcode_swab16(x) +#define __opcode_to_mem_thumb32(x) ___opcode_swahb32(x) +#define ___asm_opcode_to_mem_arm(x) ___asm_opcode_swab32(x) +#define ___asm_opcode_to_mem_thumb16(x) ___asm_opcode_swab16(x) +#define ___asm_opcode_to_mem_thumb32(x) ___asm_opcode_swahb32(x) + +#else /* ! CONFIG_CPU_ENDIAN_BE8 */ + +#define __opcode_to_mem_arm(x) ___opcode_identity32(x) +#define __opcode_to_mem_thumb16(x) ___opcode_identity16(x) +#define ___asm_opcode_to_mem_arm(x) ___asm_opcode_identity32(x) +#define ___asm_opcode_to_mem_thumb16(x) ___asm_opcode_identity16(x) +#ifndef CONFIG_CPU_ENDIAN_BE32 +/* + * On BE32 systems, using 32-bit accesses to store Thumb instructions will not + * work in all cases, due to alignment constraints. For now, a correct + * version is not provided for BE32. + */ +#define __opcode_to_mem_thumb32(x) ___opcode_swahw32(x) +#define ___asm_opcode_to_mem_thumb32(x) ___asm_opcode_swahw32(x) +#endif + +#endif /* ! CONFIG_CPU_ENDIAN_BE8 */ + +#define __mem_to_opcode_arm(x) __opcode_to_mem_arm(x) +#define __mem_to_opcode_thumb16(x) __opcode_to_mem_thumb16(x) +#ifndef CONFIG_CPU_ENDIAN_BE32 +#define __mem_to_opcode_thumb32(x) __opcode_to_mem_thumb32(x) +#endif + +/* Operations specific to Thumb opcodes */ + +/* Instruction size checks: */ +#define __opcode_is_thumb32(x) ( \ + ((x) & 0xF8000000) == 0xE8000000 \ + || ((x) & 0xF0000000) == 0xF0000000 \ +) +#define __opcode_is_thumb16(x) ( \ + ((x) & 0xFFFF0000) == 0 \ + && !(((x) & 0xF800) == 0xE800 || ((x) & 0xF000) == 0xF000) \ +) + +/* Operations to construct or split 32-bit Thumb instructions: */ +#define __opcode_thumb32_first(x) (___opcode_identity16((x) >> 16)) +#define __opcode_thumb32_second(x) (___opcode_identity16(x)) +#define __opcode_thumb32_compose(first, second) ( \ + (___opcode_identity32(___opcode_identity16(first)) << 16) \ + | ___opcode_identity32(___opcode_identity16(second)) \ +) +#define ___asm_opcode_thumb32_first(x) (___asm_opcode_identity16((x) >> 16)) +#define ___asm_opcode_thumb32_second(x) (___asm_opcode_identity16(x)) +#define ___asm_opcode_thumb32_compose(first, second) ( \ + (___asm_opcode_identity32(___asm_opcode_identity16(first)) << 16) \ + | ___asm_opcode_identity32(___asm_opcode_identity16(second)) \ +) + +/* + * Opcode injection helpers + * + * In rare cases it is necessary to assemble an opcode which the + * assembler does not support directly, or which would normally be + * rejected because of the CFLAGS or AFLAGS used to build the affected + * file. + * + * Before using these macros, consider carefully whether it is feasible + * instead to change the build flags for your file, or whether it really + * makes sense to support old assembler versions when building that + * particular kernel feature. + * + * The macros defined here should only be used where there is no viable + * alternative. + * + * + * __inst_arm(x): emit the specified ARM opcode + * __inst_thumb16(x): emit the specified 16-bit Thumb opcode + * __inst_thumb32(x): emit the specified 32-bit Thumb opcode + * + * __inst_arm_thumb16(arm, thumb): emit either the specified arm or + * 16-bit Thumb opcode, depending on whether an ARM or Thumb-2 + * kernel is being built + * + * __inst_arm_thumb32(arm, thumb): emit either the specified arm or + * 32-bit Thumb opcode, depending on whether an ARM or Thumb-2 + * kernel is being built + * + * + * Note that using these macros directly is poor practice. Instead, you + * should use them to define human-readable wrapper macros to encode the + * instructions that you care about. In code which might run on ARMv7 or + * above, you can usually use the __inst_arm_thumb{16,32} macros to + * specify the ARM and Thumb alternatives at the same time. This ensures + * that the correct opcode gets emitted depending on the instruction set + * used for the kernel build. + * + * Look at opcodes-virt.h for an example of how to use these macros. + */ +#include <linux/stringify.h> + +#define __inst_arm(x) ___inst_arm(___asm_opcode_to_mem_arm(x)) +#define __inst_thumb32(x) ___inst_thumb32( \ + ___asm_opcode_to_mem_thumb16(___asm_opcode_thumb32_first(x)), \ + ___asm_opcode_to_mem_thumb16(___asm_opcode_thumb32_second(x)) \ +) +#define __inst_thumb16(x) ___inst_thumb16(___asm_opcode_to_mem_thumb16(x)) + +#ifdef CONFIG_THUMB2_KERNEL +#define __inst_arm_thumb16(arm_opcode, thumb_opcode) \ + __inst_thumb16(thumb_opcode) +#define __inst_arm_thumb32(arm_opcode, thumb_opcode) \ + __inst_thumb32(thumb_opcode) +#else +#define __inst_arm_thumb16(arm_opcode, thumb_opcode) __inst_arm(arm_opcode) +#define __inst_arm_thumb32(arm_opcode, thumb_opcode) __inst_arm(arm_opcode) +#endif + +/* Helpers for the helpers. Don't use these directly. */ +#ifdef __ASSEMBLY__ +#define ___inst_arm(x) .long x +#define ___inst_thumb16(x) .short x +#define ___inst_thumb32(first, second) .short first, second +#else +#define ___inst_arm(x) ".long " __stringify(x) "\n\t" +#define ___inst_thumb16(x) ".short " __stringify(x) "\n\t" +#define ___inst_thumb32(first, second) \ + ".short " __stringify(first) ", " __stringify(second) "\n\t" +#endif + +#endif /* __ASM_ARM_OPCODES_H */ diff --git a/arch/arm64/include/asm/ptrace.h b/arch/arm64/include/asm/ptrace.h index 5db016e6d065..c29d88526405 100644 --- a/arch/arm64/include/asm/ptrace.h +++ b/arch/arm64/include/asm/ptrace.h @@ -61,6 +61,15 @@ #define COMPAT_PT_TEXT_ADDR 0x10000 #define COMPAT_PT_DATA_ADDR 0x10004 #define COMPAT_PT_TEXT_END_ADDR 0x10008 + +/* + * used to skip a system call when tracer changes its number to -1 + * with ptrace(PTRACE_SET_SYSCALL) + */ +#define RET_SKIP_SYSCALL -1 +#define RET_SKIP_SYSCALL_TRACE -2 +#define IS_SKIP_SYSCALL(no) ((int)(no & 0xffffffff) == -1) + #ifndef __ASSEMBLY__ /* sizeof(struct user) for AArch32 */ @@ -178,5 +187,13 @@ extern unsigned long profile_pc(struct pt_regs *regs); #define profile_pc(regs) instruction_pointer(regs) #endif +/* + * True if instr is a 32-bit thumb instruction. This works if instr + * is the first or only half-word of a thumb instruction. It also works + * when instr holds all 32-bits of a wide thumb instruction if stored + * in the form (first_half<<16)|(second_half) + */ +#define is_wide_instruction(instr) ((unsigned)(instr) >= 0xe800) + #endif /* __ASSEMBLY__ */ #endif diff --git a/arch/arm64/include/asm/seccomp.h b/arch/arm64/include/asm/seccomp.h new file mode 100644 index 000000000000..bec3a43f7b17 --- /dev/null +++ b/arch/arm64/include/asm/seccomp.h @@ -0,0 +1,25 @@ +/* + * arch/arm64/include/asm/seccomp.h + * + * Copyright (C) 2014 Linaro Limited + * Author: AKASHI Takahiro <takahiro.akashi <at> 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. + */ +#ifndef _ASM_SECCOMP_H +#define _ASM_SECCOMP_H + +#include <asm/unistd.h> + +#ifdef CONFIG_COMPAT +#define __NR_seccomp_read_32 __NR_compat_read +#define __NR_seccomp_write_32 __NR_compat_write +#define __NR_seccomp_exit_32 __NR_compat_exit +#define __NR_seccomp_sigreturn_32 __NR_compat_rt_sigreturn +#endif /* CONFIG_COMPAT */ + +#include <asm-generic/seccomp.h> + +#endif /* _ASM_SECCOMP_H */ diff --git a/arch/arm64/include/asm/syscall.h b/arch/arm64/include/asm/syscall.h index 383771eb0b87..709a574468f0 100644 --- a/arch/arm64/include/asm/syscall.h +++ b/arch/arm64/include/asm/syscall.h @@ -16,6 +16,8 @@ #ifndef __ASM_SYSCALL_H #define __ASM_SYSCALL_H +#include <uapi/linux/audit.h> +#include <linux/compat.h> #include <linux/err.h> extern const void *sys_call_table[]; @@ -105,4 +107,16 @@ static inline void syscall_set_arguments(struct task_struct *task, memcpy(®s->regs[i], args, n * sizeof(args[0])); } +/* + * We don't care about endianness (__AUDIT_ARCH_LE bit) here because + * AArch64 has the same system calls both on little- and big- endian. + */ +static inline int syscall_get_arch(void) +{ + if (is_compat_task()) + return AUDIT_ARCH_ARM; + + return AUDIT_ARCH_AARCH64; +} + #endif /* __ASM_SYSCALL_H */ diff --git a/arch/arm64/include/asm/thread_info.h b/arch/arm64/include/asm/thread_info.h index 70dd1ac01f16..e4a98b4c2020 100644 --- a/arch/arm64/include/asm/thread_info.h +++ b/arch/arm64/include/asm/thread_info.h @@ -114,6 +114,7 @@ static inline struct thread_info *current_thread_info(void) #define TIF_SIGPENDING 0 #define TIF_NEED_RESCHED 1 #define TIF_NOTIFY_RESUME 2 /* callback before returning to user */ +#define TIF_FOREIGN_FPSTATE 3 /* CPU's FP state is not current's */ #define TIF_SYSCALL_TRACE 8 #define TIF_SYSCALL_AUDIT 9 #define TIF_SYSCALL_TRACEPOINT 10 @@ -133,10 +134,15 @@ static inline struct thread_info *current_thread_info(void) #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) #define _TIF_SECCOMP (1 << TIF_SECCOMP) +#define _TIF_FOREIGN_FPSTATE (1 << TIF_FOREIGN_FPSTATE) +#define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) +#define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT) +#define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) +#define _TIF_SECCOMP (1 << TIF_SECCOMP) #define _TIF_32BIT (1 << TIF_32BIT) #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \ - _TIF_NOTIFY_RESUME) + _TIF_NOTIFY_RESUME | _TIF_FOREIGN_FPSTATE) #define _TIF_SYSCALL_WORK (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | \ _TIF_SYSCALL_TRACEPOINT | _TIF_SECCOMP) diff --git a/arch/arm64/include/asm/traps.h b/arch/arm64/include/asm/traps.h index 10ca8ff93cc2..75f51eadb9eb 100644 --- a/arch/arm64/include/asm/traps.h +++ b/arch/arm64/include/asm/traps.h @@ -18,6 +18,19 @@ #ifndef __ASM_TRAP_H #define __ASM_TRAP_H +#include <linux/list.h> + +struct undef_hook { + struct list_head node; + u32 instr_mask; + u32 instr_val; + u32 pstate_mask; + u32 pstate_val; + int (*fn)(struct pt_regs *regs, unsigned int instr); +}; + +void register_undef_hook(struct undef_hook *hook); + static inline int in_exception_text(unsigned long ptr) { extern char __exception_text_start[]; diff --git a/arch/arm64/include/asm/unistd.h b/arch/arm64/include/asm/unistd.h index c335479c2638..f67d0ec20f94 100644 --- a/arch/arm64/include/asm/unistd.h +++ b/arch/arm64/include/asm/unistd.h @@ -25,7 +25,27 @@ #define __ARCH_WANT_COMPAT_SYS_SENDFILE #define __ARCH_WANT_SYS_FORK #define __ARCH_WANT_SYS_VFORK + +/* + * Compat syscall numbers used by the AArch64 kernel. + */ +#define __NR_compat_restart_syscall 0 +#define __NR_compat_exit 1 +#define __NR_compat_read 3 +#define __NR_compat_write 4 +#define __NR_compat_sigreturn 119 +#define __NR_compat_rt_sigreturn 173 + +/* + * The following SVCs are ARM private. + */ +#define __ARM_NR_COMPAT_BASE 0x0f0000 +#define __ARM_NR_compat_cacheflush (__ARM_NR_COMPAT_BASE+2) +#define __ARM_NR_compat_set_tls (__ARM_NR_COMPAT_BASE+5) + +#define __NR_compat_syscalls 384 #endif + #define __ARCH_WANT_SYS_CLONE #include <uapi/asm/unistd.h> diff --git a/arch/arm64/include/asm/unistd32.h b/arch/arm64/include/asm/unistd32.h index 58125bf008d3..76d094565090 100644 --- a/arch/arm64/include/asm/unistd32.h +++ b/arch/arm64/include/asm/unistd32.h @@ -21,399 +21,771 @@ #define __SYSCALL(x, y) #endif -__SYSCALL(0, sys_restart_syscall) -__SYSCALL(1, sys_exit) -__SYSCALL(2, sys_fork) -__SYSCALL(3, sys_read) -__SYSCALL(4, sys_write) -__SYSCALL(5, compat_sys_open) -__SYSCALL(6, sys_close) -__SYSCALL(7, sys_ni_syscall) /* 7 was sys_waitpid */ -__SYSCALL(8, sys_creat) -__SYSCALL(9, sys_link) -__SYSCALL(10, sys_unlink) -__SYSCALL(11, compat_sys_execve) -__SYSCALL(12, sys_chdir) -__SYSCALL(13, sys_ni_syscall) /* 13 was sys_time */ -__SYSCALL(14, sys_mknod) -__SYSCALL(15, sys_chmod) -__SYSCALL(16, sys_lchown16) -__SYSCALL(17, sys_ni_syscall) /* 17 was sys_break */ -__SYSCALL(18, sys_ni_syscall) /* 18 was sys_stat */ -__SYSCALL(19, compat_sys_lseek) -__SYSCALL(20, sys_getpid) -__SYSCALL(21, compat_sys_mount) -__SYSCALL(22, sys_ni_syscall) /* 22 was sys_umount */ -__SYSCALL(23, sys_setuid16) -__SYSCALL(24, sys_getuid16) -__SYSCALL(25, sys_ni_syscall) /* 25 was sys_stime */ -__SYSCALL(26, compat_sys_ptrace) -__SYSCALL(27, sys_ni_syscall) /* 27 was sys_alarm */ -__SYSCALL(28, sys_ni_syscall) /* 28 was sys_fstat */ -__SYSCALL(29, sys_pause) -__SYSCALL(30, sys_ni_syscall) /* 30 was sys_utime */ -__SYSCALL(31, sys_ni_syscall) /* 31 was sys_stty */ -__SYSCALL(32, sys_ni_syscall) /* 32 was sys_gtty */ -__SYSCALL(33, sys_access) -__SYSCALL(34, sys_nice) -__SYSCALL(35, sys_ni_syscall) /* 35 was sys_ftime */ -__SYSCALL(36, sys_sync) -__SYSCALL(37, sys_kill) -__SYSCALL(38, sys_rename) -__SYSCALL(39, sys_mkdir) -__SYSCALL(40, sys_rmdir) -__SYSCALL(41, sys_dup) -__SYSCALL(42, sys_pipe) -__SYSCALL(43, compat_sys_times) -__SYSCALL(44, sys_ni_syscall) /* 44 was sys_prof */ -__SYSCALL(45, sys_brk) -__SYSCALL(46, sys_setgid16) -__SYSCALL(47, sys_getgid16) -__SYSCALL(48, sys_ni_syscall) /* 48 was sys_signal */ -__SYSCALL(49, sys_geteuid16) -__SYSCALL(50, sys_getegid16) -__SYSCALL(51, sys_acct) -__SYSCALL(52, sys_umount) -__SYSCALL(53, sys_ni_syscall) /* 53 was sys_lock */ -__SYSCALL(54, compat_sys_ioctl) -__SYSCALL(55, compat_sys_fcntl) -__SYSCALL(56, sys_ni_syscall) /* 56 was sys_mpx */ -__SYSCALL(57, sys_setpgid) -__SYSCALL(58, sys_ni_syscall) /* 58 was sys_ulimit */ -__SYSCALL(59, sys_ni_syscall) /* 59 was sys_olduname */ -__SYSCALL(60, sys_umask) -__SYSCALL(61, sys_chroot) -__SYSCALL(62, compat_sys_ustat) -__SYSCALL(63, sys_dup2) -__SYSCALL(64, sys_getppid) -__SYSCALL(65, sys_getpgrp) -__SYSCALL(66, sys_setsid) -__SYSCALL(67, compat_sys_sigaction) -__SYSCALL(68, sys_ni_syscall) /* 68 was sys_sgetmask */ -__SYSCALL(69, sys_ni_syscall) /* 69 was sys_ssetmask */ -__SYSCALL(70, sys_setreuid16) -__SYSCALL(71, sys_setregid16) -__SYSCALL(72, sys_sigsuspend) -__SYSCALL(73, compat_sys_sigpending) -__SYSCALL(74, sys_sethostname) -__SYSCALL(75, compat_sys_setrlimit) -__SYSCALL(76, sys_ni_syscall) /* 76 was compat_sys_getrlimit */ -__SYSCALL(77, compat_sys_getrusage) -__SYSCALL(78, compat_sys_gettimeofday) -__SYSCALL(79, compat_sys_settimeofday) -__SYSCALL(80, sys_getgroups16) -__SYSCALL(81, sys_setgroups16) -__SYSCALL(82, sys_ni_syscall) /* 82 was compat_sys_select */ -__SYSCALL(83, sys_symlink) -__SYSCALL(84, sys_ni_syscall) /* 84 was sys_lstat */ -__SYSCALL(85, sys_readlink) -__SYSCALL(86, sys_uselib) -__SYSCALL(87, sys_swapon) -__SYSCALL(88, sys_reboot) -__SYSCALL(89, sys_ni_syscall) /* 89 was sys_readdir */ -__SYSCALL(90, sys_ni_syscall) /* 90 was sys_mmap */ -__SYSCALL(91, sys_munmap) -__SYSCALL(92, compat_sys_truncate) -__SYSCALL(93, compat_sys_ftruncate) -__SYSCALL(94, sys_fchmod) -__SYSCALL(95, sys_fchown16) -__SYSCALL(96, sys_getpriority) -__SYSCALL(97, sys_setpriority) -__SYSCALL(98, sys_ni_syscall) /* 98 was sys_profil */ -__SYSCALL(99, compat_sys_statfs) -__SYSCALL(100, compat_sys_fstatfs) -__SYSCALL(101, sys_ni_syscall) /* 101 was sys_ioperm */ -__SYSCALL(102, sys_ni_syscall) /* 102 was sys_socketcall */ -__SYSCALL(103, sys_syslog) -__SYSCALL(104, compat_sys_setitimer) -__SYSCALL(105, compat_sys_getitimer) -__SYSCALL(106, compat_sys_newstat) -__SYSCALL(107, compat_sys_newlstat) -__SYSCALL(108, compat_sys_newfstat) -__SYSCALL(109, sys_ni_syscall) /* 109 was sys_uname */ -__SYSCALL(110, sys_ni_syscall) /* 110 was sys_iopl */ -__SYSCALL(111, sys_vhangup) -__SYSCALL(112, sys_ni_syscall) /* 112 was sys_idle */ -__SYSCALL(113, sys_ni_syscall) /* 113 was sys_syscall */ -__SYSCALL(114, compat_sys_wait4) -__SYSCALL(115, sys_swapoff) -__SYSCALL(116, compat_sys_sysinfo) -__SYSCALL(117, sys_ni_syscall) /* 117 was sys_ipc */ -__SYSCALL(118, sys_fsync) -__SYSCALL(119, compat_sys_sigreturn_wrapper) -__SYSCALL(120, sys_clone) -__SYSCALL(121, sys_setdomainname) -__SYSCALL(122, sys_newuname) -__SYSCALL(123, sys_ni_syscall) /* 123 was sys_modify_ldt */ -__SYSCALL(124, compat_sys_adjtimex) -__SYSCALL(125, sys_mprotect) -__SYSCALL(126, compat_sys_sigprocmask) -__SYSCALL(127, sys_ni_syscall) /* 127 was sys_create_module */ -__SYSCALL(128, sys_init_module) -__SYSCALL(129, sys_delete_module) -__SYSCALL(130, sys_ni_syscall) /* 130 was sys_get_kernel_syms */ -__SYSCALL(131, sys_quotactl) -__SYSCALL(132, sys_getpgid) -__SYSCALL(133, sys_fchdir) -__SYSCALL(134, sys_bdflush) -__SYSCALL(135, sys_sysfs) -__SYSCALL(136, sys_personality) -__SYSCALL(137, sys_ni_syscall) /* 137 was sys_afs_syscall */ -__SYSCALL(138, sys_setfsuid16) -__SYSCALL(139, sys_setfsgid16) -__SYSCALL(140, sys_llseek) -__SYSCALL(141, compat_sys_getdents) -__SYSCALL(142, compat_sys_select) -__SYSCALL(143, sys_flock) -__SYSCALL(144, sys_msync) -__SYSCALL(145, compat_sys_readv) -__SYSCALL(146, compat_sys_writev) -__SYSCALL(147, sys_getsid) -__SYSCALL(148, sys_fdatasync) -__SYSCALL(149, compat_sys_sysctl) -__SYSCALL(150, sys_mlock) -__SYSCALL(151, sys_munlock) -__SYSCALL(152, sys_mlockall) -__SYSCALL(153, sys_munlockall) -__SYSCALL(154, sys_sched_setparam) -__SYSCALL(155, sys_sched_getparam) -__SYSCALL(156, sys_sched_setscheduler) -__SYSCALL(157, sys_sched_getscheduler) -__SYSCALL(158, sys_sched_yield) -__SYSCALL(159, sys_sched_get_priority_max) -__SYSCALL(160, sys_sched_get_priority_min) -__SYSCALL(161, compat_sys_sched_rr_get_interval) -__SYSCALL(162, compat_sys_nanosleep) -__SYSCALL(163, sys_mremap) -__SYSCALL(164, sys_setresuid16) -__SYSCALL(165, sys_getresuid16) -__SYSCALL(166, sys_ni_syscall) /* 166 was sys_vm86 */ -__SYSCALL(167, sys_ni_syscall) /* 167 was sys_query_module */ -__SYSCALL(168, sys_poll) -__SYSCALL(169, sys_ni_syscall) -__SYSCALL(170, sys_setresgid16) -__SYSCALL(171, sys_getresgid16) -__SYSCALL(172, sys_prctl) -__SYSCALL(173, compat_sys_rt_sigreturn_wrapper) -__SYSCALL(174, compat_sys_rt_sigaction) -__SYSCALL(175, compat_sys_rt_sigprocmask) -__SYSCALL(176, compat_sys_rt_sigpending) -__SYSCALL(177, compat_sys_rt_sigtimedwait) -__SYSCALL(178, compat_sys_rt_sigqueueinfo) -__SYSCALL(179, compat_sys_rt_sigsuspend) -__SYSCALL(180, compat_sys_pread64_wrapper) -__SYSCALL(181, compat_sys_pwrite64_wrapper) -__SYSCALL(182, sys_chown16) -__SYSCALL(183, sys_getcwd) -__SYSCALL(184, sys_capget) -__SYSCALL(185, sys_capset) -__SYSCALL(186, compat_sys_sigaltstack) -__SYSCALL(187, compat_sys_sendfile) -__SYSCALL(188, sys_ni_syscall) /* 188 reserved */ -__SYSCALL(189, sys_ni_syscall) /* 189 reserved */ -__SYSCALL(190, sys_vfork) -__SYSCALL(191, compat_sys_getrlimit) /* SuS compliant getrlimit */ -__SYSCALL(192, sys_mmap_pgoff) -__SYSCALL(193, compat_sys_truncate64_wrapper) -__SYSCALL(194, compat_sys_ftruncate64_wrapper) -__SYSCALL(195, sys_stat64) -__SYSCALL(196, sys_lstat64) -__SYSCALL(197, sys_fstat64) -__SYSCALL(198, sys_lchown) -__SYSCALL(199, sys_getuid) -__SYSCALL(200, sys_getgid) -__SYSCALL(201, sys_geteuid) -__SYSCALL(202, sys_getegid) -__SYSCALL(203, sys_setreuid) -__SYSCALL(204, sys_setregid) -__SYSCALL(205, sys_getgroups) -__SYSCALL(206, sys_setgroups) -__SYSCALL(207, sys_fchown) -__SYSCALL(208, sys_setresuid) -__SYSCALL(209, sys_getresuid) -__SYSCALL(210, sys_setresgid) -__SYSCALL(211, sys_getresgid) -__SYSCALL(212, sys_chown) -__SYSCALL(213, sys_setuid) -__SYSCALL(214, sys_setgid) -__SYSCALL(215, sys_setfsuid) -__SYSCALL(216, sys_setfsgid) -__SYSCALL(217, compat_sys_getdents64) -__SYSCALL(218, sys_pivot_root) -__SYSCALL(219, sys_mincore) -__SYSCALL(220, sys_madvise) -__SYSCALL(221, compat_sys_fcntl64) -__SYSCALL(222, sys_ni_syscall) /* 222 for tux */ -__SYSCALL(223, sys_ni_syscall) /* 223 is unused */ -__SYSCALL(224, sys_gettid) -__SYSCALL(225, compat_sys_readahead_wrapper) -__SYSCALL(226, sys_setxattr) -__SYSCALL(227, sys_lsetxattr) -__SYSCALL(228, sys_fsetxattr) -__SYSCALL(229, sys_getxattr) -__SYSCALL(230, sys_lgetxattr) -__SYSCALL(231, sys_fgetxattr) -__SYSCALL(232, sys_listxattr) -__SYSCALL(233, sys_llistxattr) -__SYSCALL(234, sys_flistxattr) -__SYSCALL(235, sys_removexattr) -__SYSCALL(236, sys_lremovexattr) -__SYSCALL(237, sys_fremovexattr) -__SYSCALL(238, sys_tkill) -__SYSCALL(239, sys_sendfile64) -__SYSCALL(240, compat_sys_futex) -__SYSCALL(241, compat_sys_sched_setaffinity) -__SYSCALL(242, compat_sys_sched_getaffinity) -__SYSCALL(243, compat_sys_io_setup) -__SYSCALL(244, sys_io_destroy) -__SYSCALL(245, compat_sys_io_getevents) -__SYSCALL(246, compat_sys_io_submit) -__SYSCALL(247, sys_io_cancel) -__SYSCALL(248, sys_exit_group) -__SYSCALL(249, compat_sys_lookup_dcookie) -__SYSCALL(250, sys_epoll_create) -__SYSCALL(251, sys_epoll_ctl) -__SYSCALL(252, sys_epoll_wait) -__SYSCALL(253, sys_remap_file_pages) -__SYSCALL(254, sys_ni_syscall) /* 254 for set_thread_area */ -__SYSCALL(255, sys_ni_syscall) /* 255 for get_thread_area */ -__SYSCALL(256, sys_set_tid_address) -__SYSCALL(257, compat_sys_timer_create) -__SYSCALL(258, compat_sys_timer_settime) -__SYSCALL(259, compat_sys_timer_gettime) -__SYSCALL(260, sys_timer_getoverrun) -__SYSCALL(261, sys_timer_delete) -__SYSCALL(262, compat_sys_clock_settime) -__SYSCALL(263, compat_sys_clock_gettime) -__SYSCALL(264, compat_sys_clock_getres) -__SYSCALL(265, compat_sys_clock_nanosleep) -__SYSCALL(266, compat_sys_statfs64_wrapper) -__SYSCALL(267, compat_sys_fstatfs64_wrapper) -__SYSCALL(268, sys_tgkill) -__SYSCALL(269, compat_sys_utimes) -__SYSCALL(270, compat_sys_fadvise64_64_wrapper) -__SYSCALL(271, sys_pciconfig_iobase) -__SYSCALL(272, sys_pciconfig_read) -__SYSCALL(273, sys_pciconfig_write) -__SYSCALL(274, compat_sys_mq_open) -__SYSCALL(275, sys_mq_unlink) -__SYSCALL(276, compat_sys_mq_timedsend) -__SYSCALL(277, compat_sys_mq_timedreceive) -__SYSCALL(278, compat_sys_mq_notify) -__SYSCALL(279, compat_sys_mq_getsetattr) -__SYSCALL(280, compat_sys_waitid) -__SYSCALL(281, sys_socket) -__SYSCALL(282, sys_bind) -__SYSCALL(283, sys_connect) -__SYSCALL(284, sys_listen) -__SYSCALL(285, sys_accept) -__SYSCALL(286, sys_getsockname) -__SYSCALL(287, sys_getpeername) -__SYSCALL(288, sys_socketpair) -__SYSCALL(289, sys_send) -__SYSCALL(290, sys_sendto) -__SYSCALL(291, compat_sys_recv) -__SYSCALL(292, compat_sys_recvfrom) -__SYSCALL(293, sys_shutdown) -__SYSCALL(294, compat_sys_setsockopt) -__SYSCALL(295, compat_sys_getsockopt) -__SYSCALL(296, compat_sys_sendmsg) -__SYSCALL(297, compat_sys_recvmsg) -__SYSCALL(298, sys_semop) -__SYSCALL(299, sys_semget) -__SYSCALL(300, compat_sys_semctl) -__SYSCALL(301, compat_sys_msgsnd) -__SYSCALL(302, compat_sys_msgrcv) -__SYSCALL(303, sys_msgget) -__SYSCALL(304, compat_sys_msgctl) -__SYSCALL(305, compat_sys_shmat) -__SYSCALL(306, sys_shmdt) -__SYSCALL(307, sys_shmget) -__SYSCALL(308, compat_sys_shmctl) -__SYSCALL(309, sys_add_key) -__SYSCALL(310, sys_request_key) -__SYSCALL(311, compat_sys_keyctl) -__SYSCALL(312, compat_sys_semtimedop) -__SYSCALL(313, sys_ni_syscall) -__SYSCALL(314, sys_ioprio_set) -__SYSCALL(315, sys_ioprio_get) -__SYSCALL(316, sys_inotify_init) -__SYSCALL(317, sys_inotify_add_watch) -__SYSCALL(318, sys_inotify_rm_watch) -__SYSCALL(319, compat_sys_mbind) -__SYSCALL(320, compat_sys_get_mempolicy) -__SYSCALL(321, compat_sys_set_mempolicy) -__SYSCALL(322, compat_sys_openat) -__SYSCALL(323, sys_mkdirat) -__SYSCALL(324, sys_mknodat) -__SYSCALL(325, sys_fchownat) -__SYSCALL(326, compat_sys_futimesat) -__SYSCALL(327, sys_fstatat64) -__SYSCALL(328, sys_unlinkat) -__SYSCALL(329, sys_renameat) -__SYSCALL(330, sys_linkat) -__SYSCALL(331, sys_symlinkat) -__SYSCALL(332, sys_readlinkat) -__SYSCALL(333, sys_fchmodat) -__SYSCALL(334, sys_faccessat) -__SYSCALL(335, compat_sys_pselect6) -__SYSCALL(336, compat_sys_ppoll) -__SYSCALL(337, sys_unshare) -__SYSCALL(338, compat_sys_set_robust_list) -__SYSCALL(339, compat_sys_get_robust_list) -__SYSCALL(340, sys_splice) -__SYSCALL(341, compat_sys_sync_file_range2_wrapper) -__SYSCALL(342, sys_tee) -__SYSCALL(343, compat_sys_vmsplice) -__SYSCALL(344, compat_sys_move_pages) -__SYSCALL(345, sys_getcpu) -__SYSCALL(346, compat_sys_epoll_pwait) -__SYSCALL(347, compat_sys_kexec_load) -__SYSCALL(348, compat_sys_utimensat) -__SYSCALL(349, compat_sys_signalfd) -__SYSCALL(350, sys_timerfd_create) -__SYSCALL(351, sys_eventfd) -__SYSCALL(352, compat_sys_fallocate_wrapper) -__SYSCALL(353, compat_sys_timerfd_settime) -__SYSCALL(354, compat_sys_timerfd_gettime) -__SYSCALL(355, compat_sys_signalfd4) -__SYSCALL(356, sys_eventfd2) -__SYSCALL(357, sys_epoll_create1) -__SYSCALL(358, sys_dup3) -__SYSCALL(359, sys_pipe2) -__SYSCALL(360, sys_inotify_init1) -__SYSCALL(361, compat_sys_preadv) -__SYSCALL(362, compat_sys_pwritev) -__SYSCALL(363, compat_sys_rt_tgsigqueueinfo) -__SYSCALL(364, sys_perf_event_open) -__SYSCALL(365, compat_sys_recvmmsg) -__SYSCALL(366, sys_accept4) -__SYSCALL(367, sys_fanotify_init) -__SYSCALL(368, compat_sys_fanotify_mark) -__SYSCALL(369, sys_prlimit64) -__SYSCALL(370, sys_name_to_handle_at) -__SYSCALL(371, compat_sys_open_by_handle_at) -__SYSCALL(372, compat_sys_clock_adjtime) -__SYSCALL(373, sys_syncfs) -__SYSCALL(374, compat_sys_sendmmsg) -__SYSCALL(375, sys_setns) -__SYSCALL(376, compat_sys_process_vm_readv) -__SYSCALL(377, compat_sys_process_vm_writev) -__SYSCALL(378, sys_ni_syscall) /* 378 for kcmp */ - -#define __NR_compat_syscalls 379 - -/* - * Compat syscall numbers used by the AArch64 kernel. - */ -#define __NR_compat_restart_syscall 0 -#define __NR_compat_sigreturn 119 -#define __NR_compat_rt_sigreturn 173 - - -/* - * The following SVCs are ARM private. - */ -#define __ARM_NR_COMPAT_BASE 0x0f0000 -#define __ARM_NR_compat_cacheflush (__ARM_NR_COMPAT_BASE+2) -#define __ARM_NR_compat_set_tls (__ARM_NR_COMPAT_BASE+5) +#define __NR_restart_syscall 0 +__SYSCALL(__NR_restart_syscall, sys_restart_syscall) +#define __NR_exit 1 +__SYSCALL(__NR_exit, sys_exit) +#define __NR_fork 2 +__SYSCALL(__NR_fork, sys_fork) +#define __NR_read 3 +__SYSCALL(__NR_read, sys_read) +#define __NR_write 4 +__SYSCALL(__NR_write, sys_write) +#define __NR_open 5 +__SYSCALL(__NR_open, compat_sys_open) +#define __NR_close 6 +__SYSCALL(__NR_close, sys_close) + /* 7 was sys_waitpid */ +__SYSCALL(7, sys_ni_syscall) +#define __NR_creat 8 +__SYSCALL(__NR_creat, sys_creat) +#define __NR_link 9 +__SYSCALL(__NR_link, sys_link) +#define __NR_unlink 10 +__SYSCALL(__NR_unlink, sys_unlink) +#define __NR_execve 11 +__SYSCALL(__NR_execve, compat_sys_execve) +#define __NR_chdir 12 +__SYSCALL(__NR_chdir, sys_chdir) + /* 13 was sys_time */ +__SYSCALL(13, sys_ni_syscall) +#define __NR_mknod 14 +__SYSCALL(__NR_mknod, sys_mknod) +#define __NR_chmod 15 +__SYSCALL(__NR_chmod, sys_chmod) +#define __NR_lchown 16 +__SYSCALL(__NR_lchown, sys_lchown16) + /* 17 was sys_break */ +__SYSCALL(17, sys_ni_syscall) + /* 18 was sys_stat */ +__SYSCALL(18, sys_ni_syscall) +#define __NR_lseek 19 +__SYSCALL(__NR_lseek, compat_sys_lseek) +#define __NR_getpid 20 +__SYSCALL(__NR_getpid, sys_getpid) +#define __NR_mount 21 +__SYSCALL(__NR_mount, compat_sys_mount) + /* 22 was sys_umount */ +__SYSCALL(22, sys_ni_syscall) +#define __NR_setuid 23 +__SYSCALL(__NR_setuid, sys_setuid16) +#define __NR_getuid 24 +__SYSCALL(__NR_getuid, sys_getuid16) + /* 25 was sys_stime */ +__SYSCALL(25, sys_ni_syscall) +#define __NR_ptrace 26 +__SYSCALL(__NR_ptrace, compat_sys_ptrace) + /* 27 was sys_alarm */ +__SYSCALL(27, sys_ni_syscall) + /* 28 was sys_fstat */ +__SYSCALL(28, sys_ni_syscall) +#define __NR_pause 29 +__SYSCALL(__NR_pause, sys_pause) + /* 30 was sys_utime */ +__SYSCALL(30, sys_ni_syscall) + /* 31 was sys_stty */ +__SYSCALL(31, sys_ni_syscall) + /* 32 was sys_gtty */ +__SYSCALL(32, sys_ni_syscall) +#define __NR_access 33 +__SYSCALL(__NR_access, sys_access) +#define __NR_nice 34 +__SYSCALL(__NR_nice, sys_nice) + /* 35 was sys_ftime */ +__SYSCALL(35, sys_ni_syscall) +#define __NR_sync 36 +__SYSCALL(__NR_sync, sys_sync) +#define __NR_kill 37 +__SYSCALL(__NR_kill, sys_kill) +#define __NR_rename 38 +__SYSCALL(__NR_rename, sys_rename) +#define __NR_mkdir 39 +__SYSCALL(__NR_mkdir, sys_mkdir) +#define __NR_rmdir 40 +__SYSCALL(__NR_rmdir, sys_rmdir) +#define __NR_dup 41 +__SYSCALL(__NR_dup, sys_dup) +#define __NR_pipe 42 +__SYSCALL(__NR_pipe, sys_pipe) +#define __NR_times 43 +__SYSCALL(__NR_times, compat_sys_times) + /* 44 was sys_prof */ +__SYSCALL(44, sys_ni_syscall) +#define __NR_brk 45 +__SYSCALL(__NR_brk, sys_brk) +#define __NR_setgid 46 +__SYSCALL(__NR_setgid, sys_setgid16) +#define __NR_getgid 47 +__SYSCALL(__NR_getgid, sys_getgid16) + /* 48 was sys_signal */ +__SYSCALL(48, sys_ni_syscall) +#define __NR_geteuid 49 +__SYSCALL(__NR_geteuid, sys_geteuid16) +#define __NR_getegid 50 +__SYSCALL(__NR_getegid, sys_getegid16) +#define __NR_acct 51 +__SYSCALL(__NR_acct, sys_acct) +#define __NR_umount2 52 +__SYSCALL(__NR_umount2, sys_umount) + /* 53 was sys_lock */ +__SYSCALL(53, sys_ni_syscall) +#define __NR_ioctl 54 +__SYSCALL(__NR_ioctl, compat_sys_ioctl) +#define __NR_fcntl 55 +__SYSCALL(__NR_fcntl, compat_sys_fcntl) + /* 56 was sys_mpx */ +__SYSCALL(56, sys_ni_syscall) +#define __NR_setpgid 57 +__SYSCALL(__NR_setpgid, sys_setpgid) + /* 58 was sys_ulimit */ +__SYSCALL(58, sys_ni_syscall) + /* 59 was sys_olduname */ +__SYSCALL(59, sys_ni_syscall) +#define __NR_umask 60 +__SYSCALL(__NR_umask, sys_umask) +#define __NR_chroot 61 +__SYSCALL(__NR_chroot, sys_chroot) +#define __NR_ustat 62 +__SYSCALL(__NR_ustat, compat_sys_ustat) +#define __NR_dup2 63 +__SYSCALL(__NR_dup2, sys_dup2) +#define __NR_getppid 64 +__SYSCALL(__NR_getppid, sys_getppid) +#define __NR_getpgrp 65 +__SYSCALL(__NR_getpgrp, sys_getpgrp) +#define __NR_setsid 66 +__SYSCALL(__NR_setsid, sys_setsid) +#define __NR_sigaction 67 +__SYSCALL(__NR_sigaction, compat_sys_sigaction) + /* 68 was sys_sgetmask */ +__SYSCALL(68, sys_ni_syscall) + /* 69 was sys_ssetmask */ +__SYSCALL(69, sys_ni_syscall) +#define __NR_setreuid 70 +__SYSCALL(__NR_setreuid, sys_setreuid16) +#define __NR_setregid 71 +__SYSCALL(__NR_setregid, sys_setregid16) +#define __NR_sigsuspend 72 +__SYSCALL(__NR_sigsuspend, sys_sigsuspend) +#define __NR_sigpending 73 +__SYSCALL(__NR_sigpending, compat_sys_sigpending) +#define __NR_sethostname 74 +__SYSCALL(__NR_sethostname, sys_sethostname) +#define __NR_setrlimit 75 +__SYSCALL(__NR_setrlimit, compat_sys_setrlimit) + /* 76 was compat_sys_getrlimit */ +__SYSCALL(76, sys_ni_syscall) +#define __NR_getrusage 77 +__SYSCALL(__NR_getrusage, compat_sys_getrusage) +#define __NR_gettimeofday 78 +__SYSCALL(__NR_gettimeofday, compat_sys_gettimeofday) +#define __NR_settimeofday 79 +__SYSCALL(__NR_settimeofday, compat_sys_settimeofday) +#define __NR_getgroups 80 +__SYSCALL(__NR_getgroups, sys_getgroups16) +#define __NR_setgroups 81 +__SYSCALL(__NR_setgroups, sys_setgroups16) + /* 82 was compat_sys_select */ +__SYSCALL(82, sys_ni_syscall) +#define __NR_symlink 83 +__SYSCALL(__NR_symlink, sys_symlink) + /* 84 was sys_lstat */ +__SYSCALL(84, sys_ni_syscall) +#define __NR_readlink 85 +__SYSCALL(__NR_readlink, sys_readlink) +#define __NR_uselib 86 +__SYSCALL(__NR_uselib, sys_uselib) +#define __NR_swapon 87 +__SYSCALL(__NR_swapon, sys_swapon) +#define __NR_reboot 88 +__SYSCALL(__NR_reboot, sys_reboot) + /* 89 was sys_readdir */ +__SYSCALL(89, sys_ni_syscall) + /* 90 was sys_mmap */ +__SYSCALL(90, sys_ni_syscall) +#define __NR_munmap 91 +__SYSCALL(__NR_munmap, sys_munmap) +#define __NR_truncate 92 +__SYSCALL(__NR_truncate, compat_sys_truncate) +#define __NR_ftruncate 93 +__SYSCALL(__NR_ftruncate, compat_sys_ftruncate) +#define __NR_fchmod 94 +__SYSCALL(__NR_fchmod, sys_fchmod) +#define __NR_fchown 95 +__SYSCALL(__NR_fchown, sys_fchown16) +#define __NR_getpriority 96 +__SYSCALL(__NR_getpriority, sys_getpriority) +#define __NR_setpriority 97 +__SYSCALL(__NR_setpriority, sys_setpriority) + /* 98 was sys_profil */ +__SYSCALL(98, sys_ni_syscall) +#define __NR_statfs 99 +__SYSCALL(__NR_statfs, compat_sys_statfs) +#define __NR_fstatfs 100 +__SYSCALL(__NR_fstatfs, compat_sys_fstatfs) + /* 101 was sys_ioperm */ +__SYSCALL(101, sys_ni_syscall) + /* 102 was sys_socketcall */ +__SYSCALL(102, sys_ni_syscall) +#define __NR_syslog 103 +__SYSCALL(__NR_syslog, sys_syslog) +#define __NR_setitimer 104 +__SYSCALL(__NR_setitimer, compat_sys_setitimer) +#define __NR_getitimer 105 +__SYSCALL(__NR_getitimer, compat_sys_getitimer) +#define __NR_stat 106 +__SYSCALL(__NR_stat, compat_sys_newstat) +#define __NR_lstat 107 +__SYSCALL(__NR_lstat, compat_sys_newlstat) +#define __NR_fstat 108 +__SYSCALL(__NR_fstat, compat_sys_newfstat) + /* 109 was sys_uname */ +__SYSCALL(109, sys_ni_syscall) + /* 110 was sys_iopl */ +__SYSCALL(110, sys_ni_syscall) +#define __NR_vhangup 111 +__SYSCALL(__NR_vhangup, sys_vhangup) + /* 112 was sys_idle */ +__SYSCALL(112, sys_ni_syscall) + /* 113 was sys_syscall */ +__SYSCALL(113, sys_ni_syscall) +#define __NR_wait4 114 +__SYSCALL(__NR_wait4, compat_sys_wait4) +#define __NR_swapoff 115 +__SYSCALL(__NR_swapoff, sys_swapoff) +#define __NR_sysinfo 116 +__SYSCALL(__NR_sysinfo, compat_sys_sysinfo) + /* 117 was sys_ipc */ +__SYSCALL(117, sys_ni_syscall) +#define __NR_fsync 118 +__SYSCALL(__NR_fsync, sys_fsync) +#define __NR_sigreturn 119 +__SYSCALL(__NR_sigreturn, compat_sys_sigreturn_wrapper) +#define __NR_clone 120 +__SYSCALL(__NR_clone, sys_clone) +#define __NR_setdomainname 121 +__SYSCALL(__NR_setdomainname, sys_setdomainname) +#define __NR_uname 122 +__SYSCALL(__NR_uname, sys_newuname) + /* 123 was sys_modify_ldt */ +__SYSCALL(123, sys_ni_syscall) +#define __NR_adjtimex 124 +__SYSCALL(__NR_adjtimex, compat_sys_adjtimex) +#define __NR_mprotect 125 +__SYSCALL(__NR_mprotect, sys_mprotect) +#define __NR_sigprocmask 126 +__SYSCALL(__NR_sigprocmask, compat_sys_sigprocmask) + /* 127 was sys_create_module */ +__SYSCALL(127, sys_ni_syscall) +#define __NR_init_module 128 +__SYSCALL(__NR_init_module, sys_init_module) +#define __NR_delete_module 129 +__SYSCALL(__NR_delete_module, sys_delete_module) + /* 130 was sys_get_kernel_syms */ +__SYSCALL(130, sys_ni_syscall) +#define __NR_quotactl 131 +__SYSCALL(__NR_quotactl, sys_quotactl) +#define __NR_getpgid 132 +__SYSCALL(__NR_getpgid, sys_getpgid) +#define __NR_fchdir 133 +__SYSCALL(__NR_fchdir, sys_fchdir) +#define __NR_bdflush 134 +__SYSCALL(__NR_bdflush, sys_bdflush) +#define __NR_sysfs 135 +__SYSCALL(__NR_sysfs, sys_sysfs) +#define __NR_personality 136 +__SYSCALL(__NR_personality, sys_personality) + /* 137 was sys_afs_syscall */ +__SYSCALL(137, sys_ni_syscall) +#define __NR_setfsuid 138 +__SYSCALL(__NR_setfsuid, sys_setfsuid16) +#define __NR_setfsgid 139 +__SYSCALL(__NR_setfsgid, sys_setfsgid16) +#define __NR__llseek 140 +__SYSCALL(__NR__llseek, sys_llseek) +#define __NR_getdents 141 +__SYSCALL(__NR_getdents, compat_sys_getdents) +#define __NR__newselect 142 +__SYSCALL(__NR__newselect, compat_sys_select) +#define __NR_flock 143 +__SYSCALL(__NR_flock, sys_flock) +#define __NR_msync 144 +__SYSCALL(__NR_msync, sys_msync) +#define __NR_readv 145 +__SYSCALL(__NR_readv, compat_sys_readv) +#define __NR_writev 146 +__SYSCALL(__NR_writev, compat_sys_writev) +#define __NR_getsid 147 +__SYSCALL(__NR_getsid, sys_getsid) +#define __NR_fdatasync 148 +__SYSCALL(__NR_fdatasync, sys_fdatasync) +#define __NR__sysctl 149 +__SYSCALL(__NR__sysctl, compat_sys_sysctl) +#define __NR_mlock 150 +__SYSCALL(__NR_mlock, sys_mlock) +#define __NR_munlock 151 +__SYSCALL(__NR_munlock, sys_munlock) +#define __NR_mlockall 152 +__SYSCALL(__NR_mlockall, sys_mlockall) +#define __NR_munlockall 153 +__SYSCALL(__NR_munlockall, sys_munlockall) +#define __NR_sched_setparam 154 +__SYSCALL(__NR_sched_setparam, sys_sched_setparam) +#define __NR_sched_getparam 155 +__SYSCALL(__NR_sched_getparam, sys_sched_getparam) +#define __NR_sched_setscheduler 156 +__SYSCALL(__NR_sched_setscheduler, sys_sched_setscheduler) +#define __NR_sched_getscheduler 157 +__SYSCALL(__NR_sched_getscheduler, sys_sched_getscheduler) +#define __NR_sched_yield 158 +__SYSCALL(__NR_sched_yield, sys_sched_yield) +#define __NR_sched_get_priority_max 159 +__SYSCALL(__NR_sched_get_priority_max, sys_sched_get_priority_max) +#define __NR_sched_get_priority_min 160 +__SYSCALL(__NR_sched_get_priority_min, sys_sched_get_priority_min) +#define __NR_sched_rr_get_interval 161 +__SYSCALL(__NR_sched_rr_get_interval, compat_sys_sched_rr_get_interval) +#define __NR_nanosleep 162 +__SYSCALL(__NR_nanosleep, compat_sys_nanosleep) +#define __NR_mremap 163 +__SYSCALL(__NR_mremap, sys_mremap) +#define __NR_setresuid 164 +__SYSCALL(__NR_setresuid, sys_setresuid16) +#define __NR_getresuid 165 +__SYSCALL(__NR_getresuid, sys_getresuid16) + /* 166 was sys_vm86 */ +__SYSCALL(166, sys_ni_syscall) + /* 167 was sys_query_module */ +__SYSCALL(167, sys_ni_syscall) +#define __NR_poll 168 +__SYSCALL(__NR_poll, sys_poll) +#define __NR_nfsservctl 169 +__SYSCALL(__NR_nfsservctl, sys_ni_syscall) +#define __NR_setresgid 170 +__SYSCALL(__NR_setresgid, sys_setresgid16) +#define __NR_getresgid 171 +__SYSCALL(__NR_getresgid, sys_getresgid16) +#define __NR_prctl 172 +__SYSCALL(__NR_prctl, sys_prctl) +#define __NR_rt_sigreturn 173 +__SYSCALL(__NR_rt_sigreturn, compat_sys_rt_sigreturn_wrapper) +#define __NR_rt_sigaction 174 +__SYSCALL(__NR_rt_sigaction, compat_sys_rt_sigaction) +#define __NR_rt_sigprocmask 175 +__SYSCALL(__NR_rt_sigprocmask, compat_sys_rt_sigprocmask) +#define __NR_rt_sigpending 176 +__SYSCALL(__NR_rt_sigpending, compat_sys_rt_sigpending) +#define __NR_rt_sigtimedwait 177 +__SYSCALL(__NR_rt_sigtimedwait, compat_sys_rt_sigtimedwait) +#define __NR_rt_sigqueueinfo 178 +__SYSCALL(__NR_rt_sigqueueinfo, compat_sys_rt_sigqueueinfo) +#define __NR_rt_sigsuspend 179 +__SYSCALL(__NR_rt_sigsuspend, compat_sys_rt_sigsuspend) +#define __NR_pread64 180 +__SYSCALL(__NR_pread64, compat_sys_pread64_wrapper) +#define __NR_pwrite64 181 +__SYSCALL(__NR_pwrite64, compat_sys_pwrite64_wrapper) +#define __NR_chown 182 +__SYSCALL(__NR_chown, sys_chown16) +#define __NR_getcwd 183 +__SYSCALL(__NR_getcwd, sys_getcwd) +#define __NR_capget 184 +__SYSCALL(__NR_capget, sys_capget) +#define __NR_capset 185 +__SYSCALL(__NR_capset, sys_capset) +#define __NR_sigaltstack 186 +__SYSCALL(__NR_sigaltstack, compat_sys_sigaltstack) +#define __NR_sendfile 187 +__SYSCALL(__NR_sendfile, compat_sys_sendfile) + /* 188 reserved */ +__SYSCALL(188, sys_ni_syscall) + /* 189 reserved */ +__SYSCALL(189, sys_ni_syscall) +#define __NR_vfork 190 +__SYSCALL(__NR_vfork, sys_vfork) +#define __NR_ugetrlimit 191 /* SuS compliant getrlimit */ +__SYSCALL(__NR_ugetrlimit, compat_sys_getrlimit) /* SuS compliant getrlimit */ +#define __NR_mmap2 192 +__SYSCALL(__NR_mmap2, sys_mmap_pgoff) +#define __NR_truncate64 193 +__SYSCALL(__NR_truncate64, compat_sys_truncate64_wrapper) +#define __NR_ftruncate64 194 +__SYSCALL(__NR_ftruncate64, compat_sys_ftruncate64_wrapper) +#define __NR_stat64 195 +__SYSCALL(__NR_stat64, sys_stat64) +#define __NR_lstat64 196 +__SYSCALL(__NR_lstat64, sys_lstat64) +#define __NR_fstat64 197 +__SYSCALL(__NR_fstat64, sys_fstat64) +#define __NR_lchown32 198 +__SYSCALL(__NR_lchown32, sys_lchown) +#define __NR_getuid32 199 +__SYSCALL(__NR_getuid32, sys_getuid) +#define __NR_getgid32 200 +__SYSCALL(__NR_getgid32, sys_getgid) +#define __NR_geteuid32 201 +__SYSCALL(__NR_geteuid32, sys_geteuid) +#define __NR_getegid32 202 +__SYSCALL(__NR_getegid32, sys_getegid) +#define __NR_setreuid32 203 +__SYSCALL(__NR_setreuid32, sys_setreuid) +#define __NR_setregid32 204 +__SYSCALL(__NR_setregid32, sys_setregid) +#define __NR_getgroups32 205 +__SYSCALL(__NR_getgroups32, sys_getgroups) +#define __NR_setgroups32 206 +__SYSCALL(__NR_setgroups32, sys_setgroups) +#define __NR_fchown32 207 +__SYSCALL(__NR_fchown32, sys_fchown) +#define __NR_setresuid32 208 +__SYSCALL(__NR_setresuid32, sys_setresuid) +#define __NR_getresuid32 209 +__SYSCALL(__NR_getresuid32, sys_getresuid) +#define __NR_setresgid32 210 +__SYSCALL(__NR_setresgid32, sys_setresgid) +#define __NR_getresgid32 211 +__SYSCALL(__NR_getresgid32, sys_getresgid) +#define __NR_chown32 212 +__SYSCALL(__NR_chown32, sys_chown) +#define __NR_setuid32 213 +__SYSCALL(__NR_setuid32, sys_setuid) +#define __NR_setgid32 214 +__SYSCALL(__NR_setgid32, sys_setgid) +#define __NR_setfsuid32 215 +__SYSCALL(__NR_setfsuid32, sys_setfsuid) +#define __NR_setfsgid32 216 +__SYSCALL(__NR_setfsgid32, sys_setfsgid) +#define __NR_getdents64 217 +__SYSCALL(__NR_getdents64, compat_sys_getdents64) +#define __NR_pivot_root 218 +__SYSCALL(__NR_pivot_root, sys_pivot_root) +#define __NR_mincore 219 +__SYSCALL(__NR_mincore, sys_mincore) +#define __NR_madvise 220 +__SYSCALL(__NR_madvise, sys_madvise) +#define __NR_fcntl64 221 +__SYSCALL(__NR_fcntl64, compat_sys_fcntl64) + /* 222 for tux */ +__SYSCALL(222, sys_ni_syscall) + /* 223 is unused */ +__SYSCALL(223, sys_ni_syscall) +#define __NR_gettid 224 +__SYSCALL(__NR_gettid, sys_gettid) +#define __NR_readahead 225 +__SYSCALL(__NR_readahead, compat_sys_readahead_wrapper) +#define __NR_setxattr 226 +__SYSCALL(__NR_setxattr, sys_setxattr) +#define __NR_lsetxattr 227 +__SYSCALL(__NR_lsetxattr, sys_lsetxattr) +#define __NR_fsetxattr 228 +__SYSCALL(__NR_fsetxattr, sys_fsetxattr) +#define __NR_getxattr 229 +__SYSCALL(__NR_getxattr, sys_getxattr) +#define __NR_lgetxattr 230 +__SYSCALL(__NR_lgetxattr, sys_lgetxattr) +#define __NR_fgetxattr 231 +__SYSCALL(__NR_fgetxattr, sys_fgetxattr) +#define __NR_listxattr 232 +__SYSCALL(__NR_listxattr, sys_listxattr) +#define __NR_llistxattr 233 +__SYSCALL(__NR_llistxattr, sys_llistxattr) +#define __NR_flistxattr 234 +__SYSCALL(__NR_flistxattr, sys_flistxattr) +#define __NR_removexattr 235 +__SYSCALL(__NR_removexattr, sys_removexattr) +#define __NR_lremovexattr 236 +__SYSCALL(__NR_lremovexattr, sys_lremovexattr) +#define __NR_fremovexattr 237 +__SYSCALL(__NR_fremovexattr, sys_fremovexattr) +#define __NR_tkill 238 +__SYSCALL(__NR_tkill, sys_tkill) +#define __NR_sendfile64 239 +__SYSCALL(__NR_sendfile64, sys_sendfile64) +#define __NR_futex 240 +__SYSCALL(__NR_futex, compat_sys_futex) +#define __NR_sched_setaffinity 241 +__SYSCALL(__NR_sched_setaffinity, compat_sys_sched_setaffinity) +#define __NR_sched_getaffinity 242 +__SYSCALL(__NR_sched_getaffinity, compat_sys_sched_getaffinity) +#define __NR_io_setup 243 +__SYSCALL(__NR_io_setup, compat_sys_io_setup) +#define __NR_io_destroy 244 +__SYSCALL(__NR_io_destroy, sys_io_destroy) +#define __NR_io_getevents 245 +__SYSCALL(__NR_io_getevents, compat_sys_io_getevents) +#define __NR_io_submit 246 +__SYSCALL(__NR_io_submit, compat_sys_io_submit) +#define __NR_io_cancel 247 +__SYSCALL(__NR_io_cancel, sys_io_cancel) +#define __NR_exit_group 248 +__SYSCALL(__NR_exit_group, sys_exit_group) +#define __NR_lookup_dcookie 249 +__SYSCALL(__NR_lookup_dcookie, compat_sys_lookup_dcookie) +#define __NR_epoll_create 250 +__SYSCALL(__NR_epoll_create, sys_epoll_create) +#define __NR_epoll_ctl 251 +__SYSCALL(__NR_epoll_ctl, sys_epoll_ctl) +#define __NR_epoll_wait 252 +__SYSCALL(__NR_epoll_wait, sys_epoll_wait) +#define __NR_remap_file_pages 253 +__SYSCALL(__NR_remap_file_pages, sys_remap_file_pages) + /* 254 for set_thread_area */ +__SYSCALL(254, sys_ni_syscall) + /* 255 for get_thread_area */ +__SYSCALL(255, sys_ni_syscall) +#define __NR_set_tid_address 256 +__SYSCALL(__NR_set_tid_address, sys_set_tid_address) +#define __NR_timer_create 257 +__SYSCALL(__NR_timer_create, compat_sys_timer_create) +#define __NR_timer_settime 258 +__SYSCALL(__NR_timer_settime, compat_sys_timer_settime) +#define __NR_timer_gettime 259 +__SYSCALL(__NR_timer_gettime, compat_sys_timer_gettime) +#define __NR_timer_getoverrun 260 +__SYSCALL(__NR_timer_getoverrun, sys_timer_getoverrun) +#define __NR_timer_delete 261 +__SYSCALL(__NR_timer_delete, sys_timer_delete) +#define __NR_clock_settime 262 +__SYSCALL(__NR_clock_settime, compat_sys_clock_settime) +#define __NR_clock_gettime 263 +__SYSCALL(__NR_clock_gettime, compat_sys_clock_gettime) +#define __NR_clock_getres 264 +__SYSCALL(__NR_clock_getres, compat_sys_clock_getres) +#define __NR_clock_nanosleep 265 +__SYSCALL(__NR_clock_nanosleep, compat_sys_clock_nanosleep) +#define __NR_statfs64 266 +__SYSCALL(__NR_statfs64, compat_sys_statfs64_wrapper) +#define __NR_fstatfs64 267 +__SYSCALL(__NR_fstatfs64, compat_sys_fstatfs64_wrapper) +#define __NR_tgkill 268 +__SYSCALL(__NR_tgkill, sys_tgkill) +#define __NR_utimes 269 +__SYSCALL(__NR_utimes, compat_sys_utimes) +#define __NR_arm_fadvise64_64 270 +__SYSCALL(__NR_arm_fadvise64_64, compat_sys_fadvise64_64_wrapper) +#define __NR_pciconfig_iobase 271 +__SYSCALL(__NR_pciconfig_iobase, sys_pciconfig_iobase) +#define __NR_pciconfig_read 272 +__SYSCALL(__NR_pciconfig_read, sys_pciconfig_read) +#define __NR_pciconfig_write 273 +__SYSCALL(__NR_pciconfig_write, sys_pciconfig_write) +#define __NR_mq_open 274 +__SYSCALL(__NR_mq_open, compat_sys_mq_open) +#define __NR_mq_unlink 275 +__SYSCALL(__NR_mq_unlink, sys_mq_unlink) +#define __NR_mq_timedsend 276 +__SYSCALL(__NR_mq_timedsend, compat_sys_mq_timedsend) +#define __NR_mq_timedreceive 277 +__SYSCALL(__NR_mq_timedreceive, compat_sys_mq_timedreceive) +#define __NR_mq_notify 278 +__SYSCALL(__NR_mq_notify, compat_sys_mq_notify) +#define __NR_mq_getsetattr 279 +__SYSCALL(__NR_mq_getsetattr, compat_sys_mq_getsetattr) +#define __NR_waitid 280 +__SYSCALL(__NR_waitid, compat_sys_waitid) +#define __NR_socket 281 +__SYSCALL(__NR_socket, sys_socket) +#define __NR_bind 282 +__SYSCALL(__NR_bind, sys_bind) +#define __NR_connect 283 +__SYSCALL(__NR_connect, sys_connect) +#define __NR_listen 284 +__SYSCALL(__NR_listen, sys_listen) +#define __NR_accept 285 +__SYSCALL(__NR_accept, sys_accept) +#define __NR_getsockname 286 +__SYSCALL(__NR_getsockname, sys_getsockname) +#define __NR_getpeername 287 +__SYSCALL(__NR_getpeername, sys_getpeername) +#define __NR_socketpair 288 +__SYSCALL(__NR_socketpair, sys_socketpair) +#define __NR_send 289 +__SYSCALL(__NR_send, sys_send) +#define __NR_sendto 290 +__SYSCALL(__NR_sendto, sys_sendto) +#define __NR_recv 291 +__SYSCALL(__NR_recv, compat_sys_recv) +#define __NR_recvfrom 292 +__SYSCALL(__NR_recvfrom, compat_sys_recvfrom) +#define __NR_shutdown 293 +__SYSCALL(__NR_shutdown, sys_shutdown) +#define __NR_setsockopt 294 +__SYSCALL(__NR_setsockopt, compat_sys_setsockopt) +#define __NR_getsockopt 295 +__SYSCALL(__NR_getsockopt, compat_sys_getsockopt) +#define __NR_sendmsg 296 +__SYSCALL(__NR_sendmsg, compat_sys_sendmsg) +#define __NR_recvmsg 297 +__SYSCALL(__NR_recvmsg, compat_sys_recvmsg) +#define __NR_semop 298 +__SYSCALL(__NR_semop, sys_semop) +#define __NR_semget 299 +__SYSCALL(__NR_semget, sys_semget) +#define __NR_semctl 300 +__SYSCALL(__NR_semctl, compat_sys_semctl) +#define __NR_msgsnd 301 +__SYSCALL(__NR_msgsnd, compat_sys_msgsnd) +#define __NR_msgrcv 302 +__SYSCALL(__NR_msgrcv, compat_sys_msgrcv) +#define __NR_msgget 303 +__SYSCALL(__NR_msgget, sys_msgget) +#define __NR_msgctl 304 +__SYSCALL(__NR_msgctl, compat_sys_msgctl) +#define __NR_shmat 305 +__SYSCALL(__NR_shmat, compat_sys_shmat) +#define __NR_shmdt 306 +__SYSCALL(__NR_shmdt, sys_shmdt) +#define __NR_shmget 307 +__SYSCALL(__NR_shmget, sys_shmget) +#define __NR_shmctl 308 +__SYSCALL(__NR_shmctl, compat_sys_shmctl) +#define __NR_add_key 309 +__SYSCALL(__NR_add_key, sys_add_key) +#define __NR_request_key 310 +__SYSCALL(__NR_request_key, sys_request_key) +#define __NR_keyctl 311 +__SYSCALL(__NR_keyctl, compat_sys_keyctl) +#define __NR_semtimedop 312 +__SYSCALL(__NR_semtimedop, compat_sys_semtimedop) +#define __NR_vserver 313 +__SYSCALL(__NR_vserver, sys_ni_syscall) +#define __NR_ioprio_set 314 +__SYSCALL(__NR_ioprio_set, sys_ioprio_set) +#define __NR_ioprio_get 315 +__SYSCALL(__NR_ioprio_get, sys_ioprio_get) +#define __NR_inotify_init 316 +__SYSCALL(__NR_inotify_init, sys_inotify_init) +#define __NR_inotify_add_watch 317 +__SYSCALL(__NR_inotify_add_watch, sys_inotify_add_watch) +#define __NR_inotify_rm_watch 318 +__SYSCALL(__NR_inotify_rm_watch, sys_inotify_rm_watch) +#define __NR_mbind 319 +__SYSCALL(__NR_mbind, compat_sys_mbind) +#define __NR_get_mempolicy 320 +__SYSCALL(__NR_get_mempolicy, compat_sys_get_mempolicy) +#define __NR_set_mempolicy 321 +__SYSCALL(__NR_set_mempolicy, compat_sys_set_mempolicy) +#define __NR_openat 322 +__SYSCALL(__NR_openat, compat_sys_openat) +#define __NR_mkdirat 323 +__SYSCALL(__NR_mkdirat, sys_mkdirat) +#define __NR_mknodat 324 +__SYSCALL(__NR_mknodat, sys_mknodat) +#define __NR_fchownat 325 +__SYSCALL(__NR_fchownat, sys_fchownat) +#define __NR_futimesat 326 +__SYSCALL(__NR_futimesat, compat_sys_futimesat) +#define __NR_fstatat64 327 +__SYSCALL(__NR_fstatat64, sys_fstatat64) +#define __NR_unlinkat 328 +__SYSCALL(__NR_unlinkat, sys_unlinkat) +#define __NR_renameat 329 +__SYSCALL(__NR_renameat, sys_renameat) +#define __NR_linkat 330 +__SYSCALL(__NR_linkat, sys_linkat) +#define __NR_symlinkat 331 +__SYSCALL(__NR_symlinkat, sys_symlinkat) +#define __NR_readlinkat 332 +__SYSCALL(__NR_readlinkat, sys_readlinkat) +#define __NR_fchmodat 333 +__SYSCALL(__NR_fchmodat, sys_fchmodat) +#define __NR_faccessat 334 +__SYSCALL(__NR_faccessat, sys_faccessat) +#define __NR_pselect6 335 +__SYSCALL(__NR_pselect6, compat_sys_pselect6) +#define __NR_ppoll 336 +__SYSCALL(__NR_ppoll, compat_sys_ppoll) +#define __NR_unshare 337 +__SYSCALL(__NR_unshare, sys_unshare) +#define __NR_set_robust_list 338 +__SYSCALL(__NR_set_robust_list, compat_sys_set_robust_list) +#define __NR_get_robust_list 339 +__SYSCALL(__NR_get_robust_list, compat_sys_get_robust_list) +#define __NR_splice 340 +__SYSCALL(__NR_splice, sys_splice) +#define __NR_sync_file_range2 341 +__SYSCALL(__NR_sync_file_range2, compat_sys_sync_file_range2_wrapper) +#define __NR_tee 342 +__SYSCALL(__NR_tee, sys_tee) +#define __NR_vmsplice 343 +__SYSCALL(__NR_vmsplice, compat_sys_vmsplice) +#define __NR_move_pages 344 +__SYSCALL(__NR_move_pages, compat_sys_move_pages) +#define __NR_getcpu 345 +__SYSCALL(__NR_getcpu, sys_getcpu) +#define __NR_epoll_pwait 346 +__SYSCALL(__NR_epoll_pwait, compat_sys_epoll_pwait) +#define __NR_kexec_load 347 +__SYSCALL(__NR_kexec_load, compat_sys_kexec_load) +#define __NR_utimensat 348 +__SYSCALL(__NR_utimensat, compat_sys_utimensat) +#define __NR_signalfd 349 +__SYSCALL(__NR_signalfd, compat_sys_signalfd) +#define __NR_timerfd_create 350 +__SYSCALL(__NR_timerfd_create, sys_timerfd_create) +#define __NR_eventfd 351 +__SYSCALL(__NR_eventfd, sys_eventfd) +#define __NR_fallocate 352 +__SYSCALL(__NR_fallocate, compat_sys_fallocate_wrapper) +#define __NR_timerfd_settime 353 +__SYSCALL(__NR_timerfd_settime, compat_sys_timerfd_settime) +#define __NR_timerfd_gettime 354 +__SYSCALL(__NR_timerfd_gettime, compat_sys_timerfd_gettime) +#define __NR_signalfd4 355 +__SYSCALL(__NR_signalfd4, compat_sys_signalfd4) +#define __NR_eventfd2 356 +__SYSCALL(__NR_eventfd2, sys_eventfd2) +#define __NR_epoll_create1 357 +__SYSCALL(__NR_epoll_create1, sys_epoll_create1) +#define __NR_dup3 358 +__SYSCALL(__NR_dup3, sys_dup3) +#define __NR_pipe2 359 +__SYSCALL(__NR_pipe2, sys_pipe2) +#define __NR_inotify_init1 360 +__SYSCALL(__NR_inotify_init1, sys_inotify_init1) +#define __NR_preadv 361 +__SYSCALL(__NR_preadv, compat_sys_preadv) +#define __NR_pwritev 362 +__SYSCALL(__NR_pwritev, compat_sys_pwritev) +#define __NR_rt_tgsigqueueinfo 363 +__SYSCALL(__NR_rt_tgsigqueueinfo, compat_sys_rt_tgsigqueueinfo) +#define __NR_perf_event_open 364 +__SYSCALL(__NR_perf_event_open, sys_perf_event_open) +#define __NR_recvmmsg 365 +__SYSCALL(__NR_recvmmsg, compat_sys_recvmmsg) +#define __NR_accept4 366 +__SYSCALL(__NR_accept4, sys_accept4) +#define __NR_fanotify_init 367 +__SYSCALL(__NR_fanotify_init, sys_fanotify_init) +#define __NR_fanotify_mark 368 +__SYSCALL(__NR_fanotify_mark, compat_sys_fanotify_mark) +#define __NR_prlimit64 369 +__SYSCALL(__NR_prlimit64, sys_prlimit64) +#define __NR_name_to_handle_at 370 +__SYSCALL(__NR_name_to_handle_at, sys_name_to_handle_at) +#define __NR_open_by_handle_at 371 +__SYSCALL(__NR_open_by_handle_at, compat_sys_open_by_handle_at) +#define __NR_clock_adjtime 372 +__SYSCALL(__NR_clock_adjtime, compat_sys_clock_adjtime) +#define __NR_syncfs 373 +__SYSCALL(__NR_syncfs, sys_syncfs) +#define __NR_sendmmsg 374 +__SYSCALL(__NR_sendmmsg, compat_sys_sendmmsg) +#define __NR_setns 375 +__SYSCALL(__NR_setns, sys_setns) +#define __NR_process_vm_readv 376 +__SYSCALL(__NR_process_vm_readv, compat_sys_process_vm_readv) +#define __NR_process_vm_writev 377 +__SYSCALL(__NR_process_vm_writev, compat_sys_process_vm_writev) +#define __NR_kcmp 378 +__SYSCALL(__NR_kcmp, sys_kcmp) +#define __NR_finit_module 379 +__SYSCALL(__NR_finit_module, sys_finit_module) +/* #define __NR_sched_setattr 380 */ +__SYSCALL(380, sys_ni_syscall) +/* #define __NR_sched_getattr 381 */ +__SYSCALL(381, sys_ni_syscall) +/* #define __NR_renameat2 382 */ +__SYSCALL(382, sys_ni_syscall) +#define __NR_seccomp 383 +__SYSCALL(__NR_seccomp, sys_seccomp) diff --git a/arch/arm64/include/uapi/asm/ptrace.h b/arch/arm64/include/uapi/asm/ptrace.h index 6913643bbe54..49c61746297d 100644 --- a/arch/arm64/include/uapi/asm/ptrace.h +++ b/arch/arm64/include/uapi/asm/ptrace.h @@ -23,6 +23,7 @@ #include <asm/hwcap.h> +#define PTRACE_SET_SYSCALL 23 /* * PSR bits diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile index 12618cbd1bab..b52313653742 100644 --- a/arch/arm64/kernel/Makefile +++ b/arch/arm64/kernel/Makefile @@ -15,9 +15,10 @@ CFLAGS_REMOVE_return_address.o = -pg arm64-obj-y := cputable.o debug-monitors.o entry.o irq.o fpsimd.o \ entry-fpsimd.o process.o ptrace.o setup.o signal.o \ sys.o stacktrace.o time.o traps.o io.o vdso.o \ - hyp-stub.o psci.o psci-call.o cpu_ops.o insn.o return_address.o + hyp-stub.o psci.o psci-call.o cpu_ops.o insn.o return_address.o \ + opcodes.o -arm64-obj-$(CONFIG_COMPAT) += sys32.o kuser32.o signal32.o \ +arm64-obj-$(CONFIG_COMPAT) += sys32.o kuser32.o signal32.o \ sys_compat.o arm64-obj-$(CONFIG_FUNCTION_TRACER) += ftrace.o entry-ftrace.o arm64-obj-$(CONFIG_MODULES) += arm64ksyms.o module.o @@ -34,6 +35,8 @@ arm64-obj-$(CONFIG_CPU_IDLE) += cpuidle.o arm64-obj-$(CONFIG_KGDB) += kgdb.o arm64-obj-$(CONFIG_EFI) += efi.o efi-stub.o efi-entry.o +obj-$(CONFIG_SWP_EMULATE) += swp_emulate.o + obj-y += $(arm64-obj-y) vdso/ obj-m += $(arm64-obj-m) head-y := head.o diff --git a/arch/arm64/kernel/entry-fpsimd.S b/arch/arm64/kernel/entry-fpsimd.S index 6a27cd6dbfa6..d358ccacfc00 100644 --- a/arch/arm64/kernel/entry-fpsimd.S +++ b/arch/arm64/kernel/entry-fpsimd.S @@ -41,3 +41,27 @@ ENTRY(fpsimd_load_state) fpsimd_restore x0, 8 ret ENDPROC(fpsimd_load_state) + +#ifdef CONFIG_KERNEL_MODE_NEON + +/* + * Save the bottom n FP registers. + * + * x0 - pointer to struct fpsimd_partial_state + */ +ENTRY(fpsimd_save_partial_state) + fpsimd_save_partial x0, 1, 8, 9 + ret +ENDPROC(fpsimd_load_partial_state) + +/* + * Load the bottom n FP registers. + * + * x0 - pointer to struct fpsimd_partial_state + */ +ENTRY(fpsimd_load_partial_state) + fpsimd_restore_partial x0, 8, 9 + ret +ENDPROC(fpsimd_load_partial_state) + +#endif diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S index 1ee38bd61f03..bd30119fd0c3 100644 --- a/arch/arm64/kernel/entry.S +++ b/arch/arm64/kernel/entry.S @@ -25,9 +25,9 @@ #include <asm/asm-offsets.h> #include <asm/errno.h> #include <asm/esr.h> +#include <asm/ptrace.h> #include <asm/thread_info.h> #include <asm/unistd.h> -#include <asm/unistd32.h> /* * Bad Abort numbers @@ -601,7 +601,7 @@ fast_work_pending: str x0, [sp, #S_X0] // returned x0 work_pending: tbnz x1, #TIF_NEED_RESCHED, work_resched - /* TIF_SIGPENDING or TIF_NOTIFY_RESUME case */ + /* TIF_SIGPENDING, TIF_NOTIFY_RESUME or TIF_FOREIGN_FPSTATE case */ ldr x2, [sp, #S_PSTATE] mov x0, sp // 'regs' tst x2, #PSR_MODE_MASK // user mode regs? @@ -677,6 +677,10 @@ __sys_trace: mov x0, sp bl syscall_trace_enter adr lr, __sys_trace_return // return address + cmp w0, #RET_SKIP_SYSCALL_TRACE // skip syscall and tracing? + b.eq ret_to_user + cmp w0, #RET_SKIP_SYSCALL // skip syscall? + b.eq __sys_trace_return_skipped uxtw scno, w0 // syscall number (possibly new) mov x1, sp // pointer to regs cmp scno, sc_nr // check upper syscall limit @@ -690,6 +694,7 @@ __sys_trace: __sys_trace_return: str x0, [sp] // save returned x0 +__sys_trace_return_skipped: // x0 already in regs[0] mov x0, sp bl syscall_trace_exit b ret_to_user diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c index 522df9c7f3a4..5ba0217df39b 100644 --- a/arch/arm64/kernel/fpsimd.c +++ b/arch/arm64/kernel/fpsimd.c @@ -34,6 +34,60 @@ #define FPEXC_IDF (1 << 7) /* + * In order to reduce the number of times the FPSIMD state is needlessly saved + * and restored, we need to keep track of two things: + * (a) for each task, we need to remember which CPU was the last one to have + * the task's FPSIMD state loaded into its FPSIMD registers; + * (b) for each CPU, we need to remember which task's userland FPSIMD state has + * been loaded into its FPSIMD registers most recently, or whether it has + * been used to perform kernel mode NEON in the meantime. + * + * For (a), we add a 'cpu' field to struct fpsimd_state, which gets updated to + * the id of the current CPU everytime the state is loaded onto a CPU. For (b), + * we add the per-cpu variable 'fpsimd_last_state' (below), which contains the + * address of the userland FPSIMD state of the task that was loaded onto the CPU + * the most recently, or NULL if kernel mode NEON has been performed after that. + * + * With this in place, we no longer have to restore the next FPSIMD state right + * when switching between tasks. Instead, we can defer this check to userland + * resume, at which time we verify whether the CPU's fpsimd_last_state and the + * task's fpsimd_state.cpu are still mutually in sync. If this is the case, we + * can omit the FPSIMD restore. + * + * As an optimization, we use the thread_info flag TIF_FOREIGN_FPSTATE to + * indicate whether or not the userland FPSIMD state of the current task is + * present in the registers. The flag is set unless the FPSIMD registers of this + * CPU currently contain the most recent userland FPSIMD state of the current + * task. + * + * For a certain task, the sequence may look something like this: + * - the task gets scheduled in; if both the task's fpsimd_state.cpu field + * contains the id of the current CPU, and the CPU's fpsimd_last_state per-cpu + * variable points to the task's fpsimd_state, the TIF_FOREIGN_FPSTATE flag is + * cleared, otherwise it is set; + * + * - the task returns to userland; if TIF_FOREIGN_FPSTATE is set, the task's + * userland FPSIMD state is copied from memory to the registers, the task's + * fpsimd_state.cpu field is set to the id of the current CPU, the current + * CPU's fpsimd_last_state pointer is set to this task's fpsimd_state and the + * TIF_FOREIGN_FPSTATE flag is cleared; + * + * - the task executes an ordinary syscall; upon return to userland, the + * TIF_FOREIGN_FPSTATE flag will still be cleared, so no FPSIMD state is + * restored; + * + * - the task executes a syscall which executes some NEON instructions; this is + * preceded by a call to kernel_neon_begin(), which copies the task's FPSIMD + * register contents to memory, clears the fpsimd_last_state per-cpu variable + * and sets the TIF_FOREIGN_FPSTATE flag; + * + * - the task gets preempted after kernel_neon_end() is called; as we have not + * returned from the 2nd syscall yet, TIF_FOREIGN_FPSTATE is still set so + * whatever is in the FPSIMD registers is not saved to memory, but discarded. + */ +static DEFINE_PER_CPU(struct fpsimd_state *, fpsimd_last_state); + +/* * Trapped FP/ASIMD access. */ void do_fpsimd_acc(unsigned int esr, struct pt_regs *regs) @@ -71,44 +125,140 @@ void do_fpsimd_exc(unsigned int esr, struct pt_regs *regs) void fpsimd_thread_switch(struct task_struct *next) { - /* check if not kernel threads */ - if (current->mm) + /* + * Save the current FPSIMD state to memory, but only if whatever is in + * the registers is in fact the most recent userland FPSIMD state of + * 'current'. + */ + if (current->mm && !test_thread_flag(TIF_FOREIGN_FPSTATE)) fpsimd_save_state(¤t->thread.fpsimd_state); - if (next->mm) - fpsimd_load_state(&next->thread.fpsimd_state); + + if (next->mm) { + /* + * If we are switching to a task whose most recent userland + * FPSIMD state is already in the registers of *this* cpu, + * we can skip loading the state from memory. Otherwise, set + * the TIF_FOREIGN_FPSTATE flag so the state will be loaded + * upon the next return to userland. + */ + struct fpsimd_state *st = &next->thread.fpsimd_state; + + if (__this_cpu_read(fpsimd_last_state) == st + && st->cpu == smp_processor_id()) + clear_ti_thread_flag(task_thread_info(next), + TIF_FOREIGN_FPSTATE); + else + set_ti_thread_flag(task_thread_info(next), + TIF_FOREIGN_FPSTATE); + } } void fpsimd_flush_thread(void) { preempt_disable(); memset(¤t->thread.fpsimd_state, 0, sizeof(struct fpsimd_state)); - fpsimd_load_state(¤t->thread.fpsimd_state); + set_thread_flag(TIF_FOREIGN_FPSTATE); preempt_enable(); } -#ifdef CONFIG_KERNEL_MODE_NEON +/* + * Save the userland FPSIMD state of 'current' to memory, but only if the state + * currently held in the registers does in fact belong to 'current' + */ +void fpsimd_preserve_current_state(void) +{ + preempt_disable(); + if (!test_thread_flag(TIF_FOREIGN_FPSTATE)) + fpsimd_save_state(¤t->thread.fpsimd_state); + preempt_enable(); +} /* - * Kernel-side NEON support functions + * Load the userland FPSIMD state of 'current' from memory, but only if the + * FPSIMD state already held in the registers is /not/ the most recent FPSIMD + * state of 'current' */ -void kernel_neon_begin(void) +void fpsimd_restore_current_state(void) { - /* Avoid using the NEON in interrupt context */ - BUG_ON(in_interrupt()); preempt_disable(); + if (test_and_clear_thread_flag(TIF_FOREIGN_FPSTATE)) { + struct fpsimd_state *st = ¤t->thread.fpsimd_state; - if (current->mm) - fpsimd_save_state(¤t->thread.fpsimd_state); + fpsimd_load_state(st); + this_cpu_write(fpsimd_last_state, st); + st->cpu = smp_processor_id(); + } + preempt_enable(); } -EXPORT_SYMBOL(kernel_neon_begin); -void kernel_neon_end(void) +/* + * Load an updated userland FPSIMD state for 'current' from memory and set the + * flag that indicates that the FPSIMD register contents are the most recent + * FPSIMD state of 'current' + */ +void fpsimd_update_current_state(struct fpsimd_state *state) { - if (current->mm) - fpsimd_load_state(¤t->thread.fpsimd_state); + preempt_disable(); + fpsimd_load_state(state); + if (test_and_clear_thread_flag(TIF_FOREIGN_FPSTATE)) { + struct fpsimd_state *st = ¤t->thread.fpsimd_state; + this_cpu_write(fpsimd_last_state, st); + st->cpu = smp_processor_id(); + } preempt_enable(); } + +/* + * Invalidate live CPU copies of task t's FPSIMD state + */ +void fpsimd_flush_task_state(struct task_struct *t) +{ + t->thread.fpsimd_state.cpu = NR_CPUS; +} + +#ifdef CONFIG_KERNEL_MODE_NEON + +static DEFINE_PER_CPU(struct fpsimd_partial_state, hardirq_fpsimdstate); +static DEFINE_PER_CPU(struct fpsimd_partial_state, softirq_fpsimdstate); + +/* + * Kernel-side NEON support functions + */ +void kernel_neon_begin_partial(u32 num_regs) +{ + if (in_interrupt()) { + struct fpsimd_partial_state *s = this_cpu_ptr( + in_irq() ? &hardirq_fpsimdstate : &softirq_fpsimdstate); + + BUG_ON(num_regs > 32); + fpsimd_save_partial_state(s, roundup(num_regs, 2)); + } else { + /* + * Save the userland FPSIMD state if we have one and if we + * haven't done so already. Clear fpsimd_last_state to indicate + * that there is no longer userland FPSIMD state in the + * registers. + */ + preempt_disable(); + if (current->mm && + !test_and_set_thread_flag(TIF_FOREIGN_FPSTATE)) + fpsimd_save_state(¤t->thread.fpsimd_state); + this_cpu_write(fpsimd_last_state, NULL); + } +} +EXPORT_SYMBOL(kernel_neon_begin_partial); + +void kernel_neon_end(void) +{ + if (in_interrupt()) { + struct fpsimd_partial_state *s = this_cpu_ptr( + in_irq() ? &hardirq_fpsimdstate : &softirq_fpsimdstate); + fpsimd_load_partial_state(s); + } else { + preempt_enable(); + } +} EXPORT_SYMBOL(kernel_neon_end); #endif /* CONFIG_KERNEL_MODE_NEON */ @@ -119,12 +269,13 @@ static int fpsimd_cpu_pm_notifier(struct notifier_block *self, { switch (cmd) { case CPU_PM_ENTER: - if (current->mm) + if (current->mm && !test_thread_flag(TIF_FOREIGN_FPSTATE)) fpsimd_save_state(¤t->thread.fpsimd_state); + this_cpu_write(fpsimd_last_state, NULL); break; case CPU_PM_EXIT: if (current->mm) - fpsimd_load_state(¤t->thread.fpsimd_state); + set_thread_flag(TIF_FOREIGN_FPSTATE); break; case CPU_PM_ENTER_FAILED: default: diff --git a/arch/arm64/kernel/kuser32.S b/arch/arm64/kernel/kuser32.S index 7787208e8cc6..9fb6d5a3cea7 100644 --- a/arch/arm64/kernel/kuser32.S +++ b/arch/arm64/kernel/kuser32.S @@ -28,6 +28,7 @@ * See Documentation/arm/kernel_user_helpers.txt for formal definitions. */ +#include <asm/unistd.h> #include <asm/unistd32.h> .align 5 diff --git a/arch/arm64/kernel/opcodes.c b/arch/arm64/kernel/opcodes.c new file mode 100644 index 000000000000..ceb5a04a1e12 --- /dev/null +++ b/arch/arm64/kernel/opcodes.c @@ -0,0 +1,72 @@ +/* + * Copied from linux/arch/arm/kernel/opcodes.c + * + * A32 condition code lookup feature moved from nwfpe/fpopcode.c + * + * 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/module.h> +#include <asm/opcodes.h> + +#define ARM_OPCODE_CONDITION_UNCOND 0xf + +/* + * condition code lookup table + * index into the table is test code: EQ, NE, ... LT, GT, AL, NV + * + * bit position in short is condition code: NZCV + */ +static const unsigned short cc_map[16] = { + 0xF0F0, /* EQ == Z set */ + 0x0F0F, /* NE */ + 0xCCCC, /* CS == C set */ + 0x3333, /* CC */ + 0xFF00, /* MI == N set */ + 0x00FF, /* PL */ + 0xAAAA, /* VS == V set */ + 0x5555, /* VC */ + 0x0C0C, /* HI == C set && Z clear */ + 0xF3F3, /* LS == C clear || Z set */ + 0xAA55, /* GE == (N==V) */ + 0x55AA, /* LT == (N!=V) */ + 0x0A05, /* GT == (!Z && (N==V)) */ + 0xF5FA, /* LE == (Z || (N!=V)) */ + 0xFFFF, /* AL always */ + 0 /* NV */ +}; + +/* + * Returns: + * ARM_OPCODE_CONDTEST_FAIL - if condition fails + * ARM_OPCODE_CONDTEST_PASS - if condition passes (including AL) + * ARM_OPCODE_CONDTEST_UNCOND - if NV condition, or separate unconditional + * opcode space from v5 onwards + * + * Code that tests whether a conditional instruction would pass its condition + * check should check that return value == ARM_OPCODE_CONDTEST_PASS. + * + * Code that tests if a condition means that the instruction would be executed + * (regardless of conditional or unconditional) should instead check that the + * return value != ARM_OPCODE_CONDTEST_FAIL. + */ +asmlinkage unsigned int arm_check_condition(u32 opcode, u64 psr) +{ + u32 cc_bits = opcode >> 28; + u32 psr_cond = (u32)(psr & 0xffffffff) >> 28; + unsigned int ret; + + if (cc_bits != ARM_OPCODE_CONDITION_UNCOND) { + if ((cc_map[cc_bits] >> (psr_cond)) & 1) + ret = ARM_OPCODE_CONDTEST_PASS; + else + ret = ARM_OPCODE_CONDTEST_FAIL; + } else { + ret = ARM_OPCODE_CONDTEST_UNCOND; + } + + return ret; +} +EXPORT_SYMBOL_GPL(arm_check_condition); diff --git a/arch/arm64/kernel/process.c b/arch/arm64/kernel/process.c index a82af49bcc51..8e084d9cd7ec 100644 --- a/arch/arm64/kernel/process.c +++ b/arch/arm64/kernel/process.c @@ -141,6 +141,70 @@ 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; @@ -167,6 +231,8 @@ 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"); } @@ -213,7 +279,7 @@ void release_thread(struct task_struct *dead_task) int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src) { - fpsimd_save_state(¤t->thread.fpsimd_state); + fpsimd_preserve_current_state(); *dst = *src; return 0; } diff --git a/arch/arm64/kernel/ptrace.c b/arch/arm64/kernel/ptrace.c index 5cafe0e40cfa..1267de080feb 100644 --- a/arch/arm64/kernel/ptrace.c +++ b/arch/arm64/kernel/ptrace.c @@ -19,6 +19,7 @@ * along with this program. If not, see <http://www.gnu.org/licenses/>. */ +#include <linux/audit.h> #include <linux/compat.h> #include <linux/kernel.h> #include <linux/sched.h> @@ -26,6 +27,7 @@ #include <linux/smp.h> #include <linux/ptrace.h> #include <linux/user.h> +#include <linux/seccomp.h> #include <linux/security.h> #include <linux/init.h> #include <linux/signal.h> @@ -39,6 +41,7 @@ #include <asm/compat.h> #include <asm/debug-monitors.h> #include <asm/pgtable.h> +#include <asm/syscall.h> #include <asm/traps.h> #include <asm/system_misc.h> @@ -522,6 +525,7 @@ static int fpr_set(struct task_struct *target, const struct user_regset *regset, return ret; target->thread.fpsimd_state.user_fpsimd = newstate; + fpsimd_flush_task_state(target); return ret; } @@ -783,6 +787,7 @@ static int compat_vfp_set(struct task_struct *target, uregs->fpcr = fpscr & VFP_FPSCR_CTRL_MASK; } + fpsimd_flush_task_state(target); return ret; } @@ -1078,7 +1083,19 @@ const struct user_regset_view *task_user_regset_view(struct task_struct *task) long arch_ptrace(struct task_struct *child, long request, unsigned long addr, unsigned long data) { - return ptrace_request(child, request, addr, data); + int ret; + + switch (request) { + case PTRACE_SET_SYSCALL: + task_pt_regs(child)->syscallno = data; + ret = 0; + break; + default: + ret = ptrace_request(child, request, addr, data); + break; + } + + return ret; } enum ptrace_syscall_dir { @@ -1116,6 +1133,9 @@ asmlinkage int syscall_trace_enter(struct pt_regs *regs) if (test_thread_flag(TIF_SYSCALL_TRACEPOINT)) trace_sys_enter(regs, regs->syscallno); + audit_syscall_entry(syscall_get_arch(), regs->syscallno, + regs->orig_x0, regs->regs[1], regs->regs[2], regs->regs[3]); + return regs->syscallno; } @@ -1124,6 +1144,8 @@ asmlinkage void syscall_trace_exit(struct pt_regs *regs) if (test_thread_flag(TIF_SYSCALL_TRACEPOINT)) trace_sys_exit(regs, regs_return_value(regs)); + audit_syscall_exit(regs); + if (test_thread_flag(TIF_SYSCALL_TRACE)) tracehook_report_syscall(regs, PTRACE_SYSCALL_EXIT); } diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c index 9d704df65a6a..3d0b5a285061 100644 --- a/arch/arm64/kernel/setup.c +++ b/arch/arm64/kernel/setup.c @@ -547,7 +547,11 @@ static int c_show(struct seq_file *m, void *v) seq_puts(m, "\n"); seq_printf(m, "CPU implementer\t: 0x%02x\n", (midr >> 24)); - seq_printf(m, "CPU architecture: 8\n"); + seq_printf(m, "CPU architecture: %s\n", +#if IS_ENABLED(CONFIG_ARMV7_COMPAT_CPUINFO) + is_compat_task() ? "8" : +#endif + "AArch64"); seq_printf(m, "CPU variant\t: 0x%x\n", ((midr >> 20) & 0xf)); seq_printf(m, "CPU part\t: 0x%03x\n", ((midr >> 4) & 0xfff)); seq_printf(m, "CPU revision\t: %d\n\n", (midr & 0xf)); diff --git a/arch/arm64/kernel/signal.c b/arch/arm64/kernel/signal.c index e3cf09626245..bbc1aad21ce6 100644 --- a/arch/arm64/kernel/signal.c +++ b/arch/arm64/kernel/signal.c @@ -51,7 +51,7 @@ static int preserve_fpsimd_context(struct fpsimd_context __user *ctx) int err; /* dump the hardware registers to the fpsimd_state structure */ - fpsimd_save_state(fpsimd); + fpsimd_preserve_current_state(); /* copy the FP and status/control registers */ err = __copy_to_user(ctx->vregs, fpsimd->vregs, sizeof(fpsimd->vregs)); @@ -86,11 +86,8 @@ static int restore_fpsimd_context(struct fpsimd_context __user *ctx) __get_user_error(fpsimd.fpcr, &ctx->fpcr, err); /* load the hardware registers from the fpsimd_state structure */ - if (!err) { - preempt_disable(); - fpsimd_load_state(&fpsimd); - preempt_enable(); - } + if (!err) + fpsimd_update_current_state(&fpsimd); return err ? -EFAULT : 0; } @@ -423,4 +420,8 @@ asmlinkage void do_notify_resume(struct pt_regs *regs, clear_thread_flag(TIF_NOTIFY_RESUME); tracehook_notify_resume(regs); } + + if (thread_flags & _TIF_FOREIGN_FPSTATE) + fpsimd_restore_current_state(); + } diff --git a/arch/arm64/kernel/signal32.c b/arch/arm64/kernel/signal32.c index 96f545cfbb70..b2b945dcbf60 100644 --- a/arch/arm64/kernel/signal32.c +++ b/arch/arm64/kernel/signal32.c @@ -26,7 +26,7 @@ #include <asm/fpsimd.h> #include <asm/signal32.h> #include <asm/uaccess.h> -#include <asm/unistd32.h> +#include <asm/unistd.h> struct compat_sigcontext { /* We always set these two fields to 0 */ @@ -182,6 +182,14 @@ int copy_siginfo_to_user32(compat_siginfo_t __user *to, siginfo_t *from) err |= __put_user(from->si_uid, &to->si_uid); err |= __put_user(from->si_int, &to->si_int); break; +#ifdef __ARCH_SIGSYS + case __SI_SYS: + err |= __put_user((compat_uptr_t)(unsigned long) + from->si_call_addr, &to->si_call_addr); + err |= __put_user(from->si_syscall, &to->si_syscall); + err |= __put_user(from->si_arch, &to->si_arch); + break; +#endif default: /* this is just in case for now ... */ err |= __put_user(from->si_pid, &to->si_pid); err |= __put_user(from->si_uid, &to->si_uid); @@ -218,7 +226,7 @@ static int compat_preserve_vfp_context(struct compat_vfp_sigframe __user *frame) * Note that this also saves V16-31, which aren't visible * in AArch32. */ - fpsimd_save_state(fpsimd); + fpsimd_preserve_current_state(); /* Place structure header on the stack */ __put_user_error(magic, &frame->magic, err); @@ -281,11 +289,8 @@ static int compat_restore_vfp_context(struct compat_vfp_sigframe __user *frame) * We don't need to touch the exception register, so * reload the hardware state. */ - if (!err) { - preempt_disable(); - fpsimd_load_state(&fpsimd); - preempt_enable(); - } + if (!err) + fpsimd_update_current_state(&fpsimd); return err ? -EFAULT : 0; } diff --git a/arch/arm64/kernel/swp_emulate.c b/arch/arm64/kernel/swp_emulate.c new file mode 100644 index 000000000000..508fd2edb8ab --- /dev/null +++ b/arch/arm64/kernel/swp_emulate.c @@ -0,0 +1,223 @@ +/* + * Derived from from linux/arch/arm/kernel/swp_emulate.c + * + * Copyright (C) 2009 ARM Limited + * Copyright (c) 2014, NVIDIA CORPORATION. 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 as + * published by the Free Software Foundation. + * + * Implements emulation of the SWP/SWPB instructions using load-exclusive and + * store-exclusive for processors that have them disabled (or future ones that + * might not implement them). + * + * Syntax of SWP{B} instruction: SWP{B}<c> <Rt>, <Rt2>, [<Rn>] + * Where: Rt = destination + * Rt2 = source + * Rn = address + */ + +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/sched.h> +#include <linux/syscalls.h> +#include <linux/perf_event.h> + +#include <asm/opcodes.h> +#include <asm/traps.h> +#include <asm/uaccess.h> +#include <asm/system_misc.h> +#include <linux/debugfs.h> + +/* + * Error-checking SWP macros implemented using ldrex{b}/strex{b} + */ + +static int swpb(u8 in, u8 *out, u8 *addr) +{ + u8 _out; + int res; + int err; + + do { + __asm__ __volatile__( + "0: ldxrb %w1, %4\n" + "1: stxrb %w0, %w3, %4\n" + " mov %w2, #0\n" + "2:\n" + " .section .fixup,\"ax\"\n" + " .align 2\n" + "3: mov %w2, %5\n" + " b 2b\n" + " .previous\n" + " .section __ex_table,\"a\"\n" + " .align 3\n" + " .quad 0b, 3b\n" + " .quad 1b, 3b\n" + " .previous" + : "=&r" (res), "=r" (_out), "=r" (err) + : "r" (in), "Q" (*addr), "i" (-EFAULT) + : "cc", "memory"); + } while (err == 0 && res != 0); + + if (err == 0) + *out = _out; + return err; +} + +static int swp(u32 in, u32 *out, u32 *addr) +{ + u32 _out; + int res; + int err = 0; + + do { + __asm__ __volatile__( + "0: ldxr %w1, %4\n" + "1: stxr %w0, %w3, %4\n" + " mov %w2, #0\n" + "2:\n" + " .section .fixup,\"ax\"\n" + " .align 2\n" + "3: mov %w2, %5\n" + " b 2b\n" + " .previous\n" + " .section __ex_table,\"a\"\n" + " .align 3\n" + " .quad 0b, 3b\n" + " .quad 1b, 3b\n" + " .previous" + : "=&r" (res), "=r" (_out), "=r" (err) + : "r" (in), "Q" (*addr), "i" (-EFAULT) + : "cc", "memory"); + } while (err == 0 && res != 0); + + if (err == 0) + *out = _out; + return err; +} +/* + * Macros/defines for extracting register numbers from instruction. + */ +#define EXTRACT_REG_NUM(instruction, offset) \ + (((instruction) & (0xf << (offset))) >> (offset)) +#define RN_OFFSET 16 +#define RT_OFFSET 12 +#define RT2_OFFSET 0 +/* + * Bit 22 of the instruction encoding distinguishes between + * the SWP and SWPB variants (bit set means SWPB). + */ +#define TYPE_SWPB (1 << 22) + +static pid_t previous_pid; + +u64 swpb_count = 0; +u64 swp_count = 0; + +/* + * swp_handler logs the id of calling process, dissects the instruction, sanity + * checks the memory location, calls emulate_swpX for the actual operation and + * deals with fixup/error handling before returning + */ +static int swp_handler(struct pt_regs *regs, unsigned int instr) +{ + u32 destreg, data, type; + uintptr_t address; + unsigned int res = 0; + int err; + u32 temp32; + u8 temp8; + + perf_sw_event(PERF_COUNT_SW_EMULATION_FAULTS, 1, regs, regs->pc); + + res = arm_check_condition(instr, regs->pstate); + switch (res) { + case ARM_OPCODE_CONDTEST_PASS: + break; + case ARM_OPCODE_CONDTEST_FAIL: + /* Condition failed - return to next instruction */ + regs->pc += 4; + return 0; + case ARM_OPCODE_CONDTEST_UNCOND: + /* If unconditional encoding - not a SWP, undef */ + return -EFAULT; + default: + return -EINVAL; + } + + if (current->pid != previous_pid) { + pr_warn("\"%s\" (%ld) uses obsolete SWP{B} instruction\n", + current->comm, (unsigned long)current->pid); + previous_pid = current->pid; + } + + address = regs->regs[EXTRACT_REG_NUM(instr, RN_OFFSET)] & 0xffffffff; + data = regs->regs[EXTRACT_REG_NUM(instr, RT2_OFFSET)]; + destreg = EXTRACT_REG_NUM(instr, RT_OFFSET); + + type = instr & TYPE_SWPB; + + /* Check access in reasonable access range for both SWP and SWPB */ + if (!access_ok(VERIFY_WRITE, (address & ~3), 4)) { + pr_debug("SWP{B} emulation: access to %p not allowed!\n", + (void *)address); + res = -EFAULT; + } + if (type == TYPE_SWPB) { + err = swpb((u8) data, &temp8, (u8 *) address); + if (err) + return err; + regs->regs[destreg] = temp8; + regs->pc += 4; + swpb_count++; + } else if (address & 0x3) { + /* SWP to unaligned address not permitted */ + pr_debug("SWP instruction on unaligned pointer!\n"); + return -EFAULT; + } else { + err = swp((u32) data, &temp32, (u32 *) address); + if (err) + return err; + regs->regs[destreg] = temp32; + regs->pc += 4; + swp_count++; + } + + return 0; +} + +/* + * Only emulate SWP/SWPB executed in ARM state/User mode. + * The kernel must be SWP free and SWP{B} does not exist in Thumb/ThumbEE. + */ +static struct undef_hook swp_hook = { + .instr_mask = 0x0fb00ff0, + .instr_val = 0x01000090, + .pstate_mask = COMPAT_PSR_MODE_MASK | COMPAT_PSR_T_BIT, + .pstate_val = COMPAT_PSR_MODE_USR, + .fn = swp_handler +}; + +/* + * Register handler and create status file in /proc/cpu + * Invoked as late_initcall, since not needed before init spawned. + */ +static int __init swp_emulation_init(void) +{ + struct dentry *dir; + dir = debugfs_create_dir("swp_emulate", NULL); + debugfs_create_u64("swp_count", S_IRUGO | S_IWUSR, dir, &swp_count); + debugfs_create_u64("swpb_count", S_IRUGO | S_IWUSR, dir, &swpb_count); + + pr_notice("Registering SWP/SWPB emulation handler\n"); + register_undef_hook(&swp_hook); + + + return 0; +} + +late_initcall(swp_emulation_init); diff --git a/arch/arm64/kernel/sys_compat.c b/arch/arm64/kernel/sys_compat.c index 78039927c807..dc47e53e9e28 100644 --- a/arch/arm64/kernel/sys_compat.c +++ b/arch/arm64/kernel/sys_compat.c @@ -26,7 +26,7 @@ #include <linux/uaccess.h> #include <asm/cacheflush.h> -#include <asm/unistd32.h> +#include <asm/unistd.h> static inline void do_compat_cache_op(unsigned long start, unsigned long end, int flags) diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c index 12cd34ff35d0..2589ef53973a 100644 --- a/arch/arm64/kernel/traps.c +++ b/arch/arm64/kernel/traps.c @@ -3,6 +3,7 @@ * * Copyright (C) 1995-2009 Russell King * Copyright (C) 2012 ARM Ltd. + * Copyright (c) 2014, NVIDIA CORPORATION. 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 as @@ -256,15 +257,58 @@ void arm64_notify_die(const char *str, struct pt_regs *regs, die(str, regs, err); } +static LIST_HEAD(undef_hook); + +void register_undef_hook(struct undef_hook *hook) +{ + list_add(&hook->node, &undef_hook); +} + +static int call_undef_hook(struct pt_regs *regs, unsigned int instr) +{ + struct undef_hook *hook; + int (*fn)(struct pt_regs *regs, unsigned int instr) = NULL; + + list_for_each_entry(hook, &undef_hook, node) + if ((instr & hook->instr_mask) == hook->instr_val && + (regs->pstate & hook->pstate_mask) == hook->pstate_val) + fn = hook->fn; + + return fn ? fn(regs, instr) : 1; +} + asmlinkage void __exception do_undefinstr(struct pt_regs *regs) { + u32 instr; siginfo_t info; void __user *pc = (void __user *)instruction_pointer(regs); /* check for AArch32 breakpoint instructions */ if (!aarch32_break_handler(regs)) return; + if (user_mode(regs)) { + if (compat_thumb_mode(regs)) { + if (get_user(instr, (u16 __user *)pc)) + goto die_sig; + if (is_wide_instruction(instr)) { + u32 instr2; + if (get_user(instr2, (u16 __user *)pc+1)) + goto die_sig; + instr <<= 16; + instr |= instr2; + } + } else if (get_user(instr, (u32 __user *)pc)) { + goto die_sig; + } + } else { + /* kernel mode */ + instr = *((u32 *)pc); + } + + if (call_undef_hook(regs, instr) == 0) + return; +die_sig: if (show_unhandled_signals && unhandled_signal(current, SIGILL) && printk_ratelimit()) { pr_info("%s[%d]: undefined instruction: pc=%p\n", diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c index f2bded0254c3..9a9eaadb8634 100644 --- a/arch/arm64/mm/init.c +++ b/arch/arm64/mm/init.c @@ -119,9 +119,11 @@ 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_memory(pfn << PAGE_SHIFT); } EXPORT_SYMBOL(pfn_valid); #endif diff --git a/arch/arm64/mm/proc.S b/arch/arm64/mm/proc.S index c133ae53ed86..1cf94513b37f 100644 --- a/arch/arm64/mm/proc.S +++ b/arch/arm64/mm/proc.S @@ -240,6 +240,20 @@ ENTRY(__cpu_setup) ret // return to head.S ENDPROC(__cpu_setup) +#ifdef CONFIG_ARMV7_COMPAT + /* + * n n T + * U E WT T UD US IHBS + * CE0 XWHW CZ ME TEEA S + * .... .IEE .... NEAI TE.I ..AD DEN0 ACAM + * 0011 0... 1101 ..0. ..0. 10.. .... .... < hardware reserved + * .... .100 .... 01.1 11.1 ..01 0011 1101 < software settings + */ + .type crval, #object +crval: + .word 0x030802e2 // clear + .word 0x0405d03d // set +#else /* * n n T * U E WT T UD US IHBS @@ -252,3 +266,4 @@ ENDPROC(__cpu_setup) crval: .word 0x000802e2 // clear .word 0x0405d11d // set +#endif diff --git a/arch/ia64/include/asm/barrier.h b/arch/ia64/include/asm/barrier.h index 60576e06b6fb..d0a69aa35e27 100644 --- a/arch/ia64/include/asm/barrier.h +++ b/arch/ia64/include/asm/barrier.h @@ -45,14 +45,37 @@ # define smp_rmb() rmb() # define smp_wmb() wmb() # define smp_read_barrier_depends() read_barrier_depends() + #else + # define smp_mb() barrier() # define smp_rmb() barrier() # define smp_wmb() barrier() # define smp_read_barrier_depends() do { } while(0) + #endif /* + * IA64 GCC turns volatile stores into st.rel and volatile loads into ld.acq no + * need for asm trickery! + */ + +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ___p1; \ +}) + +/* * XXX check on this ---I suspect what Linus really wants here is * acquire vs release semantics but we can't discuss this stuff with * Linus just yet. Grrr... diff --git a/arch/metag/include/asm/barrier.h b/arch/metag/include/asm/barrier.h index e355a4c10968..2d6f0de77325 100644 --- a/arch/metag/include/asm/barrier.h +++ b/arch/metag/include/asm/barrier.h @@ -85,4 +85,19 @@ static inline void fence(void) #define smp_read_barrier_depends() do { } while (0) #define set_mb(var, value) do { var = value; smp_mb(); } while (0) +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ___p1; \ +}) + #endif /* _ASM_METAG_BARRIER_H */ diff --git a/arch/mips/include/asm/barrier.h b/arch/mips/include/asm/barrier.h index 314ab5532019..52c5b61d7aba 100644 --- a/arch/mips/include/asm/barrier.h +++ b/arch/mips/include/asm/barrier.h @@ -180,4 +180,19 @@ #define nudge_writes() mb() #endif +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ___p1; \ +}) + #endif /* __ASM_BARRIER_H */ diff --git a/arch/parisc/hpux/fs.c b/arch/parisc/hpux/fs.c index 838b479a42c4..88d0962de65a 100644 --- a/arch/parisc/hpux/fs.c +++ b/arch/parisc/hpux/fs.c @@ -60,6 +60,7 @@ struct hpux_dirent { }; struct getdents_callback { + struct dir_context ctx; struct hpux_dirent __user *current_dir; struct hpux_dirent __user *previous; int count; @@ -110,24 +111,23 @@ int hpux_getdents(unsigned int fd, struct hpux_dirent __user *dirent, unsigned i { struct fd arg; struct hpux_dirent __user * lastdirent; - struct getdents_callback buf; + struct getdents_callback buf = { + .ctx.actor = filldir, + .current_dir = dirent, + .count = count + }; int error; arg = fdget(fd); if (!arg.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(arg.file, filldir, &buf); + error = iterate_dir(arg.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(arg.file->f_pos, &lastdirent->d_off)) + if (put_user(buf.ctx.pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; diff --git a/arch/powerpc/include/asm/barrier.h b/arch/powerpc/include/asm/barrier.h index ae782254e731..f89da808ce31 100644 --- a/arch/powerpc/include/asm/barrier.h +++ b/arch/powerpc/include/asm/barrier.h @@ -45,11 +45,15 @@ # define SMPWMB eieio #endif +#define __lwsync() __asm__ __volatile__ (stringify_in_c(LWSYNC) : : :"memory") + #define smp_mb() mb() -#define smp_rmb() __asm__ __volatile__ (stringify_in_c(LWSYNC) : : :"memory") +#define smp_rmb() __lwsync() #define smp_wmb() __asm__ __volatile__ (stringify_in_c(SMPWMB) : : :"memory") #define smp_read_barrier_depends() read_barrier_depends() #else +#define __lwsync() barrier() + #define smp_mb() barrier() #define smp_rmb() barrier() #define smp_wmb() barrier() @@ -65,4 +69,19 @@ #define data_barrier(x) \ asm volatile("twi 0,%0,0; isync" : : "r" (x) : "memory"); +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + __lwsync(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + __lwsync(); \ + ___p1; \ +}) + #endif /* _ASM_POWERPC_BARRIER_H */ diff --git a/arch/s390/include/asm/barrier.h b/arch/s390/include/asm/barrier.h index 16760eeb79b0..578680f6207a 100644 --- a/arch/s390/include/asm/barrier.h +++ b/arch/s390/include/asm/barrier.h @@ -32,4 +32,19 @@ #define set_mb(var, value) do { var = value; mb(); } while (0) +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ___p1; \ +}) + #endif /* __ASM_BARRIER_H */ diff --git a/arch/s390/include/asm/syscall.h b/arch/s390/include/asm/syscall.h index cd29d2f4e4f3..bebc0bd8abc2 100644 --- a/arch/s390/include/asm/syscall.h +++ b/arch/s390/include/asm/syscall.h @@ -89,11 +89,10 @@ static inline void syscall_set_arguments(struct task_struct *task, regs->orig_gpr2 = args[0]; } -static inline int syscall_get_arch(struct task_struct *task, - struct pt_regs *regs) +static inline int syscall_get_arch(void) { #ifdef CONFIG_COMPAT - if (test_tsk_thread_flag(task, TIF_31BIT)) + if (test_tsk_thread_flag(current, TIF_31BIT)) return AUDIT_ARCH_S390; #endif return sizeof(long) == 8 ? AUDIT_ARCH_S390X : AUDIT_ARCH_S390; diff --git a/arch/sparc/include/asm/barrier_64.h b/arch/sparc/include/asm/barrier_64.h index 95d45986f908..b5aad964558e 100644 --- a/arch/sparc/include/asm/barrier_64.h +++ b/arch/sparc/include/asm/barrier_64.h @@ -53,4 +53,19 @@ do { __asm__ __volatile__("ba,pt %%xcc, 1f\n\t" \ #define smp_read_barrier_depends() do { } while(0) +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ___p1; \ +}) + #endif /* !(__SPARC64_BARRIER_H) */ diff --git a/arch/um/include/shared/frame_kern.h b/arch/um/include/shared/frame_kern.h index e584e40ee832..f2ca5702a4e2 100644 --- a/arch/um/include/shared/frame_kern.h +++ b/arch/um/include/shared/frame_kern.h @@ -6,13 +6,13 @@ #ifndef __FRAME_KERN_H_ #define __FRAME_KERN_H_ -extern int setup_signal_stack_sc(unsigned long stack_top, int sig, +extern int setup_signal_stack_sc(unsigned long stack_top, int sig, struct k_sigaction *ka, - struct pt_regs *regs, + struct pt_regs *regs, sigset_t *mask); -extern int setup_signal_stack_si(unsigned long stack_top, int sig, +extern int setup_signal_stack_si(unsigned long stack_top, int sig, struct k_sigaction *ka, - struct pt_regs *regs, siginfo_t *info, + struct pt_regs *regs, struct siginfo *info, sigset_t *mask); #endif diff --git a/arch/um/kernel/signal.c b/arch/um/kernel/signal.c index 3e831b3fd07b..f57e02e7910f 100644 --- a/arch/um/kernel/signal.c +++ b/arch/um/kernel/signal.c @@ -19,7 +19,7 @@ EXPORT_SYMBOL(unblock_signals); * OK, we're invoking a handler */ static void handle_signal(struct pt_regs *regs, unsigned long signr, - struct k_sigaction *ka, siginfo_t *info) + struct k_sigaction *ka, struct siginfo *info) { sigset_t *oldset = sigmask_to_save(); int singlestep = 0; @@ -71,7 +71,7 @@ static void handle_signal(struct pt_regs *regs, unsigned long signr, static int kern_do_signal(struct pt_regs *regs) { struct k_sigaction ka_copy; - siginfo_t info; + struct siginfo info; int sig, handled_sig = 0; while ((sig = get_signal_to_deliver(&info, &ka_copy, regs, NULL)) > 0) { diff --git a/arch/um/os-Linux/signal.c b/arch/um/os-Linux/signal.c index 9d9f1b4bf826..905924b773d3 100644 --- a/arch/um/os-Linux/signal.c +++ b/arch/um/os-Linux/signal.c @@ -25,7 +25,7 @@ void (*sig_info[NSIG])(int, struct siginfo *, struct uml_pt_regs *) = { [SIGIO] = sigio_handler, [SIGVTALRM] = timer_handler }; -static void sig_handler_common(int sig, siginfo_t *si, mcontext_t *mc) +static void sig_handler_common(int sig, struct siginfo *si, mcontext_t *mc) { struct uml_pt_regs r; int save_errno = errno; @@ -61,7 +61,7 @@ static void sig_handler_common(int sig, siginfo_t *si, mcontext_t *mc) static int signals_enabled; static unsigned int signals_pending; -void sig_handler(int sig, siginfo_t *si, mcontext_t *mc) +void sig_handler(int sig, struct siginfo *si, mcontext_t *mc) { int enabled; @@ -120,7 +120,7 @@ void set_sigstack(void *sig_stack, int size) panic("enabling signal stack failed, errno = %d\n", errno); } -static void (*handlers[_NSIG])(int sig, siginfo_t *si, mcontext_t *mc) = { +static void (*handlers[_NSIG])(int sig, struct siginfo *si, mcontext_t *mc) = { [SIGSEGV] = sig_handler, [SIGBUS] = sig_handler, [SIGILL] = sig_handler, @@ -162,7 +162,7 @@ static void hard_handler(int sig, siginfo_t *si, void *p) while ((sig = ffs(pending)) != 0){ sig--; pending &= ~(1 << sig); - (*handlers[sig])(sig, si, mc); + (*handlers[sig])(sig, (struct siginfo *)si, mc); } /* diff --git a/arch/um/os-Linux/skas/process.c b/arch/um/os-Linux/skas/process.c index 4625949bf1e4..908579f2b0ab 100644 --- a/arch/um/os-Linux/skas/process.c +++ b/arch/um/os-Linux/skas/process.c @@ -409,7 +409,7 @@ void userspace(struct uml_pt_regs *regs) if (WIFSTOPPED(status)) { int sig = WSTOPSIG(status); - ptrace(PTRACE_GETSIGINFO, pid, 0, &si); + ptrace(PTRACE_GETSIGINFO, pid, 0, (struct siginfo *)&si); switch (sig) { case SIGSEGV: @@ -417,7 +417,7 @@ void userspace(struct uml_pt_regs *regs) !ptrace_faultinfo) { get_skas_faultinfo(pid, ®s->faultinfo); - (*sig_info[SIGSEGV])(SIGSEGV, &si, + (*sig_info[SIGSEGV])(SIGSEGV, (struct siginfo *)&si, regs); } else handle_segv(pid, regs); @@ -426,14 +426,14 @@ void userspace(struct uml_pt_regs *regs) handle_trap(pid, regs, local_using_sysemu); break; case SIGTRAP: - relay_signal(SIGTRAP, &si, regs); + relay_signal(SIGTRAP, (struct siginfo *)&si, regs); break; case SIGVTALRM: now = os_nsecs(); if (now < nsecs) break; block_signals(); - (*sig_info[sig])(sig, &si, regs); + (*sig_info[sig])(sig, (struct siginfo *)&si, regs); unblock_signals(); nsecs = timer.it_value.tv_sec * UM_NSEC_PER_SEC + @@ -447,7 +447,7 @@ void userspace(struct uml_pt_regs *regs) case SIGFPE: case SIGWINCH: block_signals(); - (*sig_info[sig])(sig, &si, regs); + (*sig_info[sig])(sig, (struct siginfo *)&si, regs); unblock_signals(); break; default: diff --git a/arch/x86/include/asm/barrier.h b/arch/x86/include/asm/barrier.h index c6cd358a1eec..04a48903b2eb 100644 --- a/arch/x86/include/asm/barrier.h +++ b/arch/x86/include/asm/barrier.h @@ -92,12 +92,53 @@ #endif #define smp_read_barrier_depends() read_barrier_depends() #define set_mb(var, value) do { (void)xchg(&var, value); } while (0) -#else +#else /* !SMP */ #define smp_mb() barrier() #define smp_rmb() barrier() #define smp_wmb() barrier() #define smp_read_barrier_depends() do { } while (0) #define set_mb(var, value) do { var = value; barrier(); } while (0) +#endif /* SMP */ + +#if defined(CONFIG_X86_OOSTORE) || defined(CONFIG_X86_PPRO_FENCE) + +/* + * For either of these options x86 doesn't have a strong TSO memory + * model and we should fall back to full barriers. + */ + +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ___p1; \ +}) + +#else /* regular x86 TSO memory ordering */ + +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + barrier(); \ + ___p1; \ +}) + #endif /* 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/syscall.h b/arch/x86/include/asm/syscall.h index 2e188d68397c..f106908a12ec 100644 --- a/arch/x86/include/asm/syscall.h +++ b/arch/x86/include/asm/syscall.h @@ -90,8 +90,7 @@ static inline void syscall_set_arguments(struct task_struct *task, memcpy(®s->bx + i, args, n * sizeof(args[0])); } -static inline int syscall_get_arch(struct task_struct *task, - struct pt_regs *regs) +static inline int syscall_get_arch(void) { return AUDIT_ARCH_I386; } @@ -220,8 +219,7 @@ static inline void syscall_set_arguments(struct task_struct *task, } } -static inline int syscall_get_arch(struct task_struct *task, - struct pt_regs *regs) +static inline int syscall_get_arch(void) { #ifdef CONFIG_IA32_EMULATION /* @@ -233,7 +231,7 @@ static inline int syscall_get_arch(struct task_struct *task, * * x32 tasks should be considered AUDIT_ARCH_X86_64. */ - if (task_thread_info(task)->status & TS_COMPAT) + if (task_thread_info(current)->status & TS_COMPAT) return AUDIT_ARCH_I386; #endif /* Both x32 and x86_64 are considered "64-bit". */ diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 59b90379cb6a..48f439953436 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -40,19 +40,6 @@ DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_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 struct kmem_cache *task_xstate_cachep; @@ -257,14 +244,14 @@ 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/syscalls/syscall_32.tbl b/arch/x86/syscalls/syscall_32.tbl index aabfb8380a1c..01ed50255473 100644 --- a/arch/x86/syscalls/syscall_32.tbl +++ b/arch/x86/syscalls/syscall_32.tbl @@ -357,3 +357,7 @@ 348 i386 process_vm_writev sys_process_vm_writev compat_sys_process_vm_writev 349 i386 kcmp sys_kcmp 350 i386 finit_module sys_finit_module +# 351 i386 sched_setattr sys_sched_setattr +# 352 i386 sched_getattr sys_sched_getattr +# 353 i386 renameat2 sys_renameat2 +354 i386 seccomp sys_seccomp diff --git a/arch/x86/syscalls/syscall_64.tbl b/arch/x86/syscalls/syscall_64.tbl index 63a899304d27..c7b4ac76cd37 100644 --- a/arch/x86/syscalls/syscall_64.tbl +++ b/arch/x86/syscalls/syscall_64.tbl @@ -320,6 +320,10 @@ 311 64 process_vm_writev sys_process_vm_writev 312 common kcmp sys_kcmp 313 common finit_module sys_finit_module +# 314 common sched_setattr sys_sched_setattr +# 315 common sched_getattr sys_sched_getattr +# 316 common renameat2 sys_renameat2 +317 common seccomp sys_seccomp # # x32-specific system call numbers start at 512 to avoid cache impact diff --git a/arch/x86/um/signal.c b/arch/x86/um/signal.c index ae7319db18ee..5e04a1c899fa 100644 --- a/arch/x86/um/signal.c +++ b/arch/x86/um/signal.c @@ -508,7 +508,6 @@ int setup_signal_stack_si(unsigned long stack_top, int sig, { struct rt_sigframe __user *frame; int err = 0; - struct task_struct *me = current; frame = (struct rt_sigframe __user *) round_down(stack_top - sizeof(struct rt_sigframe), 16); diff --git a/block/genhd.c b/block/genhd.c index ea49f8631c22..2d350aae2eb6 100644 --- a/block/genhd.c +++ b/block/genhd.c @@ -1116,6 +1116,22 @@ 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", }; @@ -1135,6 +1151,7 @@ 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 0d9e5f97f0a8..47284e712650 100644 --- a/block/partition-generic.c +++ b/block/partition-generic.c @@ -217,10 +217,21 @@ 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 a1eba1845367..fa7e2a8cc608 100644 --- a/crypto/Kconfig +++ b/crypto/Kconfig @@ -533,6 +533,17 @@ config CRYPTO_SHA1_PPC This is the powerpc hardware accelerated implementation of the SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2). +config CRYPTO_SHA1_ARM_NEON + tristate "SHA1 digest algorithm (ARM NEON)" + depends on ARM && KERNEL_MODE_NEON && !CPU_BIG_ENDIAN + select CRYPTO_SHA1_ARM + select CRYPTO_SHA1 + select CRYPTO_HASH + help + SHA-1 secure hash standard (FIPS 180-1/DFIPS 180-2) implemented + using optimized ARM NEON assembly, when NEON instructions are + available. + config CRYPTO_SHA256 tristate "SHA224 and SHA256 digest algorithm" select CRYPTO_HASH @@ -545,6 +556,13 @@ config CRYPTO_SHA256 This code also includes SHA-224, a 224 bit hash with 112 bits of security against collision attacks. +config CRYPTO_SHA256_ARM + tristate "SHA-224/256 digest algorithm (ARM-asm and NEON)" + select CRYPTO_HASH + help + SHA-256 secure hash standard (DFIPS 180-2) implemented + using optimized ARM assembler and NEON, when available. + config CRYPTO_SHA256_SPARC64 tristate "SHA224 and SHA256 digest algorithm (SPARC64)" depends on SPARC64 @@ -566,6 +584,21 @@ config CRYPTO_SHA512 This code also includes SHA-384, a 384 bit hash with 192 bits of security against collision attacks. +config CRYPTO_SHA512_ARM_NEON + tristate "SHA384 and SHA512 digest algorithm (ARM NEON)" + depends on ARM && KERNEL_MODE_NEON && !CPU_BIG_ENDIAN + select CRYPTO_SHA512 + select CRYPTO_HASH + help + SHA-512 secure hash standard (DFIPS 180-2) implemented + using ARM NEON instructions, when available. + + This version of SHA implements a 512 bit hash with 256 bits of + security against collision attacks. + + This code also includes SHA-384, a 384 bit hash with 192 bits + of security against collision attacks. + config CRYPTO_SHA512_SPARC64 tristate "SHA384 and SHA512 digest algorithm (SPARC64)" depends on SPARC64 @@ -761,6 +794,22 @@ config CRYPTO_AES_ARM See <http://csrc.nist.gov/encryption/aes/> for more information. +config CRYPTO_AES_ARM_BS + tristate "Bit sliced AES using NEON instructions" + depends on ARM && KERNEL_MODE_NEON + select CRYPTO_ALGAPI + select CRYPTO_AES_ARM + select CRYPTO_ABLK_HELPER + help + Use a faster and more secure NEON based implementation of AES in CBC, + CTR and XTS modes + + Bit sliced AES gives around 45% speedup on Cortex-A15 for CTR mode + and for XTS mode encryption, CBC and XTS mode decryption speedup is + around 25%. (CBC encryption speed is not affected by this driver.) + This implementation does not rely on any lookup tables so it is + believed to be invulnerable to cache timing attacks. + config CRYPTO_ANUBIS tristate "Anubis cipher algorithm" select CRYPTO_ALGAPI diff --git a/crypto/ablk_helper.c b/crypto/ablk_helper.c index 62568b1fc885..ffe7278d4bd8 100644 --- a/crypto/ablk_helper.c +++ b/crypto/ablk_helper.c @@ -75,7 +75,7 @@ int ablk_encrypt(struct ablkcipher_request *req) struct ablkcipher_request *cryptd_req = ablkcipher_request_ctx(req); - memcpy(cryptd_req, req, sizeof(*req)); + *cryptd_req = *req; ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); return crypto_ablkcipher_encrypt(cryptd_req); @@ -94,7 +94,7 @@ int ablk_decrypt(struct ablkcipher_request *req) struct ablkcipher_request *cryptd_req = ablkcipher_request_ctx(req); - memcpy(cryptd_req, req, sizeof(*req)); + *cryptd_req = *req; ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); return crypto_ablkcipher_decrypt(cryptd_req); diff --git a/drivers/Kconfig b/drivers/Kconfig index d27feb5460f3..4f990ce9515a 100644 --- a/drivers/Kconfig +++ b/drivers/Kconfig @@ -100,6 +100,8 @@ source "drivers/memstick/Kconfig" source "drivers/leds/Kconfig" +source "drivers/switch/Kconfig" + source "drivers/accessibility/Kconfig" source "drivers/infiniband/Kconfig" @@ -168,4 +170,6 @@ source "drivers/reset/Kconfig" source "drivers/gator/Kconfig" +source "drivers/android/Kconfig" + endmenu diff --git a/drivers/Makefile b/drivers/Makefile index 292d528e22f1..3774b56c5d80 100644 --- a/drivers/Makefile +++ b/drivers/Makefile @@ -111,6 +111,7 @@ obj-$(CONFIG_CPU_IDLE) += cpuidle/ obj-y += mmc/ obj-$(CONFIG_MEMSTICK) += memstick/ obj-y += leds/ +obj-$(CONFIG_SWITCH) += switch/ obj-$(CONFIG_INFINIBAND) += infiniband/ obj-$(CONFIG_SGI_SN) += sn/ obj-y += firmware/ @@ -155,3 +156,4 @@ obj-$(CONFIG_NTB) += ntb/ obj-$(CONFIG_GATOR) += gator/ obj-$(CONFIG_CORESIGHT) += coresight/ +obj-$(CONFIG_ANDROID) += android/ diff --git a/drivers/android/Kconfig b/drivers/android/Kconfig new file mode 100644 index 000000000000..bdfc6c6f4f5a --- /dev/null +++ b/drivers/android/Kconfig @@ -0,0 +1,37 @@ +menu "Android" + +config ANDROID + bool "Android Drivers" + ---help--- + Enable support for various drivers needed on the Android platform + +if ANDROID + +config ANDROID_BINDER_IPC + bool "Android Binder IPC Driver" + depends on MMU + default n + ---help--- + Binder is used in Android for both communication between processes, + and remote method invocation. + + This means one Android process can call a method/routine in another + Android process, using Binder to identify, invoke and pass arguments + between said processes. + +config ANDROID_BINDER_IPC_32BIT + bool + depends on !64BIT && ANDROID_BINDER_IPC + default y + ---help--- + The Binder API has been changed to support both 32 and 64bit + applications in a mixed environment. + + Enable this to support an old 32-bit Android user-space (v4.4 and + earlier). + + Note that enabling this will break newer Android user-space. + +endif # if ANDROID + +endmenu diff --git a/drivers/android/Makefile b/drivers/android/Makefile new file mode 100644 index 000000000000..3b7e4b072c58 --- /dev/null +++ b/drivers/android/Makefile @@ -0,0 +1,3 @@ +ccflags-y += -I$(src) # needed for trace events + +obj-$(CONFIG_ANDROID_BINDER_IPC) += binder.o diff --git a/drivers/staging/android/binder.c b/drivers/android/binder.c index 0fce5fc9923b..c048624fc6b7 100644 --- a/drivers/staging/android/binder.c +++ b/drivers/android/binder.c @@ -20,11 +20,13 @@ #include <asm/cacheflush.h> #include <linux/fdtable.h> #include <linux/file.h> +#include <linux/freezer.h> #include <linux/fs.h> #include <linux/list.h> #include <linux/miscdevice.h> #include <linux/mm.h> #include <linux/module.h> +#include <linux/rtmutex.h> #include <linux/mutex.h> #include <linux/nsproxy.h> #include <linux/poll.h> @@ -36,11 +38,16 @@ #include <linux/vmalloc.h> #include <linux/slab.h> #include <linux/pid_namespace.h> +#include <linux/security.h> -#include "binder.h" +#ifdef CONFIG_ANDROID_BINDER_IPC_32BIT +#define BINDER_IPC_32BIT 1 +#endif + +#include <uapi/linux/android/binder.h> #include "binder_trace.h" -static DEFINE_MUTEX(binder_main_lock); +static DEFINE_RT_MUTEX(binder_main_lock); static DEFINE_MUTEX(binder_deferred_lock); static DEFINE_MUTEX(binder_mmap_lock); @@ -117,6 +124,7 @@ static int binder_set_stop_on_user_error(const char *val, struct kernel_param *kp) { int ret; + ret = param_set_int(val, kp); if (binder_stop_on_user_error < 2) wake_up(&binder_user_error_wait); @@ -193,6 +201,7 @@ static struct binder_transaction_log_entry *binder_transaction_log_add( struct binder_transaction_log *log) { struct binder_transaction_log_entry *e; + e = &log->entry[log->next]; memset(e, 0, sizeof(*e)); log->next++; @@ -227,8 +236,8 @@ struct binder_node { int internal_strong_refs; int local_weak_refs; int local_strong_refs; - void __user *ptr; - void __user *cookie; + binder_uintptr_t ptr; + binder_uintptr_t cookie; unsigned has_strong_ref:1; unsigned pending_strong_ref:1; unsigned has_weak_ref:1; @@ -241,7 +250,7 @@ struct binder_node { struct binder_ref_death { struct binder_work work; - void __user *cookie; + binder_uintptr_t cookie; }; struct binder_ref { @@ -418,19 +427,20 @@ static long task_close_fd(struct binder_proc *proc, unsigned int fd) static inline void binder_lock(const char *tag) { trace_binder_lock(tag); - mutex_lock(&binder_main_lock); + rt_mutex_lock(&binder_main_lock); trace_binder_locked(tag); } static inline void binder_unlock(const char *tag) { trace_binder_unlock(tag); - mutex_unlock(&binder_main_lock); + rt_mutex_unlock(&binder_main_lock); } static void binder_set_nice(long nice) { long min_nice; + if (can_nice(current, nice)) { set_user_nice(current, nice); return; @@ -450,9 +460,8 @@ static size_t binder_buffer_size(struct binder_proc *proc, { if (list_is_last(&buffer->entry, &proc->buffers)) return proc->buffer + proc->buffer_size - (void *)buffer->data; - else - return (size_t)list_entry(buffer->entry.next, - struct binder_buffer, entry) - (size_t)buffer->data; + return (size_t)list_entry(buffer->entry.next, + struct binder_buffer, entry) - (size_t)buffer->data; } static void binder_insert_free_buffer(struct binder_proc *proc, @@ -514,14 +523,14 @@ static void binder_insert_allocated_buffer(struct binder_proc *proc, } static struct binder_buffer *binder_buffer_lookup(struct binder_proc *proc, - void __user *user_ptr) + uintptr_t user_ptr) { struct rb_node *n = proc->allocated_buffers.rb_node; struct binder_buffer *buffer; struct binder_buffer *kern_ptr; - kern_ptr = user_ptr - proc->user_buffer_offset - - offsetof(struct binder_buffer, data); + 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); @@ -583,6 +592,7 @@ static int binder_update_page_range(struct binder_proc *proc, int allocate, for (page_addr = start; page_addr < end; page_addr += PAGE_SIZE) { int ret; struct page **page_array_ptr; + page = &proc->pages[(page_addr - proc->buffer) / PAGE_SIZE]; BUG_ON(*page); @@ -725,6 +735,7 @@ static struct binder_buffer *binder_alloc_buf(struct binder_proc *proc, binder_insert_allocated_buffer(proc, buffer); if (buffer_size != size) { struct binder_buffer *new_buffer = (void *)buffer->data + size; + list_add(&new_buffer->entry, &buffer->entry); new_buffer->free = 1; binder_insert_free_buffer(proc, new_buffer); @@ -790,7 +801,7 @@ static void binder_delete_free_buffer(struct binder_proc *proc, 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 with %p or %p\n", + "%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 ? @@ -837,6 +848,7 @@ static void binder_free_buf(struct binder_proc *proc, 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); @@ -845,6 +857,7 @@ static void binder_free_buf(struct binder_proc *proc, 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); @@ -855,7 +868,7 @@ static void binder_free_buf(struct binder_proc *proc, } static struct binder_node *binder_get_node(struct binder_proc *proc, - void __user *ptr) + binder_uintptr_t ptr) { struct rb_node *n = proc->nodes.rb_node; struct binder_node *node; @@ -874,8 +887,8 @@ static struct binder_node *binder_get_node(struct binder_proc *proc, } static struct binder_node *binder_new_node(struct binder_proc *proc, - void __user *ptr, - void __user *cookie) + binder_uintptr_t ptr, + binder_uintptr_t cookie) { struct rb_node **p = &proc->nodes.rb_node; struct rb_node *parent = NULL; @@ -907,9 +920,9 @@ static struct binder_node *binder_new_node(struct binder_proc *proc, INIT_LIST_HEAD(&node->work.entry); INIT_LIST_HEAD(&node->async_todo); binder_debug(BINDER_DEBUG_INTERNAL_REFS, - "%d:%d node %d u%p c%p created\n", + "%d:%d node %d u%016llx c%016llx created\n", proc->pid, current->pid, node->debug_id, - node->ptr, node->cookie); + (u64)node->ptr, (u64)node->cookie); return node; } @@ -1106,6 +1119,7 @@ static int binder_inc_ref(struct binder_ref *ref, int strong, struct list_head *target_list) { int ret; + if (strong) { if (ref->strong == 0) { ret = binder_inc_node(ref->node, 1, 1, target_list); @@ -1137,6 +1151,7 @@ static int binder_dec_ref(struct binder_ref *ref, int strong) ref->strong--; if (ref->strong == 0) { int ret; + ret = binder_dec_node(ref->node, strong, 1); if (ret) return ret; @@ -1176,6 +1191,8 @@ static void binder_send_failed_reply(struct binder_transaction *t, uint32_t error_code) { struct binder_thread *target_thread; + struct binder_transaction *next; + BUG_ON(t->flags & TF_ONE_WAY); while (1) { target_thread = t->from; @@ -1189,7 +1206,8 @@ static void binder_send_failed_reply(struct binder_transaction *t, 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, + t->debug_id, + target_thread->proc->pid, target_thread->pid); binder_pop_transaction(target_thread, t); @@ -1202,32 +1220,31 @@ static void binder_send_failed_reply(struct binder_transaction *t, target_thread->return_error); } return; - } else { - struct binder_transaction *next = t->from_parent; + } + next = t->from_parent; - binder_debug(BINDER_DEBUG_FAILED_TRANSACTION, - "send failed reply for transaction %d, target dead\n", - t->debug_id); + binder_debug(BINDER_DEBUG_FAILED_TRANSACTION, + "send failed reply for transaction %d, target dead\n", + t->debug_id); - binder_pop_transaction(target_thread, t); - if (next == NULL) { - binder_debug(BINDER_DEBUG_DEAD_BINDER, - "reply failed, no target thread at root\n"); - return; - } - t = next; + binder_pop_transaction(target_thread, t); + if (next == NULL) { binder_debug(BINDER_DEBUG_DEAD_BINDER, - "reply failed, no target thread -- retry %d\n", - t->debug_id); + "reply failed, no target thread at root\n"); + return; } + t = next; + binder_debug(BINDER_DEBUG_DEAD_BINDER, + "reply failed, no target thread -- retry %d\n", + t->debug_id); } } static void binder_transaction_buffer_release(struct binder_proc *proc, struct binder_buffer *buffer, - size_t *failed_at) + binder_size_t *failed_at) { - size_t *offp, *off_end; + binder_size_t *offp, *off_end; int debug_id = buffer->debug_id; binder_debug(BINDER_DEBUG_TRANSACTION, @@ -1238,18 +1255,20 @@ static void binder_transaction_buffer_release(struct binder_proc *proc, if (buffer->target_node) binder_dec_node(buffer->target_node, 1, 0); - offp = (size_t *)(buffer->data + ALIGN(buffer->data_size, sizeof(void *))); + offp = (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; + if (*offp > buffer->data_size - sizeof(*fp) || buffer->data_size < sizeof(*fp) || - !IS_ALIGNED(*offp, sizeof(void *))) { - pr_err("transaction release %d bad offset %zd, size %zd\n", - debug_id, *offp, buffer->data_size); + !IS_ALIGNED(*offp, sizeof(u32))) { + pr_err("transaction release %d bad offset %lld, size %zd\n", + debug_id, (u64)*offp, buffer->data_size); continue; } fp = (struct flat_binder_object *)(buffer->data + *offp); @@ -1257,21 +1276,23 @@ static void binder_transaction_buffer_release(struct binder_proc *proc, case BINDER_TYPE_BINDER: case BINDER_TYPE_WEAK_BINDER: { struct binder_node *node = binder_get_node(proc, fp->binder); + if (node == NULL) { - pr_err("transaction release %d bad node %p\n", - debug_id, fp->binder); + pr_err("transaction release %d bad node %016llx\n", + debug_id, (u64)fp->binder); break; } binder_debug(BINDER_DEBUG_TRANSACTION, - " node %d u%p\n", - node->debug_id, node->ptr); + " node %d u%016llx\n", + node->debug_id, (u64)node->ptr); binder_dec_node(node, fp->type == BINDER_TYPE_BINDER, 0); } break; case BINDER_TYPE_HANDLE: case BINDER_TYPE_WEAK_HANDLE: { struct binder_ref *ref = binder_get_ref(proc, fp->handle); + if (ref == NULL) { - pr_err("transaction release %d bad handle %ld\n", + pr_err("transaction release %d bad handle %d\n", debug_id, fp->handle); break; } @@ -1283,13 +1304,13 @@ static void binder_transaction_buffer_release(struct binder_proc *proc, case BINDER_TYPE_FD: binder_debug(BINDER_DEBUG_TRANSACTION, - " fd %ld\n", fp->handle); + " fd %d\n", fp->handle); if (failed_at) task_close_fd(proc, fp->handle); break; default: - pr_err("transaction release %d bad object type %lx\n", + pr_err("transaction release %d bad object type %x\n", debug_id, fp->type); break; } @@ -1302,7 +1323,8 @@ static void binder_transaction(struct binder_proc *proc, { struct binder_transaction *t; struct binder_work *tcomplete; - size_t *offp, *off_end; + binder_size_t *offp, *off_end; + binder_size_t off_min; struct binder_proc *target_proc; struct binder_thread *target_thread = NULL; struct binder_node *target_node = NULL; @@ -1361,6 +1383,7 @@ static void binder_transaction(struct binder_proc *proc, } else { if (tr->target.handle) { struct binder_ref *ref; + ref = binder_get_ref(proc, tr->target.handle); if (ref == NULL) { binder_user_error("%d:%d got transaction to invalid handle\n", @@ -1382,8 +1405,13 @@ static void binder_transaction(struct binder_proc *proc, return_error = BR_DEAD_REPLY; goto err_dead_binder; } + if (security_binder_transaction(proc->tsk, target_proc->tsk) < 0) { + return_error = BR_FAILED_REPLY; + goto err_invalid_target_handle; + } if (!(tr->flags & TF_ONE_WAY) && thread->transaction_stack) { struct binder_transaction *tmp; + tmp = thread->transaction_stack; if (tmp->to_thread != thread) { binder_user_error("%d:%d got new transaction with bad transaction stack, transaction %d has target %d:%d\n", @@ -1431,24 +1459,26 @@ static void binder_transaction(struct binder_proc *proc, if (reply) binder_debug(BINDER_DEBUG_TRANSACTION, - "%d:%d BC_REPLY %d -> %d:%d, data %p-%p size %zd-%zd\n", + "%d:%d BC_REPLY %d -> %d:%d, data %016llx-%016llx size %lld-%lld\n", proc->pid, thread->pid, t->debug_id, target_proc->pid, target_thread->pid, - tr->data.ptr.buffer, tr->data.ptr.offsets, - tr->data_size, tr->offsets_size); + (u64)tr->data.ptr.buffer, + (u64)tr->data.ptr.offsets, + (u64)tr->data_size, (u64)tr->offsets_size); else binder_debug(BINDER_DEBUG_TRANSACTION, - "%d:%d BC_TRANSACTION %d -> %d - node %d, data %p-%p size %zd-%zd\n", + "%d:%d BC_TRANSACTION %d -> %d - node %d, data %016llx-%016llx size %lld-%lld\n", proc->pid, thread->pid, t->debug_id, target_proc->pid, target_node->debug_id, - tr->data.ptr.buffer, tr->data.ptr.offsets, - tr->data_size, tr->offsets_size); + (u64)tr->data.ptr.buffer, + (u64)tr->data.ptr.offsets, + (u64)tr->data_size, (u64)tr->offsets_size); if (!reply && !(tr->flags & TF_ONE_WAY)) t->from = thread; else t->from = NULL; - t->sender_euid = proc->tsk->cred->euid; + t->sender_euid = task_euid(proc->tsk); t->to_proc = target_proc; t->to_thread = target_thread; t->code = tr->code; @@ -1471,43 +1501,54 @@ static void binder_transaction(struct binder_proc *proc, if (target_node) binder_inc_node(target_node, 1, 0, NULL); - offp = (size_t *)(t->buffer->data + ALIGN(tr->data_size, sizeof(void *))); + offp = (binder_size_t *)(t->buffer->data + + ALIGN(tr->data_size, sizeof(void *))); - if (copy_from_user(t->buffer->data, tr->data.ptr.buffer, tr->data_size)) { + 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; goto err_copy_data_failed; } - if (copy_from_user(offp, tr->data.ptr.offsets, tr->offsets_size)) { + if (copy_from_user(offp, (const void __user *)(uintptr_t) + tr->data.ptr.offsets, tr->offsets_size)) { binder_user_error("%d:%d got transaction with invalid offsets ptr\n", proc->pid, thread->pid); return_error = BR_FAILED_REPLY; goto err_copy_data_failed; } - if (!IS_ALIGNED(tr->offsets_size, sizeof(size_t))) { - binder_user_error("%d:%d got transaction with invalid offsets size, %zd\n", - proc->pid, thread->pid, tr->offsets_size); + 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; goto err_bad_offset; } off_end = (void *)offp + tr->offsets_size; + off_min = 0; for (; offp < off_end; offp++) { struct flat_binder_object *fp; + if (*offp > t->buffer->data_size - sizeof(*fp) || + *offp < off_min || t->buffer->data_size < sizeof(*fp) || - !IS_ALIGNED(*offp, sizeof(void *))) { - binder_user_error("%d:%d got transaction with invalid offset, %zd\n", - proc->pid, thread->pid, *offp); + !IS_ALIGNED(*offp, sizeof(u32))) { + binder_user_error("%d:%d got transaction with invalid offset, %lld (min %lld, max %lld)\n", + proc->pid, thread->pid, (u64)*offp, + (u64)off_min, + (u64)(t->buffer->data_size - + sizeof(*fp))); return_error = BR_FAILED_REPLY; goto err_bad_offset; } fp = (struct flat_binder_object *)(t->buffer->data + *offp); + off_min = *offp + sizeof(struct flat_binder_object); switch (fp->type) { case BINDER_TYPE_BINDER: case BINDER_TYPE_WEAK_BINDER: { struct binder_ref *ref; struct binder_node *node = binder_get_node(proc, fp->binder); + if (node == NULL) { node = binder_new_node(proc, fp->binder, fp->cookie); if (node == NULL) { @@ -1518,10 +1559,15 @@ static void binder_transaction(struct binder_proc *proc, node->accept_fds = !!(fp->flags & FLAT_BINDER_FLAG_ACCEPTS_FDS); } if (fp->cookie != node->cookie) { - binder_user_error("%d:%d sending u%p node %d, cookie mismatch %p != %p\n", + binder_user_error("%d:%d sending u%016llx node %d, cookie mismatch %016llx != %016llx\n", proc->pid, thread->pid, - fp->binder, node->debug_id, - fp->cookie, node->cookie); + (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)) { + return_error = BR_FAILED_REPLY; goto err_binder_get_ref_for_node_failed; } ref = binder_get_ref_for_node(target_proc, node); @@ -1539,20 +1585,25 @@ static void binder_transaction(struct binder_proc *proc, trace_binder_transaction_node_to_ref(t, node, ref); binder_debug(BINDER_DEBUG_TRANSACTION, - " node %d u%p -> ref %d desc %d\n", - node->debug_id, node->ptr, ref->debug_id, - ref->desc); + " 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 = binder_get_ref(proc, fp->handle); + if (ref == NULL) { - binder_user_error("%d:%d got transaction with invalid handle, %ld\n", + binder_user_error("%d:%d got transaction with invalid handle, %d\n", proc->pid, thread->pid, fp->handle); return_error = BR_FAILED_REPLY; goto err_binder_get_ref_failed; } + if (security_binder_transfer_binder(proc->tsk, target_proc->tsk)) { + return_error = BR_FAILED_REPLY; + goto err_binder_get_ref_failed; + } if (ref->node->proc == target_proc) { if (fp->type == BINDER_TYPE_HANDLE) fp->type = BINDER_TYPE_BINDER; @@ -1563,11 +1614,12 @@ static void binder_transaction(struct binder_proc *proc, 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%p\n", + " ref %d desc %d -> node %d u%016llx\n", ref->debug_id, ref->desc, ref->node->debug_id, - ref->node->ptr); + (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; @@ -1590,13 +1642,13 @@ static void binder_transaction(struct binder_proc *proc, if (reply) { if (!(in_reply_to->flags & TF_ACCEPT_FDS)) { - binder_user_error("%d:%d got reply with fd, %ld, but target does not allow fds\n", + 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, %ld, but target does not allow fds\n", + binder_user_error("%d:%d got transaction 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; @@ -1604,11 +1656,16 @@ static void binder_transaction(struct binder_proc *proc, file = fget(fp->handle); if (file == NULL) { - binder_user_error("%d:%d got transaction with invalid fd, %ld\n", + binder_user_error("%d:%d got transaction with invalid fd, %d\n", proc->pid, thread->pid, fp->handle); return_error = BR_FAILED_REPLY; goto err_fget_failed; } + if (security_binder_transfer_file(proc->tsk, target_proc->tsk, file) < 0) { + fput(file); + return_error = BR_FAILED_REPLY; + goto err_get_unused_fd_failed; + } target_fd = task_get_unused_fd_flags(target_proc, O_CLOEXEC); if (target_fd < 0) { fput(file); @@ -1618,13 +1675,13 @@ static void binder_transaction(struct binder_proc *proc, task_fd_install(target_proc, target_fd, file); trace_binder_transaction_fd(t, fp->handle, target_fd); binder_debug(BINDER_DEBUG_TRANSACTION, - " fd %ld -> %d\n", fp->handle, target_fd); + " fd %d -> %d\n", fp->handle, target_fd); /* TODO: fput? */ fp->handle = target_fd; } break; default: - binder_user_error("%d:%d got transaction with invalid object type, %lx\n", + binder_user_error("%d:%d got transaction with invalid object type, %x\n", proc->pid, thread->pid, fp->type); return_error = BR_FAILED_REPLY; goto err_bad_object_type; @@ -1681,12 +1738,13 @@ err_dead_binder: err_invalid_target_handle: err_no_context_mgr_node: binder_debug(BINDER_DEBUG_FAILED_TRANSACTION, - "%d:%d transaction failed %d, size %zd-%zd\n", + "%d:%d transaction failed %d, size %lld-%lld\n", proc->pid, thread->pid, return_error, - tr->data_size, tr->offsets_size); + (u64)tr->data_size, (u64)tr->offsets_size); { struct binder_transaction_log_entry *fe; + fe = binder_transaction_log_add(&binder_transaction_log_failed); *fe = *e; } @@ -1699,10 +1757,13 @@ err_no_context_mgr_node: thread->return_error = return_error; } -int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, - void __user *buffer, int size, signed long *consumed) +static int binder_thread_write(struct binder_proc *proc, + struct binder_thread *thread, + binder_uintptr_t binder_buffer, size_t size, + binder_size_t *consumed) { uint32_t cmd; + void __user *buffer = (void __user *)(uintptr_t)binder_buffer; void __user *ptr = buffer + *consumed; void __user *end = buffer + size; @@ -1771,33 +1832,33 @@ int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, } case BC_INCREFS_DONE: case BC_ACQUIRE_DONE: { - void __user *node_ptr; - void *cookie; + binder_uintptr_t node_ptr; + binder_uintptr_t cookie; struct binder_node *node; - if (get_user(node_ptr, (void * __user *)ptr)) + if (get_user(node_ptr, (binder_uintptr_t __user *)ptr)) return -EFAULT; - ptr += sizeof(void *); - if (get_user(cookie, (void * __user *)ptr)) + ptr += sizeof(binder_uintptr_t); + if (get_user(cookie, (binder_uintptr_t __user *)ptr)) return -EFAULT; - ptr += sizeof(void *); + ptr += sizeof(binder_uintptr_t); node = binder_get_node(proc, node_ptr); if (node == NULL) { - binder_user_error("%d:%d %s u%p no match\n", + binder_user_error("%d:%d %s u%016llx no match\n", proc->pid, thread->pid, cmd == BC_INCREFS_DONE ? "BC_INCREFS_DONE" : "BC_ACQUIRE_DONE", - node_ptr); + (u64)node_ptr); break; } if (cookie != node->cookie) { - binder_user_error("%d:%d %s u%p node %d cookie mismatch %p != %p\n", + binder_user_error("%d:%d %s u%016llx node %d cookie mismatch %016llx != %016llx\n", proc->pid, thread->pid, cmd == BC_INCREFS_DONE ? "BC_INCREFS_DONE" : "BC_ACQUIRE_DONE", - node_ptr, node->debug_id, - cookie, node->cookie); + (u64)node_ptr, node->debug_id, + (u64)cookie, (u64)node->cookie); break; } if (cmd == BC_ACQUIRE_DONE) { @@ -1833,27 +1894,28 @@ int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, return -EINVAL; case BC_FREE_BUFFER: { - void __user *data_ptr; + binder_uintptr_t data_ptr; struct binder_buffer *buffer; - if (get_user(data_ptr, (void * __user *)ptr)) + if (get_user(data_ptr, (binder_uintptr_t __user *)ptr)) return -EFAULT; - ptr += sizeof(void *); + ptr += sizeof(binder_uintptr_t); buffer = binder_buffer_lookup(proc, data_ptr); if (buffer == NULL) { - binder_user_error("%d:%d BC_FREE_BUFFER u%p no match\n", - proc->pid, thread->pid, data_ptr); + binder_user_error("%d:%d BC_FREE_BUFFER u%016llx no match\n", + proc->pid, thread->pid, (u64)data_ptr); break; } if (!buffer->allow_user_free) { - binder_user_error("%d:%d BC_FREE_BUFFER u%p matched unreturned buffer\n", - proc->pid, thread->pid, data_ptr); + binder_user_error("%d:%d BC_FREE_BUFFER u%016llx matched unreturned buffer\n", + proc->pid, thread->pid, (u64)data_ptr); break; } binder_debug(BINDER_DEBUG_FREE_BUFFER, - "%d:%d BC_FREE_BUFFER u%p found buffer %d for %s transaction\n", - proc->pid, thread->pid, data_ptr, buffer->debug_id, + "%d:%d BC_FREE_BUFFER u%016llx found buffer %d for %s transaction\n", + proc->pid, thread->pid, (u64)data_ptr, + buffer->debug_id, buffer->transaction ? "active" : "finished"); if (buffer->transaction) { @@ -1923,16 +1985,16 @@ int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, case BC_REQUEST_DEATH_NOTIFICATION: case BC_CLEAR_DEATH_NOTIFICATION: { uint32_t target; - void __user *cookie; + binder_uintptr_t cookie; struct binder_ref *ref; struct binder_ref_death *death; if (get_user(target, (uint32_t __user *)ptr)) return -EFAULT; ptr += sizeof(uint32_t); - if (get_user(cookie, (void __user * __user *)ptr)) + if (get_user(cookie, (binder_uintptr_t __user *)ptr)) return -EFAULT; - ptr += sizeof(void *); + ptr += sizeof(binder_uintptr_t); ref = binder_get_ref(proc, target); if (ref == NULL) { binder_user_error("%d:%d %s invalid ref %d\n", @@ -1945,12 +2007,12 @@ int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, } binder_debug(BINDER_DEBUG_DEATH_NOTIFICATION, - "%d:%d %s %p ref %d desc %d s %d w %d for node %d\n", + "%d:%d %s %016llx ref %d desc %d s %d w %d for node %d\n", proc->pid, thread->pid, cmd == BC_REQUEST_DEATH_NOTIFICATION ? "BC_REQUEST_DEATH_NOTIFICATION" : "BC_CLEAR_DEATH_NOTIFICATION", - cookie, ref->debug_id, ref->desc, + (u64)cookie, ref->debug_id, ref->desc, ref->strong, ref->weak, ref->node->debug_id); if (cmd == BC_REQUEST_DEATH_NOTIFICATION) { @@ -1988,9 +2050,10 @@ int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, } death = ref->death; if (death->cookie != cookie) { - binder_user_error("%d:%d BC_CLEAR_DEATH_NOTIFICATION death notification cookie mismatch %p != %p\n", + binder_user_error("%d:%d BC_CLEAR_DEATH_NOTIFICATION death notification cookie mismatch %016llx != %016llx\n", proc->pid, thread->pid, - death->cookie, cookie); + (u64)death->cookie, + (u64)cookie); break; } ref->death = NULL; @@ -2010,25 +2073,28 @@ int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, } break; case BC_DEAD_BINDER_DONE: { struct binder_work *w; - void __user *cookie; + binder_uintptr_t cookie; struct binder_ref_death *death = NULL; - if (get_user(cookie, (void __user * __user *)ptr)) + + if (get_user(cookie, (binder_uintptr_t __user *)ptr)) return -EFAULT; ptr += sizeof(void *); 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; break; } } binder_debug(BINDER_DEBUG_DEAD_BINDER, - "%d:%d BC_DEAD_BINDER_DONE %p found %p\n", - proc->pid, thread->pid, cookie, death); + "%d:%d BC_DEAD_BINDER_DONE %016llx found %p\n", + proc->pid, thread->pid, (u64)cookie, + death); if (death == NULL) { - binder_user_error("%d:%d BC_DEAD_BINDER_DONE %p not found\n", - proc->pid, thread->pid, cookie); + binder_user_error("%d:%d BC_DEAD_BINDER_DONE %016llx not found\n", + proc->pid, thread->pid, (u64)cookie); break; } @@ -2054,8 +2120,8 @@ int binder_thread_write(struct binder_proc *proc, struct binder_thread *thread, return 0; } -void binder_stat_br(struct binder_proc *proc, struct binder_thread *thread, - uint32_t cmd) +static void binder_stat_br(struct binder_proc *proc, + struct binder_thread *thread, uint32_t cmd) { trace_binder_return(cmd); if (_IOC_NR(cmd) < ARRAY_SIZE(binder_stats.br)) { @@ -2080,9 +2146,10 @@ static int binder_has_thread_work(struct binder_thread *thread) static int binder_thread_read(struct binder_proc *proc, struct binder_thread *thread, - void __user *buffer, int size, - signed long *consumed, int non_block) + binder_uintptr_t binder_buffer, size_t size, + binder_size_t *consumed, int non_block) { + void __user *buffer = (void __user *)(uintptr_t)binder_buffer; void __user *ptr = buffer + *consumed; void __user *end = buffer + size; @@ -2140,13 +2207,13 @@ retry: if (!binder_has_proc_work(proc, thread)) ret = -EAGAIN; } else - ret = wait_event_interruptible_exclusive(proc->wait, binder_has_proc_work(proc, thread)); + 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_interruptible(thread->wait, binder_has_thread_work(thread)); + ret = wait_event_freezable(thread->wait, binder_has_thread_work(thread)); } binder_lock(__func__); @@ -2164,12 +2231,16 @@ retry: struct binder_work *w; struct binder_transaction *t = NULL; - 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 { - if (ptr - buffer == 4 && !(thread->looper & BINDER_LOOPER_STATE_NEED_RETURN)) /* no data added */ + 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)) goto retry; break; } @@ -2202,6 +2273,7 @@ retry: 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"; @@ -2227,32 +2299,40 @@ retry: if (put_user(cmd, (uint32_t __user *)ptr)) return -EFAULT; ptr += sizeof(uint32_t); - if (put_user(node->ptr, (void * __user *)ptr)) + if (put_user(node->ptr, + (binder_uintptr_t __user *)ptr)) return -EFAULT; - ptr += sizeof(void *); - if (put_user(node->cookie, (void * __user *)ptr)) + ptr += sizeof(binder_uintptr_t); + if (put_user(node->cookie, + (binder_uintptr_t __user *)ptr)) return -EFAULT; - ptr += sizeof(void *); + ptr += sizeof(binder_uintptr_t); binder_stat_br(proc, thread, cmd); binder_debug(BINDER_DEBUG_USER_REFS, - "%d:%d %s %d u%p c%p\n", - proc->pid, thread->pid, cmd_name, node->debug_id, node->ptr, node->cookie); + "%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%p c%p deleted\n", - proc->pid, thread->pid, node->debug_id, - node->ptr, node->cookie); + "%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%p c%p state unchanged\n", - proc->pid, thread->pid, node->debug_id, node->ptr, - node->cookie); + "%d:%d node %d u%016llx c%016llx state unchanged\n", + proc->pid, thread->pid, + node->debug_id, + (u64)node->ptr, + (u64)node->cookie); } } } break; @@ -2270,17 +2350,18 @@ retry: if (put_user(cmd, (uint32_t __user *)ptr)) return -EFAULT; ptr += sizeof(uint32_t); - if (put_user(death->cookie, (void * __user *)ptr)) + if (put_user(death->cookie, + (binder_uintptr_t __user *)ptr)) return -EFAULT; - ptr += sizeof(void *); + ptr += sizeof(binder_uintptr_t); binder_stat_br(proc, thread, cmd); binder_debug(BINDER_DEBUG_DEATH_NOTIFICATION, - "%d:%d %s %p\n", + "%d:%d %s %016llx\n", proc->pid, thread->pid, cmd == BR_DEAD_BINDER ? "BR_DEAD_BINDER" : "BR_CLEAR_DEATH_NOTIFICATION_DONE", - death->cookie); + (u64)death->cookie); if (w->type == BINDER_WORK_CLEAR_DEATH_NOTIFICATION) { list_del(&w->entry); @@ -2299,6 +2380,7 @@ retry: BUG_ON(t->buffer == NULL); if (t->buffer->target_node) { struct binder_node *target_node = t->buffer->target_node; + tr.target.ptr = target_node->ptr; tr.cookie = target_node->cookie; t->saved_priority = task_nice(current); @@ -2310,8 +2392,8 @@ retry: binder_set_nice(target_node->min_priority); cmd = BR_TRANSACTION; } else { - tr.target.ptr = NULL; - tr.cookie = NULL; + tr.target.ptr = 0; + tr.cookie = 0; cmd = BR_REPLY; } tr.code = t->code; @@ -2320,6 +2402,7 @@ retry: if (t->from) { struct task_struct *sender = t->from->proc->tsk; + tr.sender_pid = task_tgid_nr_ns(sender, task_active_pid_ns(current)); } else { @@ -2328,8 +2411,9 @@ retry: tr.data_size = t->buffer->data_size; tr.offsets_size = t->buffer->offsets_size; - tr.data.ptr.buffer = (void *)t->buffer->data + - proc->user_buffer_offset; + tr.data.ptr.buffer = (binder_uintptr_t)( + (uintptr_t)t->buffer->data + + proc->user_buffer_offset); tr.data.ptr.offsets = tr.data.ptr.buffer + ALIGN(t->buffer->data_size, sizeof(void *)); @@ -2344,14 +2428,14 @@ retry: trace_binder_transaction_received(t); binder_stat_br(proc, thread, cmd); binder_debug(BINDER_DEBUG_TRANSACTION, - "%d:%d %s %d %d:%d, cmd %d size %zd-%zd ptr %p-%p\n", + "%d:%d %s %d %d:%d, cmd %d size %zd-%zd ptr %016llx-%016llx\n", 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->buffer->data_size, t->buffer->offsets_size, - tr.data.ptr.buffer, tr.data.ptr.offsets); + (u64)tr.data.ptr.buffer, (u64)tr.data.ptr.offsets); list_del(&t->work.entry); t->buffer->allow_user_free = 1; @@ -2389,6 +2473,7 @@ done: static void binder_release_work(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); @@ -2421,8 +2506,8 @@ static void binder_release_work(struct list_head *list) death = container_of(w, struct binder_ref_death, work); binder_debug(BINDER_DEBUG_DEAD_TRANSACTION, - "undelivered death notification, %p\n", - death->cookie); + "undelivered death notification, %016llx\n", + (u64)death->cookie); kfree(death); binder_stats_deleted(BINDER_STAT_DEATH); } break; @@ -2543,6 +2628,109 @@ static unsigned int binder_poll(struct file *filp, return 0; } +static int binder_ioctl_write_read(struct file *filp, + unsigned int cmd, unsigned long arg, + struct binder_thread *thread) +{ + int ret = 0; + struct binder_proc *proc = filp->private_data; + unsigned int size = _IOC_SIZE(cmd); + void __user *ubuf = (void __user *)arg; + struct binder_write_read bwr; + + if (size != sizeof(struct binder_write_read)) { + ret = -EINVAL; + goto out; + } + if (copy_from_user(&bwr, ubuf, sizeof(bwr))) { + ret = -EFAULT; + goto out; + } + binder_debug(BINDER_DEBUG_READ_WRITE, + "%d:%d write %lld at %016llx, read %lld at %016llx\n", + proc->pid, thread->pid, + (u64)bwr.write_size, (u64)bwr.write_buffer, + (u64)bwr.read_size, (u64)bwr.read_buffer); + + if (bwr.write_size > 0) { + ret = binder_thread_write(proc, thread, + bwr.write_buffer, + bwr.write_size, + &bwr.write_consumed); + trace_binder_write_done(ret); + if (ret < 0) { + bwr.read_consumed = 0; + if (copy_to_user(ubuf, &bwr, sizeof(bwr))) + ret = -EFAULT; + goto out; + } + } + if (bwr.read_size > 0) { + ret = binder_thread_read(proc, thread, bwr.read_buffer, + bwr.read_size, + &bwr.read_consumed, + filp->f_flags & O_NONBLOCK); + trace_binder_read_done(ret); + if (!list_empty(&proc->todo)) + wake_up_interruptible(&proc->wait); + if (ret < 0) { + if (copy_to_user(ubuf, &bwr, sizeof(bwr))) + ret = -EFAULT; + goto out; + } + } + binder_debug(BINDER_DEBUG_READ_WRITE, + "%d:%d wrote %lld of %lld, read return %lld of %lld\n", + proc->pid, thread->pid, + (u64)bwr.write_consumed, (u64)bwr.write_size, + (u64)bwr.read_consumed, (u64)bwr.read_size); + if (copy_to_user(ubuf, &bwr, sizeof(bwr))) { + ret = -EFAULT; + goto out; + } +out: + return ret; +} + +static int binder_ioctl_set_ctx_mgr(struct file *filp) +{ + int ret = 0; + struct binder_proc *proc = filp->private_data; + kuid_t curr_euid = current_euid(); + + if (binder_context_mgr_node != NULL) { + pr_err("BINDER_SET_CONTEXT_MGR already set\n"); + ret = -EBUSY; + goto out; + } + 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)) { + 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)); + ret = -EPERM; + goto out; + } + } else { + binder_context_mgr_uid = curr_euid; + } + binder_context_mgr_node = binder_new_node(proc, 0, 0); + if (binder_context_mgr_node == NULL) { + 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; +out: + return ret; +} + static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { int ret; @@ -2551,7 +2739,8 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) unsigned int size = _IOC_SIZE(cmd); void __user *ubuf = (void __user *)arg; - /*pr_info("binder_ioctl: %d:%d %x %lx\n", proc->pid, current->pid, cmd, arg);*/ + /*pr_info("binder_ioctl: %d:%d %x %lx\n", + proc->pid, current->pid, cmd, arg);*/ trace_binder_ioctl(cmd, arg); @@ -2567,52 +2756,11 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) } switch (cmd) { - case BINDER_WRITE_READ: { - struct binder_write_read bwr; - if (size != sizeof(struct binder_write_read)) { - ret = -EINVAL; + case BINDER_WRITE_READ: + ret = binder_ioctl_write_read(filp, cmd, arg, thread); + if (ret) goto err; - } - if (copy_from_user(&bwr, ubuf, sizeof(bwr))) { - ret = -EFAULT; - goto err; - } - binder_debug(BINDER_DEBUG_READ_WRITE, - "%d:%d write %ld at %08lx, read %ld at %08lx\n", - proc->pid, thread->pid, bwr.write_size, - bwr.write_buffer, bwr.read_size, bwr.read_buffer); - - if (bwr.write_size > 0) { - ret = binder_thread_write(proc, thread, (void __user *)bwr.write_buffer, bwr.write_size, &bwr.write_consumed); - trace_binder_write_done(ret); - if (ret < 0) { - bwr.read_consumed = 0; - if (copy_to_user(ubuf, &bwr, sizeof(bwr))) - ret = -EFAULT; - goto err; - } - } - if (bwr.read_size > 0) { - ret = binder_thread_read(proc, thread, (void __user *)bwr.read_buffer, bwr.read_size, &bwr.read_consumed, filp->f_flags & O_NONBLOCK); - trace_binder_read_done(ret); - if (!list_empty(&proc->todo)) - wake_up_interruptible(&proc->wait); - if (ret < 0) { - if (copy_to_user(ubuf, &bwr, sizeof(bwr))) - ret = -EFAULT; - goto err; - } - } - binder_debug(BINDER_DEBUG_READ_WRITE, - "%d:%d wrote %ld of %ld, read return %ld of %ld\n", - proc->pid, thread->pid, bwr.write_consumed, bwr.write_size, - bwr.read_consumed, bwr.read_size); - if (copy_to_user(ubuf, &bwr, sizeof(bwr))) { - ret = -EFAULT; - goto err; - } break; - } case BINDER_SET_MAX_THREADS: if (copy_from_user(&proc->max_threads, ubuf, sizeof(proc->max_threads))) { ret = -EINVAL; @@ -2620,30 +2768,9 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) } break; case BINDER_SET_CONTEXT_MGR: - if (binder_context_mgr_node != NULL) { - pr_err("BINDER_SET_CONTEXT_MGR already set\n"); - ret = -EBUSY; - goto err; - } - if (uid_valid(binder_context_mgr_uid)) { - if (!uid_eq(binder_context_mgr_uid, current->cred->euid)) { - pr_err("BINDER_SET_CONTEXT_MGR bad uid %d != %d\n", - from_kuid(&init_user_ns, current->cred->euid), - from_kuid(&init_user_ns, binder_context_mgr_uid)); - ret = -EPERM; - goto err; - } - } else - binder_context_mgr_uid = current->cred->euid; - binder_context_mgr_node = binder_new_node(proc, NULL, NULL); - if (binder_context_mgr_node == NULL) { - ret = -ENOMEM; + ret = binder_ioctl_set_ctx_mgr(filp); + if (ret) goto err; - } - 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; break; case BINDER_THREAD_EXIT: binder_debug(BINDER_DEBUG_THREADS, "%d:%d exit\n", @@ -2651,16 +2778,20 @@ static long binder_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) binder_free_thread(proc, thread); thread = NULL; break; - case BINDER_VERSION: + case BINDER_VERSION: { + struct binder_version __user *ver = ubuf; + if (size != sizeof(struct binder_version)) { ret = -EINVAL; goto err; } - if (put_user(BINDER_CURRENT_PROTOCOL_VERSION, &((struct binder_version *)ubuf)->protocol_version)) { + if (put_user(BINDER_CURRENT_PROTOCOL_VERSION, + &ver->protocol_version)) { ret = -EINVAL; goto err; } break; + } default: ret = -EINVAL; goto err; @@ -2681,6 +2812,7 @@ err_unlocked: static void binder_vma_open(struct vm_area_struct *vma) { struct binder_proc *proc = vma->vm_private_data; + binder_debug(BINDER_DEBUG_OPEN_CLOSE, "%d open vm area %lx-%lx (%ld K) vma %lx pagep %lx\n", proc->pid, vma->vm_start, vma->vm_end, @@ -2691,6 +2823,7 @@ static void binder_vma_open(struct vm_area_struct *vma) static void binder_vma_close(struct vm_area_struct *vma) { struct binder_proc *proc = vma->vm_private_data; + binder_debug(BINDER_DEBUG_OPEN_CLOSE, "%d close vm area %lx-%lx (%ld K) vma %lx pagep %lx\n", proc->pid, vma->vm_start, vma->vm_end, @@ -2701,9 +2834,15 @@ static void binder_vma_close(struct vm_area_struct *vma) binder_defer_work(proc, BINDER_DEFERRED_PUT_FILES); } +static int binder_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + return VM_FAULT_SIGBUS; +} + static struct vm_operations_struct binder_vm_ops = { .open = binder_vma_open, .close = binder_vma_close, + .fault = binder_vm_fault, }; static int binder_mmap(struct file *filp, struct vm_area_struct *vma) @@ -2833,6 +2972,7 @@ static int binder_open(struct inode *nodp, struct file *filp) if (binder_debugfs_dir_entry_proc) { char strbuf[11]; + snprintf(strbuf, sizeof(strbuf), "%u", proc->pid); proc->debugfs_entry = debugfs_create_file(strbuf, S_IRUGO, binder_debugfs_dir_entry_proc, proc, &binder_proc_fops); @@ -2854,8 +2994,10 @@ static void binder_deferred_flush(struct binder_proc *proc) { struct rb_node *n; int wake_count = 0; + 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; if (thread->looper & BINDER_LOOPER_STATE_WAITING) { wake_up_interruptible(&thread->wait); @@ -2872,6 +3014,7 @@ static void binder_deferred_flush(struct binder_proc *proc) static int binder_release(struct inode *nodp, struct file *filp) { struct binder_proc *proc = filp->private_data; + debugfs_remove(proc->debugfs_entry); binder_defer_work(proc, BINDER_DEFERRED_RELEASE); @@ -3033,6 +3176,7 @@ static void binder_deferred_func(struct work_struct *work) struct files_struct *files; int defer; + do { binder_lock(__func__); mutex_lock(&binder_deferred_lock); @@ -3130,8 +3274,9 @@ static void print_binder_work(struct seq_file *m, const char *prefix, break; case BINDER_WORK_NODE: node = container_of(w, struct binder_node, work); - seq_printf(m, "%snode work %d: u%p c%p\n", - prefix, node->debug_id, node->ptr, node->cookie); + seq_printf(m, "%snode work %d: u%016llx c%016llx\n", + prefix, node->debug_id, + (u64)node->ptr, (u64)node->cookie); break; case BINDER_WORK_DEAD_BINDER: seq_printf(m, "%shas dead binder\n", prefix); @@ -3191,8 +3336,8 @@ 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%p c%p hs %d hw %d ls %d lw %d is %d iw %d", - node->debug_id, node->ptr, node->cookie, + seq_printf(m, " node %d: u%016llx c%016llx hs %d hw %d ls %d lw %d is %d iw %d", + node->debug_id, (u64)node->ptr, (u64)node->cookie, node->has_strong_ref, node->has_weak_ref, node->local_strong_refs, node->local_weak_refs, node->internal_strong_refs, count); @@ -3494,6 +3639,7 @@ static const struct file_operations binder_fops = { .owner = THIS_MODULE, .poll = binder_poll, .unlocked_ioctl = binder_ioctl, + .compat_ioctl = binder_ioctl, .mmap = binder_mmap, .open = binder_open, .flush = binder_flush, diff --git a/drivers/staging/android/binder_trace.h b/drivers/android/binder_trace.h index 82a567c2af67..7f20f3dc8369 100644 --- a/drivers/staging/android/binder_trace.h +++ b/drivers/android/binder_trace.h @@ -152,7 +152,7 @@ TRACE_EVENT(binder_transaction_node_to_ref, TP_STRUCT__entry( __field(int, debug_id) __field(int, node_debug_id) - __field(void __user *, node_ptr) + __field(binder_uintptr_t, node_ptr) __field(int, ref_debug_id) __field(uint32_t, ref_desc) ), @@ -163,8 +163,9 @@ TRACE_EVENT(binder_transaction_node_to_ref, __entry->ref_debug_id = ref->debug_id; __entry->ref_desc = ref->desc; ), - TP_printk("transaction=%d node=%d src_ptr=0x%p ==> dest_ref=%d dest_desc=%d", - __entry->debug_id, __entry->node_debug_id, __entry->node_ptr, + TP_printk("transaction=%d node=%d src_ptr=0x%016llx ==> dest_ref=%d dest_desc=%d", + __entry->debug_id, __entry->node_debug_id, + (u64)__entry->node_ptr, __entry->ref_debug_id, __entry->ref_desc) ); @@ -177,7 +178,7 @@ TRACE_EVENT(binder_transaction_ref_to_node, __field(int, ref_debug_id) __field(uint32_t, ref_desc) __field(int, node_debug_id) - __field(void __user *, node_ptr) + __field(binder_uintptr_t, node_ptr) ), TP_fast_assign( __entry->debug_id = t->debug_id; @@ -186,9 +187,10 @@ TRACE_EVENT(binder_transaction_ref_to_node, __entry->node_debug_id = ref->node->debug_id; __entry->node_ptr = ref->node->ptr; ), - TP_printk("transaction=%d node=%d src_ref=%d src_desc=%d ==> dest_ptr=0x%p", + TP_printk("transaction=%d node=%d src_ref=%d src_desc=%d ==> dest_ptr=0x%016llx", __entry->debug_id, __entry->node_debug_id, - __entry->ref_debug_id, __entry->ref_desc, __entry->node_ptr) + __entry->ref_debug_id, __entry->ref_desc, + (u64)__entry->node_ptr) ); TRACE_EVENT(binder_transaction_ref_to_ref, diff --git a/drivers/base/firmware_class.c b/drivers/base/firmware_class.c index 00a565676583..9cf8dc7448e2 100644 --- a/drivers/base/firmware_class.c +++ b/drivers/base/firmware_class.c @@ -27,6 +27,7 @@ #include <linux/pm.h> #include <linux/suspend.h> #include <linux/syscore_ops.h> +#include <linux/reboot.h> #include <generated/utsrelease.h> @@ -130,6 +131,7 @@ struct firmware_buf { struct page **pages; int nr_pages; int page_array_size; + struct list_head pending_list; #endif char fw_id[]; }; @@ -171,6 +173,9 @@ static struct firmware_buf *__allocate_fw_buf(const char *fw_name, strcpy(buf->fw_id, fw_name); buf->fwc = fwc; init_completion(&buf->completion); +#ifdef CONFIG_FW_LOADER_USER_HELPER + INIT_LIST_HEAD(&buf->pending_list); +#endif pr_debug("%s: fw-%s buf=%p\n", __func__, fw_name, buf); @@ -446,10 +451,8 @@ static struct firmware_priv *to_firmware_priv(struct device *dev) return container_of(dev, struct firmware_priv, dev); } -static void fw_load_abort(struct firmware_priv *fw_priv) +static void __fw_load_abort(struct firmware_buf *buf) { - struct firmware_buf *buf = fw_priv->buf; - /* * There is a small window in which user can write to 'loading' * between loading done and disappearance of 'loading' @@ -457,8 +460,16 @@ static void fw_load_abort(struct firmware_priv *fw_priv) if (test_bit(FW_STATUS_DONE, &buf->status)) return; + list_del_init(&buf->pending_list); set_bit(FW_STATUS_ABORT, &buf->status); complete_all(&buf->completion); +} + +static void fw_load_abort(struct firmware_priv *fw_priv) +{ + struct firmware_buf *buf = fw_priv->buf; + + __fw_load_abort(buf); /* avoid user action after loading abort */ fw_priv->buf = NULL; @@ -467,6 +478,25 @@ static void fw_load_abort(struct firmware_priv *fw_priv) #define is_fw_load_aborted(buf) \ test_bit(FW_STATUS_ABORT, &(buf)->status) +static LIST_HEAD(pending_fw_head); + +/* reboot notifier for avoid deadlock with usermode_lock */ +static int fw_shutdown_notify(struct notifier_block *unused1, + unsigned long unused2, void *unused3) +{ + mutex_lock(&fw_lock); + while (!list_empty(&pending_fw_head)) + __fw_load_abort(list_first_entry(&pending_fw_head, + struct firmware_buf, + pending_list)); + mutex_unlock(&fw_lock); + return NOTIFY_DONE; +} + +static struct notifier_block fw_shutdown_nb = { + .notifier_call = fw_shutdown_notify, +}; + static ssize_t firmware_timeout_show(struct class *class, struct class_attribute *attr, char *buf) @@ -619,6 +649,7 @@ static ssize_t firmware_loading_store(struct device *dev, * is completed. * */ fw_map_pages_buf(fw_buf); + list_del_init(&fw_buf->pending_list); complete_all(&fw_buf->completion); break; } @@ -853,8 +884,15 @@ static int _request_firmware_load(struct firmware_priv *fw_priv, bool uevent, goto err_del_dev; } + mutex_lock(&fw_lock); + list_add(&buf->pending_list, &pending_fw_head); + mutex_unlock(&fw_lock); + retval = device_create_file(f_dev, &dev_attr_loading); if (retval) { + mutex_lock(&fw_lock); + list_del_init(&buf->pending_list); + mutex_unlock(&fw_lock); dev_err(f_dev, "%s: device_create_file failed\n", __func__); goto err_del_bin_attr; } @@ -1529,6 +1567,7 @@ static int __init firmware_class_init(void) { fw_cache_init(); #ifdef CONFIG_FW_LOADER_USER_HELPER + register_reboot_notifier(&fw_shutdown_nb); return class_register(&firmware_class); #else return 0; @@ -1542,6 +1581,7 @@ static void __exit firmware_class_exit(void) unregister_pm_notifier(&fw_cache.pm_notify); #endif #ifdef CONFIG_FW_LOADER_USER_HELPER + unregister_reboot_notifier(&fw_shutdown_nb); class_unregister(&firmware_class); #endif } diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index 5a9b6569dd74..5131ad8ca17f 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -29,6 +29,9 @@ #include <linux/async.h> #include <linux/suspend.h> #include <linux/cpuidle.h> +#include <linux/timer.h> +#include <linux/wakeup_reason.h> + #include "../base.h" #include "power.h" @@ -54,6 +57,12 @@ struct suspend_stats suspend_stats; static DEFINE_MUTEX(dpm_list_mtx); static pm_message_t pm_transition; +struct dpm_watchdog { + struct device *dev; + struct task_struct *tsk; + struct timer_list timer; +}; + static int async_error; /** @@ -384,6 +393,56 @@ static int dpm_run_callback(pm_callback_t cb, struct device *dev, return error; } +/** + * dpm_wd_handler - Driver suspend / resume watchdog handler. + * + * Called when a driver has timed out suspending or resuming. + * There's not much we can do here to recover so BUG() out for + * a crash-dump + */ +static void dpm_wd_handler(unsigned long data) +{ + struct dpm_watchdog *wd = (void *)data; + struct device *dev = wd->dev; + struct task_struct *tsk = wd->tsk; + + dev_emerg(dev, "**** DPM device timeout ****\n"); + show_stack(tsk, NULL); + + BUG(); +} + +/** + * dpm_wd_set - Enable pm watchdog for given device. + * @wd: Watchdog. Must be allocated on the stack. + * @dev: Device to handle. + */ +static void dpm_wd_set(struct dpm_watchdog *wd, struct device *dev) +{ + struct timer_list *timer = &wd->timer; + + wd->dev = dev; + wd->tsk = get_current(); + + init_timer_on_stack(timer); + timer->expires = jiffies + HZ * 12; + timer->function = dpm_wd_handler; + timer->data = (unsigned long)wd; + add_timer(timer); +} + +/** + * dpm_wd_clear - Disable pm watchdog. + * @wd: Watchdog to disable. + */ +static void dpm_wd_clear(struct dpm_watchdog *wd) +{ + struct timer_list *timer = &wd->timer; + + del_timer_sync(timer); + destroy_timer_on_stack(timer); +} + /*------------------------- Resume routines -------------------------*/ /** @@ -570,6 +629,7 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) pm_callback_t callback = NULL; char *info = NULL; int error = 0; + struct dpm_watchdog wd; TRACE_DEVICE(dev); TRACE_RESUME(0); @@ -585,6 +645,7 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) * a resumed device, even if the device hasn't been completed yet. */ dev->power.is_prepared = false; + dpm_wd_set(&wd, dev); if (!dev->power.is_suspended) goto Unlock; @@ -636,6 +697,7 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) Unlock: device_unlock(dev); + dpm_wd_clear(&wd); Complete: complete_all(&dev->power.completion); @@ -877,6 +939,7 @@ static int device_suspend_noirq(struct device *dev, pm_message_t state) static int dpm_suspend_noirq(pm_message_t state) { ktime_t starttime = ktime_get(); + char suspend_abort[MAX_SUSPEND_ABORT_LEN]; int error = 0; cpuidle_pause(); @@ -904,6 +967,9 @@ static int dpm_suspend_noirq(pm_message_t state) put_device(dev); if (pm_wakeup_pending()) { + pm_get_active_wakeup_sources(suspend_abort, + MAX_SUSPEND_ABORT_LEN); + log_suspend_abort_reason(suspend_abort); error = -EBUSY; break; } @@ -962,6 +1028,7 @@ static int device_suspend_late(struct device *dev, pm_message_t state) static int dpm_suspend_late(pm_message_t state) { ktime_t starttime = ktime_get(); + char suspend_abort[MAX_SUSPEND_ABORT_LEN]; int error = 0; mutex_lock(&dpm_list_mtx); @@ -987,6 +1054,9 @@ static int dpm_suspend_late(pm_message_t state) put_device(dev); if (pm_wakeup_pending()) { + pm_get_active_wakeup_sources(suspend_abort, + MAX_SUSPEND_ABORT_LEN); + log_suspend_abort_reason(suspend_abort); error = -EBUSY; break; } @@ -1053,6 +1123,8 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) pm_callback_t callback = NULL; char *info = NULL; int error = 0; + struct dpm_watchdog wd; + char suspend_abort[MAX_SUSPEND_ABORT_LEN]; dpm_wait_for_children(dev, async); @@ -1069,12 +1141,17 @@ 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; } if (dev->power.syscore) goto Complete; + + dpm_wd_set(&wd, dev); device_lock(dev); @@ -1131,6 +1208,8 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) device_unlock(dev); + dpm_wd_clear(&wd); + Complete: complete_all(&dev->power.completion); if (error) diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c index 79715e7fa43e..4b35cc08c3b6 100644 --- a/drivers/base/power/wakeup.c +++ b/drivers/base/power/wakeup.c @@ -14,6 +14,7 @@ #include <linux/suspend.h> #include <linux/seq_file.h> #include <linux/debugfs.h> +#include <linux/types.h> #include <trace/events/power.h> #include "power.h" @@ -659,6 +660,37 @@ 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) { + 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); + static void print_active_wakeup_sources(void) { struct wakeup_source *ws; diff --git a/drivers/base/syscore.c b/drivers/base/syscore.c index e8d11b6630ee..0ab546558c4e 100644 --- a/drivers/base/syscore.c +++ b/drivers/base/syscore.c @@ -10,6 +10,7 @@ #include <linux/mutex.h> #include <linux/module.h> #include <linux/interrupt.h> +#include <linux/wakeup_reason.h> static LIST_HEAD(syscore_ops_list); static DEFINE_MUTEX(syscore_ops_lock); @@ -73,6 +74,8 @@ 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/char/Kconfig b/drivers/char/Kconfig index 3bb6fa3930be..6fcb9b01e57c 100644 --- a/drivers/char/Kconfig +++ b/drivers/char/Kconfig @@ -6,6 +6,19 @@ menu "Character devices" source "drivers/tty/Kconfig" +config DEVMEM + bool "Memory device driver" + default y + help + The memory driver provides two character devices, mem and kmem, which + provide access to the system's memory. The mem device is a view of + physical memory, and each byte in the device corresponds to the + matching physical address. The kmem device is the same as mem, but + the addresses correspond to the kernel's virtual address space rather + than physical memory. These devices are standard parts of a Linux + system and most users should say Y here. You might say N if very + security conscience or memory is tight. + config DEVKMEM bool "/dev/kmem virtual device support" default y @@ -584,6 +597,10 @@ config DEVPORT depends on ISA || PCI default y +config DCC_TTY + tristate "DCC tty driver" + depends on ARM + source "drivers/s390/char/Kconfig" config MSM_SMD_PKT diff --git a/drivers/char/Makefile b/drivers/char/Makefile index 7ff1d0d208a7..e0047ed1e74c 100644 --- a/drivers/char/Makefile +++ b/drivers/char/Makefile @@ -56,6 +56,7 @@ obj-$(CONFIG_PCMCIA) += pcmcia/ obj-$(CONFIG_HANGCHECK_TIMER) += hangcheck-timer.o obj-$(CONFIG_TCG_TPM) += tpm/ +obj-$(CONFIG_DCC_TTY) += dcc_tty.o obj-$(CONFIG_PS3_FLASH) += ps3flash.o obj-$(CONFIG_JS_RTC) += js-rtc.o diff --git a/drivers/char/dcc_tty.c b/drivers/char/dcc_tty.c new file mode 100644 index 000000000000..0a62d410286f --- /dev/null +++ b/drivers/char/dcc_tty.c @@ -0,0 +1,326 @@ +/* drivers/char/dcc_tty.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/platform_device.h> +#include <linux/delay.h> +#include <linux/console.h> +#include <linux/hrtimer.h> +#include <linux/tty.h> +#include <linux/tty_driver.h> +#include <linux/tty_flip.h> + +MODULE_DESCRIPTION("DCC TTY Driver"); +MODULE_LICENSE("GPL"); +MODULE_VERSION("1.0"); + +DEFINE_SPINLOCK(g_dcc_tty_lock); +static struct hrtimer g_dcc_timer; +static char g_dcc_buffer[16]; +static int g_dcc_buffer_head; +static int g_dcc_buffer_count; +static unsigned g_dcc_write_delay_usecs = 1; +static struct tty_driver *g_dcc_tty_driver; +static struct tty_struct *g_dcc_tty; +static int g_dcc_tty_open_count; + +static void dcc_poll_locked(void) +{ + char ch; + int rch; + int written; + + while (g_dcc_buffer_count) { + ch = g_dcc_buffer[g_dcc_buffer_head]; + asm( + "mrc 14, 0, r15, c0, c1, 0\n" + "mcrcc 14, 0, %1, c0, c5, 0\n" + "movcc %0, #1\n" + "movcs %0, #0\n" + : "=r" (written) + : "r" (ch) + ); + if (written) { + if (ch == '\n') + g_dcc_buffer[g_dcc_buffer_head] = '\r'; + else { + g_dcc_buffer_head = (g_dcc_buffer_head + 1) % ARRAY_SIZE(g_dcc_buffer); + g_dcc_buffer_count--; + if (g_dcc_tty) + tty_wakeup(g_dcc_tty); + } + g_dcc_write_delay_usecs = 1; + } else { + if (g_dcc_write_delay_usecs > 0x100) + break; + g_dcc_write_delay_usecs <<= 1; + udelay(g_dcc_write_delay_usecs); + } + } + + if (g_dcc_tty && !test_bit(TTY_THROTTLED, &g_dcc_tty->flags)) { + asm( + "mrc 14, 0, %0, c0, c1, 0\n" + "tst %0, #(1 << 30)\n" + "moveq %0, #-1\n" + "mrcne 14, 0, %0, c0, c5, 0\n" + : "=r" (rch) + ); + if (rch >= 0) { + ch = rch; + tty_insert_flip_string(g_dcc_tty->port, &ch, 1); + tty_flip_buffer_push(g_dcc_tty->port); + } + } + + + if (g_dcc_buffer_count) + hrtimer_start(&g_dcc_timer, ktime_set(0, g_dcc_write_delay_usecs * NSEC_PER_USEC), HRTIMER_MODE_REL); + else + hrtimer_start(&g_dcc_timer, ktime_set(0, 20 * NSEC_PER_MSEC), HRTIMER_MODE_REL); +} + +static int dcc_tty_open(struct tty_struct * tty, struct file * filp) +{ + int ret; + unsigned long irq_flags; + + spin_lock_irqsave(&g_dcc_tty_lock, irq_flags); + if (g_dcc_tty == NULL || g_dcc_tty == tty) { + g_dcc_tty = tty; + g_dcc_tty_open_count++; + ret = 0; + } else + ret = -EBUSY; + spin_unlock_irqrestore(&g_dcc_tty_lock, irq_flags); + + printk("dcc_tty_open, tty %p, f_flags %x, returned %d\n", tty, filp->f_flags, ret); + + return ret; +} + +static void dcc_tty_close(struct tty_struct * tty, struct file * filp) +{ + printk("dcc_tty_close, tty %p, f_flags %x\n", tty, filp->f_flags); + if (g_dcc_tty == tty) { + if (--g_dcc_tty_open_count == 0) + g_dcc_tty = NULL; + } +} + +static int dcc_write(const unsigned char *buf_start, int count) +{ + const unsigned char *buf = buf_start; + unsigned long irq_flags; + int copy_len; + int space_left; + int tail; + + if (count < 1) + return 0; + + spin_lock_irqsave(&g_dcc_tty_lock, irq_flags); + do { + tail = (g_dcc_buffer_head + g_dcc_buffer_count) % ARRAY_SIZE(g_dcc_buffer); + copy_len = ARRAY_SIZE(g_dcc_buffer) - tail; + space_left = ARRAY_SIZE(g_dcc_buffer) - g_dcc_buffer_count; + if (copy_len > space_left) + copy_len = space_left; + if (copy_len > count) + copy_len = count; + memcpy(&g_dcc_buffer[tail], buf, copy_len); + g_dcc_buffer_count += copy_len; + buf += copy_len; + count -= copy_len; + if (copy_len < count && copy_len < space_left) { + space_left -= copy_len; + copy_len = count; + if (copy_len > space_left) { + copy_len = space_left; + } + memcpy(g_dcc_buffer, buf, copy_len); + buf += copy_len; + count -= copy_len; + g_dcc_buffer_count += copy_len; + } + dcc_poll_locked(); + space_left = ARRAY_SIZE(g_dcc_buffer) - g_dcc_buffer_count; + } while(count && space_left); + spin_unlock_irqrestore(&g_dcc_tty_lock, irq_flags); + return buf - buf_start; +} + +static int dcc_tty_write(struct tty_struct * tty, const unsigned char *buf, int count) +{ + int ret; + /* printk("dcc_tty_write %p, %d\n", buf, count); */ + ret = dcc_write(buf, count); + if (ret != count) + printk("dcc_tty_write %p, %d, returned %d\n", buf, count, ret); + return ret; +} + +static int dcc_tty_write_room(struct tty_struct *tty) +{ + int space_left; + unsigned long irq_flags; + + spin_lock_irqsave(&g_dcc_tty_lock, irq_flags); + space_left = ARRAY_SIZE(g_dcc_buffer) - g_dcc_buffer_count; + spin_unlock_irqrestore(&g_dcc_tty_lock, irq_flags); + return space_left; +} + +static int dcc_tty_chars_in_buffer(struct tty_struct *tty) +{ + int ret; + asm( + "mrc 14, 0, %0, c0, c1, 0\n" + "mov %0, %0, LSR #30\n" + "and %0, %0, #1\n" + : "=r" (ret) + ); + return ret; +} + +static void dcc_tty_unthrottle(struct tty_struct * tty) +{ + unsigned long irq_flags; + + spin_lock_irqsave(&g_dcc_tty_lock, irq_flags); + dcc_poll_locked(); + spin_unlock_irqrestore(&g_dcc_tty_lock, irq_flags); +} + +static enum hrtimer_restart dcc_tty_timer_func(struct hrtimer *timer) +{ + unsigned long irq_flags; + + spin_lock_irqsave(&g_dcc_tty_lock, irq_flags); + dcc_poll_locked(); + spin_unlock_irqrestore(&g_dcc_tty_lock, irq_flags); + return HRTIMER_NORESTART; +} + +void dcc_console_write(struct console *co, const char *b, unsigned count) +{ +#if 1 + dcc_write(b, count); +#else + /* blocking printk */ + while (count > 0) { + int written; + written = dcc_write(b, count); + if (written) { + b += written; + count -= written; + } + } +#endif +} + +static struct tty_driver *dcc_console_device(struct console *c, int *index) +{ + *index = 0; + return g_dcc_tty_driver; +} + +static int __init dcc_console_setup(struct console *co, char *options) +{ + if (co->index != 0) + return -ENODEV; + return 0; +} + + +static struct console dcc_console = +{ + .name = "ttyDCC", + .write = dcc_console_write, + .device = dcc_console_device, + .setup = dcc_console_setup, + .flags = CON_PRINTBUFFER, + .index = -1, +}; + +static struct tty_operations dcc_tty_ops = { + .open = dcc_tty_open, + .close = dcc_tty_close, + .write = dcc_tty_write, + .write_room = dcc_tty_write_room, + .chars_in_buffer = dcc_tty_chars_in_buffer, + .unthrottle = dcc_tty_unthrottle, +}; + +static int __init dcc_tty_init(void) +{ + int ret; + + hrtimer_init(&g_dcc_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + g_dcc_timer.function = dcc_tty_timer_func; + + g_dcc_tty_driver = alloc_tty_driver(1); + if (!g_dcc_tty_driver) { + printk(KERN_ERR "dcc_tty_probe: alloc_tty_driver failed\n"); + ret = -ENOMEM; + goto err_alloc_tty_driver_failed; + } + g_dcc_tty_driver->owner = THIS_MODULE; + g_dcc_tty_driver->driver_name = "dcc"; + g_dcc_tty_driver->name = "ttyDCC"; + g_dcc_tty_driver->major = 0; // auto assign + g_dcc_tty_driver->minor_start = 0; + g_dcc_tty_driver->type = TTY_DRIVER_TYPE_SERIAL; + g_dcc_tty_driver->subtype = SERIAL_TYPE_NORMAL; + g_dcc_tty_driver->init_termios = tty_std_termios; + g_dcc_tty_driver->flags = TTY_DRIVER_RESET_TERMIOS | TTY_DRIVER_REAL_RAW | TTY_DRIVER_DYNAMIC_DEV; + tty_set_operations(g_dcc_tty_driver, &dcc_tty_ops); + ret = tty_register_driver(g_dcc_tty_driver); + if (ret) { + printk(KERN_ERR "dcc_tty_probe: tty_register_driver failed, %d\n", ret); + goto err_tty_register_driver_failed; + } + tty_register_device(g_dcc_tty_driver, 0, NULL); + + register_console(&dcc_console); + hrtimer_start(&g_dcc_timer, ktime_set(0, 0), HRTIMER_MODE_REL); + + return 0; + +err_tty_register_driver_failed: + put_tty_driver(g_dcc_tty_driver); + g_dcc_tty_driver = NULL; +err_alloc_tty_driver_failed: + return ret; +} + +static void __exit dcc_tty_exit(void) +{ + int ret; + + tty_unregister_device(g_dcc_tty_driver, 0); + ret = tty_unregister_driver(g_dcc_tty_driver); + if (ret < 0) { + printk(KERN_ERR "dcc_tty_remove: tty_unregister_driver failed, %d\n", ret); + } else { + put_tty_driver(g_dcc_tty_driver); + } + g_dcc_tty_driver = NULL; +} + +module_init(dcc_tty_init); +module_exit(dcc_tty_exit); + + diff --git a/drivers/char/mem.c b/drivers/char/mem.c index 1ccbe9482faa..38d3069b7f0a 100644 --- a/drivers/char/mem.c +++ b/drivers/char/mem.c @@ -60,6 +60,7 @@ static inline int valid_mmap_phys_addr_range(unsigned long pfn, size_t size) } #endif +#if defined(CONFIG_DEVMEM) || defined(CONFIG_DEVKMEM) #ifdef CONFIG_STRICT_DEVMEM static inline int range_is_allowed(unsigned long pfn, unsigned long size) { @@ -85,7 +86,9 @@ static inline int range_is_allowed(unsigned long pfn, unsigned long size) return 1; } #endif +#endif +#ifdef CONFIG_DEVMEM void __weak unxlate_dev_mem_ptr(unsigned long phys, void *addr) { } @@ -212,6 +215,9 @@ static ssize_t write_mem(struct file *file, const char __user *buf, *ppos += written; return written; } +#endif /* CONFIG_DEVMEM */ + +#if defined(CONFIG_DEVMEM) || defined(CONFIG_DEVKMEM) int __weak phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, unsigned long size, pgprot_t *vma_prot) @@ -333,6 +339,7 @@ static int mmap_mem(struct file *file, struct vm_area_struct *vma) } return 0; } +#endif /* CONFIG_DEVMEM */ #ifdef CONFIG_DEVKMEM static int mmap_kmem(struct file *file, struct vm_area_struct *vma) @@ -727,6 +734,8 @@ static loff_t null_lseek(struct file *file, loff_t offset, int orig) return file->f_pos = 0; } +#if defined(CONFIG_DEVMEM) || defined(CONFIG_DEVKMEM) || defined(CONFIG_DEVPORT) + /* * The memory devices use the full 32/64 bits of the offset, and so we cannot * check against negative addresses: they are ok. The return value is weird, @@ -760,10 +769,14 @@ static loff_t memory_lseek(struct file *file, loff_t offset, int orig) return ret; } +#endif + +#if defined(CONFIG_DEVMEM) || defined(CONFIG_DEVKMEM) || defined(CONFIG_DEVPORT) static int open_port(struct inode *inode, struct file *filp) { return capable(CAP_SYS_RAWIO) ? 0 : -EPERM; } +#endif #define zero_lseek null_lseek #define full_lseek null_lseek @@ -774,6 +787,7 @@ static int open_port(struct inode *inode, struct file *filp) #define open_kmem open_mem #define open_oldmem open_mem +#ifdef CONFIG_DEVMEM static const struct file_operations mem_fops = { .llseek = memory_lseek, .read = read_mem, @@ -782,6 +796,7 @@ static const struct file_operations mem_fops = { .open = open_mem, .get_unmapped_area = get_unmapped_area_mem, }; +#endif #ifdef CONFIG_DEVKMEM static const struct file_operations kmem_fops = { @@ -851,7 +866,9 @@ static const struct memdev { const struct file_operations *fops; struct backing_dev_info *dev_info; } devlist[] = { +#ifdef CONFIG_DEVMEM [1] = { "mem", 0, &mem_fops, &directly_mappable_cdev_bdi }, +#endif #ifdef CONFIG_DEVKMEM [2] = { "kmem", 0, &kmem_fops, &directly_mappable_cdev_bdi }, #endif diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index a9c1324843eb..9e1f7d9b52b5 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -102,6 +102,16 @@ 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. + endchoice config CPU_FREQ_GOV_PERFORMANCE @@ -160,6 +170,24 @@ config CPU_FREQ_GOV_ONDEMAND If in doubt, say N. +config CPU_FREQ_GOV_INTERACTIVE + tristate "'interactive' cpufreq policy governor" + default n + 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. + + To compile this driver as a module, choose M here: the + module will be called cpufreq_interactive. + + 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 diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile index 505c62bceb9d..a8f252b32b8c 100644 --- a/drivers/cpufreq/Makefile +++ b/drivers/cpufreq/Makefile @@ -9,6 +9,7 @@ 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 # CPUfreq cross-arch helpers diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 66f6cf5064ec..016294db2a79 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -17,7 +17,9 @@ #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#include <asm/cputime.h> #include <linux/kernel.h> +#include <linux/kernel_stat.h> #include <linux/module.h> #include <linux/init.h> #include <linux/notifier.h> @@ -25,6 +27,7 @@ #include <linux/delay.h> #include <linux/interrupt.h> #include <linux/spinlock.h> +#include <linux/tick.h> #include <linux/device.h> #include <linux/slab.h> #include <linux/cpu.h> @@ -133,6 +136,51 @@ bool have_governor_per_policy(void) { return cpufreq_driver->have_governor_per_policy; } +EXPORT_SYMBOL_GPL(have_governor_per_policy); + +struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy) +{ + if (have_governor_per_policy()) + return &policy->kobj; + else + return cpufreq_global_kobject; +} +EXPORT_SYMBOL_GPL(get_governor_parent_kobj); + +static inline u64 get_cpu_idle_time_jiffy(unsigned int cpu, u64 *wall) +{ + u64 idle_time; + u64 cur_wall_time; + u64 busy_time; + + cur_wall_time = jiffies64_to_cputime64(get_jiffies_64()); + + busy_time = kcpustat_cpu(cpu).cpustat[CPUTIME_USER]; + busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_SYSTEM]; + busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_IRQ]; + busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_SOFTIRQ]; + busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_STEAL]; + busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_NICE]; + + idle_time = cur_wall_time - busy_time; + if (wall) + *wall = cputime_to_usecs(cur_wall_time); + + return cputime_to_usecs(idle_time); +} + +u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy) +{ + u64 idle_time = get_cpu_idle_time_us(cpu, io_busy ? wall : NULL); + + if (idle_time == -1ULL) + return get_cpu_idle_time_jiffy(cpu, wall); + else if (!io_busy) + idle_time += get_cpu_iowait_time_us(cpu, wall); + + return idle_time; +} +EXPORT_SYMBOL_GPL(get_cpu_idle_time); static struct cpufreq_policy *__cpufreq_cpu_get(unsigned int cpu, bool sysfs) { diff --git a/drivers/cpufreq/cpufreq_governor.c b/drivers/cpufreq/cpufreq_governor.c index 27b0e2a295ea..ece1df8eac85 100644 --- a/drivers/cpufreq/cpufreq_governor.c +++ b/drivers/cpufreq/cpufreq_governor.c @@ -23,20 +23,11 @@ #include <linux/kernel_stat.h> #include <linux/mutex.h> #include <linux/slab.h> -#include <linux/tick.h> #include <linux/types.h> #include <linux/workqueue.h> #include "cpufreq_governor.h" -static struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy) -{ - if (have_governor_per_policy()) - return &policy->kobj; - else - return cpufreq_global_kobject; -} - static struct attribute_group *get_sysfs_attr(struct dbs_data *dbs_data) { if (have_governor_per_policy()) @@ -45,41 +36,6 @@ static struct attribute_group *get_sysfs_attr(struct dbs_data *dbs_data) return dbs_data->cdata->attr_group_gov_sys; } -static inline u64 get_cpu_idle_time_jiffy(unsigned int cpu, u64 *wall) -{ - u64 idle_time; - u64 cur_wall_time; - u64 busy_time; - - cur_wall_time = jiffies64_to_cputime64(get_jiffies_64()); - - busy_time = kcpustat_cpu(cpu).cpustat[CPUTIME_USER]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_SYSTEM]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_IRQ]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_SOFTIRQ]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_STEAL]; - busy_time += kcpustat_cpu(cpu).cpustat[CPUTIME_NICE]; - - idle_time = cur_wall_time - busy_time; - if (wall) - *wall = cputime_to_usecs(cur_wall_time); - - return cputime_to_usecs(idle_time); -} - -u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy) -{ - u64 idle_time = get_cpu_idle_time_us(cpu, io_busy ? wall : NULL); - - if (idle_time == -1ULL) - return get_cpu_idle_time_jiffy(cpu, wall); - else if (!io_busy) - idle_time += get_cpu_iowait_time_us(cpu, wall); - - return idle_time; -} -EXPORT_SYMBOL_GPL(get_cpu_idle_time); - void dbs_check_cpu(struct dbs_data *dbs_data, int cpu) { struct cpu_dbs_common_info *cdbs = dbs_data->cdata->get_cpu_cdbs(cpu); @@ -396,6 +352,7 @@ int cpufreq_governor_dbs(struct cpufreq_policy *policy, mutex_lock(&dbs_data->mutex); mutex_destroy(&cpu_cdbs->timer_mutex); + cpu_cdbs->cur_policy = NULL; mutex_unlock(&dbs_data->mutex); diff --git a/drivers/cpufreq/cpufreq_governor.h b/drivers/cpufreq/cpufreq_governor.h index 4a9058aeb57e..c8028ce7554e 100644 --- a/drivers/cpufreq/cpufreq_governor.h +++ b/drivers/cpufreq/cpufreq_governor.h @@ -255,7 +255,6 @@ static ssize_t show_sampling_rate_min_gov_pol \ return sprintf(buf, "%u\n", dbs_data->min_sampling_rate); \ } -u64 get_cpu_idle_time(unsigned int cpu, u64 *wall, int io_busy); void dbs_check_cpu(struct dbs_data *dbs_data, int cpu); bool need_load_eval(struct cpu_dbs_common_info *cdbs, unsigned int sampling_rate); diff --git a/drivers/cpufreq/cpufreq_interactive.c b/drivers/cpufreq/cpufreq_interactive.c new file mode 100644 index 000000000000..cb477d6a21d6 --- /dev/null +++ b/drivers/cpufreq/cpufreq_interactive.c @@ -0,0 +1,1375 @@ +/* + * 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> +#include "cpufreq_governor.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; + unsigned int max_freq; + u64 floor_validate_time; + u64 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; + + 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->target_freq < 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->target_freq < tunables->hispeed_freq) + new_freq = tunables->hispeed_freq; + } + + if (pcpu->target_freq >= tunables->hispeed_freq && + new_freq > pcpu->target_freq && + now - pcpu->hispeed_validate_time < + freq_to_above_hispeed_delay(tunables, pcpu->target_freq)) { + 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->hispeed_validate_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. + */ + if (new_freq < pcpu->floor_freq) { + if (now - pcpu->floor_validate_time < + 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; + pcpu->floor_validate_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_if_notmax; + } + + 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_notmax: + /* + * Already set max speed and don't see a need to change that, + * wait until next idle to re-evaluate, don't need timer. + */ + if (pcpu->target_freq == pcpu->policy->max) + goto exit; + +rearm: + if (!timer_pending(&pcpu->cpu_timer)) + cpufreq_interactive_timer_resched(pcpu); + +exit: + up_read(&pcpu->enable_sem); + return; +} + +static void cpufreq_interactive_idle_start(void) +{ + struct cpufreq_interactive_cpuinfo *pcpu = + &per_cpu(cpuinfo, smp_processor_id()); + int pending; + + if (!down_read_trylock(&pcpu->enable_sem)) + return; + if (!pcpu->governor_enabled) { + up_read(&pcpu->enable_sem); + return; + } + + pending = timer_pending(&pcpu->cpu_timer); + + if (pcpu->target_freq != pcpu->policy->min) { + /* + * Entering idle while not at lowest speed. On some + * platforms this can hold the other CPU(s) at that speed + * even though the CPU is idle. Set a timer to re-evaluate + * speed so this idle CPU doesn't hold the other CPUs above + * min indefinitely. This should probably be a quirk of + * the CPUFreq driver. + */ + if (!pending) + cpufreq_interactive_timer_resched(pcpu); + } + + up_read(&pcpu->enable_sem); +} + +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 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) { + unsigned int j; + unsigned int max_freq = 0; + + pcpu = &per_cpu(cpuinfo, cpu); + if (!down_read_trylock(&pcpu->enable_sem)) + continue; + if (!pcpu->governor_enabled) { + up_read(&pcpu->enable_sem); + continue; + } + + for_each_cpu(j, pcpu->policy->cpus) { + struct cpufreq_interactive_cpuinfo *pjcpu = + &per_cpu(cpuinfo, j); + + if (pjcpu->target_freq > max_freq) + max_freq = pjcpu->target_freq; + } + + if (max_freq != pcpu->policy->cur) + __cpufreq_driver_target(pcpu->policy, + max_freq, + CPUFREQ_RELATION_H); + trace_cpufreq_interactive_setspeed(cpu, + pcpu->target_freq, + pcpu->policy->cur); + + up_read(&pcpu->enable_sem); + } + } + + 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 (tunables != pcpu->policy->governor_data) + 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->hispeed_validate_time = + ktime_to_us(ktime_get()); + anyboost = 1; + } + + /* + * Set floor freq and (re)start timer for when last + * validated. + */ + + pcpu->floor_freq = tunables->hispeed_freq; + pcpu->floor_validate_time = ktime_to_us(ktime_get()); + spin_unlock_irqrestore(&pcpu->target_freq_lock, flags[1]); + } + + 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 = strict_strtoul(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 = strict_strtoul(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 = strict_strtoul(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; + + ret = strict_strtoul(buf, 0, &val); + if (ret < 0) + return ret; + tunables->timer_rate = val; + 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) +{ + switch (val) { + case IDLE_START: + cpufreq_interactive_idle_start(); + break; + case IDLE_END: + cpufreq_interactive_idle_end(); + break; + } + + 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; + + WARN_ON(!tunables && (event != CPUFREQ_GOV_POLICY_INIT)); + + 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->floor_validate_time = + ktime_to_us(ktime_get()); + pcpu->hispeed_validate_time = + pcpu->floor_validate_time; + pcpu->max_freq = policy->max; + 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); + + /* Reschedule timer only if policy->max is raised. + * Delete the timers, else the timer callback may + * return without re-arm the timer when failed + * acquire the semaphore. This race may cause timer + * stopped unexpectedly. + */ + + if (policy->max > pcpu->max_freq) { + down_write(&pcpu->enable_sem); + del_timer_sync(&pcpu->cpu_timer); + del_timer_sync(&pcpu->cpu_slack_timer); + cpufreq_interactive_timer_start(tunables, j); + up_write(&pcpu->enable_sem); + } + + pcpu->max_freq = policy->max; + } + 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 }; + + /* 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, ¶m); + get_task_struct(speedchange_task); + + /* NB: wake up so the thread does not look hung to the freezer */ + wake_up_process(speedchange_task); + + return cpufreq_register_governor(&cpufreq_gov_interactive); +} + +#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/cpufreq/cpufreq_stats.c b/drivers/cpufreq/cpufreq_stats.c index 9823037f194b..690af18f55a9 100644 --- a/drivers/cpufreq/cpufreq_stats.c +++ b/drivers/cpufreq/cpufreq_stats.c @@ -20,6 +20,8 @@ #include <linux/kobject.h> #include <linux/spinlock.h> #include <linux/notifier.h> +#include <linux/sort.h> +#include <linux/err.h> #include <asm/cputime.h> #ifdef CONFIG_BL_SWITCHER #include <asm/bL_switcher.h> @@ -41,6 +43,20 @@ struct cpufreq_stats { #endif }; +struct all_cpufreq_stats { + unsigned int state_num; + cputime64_t *time_in_state; + unsigned int *freq_table; +}; + +struct all_freq_table { + unsigned int *freq_table; + unsigned int table_size; +}; + +static struct all_freq_table *all_freq_table; + +static DEFINE_PER_CPU(struct all_cpufreq_stats *, all_cpufreq_stats); static DEFINE_PER_CPU(struct cpufreq_stats *, cpufreq_stats_table); struct cpufreq_stats_attribute { @@ -51,14 +67,24 @@ struct cpufreq_stats_attribute { static int cpufreq_stats_update(unsigned int cpu) { struct cpufreq_stats *stat; + struct all_cpufreq_stats *all_stat; unsigned long long cur_time; cur_time = get_jiffies_64(); spin_lock(&cpufreq_stats_lock); stat = per_cpu(cpufreq_stats_table, cpu); - if (stat->time_in_state) + all_stat = per_cpu(all_cpufreq_stats, cpu); + if (!stat) { + spin_unlock(&cpufreq_stats_lock); + return 0; + } + if (stat->time_in_state) { stat->time_in_state[stat->last_index] += cur_time - stat->last_time; + if (all_stat) + all_stat->time_in_state[stat->last_index] += + cur_time - stat->last_time; + } stat->last_time = cur_time; spin_unlock(&cpufreq_stats_lock); return 0; @@ -89,6 +115,62 @@ static ssize_t show_time_in_state(struct cpufreq_policy *policy, char *buf) return len; } +static int get_index_all_cpufreq_stat(struct all_cpufreq_stats *all_stat, + unsigned int freq) +{ + int i; + if (!all_stat) + return -1; + for (i = 0; i < all_stat->state_num; i++) { + if (all_stat->freq_table[i] == freq) + return i; + } + return -1; +} + +static ssize_t show_all_time_in_state(struct kobject *kobj, + struct kobj_attribute *attr, char *buf) +{ + ssize_t len = 0; + unsigned int i, cpu, freq, index; + struct all_cpufreq_stats *all_stat; + struct cpufreq_policy *policy; + + len += scnprintf(buf + len, PAGE_SIZE - len, "freq\t\t"); + for_each_possible_cpu(cpu) { + len += scnprintf(buf + len, PAGE_SIZE - len, "cpu%d\t\t", cpu); + if (cpu_online(cpu)) + cpufreq_stats_update(cpu); + } + + if (!all_freq_table) + goto out; + for (i = 0; i < all_freq_table->table_size; i++) { + freq = all_freq_table->freq_table[i]; + len += scnprintf(buf + len, PAGE_SIZE - len, "\n%u\t\t", freq); + for_each_possible_cpu(cpu) { + policy = cpufreq_cpu_get(cpu); + if (policy == NULL) + continue; + all_stat = per_cpu(all_cpufreq_stats, policy->cpu); + index = get_index_all_cpufreq_stat(all_stat, freq); + if (index != -1) { + len += scnprintf(buf + len, PAGE_SIZE - len, + "%llu\t\t", (unsigned long long) + cputime64_to_clock_t(all_stat->time_in_state[index])); + } else { + len += scnprintf(buf + len, PAGE_SIZE - len, + "N/A\t\t"); + } + cpufreq_cpu_put(policy); + } + } + +out: + len += scnprintf(buf + len, PAGE_SIZE - len, "\n"); + return len; +} + #ifdef CONFIG_CPU_FREQ_STAT_DETAILS static ssize_t show_trans_table(struct cpufreq_policy *policy, char *buf) { @@ -152,6 +234,9 @@ static struct attribute_group stats_attr_group = { .name = "stats" }; +static struct kobj_attribute _attr_all_time_in_state = __ATTR(all_time_in_state, + 0444, show_all_time_in_state, NULL); + static int freq_table_get_index(struct cpufreq_stats *stat, unsigned int freq) { int index; @@ -198,6 +283,29 @@ put_ref: cpufreq_cpu_put(policy); } +static void cpufreq_allstats_free(void) +{ + int cpu; + struct all_cpufreq_stats *all_stat; + + sysfs_remove_file(cpufreq_global_kobject, + &_attr_all_time_in_state.attr); + + for_each_possible_cpu(cpu) { + all_stat = per_cpu(all_cpufreq_stats, cpu); + if (!all_stat) + continue; + kfree(all_stat->time_in_state); + kfree(all_stat); + per_cpu(all_cpufreq_stats, cpu) = NULL; + } + if (all_freq_table) { + kfree(all_freq_table->freq_table); + kfree(all_freq_table); + all_freq_table = NULL; + } +} + static int cpufreq_stats_create_table(struct cpufreq_policy *policy, struct cpufreq_frequency_table *table) { @@ -284,6 +392,106 @@ static void cpufreq_stats_update_policy_cpu(struct cpufreq_policy *policy) stat->cpu = policy->cpu; } +static int compare_for_sort(const void *lhs_ptr, const void *rhs_ptr) +{ + unsigned int lhs = *(const unsigned int *)(lhs_ptr); + unsigned int rhs = *(const unsigned int *)(rhs_ptr); + if (lhs < rhs) + return -1; + if (lhs > rhs) + return 1; + return 0; +} + +static bool check_all_freq_table(unsigned int freq) +{ + int i; + for (i = 0; i < all_freq_table->table_size; i++) { + if (freq == all_freq_table->freq_table[i]) + return true; + } + return false; +} + +static void create_all_freq_table(void) +{ + all_freq_table = kzalloc(sizeof(struct all_freq_table), + GFP_KERNEL); + if (!all_freq_table) + pr_warn("could not allocate memory for all_freq_table\n"); + return; +} + +static void add_all_freq_table(unsigned int freq) +{ + unsigned int size; + size = sizeof(unsigned int) * (all_freq_table->table_size + 1); + all_freq_table->freq_table = krealloc(all_freq_table->freq_table, + size, GFP_ATOMIC); + if (IS_ERR(all_freq_table->freq_table)) { + pr_warn("Could not reallocate memory for freq_table\n"); + all_freq_table->freq_table = NULL; + return; + } + all_freq_table->freq_table[all_freq_table->table_size++] = freq; +} + +static void cpufreq_allstats_create(unsigned int cpu) +{ + int i , j = 0; + unsigned int alloc_size, count = 0; + struct cpufreq_frequency_table *table = cpufreq_frequency_get_table(cpu); + struct all_cpufreq_stats *all_stat; + bool sort_needed = false; + + if (!table) + return; + + for (i = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) { + unsigned int freq = table[i].frequency; + if (freq == CPUFREQ_ENTRY_INVALID) + continue; + count++; + } + + all_stat = kzalloc(sizeof(struct all_cpufreq_stats), + GFP_KERNEL); + if (!all_stat) { + pr_warn("Cannot allocate memory for cpufreq stats\n"); + return; + } + + /*Allocate memory for freq table per cpu as well as clockticks per freq*/ + alloc_size = count * sizeof(int) + count * sizeof(cputime64_t); + all_stat->time_in_state = kzalloc(alloc_size, GFP_KERNEL); + if (!all_stat->time_in_state) { + pr_warn("Cannot allocate memory for cpufreq time_in_state\n"); + kfree(all_stat); + all_stat = NULL; + return; + } + all_stat->freq_table = (unsigned int *) + (all_stat->time_in_state + count); + + spin_lock(&cpufreq_stats_lock); + for (i = 0; table[i].frequency != CPUFREQ_TABLE_END; i++) { + unsigned int freq = table[i].frequency; + if (freq == CPUFREQ_ENTRY_INVALID) + continue; + all_stat->freq_table[j++] = freq; + if (all_freq_table && !check_all_freq_table(freq)) { + add_all_freq_table(freq); + sort_needed = true; + } + } + if (sort_needed) + sort(all_freq_table->freq_table, all_freq_table->table_size, + sizeof(unsigned int), &compare_for_sort, NULL); + all_stat->state_num = j; + per_cpu(all_cpufreq_stats, cpu) = all_stat; + spin_unlock(&cpufreq_stats_lock); +} + static int cpufreq_stat_notifier_policy(struct notifier_block *nb, unsigned long val, void *data) { @@ -302,6 +510,10 @@ static int cpufreq_stat_notifier_policy(struct notifier_block *nb, table = cpufreq_frequency_get_table(cpu); if (!table) return 0; + + if (!per_cpu(all_cpufreq_stats, cpu)) + cpufreq_allstats_create(cpu); + ret = cpufreq_stats_create_table(policy, table); if (ret) return ret; @@ -344,6 +556,30 @@ static int cpufreq_stat_notifier_trans(struct notifier_block *nb, return 0; } +static int cpufreq_stats_create_table_cpu(unsigned int cpu) +{ + struct cpufreq_policy *policy; + struct cpufreq_frequency_table *table; + int ret = -ENODEV; + + policy = cpufreq_cpu_get(cpu); + if (!policy) + return -ENODEV; + + table = cpufreq_frequency_get_table(cpu); + if (!table) + goto out; + + if (!per_cpu(all_cpufreq_stats, cpu)) + cpufreq_allstats_create(cpu); + + ret = cpufreq_stats_create_table(policy, table); + +out: + cpufreq_cpu_put(policy); + return ret; +} + static int __cpuinit cpufreq_stat_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu) @@ -363,6 +599,10 @@ static int __cpuinit cpufreq_stat_cpu_callback(struct notifier_block *nfb, case CPU_DEAD_FROZEN: cpufreq_stats_free_table(cpu); break; + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: + cpufreq_stats_create_table_cpu(cpu); + break; } return NOTIFY_OK; } @@ -407,6 +647,12 @@ static int cpufreq_stats_setup(void) return ret; } + create_all_freq_table(); + ret = sysfs_create_file(cpufreq_global_kobject, + &_attr_all_time_in_state.attr); + if (ret) + pr_warn("Error creating sysfs file for cpufreq stats\n"); + return 0; } @@ -423,6 +669,7 @@ static void cpufreq_stats_cleanup(void) cpufreq_stats_free_table(cpu); cpufreq_stats_free_sysfs(cpu); } + cpufreq_allstats_free(); } #ifdef CONFIG_BL_SWITCHER diff --git a/drivers/cpuidle/governors/menu.c b/drivers/cpuidle/governors/menu.c index bc580b67a652..efe23dd391ed 100644 --- a/drivers/cpuidle/governors/menu.c +++ b/drivers/cpuidle/governors/menu.c @@ -126,7 +126,7 @@ struct menu_device { #define LOAD_INT(x) ((x) >> FSHIFT) #define LOAD_FRAC(x) LOAD_INT(((x) & (FIXED_1-1)) * 100) -static int get_loadavg(void) +static int __maybe_unused get_loadavg(void) { unsigned long this = this_cpu_load(); @@ -173,7 +173,12 @@ static inline int performance_multiplier(void) /* for higher loadavg, we are more reluctant */ - mult += 2 * get_loadavg(); + /* + * 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_iowait_cpu(smp_processor_id()); diff --git a/drivers/gpio/gpiolib.c b/drivers/gpio/gpiolib.c index c3768fafff45..9002122c52ae 100644 --- a/drivers/gpio/gpiolib.c +++ b/drivers/gpio/gpiolib.c @@ -136,7 +136,7 @@ static struct gpio_desc *gpio_to_desc(unsigned gpio) */ static int desc_to_gpio(const struct gpio_desc *desc) { - return desc->chip->base + gpio_chip_hwgpio(desc); + return desc - &gpio_desc[0]; } @@ -1242,15 +1242,14 @@ int gpiochip_add(struct gpio_chip *chip) } } + spin_unlock_irqrestore(&gpio_lock, flags); + #ifdef CONFIG_PINCTRL INIT_LIST_HEAD(&chip->pin_ranges); #endif of_gpiochip_add(chip); -unlock: - spin_unlock_irqrestore(&gpio_lock, flags); - if (status) goto fail; @@ -1263,6 +1262,9 @@ unlock: chip->label ? : "generic"); return 0; + +unlock: + spin_unlock_irqrestore(&gpio_lock, flags); fail: /* failures here can mean systems won't boot... */ pr_err("gpiochip_add: gpios %d..%d (%s) failed to register\n", diff --git a/drivers/hid/hid-debug.c b/drivers/hid/hid-debug.c index 8453214ec376..941ab3c287ec 100644 --- a/drivers/hid/hid-debug.c +++ b/drivers/hid/hid-debug.c @@ -768,6 +768,8 @@ static const char *keys[KEY_MAX + 1] = { [KEY_ALTERASE] = "AlternateErase", [KEY_CANCEL] = "Cancel", [KEY_BRIGHTNESSDOWN] = "BrightnessDown", [KEY_BRIGHTNESSUP] = "BrightnessUp", [KEY_MEDIA] = "Media", [KEY_UNKNOWN] = "Unknown", + [BTN_DPAD_UP] = "BtnDPadUp", [BTN_DPAD_DOWN] = "BtnDPadDown", + [BTN_DPAD_LEFT] = "BtnDPadLeft", [BTN_DPAD_RIGHT] = "BtnDPadRight", [BTN_0] = "Btn0", [BTN_1] = "Btn1", [BTN_2] = "Btn2", [BTN_3] = "Btn3", [BTN_4] = "Btn4", [BTN_5] = "Btn5", @@ -797,7 +799,8 @@ static const char *keys[KEY_MAX + 1] = { [BTN_TOOL_MOUSE] = "ToolMouse", [BTN_TOOL_LENS] = "ToolLens", [BTN_TOUCH] = "Touch", [BTN_STYLUS] = "Stylus", [BTN_STYLUS2] = "Stylus2", [BTN_TOOL_DOUBLETAP] = "ToolDoubleTap", - [BTN_TOOL_TRIPLETAP] = "ToolTripleTap", [BTN_GEAR_DOWN] = "WheelBtn", + [BTN_TOOL_TRIPLETAP] = "ToolTripleTap", [BTN_TOOL_QUADTAP] = "ToolQuadrupleTap", + [BTN_GEAR_DOWN] = "WheelBtn", [BTN_GEAR_UP] = "Gear up", [KEY_OK] = "Ok", [KEY_SELECT] = "Select", [KEY_GOTO] = "Goto", [KEY_CLEAR] = "Clear", [KEY_POWER2] = "Power2", @@ -852,6 +855,16 @@ static const char *keys[KEY_MAX + 1] = { [KEY_KBDILLUMDOWN] = "KbdIlluminationDown", [KEY_KBDILLUMUP] = "KbdIlluminationUp", [KEY_SWITCHVIDEOMODE] = "SwitchVideoMode", + [KEY_BUTTONCONFIG] = "ButtonConfig", + [KEY_TASKMANAGER] = "TaskManager", + [KEY_JOURNAL] = "Journal", + [KEY_CONTROLPANEL] = "ControlPanel", + [KEY_APPSELECT] = "AppSelect", + [KEY_SCREENSAVER] = "ScreenSaver", + [KEY_VOICECOMMAND] = "VoiceCommand", + [KEY_BRIGHTNESS_MIN] = "BrightnessMin", + [KEY_BRIGHTNESS_MAX] = "BrightnessMax", + [KEY_BRIGHTNESS_AUTO] = "BrightnessAuto", }; static const char *relatives[REL_MAX + 1] = { diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c index a3915d12e746..1419f8bd11ba 100644 --- a/drivers/hid/hid-input.c +++ b/drivers/hid/hid-input.c @@ -729,6 +729,13 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x06c: map_key_clear(KEY_YELLOW); break; case 0x06d: map_key_clear(KEY_ZOOM); break; + case 0x06f: map_key_clear(KEY_BRIGHTNESSUP); break; + case 0x070: map_key_clear(KEY_BRIGHTNESSDOWN); break; + case 0x072: map_key_clear(KEY_BRIGHTNESS_TOGGLE); break; + case 0x073: map_key_clear(KEY_BRIGHTNESS_MIN); break; + case 0x074: map_key_clear(KEY_BRIGHTNESS_MAX); break; + case 0x075: map_key_clear(KEY_BRIGHTNESS_AUTO); break; + case 0x082: map_key_clear(KEY_VIDEO_NEXT); break; case 0x083: map_key_clear(KEY_LAST); break; case 0x084: map_key_clear(KEY_ENTER); break; @@ -769,6 +776,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x0bf: map_key_clear(KEY_SLOW); break; case 0x0cd: map_key_clear(KEY_PLAYPAUSE); break; + case 0x0cf: map_key_clear(KEY_VOICECOMMAND); break; case 0x0e0: map_abs_clear(ABS_VOLUME); break; case 0x0e2: map_key_clear(KEY_MUTE); break; case 0x0e5: map_key_clear(KEY_BASSBOOST); break; @@ -776,6 +784,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x0ea: map_key_clear(KEY_VOLUMEDOWN); break; case 0x0f5: map_key_clear(KEY_SLOW); break; + case 0x181: map_key_clear(KEY_BUTTONCONFIG); break; case 0x182: map_key_clear(KEY_BOOKMARKS); break; case 0x183: map_key_clear(KEY_CONFIG); break; case 0x184: map_key_clear(KEY_WORDPROCESSOR); break; @@ -789,6 +798,8 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x18c: map_key_clear(KEY_VOICEMAIL); break; case 0x18d: map_key_clear(KEY_ADDRESSBOOK); break; case 0x18e: map_key_clear(KEY_CALENDAR); break; + case 0x18f: map_key_clear(KEY_TASKMANAGER); break; + case 0x190: map_key_clear(KEY_JOURNAL); break; case 0x191: map_key_clear(KEY_FINANCE); break; case 0x192: map_key_clear(KEY_CALC); break; case 0x193: map_key_clear(KEY_PLAYER); break; @@ -797,10 +808,16 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel case 0x199: map_key_clear(KEY_CHAT); break; case 0x19c: map_key_clear(KEY_LOGOFF); break; case 0x19e: map_key_clear(KEY_COFFEE); break; + case 0x19f: map_key_clear(KEY_CONTROLPANEL); break; + case 0x1a2: map_key_clear(KEY_APPSELECT); break; + case 0x1a3: map_key_clear(KEY_NEXT); break; + case 0x1a4: map_key_clear(KEY_PREVIOUS); break; case 0x1a6: map_key_clear(KEY_HELP); break; case 0x1a7: map_key_clear(KEY_DOCUMENTS); break; case 0x1ab: map_key_clear(KEY_SPELLCHECK); break; case 0x1ae: map_key_clear(KEY_KEYBOARD); break; + case 0x1b1: map_key_clear(KEY_SCREENSAVER); break; + case 0x1b4: map_key_clear(KEY_FILE); break; case 0x1b6: map_key_clear(KEY_IMAGES); break; case 0x1b7: map_key_clear(KEY_AUDIO); break; case 0x1b8: map_key_clear(KEY_VIDEO); break; @@ -1358,8 +1375,9 @@ int hidinput_connect(struct hid_device *hid, unsigned int force) * UGCI) cram a lot of unrelated inputs into the * same interface. */ hidinput->report = report; - if (drv->input_configured) - drv->input_configured(hid, hidinput); + if (drv->input_configured && + drv->input_configured(hid, hidinput)) + goto out_cleanup; if (input_register_device(hidinput->input)) goto out_cleanup; hidinput = NULL; @@ -1380,8 +1398,9 @@ int hidinput_connect(struct hid_device *hid, unsigned int force) } if (hidinput) { - if (drv->input_configured) - drv->input_configured(hid, hidinput); + if (drv->input_configured && + drv->input_configured(hid, hidinput)) + goto out_cleanup; if (input_register_device(hidinput->input)) goto out_cleanup; } diff --git a/drivers/hid/hid-multitouch.c b/drivers/hid/hid-multitouch.c index 3d8e58ac7499..fb9ac1266719 100644 --- a/drivers/hid/hid-multitouch.c +++ b/drivers/hid/hid-multitouch.c @@ -443,6 +443,16 @@ static int mt_touch_input_mapping(struct hid_device *hdev, struct hid_input *hi, (usage->hid & HID_USAGE_PAGE) == HID_UP_BUTTON) td->mt_flags |= INPUT_MT_POINTER; + /* Only map fields from TouchScreen or TouchPad collections. + * We need to ignore fields that belong to other collections + * such as Mouse that might have the same GenericDesktop usages. */ + if (field->application == HID_DG_TOUCHSCREEN) + set_bit(INPUT_PROP_DIRECT, hi->input->propbit); + else if (field->application == HID_DG_TOUCHPAD) + set_bit(INPUT_PROP_POINTER, hi->input->propbit); + else + return 0; + if (usage->usage_index) prev_usage = &field->usage[usage->usage_index - 1]; @@ -772,12 +782,13 @@ static void mt_touch_report(struct hid_device *hid, struct hid_report *report) mt_sync_frame(td, report->field[0]->hidinput->input); } -static void mt_touch_input_configured(struct hid_device *hdev, +static int mt_touch_input_configured(struct hid_device *hdev, struct hid_input *hi) { struct mt_device *td = hid_get_drvdata(hdev); struct mt_class *cls = &td->mtclass; struct input_dev *input = hi->input; + int ret; if (!td->maxcontacts) td->maxcontacts = MT_DEFAULT_MAXCONTACT; @@ -792,9 +803,12 @@ static void mt_touch_input_configured(struct hid_device *hdev, if (cls->quirks & MT_QUIRK_NOT_SEEN_MEANS_UP) td->mt_flags |= INPUT_MT_DROP_UNUSED; - input_mt_init_slots(input, td->maxcontacts, td->mt_flags); + ret = input_mt_init_slots(input, td->maxcontacts, td->mt_flags); + if (ret) + return ret; td->mt_flags = 0; + return 0; } static int mt_input_mapping(struct hid_device *hdev, struct hid_input *hi, @@ -927,19 +941,21 @@ static void mt_post_parse(struct mt_device *td) cls->quirks &= ~MT_QUIRK_CONTACT_CNT_ACCURATE; } -static void mt_input_configured(struct hid_device *hdev, struct hid_input *hi) +static int mt_input_configured(struct hid_device *hdev, struct hid_input *hi) { struct mt_device *td = hid_get_drvdata(hdev); char *name = kstrdup(hdev->name, GFP_KERNEL); + int ret = 0; if (name) hi->input->name = name; if (hi->report->id == td->mt_report_id) - mt_touch_input_configured(hdev, hi); + ret = mt_touch_input_configured(hdev, hi); if (hi->report->id == td->pen_report_id) mt_pen_input_configured(hdev, hi); + return ret; } static int mt_probe(struct hid_device *hdev, const struct hid_device_id *id) diff --git a/drivers/iio/industrialio-event.c b/drivers/iio/industrialio-event.c index 10aa9ef86cec..145c98617936 100644 --- a/drivers/iio/industrialio-event.c +++ b/drivers/iio/industrialio-event.c @@ -35,6 +35,7 @@ */ struct iio_event_interface { wait_queue_head_t wait; + struct mutex read_lock; DECLARE_KFIFO(det_events, struct iio_event_data, 16); struct list_head dev_attr_list; @@ -97,14 +98,16 @@ static ssize_t iio_event_chrdev_read(struct file *filep, if (count < sizeof(struct iio_event_data)) return -EINVAL; - spin_lock_irq(&ev_int->wait.lock); + if (mutex_lock_interruptible(&ev_int->read_lock)) + return -ERESTARTSYS; + if (kfifo_is_empty(&ev_int->det_events)) { if (filep->f_flags & O_NONBLOCK) { ret = -EAGAIN; goto error_unlock; } /* Blocking on device; waiting for something to be there */ - ret = wait_event_interruptible_locked_irq(ev_int->wait, + ret = wait_event_interruptible(ev_int->wait, !kfifo_is_empty(&ev_int->det_events)); if (ret) goto error_unlock; @@ -114,7 +117,7 @@ static ssize_t iio_event_chrdev_read(struct file *filep, ret = kfifo_to_user(&ev_int->det_events, buf, count, &copied); error_unlock: - spin_unlock_irq(&ev_int->wait.lock); + mutex_unlock(&ev_int->read_lock); return ret ? ret : copied; } @@ -371,6 +374,7 @@ static void iio_setup_ev_int(struct iio_event_interface *ev_int) { INIT_KFIFO(ev_int->det_events); init_waitqueue_head(&ev_int->wait); + mutex_init(&ev_int->read_lock); } static const char *iio_event_group_name = "events"; @@ -434,6 +438,7 @@ int iio_device_register_eventset(struct iio_dev *indio_dev) error_free_setup_event_lines: __iio_remove_event_config_attrs(indio_dev); + mutex_destroy(&indio_dev->event_interface->read_lock); kfree(indio_dev->event_interface); error_ret: @@ -446,5 +451,6 @@ void iio_device_unregister_eventset(struct iio_dev *indio_dev) return; __iio_remove_event_config_attrs(indio_dev); kfree(indio_dev->event_interface->group.attrs); + mutex_destroy(&indio_dev->event_interface->read_lock); kfree(indio_dev->event_interface); } diff --git a/drivers/input/Kconfig b/drivers/input/Kconfig index a11ff74a5127..518efa2a9f52 100644 --- a/drivers/input/Kconfig +++ b/drivers/input/Kconfig @@ -174,6 +174,25 @@ config INPUT_APMPOWER To compile this driver as a module, choose M here: the module will be called apm-power. +config INPUT_KEYRESET + tristate "Reset key" + depends on INPUT + select INPUT_KEYCOMBO + ---help--- + Say Y here if you want to reboot when some keys are pressed; + + To compile this driver as a module, choose M here: the + module will be called keyreset. + +config INPUT_KEYCOMBO + tristate "Key combo" + depends on INPUT + ---help--- + Say Y here if you want to take action when some keys are pressed; + + To compile this driver as a module, choose M here: the + module will be called keycombo. + comment "Input Device Drivers" source "drivers/input/keyboard/Kconfig" diff --git a/drivers/input/Makefile b/drivers/input/Makefile index 5ca3f631497f..ee4c06520bb4 100644 --- a/drivers/input/Makefile +++ b/drivers/input/Makefile @@ -25,3 +25,6 @@ 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/evdev.c b/drivers/input/evdev.c index c122dd2adc22..f4897c8c1500 100644 --- a/drivers/input/evdev.c +++ b/drivers/input/evdev.c @@ -26,6 +26,7 @@ #include <linux/major.h> #include <linux/device.h> #include <linux/cdev.h> +#include <linux/wakelock.h> #include "input-compat.h" struct evdev { @@ -46,6 +47,9 @@ struct evdev_client { unsigned int tail; unsigned int packet_head; /* [future] position of the first element of next packet */ spinlock_t buffer_lock; /* protects access to buffer, head and tail */ + struct wake_lock wake_lock; + bool use_wake_lock; + char name[28]; struct fasync_struct *fasync; struct evdev *evdev; struct list_head node; @@ -73,10 +77,14 @@ static void __pass_event(struct evdev_client *client, client->buffer[client->tail].value = 0; client->packet_head = client->tail; + if (client->use_wake_lock) + wake_unlock(&client->wake_lock); } if (event->type == EV_SYN && event->code == SYN_REPORT) { client->packet_head = client->head; + if (client->use_wake_lock) + wake_lock(&client->wake_lock); kill_fasync(&client->fasync, SIGIO, POLL_IN); } } @@ -291,6 +299,8 @@ static int evdev_release(struct inode *inode, struct file *file) mutex_unlock(&evdev->mutex); evdev_detach_client(evdev, client); + if (client->use_wake_lock) + wake_lock_destroy(&client->wake_lock); if (is_vmalloc_addr(client)) vfree(client); @@ -328,6 +338,8 @@ static int evdev_open(struct inode *inode, struct file *file) client->bufsize = bufsize; spin_lock_init(&client->buffer_lock); + snprintf(client->name, sizeof(client->name), "%s-%d", + dev_name(&evdev->dev), task_tgid_vnr(current)); client->evdev = evdev; evdev_attach_client(evdev, client); @@ -394,6 +406,9 @@ static int evdev_fetch_next_event(struct evdev_client *client, if (have_event) { *event = client->buffer[client->tail++]; client->tail &= client->bufsize - 1; + if (client->use_wake_lock && + client->packet_head == client->tail) + wake_unlock(&client->wake_lock); } spin_unlock_irq(&client->buffer_lock); @@ -682,6 +697,35 @@ static int evdev_handle_mt_request(struct input_dev *dev, return 0; } +static int evdev_enable_suspend_block(struct evdev *evdev, + struct evdev_client *client) +{ + if (client->use_wake_lock) + return 0; + + spin_lock_irq(&client->buffer_lock); + wake_lock_init(&client->wake_lock, WAKE_LOCK_SUSPEND, client->name); + client->use_wake_lock = true; + if (client->packet_head != client->tail) + wake_lock(&client->wake_lock); + spin_unlock_irq(&client->buffer_lock); + return 0; +} + +static int evdev_disable_suspend_block(struct evdev *evdev, + struct evdev_client *client) +{ + if (!client->use_wake_lock) + return 0; + + spin_lock_irq(&client->buffer_lock); + client->use_wake_lock = false; + wake_lock_destroy(&client->wake_lock); + spin_unlock_irq(&client->buffer_lock); + + return 0; +} + static long evdev_do_ioctl(struct file *file, unsigned int cmd, void __user *p, int compat_mode) { @@ -763,6 +807,15 @@ static long evdev_do_ioctl(struct file *file, unsigned int cmd, case EVIOCSKEYCODE_V2: return evdev_handle_set_keycode_v2(dev, p); + + case EVIOCGSUSPENDBLOCK: + return put_user(client->use_wake_lock, ip); + + case EVIOCSSUSPENDBLOCK: + if (p) + return evdev_enable_suspend_block(evdev, client); + else + return evdev_disable_suspend_block(evdev, client); } size = _IOC_SIZE(cmd); 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..7fbf7247e65f --- /dev/null +++ b/drivers/input/keyreset.c @@ -0,0 +1,145 @@ +/* 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) +{ + sys_sync(); + kernel_restart(NULL); +} + +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 bb698e1f9e42..4abf046e30b1 100644 --- a/drivers/input/misc/Kconfig +++ b/drivers/input/misc/Kconfig @@ -299,6 +299,17 @@ 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 @@ -434,6 +445,11 @@ 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 d7fc17f11d77..6b0e8a677725 100644 --- a/drivers/input/misc/Makefile +++ b/drivers/input/misc/Makefile @@ -28,9 +28,11 @@ obj-$(CONFIG_INPUT_DA9055_ONKEY) += da9055_onkey.o obj-$(CONFIG_INPUT_DM355EVM) += dm355evm_keys.o obj-$(CONFIG_INPUT_GP2A) += gp2ap002a00f.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..a5ea27ad0e16 --- /dev/null +++ b/drivers/input/misc/keychord.c @@ -0,0 +1,391 @@ +/* + * 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]; +}; + +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; +} + +/* + * 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, *next, *end; + int ret, i, key; + unsigned long flags; + + 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; + } + + /* 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; + end = (struct input_keychord *)((char *)keychord + count); + + while (keychord < end) { + next = NEXT_KEYCHORD(keychord); + if (keychord->count <= 0 || next > end) { + pr_err("keychord: invalid keycode count %d\n", + keychord->count); + goto err_unlock_return; + } + 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; + } + + kdev->keychords = keychords; + spin_unlock_irqrestore(&kdev->lock, flags); + + ret = input_register_handler(&kdev->input_handler); + if (ret) { + kfree(keychords); + kdev->keychords = 0; + return ret; + } + kdev->registered = 1; + + return count; + +err_unlock_return: + spin_unlock_irqrestore(&kdev->lock, flags); + kfree(keychords); + 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); + + 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); + + 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/dm-verity.c b/drivers/md/dm-verity.c index 0d2e812d424b..fe07d915e20b 100644 --- a/drivers/md/dm-verity.c +++ b/drivers/md/dm-verity.c @@ -18,20 +18,36 @@ #include <linux/module.h> #include <linux/device-mapper.h> +#include <linux/reboot.h> #include <crypto/hash.h> #define DM_MSG_PREFIX "verity" +#define DM_VERITY_ENV_LENGTH 42 +#define DM_VERITY_ENV_VAR_NAME "VERITY_ERR_BLOCK_NR" + #define DM_VERITY_IO_VEC_INLINE 16 #define DM_VERITY_MEMPOOL_SIZE 4 #define DM_VERITY_DEFAULT_PREFETCH_SIZE 262144 #define DM_VERITY_MAX_LEVELS 63 +#define DM_VERITY_MAX_CORRUPTED_ERRS 100 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 = 0, + DM_VERITY_MODE_LOGGING = 1, + DM_VERITY_MODE_RESTART = 2 +}; + +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; @@ -54,6 +70,8 @@ struct dm_verity { 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 */ mempool_t *vec_mempool; /* mempool of bio vector */ @@ -180,6 +198,54 @@ static void verity_hash_at_level(struct dm_verity *v, sector_t block, int level, } /* + * Handle verification errors. + */ +static int verity_handle_err(struct dm_verity *v, enum verity_block_type type, + unsigned long long block) +{ + char verity_env[DM_VERITY_ENV_LENGTH]; + char *envp[] = { verity_env, NULL }; + const char *type_str = ""; + struct mapped_device *md = dm_table_get_md(v->ti->table); + + if (v->corrupted_errs >= DM_VERITY_MAX_CORRUPTED_ERRS) + goto out; + + ++v->corrupted_errs; + + switch (type) { + case DM_VERITY_BLOCK_TYPE_DATA: + type_str = "data"; + break; + case DM_VERITY_BLOCK_TYPE_METADATA: + type_str = "metadata"; + break; + default: + BUG(); + } + + DMERR_LIMIT("%s: %s block %llu is corrupted", v->data_dev->name, + type_str, block); + + if (v->corrupted_errs == DM_VERITY_MAX_CORRUPTED_ERRS) + DMERR("%s: reached maximum errors", v->data_dev->name); + + snprintf(verity_env, DM_VERITY_ENV_LENGTH, "%s=%d,%llu", + DM_VERITY_ENV_VAR_NAME, type, block); + + kobject_uevent_env(&disk_to_dev(dm_disk(md))->kobj, KOBJ_CHANGE, envp); + +out: + if (v->mode == DM_VERITY_MODE_LOGGING) + return 0; + + if (v->mode == DM_VERITY_MODE_RESTART) + kernel_restart("dm-verity device corrupted"); + + return 1; +} + +/* * Verify hash of a metadata block pertaining to the specified data block * ("block" argument) at a specified level ("level" argument). * @@ -256,11 +322,13 @@ static int verity_verify_level(struct dm_verity_io *io, sector_t block, goto release_ret_r; } if (unlikely(memcmp(result, io_want_digest(v, io), v->digest_size))) { - DMERR_LIMIT("metadata block %llu is corrupted", - (unsigned long long)hash_block); v->hash_failed = 1; - r = -EIO; - goto release_ret_r; + + if (verity_handle_err(v, DM_VERITY_BLOCK_TYPE_METADATA, + hash_block)) { + r = -EIO; + goto release_ret_r; + } } else aux->hash_verified = 1; } @@ -377,10 +445,11 @@ test_block_hash: return r; } if (unlikely(memcmp(result, io_want_digest(v, io), v->digest_size))) { - DMERR_LIMIT("data block %llu is corrupted", - (unsigned long long)(io->block + b)); v->hash_failed = 1; - return -EIO; + + if (verity_handle_err(v, DM_VERITY_BLOCK_TYPE_DATA, + io->block + b)) + return -EIO; } } BUG_ON(vector != io->io_vec_size); @@ -689,8 +758,8 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv) goto bad; } - if (argc != 10) { - ti->error = "Invalid argument count: exactly 10 arguments required"; + if (argc < 10 || argc > 11) { + ti->error = "Invalid argument count: 10-11 arguments required"; r = -EINVAL; goto bad; } @@ -811,6 +880,17 @@ static int verity_ctr(struct dm_target *ti, unsigned argc, char **argv) } } + if (argc > 10) { + if (sscanf(argv[10], "%d%c", &num, &dummy) != 1 || + num < DM_VERITY_MODE_EIO || + num > DM_VERITY_MODE_RESTART) { + ti->error = "Invalid mode"; + r = -EINVAL; + goto bad; + } + v->mode = num; + } + v->hash_per_block_bits = fls((1 << v->hash_dev_block_bits) / v->digest_size) - 1; diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig index c002d8660e30..181fc59fec04 100644 --- a/drivers/misc/Kconfig +++ b/drivers/misc/Kconfig @@ -424,6 +424,10 @@ config TI_DAC7512 This driver can also be built as a module. If so, the module will be called ti_dac7512. +config UID_STAT + bool "UID based statistics tracking exported to /proc/uid_stat" + default n + config VMWARE_BALLOON tristate "VMware Balloon Driver" depends on X86 && HYPERVISOR_GUEST @@ -527,6 +531,12 @@ config SRAM the genalloc API. It is supposed to be used for small on-chip SRAM areas found on many SoCs. +config UID_CPUTIME + tristate "Per-UID cpu time statistics" + depends on PROFILING + help + Per UID based cpu time statistics exported to /proc/uid_cputime + 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 c235d5b68311..89435c9a182c 100644 --- a/drivers/misc/Makefile +++ b/drivers/misc/Makefile @@ -36,6 +36,7 @@ obj-$(CONFIG_SENSORS_TSL2550) += tsl2550.o obj-$(CONFIG_EP93XX_PWM) += ep93xx_pwm.o obj-$(CONFIG_DS1682) += ds1682.o obj-$(CONFIG_TI_DAC7512) += ti_dac7512.o +obj-$(CONFIG_UID_STAT) += uid_stat.o obj-$(CONFIG_C2PORT) += c2port/ obj-$(CONFIG_HMC6352) += hmc6352.o obj-y += eeprom/ @@ -53,3 +54,4 @@ obj-$(CONFIG_INTEL_MEI) += mei/ obj-$(CONFIG_VMWARE_VMCI) += vmw_vmci/ obj-$(CONFIG_LATTICE_ECP3_CONFIG) += lattice-ecp3-config.o obj-$(CONFIG_SRAM) += sram.o +obj-$(CONFIG_UID_CPUTIME) += uid_cputime.o diff --git a/drivers/misc/uid_cputime.c b/drivers/misc/uid_cputime.c new file mode 100644 index 000000000000..acd7046ce497 --- /dev/null +++ b/drivers/misc/uid_cputime.c @@ -0,0 +1,237 @@ +/* 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/proc_fs.h> +#include <linux/profile.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_MUTEX(uid_lock); +static struct proc_dir_entry *parent; + +struct uid_entry { + uid_t uid; + cputime_t utime; + cputime_t stime; + cputime_t active_utime; + cputime_t active_stime; + struct hlist_node hash; +}; + +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; + + hash_add(hash_table, &uid_entry->hash, uid); + + return uid_entry; +} + +static int uid_stat_show(struct seq_file *m, void *v) +{ + struct uid_entry *uid_entry; + struct task_struct *task; + cputime_t utime; + cputime_t stime; + unsigned long bkt; + + 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); + for_each_process(task) { + uid_entry = find_or_register_uid(from_kuid_munged( + current_user_ns(), task_uid(task))); + if (!uid_entry) { + read_unlock(&tasklist_lock); + mutex_unlock(&uid_lock); + pr_err("%s: failed to find the uid_entry for uid %d\n", + __func__, from_kuid_munged(current_user_ns(), + task_uid(task))); + return -ENOMEM; + } + task_cputime_adjusted(task, &utime, &stime); + uid_entry->active_utime += utime; + uid_entry->active_stime += stime; + } + 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: %u %u\n", uid_entry->uid, + cputime_to_usecs(total_utime), + cputime_to_usecs(total_stime)); + } + + mutex_unlock(&uid_lock); + return 0; +} + +static int uid_stat_open(struct inode *inode, struct file *file) +{ + return single_open(file, uid_stat_show, PDE_DATA(inode)); +} + +static const struct file_operations uid_stat_fops = { + .open = uid_stat_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; + } + + mutex_lock(&uid_lock); + + for (; uid_start <= uid_end; uid_start++) { + hash_for_each_possible_safe(hash_table, uid_entry, tmp, + hash, uid_start) { + hash_del(&uid_entry->hash); + kfree(uid_entry); + } + } + + 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 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; + + 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; + +exit: + mutex_unlock(&uid_lock); + return NOTIFY_OK; +} + +static struct notifier_block process_notifier_block = { + .notifier_call = process_notifier, +}; + +static int __init proc_uid_cputime_init(void) +{ + hash_init(hash_table); + + parent = proc_mkdir("uid_cputime", NULL); + if (!parent) { + pr_err("%s: failed to create proc entry\n", __func__); + return -ENOMEM; + } + + proc_create_data("remove_uid_range", S_IWUGO, parent, &uid_remove_fops, + NULL); + + proc_create_data("show_uid_stat", S_IWUGO, parent, &uid_stat_fops, + NULL); + + profile_event_register(PROFILE_TASK_EXIT, &process_notifier_block); + + return 0; +} + +early_initcall(proc_uid_cputime_init); diff --git a/drivers/misc/uid_stat.c b/drivers/misc/uid_stat.c new file mode 100644 index 000000000000..4766c1f83b94 --- /dev/null +++ b/drivers/misc/uid_stat.c @@ -0,0 +1,152 @@ +/* drivers/misc/uid_stat.c + * + * Copyright (C) 2008 - 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. + * + */ + +#include <asm/atomic.h> + +#include <linux/err.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/list.h> +#include <linux/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/slab.h> +#include <linux/spinlock.h> +#include <linux/stat.h> +#include <linux/uid_stat.h> +#include <net/activity_stats.h> + +static DEFINE_SPINLOCK(uid_lock); +static LIST_HEAD(uid_list); +static struct proc_dir_entry *parent; + +struct uid_stat { + struct list_head link; + uid_t uid; + atomic_t tcp_rcv; + atomic_t tcp_snd; +}; + +static struct uid_stat *find_uid_stat(uid_t uid) { + struct uid_stat *entry; + + list_for_each_entry(entry, &uid_list, link) { + if (entry->uid == uid) { + return entry; + } + } + return NULL; +} + +static int uid_stat_atomic_int_show(struct seq_file *m, void *v) +{ + unsigned int bytes; + atomic_t *counter = m->private; + + bytes = (unsigned int) (atomic_read(counter) + INT_MIN); + return seq_printf(m, "%u\n", bytes); +} + +static int uid_stat_read_atomic_int_open(struct inode *inode, struct file *file) +{ + return single_open(file, uid_stat_atomic_int_show, PDE_DATA(inode)); +} + +static const struct file_operations uid_stat_read_atomic_int_fops = { + .open = uid_stat_read_atomic_int_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +/* Create a new entry for tracking the specified uid. */ +static struct uid_stat *create_stat(uid_t uid) { + struct uid_stat *new_uid; + /* Create the uid stat struct and append it to the list. */ + new_uid = kmalloc(sizeof(struct uid_stat), GFP_ATOMIC); + if (!new_uid) + return NULL; + + new_uid->uid = uid; + /* Counters start at INT_MIN, so we can track 4GB of network traffic. */ + atomic_set(&new_uid->tcp_rcv, INT_MIN); + atomic_set(&new_uid->tcp_snd, INT_MIN); + + list_add_tail(&new_uid->link, &uid_list); + return new_uid; +} + +static void create_stat_proc(struct uid_stat *new_uid) +{ + char uid_s[32]; + struct proc_dir_entry *entry; + sprintf(uid_s, "%d", new_uid->uid); + entry = proc_mkdir(uid_s, parent); + + /* Keep reference to uid_stat so we know what uid to read stats from. */ + proc_create_data("tcp_snd", S_IRUGO, entry, + &uid_stat_read_atomic_int_fops, &new_uid->tcp_snd); + + proc_create_data("tcp_rcv", S_IRUGO, entry, + &uid_stat_read_atomic_int_fops, &new_uid->tcp_rcv); +} + +static struct uid_stat *find_or_create_uid_stat(uid_t uid) +{ + struct uid_stat *entry; + unsigned long flags; + spin_lock_irqsave(&uid_lock, flags); + entry = find_uid_stat(uid); + if (entry) { + spin_unlock_irqrestore(&uid_lock, flags); + return entry; + } + entry = create_stat(uid); + spin_unlock_irqrestore(&uid_lock, flags); + if (entry) + create_stat_proc(entry); + return entry; +} + +int uid_stat_tcp_snd(uid_t uid, int size) { + struct uid_stat *entry; + activity_stats_update(); + entry = find_or_create_uid_stat(uid); + if (!entry) + return -1; + atomic_add(size, &entry->tcp_snd); + return 0; +} + +int uid_stat_tcp_rcv(uid_t uid, int size) { + struct uid_stat *entry; + activity_stats_update(); + entry = find_or_create_uid_stat(uid); + if (!entry) + return -1; + atomic_add(size, &entry->tcp_rcv); + return 0; +} + +static int __init uid_stat_init(void) +{ + parent = proc_mkdir("uid_stat", NULL); + if (!parent) { + pr_err("uid_stat: failed to create proc entry\n"); + return -1; + } + return 0; +} + +__initcall(uid_stat_init); diff --git a/drivers/mmc/card/Kconfig b/drivers/mmc/card/Kconfig index 5562308699bc..79d82124413f 100644 --- a/drivers/mmc/card/Kconfig +++ b/drivers/mmc/card/Kconfig @@ -50,6 +50,15 @@ config MMC_BLOCK_BOUNCE If unsure, say Y here. +config MMC_BLOCK_DEFERRED_RESUME + bool "Deferr MMC layer resume until I/O is requested" + depends on MMC_BLOCK + default n + help + Say Y here to enable deferred MMC resume until I/O + is requested. This will reduce overall resume latency and + save power when theres an SD card inserted but not being used. + config SDIO_UART tristate "SDIO UART/GPS class support" depends on TTY diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c index 56998eca1a8d..d813cb9a3666 100644 --- a/drivers/mmc/card/block.c +++ b/drivers/mmc/card/block.c @@ -35,6 +35,9 @@ #include <linux/capability.h> #include <linux/compat.h> +#define CREATE_TRACE_POINTS +#include <trace/events/mmc.h> + #include <linux/mmc/ioctl.h> #include <linux/mmc/card.h> #include <linux/mmc/host.h> @@ -163,11 +166,7 @@ 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; } @@ -728,18 +727,22 @@ 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: @@ -1031,9 +1034,12 @@ retry: goto out; } - if (mmc_can_sanitize(card)) + if (mmc_can_sanitize(card)) { + 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, 0); + trace_mmc_blk_erase_end(EXT_CSD_SANITIZE_START, 0, 0); + } out_retry: if (err && !mmc_blk_reset(md, card->host, type)) goto retry; @@ -1945,6 +1951,11 @@ static int mmc_blk_issue_rq(struct mmc_queue *mq, struct request *req) unsigned long flags; unsigned int cmd_flags = req ? req->cmd_flags : 0; +#ifdef CONFIG_MMC_BLOCK_DEFERRED_RESUME + if (mmc_bus_needs_resume(card->host)) + mmc_resume_bus(card->host); +#endif + if (req && !mq->mqrq_prev->req) /* claim host only for the first request */ mmc_claim_host(card->host); @@ -2067,6 +2078,7 @@ 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) md->disk->flags |= GENHD_FL_NO_PART_SCAN; @@ -2381,6 +2393,9 @@ static int mmc_blk_probe(struct mmc_card *card) mmc_set_drvdata(card, md); mmc_fixup_device(card, blk_fixups); +#ifdef CONFIG_MMC_BLOCK_DEFERRED_RESUME + mmc_set_bus_resume_policy(card->host, 1); +#endif if (mmc_add_disk(md)) goto out; @@ -2406,6 +2421,9 @@ static void mmc_blk_remove(struct mmc_card *card) mmc_release_host(card->host); mmc_blk_remove_req(md); mmc_set_drvdata(card, NULL); +#ifdef CONFIG_MMC_BLOCK_DEFERRED_RESUME + mmc_set_bus_resume_policy(card->host, 0); +#endif } #ifdef CONFIG_PM diff --git a/drivers/mmc/core/Kconfig b/drivers/mmc/core/Kconfig index 269d072ef55e..ae10a372af0f 100644 --- a/drivers/mmc/core/Kconfig +++ b/drivers/mmc/core/Kconfig @@ -26,3 +26,18 @@ config MMC_CLKGATE support handling this in order for it to be of any use. If unsure, say N. + +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 c40396f23202..6a83f4ccc108 100644 --- a/drivers/mmc/core/core.c +++ b/drivers/mmc/core/core.c @@ -27,6 +27,9 @@ #include <linux/fault-inject.h> #include <linux/random.h> #include <linux/slab.h> +#include <linux/wakelock.h> + +#include <trace/events/mmc.h> #include <linux/mmc/card.h> #include <linux/mmc/host.h> @@ -172,6 +175,7 @@ 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); + trace_mmc_blk_rw_end(cmd->opcode, cmd->arg, mrq->data); } if (mrq->stop) { @@ -536,8 +540,12 @@ struct mmc_async_req *mmc_start_req(struct mmc_host *host, mmc_start_bkops(host->card, true); } - if (!err && areq) + if (!err && areq) { + 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); @@ -1591,6 +1599,36 @@ static inline void mmc_bus_put(struct mmc_host *host) spin_unlock_irqrestore(&host->lock, flags); } +int mmc_resume_bus(struct mmc_host *host) +{ + unsigned long flags; + + if (!mmc_bus_needs_resume(host)) + return -EINVAL; + + printk("%s: Starting deferred resume\n", mmc_hostname(host)); + spin_lock_irqsave(&host->lock, flags); + host->bus_resume_flags &= ~MMC_BUSRESUME_NEEDS_RESUME; + host->rescan_disable = 0; + spin_unlock_irqrestore(&host->lock, flags); + + mmc_bus_get(host); + if (host->bus_ops && !host->bus_dead) { + mmc_power_up(host); + BUG_ON(!host->bus_ops->resume); + host->bus_ops->resume(host); + } + + if (host->bus_ops->detect && !host->bus_dead) + host->bus_ops->detect(host); + + mmc_bus_put(host); + printk("%s: Deferred resume completed\n", mmc_hostname(host)); + return 0; +} + +EXPORT_SYMBOL(mmc_resume_bus); + /* * Assign a mmc bus handler to a host. Only one bus handler may control a * host at any given time. @@ -1656,6 +1694,8 @@ void mmc_detect_change(struct mmc_host *host, unsigned long delay) spin_unlock_irqrestore(&host->lock, flags); #endif host->detect_change = 1; + + wake_lock(&host->detect_wake_lock); mmc_schedule_delayed_work(&host->detect, delay); } @@ -1815,8 +1855,13 @@ 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); + /* * qty is used to calculate the erase timeout which depends on how many * erase groups (or allocation units in SD terminology) are affected. @@ -1920,6 +1965,8 @@ static int mmc_do_erase(struct mmc_card *card, unsigned int from, } while (!(cmd.resp[0] & R1_READY_FOR_DATA) || (R1_CURRENT_STATE(cmd.resp[0]) == R1_STATE_PRG)); out: + + trace_mmc_blk_erase_end(arg, fr, nr); return err; } @@ -2351,6 +2398,7 @@ void mmc_rescan(struct work_struct *work) struct mmc_host *host = container_of(work, struct mmc_host, detect.work); int i; + bool extend_wakelock = false; if (host->rescan_disable) return; @@ -2372,6 +2420,12 @@ void mmc_rescan(struct work_struct *work) host->detect_change = 0; + /* If the card was removed the bus will be marked + * as dead - extend the wakelock so userspace + * can respond */ + if (host->bus_dead) + extend_wakelock = 1; + /* * Let mmc_bus_put() free the bus/bus_ops if we've found that * the card is no longer present. @@ -2400,16 +2454,24 @@ void mmc_rescan(struct work_struct *work) mmc_claim_host(host); for (i = 0; i < ARRAY_SIZE(freqs); i++) { - if (!mmc_rescan_try_freq(host, max(freqs[i], host->f_min))) + if (!mmc_rescan_try_freq(host, max(freqs[i], host->f_min))) { + extend_wakelock = true; break; + } if (freqs[i] <= host->f_min) break; } mmc_release_host(host); out: - if (host->caps & MMC_CAP_NEEDS_POLL) + if (extend_wakelock) + wake_lock_timeout(&host->detect_wake_lock, HZ / 2); + else + wake_unlock(&host->detect_wake_lock); + if (host->caps & MMC_CAP_NEEDS_POLL) { + wake_lock(&host->detect_wake_lock); mmc_schedule_delayed_work(&host->detect, HZ); + } } void mmc_start_host(struct mmc_host *host) @@ -2433,7 +2495,8 @@ void mmc_stop_host(struct mmc_host *host) #endif host->rescan_disable = 1; - cancel_delayed_work_sync(&host->detect); + if (cancel_delayed_work_sync(&host->detect)) + wake_unlock(&host->detect_wake_lock); mmc_flush_scheduled_work(); /* clear pm flags now and let card drivers set them as needed */ @@ -2628,7 +2691,11 @@ int mmc_suspend_host(struct mmc_host *host) { int err = 0; - cancel_delayed_work(&host->detect); + if (mmc_bus_needs_resume(host)) + return 0; + + if (cancel_delayed_work(&host->detect)) + wake_unlock(&host->detect_wake_lock); mmc_flush_scheduled_work(); mmc_bus_get(host); @@ -2679,6 +2746,12 @@ int mmc_resume_host(struct mmc_host *host) int err = 0; mmc_bus_get(host); + if (mmc_bus_manual_resume(host)) { + host->bus_resume_flags |= MMC_BUSRESUME_NEEDS_RESUME; + mmc_bus_put(host); + return 0; + } + if (host->bus_ops && !host->bus_dead) { if (!mmc_card_keep_power(host)) { mmc_power_up(host); @@ -2739,9 +2812,14 @@ int mmc_pm_notify(struct notifier_block *notify_block, } spin_lock_irqsave(&host->lock, flags); + if (mmc_bus_needs_resume(host)) { + spin_unlock_irqrestore(&host->lock, flags); + break; + } host->rescan_disable = 1; spin_unlock_irqrestore(&host->lock, flags); - cancel_delayed_work_sync(&host->detect); + if (cancel_delayed_work_sync(&host->detect)) + wake_unlock(&host->detect_wake_lock); if (!host->bus_ops || host->bus_ops->suspend) break; @@ -2762,6 +2840,10 @@ int mmc_pm_notify(struct notifier_block *notify_block, case PM_POST_RESTORE: spin_lock_irqsave(&host->lock, flags); + if (mmc_bus_manual_resume(host)) { + spin_unlock_irqrestore(&host->lock, flags); + break; + } host->rescan_disable = 0; spin_unlock_irqrestore(&host->lock, flags); mmc_detect_change(host, 0); @@ -2789,6 +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; diff --git a/drivers/mmc/core/host.c b/drivers/mmc/core/host.c index 2a3593d9f87d..56dadd667ec1 100644 --- a/drivers/mmc/core/host.c +++ b/drivers/mmc/core/host.c @@ -459,6 +459,8 @@ struct mmc_host *mmc_alloc_host(int extra, struct device *dev) spin_lock_init(&host->lock); init_waitqueue_head(&host->wq); + wake_lock_init(&host->detect_wake_lock, WAKE_LOCK_SUSPEND, + kasprintf(GFP_KERNEL, "%s_detect", mmc_hostname(host))); INIT_DELAYED_WORK(&host->detect, mmc_rescan); #ifdef CONFIG_PM host->pm_notify.notifier_call = mmc_pm_notify; @@ -511,7 +513,8 @@ int mmc_add_host(struct mmc_host *host) mmc_host_clk_sysfs_init(host); 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; } @@ -528,7 +531,9 @@ 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 @@ -555,6 +560,7 @@ void mmc_free_host(struct mmc_host *host) spin_lock(&mmc_host_lock); idr_remove(&mmc_host_idr, host->index); spin_unlock(&mmc_host_lock); + wake_lock_destroy(&host->detect_wake_lock); put_device(&host->class_dev); } diff --git a/drivers/mmc/core/sd.c b/drivers/mmc/core/sd.c index 9e645e19cec6..f008318c5c4d 100644 --- a/drivers/mmc/core/sd.c +++ b/drivers/mmc/core/sd.c @@ -805,6 +805,9 @@ 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) { /* @@ -831,7 +834,26 @@ 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; } @@ -1032,7 +1054,10 @@ 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); @@ -1042,7 +1067,23 @@ 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_release_host(host); @@ -1084,12 +1125,31 @@ static int mmc_sd_suspend(struct mmc_host *host) static int mmc_sd_resume(struct mmc_host *host) { int err; +#ifdef CONFIG_MMC_PARANOID_SD_INIT + int retries; +#endif BUG_ON(!host); BUG_ON(!host->card); mmc_claim_host(host); +#ifdef CONFIG_MMC_PARANOID_SD_INIT + retries = 5; + while (retries) { + err = mmc_sd_init_card(host, host->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->ocr, host->card); +#endif mmc_release_host(host); return err; @@ -1143,6 +1203,9 @@ int mmc_attach_sd(struct mmc_host *host) { int err; u32 ocr; +#ifdef CONFIG_MMC_PARANOID_SD_INIT + int retries; +#endif BUG_ON(!host); WARN_ON(!host->claimed); @@ -1198,9 +1261,27 @@ 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, host->ocr, 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, host->ocr, 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 6889a821c1da..46e68f125ff2 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> @@ -28,6 +29,10 @@ #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; @@ -728,19 +733,35 @@ 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 && @@ -1147,14 +1168,36 @@ 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) */ @@ -1202,3 +1245,39 @@ err: return err; } +int sdio_reset_comm(struct mmc_card *card) +{ + struct mmc_host *host = card->host; + u32 ocr; + int err; + + printk("%s():\n", __func__); + mmc_claim_host(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; + + host->ocr = mmc_select_voltage(host, ocr); + if (!host->ocr) { + err = -EINVAL; + goto err; + } + + err = mmc_sdio_init_card(host, host->ocr, 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 546c67c2bbbf..c012cf59b7d6 100644 --- a/drivers/mmc/core/sdio_bus.c +++ b/drivers/mmc/core/sdio_bus.c @@ -25,6 +25,10 @@ #include "sdio_cis.h" #include "sdio_bus.h" +#ifdef CONFIG_MMC_EMBEDDED_SDIO +#include <linux/mmc/host.h> +#endif + /* show configuration fields */ #define sdio_config_attr(field, format_string) \ static ssize_t \ @@ -270,7 +274,14 @@ 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); diff --git a/drivers/mmc/core/sdio_io.c b/drivers/mmc/core/sdio_io.c index 78cb4d5d9d58..8fdeb07723a6 100644..100755 --- 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 50543f166215..f2ab08c2c5fb 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 @@ -133,9 +140,6 @@ config BCH_CONST_T default 8 if MTD_NAND_OMAP_BCH8 endif -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..a5d3d634fd9a --- /dev/null +++ b/drivers/net/ppp/pppolac.c @@ -0,0 +1,449 @@ +/* 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_iov = (struct iovec *)&iov, + .msg_iovlen = 1, + .msg_flags = MSG_NOSIGNAL | MSG_DONTWAIT, + }; + sk_udp->sk_prot->sendmsg(NULL, 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) +{ + struct sock *sk; + + sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pppolac_proto); + 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..6016d29c0660 --- /dev/null +++ b/drivers/net/ppp/pppopns.c @@ -0,0 +1,428 @@ +/* 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, int length) +{ + 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_iov = (struct iovec *)&iov, + .msg_iovlen = 1, + .msg_flags = MSG_NOSIGNAL | MSG_DONTWAIT, + }; + sk_raw->sk_prot->sendmsg(NULL, 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) +{ + struct sock *sk; + + sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pppopns_proto); + 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 582497103fe8..aeff706f1634 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c @@ -1898,6 +1898,12 @@ static long __tun_chr_ioctl(struct file *file, unsigned int cmd, int vnet_hdr_sz; 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/Kconfig b/drivers/net/wireless/Kconfig index f8f0156dff4e..5b0a49cf1dea 100644 --- a/drivers/net/wireless/Kconfig +++ b/drivers/net/wireless/Kconfig @@ -264,6 +264,11 @@ config MWL8K To compile this driver as a module, choose M here: the module will be called mwl8k. If unsure, say N. +config WIFI_CONTROL_FUNC + bool "Enable WiFi control function abstraction" + help + Enables Power/Reset/Carddetect function abstraction + source "drivers/net/wireless/ath/Kconfig" source "drivers/net/wireless/b43/Kconfig" source "drivers/net/wireless/b43legacy/Kconfig" diff --git a/drivers/of/fdt.c b/drivers/of/fdt.c index 42fcf572ebfe..137c31f5bb05 100644 --- a/drivers/of/fdt.c +++ b/drivers/of/fdt.c @@ -703,36 +703,66 @@ 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; + 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/power/power_supply_core.c b/drivers/power/power_supply_core.c index 1c517c34e4be..082d3c2714e9 100644 --- a/drivers/power/power_supply_core.c +++ b/drivers/power/power_supply_core.c @@ -67,23 +67,40 @@ static int __power_supply_changed_work(struct device *dev, void *data) static void power_supply_changed_work(struct work_struct *work) { + unsigned long flags; struct power_supply *psy = container_of(work, struct power_supply, changed_work); dev_dbg(psy->dev, "%s\n", __func__); - class_for_each_device(power_supply_class, NULL, psy, - __power_supply_changed_work); + spin_lock_irqsave(&psy->changed_lock, flags); + if (psy->changed) { + psy->changed = false; + spin_unlock_irqrestore(&psy->changed_lock, flags); - power_supply_update_leds(psy); + class_for_each_device(power_supply_class, NULL, psy, + __power_supply_changed_work); - kobject_uevent(&psy->dev->kobj, KOBJ_CHANGE); + power_supply_update_leds(psy); + + kobject_uevent(&psy->dev->kobj, KOBJ_CHANGE); + spin_lock_irqsave(&psy->changed_lock, flags); + } + if (!psy->changed) + pm_relax(psy->dev); + spin_unlock_irqrestore(&psy->changed_lock, flags); } void power_supply_changed(struct power_supply *psy) { + unsigned long flags; + dev_dbg(psy->dev, "%s\n", __func__); + spin_lock_irqsave(&psy->changed_lock, flags); + psy->changed = true; + pm_stay_awake(psy->dev); + spin_unlock_irqrestore(&psy->changed_lock, flags); schedule_work(&psy->changed_work); } EXPORT_SYMBOL_GPL(power_supply_changed); @@ -504,6 +521,11 @@ int power_supply_register(struct device *parent, struct power_supply *psy) if (rc) goto device_add_failed; + spin_lock_init(&psy->changed_lock); + rc = device_init_wakeup(dev, true); + if (rc) + goto wakeup_init_failed; + rc = psy_register_thermal(psy); if (rc) goto register_thermal_failed; @@ -525,6 +547,7 @@ create_triggers_failed: register_cooler_failed: psy_unregister_thermal(psy); register_thermal_failed: +wakeup_init_failed: device_del(dev); kobject_set_name_failed: device_add_failed: diff --git a/drivers/power/power_supply_sysfs.c b/drivers/power/power_supply_sysfs.c index 29178f78d73c..1f7d79b03725 100644 --- a/drivers/power/power_supply_sysfs.c +++ b/drivers/power/power_supply_sysfs.c @@ -105,7 +105,10 @@ 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, @@ -189,6 +192,12 @@ static struct device_attribute power_supply_attrs[] = { POWER_SUPPLY_ATTR(time_to_full_avg), POWER_SUPPLY_ATTR(type), POWER_SUPPLY_ATTR(scope), + /* 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/staging/android/Kconfig b/drivers/staging/android/Kconfig index c0c95be0f969..b3c054e4fd58 100644 --- a/drivers/staging/android/Kconfig +++ b/drivers/staging/android/Kconfig @@ -8,17 +8,6 @@ config ANDROID if ANDROID -config ANDROID_BINDER_IPC - bool "Android Binder IPC Driver" - default n - ---help--- - Binder is used in Android for both communication between processes, - and remote method invocation. - - This means one Android process can call a method/routine in another - Android process, using Binder to identify, invoke and pass arguments - between said processes. - config ASHMEM bool "Enable the Anonymous Shared Memory Subsystem" default n @@ -31,23 +20,6 @@ config ASHMEM It is, in theory, a good memory allocator for low-memory devices, because it can discard shared memory units when under memory pressure. -config ANDROID_LOGGER - tristate "Android log driver" - default n - ---help--- - This adds support for system-wide logging using four log buffers. - - These are: - - 1: main - 2: events - 3: radio - 4: system - - Log reading and writing is performed via normal Linux reads and - optimized writes. This optimization avoids logging having too - much overhead in the system. - config ANDROID_TIMED_OUTPUT bool "Timed output class driver" default y @@ -63,14 +35,14 @@ config ANDROID_LOW_MEMORY_KILLER ---help--- Registers processes to be killed when memory is low -config ANDROID_INTF_ALARM_DEV - bool "Android alarm driver" - depends on RTC_CLASS - default n +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--- - Provides non-wakeup and rtc backed wakeup alarms based on rtc or - elapsed realtime, and a non-wakeup alarm on the monotonic clock. - Also exports the alarm interface to user-space. + Detect oom_adj values written to + /sys/module/lowmemorykiller/parameters/adj and convert them + to oom_score_adj values. config SYNC bool "Synchronization framework" @@ -99,6 +71,10 @@ config SW_SYNC_USER *WARNING* improper use of this can result in deadlocking kernel drivers from userspace. +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 c136299e05af..30ba2f0cb084 100644 --- a/drivers/staging/android/Makefile +++ b/drivers/staging/android/Makefile @@ -1,11 +1,11 @@ ccflags-y += -I$(src) # needed for trace events -obj-$(CONFIG_ANDROID_BINDER_IPC) += binder.o +obj-y += ion/ +obj-$(CONFIG_FIQ_DEBUGGER) += fiq_debugger/ + obj-$(CONFIG_ASHMEM) += ashmem.o -obj-$(CONFIG_ANDROID_LOGGER) += logger.o obj-$(CONFIG_ANDROID_TIMED_OUTPUT) += timed_output.o obj-$(CONFIG_ANDROID_TIMED_GPIO) += timed_gpio.o obj-$(CONFIG_ANDROID_LOW_MEMORY_KILLER) += lowmemorykiller.o -obj-$(CONFIG_ANDROID_INTF_ALARM_DEV) += alarm-dev.o obj-$(CONFIG_SYNC) += sync.o obj-$(CONFIG_SW_SYNC) += sw_sync.o diff --git a/drivers/staging/android/TODO b/drivers/staging/android/TODO deleted file mode 100644 index b15fb0d6b152..000000000000 --- a/drivers/staging/android/TODO +++ /dev/null @@ -1,10 +0,0 @@ -TODO: - - checkpatch.pl cleanups - - sparse fixes - - rename files to be not so "generic" - - make sure things build as modules properly - - add proper arch dependencies as needed - - audit userspace interfaces to make sure they are sane - -Please send patches to Greg Kroah-Hartman <greg@kroah.com> and Cc: -Brian Swetland <swetland@google.com> diff --git a/drivers/staging/android/alarm-dev.c b/drivers/staging/android/alarm-dev.c deleted file mode 100644 index 6dc27dac679d..000000000000 --- a/drivers/staging/android/alarm-dev.c +++ /dev/null @@ -1,443 +0,0 @@ -/* drivers/rtc/alarm-dev.c - * - * Copyright (C) 2007-2009 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. - * - */ - -#include <linux/time.h> -#include <linux/module.h> -#include <linux/device.h> -#include <linux/miscdevice.h> -#include <linux/fs.h> -#include <linux/platform_device.h> -#include <linux/sched.h> -#include <linux/spinlock.h> -#include <linux/uaccess.h> -#include <linux/alarmtimer.h> -#include "android_alarm.h" - -#define ANDROID_ALARM_PRINT_INFO (1U << 0) -#define ANDROID_ALARM_PRINT_IO (1U << 1) -#define ANDROID_ALARM_PRINT_INT (1U << 2) - -static int debug_mask = ANDROID_ALARM_PRINT_INFO; -module_param_named(debug_mask, debug_mask, int, S_IRUGO | S_IWUSR | S_IWGRP); - -#define alarm_dbg(debug_level_mask, fmt, ...) \ -do { \ - if (debug_mask & ANDROID_ALARM_PRINT_##debug_level_mask) \ - pr_info(fmt, ##__VA_ARGS__); \ -} while (0) - -#define ANDROID_ALARM_WAKEUP_MASK ( \ - ANDROID_ALARM_RTC_WAKEUP_MASK | \ - ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP_MASK) - -static int alarm_opened; -static DEFINE_SPINLOCK(alarm_slock); -static struct wakeup_source alarm_wake_lock; -static DECLARE_WAIT_QUEUE_HEAD(alarm_wait_queue); -static uint32_t alarm_pending; -static uint32_t alarm_enabled; -static uint32_t wait_pending; - -struct devalarm { - union { - struct hrtimer hrt; - struct alarm alrm; - } u; - enum android_alarm_type type; -}; - -static struct devalarm alarms[ANDROID_ALARM_TYPE_COUNT]; - - -static int is_wakeup(enum android_alarm_type type) -{ - return (type == ANDROID_ALARM_RTC_WAKEUP || - type == ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP); -} - - -static void devalarm_start(struct devalarm *alrm, ktime_t exp) -{ - if (is_wakeup(alrm->type)) - alarm_start(&alrm->u.alrm, exp); - else - hrtimer_start(&alrm->u.hrt, exp, HRTIMER_MODE_ABS); -} - - -static int devalarm_try_to_cancel(struct devalarm *alrm) -{ - if (is_wakeup(alrm->type)) - return alarm_try_to_cancel(&alrm->u.alrm); - return hrtimer_try_to_cancel(&alrm->u.hrt); -} - -static void devalarm_cancel(struct devalarm *alrm) -{ - if (is_wakeup(alrm->type)) - alarm_cancel(&alrm->u.alrm); - else - hrtimer_cancel(&alrm->u.hrt); -} - -static void alarm_clear(enum android_alarm_type alarm_type) -{ - uint32_t alarm_type_mask = 1U << alarm_type; - unsigned long flags; - - spin_lock_irqsave(&alarm_slock, flags); - alarm_dbg(IO, "alarm %d clear\n", alarm_type); - devalarm_try_to_cancel(&alarms[alarm_type]); - if (alarm_pending) { - alarm_pending &= ~alarm_type_mask; - if (!alarm_pending && !wait_pending) - __pm_relax(&alarm_wake_lock); - } - alarm_enabled &= ~alarm_type_mask; - spin_unlock_irqrestore(&alarm_slock, flags); - -} - -static void alarm_set(enum android_alarm_type alarm_type, - struct timespec *ts) -{ - uint32_t alarm_type_mask = 1U << alarm_type; - unsigned long flags; - - spin_lock_irqsave(&alarm_slock, flags); - alarm_dbg(IO, "alarm %d set %ld.%09ld\n", - alarm_type, ts->tv_sec, ts->tv_nsec); - alarm_enabled |= alarm_type_mask; - devalarm_start(&alarms[alarm_type], timespec_to_ktime(*ts)); - spin_unlock_irqrestore(&alarm_slock, flags); -} - -static int alarm_wait(void) -{ - unsigned long flags; - int rv = 0; - - spin_lock_irqsave(&alarm_slock, flags); - alarm_dbg(IO, "alarm wait\n"); - if (!alarm_pending && wait_pending) { - __pm_relax(&alarm_wake_lock); - wait_pending = 0; - } - spin_unlock_irqrestore(&alarm_slock, flags); - - rv = wait_event_interruptible(alarm_wait_queue, alarm_pending); - if (rv) - return rv; - - spin_lock_irqsave(&alarm_slock, flags); - rv = alarm_pending; - wait_pending = 1; - alarm_pending = 0; - spin_unlock_irqrestore(&alarm_slock, flags); - - return rv; -} - -static int alarm_set_rtc(struct timespec *ts) -{ - struct rtc_time new_rtc_tm; - struct rtc_device *rtc_dev; - unsigned long flags; - int rv = 0; - - rtc_time_to_tm(ts->tv_sec, &new_rtc_tm); - rtc_dev = alarmtimer_get_rtcdev(); - rv = do_settimeofday(ts); - if (rv < 0) - return rv; - if (rtc_dev) - rv = rtc_set_time(rtc_dev, &new_rtc_tm); - - spin_lock_irqsave(&alarm_slock, flags); - alarm_pending |= ANDROID_ALARM_TIME_CHANGE_MASK; - wake_up(&alarm_wait_queue); - spin_unlock_irqrestore(&alarm_slock, flags); - - return rv; -} - -static int alarm_get_time(enum android_alarm_type alarm_type, - struct timespec *ts) -{ - int rv = 0; - - switch (alarm_type) { - case ANDROID_ALARM_RTC_WAKEUP: - case ANDROID_ALARM_RTC: - getnstimeofday(ts); - break; - case ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP: - case ANDROID_ALARM_ELAPSED_REALTIME: - get_monotonic_boottime(ts); - break; - case ANDROID_ALARM_SYSTEMTIME: - ktime_get_ts(ts); - break; - default: - rv = -EINVAL; - } - return rv; -} - -static long alarm_do_ioctl(struct file *file, unsigned int cmd, - struct timespec *ts) -{ - int rv = 0; - unsigned long flags; - enum android_alarm_type alarm_type = ANDROID_ALARM_IOCTL_TO_TYPE(cmd); - - if (alarm_type >= ANDROID_ALARM_TYPE_COUNT) - return -EINVAL; - - if (ANDROID_ALARM_BASE_CMD(cmd) != ANDROID_ALARM_GET_TIME(0)) { - if ((file->f_flags & O_ACCMODE) == O_RDONLY) - return -EPERM; - if (file->private_data == NULL && - cmd != ANDROID_ALARM_SET_RTC) { - spin_lock_irqsave(&alarm_slock, flags); - if (alarm_opened) { - spin_unlock_irqrestore(&alarm_slock, flags); - return -EBUSY; - } - alarm_opened = 1; - file->private_data = (void *)1; - spin_unlock_irqrestore(&alarm_slock, flags); - } - } - - switch (ANDROID_ALARM_BASE_CMD(cmd)) { - case ANDROID_ALARM_CLEAR(0): - alarm_clear(alarm_type); - break; - case ANDROID_ALARM_SET(0): - alarm_set(alarm_type, ts); - break; - case ANDROID_ALARM_SET_AND_WAIT(0): - alarm_set(alarm_type, ts); - /* fall though */ - case ANDROID_ALARM_WAIT: - rv = alarm_wait(); - break; - case ANDROID_ALARM_SET_RTC: - rv = alarm_set_rtc(ts); - break; - case ANDROID_ALARM_GET_TIME(0): - rv = alarm_get_time(alarm_type, ts); - break; - - default: - rv = -EINVAL; - } - return rv; -} - -static long alarm_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -{ - - struct timespec ts; - int rv; - - switch (ANDROID_ALARM_BASE_CMD(cmd)) { - case ANDROID_ALARM_SET_AND_WAIT(0): - case ANDROID_ALARM_SET(0): - case ANDROID_ALARM_SET_RTC: - if (copy_from_user(&ts, (void __user *)arg, sizeof(ts))) - return -EFAULT; - break; - } - - rv = alarm_do_ioctl(file, cmd, &ts); - if (rv) - return rv; - - switch (ANDROID_ALARM_BASE_CMD(cmd)) { - case ANDROID_ALARM_GET_TIME(0): - if (copy_to_user((void __user *)arg, &ts, sizeof(ts))) - return -EFAULT; - break; - } - - return 0; -} -#ifdef CONFIG_COMPAT -static long alarm_compat_ioctl(struct file *file, unsigned int cmd, - unsigned long arg) -{ - - struct timespec ts; - int rv; - - switch (ANDROID_ALARM_BASE_CMD(cmd)) { - case ANDROID_ALARM_SET_AND_WAIT_COMPAT(0): - case ANDROID_ALARM_SET_COMPAT(0): - case ANDROID_ALARM_SET_RTC_COMPAT: - if (compat_get_timespec(&ts, (void __user *)arg)) - return -EFAULT; - /* fall through */ - case ANDROID_ALARM_GET_TIME_COMPAT(0): - cmd = ANDROID_ALARM_COMPAT_TO_NORM(cmd); - break; - } - - rv = alarm_do_ioctl(file, cmd, &ts); - if (rv) - return rv; - - switch (ANDROID_ALARM_BASE_CMD(cmd)) { - case ANDROID_ALARM_GET_TIME(0): /* NOTE: we modified cmd above */ - if (compat_put_timespec(&ts, (void __user *)arg)) - return -EFAULT; - break; - } - - return 0; -} -#endif - -static int alarm_open(struct inode *inode, struct file *file) -{ - file->private_data = NULL; - return 0; -} - -static int alarm_release(struct inode *inode, struct file *file) -{ - int i; - unsigned long flags; - - spin_lock_irqsave(&alarm_slock, flags); - if (file->private_data) { - for (i = 0; i < ANDROID_ALARM_TYPE_COUNT; i++) { - uint32_t alarm_type_mask = 1U << i; - if (alarm_enabled & alarm_type_mask) { - alarm_dbg(INFO, - "%s: clear alarm, pending %d\n", - __func__, - !!(alarm_pending & alarm_type_mask)); - alarm_enabled &= ~alarm_type_mask; - } - spin_unlock_irqrestore(&alarm_slock, flags); - devalarm_cancel(&alarms[i]); - spin_lock_irqsave(&alarm_slock, flags); - } - if (alarm_pending | wait_pending) { - if (alarm_pending) - alarm_dbg(INFO, "%s: clear pending alarms %x\n", - __func__, alarm_pending); - __pm_relax(&alarm_wake_lock); - wait_pending = 0; - alarm_pending = 0; - } - alarm_opened = 0; - } - spin_unlock_irqrestore(&alarm_slock, flags); - return 0; -} - -static void devalarm_triggered(struct devalarm *alarm) -{ - unsigned long flags; - uint32_t alarm_type_mask = 1U << alarm->type; - - alarm_dbg(INT, "%s: type %d\n", __func__, alarm->type); - spin_lock_irqsave(&alarm_slock, flags); - if (alarm_enabled & alarm_type_mask) { - __pm_wakeup_event(&alarm_wake_lock, 5000); /* 5secs */ - alarm_enabled &= ~alarm_type_mask; - alarm_pending |= alarm_type_mask; - wake_up(&alarm_wait_queue); - } - spin_unlock_irqrestore(&alarm_slock, flags); -} - - -static enum hrtimer_restart devalarm_hrthandler(struct hrtimer *hrt) -{ - struct devalarm *devalrm = container_of(hrt, struct devalarm, u.hrt); - - devalarm_triggered(devalrm); - return HRTIMER_NORESTART; -} - -static enum alarmtimer_restart devalarm_alarmhandler(struct alarm *alrm, - ktime_t now) -{ - struct devalarm *devalrm = container_of(alrm, struct devalarm, u.alrm); - - devalarm_triggered(devalrm); - return ALARMTIMER_NORESTART; -} - - -static const struct file_operations alarm_fops = { - .owner = THIS_MODULE, - .unlocked_ioctl = alarm_ioctl, - .open = alarm_open, - .release = alarm_release, -#ifdef CONFIG_COMPAT - .compat_ioctl = alarm_compat_ioctl, -#endif -}; - -static struct miscdevice alarm_device = { - .minor = MISC_DYNAMIC_MINOR, - .name = "alarm", - .fops = &alarm_fops, -}; - -static int __init alarm_dev_init(void) -{ - int err; - int i; - - err = misc_register(&alarm_device); - if (err) - return err; - - alarm_init(&alarms[ANDROID_ALARM_RTC_WAKEUP].u.alrm, - ALARM_REALTIME, devalarm_alarmhandler); - hrtimer_init(&alarms[ANDROID_ALARM_RTC].u.hrt, - CLOCK_REALTIME, HRTIMER_MODE_ABS); - alarm_init(&alarms[ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP].u.alrm, - ALARM_BOOTTIME, devalarm_alarmhandler); - hrtimer_init(&alarms[ANDROID_ALARM_ELAPSED_REALTIME].u.hrt, - CLOCK_BOOTTIME, HRTIMER_MODE_ABS); - hrtimer_init(&alarms[ANDROID_ALARM_SYSTEMTIME].u.hrt, - CLOCK_MONOTONIC, HRTIMER_MODE_ABS); - - for (i = 0; i < ANDROID_ALARM_TYPE_COUNT; i++) { - alarms[i].type = i; - if (!is_wakeup(i)) - alarms[i].u.hrt.function = devalarm_hrthandler; - } - - wakeup_source_init(&alarm_wake_lock, "alarm"); - return 0; -} - -static void __exit alarm_dev_exit(void) -{ - misc_deregister(&alarm_device); - wakeup_source_trash(&alarm_wake_lock); -} - -module_init(alarm_dev_init); -module_exit(alarm_dev_exit); - diff --git a/drivers/staging/android/android_alarm.h b/drivers/staging/android/android_alarm.h deleted file mode 100644 index 4fd32f337f9c..000000000000 --- a/drivers/staging/android/android_alarm.h +++ /dev/null @@ -1,81 +0,0 @@ -/* include/linux/android_alarm.h - * - * Copyright (C) 2006-2007 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_ANDROID_ALARM_H -#define _LINUX_ANDROID_ALARM_H - -#include <linux/ioctl.h> -#include <linux/time.h> -#include <linux/compat.h> - -enum android_alarm_type { - /* return code bit numbers or set alarm arg */ - ANDROID_ALARM_RTC_WAKEUP, - ANDROID_ALARM_RTC, - ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP, - ANDROID_ALARM_ELAPSED_REALTIME, - ANDROID_ALARM_SYSTEMTIME, - - ANDROID_ALARM_TYPE_COUNT, - - /* return code bit numbers */ - /* ANDROID_ALARM_TIME_CHANGE = 16 */ -}; - -enum android_alarm_return_flags { - ANDROID_ALARM_RTC_WAKEUP_MASK = 1U << ANDROID_ALARM_RTC_WAKEUP, - ANDROID_ALARM_RTC_MASK = 1U << ANDROID_ALARM_RTC, - ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP_MASK = - 1U << ANDROID_ALARM_ELAPSED_REALTIME_WAKEUP, - ANDROID_ALARM_ELAPSED_REALTIME_MASK = - 1U << ANDROID_ALARM_ELAPSED_REALTIME, - ANDROID_ALARM_SYSTEMTIME_MASK = 1U << ANDROID_ALARM_SYSTEMTIME, - ANDROID_ALARM_TIME_CHANGE_MASK = 1U << 16 -}; - -/* Disable alarm */ -#define ANDROID_ALARM_CLEAR(type) _IO('a', 0 | ((type) << 4)) - -/* Ack last alarm and wait for next */ -#define ANDROID_ALARM_WAIT _IO('a', 1) - -#define ALARM_IOW(c, type, size) _IOW('a', (c) | ((type) << 4), size) -/* Set alarm */ -#define ANDROID_ALARM_SET(type) ALARM_IOW(2, type, struct timespec) -#define ANDROID_ALARM_SET_AND_WAIT(type) ALARM_IOW(3, type, struct timespec) -#define ANDROID_ALARM_GET_TIME(type) ALARM_IOW(4, type, struct timespec) -#define ANDROID_ALARM_SET_RTC _IOW('a', 5, struct timespec) -#define ANDROID_ALARM_BASE_CMD(cmd) (cmd & ~(_IOC(0, 0, 0xf0, 0))) -#define ANDROID_ALARM_IOCTL_TO_TYPE(cmd) (_IOC_NR(cmd) >> 4) - - -#ifdef CONFIG_COMPAT -#define ANDROID_ALARM_SET_COMPAT(type) ALARM_IOW(2, type, \ - struct compat_timespec) -#define ANDROID_ALARM_SET_AND_WAIT_COMPAT(type) ALARM_IOW(3, type, \ - struct compat_timespec) -#define ANDROID_ALARM_GET_TIME_COMPAT(type) ALARM_IOW(4, type, \ - struct compat_timespec) -#define ANDROID_ALARM_SET_RTC_COMPAT _IOW('a', 5, \ - struct compat_timespec) -#define ANDROID_ALARM_IOCTL_NR(cmd) (_IOC_NR(cmd) & ((1<<4)-1)) -#define ANDROID_ALARM_COMPAT_TO_NORM(cmd) \ - ALARM_IOW(ANDROID_ALARM_IOCTL_NR(cmd), \ - ANDROID_ALARM_IOCTL_TO_TYPE(cmd), \ - struct timespec) - -#endif - -#endif diff --git a/drivers/staging/android/ashmem.c b/drivers/staging/android/ashmem.c index e681bdd9aa5f..ccaef8b48eba 100644 --- a/drivers/staging/android/ashmem.c +++ b/drivers/staging/android/ashmem.c @@ -224,21 +224,29 @@ static ssize_t ashmem_read(struct file *file, char __user *buf, /* If size is not set, or set to 0, always return EOF. */ if (asma->size == 0) - goto out; + goto out_unlock; if (!asma->file) { ret = -EBADF; - goto out; + goto out_unlock; } - ret = asma->file->f_op->read(asma->file, buf, len, pos); - if (ret < 0) - goto out; + mutex_unlock(&ashmem_mutex); - /** Update backing file pos, since f_ops->read() doesn't */ - asma->file->f_pos = *pos; + /* + * asma and asma->file are used outside the lock here. We assume + * once asma->file is set it will never be changed, and will not + * be destroyed until all references to the file are dropped and + * ashmem_release is called. + */ + ret = asma->file->f_op->read(asma->file, buf, len, pos); + if (ret >= 0) { + /** Update backing file pos, since f_ops->read() doesn't */ + asma->file->f_pos = *pos; + } + return ret; -out: +out_unlock: mutex_unlock(&ashmem_mutex); return ret; } @@ -317,22 +325,14 @@ 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; @@ -363,7 +363,9 @@ static int ashmem_shrink(struct shrinker *s, struct shrink_control *sc) if (!sc->nr_to_scan) return lru_count; - 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; @@ -413,6 +415,7 @@ out: static int set_name(struct ashmem_area *asma, void __user *name) { + int len; int ret = 0; char local_name[ASHMEM_NAME_LEN]; @@ -425,21 +428,19 @@ static int set_name(struct ashmem_area *asma, void __user *name) * variable that does not need protection and later copy the local * variable to the structure member with lock held. */ - if (copy_from_user(local_name, name, ASHMEM_NAME_LEN)) - return -EFAULT; - + len = strncpy_from_user(local_name, name, ASHMEM_NAME_LEN); + if (len < 0) + return len; + if (len == ASHMEM_NAME_LEN) + local_name[ASHMEM_NAME_LEN - 1] = '\0'; mutex_lock(&ashmem_mutex); /* cannot change an existing mapping's name */ - if (unlikely(asma->file)) { + if (unlikely(asma->file)) ret = -EINVAL; - goto out; - } - memcpy(asma->name + ASHMEM_NAME_PREFIX_LEN, - local_name, ASHMEM_NAME_LEN); - asma->name[ASHMEM_FULL_NAME_LEN-1] = '\0'; -out: - mutex_unlock(&ashmem_mutex); + else + strcpy(asma->name + ASHMEM_NAME_PREFIX_LEN, local_name); + mutex_unlock(&ashmem_mutex); return ret; } diff --git a/drivers/staging/android/ashmem.h b/drivers/staging/android/ashmem.h index 8dc0f0d3adf3..5abcfd7aa706 100644 --- a/drivers/staging/android/ashmem.h +++ b/drivers/staging/android/ashmem.h @@ -16,35 +16,7 @@ #include <linux/ioctl.h> #include <linux/compat.h> -#define ASHMEM_NAME_LEN 256 - -#define ASHMEM_NAME_DEF "dev/ashmem" - -/* Return values from ASHMEM_PIN: Was the mapping purged while unpinned? */ -#define ASHMEM_NOT_PURGED 0 -#define ASHMEM_WAS_PURGED 1 - -/* Return values from ASHMEM_GET_PIN_STATUS: Is the mapping pinned? */ -#define ASHMEM_IS_UNPINNED 0 -#define ASHMEM_IS_PINNED 1 - -struct ashmem_pin { - __u32 offset; /* offset into region, in bytes, page-aligned */ - __u32 len; /* length forward from offset, in bytes, page-aligned */ -}; - -#define __ASHMEMIOC 0x77 - -#define ASHMEM_SET_NAME _IOW(__ASHMEMIOC, 1, char[ASHMEM_NAME_LEN]) -#define ASHMEM_GET_NAME _IOR(__ASHMEMIOC, 2, char[ASHMEM_NAME_LEN]) -#define ASHMEM_SET_SIZE _IOW(__ASHMEMIOC, 3, size_t) -#define ASHMEM_GET_SIZE _IO(__ASHMEMIOC, 4) -#define ASHMEM_SET_PROT_MASK _IOW(__ASHMEMIOC, 5, unsigned long) -#define ASHMEM_GET_PROT_MASK _IO(__ASHMEMIOC, 6) -#define ASHMEM_PIN _IOW(__ASHMEMIOC, 7, struct ashmem_pin) -#define ASHMEM_UNPIN _IOW(__ASHMEMIOC, 8, struct ashmem_pin) -#define ASHMEM_GET_PIN_STATUS _IO(__ASHMEMIOC, 9) -#define ASHMEM_PURGE_ALL_CACHES _IO(__ASHMEMIOC, 10) +#include "uapi/ashmem.h" /* support of 32bit userspace on 64bit platforms */ #ifdef CONFIG_COMPAT diff --git a/drivers/staging/android/fiq_debugger/Kconfig b/drivers/staging/android/fiq_debugger/Kconfig new file mode 100644 index 000000000000..56f7f999377e --- /dev/null +++ b/drivers/staging/android/fiq_debugger/Kconfig @@ -0,0 +1,49 @@ +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_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..1d733624d70a --- /dev/null +++ b/drivers/staging/android/fiq_debugger/fiq_debugger.c @@ -0,0 +1,1213 @@ +/* + * 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 + +#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; + +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); + +#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"); + fiq_debugger_printf(&state->output, + " 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 (strict_strtoul(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 | IRQF_DISABLED, + "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, + }, +}; + +static int __init fiq_debugger_init(void) +{ +#if defined(CONFIG_FIQ_DEBUGGER_CONSOLE) + fiq_debugger_tty_init(); +#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..99c6584fcfa5 --- /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(&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 new file mode 100644 index 000000000000..0f8fec1f84e5 --- /dev/null +++ b/drivers/staging/android/ion/Kconfig @@ -0,0 +1,35 @@ +menuconfig ION + bool "Ion Memory Manager" + depends on HAVE_MEMBLOCK + select GENERIC_ALLOCATOR + select DMA_SHARED_BUFFER + ---help--- + Chose this option to enable the ION Memory Manager, + used by Android to efficiently allocate buffers + from userspace that can be shared between drivers. + If you're not using Android its probably safe to + say N here. + +config ION_TEST + tristate "Ion Test Device" + depends on ION + help + Choose this option to create a device that can be used to test the + kernel and device side ION functions. + +config ION_DUMMY + bool "Dummy Ion driver" + depends on ION + help + Provides a dummy ION driver that registers the + /dev/ion device and some basic heaps. This can + be used for testing the ION infrastructure if + one doesn't have access to hardware drivers that + use ION. + +config ION_TEGRA + tristate "Ion for Tegra" + depends on ARCH_TEGRA && ION + help + Choose this option if you wish to use ion on an nVidia Tegra. + diff --git a/drivers/staging/android/ion/Makefile b/drivers/staging/android/ion/Makefile new file mode 100644 index 000000000000..b56fd2bf2b4f --- /dev/null +++ b/drivers/staging/android/ion/Makefile @@ -0,0 +1,10 @@ +obj-$(CONFIG_ION) += ion.o ion_heap.o ion_page_pool.o ion_system_heap.o \ + ion_carveout_heap.o ion_chunk_heap.o ion_cma_heap.o +obj-$(CONFIG_ION_TEST) += ion_test.o +ifdef CONFIG_COMPAT +obj-$(CONFIG_ION) += compat_ion.o +endif + +obj-$(CONFIG_ION_DUMMY) += ion_dummy_driver.o +obj-$(CONFIG_ION_TEGRA) += tegra/ + diff --git a/drivers/staging/android/ion/compat_ion.c b/drivers/staging/android/ion/compat_ion.c new file mode 100644 index 000000000000..a5e8c408fd78 --- /dev/null +++ b/drivers/staging/android/ion/compat_ion.c @@ -0,0 +1,195 @@ +/* + * drivers/gpu/ion/compat_ion.c + * + * 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/compat.h> +#include <linux/fs.h> +#include <linux/uaccess.h> + +#include "ion.h" +#include "compat_ion.h" + +/* See drivers/staging/android/uapi/ion.h for the definition of these structs */ +struct compat_ion_allocation_data { + compat_size_t len; + compat_size_t align; + compat_uint_t heap_id_mask; + compat_uint_t flags; + compat_int_t handle; +}; + +struct compat_ion_custom_data { + compat_uint_t cmd; + compat_ulong_t arg; +}; + +struct compat_ion_handle_data { + compat_int_t handle; +}; + +#define COMPAT_ION_IOC_ALLOC _IOWR(ION_IOC_MAGIC, 0, \ + struct compat_ion_allocation_data) +#define COMPAT_ION_IOC_FREE _IOWR(ION_IOC_MAGIC, 1, \ + struct compat_ion_handle_data) +#define COMPAT_ION_IOC_CUSTOM _IOWR(ION_IOC_MAGIC, 6, \ + struct compat_ion_custom_data) + +static int compat_get_ion_allocation_data( + struct compat_ion_allocation_data __user *data32, + struct ion_allocation_data __user *data) +{ + compat_size_t s; + compat_uint_t u; + compat_int_t i; + int err; + + err = get_user(s, &data32->len); + err |= put_user(s, &data->len); + err |= get_user(s, &data32->align); + err |= put_user(s, &data->align); + err |= get_user(u, &data32->heap_id_mask); + err |= put_user(u, &data->heap_id_mask); + err |= get_user(u, &data32->flags); + err |= put_user(u, &data->flags); + err |= get_user(i, &data32->handle); + err |= put_user(i, &data->handle); + + return err; +} + +static int compat_get_ion_handle_data( + struct compat_ion_handle_data __user *data32, + struct ion_handle_data __user *data) +{ + compat_int_t i; + int err; + + err = get_user(i, &data32->handle); + err |= put_user(i, &data->handle); + + return err; +} + +static int compat_put_ion_allocation_data( + struct compat_ion_allocation_data __user *data32, + struct ion_allocation_data __user *data) +{ + compat_size_t s; + compat_uint_t u; + compat_int_t i; + int err; + + err = get_user(s, &data->len); + err |= put_user(s, &data32->len); + err |= get_user(s, &data->align); + err |= put_user(s, &data32->align); + err |= get_user(u, &data->heap_id_mask); + err |= put_user(u, &data32->heap_id_mask); + err |= get_user(u, &data->flags); + err |= put_user(u, &data32->flags); + err |= get_user(i, &data->handle); + err |= put_user(i, &data32->handle); + + return err; +} + +static int compat_get_ion_custom_data( + struct compat_ion_custom_data __user *data32, + struct ion_custom_data __user *data) +{ + compat_uint_t cmd; + compat_ulong_t arg; + int err; + + err = get_user(cmd, &data32->cmd); + err |= put_user(cmd, &data->cmd); + err |= get_user(arg, &data32->arg); + err |= put_user(arg, &data->arg); + + return err; +}; + +long compat_ion_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) +{ + long ret; + + if (!filp->f_op || !filp->f_op->unlocked_ioctl) + return -ENOTTY; + + switch (cmd) { + case COMPAT_ION_IOC_ALLOC: + { + struct compat_ion_allocation_data __user *data32; + struct ion_allocation_data __user *data; + int err; + + data32 = compat_ptr(arg); + data = compat_alloc_user_space(sizeof(*data)); + if (data == NULL) + return -EFAULT; + + err = compat_get_ion_allocation_data(data32, data); + if (err) + return err; + ret = filp->f_op->unlocked_ioctl(filp, ION_IOC_ALLOC, + (unsigned long)data); + err = compat_put_ion_allocation_data(data32, data); + return ret ? ret : err; + } + case COMPAT_ION_IOC_FREE: + { + struct compat_ion_handle_data __user *data32; + struct ion_handle_data __user *data; + int err; + + data32 = compat_ptr(arg); + data = compat_alloc_user_space(sizeof(*data)); + if (data == NULL) + return -EFAULT; + + err = compat_get_ion_handle_data(data32, data); + if (err) + return err; + + return filp->f_op->unlocked_ioctl(filp, ION_IOC_FREE, + (unsigned long)data); + } + case COMPAT_ION_IOC_CUSTOM: { + struct compat_ion_custom_data __user *data32; + struct ion_custom_data __user *data; + int err; + + data32 = compat_ptr(arg); + data = compat_alloc_user_space(sizeof(*data)); + if (data == NULL) + return -EFAULT; + + err = compat_get_ion_custom_data(data32, data); + if (err) + return err; + + return filp->f_op->unlocked_ioctl(filp, ION_IOC_CUSTOM, + (unsigned long)data); + } + case ION_IOC_SHARE: + case ION_IOC_MAP: + case ION_IOC_IMPORT: + case ION_IOC_SYNC: + return filp->f_op->unlocked_ioctl(filp, cmd, + (unsigned long)compat_ptr(arg)); + default: + return -ENOIOCTLCMD; + } +} diff --git a/drivers/staging/android/ion/compat_ion.h b/drivers/staging/android/ion/compat_ion.h new file mode 100644 index 000000000000..3a9c8c08c240 --- /dev/null +++ b/drivers/staging/android/ion/compat_ion.h @@ -0,0 +1,30 @@ +/* + + * drivers/gpu/ion/compat_ion.h + * + * 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 _LINUX_COMPAT_ION_H +#define _LINUX_COMPAT_ION_H + +#if IS_ENABLED(CONFIG_COMPAT) + +long compat_ion_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); + +#else + +#define compat_ion_ioctl NULL + +#endif /* CONFIG_COMPAT */ +#endif /* _LINUX_COMPAT_ION_H */ diff --git a/drivers/staging/android/ion/ion.c b/drivers/staging/android/ion/ion.c new file mode 100644 index 000000000000..698e363be551 --- /dev/null +++ b/drivers/staging/android/ion/ion.c @@ -0,0 +1,1644 @@ +/* + + * drivers/gpu/ion/ion.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/device.h> +#include <linux/file.h> +#include <linux/freezer.h> +#include <linux/fs.h> +#include <linux/anon_inodes.h> +#include <linux/kthread.h> +#include <linux/list.h> +#include <linux/memblock.h> +#include <linux/miscdevice.h> +#include <linux/export.h> +#include <linux/mm.h> +#include <linux/mm_types.h> +#include <linux/rbtree.h> +#include <linux/slab.h> +#include <linux/seq_file.h> +#include <linux/uaccess.h> +#include <linux/vmalloc.h> +#include <linux/debugfs.h> +#include <linux/dma-buf.h> +#include <linux/idr.h> + +#include "ion.h" +#include "ion_priv.h" +#include "compat_ion.h" + +/** + * struct ion_device - the metadata of the ion device node + * @dev: the actual misc device + * @buffers: an rb tree of all the existing buffers + * @buffer_lock: lock protecting the tree of buffers + * @lock: rwsem protecting the tree of heaps and clients + * @heaps: list of all the heaps in the system + * @user_clients: list of all the clients created from userspace + */ +struct ion_device { + struct miscdevice dev; + struct rb_root buffers; + struct mutex buffer_lock; + struct rw_semaphore lock; + struct plist_head heaps; + long (*custom_ioctl) (struct ion_client *client, unsigned int cmd, + unsigned long arg); + struct rb_root clients; + struct dentry *debug_root; + struct dentry *heaps_debug_root; + struct dentry *clients_debug_root; +}; + +/** + * struct ion_client - a process/hw block local address space + * @node: node in the tree of all clients + * @dev: backpointer to ion device + * @handles: an rb tree of all the handles in this client + * @idr: an idr space for allocating handle ids + * @lock: lock protecting the tree of handles + * @name: used for debugging + * @display_name: used for debugging (unique version of @name) + * @display_serial: used for debugging (to make display_name unique) + * @task: used for debugging + * + * A client represents a list of buffers this client may access. + * The mutex stored here is used to protect both handles tree + * as well as the handles themselves, and should be held while modifying either. + */ +struct ion_client { + struct rb_node node; + struct ion_device *dev; + struct rb_root handles; + struct idr idr; + struct mutex lock; + const char *name; + char *display_name; + int display_serial; + struct task_struct *task; + pid_t pid; + struct dentry *debug_root; +}; + +/** + * ion_handle - a client local reference to a buffer + * @ref: reference count + * @client: back pointer to the client the buffer resides in + * @buffer: pointer to the buffer + * @node: node in the client's handle rbtree + * @kmap_cnt: count of times this client has mapped to kernel + * @id: client-unique id allocated by client->idr + * + * Modifications to node, map_cnt or mapping should be protected by the + * lock in the client. Other fields are never changed after initialization. + */ +struct ion_handle { + struct kref ref; + struct ion_client *client; + struct ion_buffer *buffer; + struct rb_node node; + unsigned int kmap_cnt; + int id; +}; + +bool ion_buffer_fault_user_mappings(struct ion_buffer *buffer) +{ + return (buffer->flags & ION_FLAG_CACHED) && + !(buffer->flags & ION_FLAG_CACHED_NEEDS_SYNC); +} + +bool ion_buffer_cached(struct ion_buffer *buffer) +{ + return !!(buffer->flags & ION_FLAG_CACHED); +} + +static inline struct page *ion_buffer_page(struct page *page) +{ + return (struct page *)((unsigned long)page & ~(1UL)); +} + +static inline bool ion_buffer_page_is_dirty(struct page *page) +{ + return !!((unsigned long)page & 1UL); +} + +static inline void ion_buffer_page_dirty(struct page **page) +{ + *page = (struct page *)((unsigned long)(*page) | 1UL); +} + +static inline void ion_buffer_page_clean(struct page **page) +{ + *page = (struct page *)((unsigned long)(*page) & ~(1UL)); +} + +/* this function should only be called while dev->lock is held */ +static void ion_buffer_add(struct ion_device *dev, + struct ion_buffer *buffer) +{ + struct rb_node **p = &dev->buffers.rb_node; + struct rb_node *parent = NULL; + struct ion_buffer *entry; + + while (*p) { + parent = *p; + entry = rb_entry(parent, struct ion_buffer, node); + + if (buffer < entry) { + p = &(*p)->rb_left; + } else if (buffer > entry) { + p = &(*p)->rb_right; + } else { + pr_err("%s: buffer already found.", __func__); + BUG(); + } + } + + rb_link_node(&buffer->node, parent, p); + rb_insert_color(&buffer->node, &dev->buffers); +} + +/* this function should only be called while dev->lock is held */ +static struct ion_buffer *ion_buffer_create(struct ion_heap *heap, + struct ion_device *dev, + unsigned long len, + unsigned long align, + unsigned long flags) +{ + struct ion_buffer *buffer; + struct sg_table *table; + struct scatterlist *sg; + int i, ret; + + buffer = kzalloc(sizeof(struct ion_buffer), GFP_KERNEL); + if (!buffer) + return ERR_PTR(-ENOMEM); + + buffer->heap = heap; + buffer->flags = flags; + kref_init(&buffer->ref); + + ret = heap->ops->allocate(heap, buffer, len, align, flags); + + if (ret) { + if (!(heap->flags & ION_HEAP_FLAG_DEFER_FREE)) + goto err2; + + ion_heap_freelist_drain(heap, 0); + ret = heap->ops->allocate(heap, buffer, len, align, + flags); + if (ret) + goto err2; + } + + buffer->dev = dev; + buffer->size = len; + + table = heap->ops->map_dma(heap, buffer); + if (WARN_ONCE(table == NULL, + "heap->ops->map_dma should return ERR_PTR on error")) + table = ERR_PTR(-EINVAL); + if (IS_ERR(table)) { + heap->ops->free(buffer); + kfree(buffer); + return ERR_PTR(PTR_ERR(table)); + } + buffer->sg_table = table; + if (ion_buffer_fault_user_mappings(buffer)) { + int num_pages = PAGE_ALIGN(buffer->size) / PAGE_SIZE; + struct scatterlist *sg; + int i, j, k = 0; + + buffer->pages = vmalloc(sizeof(struct page *) * num_pages); + if (!buffer->pages) { + ret = -ENOMEM; + goto err1; + } + + for_each_sg(table->sgl, sg, table->nents, i) { + struct page *page = sg_page(sg); + + for (j = 0; j < sg->length / PAGE_SIZE; j++) + buffer->pages[k++] = page++; + } + + if (ret) + goto err; + } + + buffer->dev = dev; + buffer->size = len; + INIT_LIST_HEAD(&buffer->vmas); + mutex_init(&buffer->lock); + /* this will set up dma addresses for the sglist -- it is not + technically correct as per the dma api -- a specific + device isn't really taking ownership here. However, in practice on + our systems the only dma_address space is physical addresses. + Additionally, we can't afford the overhead of invalidating every + allocation via dma_map_sg. The implicit contract here is that + memory comming from the heaps is ready for dma, ie if it has a + cached mapping that mapping has been invalidated */ + for_each_sg(buffer->sg_table->sgl, sg, buffer->sg_table->nents, i) + sg_dma_address(sg) = sg_phys(sg); + mutex_lock(&dev->buffer_lock); + ion_buffer_add(dev, buffer); + mutex_unlock(&dev->buffer_lock); + return buffer; + +err: + heap->ops->unmap_dma(heap, buffer); + heap->ops->free(buffer); +err1: + if (buffer->pages) + vfree(buffer->pages); +err2: + kfree(buffer); + return ERR_PTR(ret); +} + +void ion_buffer_destroy(struct ion_buffer *buffer) +{ + if (WARN_ON(buffer->kmap_cnt > 0)) + buffer->heap->ops->unmap_kernel(buffer->heap, buffer); + buffer->heap->ops->unmap_dma(buffer->heap, buffer); + buffer->heap->ops->free(buffer); + if (buffer->pages) + vfree(buffer->pages); + kfree(buffer); +} + +static void _ion_buffer_destroy(struct kref *kref) +{ + struct ion_buffer *buffer = container_of(kref, struct ion_buffer, ref); + struct ion_heap *heap = buffer->heap; + struct ion_device *dev = buffer->dev; + + mutex_lock(&dev->buffer_lock); + rb_erase(&buffer->node, &dev->buffers); + mutex_unlock(&dev->buffer_lock); + + if (heap->flags & ION_HEAP_FLAG_DEFER_FREE) + ion_heap_freelist_add(heap, buffer); + else + ion_buffer_destroy(buffer); +} + +static void ion_buffer_get(struct ion_buffer *buffer) +{ + kref_get(&buffer->ref); +} + +static int ion_buffer_put(struct ion_buffer *buffer) +{ + return kref_put(&buffer->ref, _ion_buffer_destroy); +} + +static void ion_buffer_add_to_handle(struct ion_buffer *buffer) +{ + mutex_lock(&buffer->lock); + buffer->handle_count++; + mutex_unlock(&buffer->lock); +} + +static void ion_buffer_remove_from_handle(struct ion_buffer *buffer) +{ + /* + * when a buffer is removed from a handle, if it is not in + * any other handles, copy the taskcomm and the pid of the + * process it's being removed from into the buffer. At this + * point there will be no way to track what processes this buffer is + * being used by, it only exists as a dma_buf file descriptor. + * The taskcomm and pid can provide a debug hint as to where this fd + * is in the system + */ + mutex_lock(&buffer->lock); + buffer->handle_count--; + BUG_ON(buffer->handle_count < 0); + if (!buffer->handle_count) { + struct task_struct *task; + + task = current->group_leader; + get_task_comm(buffer->task_comm, task); + buffer->pid = task_pid_nr(task); + } + mutex_unlock(&buffer->lock); +} + +static struct ion_handle *ion_handle_create(struct ion_client *client, + struct ion_buffer *buffer) +{ + struct ion_handle *handle; + + handle = kzalloc(sizeof(struct ion_handle), GFP_KERNEL); + if (!handle) + return ERR_PTR(-ENOMEM); + kref_init(&handle->ref); + RB_CLEAR_NODE(&handle->node); + handle->client = client; + ion_buffer_get(buffer); + ion_buffer_add_to_handle(buffer); + handle->buffer = buffer; + + return handle; +} + +static void ion_handle_kmap_put(struct ion_handle *); + +static void ion_handle_destroy(struct kref *kref) +{ + struct ion_handle *handle = container_of(kref, struct ion_handle, ref); + struct ion_client *client = handle->client; + struct ion_buffer *buffer = handle->buffer; + + mutex_lock(&buffer->lock); + while (handle->kmap_cnt) + ion_handle_kmap_put(handle); + mutex_unlock(&buffer->lock); + + idr_remove(&client->idr, handle->id); + if (!RB_EMPTY_NODE(&handle->node)) + rb_erase(&handle->node, &client->handles); + + ion_buffer_remove_from_handle(buffer); + ion_buffer_put(buffer); + + kfree(handle); +} + +struct ion_buffer *ion_handle_buffer(struct ion_handle *handle) +{ + return handle->buffer; +} + +static void ion_handle_get(struct ion_handle *handle) +{ + kref_get(&handle->ref); +} + +static int ion_handle_put(struct ion_handle *handle) +{ + struct ion_client *client = handle->client; + int ret; + + mutex_lock(&client->lock); + ret = kref_put(&handle->ref, ion_handle_destroy); + mutex_unlock(&client->lock); + + return ret; +} + +static struct ion_handle *ion_handle_lookup(struct ion_client *client, + struct ion_buffer *buffer) +{ + struct rb_node *n = client->handles.rb_node; + + while (n) { + struct ion_handle *entry = rb_entry(n, struct ion_handle, node); + + if (buffer < entry->buffer) + n = n->rb_left; + else if (buffer > entry->buffer) + n = n->rb_right; + else + return entry; + } + return ERR_PTR(-EINVAL); +} + +static struct ion_handle *ion_handle_get_by_id(struct ion_client *client, + int id) +{ + struct ion_handle *handle; + + mutex_lock(&client->lock); + handle = idr_find(&client->idr, id); + if (handle) + ion_handle_get(handle); + mutex_unlock(&client->lock); + + return handle ? handle : ERR_PTR(-EINVAL); +} + +static bool ion_handle_validate(struct ion_client *client, + struct ion_handle *handle) +{ + WARN_ON(!mutex_is_locked(&client->lock)); + return (idr_find(&client->idr, handle->id) == handle); +} + +static int ion_handle_add(struct ion_client *client, struct ion_handle *handle) +{ + int id; + struct rb_node **p = &client->handles.rb_node; + struct rb_node *parent = NULL; + struct ion_handle *entry; + + id = idr_alloc(&client->idr, handle, 1, 0, GFP_KERNEL); + if (id < 0) + return id; + + handle->id = id; + + while (*p) { + parent = *p; + entry = rb_entry(parent, struct ion_handle, node); + + if (handle->buffer < entry->buffer) + p = &(*p)->rb_left; + else if (handle->buffer > entry->buffer) + p = &(*p)->rb_right; + else + WARN(1, "%s: buffer already found.", __func__); + } + + rb_link_node(&handle->node, parent, p); + rb_insert_color(&handle->node, &client->handles); + + return 0; +} + +struct ion_handle *ion_alloc(struct ion_client *client, size_t len, + size_t align, unsigned int heap_id_mask, + unsigned int flags) +{ + struct ion_handle *handle; + struct ion_device *dev = client->dev; + struct ion_buffer *buffer = NULL; + struct ion_heap *heap; + int ret; + + pr_debug("%s: len %zu align %zu heap_id_mask %u flags %x\n", __func__, + len, align, heap_id_mask, flags); + /* + * traverse the list of heaps available in this system in priority + * order. If the heap type is supported by the client, and matches the + * request of the caller allocate from it. Repeat until allocate has + * succeeded or all heaps have been tried + */ + len = PAGE_ALIGN(len); + + if (!len) + return ERR_PTR(-EINVAL); + + down_read(&dev->lock); + plist_for_each_entry(heap, &dev->heaps, node) { + /* if the caller didn't specify this heap id */ + if (!((1 << heap->id) & heap_id_mask)) + continue; + buffer = ion_buffer_create(heap, dev, len, align, flags); + if (!IS_ERR(buffer)) + break; + } + up_read(&dev->lock); + + if (buffer == NULL) + return ERR_PTR(-ENODEV); + + if (IS_ERR(buffer)) + return ERR_PTR(PTR_ERR(buffer)); + + handle = ion_handle_create(client, buffer); + + /* + * ion_buffer_create will create a buffer with a ref_cnt of 1, + * and ion_handle_create will take a second reference, drop one here + */ + ion_buffer_put(buffer); + + if (IS_ERR(handle)) + return handle; + + mutex_lock(&client->lock); + ret = ion_handle_add(client, handle); + mutex_unlock(&client->lock); + if (ret) { + ion_handle_put(handle); + handle = ERR_PTR(ret); + } + + return handle; +} +EXPORT_SYMBOL(ion_alloc); + +void ion_free(struct ion_client *client, struct ion_handle *handle) +{ + bool valid_handle; + + BUG_ON(client != handle->client); + + mutex_lock(&client->lock); + valid_handle = ion_handle_validate(client, handle); + + if (!valid_handle) { + WARN(1, "%s: invalid handle passed to free.\n", __func__); + mutex_unlock(&client->lock); + return; + } + mutex_unlock(&client->lock); + ion_handle_put(handle); +} +EXPORT_SYMBOL(ion_free); + +int ion_phys(struct ion_client *client, struct ion_handle *handle, + ion_phys_addr_t *addr, size_t *len) +{ + struct ion_buffer *buffer; + int ret; + + mutex_lock(&client->lock); + if (!ion_handle_validate(client, handle)) { + mutex_unlock(&client->lock); + return -EINVAL; + } + + buffer = handle->buffer; + + if (!buffer->heap->ops->phys) { + pr_err("%s: ion_phys is not implemented by this heap.\n", + __func__); + mutex_unlock(&client->lock); + return -ENODEV; + } + mutex_unlock(&client->lock); + ret = buffer->heap->ops->phys(buffer->heap, buffer, addr, len); + return ret; +} +EXPORT_SYMBOL(ion_phys); + +static void *ion_buffer_kmap_get(struct ion_buffer *buffer) +{ + void *vaddr; + + if (buffer->kmap_cnt) { + buffer->kmap_cnt++; + return buffer->vaddr; + } + vaddr = buffer->heap->ops->map_kernel(buffer->heap, buffer); + if (WARN_ONCE(vaddr == NULL, + "heap->ops->map_kernel should return ERR_PTR on error")) + return ERR_PTR(-EINVAL); + if (IS_ERR(vaddr)) + return vaddr; + buffer->vaddr = vaddr; + buffer->kmap_cnt++; + return vaddr; +} + +static void *ion_handle_kmap_get(struct ion_handle *handle) +{ + struct ion_buffer *buffer = handle->buffer; + void *vaddr; + + if (handle->kmap_cnt) { + handle->kmap_cnt++; + return buffer->vaddr; + } + vaddr = ion_buffer_kmap_get(buffer); + if (IS_ERR(vaddr)) + return vaddr; + handle->kmap_cnt++; + return vaddr; +} + +static void ion_buffer_kmap_put(struct ion_buffer *buffer) +{ + buffer->kmap_cnt--; + if (!buffer->kmap_cnt) { + buffer->heap->ops->unmap_kernel(buffer->heap, buffer); + buffer->vaddr = NULL; + } +} + +static void ion_handle_kmap_put(struct ion_handle *handle) +{ + struct ion_buffer *buffer = handle->buffer; + + handle->kmap_cnt--; + if (!handle->kmap_cnt) + ion_buffer_kmap_put(buffer); +} + +void *ion_map_kernel(struct ion_client *client, struct ion_handle *handle) +{ + struct ion_buffer *buffer; + void *vaddr; + + mutex_lock(&client->lock); + if (!ion_handle_validate(client, handle)) { + pr_err("%s: invalid handle passed to map_kernel.\n", + __func__); + mutex_unlock(&client->lock); + return ERR_PTR(-EINVAL); + } + + buffer = handle->buffer; + + if (!handle->buffer->heap->ops->map_kernel) { + pr_err("%s: map_kernel is not implemented by this heap.\n", + __func__); + mutex_unlock(&client->lock); + return ERR_PTR(-ENODEV); + } + + mutex_lock(&buffer->lock); + vaddr = ion_handle_kmap_get(handle); + mutex_unlock(&buffer->lock); + mutex_unlock(&client->lock); + return vaddr; +} +EXPORT_SYMBOL(ion_map_kernel); + +void ion_unmap_kernel(struct ion_client *client, struct ion_handle *handle) +{ + struct ion_buffer *buffer; + + mutex_lock(&client->lock); + buffer = handle->buffer; + mutex_lock(&buffer->lock); + ion_handle_kmap_put(handle); + mutex_unlock(&buffer->lock); + mutex_unlock(&client->lock); +} +EXPORT_SYMBOL(ion_unmap_kernel); + +static int ion_debug_client_show(struct seq_file *s, void *unused) +{ + struct ion_client *client = s->private; + struct rb_node *n; + size_t sizes[ION_NUM_HEAP_IDS] = {0}; + const char *names[ION_NUM_HEAP_IDS] = {NULL}; + int i; + + mutex_lock(&client->lock); + for (n = rb_first(&client->handles); n; n = rb_next(n)) { + struct ion_handle *handle = rb_entry(n, struct ion_handle, + node); + unsigned int id = handle->buffer->heap->id; + + if (!names[id]) + names[id] = handle->buffer->heap->name; + sizes[id] += handle->buffer->size; + } + mutex_unlock(&client->lock); + + seq_printf(s, "%16.16s: %16.16s\n", "heap_name", "size_in_bytes"); + for (i = 0; i < ION_NUM_HEAP_IDS; i++) { + if (!names[i]) + continue; + seq_printf(s, "%16.16s: %16zu\n", names[i], sizes[i]); + } + return 0; +} + +static int ion_debug_client_open(struct inode *inode, struct file *file) +{ + return single_open(file, ion_debug_client_show, inode->i_private); +} + +static const struct file_operations debug_client_fops = { + .open = ion_debug_client_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int ion_get_client_serial(const struct rb_root *root, + const unsigned char *name) +{ + int serial = -1; + struct rb_node *node; + + for (node = rb_first(root); node; node = rb_next(node)) { + struct ion_client *client = rb_entry(node, struct ion_client, + node); + + if (strcmp(client->name, name)) + continue; + serial = max(serial, client->display_serial); + } + return serial + 1; +} + +struct ion_client *ion_client_create(struct ion_device *dev, + const char *name) +{ + struct ion_client *client; + struct task_struct *task; + struct rb_node **p; + struct rb_node *parent = NULL; + struct ion_client *entry; + pid_t pid; + + if (!name) { + pr_err("%s: Name cannot be null\n", __func__); + return ERR_PTR(-EINVAL); + } + + get_task_struct(current->group_leader); + task_lock(current->group_leader); + pid = task_pid_nr(current->group_leader); + /* don't bother to store task struct for kernel threads, + they can't be killed anyway */ + if (current->group_leader->flags & PF_KTHREAD) { + put_task_struct(current->group_leader); + task = NULL; + } else { + task = current->group_leader; + } + task_unlock(current->group_leader); + + client = kzalloc(sizeof(struct ion_client), GFP_KERNEL); + if (!client) + goto err_put_task_struct; + + client->dev = dev; + client->handles = RB_ROOT; + idr_init(&client->idr); + mutex_init(&client->lock); + client->task = task; + client->pid = pid; + client->name = kstrdup(name, GFP_KERNEL); + if (!client->name) + goto err_free_client; + + down_write(&dev->lock); + client->display_serial = ion_get_client_serial(&dev->clients, name); + client->display_name = kasprintf( + GFP_KERNEL, "%s-%d", name, client->display_serial); + if (!client->display_name) { + up_write(&dev->lock); + goto err_free_client_name; + } + p = &dev->clients.rb_node; + while (*p) { + parent = *p; + entry = rb_entry(parent, struct ion_client, node); + + if (client < entry) + p = &(*p)->rb_left; + else if (client > entry) + p = &(*p)->rb_right; + } + rb_link_node(&client->node, parent, p); + rb_insert_color(&client->node, &dev->clients); + + client->debug_root = debugfs_create_file(client->display_name, 0664, + dev->clients_debug_root, + client, &debug_client_fops); + if (!client->debug_root) { + char buf[256], *path; + path = dentry_path(dev->clients_debug_root, buf, 256); + pr_err("Failed to create client debugfs at %s/%s\n", + path, client->display_name); + } + + up_write(&dev->lock); + + return client; + +err_free_client_name: + kfree(client->name); +err_free_client: + kfree(client); +err_put_task_struct: + if (task) + put_task_struct(current->group_leader); + return ERR_PTR(-ENOMEM); +} +EXPORT_SYMBOL(ion_client_create); + +void ion_client_destroy(struct ion_client *client) +{ + struct ion_device *dev = client->dev; + struct rb_node *n; + + pr_debug("%s: %d\n", __func__, __LINE__); + while ((n = rb_first(&client->handles))) { + struct ion_handle *handle = rb_entry(n, struct ion_handle, + node); + ion_handle_destroy(&handle->ref); + } + + idr_destroy(&client->idr); + + down_write(&dev->lock); + if (client->task) + put_task_struct(client->task); + rb_erase(&client->node, &dev->clients); + debugfs_remove_recursive(client->debug_root); + up_write(&dev->lock); + + kfree(client->display_name); + kfree(client->name); + kfree(client); +} +EXPORT_SYMBOL(ion_client_destroy); + +struct sg_table *ion_sg_table(struct ion_client *client, + struct ion_handle *handle) +{ + struct ion_buffer *buffer; + struct sg_table *table; + + mutex_lock(&client->lock); + if (!ion_handle_validate(client, handle)) { + pr_err("%s: invalid handle passed to map_dma.\n", + __func__); + mutex_unlock(&client->lock); + return ERR_PTR(-EINVAL); + } + buffer = handle->buffer; + table = buffer->sg_table; + mutex_unlock(&client->lock); + return table; +} +EXPORT_SYMBOL(ion_sg_table); + +static void ion_buffer_sync_for_device(struct ion_buffer *buffer, + struct device *dev, + enum dma_data_direction direction); + +static struct sg_table *ion_map_dma_buf(struct dma_buf_attachment *attachment, + enum dma_data_direction direction) +{ + struct dma_buf *dmabuf = attachment->dmabuf; + struct ion_buffer *buffer = dmabuf->priv; + + ion_buffer_sync_for_device(buffer, attachment->dev, direction); + return buffer->sg_table; +} + +static void ion_unmap_dma_buf(struct dma_buf_attachment *attachment, + struct sg_table *table, + enum dma_data_direction direction) +{ +} + +void ion_pages_sync_for_device(struct device *dev, struct page *page, + size_t size, enum dma_data_direction dir) +{ + struct scatterlist sg; + + sg_init_table(&sg, 1); + sg_set_page(&sg, page, size, 0); + /* + * This is not correct - sg_dma_address needs a dma_addr_t that is valid + * for the the targeted device, but this works on the currently targeted + * hardware. + */ + sg_dma_address(&sg) = page_to_phys(page); + dma_sync_sg_for_device(dev, &sg, 1, dir); +} + +struct ion_vma_list { + struct list_head list; + struct vm_area_struct *vma; +}; + +static void ion_buffer_sync_for_device(struct ion_buffer *buffer, + struct device *dev, + enum dma_data_direction dir) +{ + struct ion_vma_list *vma_list; + int pages = PAGE_ALIGN(buffer->size) / PAGE_SIZE; + int i; + + pr_debug("%s: syncing for device %s\n", __func__, + dev ? dev_name(dev) : "null"); + + if (!ion_buffer_fault_user_mappings(buffer)) + return; + + mutex_lock(&buffer->lock); + for (i = 0; i < pages; i++) { + struct page *page = buffer->pages[i]; + + if (ion_buffer_page_is_dirty(page)) + ion_pages_sync_for_device(dev, ion_buffer_page(page), + PAGE_SIZE, dir); + + ion_buffer_page_clean(buffer->pages + i); + } + list_for_each_entry(vma_list, &buffer->vmas, list) { + struct vm_area_struct *vma = vma_list->vma; + + zap_page_range(vma, vma->vm_start, vma->vm_end - vma->vm_start, + NULL); + } + mutex_unlock(&buffer->lock); +} + +static int ion_vm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + struct ion_buffer *buffer = vma->vm_private_data; + unsigned long pfn; + int ret; + + mutex_lock(&buffer->lock); + ion_buffer_page_dirty(buffer->pages + vmf->pgoff); + BUG_ON(!buffer->pages || !buffer->pages[vmf->pgoff]); + + pfn = page_to_pfn(ion_buffer_page(buffer->pages[vmf->pgoff])); + ret = vm_insert_pfn(vma, (unsigned long)vmf->virtual_address, pfn); + mutex_unlock(&buffer->lock); + if (ret) + return VM_FAULT_ERROR; + + return VM_FAULT_NOPAGE; +} + +static void ion_vm_open(struct vm_area_struct *vma) +{ + struct ion_buffer *buffer = vma->vm_private_data; + struct ion_vma_list *vma_list; + + vma_list = kmalloc(sizeof(struct ion_vma_list), GFP_KERNEL); + if (!vma_list) + return; + vma_list->vma = vma; + mutex_lock(&buffer->lock); + list_add(&vma_list->list, &buffer->vmas); + mutex_unlock(&buffer->lock); + pr_debug("%s: adding %p\n", __func__, vma); +} + +static void ion_vm_close(struct vm_area_struct *vma) +{ + struct ion_buffer *buffer = vma->vm_private_data; + struct ion_vma_list *vma_list, *tmp; + + pr_debug("%s\n", __func__); + mutex_lock(&buffer->lock); + list_for_each_entry_safe(vma_list, tmp, &buffer->vmas, list) { + if (vma_list->vma != vma) + continue; + list_del(&vma_list->list); + kfree(vma_list); + pr_debug("%s: deleting %p\n", __func__, vma); + break; + } + mutex_unlock(&buffer->lock); +} + +static struct vm_operations_struct ion_vma_ops = { + .open = ion_vm_open, + .close = ion_vm_close, + .fault = ion_vm_fault, +}; + +static int ion_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma) +{ + struct ion_buffer *buffer = dmabuf->priv; + int ret = 0; + + if (!buffer->heap->ops->map_user) { + pr_err("%s: this heap does not define a method for mapping " + "to userspace\n", __func__); + return -EINVAL; + } + + if (ion_buffer_fault_user_mappings(buffer)) { + vma->vm_flags |= VM_IO | VM_PFNMAP | VM_DONTEXPAND | + VM_DONTDUMP; + vma->vm_private_data = buffer; + vma->vm_ops = &ion_vma_ops; + ion_vm_open(vma); + return 0; + } + + if (!(buffer->flags & ION_FLAG_CACHED)) + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); + + mutex_lock(&buffer->lock); + /* now map it to userspace */ + ret = buffer->heap->ops->map_user(buffer->heap, buffer, vma); + mutex_unlock(&buffer->lock); + + if (ret) + pr_err("%s: failure mapping buffer to userspace\n", + __func__); + + return ret; +} + +static void ion_dma_buf_release(struct dma_buf *dmabuf) +{ + struct ion_buffer *buffer = dmabuf->priv; + + ion_buffer_put(buffer); +} + +static void *ion_dma_buf_kmap(struct dma_buf *dmabuf, unsigned long offset) +{ + struct ion_buffer *buffer = dmabuf->priv; + + return buffer->vaddr + offset * PAGE_SIZE; +} + +static void ion_dma_buf_kunmap(struct dma_buf *dmabuf, unsigned long offset, + void *ptr) +{ + return; +} + +static int ion_dma_buf_begin_cpu_access(struct dma_buf *dmabuf, size_t start, + size_t len, + enum dma_data_direction direction) +{ + struct ion_buffer *buffer = dmabuf->priv; + void *vaddr; + + if (!buffer->heap->ops->map_kernel) { + pr_err("%s: map kernel is not implemented by this heap.\n", + __func__); + return -ENODEV; + } + + mutex_lock(&buffer->lock); + vaddr = ion_buffer_kmap_get(buffer); + mutex_unlock(&buffer->lock); + if (IS_ERR(vaddr)) + return PTR_ERR(vaddr); + return 0; +} + +static void ion_dma_buf_end_cpu_access(struct dma_buf *dmabuf, size_t start, + size_t len, + enum dma_data_direction direction) +{ + struct ion_buffer *buffer = dmabuf->priv; + + mutex_lock(&buffer->lock); + ion_buffer_kmap_put(buffer); + mutex_unlock(&buffer->lock); +} + +static struct dma_buf_ops dma_buf_ops = { + .map_dma_buf = ion_map_dma_buf, + .unmap_dma_buf = ion_unmap_dma_buf, + .mmap = ion_mmap, + .release = ion_dma_buf_release, + .begin_cpu_access = ion_dma_buf_begin_cpu_access, + .end_cpu_access = ion_dma_buf_end_cpu_access, + .kmap_atomic = ion_dma_buf_kmap, + .kunmap_atomic = ion_dma_buf_kunmap, + .kmap = ion_dma_buf_kmap, + .kunmap = ion_dma_buf_kunmap, +}; + +struct dma_buf *ion_share_dma_buf(struct ion_client *client, + struct ion_handle *handle) +{ + struct ion_buffer *buffer; + struct dma_buf *dmabuf; + bool valid_handle; + + mutex_lock(&client->lock); + valid_handle = ion_handle_validate(client, handle); + if (!valid_handle) { + WARN(1, "%s: invalid handle passed to share.\n", __func__); + mutex_unlock(&client->lock); + return ERR_PTR(-EINVAL); + } + buffer = handle->buffer; + ion_buffer_get(buffer); + mutex_unlock(&client->lock); + + dmabuf = dma_buf_export(buffer, &dma_buf_ops, buffer->size, O_RDWR); + if (IS_ERR(dmabuf)) { + ion_buffer_put(buffer); + return dmabuf; + } + + return dmabuf; +} +EXPORT_SYMBOL(ion_share_dma_buf); + +int ion_share_dma_buf_fd(struct ion_client *client, struct ion_handle *handle) +{ + struct dma_buf *dmabuf; + int fd; + + dmabuf = ion_share_dma_buf(client, handle); + if (IS_ERR(dmabuf)) + return PTR_ERR(dmabuf); + + fd = dma_buf_fd(dmabuf, O_CLOEXEC); + if (fd < 0) + dma_buf_put(dmabuf); + + return fd; +} +EXPORT_SYMBOL(ion_share_dma_buf_fd); + +struct ion_handle *ion_import_dma_buf(struct ion_client *client, int fd) +{ + struct dma_buf *dmabuf; + struct ion_buffer *buffer; + struct ion_handle *handle; + int ret; + + dmabuf = dma_buf_get(fd); + if (IS_ERR(dmabuf)) + return ERR_PTR(PTR_ERR(dmabuf)); + /* if this memory came from ion */ + + if (dmabuf->ops != &dma_buf_ops) { + pr_err("%s: can not import dmabuf from another exporter\n", + __func__); + dma_buf_put(dmabuf); + return ERR_PTR(-EINVAL); + } + buffer = dmabuf->priv; + + mutex_lock(&client->lock); + /* if a handle exists for this buffer just take a reference to it */ + handle = ion_handle_lookup(client, buffer); + if (!IS_ERR(handle)) { + ion_handle_get(handle); + mutex_unlock(&client->lock); + goto end; + } + mutex_unlock(&client->lock); + + handle = ion_handle_create(client, buffer); + if (IS_ERR(handle)) + goto end; + + mutex_lock(&client->lock); + ret = ion_handle_add(client, handle); + mutex_unlock(&client->lock); + if (ret) { + ion_handle_put(handle); + handle = ERR_PTR(ret); + } + +end: + dma_buf_put(dmabuf); + return handle; +} +EXPORT_SYMBOL(ion_import_dma_buf); + +static int ion_sync_for_device(struct ion_client *client, int fd) +{ + struct dma_buf *dmabuf; + struct ion_buffer *buffer; + + dmabuf = dma_buf_get(fd); + if (IS_ERR(dmabuf)) + return PTR_ERR(dmabuf); + + /* if this memory came from ion */ + if (dmabuf->ops != &dma_buf_ops) { + pr_err("%s: can not sync dmabuf from another exporter\n", + __func__); + dma_buf_put(dmabuf); + return -EINVAL; + } + buffer = dmabuf->priv; + + dma_sync_sg_for_device(NULL, buffer->sg_table->sgl, + buffer->sg_table->nents, DMA_BIDIRECTIONAL); + dma_buf_put(dmabuf); + return 0; +} + +/* fix up the cases where the ioctl direction bits are incorrect */ +static unsigned int ion_ioctl_dir(unsigned int cmd) +{ + switch (cmd) { + case ION_IOC_SYNC: + case ION_IOC_FREE: + case ION_IOC_CUSTOM: + return _IOC_WRITE; + default: + return _IOC_DIR(cmd); + } +} + +static long ion_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) +{ + struct ion_client *client = filp->private_data; + struct ion_device *dev = client->dev; + struct ion_handle *cleanup_handle = NULL; + int ret = 0; + unsigned int dir; + + union { + struct ion_fd_data fd; + struct ion_allocation_data allocation; + struct ion_handle_data handle; + struct ion_custom_data custom; + } data; + + dir = ion_ioctl_dir(cmd); + + if (_IOC_SIZE(cmd) > sizeof(data)) + return -EINVAL; + + if (dir & _IOC_WRITE) + if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd))) + return -EFAULT; + + switch (cmd) { + case ION_IOC_ALLOC: + { + struct ion_handle *handle; + + handle = ion_alloc(client, data.allocation.len, + data.allocation.align, + data.allocation.heap_id_mask, + data.allocation.flags); + if (IS_ERR(handle)) + return PTR_ERR(handle); + + data.allocation.handle = handle->id; + + cleanup_handle = handle; + break; + } + case ION_IOC_FREE: + { + struct ion_handle *handle; + + handle = ion_handle_get_by_id(client, data.handle.handle); + if (IS_ERR(handle)) + return PTR_ERR(handle); + ion_free(client, handle); + ion_handle_put(handle); + break; + } + case ION_IOC_SHARE: + case ION_IOC_MAP: + { + struct ion_handle *handle; + + handle = ion_handle_get_by_id(client, data.handle.handle); + if (IS_ERR(handle)) + return PTR_ERR(handle); + data.fd.fd = ion_share_dma_buf_fd(client, handle); + ion_handle_put(handle); + if (data.fd.fd < 0) + ret = data.fd.fd; + break; + } + case ION_IOC_IMPORT: + { + struct ion_handle *handle; + + handle = ion_import_dma_buf(client, data.fd.fd); + if (IS_ERR(handle)) + ret = PTR_ERR(handle); + else + data.handle.handle = handle->id; + break; + } + case ION_IOC_SYNC: + { + ret = ion_sync_for_device(client, data.fd.fd); + break; + } + case ION_IOC_CUSTOM: + { + if (!dev->custom_ioctl) + return -ENOTTY; + ret = dev->custom_ioctl(client, data.custom.cmd, + data.custom.arg); + break; + } + default: + return -ENOTTY; + } + + if (dir & _IOC_READ) { + if (copy_to_user((void __user *)arg, &data, _IOC_SIZE(cmd))) { + if (cleanup_handle) + ion_free(client, cleanup_handle); + return -EFAULT; + } + } + return ret; +} + +static int ion_release(struct inode *inode, struct file *file) +{ + struct ion_client *client = file->private_data; + + pr_debug("%s: %d\n", __func__, __LINE__); + ion_client_destroy(client); + return 0; +} + +static int ion_open(struct inode *inode, struct file *file) +{ + struct miscdevice *miscdev = file->private_data; + struct ion_device *dev = container_of(miscdev, struct ion_device, dev); + struct ion_client *client; + char debug_name[64]; + + pr_debug("%s: %d\n", __func__, __LINE__); + snprintf(debug_name, 64, "%u", task_pid_nr(current->group_leader)); + client = ion_client_create(dev, debug_name); + if (IS_ERR(client)) + return PTR_ERR(client); + file->private_data = client; + + return 0; +} + +static const struct file_operations ion_fops = { + .owner = THIS_MODULE, + .open = ion_open, + .release = ion_release, + .unlocked_ioctl = ion_ioctl, + .compat_ioctl = compat_ion_ioctl, +}; + +static size_t ion_debug_heap_total(struct ion_client *client, + unsigned int id) +{ + size_t size = 0; + struct rb_node *n; + + mutex_lock(&client->lock); + for (n = rb_first(&client->handles); n; n = rb_next(n)) { + struct ion_handle *handle = rb_entry(n, + struct ion_handle, + node); + if (handle->buffer->heap->id == id) + size += handle->buffer->size; + } + mutex_unlock(&client->lock); + return size; +} + +static int ion_debug_heap_show(struct seq_file *s, void *unused) +{ + struct ion_heap *heap = s->private; + struct ion_device *dev = heap->dev; + struct rb_node *n; + size_t total_size = 0; + size_t total_orphaned_size = 0; + + seq_printf(s, "%16.s %16.s %16.s\n", "client", "pid", "size"); + seq_printf(s, "----------------------------------------------------\n"); + + for (n = rb_first(&dev->clients); n; n = rb_next(n)) { + struct ion_client *client = rb_entry(n, struct ion_client, + node); + size_t size = ion_debug_heap_total(client, heap->id); + + if (!size) + continue; + if (client->task) { + char task_comm[TASK_COMM_LEN]; + + get_task_comm(task_comm, client->task); + seq_printf(s, "%16.s %16u %16zu\n", task_comm, + client->pid, size); + } else { + seq_printf(s, "%16.s %16u %16zu\n", client->name, + client->pid, size); + } + } + seq_printf(s, "----------------------------------------------------\n"); + seq_printf(s, "orphaned allocations (info is from last known client):" + "\n"); + mutex_lock(&dev->buffer_lock); + for (n = rb_first(&dev->buffers); n; n = rb_next(n)) { + struct ion_buffer *buffer = rb_entry(n, struct ion_buffer, + node); + if (buffer->heap->id != heap->id) + continue; + total_size += buffer->size; + if (!buffer->handle_count) { + seq_printf(s, "%16.s %16u %16zu %d %d\n", + buffer->task_comm, buffer->pid, + buffer->size, buffer->kmap_cnt, + atomic_read(&buffer->ref.refcount)); + total_orphaned_size += buffer->size; + } + } + mutex_unlock(&dev->buffer_lock); + seq_printf(s, "----------------------------------------------------\n"); + seq_printf(s, "%16.s %16zu\n", "total orphaned", + total_orphaned_size); + seq_printf(s, "%16.s %16zu\n", "total ", total_size); + if (heap->flags & ION_HEAP_FLAG_DEFER_FREE) + seq_printf(s, "%16.s %16zu\n", "deferred free", + heap->free_list_size); + seq_printf(s, "----------------------------------------------------\n"); + + if (heap->debug_show) + heap->debug_show(heap, s, unused); + + return 0; +} + +static int ion_debug_heap_open(struct inode *inode, struct file *file) +{ + return single_open(file, ion_debug_heap_show, inode->i_private); +} + +static const struct file_operations debug_heap_fops = { + .open = ion_debug_heap_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +#ifdef DEBUG_HEAP_SHRINKER +static int debug_shrink_set(void *data, u64 val) +{ + struct ion_heap *heap = data; + struct shrink_control sc; + int objs; + + sc.gfp_mask = -1; + sc.nr_to_scan = 0; + + if (!val) + return 0; + + objs = heap->shrinker.shrink(&heap->shrinker, &sc); + sc.nr_to_scan = objs; + + heap->shrinker.shrink(&heap->shrinker, &sc); + return 0; +} + +static int debug_shrink_get(void *data, u64 *val) +{ + struct ion_heap *heap = data; + struct shrink_control sc; + int objs; + + sc.gfp_mask = -1; + sc.nr_to_scan = 0; + + objs = heap->shrinker.shrink(&heap->shrinker, &sc); + *val = objs; + return 0; +} + +DEFINE_SIMPLE_ATTRIBUTE(debug_shrink_fops, debug_shrink_get, + debug_shrink_set, "%llu\n"); +#endif + +void ion_device_add_heap(struct ion_device *dev, struct ion_heap *heap) +{ + struct dentry *debug_file; + + if (!heap->ops->allocate || !heap->ops->free || !heap->ops->map_dma || + !heap->ops->unmap_dma) + pr_err("%s: can not add heap with invalid ops struct.\n", + __func__); + + if (heap->flags & ION_HEAP_FLAG_DEFER_FREE) + ion_heap_init_deferred_free(heap); + + if ((heap->flags & ION_HEAP_FLAG_DEFER_FREE) || heap->ops->shrink) + ion_heap_init_shrinker(heap); + + heap->dev = dev; + down_write(&dev->lock); + /* use negative heap->id to reverse the priority -- when traversing + the list later attempt higher id numbers first */ + plist_node_init(&heap->node, -heap->id); + plist_add(&heap->node, &dev->heaps); + debug_file = debugfs_create_file(heap->name, 0664, + dev->heaps_debug_root, heap, + &debug_heap_fops); + + if (!debug_file) { + char buf[256], *path; + + path = dentry_path(dev->heaps_debug_root, buf, 256); + pr_err("Failed to create heap debugfs at %s/%s\n", + path, heap->name); + } + +#ifdef DEBUG_HEAP_SHRINKER + if (heap->shrinker.shrink) { + char debug_name[64]; + + snprintf(debug_name, 64, "%s_shrink", heap->name); + debug_file = debugfs_create_file( + debug_name, 0644, dev->heaps_debug_root, heap, + &debug_shrink_fops); + if (!debug_file) { + char buf[256], *path; + + path = dentry_path(dev->heaps_debug_root, buf, 256); + pr_err("Failed to create heap shrinker debugfs at %s/%s\n", + path, debug_name); + } + } +#endif + up_write(&dev->lock); +} + +struct ion_device *ion_device_create(long (*custom_ioctl) + (struct ion_client *client, + unsigned int cmd, + unsigned long arg)) +{ + struct ion_device *idev; + int ret; + + idev = kzalloc(sizeof(struct ion_device), GFP_KERNEL); + if (!idev) + return ERR_PTR(-ENOMEM); + + idev->dev.minor = MISC_DYNAMIC_MINOR; + idev->dev.name = "ion"; + idev->dev.fops = &ion_fops; + idev->dev.parent = NULL; + ret = misc_register(&idev->dev); + if (ret) { + pr_err("ion: failed to register misc device.\n"); + return ERR_PTR(ret); + } + + idev->debug_root = debugfs_create_dir("ion", NULL); + if (!idev->debug_root) { + pr_err("ion: failed to create debugfs root directory.\n"); + goto debugfs_done; + } + idev->heaps_debug_root = debugfs_create_dir("heaps", idev->debug_root); + if (!idev->heaps_debug_root) { + pr_err("ion: failed to create debugfs heaps directory.\n"); + goto debugfs_done; + } + idev->clients_debug_root = debugfs_create_dir("clients", + idev->debug_root); + if (!idev->clients_debug_root) + pr_err("ion: failed to create debugfs clients directory.\n"); + +debugfs_done: + + idev->custom_ioctl = custom_ioctl; + idev->buffers = RB_ROOT; + mutex_init(&idev->buffer_lock); + init_rwsem(&idev->lock); + plist_head_init(&idev->heaps); + idev->clients = RB_ROOT; + return idev; +} + +void ion_device_destroy(struct ion_device *dev) +{ + misc_deregister(&dev->dev); + debugfs_remove_recursive(dev->debug_root); + /* XXX need to free the heaps and clients ? */ + kfree(dev); +} + +void __init ion_reserve(struct ion_platform_data *data) +{ + int i; + + for (i = 0; i < data->nr; i++) { + if (data->heaps[i].size == 0) + continue; + + if (data->heaps[i].base == 0) { + phys_addr_t paddr; + + paddr = memblock_alloc_base(data->heaps[i].size, + data->heaps[i].align, + MEMBLOCK_ALLOC_ANYWHERE); + if (!paddr) { + pr_err("%s: error allocating memblock for " + "heap %d\n", + __func__, i); + continue; + } + data->heaps[i].base = paddr; + } else { + int ret = memblock_reserve(data->heaps[i].base, + data->heaps[i].size); + if (ret) + pr_err("memblock reserve of %zx@%lx failed\n", + data->heaps[i].size, + data->heaps[i].base); + } + pr_info("%s: %s reserved base %lx size %zu\n", __func__, + data->heaps[i].name, + data->heaps[i].base, + data->heaps[i].size); + } +} diff --git a/drivers/staging/android/ion/ion.h b/drivers/staging/android/ion/ion.h new file mode 100644 index 000000000000..dcd2a0cdb192 --- /dev/null +++ b/drivers/staging/android/ion/ion.h @@ -0,0 +1,204 @@ +/* + * drivers/staging/android/ion/ion.h + * + * 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. + * + */ + +#ifndef _LINUX_ION_H +#define _LINUX_ION_H + +#include <linux/types.h> + +#include "../uapi/ion.h" + +struct ion_handle; +struct ion_device; +struct ion_heap; +struct ion_mapper; +struct ion_client; +struct ion_buffer; + +/* This should be removed some day when phys_addr_t's are fully + plumbed in the kernel, and all instances of ion_phys_addr_t should + be converted to phys_addr_t. For the time being many kernel interfaces + do not accept phys_addr_t's that would have to */ +#define ion_phys_addr_t unsigned long + +/** + * struct ion_platform_heap - defines a heap in the given platform + * @type: type of the heap from ion_heap_type enum + * @id: unique identifier for heap. When allocating higher numbers + * will be allocated from first. At allocation these are passed + * as a bit mask and therefore can not exceed ION_NUM_HEAP_IDS. + * @name: used for debug purposes + * @base: base address of heap in physical memory if applicable + * @size: size of the heap in bytes if applicable + * @align: required alignment in physical memory if applicable + * @priv: private info passed from the board file + * + * Provided by the board file. + */ +struct ion_platform_heap { + enum ion_heap_type type; + unsigned int id; + const char *name; + ion_phys_addr_t base; + size_t size; + ion_phys_addr_t align; + void *priv; +}; + +/** + * struct ion_platform_data - array of platform heaps passed from board file + * @nr: number of structures in the array + * @heaps: array of platform_heap structions + * + * Provided by the board file in the form of platform data to a platform device. + */ +struct ion_platform_data { + int nr; + struct ion_platform_heap *heaps; +}; + +/** + * ion_reserve() - reserve memory for ion heaps if applicable + * @data: platform data specifying starting physical address and + * size + * + * Calls memblock reserve to set aside memory for heaps that are + * located at specific memory addresses or of specfic sizes not + * managed by the kernel + */ +void ion_reserve(struct ion_platform_data *data); + +/** + * ion_client_create() - allocate a client and returns it + * @dev: the global ion device + * @heap_type_mask: mask of heaps this client can allocate from + * @name: used for debugging + */ +struct ion_client *ion_client_create(struct ion_device *dev, + const char *name); + +/** + * ion_client_destroy() - free's a client and all it's handles + * @client: the client + * + * Free the provided client and all it's resources including + * any handles it is holding. + */ +void ion_client_destroy(struct ion_client *client); + +/** + * ion_alloc - allocate ion memory + * @client: the client + * @len: size of the allocation + * @align: requested allocation alignment, lots of hardware blocks + * have alignment requirements of some kind + * @heap_id_mask: mask of heaps to allocate from, if multiple bits are set + * heaps will be tried in order from highest to lowest + * id + * @flags: heap flags, the low 16 bits are consumed by ion, the + * high 16 bits are passed on to the respective heap and + * can be heap custom + * + * Allocate memory in one of the heaps provided in heap mask and return + * an opaque handle to it. + */ +struct ion_handle *ion_alloc(struct ion_client *client, size_t len, + size_t align, unsigned int heap_id_mask, + unsigned int flags); + +/** + * ion_free - free a handle + * @client: the client + * @handle: the handle to free + * + * Free the provided handle. + */ +void ion_free(struct ion_client *client, struct ion_handle *handle); + +/** + * ion_phys - returns the physical address and len of a handle + * @client: the client + * @handle: the handle + * @addr: a pointer to put the address in + * @len: a pointer to put the length in + * + * This function queries the heap for a particular handle to get the + * handle's physical address. It't output is only correct if + * a heap returns physically contiguous memory -- in other cases + * this api should not be implemented -- ion_sg_table should be used + * instead. Returns -EINVAL if the handle is invalid. This has + * no implications on the reference counting of the handle -- + * the returned value may not be valid if the caller is not + * holding a reference. + */ +int ion_phys(struct ion_client *client, struct ion_handle *handle, + ion_phys_addr_t *addr, size_t *len); + +/** + * ion_map_dma - return an sg_table describing a handle + * @client: the client + * @handle: the handle + * + * This function returns the sg_table describing + * a particular ion handle. + */ +struct sg_table *ion_sg_table(struct ion_client *client, + struct ion_handle *handle); + +/** + * ion_map_kernel - create mapping for the given handle + * @client: the client + * @handle: handle to map + * + * Map the given handle into the kernel and return a kernel address that + * can be used to access this address. + */ +void *ion_map_kernel(struct ion_client *client, struct ion_handle *handle); + +/** + * ion_unmap_kernel() - destroy a kernel mapping for a handle + * @client: the client + * @handle: handle to unmap + */ +void ion_unmap_kernel(struct ion_client *client, struct ion_handle *handle); + +/** + * ion_share_dma_buf() - share buffer as dma-buf + * @client: the client + * @handle: the handle + */ +struct dma_buf *ion_share_dma_buf(struct ion_client *client, + struct ion_handle *handle); + +/** + * ion_share_dma_buf_fd() - given an ion client, create a dma-buf fd + * @client: the client + * @handle: the handle + */ +int ion_share_dma_buf_fd(struct ion_client *client, struct ion_handle *handle); + +/** + * ion_import_dma_buf() - given an dma-buf fd from the ion exporter get handle + * @client: the client + * @fd: the dma-buf fd + * + * Given an dma-buf fd that was allocated through ion via ion_share_dma_buf, + * import that fd and return a handle representing it. If a dma-buf from + * another exporter is passed in this function will return ERR_PTR(-EINVAL) + */ +struct ion_handle *ion_import_dma_buf(struct ion_client *client, int fd); + +#endif /* _LINUX_ION_H */ diff --git a/drivers/staging/android/ion/ion_carveout_heap.c b/drivers/staging/android/ion/ion_carveout_heap.c new file mode 100644 index 000000000000..5165de2ce343 --- /dev/null +++ b/drivers/staging/android/ion/ion_carveout_heap.c @@ -0,0 +1,194 @@ +/* + * drivers/gpu/ion/ion_carveout_heap.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/spinlock.h> +#include <linux/dma-mapping.h> +#include <linux/err.h> +#include <linux/genalloc.h> +#include <linux/io.h> +#include <linux/mm.h> +#include <linux/scatterlist.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include "ion.h" +#include "ion_priv.h" + +struct ion_carveout_heap { + struct ion_heap heap; + struct gen_pool *pool; + ion_phys_addr_t base; +}; + +ion_phys_addr_t ion_carveout_allocate(struct ion_heap *heap, + unsigned long size, + unsigned long align) +{ + struct ion_carveout_heap *carveout_heap = + container_of(heap, struct ion_carveout_heap, heap); + unsigned long offset = gen_pool_alloc(carveout_heap->pool, size); + + if (!offset) + return ION_CARVEOUT_ALLOCATE_FAIL; + + return offset; +} + +void ion_carveout_free(struct ion_heap *heap, ion_phys_addr_t addr, + unsigned long size) +{ + struct ion_carveout_heap *carveout_heap = + container_of(heap, struct ion_carveout_heap, heap); + + if (addr == ION_CARVEOUT_ALLOCATE_FAIL) + return; + gen_pool_free(carveout_heap->pool, addr, size); +} + +static int ion_carveout_heap_phys(struct ion_heap *heap, + struct ion_buffer *buffer, + ion_phys_addr_t *addr, size_t *len) +{ + struct sg_table *table = buffer->priv_virt; + struct page *page = sg_page(table->sgl); + ion_phys_addr_t paddr = PFN_PHYS(page_to_pfn(page)); + + *addr = paddr; + *len = buffer->size; + return 0; +} + +static int ion_carveout_heap_allocate(struct ion_heap *heap, + struct ion_buffer *buffer, + unsigned long size, unsigned long align, + unsigned long flags) +{ + struct sg_table *table; + ion_phys_addr_t paddr; + int ret; + + if (align > PAGE_SIZE) + return -EINVAL; + + table = kzalloc(sizeof(struct sg_table), GFP_KERNEL); + if (!table) + return -ENOMEM; + ret = sg_alloc_table(table, 1, GFP_KERNEL); + if (ret) + goto err_free; + + paddr = ion_carveout_allocate(heap, size, align); + if (paddr == ION_CARVEOUT_ALLOCATE_FAIL) { + ret = -ENOMEM; + goto err_free_table; + } + + sg_set_page(table->sgl, pfn_to_page(PFN_DOWN(paddr)), size, 0); + buffer->priv_virt = table; + + return 0; + +err_free_table: + sg_free_table(table); +err_free: + kfree(table); + return ret; +} + +static void ion_carveout_heap_free(struct ion_buffer *buffer) +{ + struct ion_heap *heap = buffer->heap; + struct sg_table *table = buffer->priv_virt; + struct page *page = sg_page(table->sgl); + ion_phys_addr_t paddr = PFN_PHYS(page_to_pfn(page)); + + ion_heap_buffer_zero(buffer); + + if (ion_buffer_cached(buffer)) + dma_sync_sg_for_device(NULL, table->sgl, table->nents, + DMA_BIDIRECTIONAL); + + ion_carveout_free(heap, paddr, buffer->size); + sg_free_table(table); + kfree(table); +} + +static struct sg_table *ion_carveout_heap_map_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return buffer->priv_virt; +} + +static void ion_carveout_heap_unmap_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return; +} + +static struct ion_heap_ops carveout_heap_ops = { + .allocate = ion_carveout_heap_allocate, + .free = ion_carveout_heap_free, + .phys = ion_carveout_heap_phys, + .map_dma = ion_carveout_heap_map_dma, + .unmap_dma = ion_carveout_heap_unmap_dma, + .map_user = ion_heap_map_user, + .map_kernel = ion_heap_map_kernel, + .unmap_kernel = ion_heap_unmap_kernel, +}; + +struct ion_heap *ion_carveout_heap_create(struct ion_platform_heap *heap_data) +{ + struct ion_carveout_heap *carveout_heap; + int ret; + + struct page *page; + size_t size; + + page = pfn_to_page(PFN_DOWN(heap_data->base)); + size = heap_data->size; + + ion_pages_sync_for_device(NULL, page, size, DMA_BIDIRECTIONAL); + + ret = ion_heap_pages_zero(page, size, pgprot_writecombine(PAGE_KERNEL)); + if (ret) + return ERR_PTR(ret); + + carveout_heap = kzalloc(sizeof(struct ion_carveout_heap), GFP_KERNEL); + if (!carveout_heap) + return ERR_PTR(-ENOMEM); + + carveout_heap->pool = gen_pool_create(12, -1); + if (!carveout_heap->pool) { + kfree(carveout_heap); + return ERR_PTR(-ENOMEM); + } + carveout_heap->base = heap_data->base; + gen_pool_add(carveout_heap->pool, carveout_heap->base, heap_data->size, + -1); + carveout_heap->heap.ops = &carveout_heap_ops; + carveout_heap->heap.type = ION_HEAP_TYPE_CARVEOUT; + carveout_heap->heap.flags = ION_HEAP_FLAG_DEFER_FREE; + + return &carveout_heap->heap; +} + +void ion_carveout_heap_destroy(struct ion_heap *heap) +{ + struct ion_carveout_heap *carveout_heap = + container_of(heap, struct ion_carveout_heap, heap); + + gen_pool_destroy(carveout_heap->pool); + kfree(carveout_heap); + carveout_heap = NULL; +} diff --git a/drivers/staging/android/ion/ion_chunk_heap.c b/drivers/staging/android/ion/ion_chunk_heap.c new file mode 100644 index 000000000000..ca20d6279602 --- /dev/null +++ b/drivers/staging/android/ion/ion_chunk_heap.c @@ -0,0 +1,195 @@ +/* + * drivers/gpu/ion/ion_chunk_heap.c + * + * 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/dma-mapping.h> +#include <linux/err.h> +#include <linux/genalloc.h> +#include <linux/io.h> +#include <linux/mm.h> +#include <linux/scatterlist.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include "ion.h" +#include "ion_priv.h" + +struct ion_chunk_heap { + struct ion_heap heap; + struct gen_pool *pool; + ion_phys_addr_t base; + unsigned long chunk_size; + unsigned long size; + unsigned long allocated; +}; + +static int ion_chunk_heap_allocate(struct ion_heap *heap, + struct ion_buffer *buffer, + unsigned long size, unsigned long align, + unsigned long flags) +{ + struct ion_chunk_heap *chunk_heap = + container_of(heap, struct ion_chunk_heap, heap); + struct sg_table *table; + struct scatterlist *sg; + int ret, i; + unsigned long num_chunks; + unsigned long allocated_size; + + if (align > chunk_heap->chunk_size) + return -EINVAL; + + allocated_size = ALIGN(size, chunk_heap->chunk_size); + num_chunks = allocated_size / chunk_heap->chunk_size; + + if (allocated_size > chunk_heap->size - chunk_heap->allocated) + return -ENOMEM; + + table = kzalloc(sizeof(struct sg_table), GFP_KERNEL); + if (!table) + return -ENOMEM; + ret = sg_alloc_table(table, num_chunks, GFP_KERNEL); + if (ret) { + kfree(table); + return ret; + } + + sg = table->sgl; + for (i = 0; i < num_chunks; i++) { + unsigned long paddr = gen_pool_alloc(chunk_heap->pool, + chunk_heap->chunk_size); + if (!paddr) + goto err; + sg_set_page(sg, pfn_to_page(PFN_DOWN(paddr)), + chunk_heap->chunk_size, 0); + sg = sg_next(sg); + } + + buffer->priv_virt = table; + chunk_heap->allocated += allocated_size; + return 0; +err: + sg = table->sgl; + for (i -= 1; i >= 0; i--) { + gen_pool_free(chunk_heap->pool, page_to_phys(sg_page(sg)), + sg->length); + sg = sg_next(sg); + } + sg_free_table(table); + kfree(table); + return -ENOMEM; +} + +static void ion_chunk_heap_free(struct ion_buffer *buffer) +{ + struct ion_heap *heap = buffer->heap; + struct ion_chunk_heap *chunk_heap = + container_of(heap, struct ion_chunk_heap, heap); + struct sg_table *table = buffer->priv_virt; + struct scatterlist *sg; + int i; + unsigned long allocated_size; + + allocated_size = ALIGN(buffer->size, chunk_heap->chunk_size); + + ion_heap_buffer_zero(buffer); + + if (ion_buffer_cached(buffer)) + dma_sync_sg_for_device(NULL, table->sgl, table->nents, + DMA_BIDIRECTIONAL); + + for_each_sg(table->sgl, sg, table->nents, i) { + gen_pool_free(chunk_heap->pool, page_to_phys(sg_page(sg)), + sg->length); + } + chunk_heap->allocated -= allocated_size; + sg_free_table(table); + kfree(table); +} + +static struct sg_table *ion_chunk_heap_map_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return buffer->priv_virt; +} + +static void ion_chunk_heap_unmap_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return; +} + +static struct ion_heap_ops chunk_heap_ops = { + .allocate = ion_chunk_heap_allocate, + .free = ion_chunk_heap_free, + .map_dma = ion_chunk_heap_map_dma, + .unmap_dma = ion_chunk_heap_unmap_dma, + .map_user = ion_heap_map_user, + .map_kernel = ion_heap_map_kernel, + .unmap_kernel = ion_heap_unmap_kernel, +}; + +struct ion_heap *ion_chunk_heap_create(struct ion_platform_heap *heap_data) +{ + struct ion_chunk_heap *chunk_heap; + int ret; + struct page *page; + size_t size; + + page = pfn_to_page(PFN_DOWN(heap_data->base)); + size = heap_data->size; + + ion_pages_sync_for_device(NULL, page, size, DMA_BIDIRECTIONAL); + + ret = ion_heap_pages_zero(page, size, pgprot_writecombine(PAGE_KERNEL)); + if (ret) + return ERR_PTR(ret); + + chunk_heap = kzalloc(sizeof(struct ion_chunk_heap), GFP_KERNEL); + if (!chunk_heap) + return ERR_PTR(-ENOMEM); + + chunk_heap->chunk_size = (unsigned long)heap_data->priv; + chunk_heap->pool = gen_pool_create(get_order(chunk_heap->chunk_size) + + PAGE_SHIFT, -1); + if (!chunk_heap->pool) { + ret = -ENOMEM; + goto error_gen_pool_create; + } + chunk_heap->base = heap_data->base; + chunk_heap->size = heap_data->size; + chunk_heap->allocated = 0; + + gen_pool_add(chunk_heap->pool, chunk_heap->base, heap_data->size, -1); + chunk_heap->heap.ops = &chunk_heap_ops; + chunk_heap->heap.type = ION_HEAP_TYPE_CHUNK; + chunk_heap->heap.flags = ION_HEAP_FLAG_DEFER_FREE; + pr_info("%s: base %lu size %zu align %ld\n", __func__, chunk_heap->base, + heap_data->size, heap_data->align); + + return &chunk_heap->heap; + +error_gen_pool_create: + kfree(chunk_heap); + return ERR_PTR(ret); +} + +void ion_chunk_heap_destroy(struct ion_heap *heap) +{ + struct ion_chunk_heap *chunk_heap = + container_of(heap, struct ion_chunk_heap, heap); + + gen_pool_destroy(chunk_heap->pool); + kfree(chunk_heap); + chunk_heap = NULL; +} diff --git a/drivers/staging/android/ion/ion_cma_heap.c b/drivers/staging/android/ion/ion_cma_heap.c new file mode 100644 index 000000000000..4418bda76474 --- /dev/null +++ b/drivers/staging/android/ion/ion_cma_heap.c @@ -0,0 +1,218 @@ +/* + * drivers/gpu/ion/ion_cma_heap.c + * + * Copyright (C) Linaro 2012 + * Author: <benjamin.gaignard@linaro.org> for ST-Ericsson. + * + * 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/slab.h> +#include <linux/errno.h> +#include <linux/err.h> +#include <linux/dma-mapping.h> + +#include "ion.h" +#include "ion_priv.h" + +#define ION_CMA_ALLOCATE_FAILED -1 + +struct ion_cma_heap { + struct ion_heap heap; + struct device *dev; +}; + +#define to_cma_heap(x) container_of(x, struct ion_cma_heap, heap) + +struct ion_cma_buffer_info { + void *cpu_addr; + dma_addr_t handle; + struct sg_table *table; +}; + +/* + * Create scatter-list for the already allocated DMA buffer. + * This function could be replaced by dma_common_get_sgtable + * as soon as it will avalaible. + */ +static int ion_cma_get_sgtable(struct device *dev, struct sg_table *sgt, + void *cpu_addr, dma_addr_t handle, size_t size) +{ + struct page *page = virt_to_page(cpu_addr); + int ret; + + ret = sg_alloc_table(sgt, 1, GFP_KERNEL); + if (unlikely(ret)) + return ret; + + sg_set_page(sgt->sgl, page, PAGE_ALIGN(size), 0); + return 0; +} + +/* ION CMA heap operations functions */ +static int ion_cma_allocate(struct ion_heap *heap, struct ion_buffer *buffer, + unsigned long len, unsigned long align, + unsigned long flags) +{ + struct ion_cma_heap *cma_heap = to_cma_heap(heap); + struct device *dev = cma_heap->dev; + struct ion_cma_buffer_info *info; + + dev_dbg(dev, "Request buffer allocation len %ld\n", len); + + if (buffer->flags & ION_FLAG_CACHED) + return -EINVAL; + + if (align > PAGE_SIZE) + return -EINVAL; + + info = kzalloc(sizeof(struct ion_cma_buffer_info), GFP_KERNEL); + if (!info) { + dev_err(dev, "Can't allocate buffer info\n"); + return ION_CMA_ALLOCATE_FAILED; + } + + info->cpu_addr = dma_alloc_coherent(dev, len, &(info->handle), + GFP_HIGHUSER | __GFP_ZERO); + + if (!info->cpu_addr) { + dev_err(dev, "Fail to allocate buffer\n"); + goto err; + } + + info->table = kmalloc(sizeof(struct sg_table), GFP_KERNEL); + if (!info->table) { + dev_err(dev, "Fail to allocate sg table\n"); + goto free_mem; + } + + if (ion_cma_get_sgtable + (dev, info->table, info->cpu_addr, info->handle, len)) + goto free_table; + /* keep this for memory release */ + buffer->priv_virt = info; + dev_dbg(dev, "Allocate buffer %p\n", buffer); + return 0; + +free_table: + kfree(info->table); +free_mem: + dma_free_coherent(dev, len, info->cpu_addr, info->handle); +err: + kfree(info); + return ION_CMA_ALLOCATE_FAILED; +} + +static void ion_cma_free(struct ion_buffer *buffer) +{ + struct ion_cma_heap *cma_heap = to_cma_heap(buffer->heap); + struct device *dev = cma_heap->dev; + struct ion_cma_buffer_info *info = buffer->priv_virt; + + dev_dbg(dev, "Release buffer %p\n", buffer); + /* release memory */ + dma_free_coherent(dev, buffer->size, info->cpu_addr, info->handle); + /* release sg table */ + sg_free_table(info->table); + kfree(info->table); + kfree(info); +} + +/* return physical address in addr */ +static int ion_cma_phys(struct ion_heap *heap, struct ion_buffer *buffer, + ion_phys_addr_t *addr, size_t *len) +{ + struct ion_cma_heap *cma_heap = to_cma_heap(buffer->heap); + struct device *dev = cma_heap->dev; + struct ion_cma_buffer_info *info = buffer->priv_virt; + + dev_dbg(dev, "Return buffer %p physical address 0x%pa\n", buffer, + &info->handle); + + *addr = info->handle; + *len = buffer->size; + + return 0; +} + +static struct sg_table *ion_cma_heap_map_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + struct ion_cma_buffer_info *info = buffer->priv_virt; + + return info->table; +} + +static void ion_cma_heap_unmap_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return; +} + +static int ion_cma_mmap(struct ion_heap *mapper, struct ion_buffer *buffer, + struct vm_area_struct *vma) +{ + struct ion_cma_heap *cma_heap = to_cma_heap(buffer->heap); + struct device *dev = cma_heap->dev; + struct ion_cma_buffer_info *info = buffer->priv_virt; + + return dma_mmap_coherent(dev, vma, info->cpu_addr, info->handle, + buffer->size); +} + +static void *ion_cma_map_kernel(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + struct ion_cma_buffer_info *info = buffer->priv_virt; + /* kernel memory mapping has been done at allocation time */ + return info->cpu_addr; +} + +static void ion_cma_unmap_kernel(struct ion_heap *heap, + struct ion_buffer *buffer) +{ +} + +static struct ion_heap_ops ion_cma_ops = { + .allocate = ion_cma_allocate, + .free = ion_cma_free, + .map_dma = ion_cma_heap_map_dma, + .unmap_dma = ion_cma_heap_unmap_dma, + .phys = ion_cma_phys, + .map_user = ion_cma_mmap, + .map_kernel = ion_cma_map_kernel, + .unmap_kernel = ion_cma_unmap_kernel, +}; + +struct ion_heap *ion_cma_heap_create(struct ion_platform_heap *data) +{ + struct ion_cma_heap *cma_heap; + + cma_heap = kzalloc(sizeof(struct ion_cma_heap), GFP_KERNEL); + + if (!cma_heap) + return ERR_PTR(-ENOMEM); + + cma_heap->heap.ops = &ion_cma_ops; + /* get device from private heaps data, later it will be + * used to make the link with reserved CMA memory */ + cma_heap->dev = data->priv; + cma_heap->heap.type = ION_HEAP_TYPE_DMA; + return &cma_heap->heap; +} + +void ion_cma_heap_destroy(struct ion_heap *heap) +{ + struct ion_cma_heap *cma_heap = to_cma_heap(heap); + + kfree(cma_heap); +} diff --git a/drivers/staging/android/ion/ion_dummy_driver.c b/drivers/staging/android/ion/ion_dummy_driver.c new file mode 100644 index 000000000000..55b2002753f2 --- /dev/null +++ b/drivers/staging/android/ion/ion_dummy_driver.c @@ -0,0 +1,158 @@ +/* + * drivers/gpu/ion/ion_dummy_driver.c + * + * Copyright (C) 2013 Linaro, 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/err.h> +#include <linux/platform_device.h> +#include <linux/slab.h> +#include <linux/bootmem.h> +#include <linux/memblock.h> +#include <linux/sizes.h> +#include "ion.h" +#include "ion_priv.h" + +struct ion_device *idev; +struct ion_heap **heaps; + +void *carveout_ptr; +void *chunk_ptr; + +struct ion_platform_heap dummy_heaps[] = { + { + .id = ION_HEAP_TYPE_SYSTEM, + .type = ION_HEAP_TYPE_SYSTEM, + .name = "system", + }, + { + .id = ION_HEAP_TYPE_SYSTEM_CONTIG, + .type = ION_HEAP_TYPE_SYSTEM_CONTIG, + .name = "system contig", + }, + { + .id = ION_HEAP_TYPE_CARVEOUT, + .type = ION_HEAP_TYPE_CARVEOUT, + .name = "carveout", + .size = SZ_4M, + }, + { + .id = ION_HEAP_TYPE_CHUNK, + .type = ION_HEAP_TYPE_CHUNK, + .name = "chunk", + .size = SZ_4M, + .align = SZ_16K, + .priv = (void *)(SZ_16K), + }, +}; + +struct ion_platform_data dummy_ion_pdata = { + .nr = 4, + .heaps = dummy_heaps, +}; + +static int __init ion_dummy_init(void) +{ + int i, err; + + idev = ion_device_create(NULL); + heaps = kzalloc(sizeof(struct ion_heap *) * dummy_ion_pdata.nr, + GFP_KERNEL); + if (!heaps) + return PTR_ERR(heaps); + + + /* Allocate a dummy carveout heap */ + carveout_ptr = alloc_pages_exact( + dummy_heaps[ION_HEAP_TYPE_CARVEOUT].size, + GFP_KERNEL); + if (carveout_ptr) + dummy_heaps[ION_HEAP_TYPE_CARVEOUT].base = + virt_to_phys(carveout_ptr); + else + pr_err("ion_dummy: Could not allocate carveout\n"); + + /* Allocate a dummy chunk heap */ + chunk_ptr = alloc_pages_exact( + dummy_heaps[ION_HEAP_TYPE_CHUNK].size, + GFP_KERNEL); + if (chunk_ptr) + dummy_heaps[ION_HEAP_TYPE_CHUNK].base = virt_to_phys(chunk_ptr); + else + pr_err("ion_dummy: Could not allocate chunk\n"); + + for (i = 0; i < dummy_ion_pdata.nr; i++) { + struct ion_platform_heap *heap_data = &dummy_ion_pdata.heaps[i]; + + if (heap_data->type == ION_HEAP_TYPE_CARVEOUT && + !heap_data->base) + continue; + + if (heap_data->type == ION_HEAP_TYPE_CHUNK && !heap_data->base) + continue; + + heaps[i] = ion_heap_create(heap_data); + if (IS_ERR_OR_NULL(heaps[i])) { + err = PTR_ERR(heaps[i]); + goto err; + } + ion_device_add_heap(idev, heaps[i]); + } + return 0; +err: + for (i = 0; i < dummy_ion_pdata.nr; i++) { + if (heaps[i]) + ion_heap_destroy(heaps[i]); + } + kfree(heaps); + + if (carveout_ptr) { + free_pages_exact(carveout_ptr, + dummy_heaps[ION_HEAP_TYPE_CARVEOUT].size); + carveout_ptr = NULL; + } + if (chunk_ptr) { + free_pages_exact(chunk_ptr, + dummy_heaps[ION_HEAP_TYPE_CHUNK].size); + chunk_ptr = NULL; + } + return err; +} + +static void __exit ion_dummy_exit(void) +{ + int i; + + ion_device_destroy(idev); + + for (i = 0; i < dummy_ion_pdata.nr; i++) + ion_heap_destroy(heaps[i]); + kfree(heaps); + + if (carveout_ptr) { + free_pages_exact(carveout_ptr, + dummy_heaps[ION_HEAP_TYPE_CARVEOUT].size); + carveout_ptr = NULL; + } + if (chunk_ptr) { + free_pages_exact(chunk_ptr, + dummy_heaps[ION_HEAP_TYPE_CHUNK].size); + chunk_ptr = NULL; + } + + return; +} + +module_init(ion_dummy_init); +module_exit(ion_dummy_exit); + diff --git a/drivers/staging/android/ion/ion_heap.c b/drivers/staging/android/ion/ion_heap.c new file mode 100644 index 000000000000..e1bbd8b1b4f4 --- /dev/null +++ b/drivers/staging/android/ion/ion_heap.c @@ -0,0 +1,371 @@ +/* + * drivers/gpu/ion/ion_heap.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/err.h> +#include <linux/freezer.h> +#include <linux/kthread.h> +#include <linux/mm.h> +#include <linux/rtmutex.h> +#include <linux/sched.h> +#include <linux/scatterlist.h> +#include <linux/vmalloc.h> +#include "ion.h" +#include "ion_priv.h" + +void *ion_heap_map_kernel(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + struct scatterlist *sg; + int i, j; + void *vaddr; + pgprot_t pgprot; + struct sg_table *table = buffer->sg_table; + int npages = PAGE_ALIGN(buffer->size) / PAGE_SIZE; + struct page **pages = vmalloc(sizeof(struct page *) * npages); + struct page **tmp = pages; + + if (!pages) + return NULL; + + if (buffer->flags & ION_FLAG_CACHED) + pgprot = PAGE_KERNEL; + else + pgprot = pgprot_writecombine(PAGE_KERNEL); + + for_each_sg(table->sgl, sg, table->nents, i) { + int npages_this_entry = PAGE_ALIGN(sg->length) / PAGE_SIZE; + struct page *page = sg_page(sg); + + BUG_ON(i >= npages); + for (j = 0; j < npages_this_entry; j++) + *(tmp++) = page++; + } + vaddr = vmap(pages, npages, VM_MAP, pgprot); + vfree(pages); + + if (vaddr == NULL) + return ERR_PTR(-ENOMEM); + + return vaddr; +} + +void ion_heap_unmap_kernel(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + vunmap(buffer->vaddr); +} + +int ion_heap_map_user(struct ion_heap *heap, struct ion_buffer *buffer, + struct vm_area_struct *vma) +{ + struct sg_table *table = buffer->sg_table; + unsigned long addr = vma->vm_start; + unsigned long offset = vma->vm_pgoff * PAGE_SIZE; + struct scatterlist *sg; + int i; + int ret; + + for_each_sg(table->sgl, sg, table->nents, i) { + struct page *page = sg_page(sg); + unsigned long remainder = vma->vm_end - addr; + unsigned long len = sg->length; + + if (offset >= sg->length) { + offset -= sg->length; + continue; + } else if (offset) { + page += offset / PAGE_SIZE; + len = sg->length - offset; + offset = 0; + } + len = min(len, remainder); + ret = remap_pfn_range(vma, addr, page_to_pfn(page), len, + vma->vm_page_prot); + if (ret) + return ret; + addr += len; + if (addr >= vma->vm_end) + return 0; + } + return 0; +} + +static int ion_heap_clear_pages(struct page **pages, int num, pgprot_t pgprot) +{ + void *addr = vm_map_ram(pages, num, -1, pgprot); + + if (!addr) + return -ENOMEM; + memset(addr, 0, PAGE_SIZE * num); + vm_unmap_ram(addr, num); + + return 0; +} + +static int ion_heap_sglist_zero(struct scatterlist *sgl, unsigned int nents, + pgprot_t pgprot) +{ + int p = 0; + int ret = 0; + struct sg_page_iter piter; + struct page *pages[32]; + + for_each_sg_page(sgl, &piter, nents, 0) { + pages[p++] = sg_page_iter_page(&piter); + if (p == ARRAY_SIZE(pages)) { + ret = ion_heap_clear_pages(pages, p, pgprot); + if (ret) + return ret; + p = 0; + } + } + if (p) + ret = ion_heap_clear_pages(pages, p, pgprot); + + return ret; +} + +int ion_heap_buffer_zero(struct ion_buffer *buffer) +{ + struct sg_table *table = buffer->sg_table; + pgprot_t pgprot; + + if (buffer->flags & ION_FLAG_CACHED) + pgprot = PAGE_KERNEL; + else + pgprot = pgprot_writecombine(PAGE_KERNEL); + + return ion_heap_sglist_zero(table->sgl, table->nents, pgprot); +} + +int ion_heap_pages_zero(struct page *page, size_t size, pgprot_t pgprot) +{ + struct scatterlist sg; + + sg_init_table(&sg, 1); + sg_set_page(&sg, page, size, 0); + return ion_heap_sglist_zero(&sg, 1, pgprot); +} + +void ion_heap_freelist_add(struct ion_heap *heap, struct ion_buffer *buffer) +{ + spin_lock(&heap->free_lock); + list_add(&buffer->list, &heap->free_list); + heap->free_list_size += buffer->size; + spin_unlock(&heap->free_lock); + wake_up(&heap->waitqueue); +} + +size_t ion_heap_freelist_size(struct ion_heap *heap) +{ + size_t size; + + spin_lock(&heap->free_lock); + size = heap->free_list_size; + spin_unlock(&heap->free_lock); + + return size; +} + +static size_t _ion_heap_freelist_drain(struct ion_heap *heap, size_t size, + bool skip_pools) +{ + struct ion_buffer *buffer; + size_t total_drained = 0; + + if (ion_heap_freelist_size(heap) == 0) + return 0; + + spin_lock(&heap->free_lock); + if (size == 0) + size = heap->free_list_size; + + while (!list_empty(&heap->free_list)) { + if (total_drained >= size) + break; + buffer = list_first_entry(&heap->free_list, struct ion_buffer, + list); + list_del(&buffer->list); + heap->free_list_size -= buffer->size; + if (skip_pools) + buffer->private_flags |= ION_PRIV_FLAG_SHRINKER_FREE; + total_drained += buffer->size; + spin_unlock(&heap->free_lock); + ion_buffer_destroy(buffer); + spin_lock(&heap->free_lock); + } + spin_unlock(&heap->free_lock); + + return total_drained; +} + +size_t ion_heap_freelist_drain(struct ion_heap *heap, size_t size) +{ + return _ion_heap_freelist_drain(heap, size, false); +} + +size_t ion_heap_freelist_shrink(struct ion_heap *heap, size_t size) +{ + return _ion_heap_freelist_drain(heap, size, true); +} + +static int ion_heap_deferred_free(void *data) +{ + struct ion_heap *heap = data; + + while (true) { + struct ion_buffer *buffer; + + wait_event_freezable(heap->waitqueue, + ion_heap_freelist_size(heap) > 0); + + spin_lock(&heap->free_lock); + if (list_empty(&heap->free_list)) { + spin_unlock(&heap->free_lock); + continue; + } + buffer = list_first_entry(&heap->free_list, struct ion_buffer, + list); + list_del(&buffer->list); + heap->free_list_size -= buffer->size; + spin_unlock(&heap->free_lock); + ion_buffer_destroy(buffer); + } + + return 0; +} + +int ion_heap_init_deferred_free(struct ion_heap *heap) +{ + struct sched_param param = { .sched_priority = 0 }; + + INIT_LIST_HEAD(&heap->free_list); + heap->free_list_size = 0; + spin_lock_init(&heap->free_lock); + init_waitqueue_head(&heap->waitqueue); + heap->task = kthread_run(ion_heap_deferred_free, heap, + "%s", heap->name); + sched_setscheduler(heap->task, SCHED_IDLE, ¶m); + if (IS_ERR(heap->task)) { + pr_err("%s: creating thread for deferred free failed\n", + __func__); + return PTR_RET(heap->task); + } + return 0; +} + +static int ion_heap_shrink(struct shrinker *shrinker, struct shrink_control *sc) +{ + struct ion_heap *heap = container_of(shrinker, struct ion_heap, + shrinker); + int total = 0; + int freed = 0; + int to_scan = sc->nr_to_scan; + + if (to_scan == 0) + goto out; + + /* + * shrink the free list first, no point in zeroing the memory if we're + * just going to reclaim it. Also, skip any possible page pooling. + */ + if (heap->flags & ION_HEAP_FLAG_DEFER_FREE) + freed = ion_heap_freelist_shrink(heap, to_scan * PAGE_SIZE) / + PAGE_SIZE; + + to_scan -= freed; + if (to_scan < 0) + to_scan = 0; + +out: + total = ion_heap_freelist_size(heap) / PAGE_SIZE; + if (heap->ops->shrink) + total += heap->ops->shrink(heap, sc->gfp_mask, to_scan); + return total; +} + +void ion_heap_init_shrinker(struct ion_heap *heap) +{ + heap->shrinker.shrink = ion_heap_shrink; + heap->shrinker.seeks = DEFAULT_SEEKS; + heap->shrinker.batch = 0; + register_shrinker(&heap->shrinker); +} + +struct ion_heap *ion_heap_create(struct ion_platform_heap *heap_data) +{ + struct ion_heap *heap = NULL; + + switch (heap_data->type) { + case ION_HEAP_TYPE_SYSTEM_CONTIG: + heap = ion_system_contig_heap_create(heap_data); + break; + case ION_HEAP_TYPE_SYSTEM: + heap = ion_system_heap_create(heap_data); + break; + case ION_HEAP_TYPE_CARVEOUT: + heap = ion_carveout_heap_create(heap_data); + break; + case ION_HEAP_TYPE_CHUNK: + heap = ion_chunk_heap_create(heap_data); + break; + case ION_HEAP_TYPE_DMA: + heap = ion_cma_heap_create(heap_data); + break; + default: + pr_err("%s: Invalid heap type %d\n", __func__, + heap_data->type); + return ERR_PTR(-EINVAL); + } + + if (IS_ERR_OR_NULL(heap)) { + pr_err("%s: error creating heap %s type %d base %lu size %zu\n", + __func__, heap_data->name, heap_data->type, + heap_data->base, heap_data->size); + return ERR_PTR(-EINVAL); + } + + heap->name = heap_data->name; + heap->id = heap_data->id; + return heap; +} + +void ion_heap_destroy(struct ion_heap *heap) +{ + if (!heap) + return; + + switch (heap->type) { + case ION_HEAP_TYPE_SYSTEM_CONTIG: + ion_system_contig_heap_destroy(heap); + break; + case ION_HEAP_TYPE_SYSTEM: + ion_system_heap_destroy(heap); + break; + case ION_HEAP_TYPE_CARVEOUT: + ion_carveout_heap_destroy(heap); + break; + case ION_HEAP_TYPE_CHUNK: + ion_chunk_heap_destroy(heap); + break; + case ION_HEAP_TYPE_DMA: + ion_cma_heap_destroy(heap); + break; + default: + pr_err("%s: Invalid heap type %d\n", __func__, + heap->type); + } +} diff --git a/drivers/staging/android/ion/ion_page_pool.c b/drivers/staging/android/ion/ion_page_pool.c new file mode 100644 index 000000000000..0e20e62ec4bc --- /dev/null +++ b/drivers/staging/android/ion/ion_page_pool.c @@ -0,0 +1,190 @@ +/* + * drivers/gpu/ion/ion_mem_pool.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/debugfs.h> +#include <linux/dma-mapping.h> +#include <linux/err.h> +#include <linux/fs.h> +#include <linux/list.h> +#include <linux/module.h> +#include <linux/slab.h> +#include "ion_priv.h" + +struct ion_page_pool_item { + struct page *page; + struct list_head list; +}; + +static void *ion_page_pool_alloc_pages(struct ion_page_pool *pool) +{ + struct page *page = alloc_pages(pool->gfp_mask, pool->order); + + if (!page) + return NULL; + ion_pages_sync_for_device(NULL, page, PAGE_SIZE << pool->order, + DMA_BIDIRECTIONAL); + return page; +} + +static void ion_page_pool_free_pages(struct ion_page_pool *pool, + struct page *page) +{ + __free_pages(page, pool->order); +} + +static int ion_page_pool_add(struct ion_page_pool *pool, struct page *page) +{ + struct ion_page_pool_item *item; + + item = kmalloc(sizeof(struct ion_page_pool_item), GFP_KERNEL); + if (!item) + return -ENOMEM; + + mutex_lock(&pool->mutex); + item->page = page; + if (PageHighMem(page)) { + list_add_tail(&item->list, &pool->high_items); + pool->high_count++; + } else { + list_add_tail(&item->list, &pool->low_items); + pool->low_count++; + } + mutex_unlock(&pool->mutex); + return 0; +} + +static struct page *ion_page_pool_remove(struct ion_page_pool *pool, bool high) +{ + struct ion_page_pool_item *item; + struct page *page; + + if (high) { + BUG_ON(!pool->high_count); + item = list_first_entry(&pool->high_items, + struct ion_page_pool_item, list); + pool->high_count--; + } else { + BUG_ON(!pool->low_count); + item = list_first_entry(&pool->low_items, + struct ion_page_pool_item, list); + pool->low_count--; + } + + list_del(&item->list); + page = item->page; + kfree(item); + return page; +} + +void *ion_page_pool_alloc(struct ion_page_pool *pool) +{ + struct page *page = NULL; + + BUG_ON(!pool); + + mutex_lock(&pool->mutex); + if (pool->high_count) + page = ion_page_pool_remove(pool, true); + else if (pool->low_count) + page = ion_page_pool_remove(pool, false); + mutex_unlock(&pool->mutex); + + if (!page) + page = ion_page_pool_alloc_pages(pool); + + return page; +} + +void ion_page_pool_free(struct ion_page_pool *pool, struct page *page) +{ + int ret; + + ret = ion_page_pool_add(pool, page); + if (ret) + ion_page_pool_free_pages(pool, page); +} + +static int ion_page_pool_total(struct ion_page_pool *pool, bool high) +{ + int total = 0; + + total += high ? (pool->high_count + pool->low_count) * + (1 << pool->order) : + pool->low_count * (1 << pool->order); + return total; +} + +int ion_page_pool_shrink(struct ion_page_pool *pool, gfp_t gfp_mask, + int nr_to_scan) +{ + int i; + bool high; + + high = !!(gfp_mask & __GFP_HIGHMEM); + + for (i = 0; i < nr_to_scan; i++) { + struct page *page; + + mutex_lock(&pool->mutex); + if (pool->low_count) { + page = ion_page_pool_remove(pool, false); + } else if (high && pool->high_count) { + page = ion_page_pool_remove(pool, true); + } else { + mutex_unlock(&pool->mutex); + break; + } + mutex_unlock(&pool->mutex); + ion_page_pool_free_pages(pool, page); + } + + return ion_page_pool_total(pool, high); +} + +struct ion_page_pool *ion_page_pool_create(gfp_t gfp_mask, unsigned int order) +{ + struct ion_page_pool *pool = kmalloc(sizeof(struct ion_page_pool), + GFP_KERNEL); + if (!pool) + return NULL; + pool->high_count = 0; + pool->low_count = 0; + INIT_LIST_HEAD(&pool->low_items); + INIT_LIST_HEAD(&pool->high_items); + pool->gfp_mask = gfp_mask; + pool->order = order; + mutex_init(&pool->mutex); + plist_node_init(&pool->list, order); + + return pool; +} + +void ion_page_pool_destroy(struct ion_page_pool *pool) +{ + kfree(pool); +} + +static int __init ion_page_pool_init(void) +{ + return 0; +} + +static void __exit ion_page_pool_exit(void) +{ +} + +module_init(ion_page_pool_init); +module_exit(ion_page_pool_exit); diff --git a/drivers/staging/android/ion/ion_priv.h b/drivers/staging/android/ion/ion_priv.h new file mode 100644 index 000000000000..8871f35f384b --- /dev/null +++ b/drivers/staging/android/ion/ion_priv.h @@ -0,0 +1,406 @@ +/* + * drivers/gpu/ion/ion_priv.h + * + * 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. + * + */ + +#ifndef _ION_PRIV_H +#define _ION_PRIV_H + +#include <linux/device.h> +#include <linux/dma-direction.h> +#include <linux/kref.h> +#include <linux/mm_types.h> +#include <linux/mutex.h> +#include <linux/rbtree.h> +#include <linux/sched.h> +#include <linux/shrinker.h> +#include <linux/types.h> +#include <linux/device.h> + +#include "ion.h" + +struct ion_buffer *ion_handle_buffer(struct ion_handle *handle); + +/** + * struct ion_buffer - metadata for a particular buffer + * @ref: refernce count + * @node: node in the ion_device buffers tree + * @dev: back pointer to the ion_device + * @heap: back pointer to the heap the buffer came from + * @flags: buffer specific flags + * @private_flags: internal buffer specific flags + * @size: size of the buffer + * @priv_virt: private data to the buffer representable as + * a void * + * @priv_phys: private data to the buffer representable as + * an ion_phys_addr_t (and someday a phys_addr_t) + * @lock: protects the buffers cnt fields + * @kmap_cnt: number of times the buffer is mapped to the kernel + * @vaddr: the kenrel mapping if kmap_cnt is not zero + * @dmap_cnt: number of times the buffer is mapped for dma + * @sg_table: the sg table for the buffer if dmap_cnt is not zero + * @pages: flat array of pages in the buffer -- used by fault + * handler and only valid for buffers that are faulted in + * @vmas: list of vma's mapping this buffer + * @handle_count: count of handles referencing this buffer + * @task_comm: taskcomm of last client to reference this buffer in a + * handle, used for debugging + * @pid: pid of last client to reference this buffer in a + * handle, used for debugging +*/ +struct ion_buffer { + struct kref ref; + union { + struct rb_node node; + struct list_head list; + }; + struct ion_device *dev; + struct ion_heap *heap; + unsigned long flags; + unsigned long private_flags; + size_t size; + union { + void *priv_virt; + ion_phys_addr_t priv_phys; + }; + struct mutex lock; + int kmap_cnt; + void *vaddr; + int dmap_cnt; + struct sg_table *sg_table; + struct page **pages; + struct list_head vmas; + /* used to track orphaned buffers */ + int handle_count; + char task_comm[TASK_COMM_LEN]; + pid_t pid; +}; +void ion_buffer_destroy(struct ion_buffer *buffer); + +/** + * struct ion_heap_ops - ops to operate on a given heap + * @allocate: allocate memory + * @free: free memory + * @phys get physical address of a buffer (only define on + * physically contiguous heaps) + * @map_dma map the memory for dma to a scatterlist + * @unmap_dma unmap the memory for dma + * @map_kernel map memory to the kernel + * @unmap_kernel unmap memory to the kernel + * @map_user map memory to userspace + * + * allocate, phys, and map_user return 0 on success, -errno on error. + * map_dma and map_kernel return pointer on success, ERR_PTR on + * error. @free will be called with ION_PRIV_FLAG_SHRINKER_FREE set in + * the buffer's private_flags when called from a shrinker. In that + * case, the pages being free'd must be truly free'd back to the + * system, not put in a page pool or otherwise cached. + */ +struct ion_heap_ops { + int (*allocate) (struct ion_heap *heap, + struct ion_buffer *buffer, unsigned long len, + unsigned long align, unsigned long flags); + void (*free) (struct ion_buffer *buffer); + int (*phys) (struct ion_heap *heap, struct ion_buffer *buffer, + ion_phys_addr_t *addr, size_t *len); + struct sg_table *(*map_dma) (struct ion_heap *heap, + struct ion_buffer *buffer); + void (*unmap_dma) (struct ion_heap *heap, struct ion_buffer *buffer); + void * (*map_kernel) (struct ion_heap *heap, struct ion_buffer *buffer); + void (*unmap_kernel) (struct ion_heap *heap, struct ion_buffer *buffer); + int (*map_user) (struct ion_heap *mapper, struct ion_buffer *buffer, + struct vm_area_struct *vma); + int (*shrink)(struct ion_heap *heap, gfp_t gfp_mask, int nr_to_scan); +}; + +/** + * heap flags - flags between the heaps and core ion code + */ +#define ION_HEAP_FLAG_DEFER_FREE (1 << 0) + +/** + * private flags - flags internal to ion + */ +/* + * Buffer is being freed from a shrinker function. Skip any possible + * heap-specific caching mechanism (e.g. page pools). Guarantees that + * any buffer storage that came from the system allocator will be + * returned to the system allocator. + */ +#define ION_PRIV_FLAG_SHRINKER_FREE (1 << 0) + +/** + * struct ion_heap - represents a heap in the system + * @node: rb node to put the heap on the device's tree of heaps + * @dev: back pointer to the ion_device + * @type: type of heap + * @ops: ops struct as above + * @flags: flags + * @id: id of heap, also indicates priority of this heap when + * allocating. These are specified by platform data and + * MUST be unique + * @name: used for debugging + * @shrinker: a shrinker for the heap + * @free_list: free list head if deferred free is used + * @free_list_size size of the deferred free list in bytes + * @lock: protects the free list + * @waitqueue: queue to wait on from deferred free thread + * @task: task struct of deferred free thread + * @debug_show: called when heap debug file is read to add any + * heap specific debug info to output + * + * Represents a pool of memory from which buffers can be made. In some + * systems the only heap is regular system memory allocated via vmalloc. + * On others, some blocks might require large physically contiguous buffers + * that are allocated from a specially reserved heap. + */ +struct ion_heap { + struct plist_node node; + struct ion_device *dev; + enum ion_heap_type type; + struct ion_heap_ops *ops; + unsigned long flags; + unsigned int id; + const char *name; + struct shrinker shrinker; + struct list_head free_list; + size_t free_list_size; + spinlock_t free_lock; + wait_queue_head_t waitqueue; + struct task_struct *task; + + int (*debug_show)(struct ion_heap *heap, struct seq_file *, void *); +}; + +/** + * ion_buffer_cached - this ion buffer is cached + * @buffer: buffer + * + * indicates whether this ion buffer is cached + */ +bool ion_buffer_cached(struct ion_buffer *buffer); + +/** + * ion_buffer_fault_user_mappings - fault in user mappings of this buffer + * @buffer: buffer + * + * indicates whether userspace mappings of this buffer will be faulted + * in, this can affect how buffers are allocated from the heap. + */ +bool ion_buffer_fault_user_mappings(struct ion_buffer *buffer); + +/** + * ion_device_create - allocates and returns an ion device + * @custom_ioctl: arch specific ioctl function if applicable + * + * returns a valid device or -PTR_ERR + */ +struct ion_device *ion_device_create(long (*custom_ioctl) + (struct ion_client *client, + unsigned int cmd, + unsigned long arg)); + +/** + * ion_device_destroy - free and device and it's resource + * @dev: the device + */ +void ion_device_destroy(struct ion_device *dev); + +/** + * ion_device_add_heap - adds a heap to the ion device + * @dev: the device + * @heap: the heap to add + */ +void ion_device_add_heap(struct ion_device *dev, struct ion_heap *heap); + +/** + * some helpers for common operations on buffers using the sg_table + * and vaddr fields + */ +void *ion_heap_map_kernel(struct ion_heap *, struct ion_buffer *); +void ion_heap_unmap_kernel(struct ion_heap *, struct ion_buffer *); +int ion_heap_map_user(struct ion_heap *, struct ion_buffer *, + struct vm_area_struct *); +int ion_heap_buffer_zero(struct ion_buffer *buffer); +int ion_heap_pages_zero(struct page *page, size_t size, pgprot_t pgprot); + +/** + * ion_heap_init_shrinker + * @heap: the heap + * + * If a heap sets the ION_HEAP_FLAG_DEFER_FREE flag or defines the shrink op + * this function will be called to setup a shrinker to shrink the freelists + * and call the heap's shrink op. + */ +void ion_heap_init_shrinker(struct ion_heap *heap); + +/** + * ion_heap_init_deferred_free -- initialize deferred free functionality + * @heap: the heap + * + * If a heap sets the ION_HEAP_FLAG_DEFER_FREE flag this function will + * be called to setup deferred frees. Calls to free the buffer will + * return immediately and the actual free will occur some time later + */ +int ion_heap_init_deferred_free(struct ion_heap *heap); + +/** + * ion_heap_freelist_add - add a buffer to the deferred free list + * @heap: the heap + * @buffer: the buffer + * + * Adds an item to the deferred freelist. + */ +void ion_heap_freelist_add(struct ion_heap *heap, struct ion_buffer *buffer); + +/** + * ion_heap_freelist_drain - drain the deferred free list + * @heap: the heap + * @size: ammount of memory to drain in bytes + * + * Drains the indicated amount of memory from the deferred freelist immediately. + * Returns the total amount freed. The total freed may be higher depending + * on the size of the items in the list, or lower if there is insufficient + * total memory on the freelist. + */ +size_t ion_heap_freelist_drain(struct ion_heap *heap, size_t size); + +/** + * ion_heap_freelist_shrink - drain the deferred free + * list, skipping any heap-specific + * pooling or caching mechanisms + * + * @heap: the heap + * @size: amount of memory to drain in bytes + * + * Drains the indicated amount of memory from the deferred freelist immediately. + * Returns the total amount freed. The total freed may be higher depending + * on the size of the items in the list, or lower if there is insufficient + * total memory on the freelist. + * + * Unlike with @ion_heap_freelist_drain, don't put any pages back into + * page pools or otherwise cache the pages. Everything must be + * genuinely free'd back to the system. If you're free'ing from a + * shrinker you probably want to use this. Note that this relies on + * the heap.ops.free callback honoring the ION_PRIV_FLAG_SHRINKER_FREE + * flag. + */ +size_t ion_heap_freelist_shrink(struct ion_heap *heap, + size_t size); + +/** + * ion_heap_freelist_size - returns the size of the freelist in bytes + * @heap: the heap + */ +size_t ion_heap_freelist_size(struct ion_heap *heap); + + +/** + * functions for creating and destroying the built in ion heaps. + * architectures can add their own custom architecture specific + * heaps as appropriate. + */ + +struct ion_heap *ion_heap_create(struct ion_platform_heap *); +void ion_heap_destroy(struct ion_heap *); +struct ion_heap *ion_system_heap_create(struct ion_platform_heap *); +void ion_system_heap_destroy(struct ion_heap *); + +struct ion_heap *ion_system_contig_heap_create(struct ion_platform_heap *); +void ion_system_contig_heap_destroy(struct ion_heap *); + +struct ion_heap *ion_carveout_heap_create(struct ion_platform_heap *); +void ion_carveout_heap_destroy(struct ion_heap *); + +struct ion_heap *ion_chunk_heap_create(struct ion_platform_heap *); +void ion_chunk_heap_destroy(struct ion_heap *); +struct ion_heap *ion_cma_heap_create(struct ion_platform_heap *); +void ion_cma_heap_destroy(struct ion_heap *); + +/** + * kernel api to allocate/free from carveout -- used when carveout is + * used to back an architecture specific custom heap + */ +ion_phys_addr_t ion_carveout_allocate(struct ion_heap *heap, unsigned long size, + unsigned long align); +void ion_carveout_free(struct ion_heap *heap, ion_phys_addr_t addr, + unsigned long size); +/** + * The carveout heap returns physical addresses, since 0 may be a valid + * physical address, this is used to indicate allocation failed + */ +#define ION_CARVEOUT_ALLOCATE_FAIL -1 + +/** + * functions for creating and destroying a heap pool -- allows you + * to keep a pool of pre allocated memory to use from your heap. Keeping + * a pool of memory that is ready for dma, ie any cached mapping have been + * invalidated from the cache, provides a significant peformance benefit on + * many systems */ + +/** + * struct ion_page_pool - pagepool struct + * @high_count: number of highmem items in the pool + * @low_count: number of lowmem items in the pool + * @high_items: list of highmem items + * @low_items: list of lowmem items + * @mutex: lock protecting this struct and especially the count + * item list + * @gfp_mask: gfp_mask to use from alloc + * @order: order of pages in the pool + * @list: plist node for list of pools + * + * Allows you to keep a pool of pre allocated pages to use from your heap. + * Keeping a pool of pages that is ready for dma, ie any cached mapping have + * been invalidated from the cache, provides a significant peformance benefit + * on many systems + */ +struct ion_page_pool { + int high_count; + int low_count; + struct list_head high_items; + struct list_head low_items; + struct mutex mutex; + gfp_t gfp_mask; + unsigned int order; + struct plist_node list; +}; + +struct ion_page_pool *ion_page_pool_create(gfp_t gfp_mask, unsigned int order); +void ion_page_pool_destroy(struct ion_page_pool *); +void *ion_page_pool_alloc(struct ion_page_pool *); +void ion_page_pool_free(struct ion_page_pool *, struct page *); + +/** ion_page_pool_shrink - shrinks the size of the memory cached in the pool + * @pool: the pool + * @gfp_mask: the memory type to reclaim + * @nr_to_scan: number of items to shrink in pages + * + * returns the number of items freed in pages + */ +int ion_page_pool_shrink(struct ion_page_pool *pool, gfp_t gfp_mask, + int nr_to_scan); + +/** + * ion_pages_sync_for_device - cache flush pages for use with the specified + * device + * @dev: the device the pages will be used with + * @page: the first page to be flushed + * @size: size in bytes of region to be flushed + * @dir: direction of dma transfer + */ +void ion_pages_sync_for_device(struct device *dev, struct page *page, + size_t size, enum dma_data_direction dir); + +#endif /* _ION_PRIV_H */ diff --git a/drivers/staging/android/ion/ion_system_heap.c b/drivers/staging/android/ion/ion_system_heap.c new file mode 100644 index 000000000000..5945445ee5f3 --- /dev/null +++ b/drivers/staging/android/ion/ion_system_heap.c @@ -0,0 +1,451 @@ +/* + * drivers/gpu/ion/ion_system_heap.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 <asm/page.h> +#include <linux/dma-mapping.h> +#include <linux/err.h> +#include <linux/highmem.h> +#include <linux/mm.h> +#include <linux/scatterlist.h> +#include <linux/seq_file.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include "ion.h" +#include "ion_priv.h" + +static gfp_t high_order_gfp_flags = (GFP_HIGHUSER | __GFP_ZERO | __GFP_NOWARN | + __GFP_NORETRY) & ~__GFP_WAIT; +static gfp_t low_order_gfp_flags = (GFP_HIGHUSER | __GFP_ZERO | __GFP_NOWARN); +static const unsigned int orders[] = {8, 4, 0}; +static const int num_orders = ARRAY_SIZE(orders); +static int order_to_index(unsigned int order) +{ + int i; + + for (i = 0; i < num_orders; i++) + if (order == orders[i]) + return i; + BUG(); + return -1; +} + +static unsigned int order_to_size(int order) +{ + return PAGE_SIZE << order; +} + +struct ion_system_heap { + struct ion_heap heap; + struct ion_page_pool **pools; +}; + +struct page_info { + struct page *page; + unsigned int order; + struct list_head list; +}; + +static struct page *alloc_buffer_page(struct ion_system_heap *heap, + struct ion_buffer *buffer, + unsigned long order) +{ + bool cached = ion_buffer_cached(buffer); + struct ion_page_pool *pool = heap->pools[order_to_index(order)]; + struct page *page; + + if (!cached) { + page = ion_page_pool_alloc(pool); + } else { + gfp_t gfp_flags = low_order_gfp_flags; + + if (order > 4) + gfp_flags = high_order_gfp_flags; + page = alloc_pages(gfp_flags, order); + if (!page) + return NULL; + ion_pages_sync_for_device(NULL, page, PAGE_SIZE << order, + DMA_BIDIRECTIONAL); + } + if (!page) + return NULL; + + return page; +} + +static void free_buffer_page(struct ion_system_heap *heap, + struct ion_buffer *buffer, struct page *page, + unsigned int order) +{ + bool cached = ion_buffer_cached(buffer); + + if (!cached && !(buffer->private_flags & ION_PRIV_FLAG_SHRINKER_FREE)) { + struct ion_page_pool *pool = heap->pools[order_to_index(order)]; + + ion_page_pool_free(pool, page); + } else { + __free_pages(page, order); + } +} + + +static struct page_info *alloc_largest_available(struct ion_system_heap *heap, + struct ion_buffer *buffer, + unsigned long size, + unsigned int max_order) +{ + struct page *page; + struct page_info *info; + int i; + + info = kmalloc(sizeof(struct page_info), GFP_KERNEL); + if (!info) + return NULL; + + for (i = 0; i < num_orders; i++) { + if (size < order_to_size(orders[i])) + continue; + if (max_order < orders[i]) + continue; + + page = alloc_buffer_page(heap, buffer, orders[i]); + if (!page) + continue; + + info->page = page; + info->order = orders[i]; + INIT_LIST_HEAD(&info->list); + return info; + } + kfree(info); + + return NULL; +} + +static int ion_system_heap_allocate(struct ion_heap *heap, + struct ion_buffer *buffer, + unsigned long size, unsigned long align, + unsigned long flags) +{ + struct ion_system_heap *sys_heap = container_of(heap, + struct ion_system_heap, + heap); + struct sg_table *table; + struct scatterlist *sg; + int ret; + struct list_head pages; + struct page_info *info, *tmp_info; + int i = 0; + unsigned long size_remaining = PAGE_ALIGN(size); + unsigned int max_order = orders[0]; + + if (align > PAGE_SIZE) + return -EINVAL; + + if (size / PAGE_SIZE > totalram_pages / 2) + return -ENOMEM; + + INIT_LIST_HEAD(&pages); + while (size_remaining > 0) { + info = alloc_largest_available(sys_heap, buffer, size_remaining, + max_order); + if (!info) + goto err; + list_add_tail(&info->list, &pages); + size_remaining -= (1 << info->order) * PAGE_SIZE; + max_order = info->order; + i++; + } + table = kzalloc(sizeof(struct sg_table), GFP_KERNEL); + if (!table) + goto err; + + ret = sg_alloc_table(table, i, GFP_KERNEL); + if (ret) + goto err1; + + sg = table->sgl; + list_for_each_entry_safe(info, tmp_info, &pages, list) { + struct page *page = info->page; + sg_set_page(sg, page, (1 << info->order) * PAGE_SIZE, 0); + sg = sg_next(sg); + list_del(&info->list); + kfree(info); + } + + buffer->priv_virt = table; + return 0; +err1: + kfree(table); +err: + list_for_each_entry_safe(info, tmp_info, &pages, list) { + free_buffer_page(sys_heap, buffer, info->page, info->order); + kfree(info); + } + return -ENOMEM; +} + +static void ion_system_heap_free(struct ion_buffer *buffer) +{ + struct ion_heap *heap = buffer->heap; + struct ion_system_heap *sys_heap = container_of(heap, + struct ion_system_heap, + heap); + struct sg_table *table = buffer->sg_table; + bool cached = ion_buffer_cached(buffer); + struct scatterlist *sg; + LIST_HEAD(pages); + int i; + + /* uncached pages come from the page pools, zero them before returning + for security purposes (other allocations are zerod at alloc time */ + if (!cached && !(buffer->private_flags & ION_PRIV_FLAG_SHRINKER_FREE)) + ion_heap_buffer_zero(buffer); + + for_each_sg(table->sgl, sg, table->nents, i) + free_buffer_page(sys_heap, buffer, sg_page(sg), + get_order(sg->length)); + sg_free_table(table); + kfree(table); +} + +static struct sg_table *ion_system_heap_map_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return buffer->priv_virt; +} + +static void ion_system_heap_unmap_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return; +} + +static int ion_system_heap_shrink(struct ion_heap *heap, gfp_t gfp_mask, + int nr_to_scan) +{ + struct ion_system_heap *sys_heap; + int nr_total = 0; + int i; + + sys_heap = container_of(heap, struct ion_system_heap, heap); + + for (i = 0; i < num_orders; i++) { + struct ion_page_pool *pool = sys_heap->pools[i]; + + nr_total += ion_page_pool_shrink(pool, gfp_mask, nr_to_scan); + } + + return nr_total; +} + +static struct ion_heap_ops system_heap_ops = { + .allocate = ion_system_heap_allocate, + .free = ion_system_heap_free, + .map_dma = ion_system_heap_map_dma, + .unmap_dma = ion_system_heap_unmap_dma, + .map_kernel = ion_heap_map_kernel, + .unmap_kernel = ion_heap_unmap_kernel, + .map_user = ion_heap_map_user, + .shrink = ion_system_heap_shrink, +}; + +static int ion_system_heap_debug_show(struct ion_heap *heap, struct seq_file *s, + void *unused) +{ + + struct ion_system_heap *sys_heap = container_of(heap, + struct ion_system_heap, + heap); + int i; + + for (i = 0; i < num_orders; i++) { + struct ion_page_pool *pool = sys_heap->pools[i]; + + seq_printf(s, "%d order %u highmem pages in pool = %lu total\n", + pool->high_count, pool->order, + (1 << pool->order) * PAGE_SIZE * pool->high_count); + seq_printf(s, "%d order %u lowmem pages in pool = %lu total\n", + pool->low_count, pool->order, + (1 << pool->order) * PAGE_SIZE * pool->low_count); + } + return 0; +} + +struct ion_heap *ion_system_heap_create(struct ion_platform_heap *unused) +{ + struct ion_system_heap *heap; + int i; + + heap = kzalloc(sizeof(struct ion_system_heap), GFP_KERNEL); + if (!heap) + return ERR_PTR(-ENOMEM); + heap->heap.ops = &system_heap_ops; + heap->heap.type = ION_HEAP_TYPE_SYSTEM; + heap->heap.flags = ION_HEAP_FLAG_DEFER_FREE; + heap->pools = kzalloc(sizeof(struct ion_page_pool *) * num_orders, + GFP_KERNEL); + if (!heap->pools) + goto err_alloc_pools; + for (i = 0; i < num_orders; i++) { + struct ion_page_pool *pool; + gfp_t gfp_flags = low_order_gfp_flags; + + if (orders[i] > 4) + gfp_flags = high_order_gfp_flags; + pool = ion_page_pool_create(gfp_flags, orders[i]); + if (!pool) + goto err_create_pool; + heap->pools[i] = pool; + } + + heap->heap.debug_show = ion_system_heap_debug_show; + return &heap->heap; +err_create_pool: + for (i = 0; i < num_orders; i++) + if (heap->pools[i]) + ion_page_pool_destroy(heap->pools[i]); + kfree(heap->pools); +err_alloc_pools: + kfree(heap); + return ERR_PTR(-ENOMEM); +} + +void ion_system_heap_destroy(struct ion_heap *heap) +{ + struct ion_system_heap *sys_heap = container_of(heap, + struct ion_system_heap, + heap); + int i; + + for (i = 0; i < num_orders; i++) + ion_page_pool_destroy(sys_heap->pools[i]); + kfree(sys_heap->pools); + kfree(sys_heap); +} + +static int ion_system_contig_heap_allocate(struct ion_heap *heap, + struct ion_buffer *buffer, + unsigned long len, + unsigned long align, + unsigned long flags) +{ + int order = get_order(len); + struct page *page; + struct sg_table *table; + unsigned long i; + int ret; + + if (align > (PAGE_SIZE << order)) + return -EINVAL; + + page = alloc_pages(low_order_gfp_flags, order); + if (!page) + return -ENOMEM; + + split_page(page, order); + + len = PAGE_ALIGN(len); + for (i = len >> PAGE_SHIFT; i < (1 << order); i++) + __free_page(page + i); + + table = kzalloc(sizeof(struct sg_table), GFP_KERNEL); + if (!table) { + ret = -ENOMEM; + goto out; + } + + ret = sg_alloc_table(table, 1, GFP_KERNEL); + if (ret) + goto out; + + sg_set_page(table->sgl, page, len, 0); + + buffer->priv_virt = table; + + ion_pages_sync_for_device(NULL, page, len, DMA_BIDIRECTIONAL); + + return 0; + +out: + for (i = 0; i < len >> PAGE_SHIFT; i++) + __free_page(page + i); + kfree(table); + return ret; +} + +static void ion_system_contig_heap_free(struct ion_buffer *buffer) +{ + struct sg_table *table = buffer->priv_virt; + struct page *page = sg_page(table->sgl); + unsigned long pages = PAGE_ALIGN(buffer->size) >> PAGE_SHIFT; + unsigned long i; + + for (i = 0; i < pages; i++) + __free_page(page + i); + sg_free_table(table); + kfree(table); +} + +static int ion_system_contig_heap_phys(struct ion_heap *heap, + struct ion_buffer *buffer, + ion_phys_addr_t *addr, size_t *len) +{ + struct sg_table *table = buffer->priv_virt; + struct page *page = sg_page(table->sgl); + *addr = page_to_phys(page); + *len = buffer->size; + return 0; +} + +static struct sg_table *ion_system_contig_heap_map_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ + return buffer->priv_virt; +} + +static void ion_system_contig_heap_unmap_dma(struct ion_heap *heap, + struct ion_buffer *buffer) +{ +} + +static struct ion_heap_ops kmalloc_ops = { + .allocate = ion_system_contig_heap_allocate, + .free = ion_system_contig_heap_free, + .phys = ion_system_contig_heap_phys, + .map_dma = ion_system_contig_heap_map_dma, + .unmap_dma = ion_system_contig_heap_unmap_dma, + .map_kernel = ion_heap_map_kernel, + .unmap_kernel = ion_heap_unmap_kernel, + .map_user = ion_heap_map_user, +}; + +struct ion_heap *ion_system_contig_heap_create(struct ion_platform_heap *unused) +{ + struct ion_heap *heap; + + heap = kzalloc(sizeof(struct ion_heap), GFP_KERNEL); + if (!heap) + return ERR_PTR(-ENOMEM); + heap->ops = &kmalloc_ops; + heap->type = ION_HEAP_TYPE_SYSTEM_CONTIG; + return heap; +} + +void ion_system_contig_heap_destroy(struct ion_heap *heap) +{ + kfree(heap); +} + diff --git a/drivers/staging/android/ion/ion_test.c b/drivers/staging/android/ion/ion_test.c new file mode 100644 index 000000000000..654acb5c8eba --- /dev/null +++ b/drivers/staging/android/ion/ion_test.c @@ -0,0 +1,282 @@ +/* + * + * 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. + * + */ + +#define pr_fmt(fmt) "ion-test: " fmt + +#include <linux/dma-buf.h> +#include <linux/dma-direction.h> +#include <linux/fs.h> +#include <linux/miscdevice.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/sched.h> +#include <linux/slab.h> +#include <linux/uaccess.h> +#include <linux/vmalloc.h> + +#include "ion.h" +#include "../uapi/ion_test.h" + +#define u64_to_uptr(x) ((void __user *)(unsigned long)(x)) + +struct ion_test_device { + struct miscdevice misc; +}; + +struct ion_test_data { + struct dma_buf *dma_buf; + struct device *dev; +}; + +static int ion_handle_test_dma(struct device *dev, struct dma_buf *dma_buf, + void __user *ptr, size_t offset, size_t size, bool write) +{ + int ret = 0; + struct dma_buf_attachment *attach; + struct sg_table *table; + pgprot_t pgprot = pgprot_writecombine(PAGE_KERNEL); + enum dma_data_direction dir = write ? DMA_FROM_DEVICE : DMA_TO_DEVICE; + struct sg_page_iter sg_iter; + unsigned long offset_page; + + attach = dma_buf_attach(dma_buf, dev); + if (IS_ERR(attach)) + return PTR_ERR(attach); + + table = dma_buf_map_attachment(attach, dir); + if (IS_ERR(table)) + return PTR_ERR(table); + + offset_page = offset >> PAGE_SHIFT; + offset %= PAGE_SIZE; + + for_each_sg_page(table->sgl, &sg_iter, table->nents, offset_page) { + struct page *page = sg_page_iter_page(&sg_iter); + void *vaddr = vmap(&page, 1, VM_MAP, pgprot); + size_t to_copy = PAGE_SIZE - offset; + + to_copy = min(to_copy, size); + if (!vaddr) { + ret = -ENOMEM; + goto err; + } + + if (write) + ret = copy_from_user(vaddr + offset, ptr, to_copy); + else + ret = copy_to_user(ptr, vaddr + offset, to_copy); + + vunmap(vaddr); + if (ret) { + ret = -EFAULT; + goto err; + } + size -= to_copy; + if (!size) + break; + ptr += to_copy; + offset = 0; + } + +err: + dma_buf_unmap_attachment(attach, table, dir); + dma_buf_detach(dma_buf, attach); + return ret; +} + +static int ion_handle_test_kernel(struct dma_buf *dma_buf, void __user *ptr, + size_t offset, size_t size, bool write) +{ + int ret; + unsigned long page_offset = offset >> PAGE_SHIFT; + size_t copy_offset = offset % PAGE_SIZE; + size_t copy_size = size; + enum dma_data_direction dir = write ? DMA_FROM_DEVICE : DMA_TO_DEVICE; + + if (offset > dma_buf->size || size > dma_buf->size - offset) + return -EINVAL; + + ret = dma_buf_begin_cpu_access(dma_buf, offset, size, dir); + if (ret) + return ret; + + while (copy_size > 0) { + size_t to_copy; + void *vaddr = dma_buf_kmap(dma_buf, page_offset); + + if (!vaddr) + goto err; + + to_copy = min_t(size_t, PAGE_SIZE - copy_offset, copy_size); + + if (write) + ret = copy_from_user(vaddr + copy_offset, ptr, to_copy); + else + ret = copy_to_user(ptr, vaddr + copy_offset, to_copy); + + dma_buf_kunmap(dma_buf, page_offset, vaddr); + if (ret) { + ret = -EFAULT; + goto err; + } + + copy_size -= to_copy; + ptr += to_copy; + page_offset++; + copy_offset = 0; + } +err: + dma_buf_end_cpu_access(dma_buf, offset, size, dir); + return ret; +} + +static long ion_test_ioctl(struct file *filp, unsigned int cmd, + unsigned long arg) +{ + struct ion_test_data *test_data = filp->private_data; + int ret = 0; + + union { + struct ion_test_rw_data test_rw; + } data; + + if (_IOC_SIZE(cmd) > sizeof(data)) + return -EINVAL; + + if (_IOC_DIR(cmd) & _IOC_WRITE) + if (copy_from_user(&data, (void __user *)arg, _IOC_SIZE(cmd))) + return -EFAULT; + + switch (cmd) { + case ION_IOC_TEST_SET_FD: + { + struct dma_buf *dma_buf = NULL; + int fd = arg; + + if (fd >= 0) { + dma_buf = dma_buf_get((int)arg); + if (IS_ERR(dma_buf)) + return PTR_ERR(dma_buf); + } + if (test_data->dma_buf) + dma_buf_put(test_data->dma_buf); + test_data->dma_buf = dma_buf; + break; + } + case ION_IOC_TEST_DMA_MAPPING: + { + ret = ion_handle_test_dma(test_data->dev, test_data->dma_buf, + u64_to_uptr(data.test_rw.ptr), + data.test_rw.offset, data.test_rw.size, + data.test_rw.write); + break; + } + case ION_IOC_TEST_KERNEL_MAPPING: + { + ret = ion_handle_test_kernel(test_data->dma_buf, + u64_to_uptr(data.test_rw.ptr), + data.test_rw.offset, data.test_rw.size, + data.test_rw.write); + break; + } + default: + return -ENOTTY; + } + + if (_IOC_DIR(cmd) & _IOC_READ) { + if (copy_to_user((void __user *)arg, &data, sizeof(data))) + return -EFAULT; + } + return ret; +} + +static int ion_test_open(struct inode *inode, struct file *file) +{ + struct ion_test_data *data; + struct miscdevice *miscdev = file->private_data; + + data = kzalloc(sizeof(struct ion_test_data), GFP_KERNEL); + if (!data) + return -ENOMEM; + + data->dev = miscdev->parent; + + file->private_data = data; + + return 0; +} + +static int ion_test_release(struct inode *inode, struct file *file) +{ + struct ion_test_data *data = file->private_data; + + kfree(data); + + return 0; +} + +static const struct file_operations ion_test_fops = { + .owner = THIS_MODULE, + .unlocked_ioctl = ion_test_ioctl, + .compat_ioctl = ion_test_ioctl, + .open = ion_test_open, + .release = ion_test_release, +}; + +static int __init ion_test_probe(struct platform_device *pdev) +{ + int ret; + struct ion_test_device *testdev; + + testdev = devm_kzalloc(&pdev->dev, sizeof(struct ion_test_device), + GFP_KERNEL); + if (!testdev) + return -ENOMEM; + + testdev->misc.minor = MISC_DYNAMIC_MINOR; + testdev->misc.name = "ion-test"; + testdev->misc.fops = &ion_test_fops; + testdev->misc.parent = &pdev->dev; + ret = misc_register(&testdev->misc); + if (ret) { + pr_err("failed to register misc device.\n"); + return ret; + } + + platform_set_drvdata(pdev, testdev); + + return 0; +} + +static struct platform_driver ion_test_platform_driver = { + .driver = { + .name = "ion-test", + }, +}; + +static int __init ion_test_init(void) +{ + platform_device_register_simple("ion-test", -1, NULL, 0); + return platform_driver_probe(&ion_test_platform_driver, ion_test_probe); +} + +static void __exit ion_test_exit(void) +{ + platform_driver_unregister(&ion_test_platform_driver); +} + +module_init(ion_test_init); +module_exit(ion_test_exit); diff --git a/drivers/staging/android/ion/tegra/Makefile b/drivers/staging/android/ion/tegra/Makefile new file mode 100644 index 000000000000..11cd003fb08f --- /dev/null +++ b/drivers/staging/android/ion/tegra/Makefile @@ -0,0 +1 @@ +obj-y += tegra_ion.o diff --git a/drivers/staging/android/ion/tegra/tegra_ion.c b/drivers/staging/android/ion/tegra/tegra_ion.c new file mode 100644 index 000000000000..3474c65f87fa --- /dev/null +++ b/drivers/staging/android/ion/tegra/tegra_ion.c @@ -0,0 +1,84 @@ +/* + * drivers/gpu/tegra/tegra_ion.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/err.h> +#include <linux/platform_device.h> +#include <linux/slab.h> +#include "../ion.h" +#include "../ion_priv.h" + +static struct ion_device *idev; +static int num_heaps; +static struct ion_heap **heaps; + +static int tegra_ion_probe(struct platform_device *pdev) +{ + struct ion_platform_data *pdata = pdev->dev.platform_data; + int err; + int i; + + num_heaps = pdata->nr; + + heaps = kzalloc(sizeof(struct ion_heap *) * pdata->nr, GFP_KERNEL); + + idev = ion_device_create(NULL); + if (IS_ERR_OR_NULL(idev)) { + kfree(heaps); + return PTR_ERR(idev); + } + + /* create the heaps as specified in the board file */ + for (i = 0; i < num_heaps; i++) { + struct ion_platform_heap *heap_data = &pdata->heaps[i]; + + heaps[i] = ion_heap_create(heap_data); + if (IS_ERR_OR_NULL(heaps[i])) { + err = PTR_ERR(heaps[i]); + goto err; + } + ion_device_add_heap(idev, heaps[i]); + } + platform_set_drvdata(pdev, idev); + return 0; +err: + for (i = 0; i < num_heaps; i++) { + if (heaps[i]) + ion_heap_destroy(heaps[i]); + } + kfree(heaps); + return err; +} + +static int tegra_ion_remove(struct platform_device *pdev) +{ + struct ion_device *idev = platform_get_drvdata(pdev); + int i; + + ion_device_destroy(idev); + for (i = 0; i < num_heaps; i++) + ion_heap_destroy(heaps[i]); + kfree(heaps); + return 0; +} + +static struct platform_driver ion_driver = { + .probe = tegra_ion_probe, + .remove = tegra_ion_remove, + .driver = { .name = "ion-tegra" } +}; + +module_platform_driver(ion_driver); + diff --git a/drivers/staging/android/logger.c b/drivers/staging/android/logger.c deleted file mode 100644 index 34519ea14b54..000000000000 --- a/drivers/staging/android/logger.c +++ /dev/null @@ -1,851 +0,0 @@ -/* - * drivers/misc/logger.c - * - * A Logging Subsystem - * - * Copyright (C) 2007-2008 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. - */ - -#define pr_fmt(fmt) "logger: " fmt - -#include <linux/sched.h> -#include <linux/module.h> -#include <linux/fs.h> -#include <linux/miscdevice.h> -#include <linux/uaccess.h> -#include <linux/poll.h> -#include <linux/slab.h> -#include <linux/time.h> -#include <linux/vmalloc.h> -#include <linux/aio.h> -#include "logger.h" - -#include <asm/ioctls.h> - -/** - * struct logger_log - represents a specific log, such as 'main' or 'radio' - * @buffer: The actual ring buffer - * @misc: The "misc" device representing the log - * @wq: The wait queue for @readers - * @readers: This log's readers - * @mutex: The mutex that protects the @buffer - * @w_off: The current write head offset - * @head: The head, or location that readers start reading at. - * @size: The size of the log - * @logs: The list of log channels - * - * This structure lives from module insertion until module removal, so it does - * not need additional reference counting. The structure is protected by the - * mutex 'mutex'. - */ -struct logger_log { - unsigned char *buffer; - struct miscdevice misc; - wait_queue_head_t wq; - struct list_head readers; - struct mutex mutex; - size_t w_off; - size_t head; - size_t size; - struct list_head logs; -}; - -static LIST_HEAD(log_list); - - -/** - * struct logger_reader - a logging device open for reading - * @log: The associated log - * @list: The associated entry in @logger_log's list - * @r_off: The current read head offset. - * @r_all: Reader can read all entries - * @r_ver: Reader ABI version - * - * This object lives from open to release, so we don't need additional - * reference counting. The structure is protected by log->mutex. - */ -struct logger_reader { - struct logger_log *log; - struct list_head list; - size_t r_off; - bool r_all; - int r_ver; -}; - -/* logger_offset - returns index 'n' into the log via (optimized) modulus */ -static size_t logger_offset(struct logger_log *log, size_t n) -{ - return n & (log->size - 1); -} - - -/* - * file_get_log - Given a file structure, return the associated log - * - * This isn't aesthetic. We have several goals: - * - * 1) Need to quickly obtain the associated log during an I/O operation - * 2) Readers need to maintain state (logger_reader) - * 3) Writers need to be very fast (open() should be a near no-op) - * - * In the reader case, we can trivially go file->logger_reader->logger_log. - * For a writer, we don't want to maintain a logger_reader, so we just go - * file->logger_log. Thus what file->private_data points at depends on whether - * or not the file was opened for reading. This function hides that dirtiness. - */ -static inline struct logger_log *file_get_log(struct file *file) -{ - if (file->f_mode & FMODE_READ) { - struct logger_reader *reader = file->private_data; - return reader->log; - } else - return file->private_data; -} - -/* - * get_entry_header - returns a pointer to the logger_entry header within - * 'log' starting at offset 'off'. A temporary logger_entry 'scratch' must - * be provided. Typically the return value will be a pointer within - * 'logger->buf'. However, a pointer to 'scratch' may be returned if - * the log entry spans the end and beginning of the circular buffer. - */ -static struct logger_entry *get_entry_header(struct logger_log *log, - size_t off, struct logger_entry *scratch) -{ - size_t len = min(sizeof(struct logger_entry), log->size - off); - if (len != sizeof(struct logger_entry)) { - memcpy(((void *) scratch), log->buffer + off, len); - memcpy(((void *) scratch) + len, log->buffer, - sizeof(struct logger_entry) - len); - return scratch; - } - - return (struct logger_entry *) (log->buffer + off); -} - -/* - * get_entry_msg_len - Grabs the length of the message of the entry - * starting from from 'off'. - * - * An entry length is 2 bytes (16 bits) in host endian order. - * In the log, the length does not include the size of the log entry structure. - * This function returns the size including the log entry structure. - * - * Caller needs to hold log->mutex. - */ -static __u32 get_entry_msg_len(struct logger_log *log, size_t off) -{ - struct logger_entry scratch; - struct logger_entry *entry; - - entry = get_entry_header(log, off, &scratch); - return entry->len; -} - -static size_t get_user_hdr_len(int ver) -{ - if (ver < 2) - return sizeof(struct user_logger_entry_compat); - else - return sizeof(struct logger_entry); -} - -static ssize_t copy_header_to_user(int ver, struct logger_entry *entry, - char __user *buf) -{ - void *hdr; - size_t hdr_len; - struct user_logger_entry_compat v1; - - if (ver < 2) { - v1.len = entry->len; - v1.__pad = 0; - v1.pid = entry->pid; - v1.tid = entry->tid; - v1.sec = entry->sec; - v1.nsec = entry->nsec; - hdr = &v1; - hdr_len = sizeof(struct user_logger_entry_compat); - } else { - hdr = entry; - hdr_len = sizeof(struct logger_entry); - } - - return copy_to_user(buf, hdr, hdr_len); -} - -/* - * do_read_log_to_user - reads exactly 'count' bytes from 'log' into the - * user-space buffer 'buf'. Returns 'count' on success. - * - * Caller must hold log->mutex. - */ -static ssize_t do_read_log_to_user(struct logger_log *log, - struct logger_reader *reader, - char __user *buf, - size_t count) -{ - struct logger_entry scratch; - struct logger_entry *entry; - size_t len; - size_t msg_start; - - /* - * First, copy the header to userspace, using the version of - * the header requested - */ - entry = get_entry_header(log, reader->r_off, &scratch); - if (copy_header_to_user(reader->r_ver, entry, buf)) - return -EFAULT; - - count -= get_user_hdr_len(reader->r_ver); - buf += get_user_hdr_len(reader->r_ver); - msg_start = logger_offset(log, - reader->r_off + sizeof(struct logger_entry)); - - /* - * We read from the msg in two disjoint operations. First, we read from - * the current msg head offset up to 'count' bytes or to the end of - * the log, whichever comes first. - */ - len = min(count, log->size - msg_start); - if (copy_to_user(buf, log->buffer + msg_start, len)) - return -EFAULT; - - /* - * Second, we read any remaining bytes, starting back at the head of - * the log. - */ - if (count != len) - if (copy_to_user(buf + len, log->buffer, count - len)) - return -EFAULT; - - reader->r_off = logger_offset(log, reader->r_off + - sizeof(struct logger_entry) + count); - - return count + get_user_hdr_len(reader->r_ver); -} - -/* - * get_next_entry_by_uid - Starting at 'off', returns an offset into - * 'log->buffer' which contains the first entry readable by 'euid' - */ -static size_t get_next_entry_by_uid(struct logger_log *log, - size_t off, kuid_t euid) -{ - while (off != log->w_off) { - struct logger_entry *entry; - struct logger_entry scratch; - size_t next_len; - - entry = get_entry_header(log, off, &scratch); - - if (uid_eq(entry->euid, euid)) - return off; - - next_len = sizeof(struct logger_entry) + entry->len; - off = logger_offset(log, off + next_len); - } - - return off; -} - -/* - * logger_read - our log's read() method - * - * Behavior: - * - * - O_NONBLOCK works - * - If there are no log entries to read, blocks until log is written to - * - Atomically reads exactly one log entry - * - * Will set errno to EINVAL if read - * buffer is insufficient to hold next entry. - */ -static ssize_t logger_read(struct file *file, char __user *buf, - size_t count, loff_t *pos) -{ - struct logger_reader *reader = file->private_data; - struct logger_log *log = reader->log; - ssize_t ret; - DEFINE_WAIT(wait); - -start: - while (1) { - mutex_lock(&log->mutex); - - prepare_to_wait(&log->wq, &wait, TASK_INTERRUPTIBLE); - - ret = (log->w_off == reader->r_off); - mutex_unlock(&log->mutex); - if (!ret) - break; - - if (file->f_flags & O_NONBLOCK) { - ret = -EAGAIN; - break; - } - - if (signal_pending(current)) { - ret = -EINTR; - break; - } - - schedule(); - } - - finish_wait(&log->wq, &wait); - if (ret) - return ret; - - mutex_lock(&log->mutex); - - if (!reader->r_all) - reader->r_off = get_next_entry_by_uid(log, - reader->r_off, current_euid()); - - /* is there still something to read or did we race? */ - if (unlikely(log->w_off == reader->r_off)) { - mutex_unlock(&log->mutex); - goto start; - } - - /* get the size of the next entry */ - ret = get_user_hdr_len(reader->r_ver) + - get_entry_msg_len(log, reader->r_off); - if (count < ret) { - ret = -EINVAL; - goto out; - } - - /* get exactly one entry from the log */ - ret = do_read_log_to_user(log, reader, buf, ret); - -out: - mutex_unlock(&log->mutex); - - return ret; -} - -/* - * get_next_entry - return the offset of the first valid entry at least 'len' - * bytes after 'off'. - * - * Caller must hold log->mutex. - */ -static size_t get_next_entry(struct logger_log *log, size_t off, size_t len) -{ - size_t count = 0; - - do { - size_t nr = sizeof(struct logger_entry) + - get_entry_msg_len(log, off); - off = logger_offset(log, off + nr); - count += nr; - } while (count < len); - - return off; -} - -/* - * is_between - is a < c < b, accounting for wrapping of a, b, and c - * positions in the buffer - * - * That is, if a<b, check for c between a and b - * and if a>b, check for c outside (not between) a and b - * - * |------- a xxxxxxxx b --------| - * c^ - * - * |xxxxx b --------- a xxxxxxxxx| - * c^ - * or c^ - */ -static inline int is_between(size_t a, size_t b, size_t c) -{ - if (a < b) { - /* is c between a and b? */ - if (a < c && c <= b) - return 1; - } else { - /* is c outside of b through a? */ - if (c <= b || a < c) - return 1; - } - - return 0; -} - -/* - * fix_up_readers - walk the list of all readers and "fix up" any who were - * lapped by the writer; also do the same for the default "start head". - * We do this by "pulling forward" the readers and start head to the first - * entry after the new write head. - * - * The caller needs to hold log->mutex. - */ -static void fix_up_readers(struct logger_log *log, size_t len) -{ - size_t old = log->w_off; - size_t new = logger_offset(log, old + len); - struct logger_reader *reader; - - if (is_between(old, new, log->head)) - log->head = get_next_entry(log, log->head, len); - - list_for_each_entry(reader, &log->readers, list) - if (is_between(old, new, reader->r_off)) - reader->r_off = get_next_entry(log, reader->r_off, len); -} - -/* - * do_write_log - writes 'len' bytes from 'buf' to 'log' - * - * The caller needs to hold log->mutex. - */ -static void do_write_log(struct logger_log *log, const void *buf, size_t count) -{ - size_t len; - - len = min(count, log->size - log->w_off); - memcpy(log->buffer + log->w_off, buf, len); - - if (count != len) - memcpy(log->buffer, buf + len, count - len); - - log->w_off = logger_offset(log, log->w_off + count); - -} - -/* - * do_write_log_user - writes 'len' bytes from the user-space buffer 'buf' to - * the log 'log' - * - * The caller needs to hold log->mutex. - * - * Returns 'count' on success, negative error code on failure. - */ -static ssize_t do_write_log_from_user(struct logger_log *log, - const void __user *buf, size_t count) -{ - size_t len; - - len = min(count, log->size - log->w_off); - if (len && copy_from_user(log->buffer + log->w_off, buf, len)) - return -EFAULT; - - if (count != len) - if (copy_from_user(log->buffer, buf + len, count - len)) - /* - * Note that by not updating w_off, this abandons the - * portion of the new entry that *was* successfully - * copied, just above. This is intentional to avoid - * message corruption from missing fragments. - */ - return -EFAULT; - - log->w_off = logger_offset(log, log->w_off + count); - - return count; -} - -/* - * logger_aio_write - our write method, implementing support for write(), - * writev(), and aio_write(). Writes are our fast path, and we try to optimize - * them above all else. - */ -static ssize_t logger_aio_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t ppos) -{ - struct logger_log *log = file_get_log(iocb->ki_filp); - size_t orig; - struct logger_entry header; - struct timespec now; - ssize_t ret = 0; - - now = current_kernel_time(); - - header.pid = current->tgid; - header.tid = current->pid; - header.sec = now.tv_sec; - header.nsec = now.tv_nsec; - header.euid = current_euid(); - header.len = min_t(size_t, iocb->ki_left, LOGGER_ENTRY_MAX_PAYLOAD); - header.hdr_size = sizeof(struct logger_entry); - - /* null writes succeed, return zero */ - if (unlikely(!header.len)) - return 0; - - mutex_lock(&log->mutex); - - orig = log->w_off; - - /* - * Fix up any readers, pulling them forward to the first readable - * entry after (what will be) the new write offset. We do this now - * because if we partially fail, we can end up with clobbered log - * entries that encroach on readable buffer. - */ - fix_up_readers(log, sizeof(struct logger_entry) + header.len); - - do_write_log(log, &header, sizeof(struct logger_entry)); - - while (nr_segs-- > 0) { - size_t len; - ssize_t nr; - - /* figure out how much of this vector we can keep */ - len = min_t(size_t, iov->iov_len, header.len - ret); - - /* write out this segment's payload */ - nr = do_write_log_from_user(log, iov->iov_base, len); - if (unlikely(nr < 0)) { - log->w_off = orig; - mutex_unlock(&log->mutex); - return nr; - } - - iov++; - ret += nr; - } - - mutex_unlock(&log->mutex); - - /* wake up any blocked readers */ - wake_up_interruptible(&log->wq); - - return ret; -} - -static struct logger_log *get_log_from_minor(int minor) -{ - struct logger_log *log; - - list_for_each_entry(log, &log_list, logs) - if (log->misc.minor == minor) - return log; - return NULL; -} - -/* - * logger_open - the log's open() file operation - * - * Note how near a no-op this is in the write-only case. Keep it that way! - */ -static int logger_open(struct inode *inode, struct file *file) -{ - struct logger_log *log; - int ret; - - ret = nonseekable_open(inode, file); - if (ret) - return ret; - - log = get_log_from_minor(MINOR(inode->i_rdev)); - if (!log) - return -ENODEV; - - if (file->f_mode & FMODE_READ) { - struct logger_reader *reader; - - reader = kmalloc(sizeof(struct logger_reader), GFP_KERNEL); - if (!reader) - return -ENOMEM; - - reader->log = log; - reader->r_ver = 1; - reader->r_all = in_egroup_p(inode->i_gid) || - capable(CAP_SYSLOG); - - INIT_LIST_HEAD(&reader->list); - - mutex_lock(&log->mutex); - reader->r_off = log->head; - list_add_tail(&reader->list, &log->readers); - mutex_unlock(&log->mutex); - - file->private_data = reader; - } else - file->private_data = log; - - return 0; -} - -/* - * logger_release - the log's release file operation - * - * Note this is a total no-op in the write-only case. Keep it that way! - */ -static int logger_release(struct inode *ignored, struct file *file) -{ - if (file->f_mode & FMODE_READ) { - struct logger_reader *reader = file->private_data; - struct logger_log *log = reader->log; - - mutex_lock(&log->mutex); - list_del(&reader->list); - mutex_unlock(&log->mutex); - - kfree(reader); - } - - return 0; -} - -/* - * logger_poll - the log's poll file operation, for poll/select/epoll - * - * Note we always return POLLOUT, because you can always write() to the log. - * Note also that, strictly speaking, a return value of POLLIN does not - * guarantee that the log is readable without blocking, as there is a small - * chance that the writer can lap the reader in the interim between poll() - * returning and the read() request. - */ -static unsigned int logger_poll(struct file *file, poll_table *wait) -{ - struct logger_reader *reader; - struct logger_log *log; - unsigned int ret = POLLOUT | POLLWRNORM; - - if (!(file->f_mode & FMODE_READ)) - return ret; - - reader = file->private_data; - log = reader->log; - - poll_wait(file, &log->wq, wait); - - mutex_lock(&log->mutex); - if (!reader->r_all) - reader->r_off = get_next_entry_by_uid(log, - reader->r_off, current_euid()); - - if (log->w_off != reader->r_off) - ret |= POLLIN | POLLRDNORM; - mutex_unlock(&log->mutex); - - return ret; -} - -static long logger_set_version(struct logger_reader *reader, void __user *arg) -{ - int version; - if (copy_from_user(&version, arg, sizeof(int))) - return -EFAULT; - - if ((version < 1) || (version > 2)) - return -EINVAL; - - reader->r_ver = version; - return 0; -} - -static long logger_ioctl(struct file *file, unsigned int cmd, unsigned long arg) -{ - struct logger_log *log = file_get_log(file); - struct logger_reader *reader; - long ret = -EINVAL; - void __user *argp = (void __user *) arg; - - mutex_lock(&log->mutex); - - switch (cmd) { - case LOGGER_GET_LOG_BUF_SIZE: - ret = log->size; - break; - case LOGGER_GET_LOG_LEN: - if (!(file->f_mode & FMODE_READ)) { - ret = -EBADF; - break; - } - reader = file->private_data; - if (log->w_off >= reader->r_off) - ret = log->w_off - reader->r_off; - else - ret = (log->size - reader->r_off) + log->w_off; - break; - case LOGGER_GET_NEXT_ENTRY_LEN: - if (!(file->f_mode & FMODE_READ)) { - ret = -EBADF; - break; - } - reader = file->private_data; - - if (!reader->r_all) - reader->r_off = get_next_entry_by_uid(log, - reader->r_off, current_euid()); - - if (log->w_off != reader->r_off) - ret = get_user_hdr_len(reader->r_ver) + - get_entry_msg_len(log, reader->r_off); - else - ret = 0; - break; - case LOGGER_FLUSH_LOG: - if (!(file->f_mode & FMODE_WRITE)) { - ret = -EBADF; - break; - } - if (!(in_egroup_p(file->f_dentry->d_inode->i_gid) || - capable(CAP_SYSLOG))) { - ret = -EPERM; - break; - } - list_for_each_entry(reader, &log->readers, list) - reader->r_off = log->w_off; - log->head = log->w_off; - ret = 0; - break; - case LOGGER_GET_VERSION: - if (!(file->f_mode & FMODE_READ)) { - ret = -EBADF; - break; - } - reader = file->private_data; - ret = reader->r_ver; - break; - case LOGGER_SET_VERSION: - if (!(file->f_mode & FMODE_READ)) { - ret = -EBADF; - break; - } - reader = file->private_data; - ret = logger_set_version(reader, argp); - break; - } - - mutex_unlock(&log->mutex); - - return ret; -} - -static const struct file_operations logger_fops = { - .owner = THIS_MODULE, - .read = logger_read, - .aio_write = logger_aio_write, - .poll = logger_poll, - .unlocked_ioctl = logger_ioctl, - .compat_ioctl = logger_ioctl, - .open = logger_open, - .release = logger_release, -}; - -/* - * Log size must must be a power of two, and greater than - * (LOGGER_ENTRY_MAX_PAYLOAD + sizeof(struct logger_entry)). - */ -static int __init create_log(char *log_name, int size) -{ - int ret = 0; - struct logger_log *log; - unsigned char *buffer; - - buffer = vmalloc(size); - if (buffer == NULL) - return -ENOMEM; - - log = kzalloc(sizeof(struct logger_log), GFP_KERNEL); - if (log == NULL) { - ret = -ENOMEM; - goto out_free_buffer; - } - log->buffer = buffer; - - log->misc.minor = MISC_DYNAMIC_MINOR; - log->misc.name = kstrdup(log_name, GFP_KERNEL); - if (log->misc.name == NULL) { - ret = -ENOMEM; - goto out_free_log; - } - - log->misc.fops = &logger_fops; - log->misc.parent = NULL; - - init_waitqueue_head(&log->wq); - INIT_LIST_HEAD(&log->readers); - mutex_init(&log->mutex); - log->w_off = 0; - log->head = 0; - log->size = size; - - INIT_LIST_HEAD(&log->logs); - list_add_tail(&log->logs, &log_list); - - /* finally, initialize the misc device for this log */ - ret = misc_register(&log->misc); - if (unlikely(ret)) { - pr_err("failed to register misc device for log '%s'!\n", - log->misc.name); - goto out_free_log; - } - - pr_info("created %luK log '%s'\n", - (unsigned long) log->size >> 10, log->misc.name); - - return 0; - -out_free_log: - kfree(log); - -out_free_buffer: - vfree(buffer); - return ret; -} - -static int __init logger_init(void) -{ - int ret; - - ret = create_log(LOGGER_LOG_MAIN, 256*1024); - if (unlikely(ret)) - goto out; - - ret = create_log(LOGGER_LOG_EVENTS, 256*1024); - if (unlikely(ret)) - goto out; - - ret = create_log(LOGGER_LOG_RADIO, 256*1024); - if (unlikely(ret)) - goto out; - - ret = create_log(LOGGER_LOG_SYSTEM, 256*1024); - if (unlikely(ret)) - goto out; - -out: - return ret; -} - -static void __exit logger_exit(void) -{ - struct logger_log *current_log, *next_log; - - list_for_each_entry_safe(current_log, next_log, &log_list, logs) { - /* we have to delete all the entry inside log_list */ - misc_deregister(¤t_log->misc); - vfree(current_log->buffer); - kfree(current_log->misc.name); - list_del(¤t_log->logs); - kfree(current_log); - } -} - - -device_initcall(logger_init); -module_exit(logger_exit); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Robert Love, <rlove@google.com>"); -MODULE_DESCRIPTION("Android Logger"); diff --git a/drivers/staging/android/logger.h b/drivers/staging/android/logger.h deleted file mode 100644 index 70af7d805dff..000000000000 --- a/drivers/staging/android/logger.h +++ /dev/null @@ -1,89 +0,0 @@ -/* include/linux/logger.h - * - * Copyright (C) 2007-2008 Google, Inc. - * Author: Robert Love <rlove@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_LOGGER_H -#define _LINUX_LOGGER_H - -#include <linux/types.h> -#include <linux/ioctl.h> - -/** - * struct user_logger_entry_compat - defines a single entry that is given to a logger - * @len: The length of the payload - * @__pad: Two bytes of padding that appear to be required - * @pid: The generating process' process ID - * @tid: The generating process' thread ID - * @sec: The number of seconds that have elapsed since the Epoch - * @nsec: The number of nanoseconds that have elapsed since @sec - * @msg: The message that is to be logged - * - * The userspace structure for version 1 of the logger_entry ABI. - * This structure is returned to userspace unless the caller requests - * an upgrade to a newer ABI version. - */ -struct user_logger_entry_compat { - __u16 len; - __u16 __pad; - __s32 pid; - __s32 tid; - __s32 sec; - __s32 nsec; - char msg[0]; -}; - -/** - * struct logger_entry - defines a single entry that is given to a logger - * @len: The length of the payload - * @hdr_size: sizeof(struct logger_entry_v2) - * @pid: The generating process' process ID - * @tid: The generating process' thread ID - * @sec: The number of seconds that have elapsed since the Epoch - * @nsec: The number of nanoseconds that have elapsed since @sec - * @euid: Effective UID of logger - * @msg: The message that is to be logged - * - * The structure for version 2 of the logger_entry ABI. - * This structure is returned to userspace if ioctl(LOGGER_SET_VERSION) - * is called with version >= 2 - */ -struct logger_entry { - __u16 len; - __u16 hdr_size; - __s32 pid; - __s32 tid; - __s32 sec; - __s32 nsec; - kuid_t euid; - char msg[0]; -}; - -#define LOGGER_LOG_RADIO "log_radio" /* radio-related messages */ -#define LOGGER_LOG_EVENTS "log_events" /* system/hardware events */ -#define LOGGER_LOG_SYSTEM "log_system" /* system/framework messages */ -#define LOGGER_LOG_MAIN "log_main" /* everything else */ - -#define LOGGER_ENTRY_MAX_PAYLOAD 4076 - -#define __LOGGERIO 0xAE - -#define LOGGER_GET_LOG_BUF_SIZE _IO(__LOGGERIO, 1) /* size of log */ -#define LOGGER_GET_LOG_LEN _IO(__LOGGERIO, 2) /* used log len */ -#define LOGGER_GET_NEXT_ENTRY_LEN _IO(__LOGGERIO, 3) /* next entry len */ -#define LOGGER_FLUSH_LOG _IO(__LOGGERIO, 4) /* flush log */ -#define LOGGER_GET_VERSION _IO(__LOGGERIO, 5) /* abi version */ -#define LOGGER_SET_VERSION _IO(__LOGGERIO, 6) /* abi version */ - -#endif /* _LINUX_LOGGER_H */ diff --git a/drivers/staging/android/lowmemorykiller.c b/drivers/staging/android/lowmemorykiller.c index fe74494868ef..a56e0894f668 100644 --- a/drivers/staging/android/lowmemorykiller.c +++ b/drivers/staging/android/lowmemorykiller.c @@ -39,7 +39,6 @@ #include <linux/sched.h> #include <linux/swap.h> #include <linux/rcupdate.h> -#include <linux/profile.h> #include <linux/notifier.h> static uint32_t lowmem_debug_level = 1; @@ -74,6 +73,7 @@ static int lowmem_shrink(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); @@ -86,8 +86,8 @@ static int lowmem_shrink(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; } @@ -144,13 +144,22 @@ static int lowmem_shrink(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) { - lowmem_print(1, "send sigkill to %d (%s), adj %hd, size %d\n", - selected->pid, selected->comm, - selected_oom_score_adj, selected_tasksize); + lowmem_print(1, "Killing '%s' (%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_oom_score_adj, + selected_tasksize * (long)(PAGE_SIZE / 1024), + current->comm, current->pid, + other_file * (long)(PAGE_SIZE / 1024), + minfree * (long)(PAGE_SIZE / 1024), + min_score_adj, + other_free * (long)(PAGE_SIZE / 1024)); lowmem_deathpending_timeout = jiffies + HZ; send_sig(SIGKILL, selected, 0); set_tsk_thread_flag(selected, TIF_MEMDIE); @@ -178,9 +187,94 @@ static void __exit lowmem_exit(void) unregister_shrinker(&lowmem_shrinker); } +#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 = ¶m_ops_short, + .elemsize = sizeof(lowmem_adj[0]), + .elem = lowmem_adj, +}; +#endif + module_param_named(cost, lowmem_shrinker.seeks, int, S_IRUGO | S_IWUSR); +#ifdef CONFIG_ANDROID_LOW_MEMORY_KILLER_AUTODETECT_OOM_ADJ_VALUES +__module_param_call(MODULE_PARAM_PREFIX, adj, + &lowmem_adj_array_ops, + .arr = &__param_arr_adj, + S_IRUGO | S_IWUSR, -1); +__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/sw_sync.c b/drivers/staging/android/sw_sync.c index 4928f93bdf3d..820af5cfb830 100644 --- a/drivers/staging/android/sw_sync.c +++ b/drivers/staging/android/sw_sync.c @@ -97,6 +97,7 @@ static void sw_sync_pt_value_str(struct sync_pt *sync_pt, char *str, int size) { struct sw_sync_pt *pt = (struct sw_sync_pt *)sync_pt; + snprintf(str, size, "%d", pt->value); } @@ -156,6 +157,7 @@ static int sw_sync_open(struct inode *inode, struct file *file) static int sw_sync_release(struct inode *inode, struct file *file) { struct sw_sync_timeline *obj = file->private_data; + sync_timeline_destroy(&obj->obj); return 0; } diff --git a/drivers/staging/android/sw_sync.h b/drivers/staging/android/sw_sync.h index 585040be5f18..1a50669ec8a9 100644 --- a/drivers/staging/android/sw_sync.h +++ b/drivers/staging/android/sw_sync.h @@ -18,10 +18,9 @@ #define _LINUX_SW_SYNC_H #include <linux/types.h> - -#ifdef __KERNEL__ - +#include <linux/kconfig.h> #include "sync.h" +#include "uapi/sw_sync.h" struct sw_sync_timeline { struct sync_timeline obj; @@ -35,24 +34,26 @@ struct sw_sync_pt { u32 value; }; +#if IS_ENABLED(CONFIG_SW_SYNC) struct sw_sync_timeline *sw_sync_timeline_create(const char *name); void sw_sync_timeline_inc(struct sw_sync_timeline *obj, u32 inc); struct sync_pt *sw_sync_pt_create(struct sw_sync_timeline *obj, u32 value); - -#endif /* __KERNEL __ */ - -struct sw_sync_create_fence_data { - __u32 value; - char name[32]; - __s32 fence; /* fd of new fence */ -}; - -#define SW_SYNC_IOC_MAGIC 'W' - -#define SW_SYNC_IOC_CREATE_FENCE _IOWR(SW_SYNC_IOC_MAGIC, 0,\ - struct sw_sync_create_fence_data) -#define SW_SYNC_IOC_INC _IOW(SW_SYNC_IOC_MAGIC, 1, __u32) - +#else +static inline struct sw_sync_timeline *sw_sync_timeline_create(const char *name) +{ + return NULL; +} + +static inline void sw_sync_timeline_inc(struct sw_sync_timeline *obj, u32 inc) +{ +} + +static inline struct sync_pt *sw_sync_pt_create(struct sw_sync_timeline *obj, + u32 value) +{ + return NULL; +} +#endif /* IS_ENABLED(CONFIG_SW_SYNC) */ #endif /* _LINUX_SW_SYNC_H */ diff --git a/drivers/staging/android/sync.c b/drivers/staging/android/sync.c index 3893a3574769..61e624905916 100644 --- a/drivers/staging/android/sync.c +++ b/drivers/staging/android/sync.c @@ -79,27 +79,27 @@ static void sync_timeline_free(struct kref *kref) container_of(kref, struct sync_timeline, kref); unsigned long flags; - if (obj->ops->release_obj) - obj->ops->release_obj(obj); - spin_lock_irqsave(&sync_timeline_list_lock, flags); list_del(&obj->sync_timeline_list); spin_unlock_irqrestore(&sync_timeline_list_lock, flags); + if (obj->ops->release_obj) + obj->ops->release_obj(obj); + kfree(obj); } void sync_timeline_destroy(struct sync_timeline *obj) { obj->destroyed = true; + smp_wmb(); /* - * If this is not the last reference, signal any children - * that their parent is going away. + * signal any children that their parent is going away. */ + sync_timeline_signal(obj); - if (!kref_put(&obj->kref, sync_timeline_free)) - sync_timeline_signal(obj); + kref_put(&obj->kref, sync_timeline_free); } EXPORT_SYMBOL(sync_timeline_destroy); @@ -384,6 +384,7 @@ static void sync_fence_detach_pts(struct sync_fence *fence) list_for_each_safe(pos, n, &fence->pt_list_head) { struct sync_pt *pt = container_of(pos, struct sync_pt, pt_list); + sync_timeline_remove_pt(pt); } } @@ -394,6 +395,7 @@ static void sync_fence_free_pts(struct sync_fence *fence) list_for_each_safe(pos, n, &fence->pt_list_head) { struct sync_pt *pt = container_of(pos, struct sync_pt, pt_list); + sync_pt_free(pt); } } @@ -827,6 +829,7 @@ static long sync_fence_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct sync_fence *fence = file->private_data; + switch (cmd) { case SYNC_IOC_WAIT: return sync_fence_ioctl_wait(fence, arg); @@ -856,18 +859,21 @@ static const char *sync_status_str(int status) static void sync_print_pt(struct seq_file *s, struct sync_pt *pt, bool fence) { int status = pt->status; + seq_printf(s, " %s%spt %s", fence ? pt->parent->name : "", fence ? "_" : "", sync_status_str(status)); if (pt->status) { struct timeval tv = ktime_to_timeval(pt->timestamp); + seq_printf(s, "@%ld.%06ld", tv.tv_sec, tv.tv_usec); } if (pt->parent->ops->timeline_value_str && pt->parent->ops->pt_value_str) { char value[64]; + pt->parent->ops->pt_value_str(pt, value, sizeof(value)); seq_printf(s, ": %s", value); if (fence) { @@ -892,6 +898,7 @@ static void sync_print_obj(struct seq_file *s, struct sync_timeline *obj) if (obj->ops->timeline_value_str) { char value[64]; + obj->ops->timeline_value_str(obj, value, sizeof(value)); seq_printf(s, ": %s", value); } else if (obj->ops->print_obj) { @@ -1001,6 +1008,7 @@ void sync_dump(void) for (i = 0; i < s.count; i += DUMP_CHUNK) { if ((s.count - i) > DUMP_CHUNK) { char c = s.buf[i + DUMP_CHUNK]; + s.buf[i + DUMP_CHUNK] = 0; pr_cont("%s", s.buf + i); s.buf[i + DUMP_CHUNK] = c; diff --git a/drivers/staging/android/sync.h b/drivers/staging/android/sync.h index 38ea986dc70f..75da9e85ac69 100644 --- a/drivers/staging/android/sync.h +++ b/drivers/staging/android/sync.h @@ -14,14 +14,14 @@ #define _LINUX_SYNC_H #include <linux/types.h> -#ifdef __KERNEL__ - #include <linux/kref.h> #include <linux/ktime.h> #include <linux/list.h> #include <linux/spinlock.h> #include <linux/wait.h> +#include "uapi/sync.h" + struct sync_timeline; struct sync_pt; struct sync_fence; @@ -341,86 +341,4 @@ int sync_fence_cancel_async(struct sync_fence *fence, */ int sync_fence_wait(struct sync_fence *fence, long timeout); -#endif /* __KERNEL__ */ - -/** - * struct sync_merge_data - data passed to merge ioctl - * @fd2: file descriptor of second fence - * @name: name of new fence - * @fence: returns the fd of the new fence to userspace - */ -struct sync_merge_data { - __s32 fd2; /* fd of second fence */ - char name[32]; /* name of new fence */ - __s32 fence; /* fd on newly created fence */ -}; - -/** - * struct sync_pt_info - detailed sync_pt information - * @len: length of sync_pt_info including any driver_data - * @obj_name: name of parent sync_timeline - * @driver_name: name of driver implmenting the parent - * @status: status of the sync_pt 0:active 1:signaled <0:error - * @timestamp_ns: timestamp of status change in nanoseconds - * @driver_data: any driver dependant data - */ -struct sync_pt_info { - __u32 len; - char obj_name[32]; - char driver_name[32]; - __s32 status; - __u64 timestamp_ns; - - __u8 driver_data[0]; -}; - -/** - * struct sync_fence_info_data - data returned from fence info ioctl - * @len: ioctl caller writes the size of the buffer its passing in. - * ioctl returns length of sync_fence_data reutnred to userspace - * including pt_info. - * @name: name of fence - * @status: status of fence. 1: signaled 0:active <0:error - * @pt_info: a sync_pt_info struct for every sync_pt in the fence - */ -struct sync_fence_info_data { - __u32 len; - char name[32]; - __s32 status; - - __u8 pt_info[0]; -}; - -#define SYNC_IOC_MAGIC '>' - -/** - * DOC: SYNC_IOC_WAIT - wait for a fence to signal - * - * pass timeout in milliseconds. Waits indefinitely timeout < 0. - */ -#define SYNC_IOC_WAIT _IOW(SYNC_IOC_MAGIC, 0, __s32) - -/** - * DOC: SYNC_IOC_MERGE - merge two fences - * - * Takes a struct sync_merge_data. Creates a new fence containing copies of - * the sync_pts in both the calling fd and sync_merge_data.fd2. Returns the - * new fence's fd in sync_merge_data.fence - */ -#define SYNC_IOC_MERGE _IOWR(SYNC_IOC_MAGIC, 1, struct sync_merge_data) - -/** - * DOC: SYNC_IOC_FENCE_INFO - get detailed information on a fence - * - * Takes a struct sync_fence_info_data with extra space allocated for pt_info. - * Caller should write the size of the buffer into len. On return, len is - * updated to reflect the total size of the sync_fence_info_data including - * pt_info. - * - * pt_info is a buffer containing sync_pt_infos for every sync_pt in the fence. - * To itterate over the sync_pt_infos, use the sync_pt_info.len field. - */ -#define SYNC_IOC_FENCE_INFO _IOWR(SYNC_IOC_MAGIC, 2,\ - struct sync_fence_info_data) - #endif /* _LINUX_SYNC_H */ diff --git a/drivers/staging/android/timed_gpio.c b/drivers/staging/android/timed_gpio.c index e81451425c01..ae9966d1f7cc 100644 --- a/drivers/staging/android/timed_gpio.c +++ b/drivers/staging/android/timed_gpio.c @@ -51,6 +51,7 @@ static int gpio_get_time(struct timed_output_dev *dev) if (hrtimer_active(&data->timer)) { ktime_t r = hrtimer_get_remaining(&data->timer); struct timeval t = ktime_to_timeval(r); + return t.tv_sec * 1000 + t.tv_usec / 1000; } else return 0; diff --git a/drivers/staging/android/uapi/ashmem.h b/drivers/staging/android/uapi/ashmem.h new file mode 100644 index 000000000000..13df42d200b7 --- /dev/null +++ b/drivers/staging/android/uapi/ashmem.h @@ -0,0 +1,48 @@ +/* + * drivers/staging/android/uapi/ashmem.h + * + * Copyright 2008 Google Inc. + * Author: Robert Love + * + * 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 _UAPI_LINUX_ASHMEM_H +#define _UAPI_LINUX_ASHMEM_H + +#include <linux/ioctl.h> +#include <linux/types.h> + +#define ASHMEM_NAME_LEN 256 + +#define ASHMEM_NAME_DEF "dev/ashmem" + +/* Return values from ASHMEM_PIN: Was the mapping purged while unpinned? */ +#define ASHMEM_NOT_PURGED 0 +#define ASHMEM_WAS_PURGED 1 + +/* Return values from ASHMEM_GET_PIN_STATUS: Is the mapping pinned? */ +#define ASHMEM_IS_UNPINNED 0 +#define ASHMEM_IS_PINNED 1 + +struct ashmem_pin { + __u32 offset; /* offset into region, in bytes, page-aligned */ + __u32 len; /* length forward from offset, in bytes, page-aligned */ +}; + +#define __ASHMEMIOC 0x77 + +#define ASHMEM_SET_NAME _IOW(__ASHMEMIOC, 1, char[ASHMEM_NAME_LEN]) +#define ASHMEM_GET_NAME _IOR(__ASHMEMIOC, 2, char[ASHMEM_NAME_LEN]) +#define ASHMEM_SET_SIZE _IOW(__ASHMEMIOC, 3, size_t) +#define ASHMEM_GET_SIZE _IO(__ASHMEMIOC, 4) +#define ASHMEM_SET_PROT_MASK _IOW(__ASHMEMIOC, 5, unsigned long) +#define ASHMEM_GET_PROT_MASK _IO(__ASHMEMIOC, 6) +#define ASHMEM_PIN _IOW(__ASHMEMIOC, 7, struct ashmem_pin) +#define ASHMEM_UNPIN _IOW(__ASHMEMIOC, 8, struct ashmem_pin) +#define ASHMEM_GET_PIN_STATUS _IO(__ASHMEMIOC, 9) +#define ASHMEM_PURGE_ALL_CACHES _IO(__ASHMEMIOC, 10) + +#endif /* _UAPI_LINUX_ASHMEM_H */ diff --git a/drivers/staging/android/uapi/ion.h b/drivers/staging/android/uapi/ion.h new file mode 100644 index 000000000000..f09e7c154d69 --- /dev/null +++ b/drivers/staging/android/uapi/ion.h @@ -0,0 +1,196 @@ +/* + * drivers/staging/android/uapi/ion.h + * + * 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. + * + */ + +#ifndef _UAPI_LINUX_ION_H +#define _UAPI_LINUX_ION_H + +#include <linux/ioctl.h> +#include <linux/types.h> + +typedef int ion_user_handle_t; + +/** + * enum ion_heap_types - list of all possible types of heaps + * @ION_HEAP_TYPE_SYSTEM: memory allocated via vmalloc + * @ION_HEAP_TYPE_SYSTEM_CONTIG: memory allocated via kmalloc + * @ION_HEAP_TYPE_CARVEOUT: memory allocated from a prereserved + * carveout heap, allocations are physically + * contiguous + * @ION_HEAP_TYPE_DMA: memory allocated via DMA API + * @ION_NUM_HEAPS: helper for iterating over heaps, a bit mask + * is used to identify the heaps, so only 32 + * total heap types are supported + */ +enum ion_heap_type { + ION_HEAP_TYPE_SYSTEM, + ION_HEAP_TYPE_SYSTEM_CONTIG, + ION_HEAP_TYPE_CARVEOUT, + ION_HEAP_TYPE_CHUNK, + ION_HEAP_TYPE_DMA, + ION_HEAP_TYPE_CUSTOM, /* must be last so device specific heaps always + are at the end of this enum */ + ION_NUM_HEAPS = 16, +}; + +#define ION_HEAP_SYSTEM_MASK (1 << ION_HEAP_TYPE_SYSTEM) +#define ION_HEAP_SYSTEM_CONTIG_MASK (1 << ION_HEAP_TYPE_SYSTEM_CONTIG) +#define ION_HEAP_CARVEOUT_MASK (1 << ION_HEAP_TYPE_CARVEOUT) +#define ION_HEAP_TYPE_DMA_MASK (1 << ION_HEAP_TYPE_DMA) + +#define ION_NUM_HEAP_IDS sizeof(unsigned int) * 8 + +/** + * allocation flags - the lower 16 bits are used by core ion, the upper 16 + * bits are reserved for use by the heaps themselves. + */ +#define ION_FLAG_CACHED 1 /* mappings of this buffer should be + cached, ion will do cache + maintenance when the buffer is + mapped for dma */ +#define ION_FLAG_CACHED_NEEDS_SYNC 2 /* mappings of this buffer will created + at mmap time, if this is set + caches must be managed manually */ + +/** + * DOC: Ion Userspace API + * + * create a client by opening /dev/ion + * most operations handled via following ioctls + * + */ + +/** + * struct ion_allocation_data - metadata passed from userspace for allocations + * @len: size of the allocation + * @align: required alignment of the allocation + * @heap_id_mask: mask of heap ids to allocate from + * @flags: flags passed to heap + * @handle: pointer that will be populated with a cookie to use to + * refer to this allocation + * + * Provided by userspace as an argument to the ioctl + */ +struct ion_allocation_data { + size_t len; + size_t align; + unsigned int heap_id_mask; + unsigned int flags; + ion_user_handle_t handle; +}; + +/** + * struct ion_fd_data - metadata passed to/from userspace for a handle/fd pair + * @handle: a handle + * @fd: a file descriptor representing that handle + * + * For ION_IOC_SHARE or ION_IOC_MAP userspace populates the handle field with + * the handle returned from ion alloc, and the kernel returns the file + * descriptor to share or map in the fd field. For ION_IOC_IMPORT, userspace + * provides the file descriptor and the kernel returns the handle. + */ +struct ion_fd_data { + ion_user_handle_t handle; + int fd; +}; + +/** + * struct ion_handle_data - a handle passed to/from the kernel + * @handle: a handle + */ +struct ion_handle_data { + ion_user_handle_t handle; +}; + +/** + * struct ion_custom_data - metadata passed to/from userspace for a custom ioctl + * @cmd: the custom ioctl function to call + * @arg: additional data to pass to the custom ioctl, typically a user + * pointer to a predefined structure + * + * This works just like the regular cmd and arg fields of an ioctl. + */ +struct ion_custom_data { + unsigned int cmd; + unsigned long arg; +}; + +#define ION_IOC_MAGIC 'I' + +/** + * DOC: ION_IOC_ALLOC - allocate memory + * + * Takes an ion_allocation_data struct and returns it with the handle field + * populated with the opaque handle for the allocation. + */ +#define ION_IOC_ALLOC _IOWR(ION_IOC_MAGIC, 0, \ + struct ion_allocation_data) + +/** + * DOC: ION_IOC_FREE - free memory + * + * Takes an ion_handle_data struct and frees the handle. + */ +#define ION_IOC_FREE _IOWR(ION_IOC_MAGIC, 1, struct ion_handle_data) + +/** + * DOC: ION_IOC_MAP - get a file descriptor to mmap + * + * Takes an ion_fd_data struct with the handle field populated with a valid + * opaque handle. Returns the struct with the fd field set to a file + * descriptor open in the current address space. This file descriptor + * can then be used as an argument to mmap. + */ +#define ION_IOC_MAP _IOWR(ION_IOC_MAGIC, 2, struct ion_fd_data) + +/** + * DOC: ION_IOC_SHARE - creates a file descriptor to use to share an allocation + * + * Takes an ion_fd_data struct with the handle field populated with a valid + * opaque handle. Returns the struct with the fd field set to a file + * descriptor open in the current address space. This file descriptor + * can then be passed to another process. The corresponding opaque handle can + * be retrieved via ION_IOC_IMPORT. + */ +#define ION_IOC_SHARE _IOWR(ION_IOC_MAGIC, 4, struct ion_fd_data) + +/** + * DOC: ION_IOC_IMPORT - imports a shared file descriptor + * + * Takes an ion_fd_data struct with the fd field populated with a valid file + * descriptor obtained from ION_IOC_SHARE and returns the struct with the handle + * filed set to the corresponding opaque handle. + */ +#define ION_IOC_IMPORT _IOWR(ION_IOC_MAGIC, 5, struct ion_fd_data) + +/** + * DOC: ION_IOC_SYNC - syncs a shared file descriptors to memory + * + * Deprecated in favor of using the dma_buf api's correctly (syncing + * will happend automatically when the buffer is mapped to a device). + * If necessary should be used after touching a cached buffer from the cpu, + * this will make the buffer in memory coherent. + */ +#define ION_IOC_SYNC _IOWR(ION_IOC_MAGIC, 7, struct ion_fd_data) + +/** + * DOC: ION_IOC_CUSTOM - call architecture specific ion ioctl + * + * Takes the argument of the architecture specific ioctl to call and + * passes appropriate userdata for that ioctl + */ +#define ION_IOC_CUSTOM _IOWR(ION_IOC_MAGIC, 6, struct ion_custom_data) + +#endif /* _UAPI_LINUX_ION_H */ diff --git a/drivers/staging/android/uapi/ion_test.h b/drivers/staging/android/uapi/ion_test.h new file mode 100644 index 000000000000..ffef06f63133 --- /dev/null +++ b/drivers/staging/android/uapi/ion_test.h @@ -0,0 +1,70 @@ +/* + * drivers/staging/android/uapi/ion.h + * + * 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. + * + */ + +#ifndef _UAPI_LINUX_ION_TEST_H +#define _UAPI_LINUX_ION_TEST_H + +#include <linux/ioctl.h> +#include <linux/types.h> + +/** + * struct ion_test_rw_data - metadata passed to the kernel to read handle + * @ptr: a pointer to an area at least as large as size + * @offset: offset into the ion buffer to start reading + * @size: size to read or write + * @write: 1 to write, 0 to read + */ +struct ion_test_rw_data { + __u64 ptr; + __u64 offset; + __u64 size; + int write; + int __padding; +}; + +#define ION_IOC_MAGIC 'I' + +/** + * DOC: ION_IOC_TEST_SET_DMA_BUF - attach a dma buf to the test driver + * + * Attaches a dma buf fd to the test driver. Passing a second fd or -1 will + * release the first fd. + */ +#define ION_IOC_TEST_SET_FD \ + _IO(ION_IOC_MAGIC, 0xf0) + +/** + * DOC: ION_IOC_TEST_DMA_MAPPING - read or write memory from a handle as DMA + * + * Reads or writes the memory from a handle using an uncached mapping. Can be + * used by unit tests to emulate a DMA engine as close as possible. Only + * expected to be used for debugging and testing, may not always be available. + */ +#define ION_IOC_TEST_DMA_MAPPING \ + _IOW(ION_IOC_MAGIC, 0xf1, struct ion_test_rw_data) + +/** + * DOC: ION_IOC_TEST_KERNEL_MAPPING - read or write memory from a handle + * + * Reads or writes the memory from a handle using a kernel mapping. Can be + * used by unit tests to test heap map_kernel functions. Only expected to be + * used for debugging and testing, may not always be available. + */ +#define ION_IOC_TEST_KERNEL_MAPPING \ + _IOW(ION_IOC_MAGIC, 0xf2, struct ion_test_rw_data) + + +#endif /* _UAPI_LINUX_ION_H */ diff --git a/drivers/staging/android/uapi/sw_sync.h b/drivers/staging/android/uapi/sw_sync.h new file mode 100644 index 000000000000..9b5d4869505c --- /dev/null +++ b/drivers/staging/android/uapi/sw_sync.h @@ -0,0 +1,32 @@ +/* + * 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 _UAPI_LINUX_SW_SYNC_H +#define _UAPI_LINUX_SW_SYNC_H + +#include <linux/types.h> + +struct sw_sync_create_fence_data { + __u32 value; + char name[32]; + __s32 fence; /* fd of new fence */ +}; + +#define SW_SYNC_IOC_MAGIC 'W' + +#define SW_SYNC_IOC_CREATE_FENCE _IOWR(SW_SYNC_IOC_MAGIC, 0,\ + struct sw_sync_create_fence_data) +#define SW_SYNC_IOC_INC _IOW(SW_SYNC_IOC_MAGIC, 1, __u32) + +#endif /* _UAPI_LINUX_SW_SYNC_H */ diff --git a/drivers/staging/android/uapi/sync.h b/drivers/staging/android/uapi/sync.h new file mode 100644 index 000000000000..57fdaadc4b04 --- /dev/null +++ b/drivers/staging/android/uapi/sync.h @@ -0,0 +1,97 @@ +/* + * Copyright (C) 2012 Google, Inc. + * + * 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_SYNC_H +#define _UAPI_LINUX_SYNC_H + +#include <linux/ioctl.h> +#include <linux/types.h> + +/** + * struct sync_merge_data - data passed to merge ioctl + * @fd2: file descriptor of second fence + * @name: name of new fence + * @fence: returns the fd of the new fence to userspace + */ +struct sync_merge_data { + __s32 fd2; /* fd of second fence */ + char name[32]; /* name of new fence */ + __s32 fence; /* fd on newly created fence */ +}; + +/** + * struct sync_pt_info - detailed sync_pt information + * @len: length of sync_pt_info including any driver_data + * @obj_name: name of parent sync_timeline + * @driver_name: name of driver implmenting the parent + * @status: status of the sync_pt 0:active 1:signaled <0:error + * @timestamp_ns: timestamp of status change in nanoseconds + * @driver_data: any driver dependant data + */ +struct sync_pt_info { + __u32 len; + char obj_name[32]; + char driver_name[32]; + __s32 status; + __u64 timestamp_ns; + + __u8 driver_data[0]; +}; + +/** + * struct sync_fence_info_data - data returned from fence info ioctl + * @len: ioctl caller writes the size of the buffer its passing in. + * ioctl returns length of sync_fence_data reutnred to userspace + * including pt_info. + * @name: name of fence + * @status: status of fence. 1: signaled 0:active <0:error + * @pt_info: a sync_pt_info struct for every sync_pt in the fence + */ +struct sync_fence_info_data { + __u32 len; + char name[32]; + __s32 status; + + __u8 pt_info[0]; +}; + +#define SYNC_IOC_MAGIC '>' + +/** + * DOC: SYNC_IOC_WAIT - wait for a fence to signal + * + * pass timeout in milliseconds. Waits indefinitely timeout < 0. + */ +#define SYNC_IOC_WAIT _IOW(SYNC_IOC_MAGIC, 0, __s32) + +/** + * DOC: SYNC_IOC_MERGE - merge two fences + * + * Takes a struct sync_merge_data. Creates a new fence containing copies of + * the sync_pts in both the calling fd and sync_merge_data.fd2. Returns the + * new fence's fd in sync_merge_data.fence + */ +#define SYNC_IOC_MERGE _IOWR(SYNC_IOC_MAGIC, 1, struct sync_merge_data) + +/** + * DOC: SYNC_IOC_FENCE_INFO - get detailed information on a fence + * + * Takes a struct sync_fence_info_data with extra space allocated for pt_info. + * Caller should write the size of the buffer into len. On return, len is + * updated to reflect the total size of the sync_fence_info_data including + * pt_info. + * + * pt_info is a buffer containing sync_pt_infos for every sync_pt in the fence. + * To itterate over the sync_pt_infos, use the sync_pt_info.len field. + */ +#define SYNC_IOC_FENCE_INFO _IOWR(SYNC_IOC_MAGIC, 2,\ + struct sync_fence_info_data) + +#endif /* _UAPI_LINUX_SYNC_H */ diff --git a/drivers/switch/Kconfig b/drivers/switch/Kconfig new file mode 100644 index 000000000000..19404b6f7778 --- /dev/null +++ b/drivers/switch/Kconfig @@ -0,0 +1,15 @@ +menuconfig SWITCH + tristate "Switch class support" + help + Say Y here to enable switch class support. This allows + monitoring switches by userspace via sysfs and uevent. + +if SWITCH + +config SWITCH_GPIO + tristate "GPIO Swith support" + depends on GPIOLIB + help + Say Y here to enable GPIO based switch support. + +endif # SWITCH diff --git a/drivers/switch/Makefile b/drivers/switch/Makefile new file mode 100644 index 000000000000..f7606ed4a719 --- /dev/null +++ b/drivers/switch/Makefile @@ -0,0 +1,4 @@ +# Switch Class Driver +obj-$(CONFIG_SWITCH) += switch_class.o +obj-$(CONFIG_SWITCH_GPIO) += switch_gpio.o + diff --git a/drivers/switch/switch_class.c b/drivers/switch/switch_class.c new file mode 100644 index 000000000000..e373b625806e --- /dev/null +++ b/drivers/switch/switch_class.c @@ -0,0 +1,174 @@ +/* + * drivers/switch/switch_class.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/module.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/device.h> +#include <linux/fs.h> +#include <linux/err.h> +#include <linux/switch.h> + +struct class *switch_class; +static atomic_t device_count; + +static ssize_t state_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct switch_dev *sdev = (struct switch_dev *) + dev_get_drvdata(dev); + + if (sdev->print_state) { + int ret = sdev->print_state(sdev, buf); + if (ret >= 0) + return ret; + } + return sprintf(buf, "%d\n", sdev->state); +} + +static ssize_t name_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + struct switch_dev *sdev = (struct switch_dev *) + dev_get_drvdata(dev); + + if (sdev->print_name) { + int ret = sdev->print_name(sdev, buf); + if (ret >= 0) + return ret; + } + return sprintf(buf, "%s\n", sdev->name); +} + +static DEVICE_ATTR(state, S_IRUGO, state_show, NULL); +static DEVICE_ATTR(name, S_IRUGO, name_show, NULL); + +void switch_set_state(struct switch_dev *sdev, int state) +{ + char name_buf[120]; + char state_buf[120]; + char *prop_buf; + char *envp[3]; + int env_offset = 0; + int length; + + if (sdev->state != state) { + sdev->state = state; + + prop_buf = (char *)get_zeroed_page(GFP_KERNEL); + if (prop_buf) { + length = name_show(sdev->dev, NULL, prop_buf); + if (length > 0) { + if (prop_buf[length - 1] == '\n') + prop_buf[length - 1] = 0; + snprintf(name_buf, sizeof(name_buf), + "SWITCH_NAME=%s", prop_buf); + envp[env_offset++] = name_buf; + } + length = state_show(sdev->dev, NULL, prop_buf); + if (length > 0) { + if (prop_buf[length - 1] == '\n') + prop_buf[length - 1] = 0; + snprintf(state_buf, sizeof(state_buf), + "SWITCH_STATE=%s", prop_buf); + envp[env_offset++] = state_buf; + } + envp[env_offset] = NULL; + kobject_uevent_env(&sdev->dev->kobj, KOBJ_CHANGE, envp); + free_page((unsigned long)prop_buf); + } else { + printk(KERN_ERR "out of memory in switch_set_state\n"); + kobject_uevent(&sdev->dev->kobj, KOBJ_CHANGE); + } + } +} +EXPORT_SYMBOL_GPL(switch_set_state); + +static int create_switch_class(void) +{ + if (!switch_class) { + switch_class = class_create(THIS_MODULE, "switch"); + if (IS_ERR(switch_class)) + return PTR_ERR(switch_class); + atomic_set(&device_count, 0); + } + + return 0; +} + +int switch_dev_register(struct switch_dev *sdev) +{ + int ret; + + if (!switch_class) { + ret = create_switch_class(); + if (ret < 0) + return ret; + } + + sdev->index = atomic_inc_return(&device_count); + sdev->dev = device_create(switch_class, NULL, + MKDEV(0, sdev->index), NULL, sdev->name); + if (IS_ERR(sdev->dev)) + return PTR_ERR(sdev->dev); + + ret = device_create_file(sdev->dev, &dev_attr_state); + if (ret < 0) + goto err_create_file_1; + ret = device_create_file(sdev->dev, &dev_attr_name); + if (ret < 0) + goto err_create_file_2; + + dev_set_drvdata(sdev->dev, sdev); + sdev->state = 0; + return 0; + +err_create_file_2: + device_remove_file(sdev->dev, &dev_attr_state); +err_create_file_1: + device_destroy(switch_class, MKDEV(0, sdev->index)); + printk(KERN_ERR "switch: Failed to register driver %s\n", sdev->name); + + return ret; +} +EXPORT_SYMBOL_GPL(switch_dev_register); + +void switch_dev_unregister(struct switch_dev *sdev) +{ + device_remove_file(sdev->dev, &dev_attr_name); + device_remove_file(sdev->dev, &dev_attr_state); + device_destroy(switch_class, MKDEV(0, sdev->index)); + dev_set_drvdata(sdev->dev, NULL); +} +EXPORT_SYMBOL_GPL(switch_dev_unregister); + +static int __init switch_class_init(void) +{ + return create_switch_class(); +} + +static void __exit switch_class_exit(void) +{ + class_destroy(switch_class); +} + +module_init(switch_class_init); +module_exit(switch_class_exit); + +MODULE_AUTHOR("Mike Lockwood <lockwood@android.com>"); +MODULE_DESCRIPTION("Switch class driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/switch/switch_gpio.c b/drivers/switch/switch_gpio.c new file mode 100644 index 000000000000..621d62d20c99 --- /dev/null +++ b/drivers/switch/switch_gpio.c @@ -0,0 +1,172 @@ +/* + * drivers/switch/switch_gpio.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/module.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/platform_device.h> +#include <linux/slab.h> +#include <linux/switch.h> +#include <linux/workqueue.h> +#include <linux/gpio.h> + +struct gpio_switch_data { + struct switch_dev sdev; + unsigned gpio; + const char *name_on; + const char *name_off; + const char *state_on; + const char *state_off; + int irq; + struct work_struct work; +}; + +static void gpio_switch_work(struct work_struct *work) +{ + int state; + struct gpio_switch_data *data = + container_of(work, struct gpio_switch_data, work); + + state = gpio_get_value(data->gpio); + switch_set_state(&data->sdev, state); +} + +static irqreturn_t gpio_irq_handler(int irq, void *dev_id) +{ + struct gpio_switch_data *switch_data = + (struct gpio_switch_data *)dev_id; + + schedule_work(&switch_data->work); + return IRQ_HANDLED; +} + +static ssize_t switch_gpio_print_state(struct switch_dev *sdev, char *buf) +{ + struct gpio_switch_data *switch_data = + container_of(sdev, struct gpio_switch_data, sdev); + const char *state; + if (switch_get_state(sdev)) + state = switch_data->state_on; + else + state = switch_data->state_off; + + if (state) + return sprintf(buf, "%s\n", state); + return -1; +} + +static int gpio_switch_probe(struct platform_device *pdev) +{ + struct gpio_switch_platform_data *pdata = pdev->dev.platform_data; + struct gpio_switch_data *switch_data; + int ret = 0; + + if (!pdata) + return -EBUSY; + + switch_data = kzalloc(sizeof(struct gpio_switch_data), GFP_KERNEL); + if (!switch_data) + return -ENOMEM; + + switch_data->sdev.name = pdata->name; + switch_data->gpio = pdata->gpio; + switch_data->name_on = pdata->name_on; + switch_data->name_off = pdata->name_off; + switch_data->state_on = pdata->state_on; + switch_data->state_off = pdata->state_off; + switch_data->sdev.print_state = switch_gpio_print_state; + + ret = switch_dev_register(&switch_data->sdev); + if (ret < 0) + goto err_switch_dev_register; + + ret = gpio_request(switch_data->gpio, pdev->name); + if (ret < 0) + goto err_request_gpio; + + ret = gpio_direction_input(switch_data->gpio); + if (ret < 0) + goto err_set_gpio_input; + + INIT_WORK(&switch_data->work, gpio_switch_work); + + switch_data->irq = gpio_to_irq(switch_data->gpio); + if (switch_data->irq < 0) { + ret = switch_data->irq; + goto err_detect_irq_num_failed; + } + + ret = request_irq(switch_data->irq, gpio_irq_handler, + IRQF_TRIGGER_LOW, pdev->name, switch_data); + if (ret < 0) + goto err_request_irq; + + /* Perform initial detection */ + gpio_switch_work(&switch_data->work); + + return 0; + +err_request_irq: +err_detect_irq_num_failed: +err_set_gpio_input: + gpio_free(switch_data->gpio); +err_request_gpio: + switch_dev_unregister(&switch_data->sdev); +err_switch_dev_register: + kfree(switch_data); + + return ret; +} + +static int gpio_switch_remove(struct platform_device *pdev) +{ + struct gpio_switch_data *switch_data = platform_get_drvdata(pdev); + + cancel_work_sync(&switch_data->work); + gpio_free(switch_data->gpio); + switch_dev_unregister(&switch_data->sdev); + kfree(switch_data); + + return 0; +} + +static struct platform_driver gpio_switch_driver = { + .probe = gpio_switch_probe, + .remove = gpio_switch_remove, + .driver = { + .name = "switch-gpio", + .owner = THIS_MODULE, + }, +}; + +static int __init gpio_switch_init(void) +{ + return platform_driver_register(&gpio_switch_driver); +} + +static void __exit gpio_switch_exit(void) +{ + platform_driver_unregister(&gpio_switch_driver); +} + +module_init(gpio_switch_init); +module_exit(gpio_switch_exit); + +MODULE_AUTHOR("Mike Lockwood <lockwood@android.com>"); +MODULE_DESCRIPTION("GPIO Switch driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/tty/serial/serial_core.c b/drivers/tty/serial/serial_core.c index 1fabb22ae615..57977ec53a7a 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_circ_empty(&state->xmit) && state->xmit.buf && !tty->stopped && !tty->hw_stopped) port->ops->start_tx(port); diff --git a/drivers/usb/gadget/Kconfig b/drivers/usb/gadget/Kconfig index f41aa0d0c414..9de2eb28085e 100644 --- a/drivers/usb/gadget/Kconfig +++ b/drivers/usb/gadget/Kconfig @@ -818,6 +818,27 @@ config USB_G_PRINTER For more information, see Documentation/usb/gadget_printer.txt which includes sample code for accessing the device file. +config USB_G_ANDROID + boolean "Android Composite Gadget" + depends on SND + select SND_PCM + select SND_RAWMIDI + select USB_F_ACM + select USB_LIBCOMPOSITE + select USB_U_SERIAL + help + The Android Composite Gadget supports multiple USB + functions: adb, acm, mass storage, mtp, accessory + and rndis. + Each function can be configured and enabled/disabled + dynamically from userspace through a sysfs interface. + +config USB_ANDROID_RNDIS_DWORD_ALIGNED + boolean "Use double word aligned" + depends on USB_G_ANDROID + help + Provides dword aligned for DMA controller. + if TTY config USB_CDC_COMPOSITE diff --git a/drivers/usb/gadget/Makefile b/drivers/usb/gadget/Makefile index 6afd16659e78..0ec50ae5b639 100644 --- a/drivers/usb/gadget/Makefile +++ b/drivers/usb/gadget/Makefile @@ -65,6 +65,7 @@ g_nokia-y := nokia.o g_webcam-y := webcam.o g_ncm-y := ncm.o g_acm_ms-y := acm_ms.o +g_android-y := android.o g_tcm_usb_gadget-y := tcm_usb_gadget.o obj-$(CONFIG_USB_ZERO) += g_zero.o @@ -84,4 +85,5 @@ obj-$(CONFIG_USB_G_NOKIA) += g_nokia.o obj-$(CONFIG_USB_G_WEBCAM) += g_webcam.o obj-$(CONFIG_USB_G_NCM) += g_ncm.o obj-$(CONFIG_USB_G_ACM_MS) += g_acm_ms.o +obj-$(CONFIG_USB_G_ANDROID) += g_android.o obj-$(CONFIG_USB_GADGET_TARGET) += tcm_usb_gadget.o diff --git a/drivers/usb/gadget/android.c b/drivers/usb/gadget/android.c new file mode 100644 index 000000000000..61c5e582bf47 --- /dev/null +++ b/drivers/usb/gadget/android.c @@ -0,0 +1,1587 @@ +/* + * Gadget Driver for Android + * + * Copyright (C) 2008 Google, Inc. + * Author: Mike Lockwood <lockwood@android.com> + * Benoit Goby <benoit@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/init.h> +#include <linux/module.h> +#include <linux/fs.h> +#include <linux/delay.h> +#include <linux/kernel.h> +#include <linux/utsname.h> +#include <linux/platform_device.h> + +#include <linux/usb/ch9.h> +#include <linux/usb/composite.h> +#include <linux/usb/gadget.h> + +#include "gadget_chips.h" + +#include "f_fs.c" +#include "f_audio_source.c" +#include "f_midi.c" +#include "f_mass_storage.c" +#include "f_mtp.c" +#include "f_accessory.c" +#define USB_ETH_RNDIS y +#include "f_rndis.c" +#include "rndis.c" +#include "u_ether.c" + +MODULE_AUTHOR("Mike Lockwood"); +MODULE_DESCRIPTION("Android Composite USB Driver"); +MODULE_LICENSE("GPL"); +MODULE_VERSION("1.0"); + +static const char longname[] = "Gadget Android"; + +/* Default vendor and product IDs, overridden by userspace */ +#define VENDOR_ID 0x18D1 +#define PRODUCT_ID 0x0001 + +/* f_midi configuration */ +#define MIDI_INPUT_PORTS 1 +#define MIDI_OUTPUT_PORTS 1 +#define MIDI_BUFFER_SIZE 256 +#define MIDI_QUEUE_LENGTH 32 + +struct android_usb_function { + char *name; + void *config; + + struct device *dev; + char *dev_name; + struct device_attribute **attributes; + + /* for android_dev.enabled_functions */ + struct list_head enabled_list; + + /* Optional: initialization during gadget bind */ + int (*init)(struct android_usb_function *, struct usb_composite_dev *); + /* Optional: cleanup during gadget unbind */ + void (*cleanup)(struct android_usb_function *); + /* Optional: called when the function is added the list of + * enabled functions */ + void (*enable)(struct android_usb_function *); + /* Optional: called when it is removed */ + void (*disable)(struct android_usb_function *); + + int (*bind_config)(struct android_usb_function *, + struct usb_configuration *); + + /* Optional: called when the configuration is removed */ + void (*unbind_config)(struct android_usb_function *, + struct usb_configuration *); + /* Optional: handle ctrl requests before the device is configured */ + int (*ctrlrequest)(struct android_usb_function *, + struct usb_composite_dev *, + const struct usb_ctrlrequest *); +}; + +struct android_dev { + struct android_usb_function **functions; + struct list_head enabled_functions; + struct usb_composite_dev *cdev; + struct device *dev; + + void (*setup_complete)(struct usb_ep *ep, + struct usb_request *req); + + bool enabled; + int disable_depth; + struct mutex mutex; + bool connected; + bool sw_connected; + struct work_struct work; + char ffs_aliases[256]; +}; + +static struct class *android_class; +static struct android_dev *_android_dev; +static int android_bind_config(struct usb_configuration *c); +static void android_unbind_config(struct usb_configuration *c); + +/* string IDs are assigned dynamically */ +#define STRING_MANUFACTURER_IDX 0 +#define STRING_PRODUCT_IDX 1 +#define STRING_SERIAL_IDX 2 + +static char manufacturer_string[256]; +static char product_string[256]; +static char serial_string[256]; + +/* String Table */ +static struct usb_string strings_dev[] = { + [STRING_MANUFACTURER_IDX].s = manufacturer_string, + [STRING_PRODUCT_IDX].s = product_string, + [STRING_SERIAL_IDX].s = serial_string, + { } /* end of list */ +}; + +static struct usb_gadget_strings stringtab_dev = { + .language = 0x0409, /* en-us */ + .strings = strings_dev, +}; + +static struct usb_gadget_strings *dev_strings[] = { + &stringtab_dev, + NULL, +}; + +static struct usb_device_descriptor device_desc = { + .bLength = sizeof(device_desc), + .bDescriptorType = USB_DT_DEVICE, + .bcdUSB = __constant_cpu_to_le16(0x0200), + .bDeviceClass = USB_CLASS_PER_INTERFACE, + .idVendor = __constant_cpu_to_le16(VENDOR_ID), + .idProduct = __constant_cpu_to_le16(PRODUCT_ID), + .bcdDevice = __constant_cpu_to_le16(0xffff), + .bNumConfigurations = 1, +}; + +static struct usb_configuration android_config_driver = { + .label = "android", + .unbind = android_unbind_config, + .bConfigurationValue = 1, + .bmAttributes = USB_CONFIG_ATT_ONE | USB_CONFIG_ATT_SELFPOWER, + .MaxPower = 500, /* 500ma */ +}; + +static void android_work(struct work_struct *data) +{ + struct android_dev *dev = container_of(data, struct android_dev, work); + struct usb_composite_dev *cdev = dev->cdev; + char *disconnected[2] = { "USB_STATE=DISCONNECTED", NULL }; + char *connected[2] = { "USB_STATE=CONNECTED", NULL }; + char *configured[2] = { "USB_STATE=CONFIGURED", NULL }; + char **uevent_envp = NULL; + unsigned long flags; + + spin_lock_irqsave(&cdev->lock, flags); + if (cdev->config) + uevent_envp = configured; + else if (dev->connected != dev->sw_connected) + uevent_envp = dev->connected ? connected : disconnected; + dev->sw_connected = dev->connected; + spin_unlock_irqrestore(&cdev->lock, flags); + + if (uevent_envp) { + kobject_uevent_env(&dev->dev->kobj, KOBJ_CHANGE, uevent_envp); + pr_info("%s: sent uevent %s\n", __func__, uevent_envp[0]); + } else { + pr_info("%s: did not send uevent (%d %d %p)\n", __func__, + dev->connected, dev->sw_connected, cdev->config); + } +} + +static void android_enable(struct android_dev *dev) +{ + struct usb_composite_dev *cdev = dev->cdev; + + if (WARN_ON(!dev->disable_depth)) + return; + + if (--dev->disable_depth == 0) { + usb_add_config(cdev, &android_config_driver, + android_bind_config); + usb_gadget_connect(cdev->gadget); + } +} + +static void android_disable(struct android_dev *dev) +{ + struct usb_composite_dev *cdev = dev->cdev; + + if (dev->disable_depth++ == 0) { + usb_gadget_disconnect(cdev->gadget); + /* Cancel pending control requests */ + usb_ep_dequeue(cdev->gadget->ep0, cdev->req); + usb_remove_config(cdev, &android_config_driver); + } +} + +/*-------------------------------------------------------------------------*/ +/* Supported functions initialization */ + +struct functionfs_config { + bool opened; + bool enabled; + struct ffs_data *data; +}; + +static int ffs_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + f->config = kzalloc(sizeof(struct functionfs_config), GFP_KERNEL); + if (!f->config) + return -ENOMEM; + + return functionfs_init(); +} + +static void ffs_function_cleanup(struct android_usb_function *f) +{ + functionfs_cleanup(); + kfree(f->config); +} + +static void ffs_function_enable(struct android_usb_function *f) +{ + struct android_dev *dev = _android_dev; + struct functionfs_config *config = f->config; + + config->enabled = true; + + /* Disable the gadget until the function is ready */ + if (!config->opened) + android_disable(dev); +} + +static void ffs_function_disable(struct android_usb_function *f) +{ + struct android_dev *dev = _android_dev; + struct functionfs_config *config = f->config; + + config->enabled = false; + + /* Balance the disable that was called in closed_callback */ + if (!config->opened) + android_enable(dev); +} + +static int ffs_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + struct functionfs_config *config = f->config; + return functionfs_bind_config(c->cdev, c, config->data); +} + +static ssize_t +ffs_aliases_show(struct device *pdev, struct device_attribute *attr, char *buf) +{ + struct android_dev *dev = _android_dev; + int ret; + + mutex_lock(&dev->mutex); + ret = sprintf(buf, "%s\n", dev->ffs_aliases); + mutex_unlock(&dev->mutex); + + return ret; +} + +static ssize_t +ffs_aliases_store(struct device *pdev, struct device_attribute *attr, + const char *buf, size_t size) +{ + struct android_dev *dev = _android_dev; + char buff[256]; + + mutex_lock(&dev->mutex); + + if (dev->enabled) { + mutex_unlock(&dev->mutex); + return -EBUSY; + } + + strlcpy(buff, buf, sizeof(buff)); + strlcpy(dev->ffs_aliases, strim(buff), sizeof(dev->ffs_aliases)); + + mutex_unlock(&dev->mutex); + + return size; +} + +static DEVICE_ATTR(aliases, S_IRUGO | S_IWUSR, ffs_aliases_show, + ffs_aliases_store); +static struct device_attribute *ffs_function_attributes[] = { + &dev_attr_aliases, + NULL +}; + +static struct android_usb_function ffs_function = { + .name = "ffs", + .init = ffs_function_init, + .enable = ffs_function_enable, + .disable = ffs_function_disable, + .cleanup = ffs_function_cleanup, + .bind_config = ffs_function_bind_config, + .attributes = ffs_function_attributes, +}; + +static int functionfs_ready_callback(struct ffs_data *ffs) +{ + struct android_dev *dev = _android_dev; + struct functionfs_config *config = ffs_function.config; + int ret = 0; + + mutex_lock(&dev->mutex); + + ret = functionfs_bind(ffs, dev->cdev); + if (ret) + goto err; + + config->data = ffs; + config->opened = true; + + if (config->enabled) + android_enable(dev); + +err: + mutex_unlock(&dev->mutex); + return ret; +} + +static void functionfs_closed_callback(struct ffs_data *ffs) +{ + struct android_dev *dev = _android_dev; + struct functionfs_config *config = ffs_function.config; + + mutex_lock(&dev->mutex); + + if (config->enabled) + android_disable(dev); + + config->opened = false; + config->data = NULL; + + functionfs_unbind(ffs); + + mutex_unlock(&dev->mutex); +} + +static void *functionfs_acquire_dev_callback(const char *dev_name) +{ + return 0; +} + +static void functionfs_release_dev_callback(struct ffs_data *ffs_data) +{ +} + +#define MAX_ACM_INSTANCES 4 +struct acm_function_config { + int instances; + int instances_on; + struct usb_function *f_acm[MAX_ACM_INSTANCES]; + struct usb_function_instance *f_acm_inst[MAX_ACM_INSTANCES]; +}; + +static int +acm_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + int i; + int ret; + struct acm_function_config *config; + + config = kzalloc(sizeof(struct acm_function_config), GFP_KERNEL); + if (!config) + return -ENOMEM; + f->config = config; + + for (i = 0; i < MAX_ACM_INSTANCES; i++) { + config->f_acm_inst[i] = usb_get_function_instance("acm"); + if (IS_ERR(config->f_acm_inst[i])) { + ret = PTR_ERR(config->f_acm_inst[i]); + goto err_usb_get_function_instance; + } + config->f_acm[i] = usb_get_function(config->f_acm_inst[i]); + if (IS_ERR(config->f_acm[i])) { + ret = PTR_ERR(config->f_acm[i]); + goto err_usb_get_function; + } + } + return 0; +err_usb_get_function_instance: + while (i-- > 0) { + usb_put_function(config->f_acm[i]); +err_usb_get_function: + usb_put_function_instance(config->f_acm_inst[i]); + } + return ret; +} + +static void acm_function_cleanup(struct android_usb_function *f) +{ + int i; + struct acm_function_config *config = f->config; + + for (i = 0; i < MAX_ACM_INSTANCES; i++) { + usb_put_function(config->f_acm[i]); + usb_put_function_instance(config->f_acm_inst[i]); + } + kfree(f->config); + f->config = NULL; +} + +static int +acm_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + int i; + int ret = 0; + struct acm_function_config *config = f->config; + + config->instances_on = config->instances; + for (i = 0; i < config->instances_on; i++) { + ret = usb_add_function(c, config->f_acm[i]); + if (ret) { + pr_err("Could not bind acm%u config\n", i); + goto err_usb_add_function; + } + } + + return 0; + +err_usb_add_function: + while (i-- > 0) + usb_remove_function(c, config->f_acm[i]); + return ret; +} + +static void acm_function_unbind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + int i; + struct acm_function_config *config = f->config; + + for (i = 0; i < config->instances_on; i++) + usb_remove_function(c, config->f_acm[i]); +} + +static ssize_t acm_instances_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct acm_function_config *config = f->config; + return sprintf(buf, "%d\n", config->instances); +} + +static ssize_t acm_instances_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct acm_function_config *config = f->config; + int value; + + sscanf(buf, "%d", &value); + if (value > MAX_ACM_INSTANCES) + value = MAX_ACM_INSTANCES; + config->instances = value; + return size; +} + +static DEVICE_ATTR(instances, S_IRUGO | S_IWUSR, acm_instances_show, + acm_instances_store); +static struct device_attribute *acm_function_attributes[] = { + &dev_attr_instances, + NULL +}; + +static struct android_usb_function acm_function = { + .name = "acm", + .init = acm_function_init, + .cleanup = acm_function_cleanup, + .bind_config = acm_function_bind_config, + .unbind_config = acm_function_unbind_config, + .attributes = acm_function_attributes, +}; + + +static int +mtp_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + return mtp_setup(); +} + +static void mtp_function_cleanup(struct android_usb_function *f) +{ + mtp_cleanup(); +} + +static int +mtp_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + return mtp_bind_config(c, false); +} + +static int +ptp_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + /* nothing to do - initialization is handled by mtp_function_init */ + return 0; +} + +static void ptp_function_cleanup(struct android_usb_function *f) +{ + /* nothing to do - cleanup is handled by mtp_function_cleanup */ +} + +static int +ptp_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + return mtp_bind_config(c, true); +} + +static int mtp_function_ctrlrequest(struct android_usb_function *f, + struct usb_composite_dev *cdev, + const struct usb_ctrlrequest *c) +{ + return mtp_ctrlrequest(cdev, c); +} + +static struct android_usb_function mtp_function = { + .name = "mtp", + .init = mtp_function_init, + .cleanup = mtp_function_cleanup, + .bind_config = mtp_function_bind_config, + .ctrlrequest = mtp_function_ctrlrequest, +}; + +/* PTP function is same as MTP with slightly different interface descriptor */ +static struct android_usb_function ptp_function = { + .name = "ptp", + .init = ptp_function_init, + .cleanup = ptp_function_cleanup, + .bind_config = ptp_function_bind_config, +}; + + +struct rndis_function_config { + u8 ethaddr[ETH_ALEN]; + u32 vendorID; + char manufacturer[256]; + /* "Wireless" RNDIS; auto-detected by Windows */ + bool wceis; + struct eth_dev *dev; +}; + +static int +rndis_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + f->config = kzalloc(sizeof(struct rndis_function_config), GFP_KERNEL); + if (!f->config) + return -ENOMEM; + return 0; +} + +static void rndis_function_cleanup(struct android_usb_function *f) +{ + kfree(f->config); + f->config = NULL; +} + +static int +rndis_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + int ret; + struct eth_dev *dev; + struct rndis_function_config *rndis = f->config; + + if (!rndis) { + pr_err("%s: rndis_pdata\n", __func__); + return -1; + } + + pr_info("%s MAC: %02X:%02X:%02X:%02X:%02X:%02X\n", __func__, + rndis->ethaddr[0], rndis->ethaddr[1], rndis->ethaddr[2], + rndis->ethaddr[3], rndis->ethaddr[4], rndis->ethaddr[5]); + + dev = gether_setup_name(c->cdev->gadget, rndis->ethaddr, "rndis"); + if (IS_ERR(dev)) { + ret = PTR_ERR(dev); + pr_err("%s: gether_setup failed\n", __func__); + return ret; + } + rndis->dev = dev; + + if (rndis->wceis) { + /* "Wireless" RNDIS; auto-detected by Windows */ + rndis_iad_descriptor.bFunctionClass = + USB_CLASS_WIRELESS_CONTROLLER; + rndis_iad_descriptor.bFunctionSubClass = 0x01; + rndis_iad_descriptor.bFunctionProtocol = 0x03; + rndis_control_intf.bInterfaceClass = + USB_CLASS_WIRELESS_CONTROLLER; + rndis_control_intf.bInterfaceSubClass = 0x01; + rndis_control_intf.bInterfaceProtocol = 0x03; + } + + return rndis_bind_config_vendor(c, rndis->ethaddr, rndis->vendorID, + rndis->manufacturer, rndis->dev); +} + +static void rndis_function_unbind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + struct rndis_function_config *rndis = f->config; + gether_cleanup(rndis->dev); +} + +static ssize_t rndis_manufacturer_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *config = f->config; + return sprintf(buf, "%s\n", config->manufacturer); +} + +static ssize_t rndis_manufacturer_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *config = f->config; + + if (size >= sizeof(config->manufacturer)) + return -EINVAL; + if (sscanf(buf, "%s", config->manufacturer) == 1) + return size; + return -1; +} + +static DEVICE_ATTR(manufacturer, S_IRUGO | S_IWUSR, rndis_manufacturer_show, + rndis_manufacturer_store); + +static ssize_t rndis_wceis_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *config = f->config; + return sprintf(buf, "%d\n", config->wceis); +} + +static ssize_t rndis_wceis_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *config = f->config; + int value; + + if (sscanf(buf, "%d", &value) == 1) { + config->wceis = value; + return size; + } + return -EINVAL; +} + +static DEVICE_ATTR(wceis, S_IRUGO | S_IWUSR, rndis_wceis_show, + rndis_wceis_store); + +static ssize_t rndis_ethaddr_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *rndis = f->config; + return sprintf(buf, "%02x:%02x:%02x:%02x:%02x:%02x\n", + rndis->ethaddr[0], rndis->ethaddr[1], rndis->ethaddr[2], + rndis->ethaddr[3], rndis->ethaddr[4], rndis->ethaddr[5]); +} + +static ssize_t rndis_ethaddr_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *rndis = f->config; + + if (sscanf(buf, "%02x:%02x:%02x:%02x:%02x:%02x\n", + (int *)&rndis->ethaddr[0], (int *)&rndis->ethaddr[1], + (int *)&rndis->ethaddr[2], (int *)&rndis->ethaddr[3], + (int *)&rndis->ethaddr[4], (int *)&rndis->ethaddr[5]) == 6) + return size; + return -EINVAL; +} + +static DEVICE_ATTR(ethaddr, S_IRUGO | S_IWUSR, rndis_ethaddr_show, + rndis_ethaddr_store); + +static ssize_t rndis_vendorID_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *config = f->config; + return sprintf(buf, "%04x\n", config->vendorID); +} + +static ssize_t rndis_vendorID_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct rndis_function_config *config = f->config; + int value; + + if (sscanf(buf, "%04x", &value) == 1) { + config->vendorID = value; + return size; + } + return -EINVAL; +} + +static DEVICE_ATTR(vendorID, S_IRUGO | S_IWUSR, rndis_vendorID_show, + rndis_vendorID_store); + +static struct device_attribute *rndis_function_attributes[] = { + &dev_attr_manufacturer, + &dev_attr_wceis, + &dev_attr_ethaddr, + &dev_attr_vendorID, + NULL +}; + +static struct android_usb_function rndis_function = { + .name = "rndis", + .init = rndis_function_init, + .cleanup = rndis_function_cleanup, + .bind_config = rndis_function_bind_config, + .unbind_config = rndis_function_unbind_config, + .attributes = rndis_function_attributes, +}; + + +struct mass_storage_function_config { + struct fsg_config fsg; + struct fsg_common *common; +}; + +static int mass_storage_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + struct mass_storage_function_config *config; + struct fsg_common *common; + int err; + + config = kzalloc(sizeof(struct mass_storage_function_config), + GFP_KERNEL); + if (!config) + return -ENOMEM; + + config->fsg.nluns = 1; + config->fsg.luns[0].removable = 1; + + common = fsg_common_init(NULL, cdev, &config->fsg); + if (IS_ERR(common)) { + kfree(config); + return PTR_ERR(common); + } + + err = sysfs_create_link(&f->dev->kobj, + &common->luns[0].dev.kobj, + "lun"); + if (err) { + kfree(config); + return err; + } + + config->common = common; + f->config = config; + return 0; +} + +static void mass_storage_function_cleanup(struct android_usb_function *f) +{ + kfree(f->config); + f->config = NULL; +} + +static int mass_storage_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + struct mass_storage_function_config *config = f->config; + return fsg_bind_config(c->cdev, c, config->common); +} + +static ssize_t mass_storage_inquiry_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct mass_storage_function_config *config = f->config; + return sprintf(buf, "%s\n", config->common->inquiry_string); +} + +static ssize_t mass_storage_inquiry_store(struct device *dev, + struct device_attribute *attr, const char *buf, size_t size) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct mass_storage_function_config *config = f->config; + if (size >= sizeof(config->common->inquiry_string)) + return -EINVAL; + if (sscanf(buf, "%s", config->common->inquiry_string) != 1) + return -EINVAL; + return size; +} + +static DEVICE_ATTR(inquiry_string, S_IRUGO | S_IWUSR, + mass_storage_inquiry_show, + mass_storage_inquiry_store); + +static struct device_attribute *mass_storage_function_attributes[] = { + &dev_attr_inquiry_string, + NULL +}; + +static struct android_usb_function mass_storage_function = { + .name = "mass_storage", + .init = mass_storage_function_init, + .cleanup = mass_storage_function_cleanup, + .bind_config = mass_storage_function_bind_config, + .attributes = mass_storage_function_attributes, +}; + + +static int accessory_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + return acc_setup(); +} + +static void accessory_function_cleanup(struct android_usb_function *f) +{ + acc_cleanup(); +} + +static int accessory_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + return acc_bind_config(c); +} + +static int accessory_function_ctrlrequest(struct android_usb_function *f, + struct usb_composite_dev *cdev, + const struct usb_ctrlrequest *c) +{ + return acc_ctrlrequest(cdev, c); +} + +static struct android_usb_function accessory_function = { + .name = "accessory", + .init = accessory_function_init, + .cleanup = accessory_function_cleanup, + .bind_config = accessory_function_bind_config, + .ctrlrequest = accessory_function_ctrlrequest, +}; + +static int audio_source_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + struct audio_source_config *config; + + config = kzalloc(sizeof(struct audio_source_config), GFP_KERNEL); + if (!config) + return -ENOMEM; + config->card = -1; + config->device = -1; + f->config = config; + return 0; +} + +static void audio_source_function_cleanup(struct android_usb_function *f) +{ + kfree(f->config); +} + +static int audio_source_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + struct audio_source_config *config = f->config; + + return audio_source_bind_config(c, config); +} + +static void audio_source_function_unbind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + struct audio_source_config *config = f->config; + + config->card = -1; + config->device = -1; +} + +static ssize_t audio_source_pcm_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct audio_source_config *config = f->config; + + /* print PCM card and device numbers */ + return sprintf(buf, "%d %d\n", config->card, config->device); +} + +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 android_usb_function audio_source_function = { + .name = "audio_source", + .init = audio_source_function_init, + .cleanup = audio_source_function_cleanup, + .bind_config = audio_source_function_bind_config, + .unbind_config = audio_source_function_unbind_config, + .attributes = audio_source_function_attributes, +}; + +static int midi_function_init(struct android_usb_function *f, + struct usb_composite_dev *cdev) +{ + struct midi_alsa_config *config; + + config = kzalloc(sizeof(struct midi_alsa_config), GFP_KERNEL); + f->config = config; + if (!config) + return -ENOMEM; + config->card = -1; + config->device = -1; + return 0; +} + +static void midi_function_cleanup(struct android_usb_function *f) +{ + kfree(f->config); +} + +static int midi_function_bind_config(struct android_usb_function *f, + struct usb_configuration *c) +{ + struct midi_alsa_config *config = f->config; + + return f_midi_bind_config(c, SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1, + MIDI_INPUT_PORTS, MIDI_OUTPUT_PORTS, MIDI_BUFFER_SIZE, + MIDI_QUEUE_LENGTH, config); +} + +static ssize_t midi_alsa_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + struct android_usb_function *f = dev_get_drvdata(dev); + struct midi_alsa_config *config = f->config; + + /* print ALSA card and device numbers */ + return sprintf(buf, "%d %d\n", config->card, config->device); +} + +static DEVICE_ATTR(alsa, S_IRUGO, midi_alsa_show, NULL); + +static struct device_attribute *midi_function_attributes[] = { + &dev_attr_alsa, + NULL +}; + +static struct android_usb_function midi_function = { + .name = "midi", + .init = midi_function_init, + .cleanup = midi_function_cleanup, + .bind_config = midi_function_bind_config, + .attributes = midi_function_attributes, +}; + +static struct android_usb_function *supported_functions[] = { + &ffs_function, + &acm_function, + &mtp_function, + &ptp_function, + &rndis_function, + &mass_storage_function, + &accessory_function, + &audio_source_function, + &midi_function, + NULL +}; + +static int android_init_functions(struct android_usb_function **functions, + struct usb_composite_dev *cdev) +{ + struct android_dev *dev = _android_dev; + struct android_usb_function *f; + struct device_attribute **attrs; + struct device_attribute *attr; + int err; + int index = 0; + + for (; (f = *functions++); index++) { + f->dev_name = kasprintf(GFP_KERNEL, "f_%s", f->name); + f->dev = device_create(android_class, dev->dev, + MKDEV(0, index), f, f->dev_name); + if (IS_ERR(f->dev)) { + pr_err("%s: Failed to create dev %s", __func__, + f->dev_name); + err = PTR_ERR(f->dev); + goto err_create; + } + + if (f->init) { + err = f->init(f, cdev); + if (err) { + pr_err("%s: Failed to init %s", __func__, + f->name); + goto err_out; + } + } + + attrs = f->attributes; + if (attrs) { + while ((attr = *attrs++) && !err) + err = device_create_file(f->dev, attr); + } + if (err) { + pr_err("%s: Failed to create function %s attributes", + __func__, f->name); + goto err_out; + } + } + return 0; + +err_out: + device_destroy(android_class, f->dev->devt); +err_create: + kfree(f->dev_name); + return err; +} + +static void android_cleanup_functions(struct android_usb_function **functions) +{ + struct android_usb_function *f; + + while (*functions) { + f = *functions++; + + if (f->dev) { + device_destroy(android_class, f->dev->devt); + kfree(f->dev_name); + } + + if (f->cleanup) + f->cleanup(f); + } +} + +static int +android_bind_enabled_functions(struct android_dev *dev, + struct usb_configuration *c) +{ + struct android_usb_function *f; + int ret; + + list_for_each_entry(f, &dev->enabled_functions, enabled_list) { + ret = f->bind_config(f, c); + if (ret) { + pr_err("%s: %s failed", __func__, f->name); + return ret; + } + } + return 0; +} + +static void +android_unbind_enabled_functions(struct android_dev *dev, + struct usb_configuration *c) +{ + struct android_usb_function *f; + + list_for_each_entry(f, &dev->enabled_functions, enabled_list) { + if (f->unbind_config) + f->unbind_config(f, c); + } +} + +static int android_enable_function(struct android_dev *dev, char *name) +{ + struct android_usb_function **functions = dev->functions; + struct android_usb_function *f; + while ((f = *functions++)) { + if (!strcmp(name, f->name)) { + list_add_tail(&f->enabled_list, + &dev->enabled_functions); + return 0; + } + } + return -EINVAL; +} + +/*-------------------------------------------------------------------------*/ +/* /sys/class/android_usb/android%d/ interface */ + +static ssize_t +functions_show(struct device *pdev, struct device_attribute *attr, char *buf) +{ + struct android_dev *dev = dev_get_drvdata(pdev); + struct android_usb_function *f; + char *buff = buf; + + mutex_lock(&dev->mutex); + + list_for_each_entry(f, &dev->enabled_functions, enabled_list) + buff += sprintf(buff, "%s,", f->name); + + mutex_unlock(&dev->mutex); + + if (buff != buf) + *(buff-1) = '\n'; + return buff - buf; +} + +static ssize_t +functions_store(struct device *pdev, struct device_attribute *attr, + const char *buff, size_t size) +{ + struct android_dev *dev = dev_get_drvdata(pdev); + char *name; + char buf[256], *b; + char aliases[256], *a; + int err; + int is_ffs; + int ffs_enabled = 0; + + mutex_lock(&dev->mutex); + + if (dev->enabled) { + mutex_unlock(&dev->mutex); + return -EBUSY; + } + + INIT_LIST_HEAD(&dev->enabled_functions); + + strlcpy(buf, buff, sizeof(buf)); + b = strim(buf); + + while (b) { + name = strsep(&b, ","); + if (!name) + continue; + + is_ffs = 0; + strlcpy(aliases, dev->ffs_aliases, sizeof(aliases)); + a = aliases; + + while (a) { + char *alias = strsep(&a, ","); + if (alias && !strcmp(name, alias)) { + is_ffs = 1; + break; + } + } + + if (is_ffs) { + if (ffs_enabled) + continue; + err = android_enable_function(dev, "ffs"); + if (err) + pr_err("android_usb: Cannot enable ffs (%d)", + err); + else + ffs_enabled = 1; + continue; + } + + err = android_enable_function(dev, name); + if (err) + pr_err("android_usb: Cannot enable '%s' (%d)", + name, err); + } + + mutex_unlock(&dev->mutex); + + return size; +} + +static ssize_t enable_show(struct device *pdev, struct device_attribute *attr, + char *buf) +{ + struct android_dev *dev = dev_get_drvdata(pdev); + return sprintf(buf, "%d\n", dev->enabled); +} + +static ssize_t enable_store(struct device *pdev, struct device_attribute *attr, + const char *buff, size_t size) +{ + struct android_dev *dev = dev_get_drvdata(pdev); + struct usb_composite_dev *cdev = dev->cdev; + struct android_usb_function *f; + int enabled = 0; + + + if (!cdev) + return -ENODEV; + + mutex_lock(&dev->mutex); + + sscanf(buff, "%d", &enabled); + if (enabled && !dev->enabled) { + /* + * Update values in composite driver's copy of + * device descriptor. + */ + cdev->desc.idVendor = device_desc.idVendor; + cdev->desc.idProduct = device_desc.idProduct; + cdev->desc.bcdDevice = device_desc.bcdDevice; + cdev->desc.bDeviceClass = device_desc.bDeviceClass; + cdev->desc.bDeviceSubClass = device_desc.bDeviceSubClass; + cdev->desc.bDeviceProtocol = device_desc.bDeviceProtocol; + list_for_each_entry(f, &dev->enabled_functions, enabled_list) { + if (f->enable) + f->enable(f); + } + android_enable(dev); + dev->enabled = true; + } else if (!enabled && dev->enabled) { + android_disable(dev); + list_for_each_entry(f, &dev->enabled_functions, enabled_list) { + if (f->disable) + f->disable(f); + } + dev->enabled = false; + } else { + pr_err("android_usb: already %s\n", + dev->enabled ? "enabled" : "disabled"); + } + + mutex_unlock(&dev->mutex); + return size; +} + +static ssize_t state_show(struct device *pdev, struct device_attribute *attr, + char *buf) +{ + struct android_dev *dev = dev_get_drvdata(pdev); + struct usb_composite_dev *cdev = dev->cdev; + char *state = "DISCONNECTED"; + unsigned long flags; + + 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); +} + +#define DESCRIPTOR_ATTR(field, format_string) \ +static ssize_t \ +field ## _show(struct device *dev, struct device_attribute *attr, \ + char *buf) \ +{ \ + return sprintf(buf, format_string, device_desc.field); \ +} \ +static ssize_t \ +field ## _store(struct device *dev, struct device_attribute *attr, \ + const char *buf, size_t size) \ +{ \ + int value; \ + if (sscanf(buf, format_string, &value) == 1) { \ + device_desc.field = value; \ + return size; \ + } \ + return -1; \ +} \ +static DEVICE_ATTR(field, S_IRUGO | S_IWUSR, field ## _show, field ## _store); + +#define DESCRIPTOR_STRING_ATTR(field, buffer) \ +static ssize_t \ +field ## _show(struct device *dev, struct device_attribute *attr, \ + char *buf) \ +{ \ + return sprintf(buf, "%s", buffer); \ +} \ +static ssize_t \ +field ## _store(struct device *dev, struct device_attribute *attr, \ + const char *buf, size_t size) \ +{ \ + if (size >= sizeof(buffer)) \ + return -EINVAL; \ + return strlcpy(buffer, buf, sizeof(buffer)); \ +} \ +static DEVICE_ATTR(field, S_IRUGO | S_IWUSR, field ## _show, field ## _store); + + +DESCRIPTOR_ATTR(idVendor, "%04x\n") +DESCRIPTOR_ATTR(idProduct, "%04x\n") +DESCRIPTOR_ATTR(bcdDevice, "%04x\n") +DESCRIPTOR_ATTR(bDeviceClass, "%d\n") +DESCRIPTOR_ATTR(bDeviceSubClass, "%d\n") +DESCRIPTOR_ATTR(bDeviceProtocol, "%d\n") +DESCRIPTOR_STRING_ATTR(iManufacturer, manufacturer_string) +DESCRIPTOR_STRING_ATTR(iProduct, product_string) +DESCRIPTOR_STRING_ATTR(iSerial, serial_string) + +static DEVICE_ATTR(functions, S_IRUGO | S_IWUSR, functions_show, + functions_store); +static DEVICE_ATTR(enable, S_IRUGO | S_IWUSR, enable_show, enable_store); +static DEVICE_ATTR(state, S_IRUGO, state_show, NULL); + +static struct device_attribute *android_usb_attributes[] = { + &dev_attr_idVendor, + &dev_attr_idProduct, + &dev_attr_bcdDevice, + &dev_attr_bDeviceClass, + &dev_attr_bDeviceSubClass, + &dev_attr_bDeviceProtocol, + &dev_attr_iManufacturer, + &dev_attr_iProduct, + &dev_attr_iSerial, + &dev_attr_functions, + &dev_attr_enable, + &dev_attr_state, + NULL +}; + +/*-------------------------------------------------------------------------*/ +/* Composite driver */ + +static int android_bind_config(struct usb_configuration *c) +{ + struct android_dev *dev = _android_dev; + int ret = 0; + + ret = android_bind_enabled_functions(dev, c); + if (ret) + return ret; + + return 0; +} + +static void android_unbind_config(struct usb_configuration *c) +{ + struct android_dev *dev = _android_dev; + + android_unbind_enabled_functions(dev, c); +} + +static int android_bind(struct usb_composite_dev *cdev) +{ + struct android_dev *dev = _android_dev; + struct usb_gadget *gadget = cdev->gadget; + int id, ret; + + /* Save the default handler */ + dev->setup_complete = cdev->req->complete; + + /* + * Start disconnected. Userspace will connect the gadget once + * it is done configuring the functions. + */ + usb_gadget_disconnect(gadget); + + ret = android_init_functions(dev->functions, cdev); + if (ret) + return ret; + + /* Allocate string descriptor numbers ... note that string + * contents can be overridden by the composite_dev glue. + */ + id = usb_string_id(cdev); + if (id < 0) + return id; + strings_dev[STRING_MANUFACTURER_IDX].id = id; + device_desc.iManufacturer = id; + + id = usb_string_id(cdev); + if (id < 0) + return id; + strings_dev[STRING_PRODUCT_IDX].id = id; + device_desc.iProduct = id; + + /* Default strings - should be updated by userspace */ + strncpy(manufacturer_string, "Android", sizeof(manufacturer_string)-1); + strncpy(product_string, "Android", sizeof(product_string) - 1); + strncpy(serial_string, "0123456789ABCDEF", sizeof(serial_string) - 1); + + id = usb_string_id(cdev); + if (id < 0) + return id; + strings_dev[STRING_SERIAL_IDX].id = id; + device_desc.iSerialNumber = id; + + usb_gadget_set_selfpowered(gadget); + dev->cdev = cdev; + + return 0; +} + +static int android_usb_unbind(struct usb_composite_dev *cdev) +{ + struct android_dev *dev = _android_dev; + + cancel_work_sync(&dev->work); + android_cleanup_functions(dev->functions); + return 0; +} + +/* HACK: android needs to override setup for accessory to work */ +static int (*composite_setup_func)(struct usb_gadget *gadget, const struct usb_ctrlrequest *c); + +static int +android_setup(struct usb_gadget *gadget, const struct usb_ctrlrequest *c) +{ + struct android_dev *dev = _android_dev; + struct usb_composite_dev *cdev = get_gadget_data(gadget); + struct usb_request *req = cdev->req; + struct android_usb_function *f; + int value = -EOPNOTSUPP; + unsigned long flags; + + req->zero = 0; + req->length = 0; + req->complete = dev->setup_complete; + gadget->ep0->driver_data = cdev; + + list_for_each_entry(f, &dev->enabled_functions, enabled_list) { + if (f->ctrlrequest) { + value = f->ctrlrequest(f, cdev, c); + if (value >= 0) + break; + } + } + + /* Special case the accessory function. + * It needs to handle control requests before it is enabled. + */ + if (value < 0) + value = acc_ctrlrequest(cdev, c); + + if (value < 0) + value = composite_setup_func(gadget, c); + + spin_lock_irqsave(&cdev->lock, flags); + if (!dev->connected) { + dev->connected = 1; + schedule_work(&dev->work); + } else if (c->bRequest == USB_REQ_SET_CONFIGURATION && + cdev->config) { + schedule_work(&dev->work); + } + spin_unlock_irqrestore(&cdev->lock, flags); + + return value; +} + +static void android_disconnect(struct usb_composite_dev *cdev) +{ + struct android_dev *dev = _android_dev; + + /* accessory HID support can be active while the + accessory function is not actually enabled, + so we need to inform it when we are disconnected. + */ + acc_disconnect(); + + dev->connected = 0; + schedule_work(&dev->work); +} + +static struct usb_composite_driver android_usb_driver = { + .name = "android_usb", + .dev = &device_desc, + .strings = dev_strings, + .bind = android_bind, + .unbind = android_usb_unbind, + .disconnect = android_disconnect, + .max_speed = USB_SPEED_HIGH, +}; + +static int android_create_device(struct android_dev *dev) +{ + struct device_attribute **attrs = android_usb_attributes; + struct device_attribute *attr; + int err; + + dev->dev = device_create(android_class, NULL, + MKDEV(0, 0), NULL, "android0"); + if (IS_ERR(dev->dev)) + return PTR_ERR(dev->dev); + + dev_set_drvdata(dev->dev, dev); + + while ((attr = *attrs++)) { + err = device_create_file(dev->dev, attr); + if (err) { + device_destroy(android_class, dev->dev->devt); + return err; + } + } + return 0; +} + + +static int __init init(void) +{ + struct android_dev *dev; + int err; + + android_class = class_create(THIS_MODULE, "android_usb"); + if (IS_ERR(android_class)) + return PTR_ERR(android_class); + + dev = kzalloc(sizeof(*dev), GFP_KERNEL); + if (!dev) { + err = -ENOMEM; + goto err_dev; + } + + dev->disable_depth = 1; + dev->functions = supported_functions; + INIT_LIST_HEAD(&dev->enabled_functions); + INIT_WORK(&dev->work, android_work); + mutex_init(&dev->mutex); + + err = android_create_device(dev); + if (err) { + pr_err("%s: failed to create android device %d", __func__, err); + goto err_create; + } + + _android_dev = dev; + + err = usb_composite_probe(&android_usb_driver); + if (err) { + pr_err("%s: failed to probe driver %d", __func__, err); + goto err_probe; + } + + /* HACK: exchange composite's setup with ours */ + composite_setup_func = android_usb_driver.gadget_driver.setup; + android_usb_driver.gadget_driver.setup = android_setup; + + return 0; + +err_probe: + device_destroy(android_class, dev->dev->devt); +err_create: + kfree(dev); +err_dev: + class_destroy(android_class); + return err; +} +late_initcall(init); + +static void __exit cleanup(void) +{ + usb_composite_unregister(&android_usb_driver); + class_destroy(android_class); + kfree(_android_dev); + _android_dev = NULL; +} +module_exit(cleanup); diff --git a/drivers/usb/gadget/composite.c b/drivers/usb/gadget/composite.c index a660716f9331..6a87860a32b9 100644 --- a/drivers/usb/gadget/composite.c +++ b/drivers/usb/gadget/composite.c @@ -812,7 +812,7 @@ done: } EXPORT_SYMBOL_GPL(usb_add_config); -static void remove_config(struct usb_composite_dev *cdev, +static void unbind_config(struct usb_composite_dev *cdev, struct usb_configuration *config) { while (!list_empty(&config->functions)) { @@ -827,7 +827,6 @@ static void remove_config(struct usb_composite_dev *cdev, /* may free memory for "f" */ } } - list_del(&config->list); if (config->unbind) { DBG(cdev, "unbind config '%s'/%p\n", config->label, config); config->unbind(config); @@ -854,9 +853,11 @@ void usb_remove_config(struct usb_composite_dev *cdev, if (cdev->config == config) reset_config(cdev); + list_del(&config->list); + spin_unlock_irqrestore(&cdev->lock, flags); - remove_config(cdev, config); + unbind_config(cdev, config); } /*-------------------------------------------------------------------------*/ @@ -1525,7 +1526,8 @@ static void __composite_unbind(struct usb_gadget *gadget, bool unbind_driver) struct usb_configuration *c; c = list_first_entry(&cdev->configs, struct usb_configuration, list); - remove_config(cdev, c); + list_del(&c->list); + unbind_config(cdev, c); } if (cdev->driver->unbind && unbind_driver) cdev->driver->unbind(cdev); diff --git a/drivers/usb/gadget/f_accessory.c b/drivers/usb/gadget/f_accessory.c new file mode 100644 index 000000000000..0237f1e059b4 --- /dev/null +++ b/drivers/usb/gadget/f_accessory.c @@ -0,0 +1,1215 @@ +/* + * 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> + +#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; + + /* set to 1 when we connect */ + int online:1; + /* 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; + +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->online = 0; + 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 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, +}; + +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; + + 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)); + _acc_dev->disconnected = 0; + 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 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; + } 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; + value = 0; + } else if (b_request == ACCESSORY_REGISTER_HID) { + value = acc_register_hid(dev, w_value, w_index); + } else if (b_request == ACCESSORY_UNREGISTER_HID) { + 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); + + /* 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; +} + +static int +acc_function_bind(struct usb_configuration *c, struct usb_function *f) +{ + 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); + + 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 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; + + 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; + + /* 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); + 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_bind_config(struct usb_configuration *c) +{ + struct acc_dev *dev = _acc_dev; + int ret; + + printk(KERN_INFO "acc_bind_config\n"); + + /* allocate a string ID for our interface */ + 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; + 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; + dev->function.unbind = acc_function_unbind; + dev->function.set_alt = acc_function_set_alt; + dev->function.disable = acc_function_disable; + + return usb_add_function(c, &dev->function); +} + +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; +} + +static void acc_disconnect(void) +{ + /* unregister all HID devices if USB is disconnected */ + kill_all_hid_devices(_acc_dev); +} + +static void acc_cleanup(void) +{ + misc_deregister(&acc_device); + kfree(_acc_dev); + _acc_dev = NULL; +} diff --git a/drivers/usb/gadget/f_audio_source.c b/drivers/usb/gadget/f_audio_source.c new file mode 100644 index 000000000000..07a1a3c28f3d --- /dev/null +++ b/drivers/usb/gadget/f_audio_source.c @@ -0,0 +1,834 @@ +/* + * 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 <sound/core.h> +#include <sound/initval.h> +#include <sound/pcm.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 + +/* B.3.1 Standard AC Interface Descriptor */ +static struct usb_interface_descriptor audio_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 audio_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 audio_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 audio_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 audio_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 audio_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 audio_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 audio_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 audio_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 audio_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 audio_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 audio_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 *)&audio_ac_interface_desc, + (struct usb_descriptor_header *)&audio_ac_header_desc, + + (struct usb_descriptor_header *)&audio_input_terminal_desc, + (struct usb_descriptor_header *)&audio_output_terminal_desc, + (struct usb_descriptor_header *)&audio_feature_unit_desc, + + (struct usb_descriptor_header *)&audio_as_interface_alt_0_desc, + (struct usb_descriptor_header *)&audio_as_interface_alt_1_desc, + (struct usb_descriptor_header *)&audio_as_header_desc, + + (struct usb_descriptor_header *)&audio_as_type_i_desc, + + (struct usb_descriptor_header *)&audio_hs_as_in_ep_desc, + (struct usb_descriptor_header *)&audio_as_iso_in_desc, + NULL, +}; + +static struct usb_descriptor_header *fs_audio_desc[] = { + (struct usb_descriptor_header *)&audio_ac_interface_desc, + (struct usb_descriptor_header *)&audio_ac_header_desc, + + (struct usb_descriptor_header *)&audio_input_terminal_desc, + (struct usb_descriptor_header *)&audio_output_terminal_desc, + (struct usb_descriptor_header *)&audio_feature_unit_desc, + + (struct usb_descriptor_header *)&audio_as_interface_alt_0_desc, + (struct usb_descriptor_header *)&audio_as_interface_alt_1_desc, + (struct usb_descriptor_header *)&audio_as_header_desc, + + (struct usb_descriptor_header *)&audio_as_type_i_desc, + + (struct usb_descriptor_header *)&audio_fs_as_in_ep_desc, + (struct usb_descriptor_header *)&audio_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; +}; + +static inline struct audio_dev *func_to_audio(struct usb_function *f) +{ + return container_of(f, struct audio_dev, func); +} + +/*-------------------------------------------------------------------------*/ + +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 = ktime_to_ns(now) - ktime_to_ns(audio->start_time); + do_div(msecs, 1000000); + frames = msecs * SAMPLE_RATE; + do_div(frames, 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_build_desc(struct audio_dev *audio) +{ + u8 *sam_freq; + int rate; + + /* Set channel numbers */ + audio_input_terminal_desc.bNrChannels = 2; + audio_as_type_i_desc.bNrChannels = 2; + + /* Set sample rates */ + rate = SAMPLE_RATE; + sam_freq = audio_as_type_i_desc.tSamFreq[0]; + memcpy(sam_freq, &rate, 3); +} + +/* 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; + + audio_build_desc(audio); + + /* allocate instance-specific interface IDs, and patch descriptors */ + status = usb_interface_id(c, f); + if (status < 0) + goto fail; + audio_ac_interface_desc.bInterfaceNumber = status; + + /* AUDIO_AC_INTERFACE */ + audio_ac_header_desc.baInterfaceNr[0] = status; + + status = usb_interface_id(c, f); + if (status < 0) + goto fail; + audio_as_interface_alt_0_desc.bInterfaceNumber = status; + audio_as_interface_alt_1_desc.bInterfaceNumber = status; + + /* AUDIO_AS_INTERFACE */ + audio_ac_header_desc.baInterfaceNr[1] = status; + + status = -ENODEV; + + /* allocate our endpoint */ + ep = usb_ep_autoconfig(cdev->gadget, &audio_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)) + audio_hs_as_in_ep_desc.bEndpointAddress = + audio_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; +} + +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; + 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); + 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, + }, + .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; + struct snd_card *card; + struct snd_pcm *pcm; + int err; + + config->card = -1; + config->device = -1; + + audio = &_audio_dev; + + err = snd_card_create(SNDRV_DEFAULT_IDX1, SNDRV_DEFAULT_STR1, + THIS_MODULE, 0, &card); + if (err) + return err; + + snd_card_set_dev(card, &c->cdev->gadget->dev); + + 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; + + err = usb_add_function(c, &audio->func); + if (err) + goto add_fail; + + config->card = pcm->card->number; + config->device = pcm->device; + audio->card = card; + return 0; + +add_fail: +register_fail: +pcm_fail: + snd_card_free(audio->card); + return err; +} diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c index 84219f656051..c36f660dcc2e 100644 --- a/drivers/usb/gadget/f_fs.c +++ b/drivers/usb/gadget/f_fs.c @@ -1563,7 +1563,12 @@ static int ffs_func_eps_enable(struct ffs_function *func) spin_lock_irqsave(&func->ffs->eps_lock, flags); do { struct usb_endpoint_descriptor *ds; - ds = ep->descs[ep->descs[1] ? 1 : 0]; + int desc_idx = ffs->gadget->speed == USB_SPEED_HIGH ? 1 : 0; + ds = ep->descs[desc_idx]; + if (!ds) { + ret = -EINVAL; + break; + } ep->ep->driver_data = ep; ep->ep->desc = ds; diff --git a/drivers/usb/gadget/f_midi.c b/drivers/usb/gadget/f_midi.c index 263e721c2694..b0bce7ea31fa 100644 --- a/drivers/usb/gadget/f_midi.c +++ b/drivers/usb/gadget/f_midi.c @@ -65,6 +65,11 @@ struct gmidi_in_port { uint8_t data[2]; }; +struct midi_alsa_config { + int card; + int device; +}; + struct f_midi { struct usb_function func; struct usb_gadget *gadget; @@ -97,7 +102,7 @@ DECLARE_USB_MIDI_OUT_JACK_DESCRIPTOR(1); DECLARE_USB_MS_ENDPOINT_DESCRIPTOR(16); /* B.3.1 Standard AC Interface Descriptor */ -static struct usb_interface_descriptor ac_interface_desc __initdata = { +static struct usb_interface_descriptor ac_interface_desc /* __initdata */ = { .bLength = USB_DT_INTERFACE_SIZE, .bDescriptorType = USB_DT_INTERFACE, /* .bInterfaceNumber = DYNAMIC */ @@ -108,7 +113,7 @@ static struct usb_interface_descriptor ac_interface_desc __initdata = { }; /* B.3.2 Class-Specific AC Interface Descriptor */ -static struct uac1_ac_header_descriptor_1 ac_header_desc __initdata = { +static struct uac1_ac_header_descriptor_1 ac_header_desc /* __initdata */ = { .bLength = UAC_DT_AC_HEADER_SIZE(1), .bDescriptorType = USB_DT_CS_INTERFACE, .bDescriptorSubtype = USB_MS_HEADER, @@ -119,7 +124,7 @@ static struct uac1_ac_header_descriptor_1 ac_header_desc __initdata = { }; /* B.4.1 Standard MS Interface Descriptor */ -static struct usb_interface_descriptor ms_interface_desc __initdata = { +static struct usb_interface_descriptor ms_interface_desc /* __initdata */ = { .bLength = USB_DT_INTERFACE_SIZE, .bDescriptorType = USB_DT_INTERFACE, /* .bInterfaceNumber = DYNAMIC */ @@ -130,7 +135,7 @@ static struct usb_interface_descriptor ms_interface_desc __initdata = { }; /* B.4.2 Class-Specific MS Interface Descriptor */ -static struct usb_ms_header_descriptor ms_header_desc __initdata = { +static struct usb_ms_header_descriptor ms_header_desc /* __initdata */ = { .bLength = USB_DT_MS_HEADER_SIZE, .bDescriptorType = USB_DT_CS_INTERFACE, .bDescriptorSubtype = USB_MS_HEADER, @@ -191,7 +196,7 @@ static struct usb_gadget_strings *midi_strings[] = { NULL, }; -static struct usb_request *alloc_ep_req(struct usb_ep *ep, unsigned length) +static struct usb_request *midi_alloc_ep_req(struct usb_ep *ep, unsigned length) { struct usb_request *req; @@ -207,7 +212,7 @@ static struct usb_request *alloc_ep_req(struct usb_ep *ep, unsigned length) return req; } -static void free_ep_req(struct usb_ep *ep, struct usb_request *req) +static void midi_free_ep_req(struct usb_ep *ep, struct usb_request *req) { kfree(req->buf); usb_ep_free_request(ep, req); @@ -278,7 +283,7 @@ f_midi_complete(struct usb_ep *ep, struct usb_request *req) if (ep == midi->out_ep) f_midi_handle_out_data(ep, req); - free_ep_req(ep, req); + midi_free_ep_req(ep, req); return; case -EOVERFLOW: /* buffer overrun on read means that @@ -365,7 +370,7 @@ static int f_midi_set_alt(struct usb_function *f, unsigned intf, unsigned alt) /* allocate a bunch of read buffers and queue them all at once. */ for (i = 0; i < midi->qlen && err == 0; i++) { struct usb_request *req = - alloc_ep_req(midi->out_ep, midi->buflen); + midi_alloc_ep_req(midi->out_ep, midi->buflen); if (req == NULL) return -ENOMEM; @@ -409,7 +414,7 @@ static void f_midi_unbind(struct usb_configuration *c, struct usb_function *f) card = midi->card; midi->card = NULL; if (card) - snd_card_free(card); + snd_card_free_when_closed(card); kfree(midi->id); midi->id = NULL; @@ -546,10 +551,10 @@ static void f_midi_transmit(struct f_midi *midi, struct usb_request *req) return; if (!req) - req = alloc_ep_req(ep, midi->buflen); + req = midi_alloc_ep_req(ep, midi->buflen); if (!req) { - ERROR(midi, "gmidi_transmit: alloc_ep_request failed\n"); + ERROR(midi, "gmidi_transmit: midi_alloc_ep_request failed\n"); return; } req->length = 0; @@ -575,7 +580,7 @@ static void f_midi_transmit(struct f_midi *midi, struct usb_request *req) if (req->length > 0) usb_ep_queue(ep, req, GFP_ATOMIC); else - free_ep_req(ep, req); + midi_free_ep_req(ep, req); } static void f_midi_in_tasklet(unsigned long data) @@ -733,7 +738,7 @@ fail: /* MIDI function driver setup/binding */ -static int __init +static int /* __init */ f_midi_bind(struct usb_configuration *c, struct usb_function *f) { struct usb_descriptor_header **midi_function; @@ -923,16 +928,22 @@ fail: * * Returns zero on success, else negative errno. */ -int __init f_midi_bind_config(struct usb_configuration *c, +int /* __init */ f_midi_bind_config(struct usb_configuration *c, int index, char *id, unsigned int in_ports, unsigned int out_ports, unsigned int buflen, - unsigned int qlen) + unsigned int qlen, + struct midi_alsa_config* config) { struct f_midi *midi; int status, i; + if (config) { + config->card = -1; + config->device = -1; + } + /* sanity check */ if (in_ports > MAX_PORTS || out_ports > MAX_PORTS) return -EINVAL; @@ -961,6 +972,10 @@ int __init f_midi_bind_config(struct usb_configuration *c, tasklet_init(&midi->tasklet, f_midi_in_tasklet, (unsigned long) midi); /* set up ALSA midi devices */ + midi->id = kstrdup(id, GFP_KERNEL); + midi->index = index; + midi->buflen = buflen; + midi->qlen = qlen; midi->in_ports = in_ports; midi->out_ports = out_ports; status = f_midi_register_card(midi); @@ -974,15 +989,16 @@ int __init f_midi_bind_config(struct usb_configuration *c, midi->func.set_alt = f_midi_set_alt; midi->func.disable = f_midi_disable; - midi->id = kstrdup(id, GFP_KERNEL); - midi->index = index; - midi->buflen = buflen; - midi->qlen = qlen; - status = usb_add_function(c, &midi->func); if (status) goto setup_fail; + + if (config) { + config->card = midi->rmidi->card->number; + config->device = midi->rmidi->device; + } + return 0; setup_fail: diff --git a/drivers/usb/gadget/f_mtp.c b/drivers/usb/gadget/f_mtp.c new file mode 100644 index 000000000000..620aeaaf2d72 --- /dev/null +++ b/drivers/usb/gadget/f_mtp.c @@ -0,0 +1,1287 @@ +/* + * 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> + +#define MTP_BULK_BUFFER_SIZE 16384 +#define INTR_BUFFER_SIZE 28 + +/* 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 + +static const char mtp_shortname[] = "mtp_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_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_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 *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_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 = __constant_cpu_to_le16(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; +}; + +/* 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: + printk(KERN_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; + + DBG(cdev, "mtp_read(%zu)\n", count); + + if (count > MTP_BULK_BUFFER_SIZE) + return -EINVAL; + + /* 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->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 = count; + 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; + header->length = __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 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; + + 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; + + /* 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; + } + + 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 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; + + 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; +} + +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_bind_config(struct usb_configuration *c, bool ptp_config) +{ + struct mtp_dev *dev = _mtp_dev; + int ret = 0; + + printk(KERN_INFO "mtp_bind_config\n"); + + /* allocate a string ID for our interface */ + 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; + } + + dev->cdev = c->cdev; + dev->function.name = "mtp"; + dev->function.strings = mtp_strings; + if (ptp_config) { + dev->function.fs_descriptors = fs_ptp_descs; + dev->function.hs_descriptors = hs_ptp_descs; + } else { + dev->function.fs_descriptors = fs_mtp_descs; + dev->function.hs_descriptors = hs_mtp_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; + + return usb_add_function(c, &dev->function); +} + +static int mtp_setup(void) +{ + struct mtp_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); + 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 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); +} diff --git a/drivers/usb/gadget/f_rndis.c b/drivers/usb/gadget/f_rndis.c index 36e8c44d8e5e..5c274f1d7b7a 100644 --- a/drivers/usb/gadget/f_rndis.c +++ b/drivers/usb/gadget/f_rndis.c @@ -25,7 +25,6 @@ #include "u_ether.h" #include "rndis.h" - /* * This function is an RNDIS Ethernet port -- a Microsoft protocol that's * been promoted instead of the standard CDC Ethernet. The published RNDIS @@ -67,6 +66,16 @@ * - 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; @@ -448,6 +457,7 @@ static void rndis_command_complete(struct usb_ep *ep, struct usb_request *req) { struct f_rndis *rndis = req->context; int status; + rndis_init_msg_type *buf; /* received RNDIS command from USB_CDC_SEND_ENCAPSULATED_COMMAND */ // spin_lock(&dev->lock); @@ -455,6 +465,21 @@ 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; + pr_info("%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); } @@ -748,6 +773,7 @@ rndis_bind(struct usb_configuration *c, struct usb_function *f) rndis_set_param_medium(rndis->config, RNDIS_MEDIUM_802_3, 0); rndis_set_host_mac(rndis->config, rndis->ethaddr); + rndis_set_max_pkt_xfer(rndis->config, rndis_ul_max_pkt_per_xfer); if (rndis->manufacturer && rndis->vendorID && rndis_set_param_vendor(rndis->config, rndis->vendorID, @@ -821,12 +847,12 @@ rndis_bind_config_vendor(struct usb_configuration *c, u8 ethaddr[ETH_ALEN], if (!can_support_rndis(c) || !ethaddr) return -EINVAL; - if (rndis_string_defs[0].id == 0) { - /* ... and setup RNDIS itself */ - status = rndis_init(); - if (status < 0) - return status; + /* setup RNDIS itself */ + status = rndis_init(); + if (status < 0) + return status; + if (rndis_string_defs[0].id == 0) { status = usb_string_ids_tab(c->cdev, rndis_string_defs); if (status) return status; @@ -854,6 +880,8 @@ rndis_bind_config_vendor(struct usb_configuration *c, u8 ethaddr[ETH_ALEN], 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"; rndis->port.func.strings = rndis_strings; diff --git a/drivers/usb/gadget/rndis.c b/drivers/usb/gadget/rndis.c index 1e4cfb05f70b..cb2767df3fba 100644 --- a/drivers/usb/gadget/rndis.c +++ b/drivers/usb/gadget/rndis.c @@ -59,6 +59,16 @@ MODULE_PARM_DESC (rndis_debug, "enable debugging"); #define RNDIS_MAX_CONFIGS 1 +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"); + static rndis_params rndis_per_dev_params[RNDIS_MAX_CONFIGS]; @@ -585,12 +595,12 @@ static int rndis_init_response(int configNr, rndis_init_msg_type *buf) 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); @@ -686,6 +696,12 @@ static int rndis_reset_response(int configNr, rndis_reset_msg_type *buf) rndis_reset_cmplt_type *resp; rndis_resp_t *r; struct rndis_params *params = rndis_per_dev_params + configNr; + u32 length; + u8 *xbuf; + + /* drain the response queue */ + while ((xbuf = rndis_get_next_response(configNr, &length))) + rndis_free_response(configNr, xbuf); r = rndis_add_response(configNr, sizeof(rndis_reset_cmplt_type)); if (!r) @@ -910,6 +926,8 @@ int rndis_set_param_dev(u8 configNr, struct net_device *dev, u16 *cdc_filter) rndis_per_dev_params[configNr].dev = dev; rndis_per_dev_params[configNr].filter = cdc_filter; + rndis_ul_max_xfer_size_rcvd = 0; + rndis_ul_max_pkt_per_xfer_rcvd = 0; return 0; } @@ -936,6 +954,13 @@ int rndis_set_param_medium(u8 configNr, u32 medium, u32 speed) return 0; } +void rndis_set_max_pkt_xfer(u8 configNr, u8 max_pkt_per_xfer) +{ + pr_debug("%s:\n", __func__); + + rndis_per_dev_params[configNr].max_pkt_per_xfer = max_pkt_per_xfer; +} + void rndis_add_hdr(struct sk_buff *skb) { struct rndis_packet_msg_type *header; @@ -1008,23 +1033,73 @@ 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; + + /* 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:%u", + skb->len, sizeof *hdr); + dev_kfree_skb_any(skb); + return -EINVAL; + } - /* DataOffset, DataLength */ - if (!skb_pull(skb, get_unaligned_le32(tmp++) + 8)) { - dev_kfree_skb_any(skb); - return -EOVERFLOW; + 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; @@ -1127,11 +1202,15 @@ static struct proc_dir_entry *rndis_connect_state [RNDIS_MAX_CONFIGS]; #endif /* CONFIG_USB_GADGET_DEBUG_FILES */ +static bool rndis_initialized; int rndis_init(void) { u8 i; + if (rndis_initialized) + return 0; + for (i = 0; i < RNDIS_MAX_CONFIGS; i++) { #ifdef CONFIG_USB_GADGET_DEBUG_FILES char name [20]; @@ -1158,6 +1237,7 @@ int rndis_init(void) INIT_LIST_HEAD(&(rndis_per_dev_params[i].resp_queue)); } + rndis_initialized = true; return 0; } @@ -1166,7 +1246,13 @@ void rndis_exit(void) #ifdef CONFIG_USB_GADGET_DEBUG_FILES u8 i; char name[20]; +#endif + + if (!rndis_initialized) + return; + rndis_initialized = false; +#ifdef CONFIG_USB_GADGET_DEBUG_FILES for (i = 0; i < RNDIS_MAX_CONFIGS; i++) { sprintf(name, NAME_TEMPLATE, i); remove_proc_entry(name, NULL); diff --git a/drivers/usb/gadget/rndis.h b/drivers/usb/gadget/rndis.h index 0647f2f34e89..12045b31a311 100644 --- a/drivers/usb/gadget/rndis.h +++ b/drivers/usb/gadget/rndis.h @@ -189,6 +189,7 @@ 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; @@ -204,6 +205,7 @@ int rndis_set_param_dev (u8 configNr, struct net_device *dev, int rndis_set_param_vendor (u8 configNr, u32 vendorID, const char *vendorDescr); int rndis_set_param_medium (u8 configNr, u32 medium, u32 speed); +void rndis_set_max_pkt_xfer(u8 configNr, 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/u_ether.c b/drivers/usb/gadget/u_ether.c index 4b76124ce96b..14f587efc0f6 100644 --- a/drivers/usb/gadget/u_ether.c +++ b/drivers/usb/gadget/u_ether.c @@ -48,6 +48,8 @@ #define UETH__VERSION "29-May-2008" +static struct workqueue_struct *uether_wq; + struct eth_dev { /* lock is held while accessing port_usb */ @@ -59,17 +61,25 @@ struct eth_dev { spinlock_t req_lock; /* guard {rx,tx}_reqs */ struct list_head tx_reqs, rx_reqs; - atomic_t tx_qlen; + unsigned 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 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 @@ -84,7 +94,7 @@ struct eth_dev { #define DEFAULT_QLEN 2 /* double buffering by default */ -static unsigned qmult = 5; +static unsigned qmult = 10; module_param(qmult, uint, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(qmult, "queue length multiplier at high/super speed"); @@ -226,9 +236,13 @@ 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: %d\n", __func__, size); skb = alloc_skb(size + NET_IP_ALIGN, gfp_flags); if (skb == NULL) { DBG(dev, "no rx skb\n"); @@ -254,18 +268,16 @@ 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) { @@ -281,6 +293,10 @@ 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; @@ -289,30 +305,9 @@ 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 > VLAN_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 */ @@ -335,22 +330,20 @@ 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) @@ -415,16 +408,24 @@ 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)) + 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; } @@ -434,6 +435,36 @@ 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); @@ -451,6 +482,11 @@ 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: @@ -461,16 +497,74 @@ 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)) netif_wake_queue(dev->net); } @@ -480,6 +574,26 @@ 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) { @@ -506,6 +620,10 @@ 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 (!is_promisc(cdc_filter)) { u8 *dest = skb->data; @@ -560,11 +678,39 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb, spin_unlock_irqrestore(&dev->lock, flags); if (!skb) goto drop; + } + + 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->buf = skb->data; - req->context = skb; + req->complete = tx_complete; /* NCM requires no zlp if transfer is dwNtbInMaxSize */ @@ -579,17 +725,26 @@ 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) - ? ((atomic_read(&dev->tx_qlen) % qmult) != 0) - : 0; + /* throttle highspeed IRQ rate back slightly */ + if (gadget_is_dualspeed(dev->gadget) && + (dev->gadget->speed == USB_SPEED_HIGH)) { + dev->tx_qlen++; + if (dev->tx_qlen == (qmult/2)) { + req->no_interrupt = 0; + dev->tx_qlen = 0; + } else { + req->no_interrupt = 1; + } + } else { + req->no_interrupt = 0; + } retval = usb_ep_queue(in, req, GFP_ATOMIC); switch (retval) { @@ -598,11 +753,11 @@ static netdev_tx_t eth_start_xmit(struct sk_buff *skb, break; case 0: net->trans_start = jiffies; - atomic_inc(&dev->tx_qlen); } 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++; spin_lock_irqsave(&dev->req_lock, flags); @@ -611,6 +766,7 @@ drop: list_add(&req->list, &dev->tx_reqs); spin_unlock_irqrestore(&dev->req_lock, flags); } +success: return NETDEV_TX_OK; } @@ -624,7 +780,7 @@ static void eth_start(struct eth_dev *dev, gfp_t gfp_flags) rx_fill(dev, gfp_flags); /* and open the tx floodgates */ - atomic_set(&dev->tx_qlen, 0); + dev->tx_qlen = 0; netif_wake_queue(dev->net); } @@ -697,6 +853,8 @@ static int eth_stop(struct net_device *net) /*-------------------------------------------------------------------------*/ +static u8 host_ethaddr[ETH_ALEN]; + /* initial value, changed by "ifconfig usb0 hw ether xx:xx:xx:xx:xx:xx" */ static char *dev_addr; module_param(dev_addr, charp, S_IRUGO); @@ -728,6 +886,17 @@ static int get_ether_addr(const char *str, u8 *dev_addr) return 1; } +static int get_host_ether_addr(u8 *str, u8 *dev_addr) +{ + memcpy(dev_addr, str, ETH_ALEN); + if (is_valid_ether_addr(dev_addr)) + return 0; + + random_ether_addr(dev_addr); + memcpy(str, dev_addr, ETH_ALEN); + return 1; +} + static const struct net_device_ops eth_netdev_ops = { .ndo_open = eth_open, .ndo_stop = eth_stop, @@ -770,6 +939,7 @@ struct eth_dev *gether_setup_name(struct usb_gadget *g, u8 ethaddr[ETH_ALEN], 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); @@ -782,9 +952,11 @@ struct eth_dev *gether_setup_name(struct usb_gadget *g, u8 ethaddr[ETH_ALEN], if (get_ether_addr(dev_addr, net->dev_addr)) dev_warn(&g->dev, "using random %s ethernet address\n", "self"); - if (get_ether_addr(host_addr, dev->host_mac)) - dev_warn(&g->dev, - "using random %s ethernet address\n", "host"); + + if (get_host_ether_addr(host_ethaddr, dev->host_mac)) + dev_warn(&g->dev, "using random %s ethernet address\n", "host"); + else + dev_warn(&g->dev, "using previous %s ethernet address\n", "host"); if (ethaddr) memcpy(ethaddr, dev->host_mac, ETH_ALEN); @@ -882,8 +1054,13 @@ 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) @@ -927,6 +1104,7 @@ 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) @@ -949,6 +1127,8 @@ 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); } @@ -968,6 +1148,12 @@ 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->driver_data = NULL; link->out_ep->desc = NULL; @@ -980,3 +1166,23 @@ void gether_disconnect(struct gether *link) dev->port_usb = NULL; spin_unlock(&dev->lock); } + +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_DESCRIPTION("ethernet over USB driver"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/usb/gadget/u_ether.h b/drivers/usb/gadget/u_ether.h index 02522338a708..67eda50ae995 100644 --- a/drivers/usb/gadget/u_ether.h +++ b/drivers/usb/gadget/u_ether.h @@ -54,6 +54,9 @@ 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; struct sk_buff *(*wrap)(struct gether *port, struct sk_buff *skb); int (*unwrap)(struct gether *port, diff --git a/drivers/usb/gadget/u_serial.c b/drivers/usb/gadget/u_serial.c index b369292d4b90..72068082e83c 100644 --- a/drivers/usb/gadget/u_serial.c +++ b/drivers/usb/gadget/u_serial.c @@ -1126,6 +1126,7 @@ int gserial_alloc_line(unsigned char *line_num) tty_dev = tty_port_register_device(&ports[port_num].port->port, gs_tty_driver, port_num, NULL); + if (IS_ERR(tty_dev)) { struct gs_port *port; pr_err("%s: failed to register tty for port %d, err %ld\n", diff --git a/drivers/usb/gadget/udc-core.c b/drivers/usb/gadget/udc-core.c index 817a26cbfab1..58a861395ea0 100644 --- a/drivers/usb/gadget/udc-core.c +++ b/drivers/usb/gadget/udc-core.c @@ -335,7 +335,15 @@ static int udc_bind_to_driver(struct usb_udc *udc, struct usb_gadget_driver *dri driver->unbind(udc->gadget); goto err1; } - usb_gadget_connect(udc->gadget); + /* + * HACK: The Android gadget driver disconnects the gadget + * on bind and expects the gadget to stay disconnected until + * it calls usb_gadget_connect when userspace is ready. Remove + * the call to usb_gadget_connect bellow to avoid enabling the + * pullup before userspace is ready. + * + * usb_gadget_connect(udc->gadget); + */ kobject_uevent(&udc->dev.kobj, KOBJ_CHANGE); return 0; diff --git a/drivers/usb/phy/Kconfig b/drivers/usb/phy/Kconfig index be41733a5eac..0dd0da0d0ac5 100644 --- a/drivers/usb/phy/Kconfig +++ b/drivers/usb/phy/Kconfig @@ -16,6 +16,14 @@ menuconfig USB_PHY If you're not sure if this applies to you, it probably doesn't; say N here. +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. + if USB_PHY # diff --git a/drivers/usb/phy/Makefile b/drivers/usb/phy/Makefile index a9169cb1e6fc..a0a6cbad8806 100644 --- a/drivers/usb/phy/Makefile +++ b/drivers/usb/phy/Makefile @@ -5,6 +5,7 @@ ccflags-$(CONFIG_USB_DEBUG) := -DDEBUG obj-$(CONFIG_USB_PHY) += phy.o +obj-$(CONFIG_USB_OTG_WAKELOCK) += otg-wakelock.o # transceiver drivers, keep the list sorted 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 29a5121ce7fd..a3279c7def71 100644 --- a/drivers/video/Kconfig +++ b/drivers/video/Kconfig @@ -2496,6 +2496,7 @@ source "drivers/video/omap2/Kconfig" source "drivers/video/exynos/Kconfig" source "drivers/video/mmp/Kconfig" source "drivers/video/backlight/Kconfig" +source "drivers/video/adf/Kconfig" if VT source "drivers/video/console/Kconfig" diff --git a/drivers/video/Makefile b/drivers/video/Makefile index 33869eea4981..3adbd32eb091 100644 --- a/drivers/video/Makefile +++ b/drivers/video/Makefile @@ -12,6 +12,7 @@ fb-y := fbmem.o fbmon.o fbcmap.o fbsysfs.o \ modedb.o fbcvt.o fb-objs := $(fb-y) +obj-$(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..33858b73d8bb --- /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 + tristate "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..78d0915122f4 --- /dev/null +++ b/drivers/video/adf/Makefile @@ -0,0 +1,15 @@ +ccflags-y := -Idrivers/staging/android + +CFLAGS_adf.o := -I$(src) + +obj-$(CONFIG_ADF) += adf.o \ + adf_client.o \ + adf_fops.o \ + adf_format.o \ + adf_sysfs.o + +obj-$(CONFIG_COMPAT) += adf_fops32.o + +obj-$(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..7fbf33e1cb39 --- /dev/null +++ b/drivers/video/adf/adf_fops.c @@ -0,0 +1,957 @@ +/* + * 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, arg->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(arg->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; + size_t n_intfs, n_bufs, i; + void *custom_data = NULL; + size_t custom_data_size; + int ret = 0; + + complete_fence_fd = get_unused_fd(); + if (complete_fence_fd < 0) + return complete_fence_fd; + + if (get_user(n_intfs, &arg->n_interfaces)) { + ret = -EFAULT; + goto err_get_user; + } + + if (n_intfs > ADF_MAX_INTERFACES) { + ret = -EINVAL; + goto err_get_user; + } + + if (get_user(n_bufs, &arg->n_bufs)) { + ret = -EFAULT; + goto err_get_user; + } + + if (n_bufs > ADF_MAX_BUFFERS) { + ret = -EINVAL; + goto err_get_user; + } + + if (get_user(custom_data_size, &arg->custom_data_size)) { + ret = -EFAULT; + goto err_get_user; + } + + if (custom_data_size > ADF_MAX_CUSTOM_DATA_SIZE) { + ret = -EINVAL; + goto err_get_user; + } + + if (n_intfs) { + intfs = kmalloc(sizeof(intfs[0]) * n_intfs, GFP_KERNEL); + if (!intfs) { + ret = -ENOMEM; + goto err_get_user; + } + } + + for (i = 0; i < n_intfs; i++) { + u32 intf_id; + if (get_user(intf_id, &arg->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 (n_bufs) { + bufs = kzalloc(sizeof(bufs[0]) * n_bufs, GFP_KERNEL); + if (!bufs) { + ret = -ENOMEM; + goto err_get_user; + } + } + + for (i = 0; i < n_bufs; i++) { + ret = adf_buffer_import(dev, &arg->bufs[i], &bufs[i]); + if (ret < 0) { + memset(&bufs[i], 0, sizeof(bufs[i])); + goto err_import; + } + } + + if (custom_data_size) { + custom_data = kzalloc(custom_data_size, GFP_KERNEL); + if (!custom_data) { + ret = -ENOMEM; + goto err_import; + } + + if (copy_from_user(custom_data, arg->custom_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, n_intfs, bufs, + n_bufs, custom_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 < 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(); + 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, arg->custom_data, + &data.custom_data_size); + mutex_unlock(&dev->client_lock); + + if (ret < 0) + goto done; + + ret = adf_copy_attachment_list_to_user(arg->attachments, + data.n_attachments, attach, n_attach); + if (ret < 0) + goto done; + + ret = adf_copy_attachment_list_to_user(arg->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(arg->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, arg->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..e73a7d59f1e6 --- /dev/null +++ b/drivers/video/adf/adf_memblock.c @@ -0,0 +1,160 @@ +/* + * 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; + + 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; + buf = dma_buf_export(pdata, &adf_memblock_ops, size, flags); + 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(×tamp, &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/w1/masters/ds2482.c b/drivers/w1/masters/ds2482.c index e033491fe308..ab29939ea4cd 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) @@ -94,10 +97,16 @@ static const struct i2c_device_id ds2482_id[] = { { } }; +static const struct dev_pm_ops ds2482_pm_ops = { + .suspend = ds2482_suspend, + .resume = ds2482_resume, +}; + static struct i2c_driver ds2482_driver = { .driver = { .owner = THIS_MODULE, .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/adfs/super.c b/fs/adfs/super.c index 0ff4bae2c2a2..479ef5a9c5d3 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -213,6 +213,7 @@ static int parse_options(struct super_block *sb, char *options) static int adfs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_NODIRATIME; return parse_options(sb, data); } diff --git a/fs/affs/super.c b/fs/affs/super.c index 45161a832bbc..01c6e07acbff 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -549,6 +549,7 @@ affs_remount(struct super_block *sb, int *flags, char *data) pr_debug("AFFS: remount(flags=0x%x,opts=\"%s\")\n",*flags,data); + sync_filesystem(sb); *flags |= MS_NODIRATIME; memcpy(volume, sbi->s_volume, 32); diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index f95dddced968..7192a7e99373 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -908,6 +908,7 @@ befs_fill_super(struct super_block *sb, void *data, int silent) static int befs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); if (!(*flags & MS_RDONLY)) return -EINVAL; return 0; diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index f0857e092a3c..461731c7fa3a 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1239,6 +1239,7 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data) unsigned int old_metadata_ratio = fs_info->metadata_ratio; int ret; + sync_filesystem(sb); btrfs_remount_prepare(fs_info); ret = btrfs_parse_options(root, data); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 3752b9f6d9e4..0d7daf714c4b 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -521,6 +521,7 @@ static int cifs_show_stats(struct seq_file *s, struct dentry *root) static int cifs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_NODIRATIME; return 0; } diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index bfbf4700d160..b70aa7c91394 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -447,7 +447,7 @@ wait_for_response(struct TCP_Server_Info *server, struct mid_q_entry *midQ) { int error; - error = wait_event_freezekillable(server->response_q, + error = wait_event_freezekillable_unsafe(server->response_q, midQ->mid_state != MID_REQUEST_SUBMITTED); if (error < 0) return -ERESTARTSYS; diff --git a/fs/coda/dir.c b/fs/coda/dir.c index b7d3a05c062c..fc66861b3598 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -391,8 +391,7 @@ static int coda_readdir(struct file *coda_file, void *buf, filldir_t filldir) if (!host_file->f_op) return -ENOTDIR; - if (host_file->f_op->readdir) - { + if (host_file->f_op->readdir) { /* potemkin case: we were handed a directory inode. * We can't use vfs_readdir because we have to keep the file * position in sync between the coda_file and the host_file. @@ -410,8 +409,20 @@ static int coda_readdir(struct file *coda_file, void *buf, filldir_t filldir) coda_file->f_pos = host_file->f_pos; mutex_unlock(&host_inode->i_mutex); - } - else /* Venus: we must read Venus dirents from a file */ + } else if (host_file->f_op->iterate) { + struct inode *host_inode = file_inode(host_file); + struct dir_context *ctx = buf; + + mutex_lock(&host_inode->i_mutex); + ret = -ENOENT; + if (!IS_DEADDIR(host_inode)) { + ret = host_file->f_op->iterate(host_file, ctx); + file_accessed(host_file); + } + mutex_unlock(&host_inode->i_mutex); + + coda_file->f_pos = ctx->pos; + } else /* Venus: we must read Venus dirents from a file */ ret = coda_venus_readdir(coda_file, buf, filldir); return ret; diff --git a/fs/coda/inode.c b/fs/coda/inode.c index 4dcc0d81a7aa..0aa4c4d75ec6 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -96,6 +96,7 @@ void coda_destroy_inodecache(void) static int coda_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_NOATIME; return 0; } diff --git a/fs/compat.c b/fs/compat.c index fc3b55dce184..6af20de2c1a3 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -832,6 +832,7 @@ struct compat_old_linux_dirent { }; struct compat_readdir_callback { + struct dir_context ctx; struct compat_old_linux_dirent __user *dirent; int result; }; @@ -873,15 +874,15 @@ asmlinkage long compat_sys_old_readdir(unsigned int fd, { int error; struct fd f = fdget(fd); - struct compat_readdir_callback buf; + struct compat_readdir_callback buf = { + .ctx.actor = compat_fillonedir, + .dirent = dirent + }; if (!f.file) return -EBADF; - buf.result = 0; - buf.dirent = dirent; - - error = vfs_readdir(f.file, compat_fillonedir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (buf.result) error = buf.result; @@ -897,6 +898,7 @@ struct compat_linux_dirent { }; struct compat_getdents_callback { + struct dir_context ctx; struct compat_linux_dirent __user *current_dir; struct compat_linux_dirent __user *previous; int count; @@ -951,7 +953,11 @@ asmlinkage long compat_sys_getdents(unsigned int fd, { struct fd f; struct compat_linux_dirent __user * lastdirent; - struct compat_getdents_callback buf; + struct compat_getdents_callback buf = { + .ctx.actor = compat_filldir, + .current_dir = dirent, + .count = count + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -961,17 +967,12 @@ asmlinkage long compat_sys_getdents(unsigned int fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, compat_filldir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(f.file->f_pos, &lastdirent->d_off)) + if (put_user(buf.ctx.pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; @@ -983,6 +984,7 @@ asmlinkage long compat_sys_getdents(unsigned int fd, #ifndef __ARCH_OMIT_COMPAT_SYS_GETDENTS64 struct compat_getdents_callback64 { + struct dir_context ctx; struct linux_dirent64 __user *current_dir; struct linux_dirent64 __user *previous; int count; @@ -1036,7 +1038,11 @@ asmlinkage long compat_sys_getdents64(unsigned int fd, { struct fd f; struct linux_dirent64 __user * lastdirent; - struct compat_getdents_callback64 buf; + struct compat_getdents_callback64 buf = { + .ctx.actor = compat_filldir64, + .current_dir = dirent, + .count = count + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -1046,17 +1052,12 @@ asmlinkage long compat_sys_getdents64(unsigned int fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, compat_filldir64, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - typeof(lastdirent->d_off) d_off = f.file->f_pos; + typeof(lastdirent->d_off) d_off = buf.ctx.pos; if (__put_user_unaligned(d_off, &lastdirent->d_off)) error = -EFAULT; else diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 35b1c7bd18b7..c0148585670d 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -227,6 +227,7 @@ static void cramfs_put_super(struct super_block *sb) static int cramfs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_RDONLY; return 0; } diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 26d7fff8d78e..84378f0ffd6d 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -218,6 +218,7 @@ static int debugfs_remount(struct super_block *sb, int *flags, char *data) int err; struct debugfs_fs_info *fsi = sb->s_fs_info; + sync_filesystem(sb); err = debugfs_parse_options(data, &fsi->mount_opts); if (err) goto fail; diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c index a726b9f29cb7..c71038079b47 100644 --- a/fs/devpts/inode.c +++ b/fs/devpts/inode.c @@ -313,6 +313,7 @@ static int devpts_remount(struct super_block *sb, int *flags, char *data) struct pts_fs_info *fsi = DEVPTS_SB(sb); struct pts_mount_opts *opts = &fsi->mount_opts; + sync_filesystem(sb); err = parse_mount_options(data, PARSE_REMOUNT, opts); /* diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 9ff3664bb3ea..4912bf47226c 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -68,6 +68,7 @@ static ssize_t ecryptfs_read_update_atime(struct kiocb *iocb, } struct ecryptfs_getdents_callback { + struct dir_context ctx; void *dirent; struct dentry *dentry; filldir_t filldir; @@ -115,18 +116,19 @@ static int ecryptfs_readdir(struct file *file, void *dirent, filldir_t filldir) int rc; struct file *lower_file; struct inode *inode; - struct ecryptfs_getdents_callback buf; + struct ecryptfs_getdents_callback buf = { + .dirent = dirent, + .dentry = file->f_path.dentry, + .filldir = filldir, + .filldir_called = 0, + .entries_written = 0, + .ctx.actor = ecryptfs_filldir + }; lower_file = ecryptfs_file_to_lower(file); lower_file->f_pos = file->f_pos; inode = file_inode(file); - memset(&buf, 0, sizeof(buf)); - buf.dirent = dirent; - buf.dentry = file->f_path.dentry; - buf.filldir = filldir; - buf.filldir_called = 0; - buf.entries_written = 0; - rc = vfs_readdir(lower_file, ecryptfs_filldir, (void *)&buf); + rc = iterate_dir(lower_file, &buf.ctx); file->f_pos = lower_file->f_pos; if (rc < 0) goto out; diff --git a/fs/efs/super.c b/fs/efs/super.c index c6f57a74a559..4709692cd7a0 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -113,6 +113,7 @@ static void efs_put_super(struct super_block *s) static int efs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_RDONLY; return 0; } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index deecc7294a67..db10e00c971a 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> @@ -1602,7 +1603,8 @@ 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); @@ -1817,8 +1819,7 @@ SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd, goto error_tgt_fput; /* Check if EPOLLWAKEUP is allowed */ - if ((epds.events & EPOLLWAKEUP) && !capable(CAP_BLOCK_SUSPEND)) - epds.events &= ~EPOLLWAKEUP; + ep_take_care_of_epollwakeup(&epds); /* * We have to check that the file structure underneath the file descriptor diff --git a/fs/exec.c b/fs/exec.c index dd6aa61c8548..cb7f31c71c6b 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1220,7 +1220,7 @@ EXPORT_SYMBOL(install_exec_creds); /* * determine how safe it is to execute the proposed program * - the caller must hold ->cred_guard_mutex to protect against - * PTRACE_ATTACH + * PTRACE_ATTACH or seccomp thread-sync */ static int check_unsafe_exec(struct linux_binprm *bprm) { @@ -1239,7 +1239,7 @@ static int check_unsafe_exec(struct linux_binprm *bprm) * This isn't strictly necessary, but it makes it harder for LSMs to * mess up. */ - if (current->no_new_privs) + if (task_no_new_privs(current)) bprm->unsafe |= LSM_UNSAFE_NO_NEW_PRIVS; n_fs = 1; @@ -1286,7 +1286,7 @@ int prepare_binprm(struct linux_binprm *bprm) bprm->cred->egid = current_egid(); if (!(bprm->file->f_path.mnt->mnt_flags & MNT_NOSUID) && - !current->no_new_privs && + !task_no_new_privs(current) && kuid_has_mapping(bprm->cred->user_ns, inode->i_uid) && kgid_has_mapping(bprm->cred->user_ns, inode->i_gid)) { /* Set-uid? */ diff --git a/fs/exportfs/expfs.c b/fs/exportfs/expfs.c index b4eec4c9a790..dc4784b52d92 100644 --- a/fs/exportfs/expfs.c +++ b/fs/exportfs/expfs.c @@ -212,6 +212,7 @@ reconnect_path(struct vfsmount *mnt, struct dentry *target_dir, char *nbuf) } struct getdents_callback { + struct dir_context ctx; char *name; /* name that was found. It already points to a buffer NAME_MAX+1 is size */ unsigned long ino; /* the inum we are looking for */ @@ -254,7 +255,11 @@ static int get_name(const struct path *path, char *name, struct dentry *child) struct inode *dir = path->dentry->d_inode; int error; struct file *file; - struct getdents_callback buffer; + struct getdents_callback buffer = { + .ctx.actor = filldir_one, + .name = name, + .ino = child->d_inode->i_ino + }; error = -ENOTDIR; if (!dir || !S_ISDIR(dir->i_mode)) @@ -271,17 +276,14 @@ static int get_name(const struct path *path, char *name, struct dentry *child) goto out; error = -EINVAL; - if (!file->f_op->readdir) + if (!file->f_op->readdir && !file->f_op->iterate) goto out_close; - buffer.name = name; - buffer.ino = child->d_inode->i_ino; - buffer.found = 0; buffer.sequence = 0; while (1) { int old_seq = buffer.sequence; - error = vfs_readdir(file, filldir_one, &buffer); + error = iterate_dir(file, &buffer.ctx); if (buffer.found) { error = 0; break; diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 20d6697bd638..d260115c0350 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -1254,6 +1254,7 @@ static int ext2_remount (struct super_block * sb, int * flags, char * data) unsigned long old_sb_flags; int err; + sync_filesystem(sb); spin_lock(&sbi->s_lock); /* Store the old options */ diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 882d4bdfd428..b1e30be1945b 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -2588,6 +2588,8 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) int i; #endif + sync_filesystem(sb); + /* Store the original options */ old_sb_flags = sb->s_flags; old_opts.s_mount_opt = sbi->s_mount_opt; diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 2a71466b0115..e5a77bb30ba3 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -2063,7 +2063,8 @@ 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 */ struct buffer_head *ext4_getblk(handle_t *, struct inode *, diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index d4fd81c44f55..8855db670425 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -605,11 +605,13 @@ 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; @@ -617,13 +619,15 @@ 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; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index df5050f9080b..aed1a67ab3a2 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2707,7 +2707,8 @@ 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; @@ -2716,7 +2717,7 @@ 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); } /* @@ -2738,7 +2739,7 @@ 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" @@ -4790,7 +4791,8 @@ 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" @@ -4997,13 +4999,15 @@ 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) { struct ext4_free_extent ex; int ret = 0; @@ -5022,7 +5026,7 @@ static int ext4_trim_extent(struct super_block *sb, int start, int count, */ 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; @@ -5035,6 +5039,7 @@ static int ext4_trim_extent(struct super_block *sb, int start, int count, * @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 @@ -5049,7 +5054,7 @@ static int ext4_trim_extent(struct super_block *sb, int start, int count, 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; @@ -5082,7 +5087,8 @@ 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; @@ -5124,6 +5130,7 @@ 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 @@ -5132,7 +5139,8 @@ 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; @@ -5188,7 +5196,7 @@ 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/super.c b/fs/ext4/super.c index 21a0b43a7d31..913a6b279220 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -4622,6 +4622,8 @@ static int ext4_remount(struct super_block *sb, int *flags, char *data) #endif char *orig_data = kstrdup(data, GFP_KERNEL); + sync_filesystem(sb); + /* Store the original options */ old_sb_flags = sb->s_flags; old_opts.s_mount_opt = sbi->s_mount_opt; diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 8555f7df82c7..03ab8b830940 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -239,7 +239,6 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) if (test_opt(sbi, DISABLE_EXT_IDENTIFY)) seq_puts(seq, ",disable_ext_identify"); - seq_printf(seq, ",active_logs=%u", sbi->active_logs); return 0; } diff --git a/fs/fat/dir.c b/fs/fat/dir.c index 7a6f02caf286..4b775e606fc8 100644 --- a/fs/fat/dir.c +++ b/fs/fat/dir.c @@ -776,6 +776,13 @@ static int fat_ioctl_readdir(struct inode *inode, struct file *filp, return ret; } +static int fat_ioctl_volume_id(struct inode *dir) +{ + struct super_block *sb = dir->i_sb; + struct msdos_sb_info *sbi = MSDOS_SB(sb); + return sbi->vol_id; +} + static long fat_dir_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { @@ -792,6 +799,8 @@ static long fat_dir_ioctl(struct file *filp, unsigned int cmd, short_only = 0; both = 1; break; + case VFAT_IOCTL_GET_VOLUME_ID: + return fat_ioctl_volume_id(inode); default: return fat_generic_ioctl(filp, cmd, arg); } @@ -832,6 +841,8 @@ static long fat_compat_dir_ioctl(struct file *filp, unsigned cmd, short_only = 0; both = 1; break; + case VFAT_IOCTL_GET_VOLUME_ID: + return fat_ioctl_volume_id(inode); default: return fat_generic_ioctl(filp, cmd, (unsigned long)arg); } diff --git a/fs/fat/fat.h b/fs/fat/fat.h index 21664fcf3616..8180ecef59fc 100644 --- a/fs/fat/fat.h +++ b/fs/fat/fat.h @@ -86,6 +86,7 @@ struct msdos_sb_info { const void *dir_ops; /* Opaque; default directory operations */ int dir_per_block; /* dir entries per block */ int dir_per_block_bits; /* log2(dir_per_block) */ + unsigned long vol_id; /* volume ID */ int fatent_shift; struct fatent_operations *fatent_ops; diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 5d4513cb1b3c..ebdc6656c067 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -632,6 +632,8 @@ static int fat_remount(struct super_block *sb, int *flags, char *data) struct msdos_sb_info *sbi = MSDOS_SB(sb); *flags |= MS_NODIRATIME | (sbi->options.isvfat ? 0 : MS_NOATIME); + sync_filesystem(sb); + /* make sure we update state on remount. */ new_rdonly = *flags & MS_RDONLY; if (new_rdonly != (sb->s_flags & MS_RDONLY)) { @@ -1252,6 +1254,7 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, struct inode *fsinfo_inode = NULL; struct buffer_head *bh; struct fat_boot_sector *b; + struct fat_boot_bsx *bsx; struct msdos_sb_info *sbi; u16 logical_sector_size; u32 total_sectors, total_clusters, fat_clusters, rootdir_sectors; @@ -1398,6 +1401,8 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, goto out_fail; } + bsx = (struct fat_boot_bsx *)(bh->b_data + FAT32_BSX_OFFSET); + fsinfo = (struct fat_boot_fsinfo *)fsinfo_bh->b_data; if (!IS_FSINFO(fsinfo)) { fat_msg(sb, KERN_WARNING, "Invalid FSINFO signature: " @@ -1413,8 +1418,14 @@ int fat_fill_super(struct super_block *sb, void *data, int silent, int isvfat, } brelse(fsinfo_bh); + } else { + bsx = (struct fat_boot_bsx *)(bh->b_data + FAT16_BSX_OFFSET); } + /* interpret volume ID as a little endian 32 bit integer */ + sbi->vol_id = (((u32)bsx->vol_id[0]) | ((u32)bsx->vol_id[1] << 8) | + ((u32)bsx->vol_id[2] << 16) | ((u32)bsx->vol_id[3] << 24)); + sbi->dir_per_block = sb->s_blocksize / sizeof(struct msdos_dir_entry); sbi->dir_per_block_bits = ffs(sbi->dir_per_block) - 1; diff --git a/fs/freevxfs/vxfs_super.c b/fs/freevxfs/vxfs_super.c index e37eb274e492..7ca8c75d50d3 100644 --- a/fs/freevxfs/vxfs_super.c +++ b/fs/freevxfs/vxfs_super.c @@ -124,6 +124,7 @@ vxfs_statfs(struct dentry *dentry, struct kstatfs *bufp) static int vxfs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_RDONLY; return 0; } diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index b44306378193..62d426d1d5e3 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1170,7 +1170,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) if ((inode->i_state & flags) == flags) 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/fuse/dev.c b/fs/fuse/dev.c index b535008b6c4c..f4e875a29e7a 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -20,6 +20,7 @@ #include <linux/swap.h> #include <linux/splice.h> #include <linux/aio.h> +#include <linux/freezer.h> MODULE_ALIAS_MISCDEV(FUSE_MINOR); MODULE_ALIAS("devname:fuse"); @@ -464,7 +465,10 @@ __acquires(fc->lock) * Wait it out. */ spin_unlock(&fc->lock); - wait_event(req->waitq, req->state == FUSE_REQ_FINISHED); + + while (req->state != FUSE_REQ_FINISHED) + wait_event_freezable(req->waitq, + req->state == FUSE_REQ_FINISHED); spin_lock(&fc->lock); if (!req->aborted) diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 39a986e1da9e..8aabbe2a3989 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -135,6 +135,7 @@ static void fuse_evict_inode(struct inode *inode) static int fuse_remount_fs(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); if (*flags & MS_MANDLOCK) return -EINVAL; diff --git a/fs/gfs2/export.c b/fs/gfs2/export.c index 9973df4ff565..7102c7a5af4d 100644 --- a/fs/gfs2/export.c +++ b/fs/gfs2/export.c @@ -88,7 +88,10 @@ static int gfs2_get_name(struct dentry *parent, char *name, struct inode *dir = parent->d_inode; struct inode *inode = child->d_inode; struct gfs2_inode *dip, *ip; - struct get_name_filldir gnfd; + struct get_name_filldir gnfd = { + .ctx.actor = get_name_filldir, + .name = name + }; struct gfs2_holder gh; u64 offset = 0; int error; @@ -106,7 +109,6 @@ static int gfs2_get_name(struct dentry *parent, char *name, *name = 0; gnfd.inum.no_addr = ip->i_no_addr; gnfd.inum.no_formal_ino = ip->i_no_formal_ino; - gnfd.name = name; error = gfs2_glock_nq_init(dip->i_gl, LM_ST_SHARED, 0, &gh); if (error) diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index e5639dec66c4..db7fff5a29c1 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -1142,6 +1142,8 @@ static int gfs2_remount_fs(struct super_block *sb, int *flags, char *data) struct gfs2_tune *gt = &sdp->sd_tune; int error; + sync_filesystem(sb); + spin_lock(>->gt_spin); args.ar_commit = gt->gt_logd_secs; args.ar_quota_quantum = gt->gt_quota_quantum; diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 2d2039e754cd..eee7206c38d1 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -112,6 +112,7 @@ static int hfs_statfs(struct dentry *dentry, struct kstatfs *buf) static int hfs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_NODIRATIME; if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) return 0; diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 4c4d142cf890..1b9414f701e6 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -323,6 +323,7 @@ static int hfsplus_statfs(struct dentry *dentry, struct kstatfs *buf) static int hfsplus_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) return 0; if (!(*flags & MS_RDONLY)) { diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index 962e90c37aec..d37f6ff8e07f 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -395,6 +395,8 @@ static int hpfs_remount_fs(struct super_block *s, int *flags, char *data) struct hpfs_sb_info *sbi = hpfs_sb(s); char *new_opts = kstrdup(data, GFP_KERNEL); + sync_filesystem(s); + *flags |= MS_NOATIME; hpfs_lock(s); diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 0defb1cc2a35..0918f0e2e266 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -243,6 +243,7 @@ static int jffs2_remount_fs(struct super_block *sb, int *flags, char *data) struct jffs2_sb_info *c = JFFS2_SB_INFO(sb); int err; + sync_filesystem(sb); err = jffs2_parse_options(c, data); if (err) return -EINVAL; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 788e0a9c1fb0..b7486dafa1cd 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -413,6 +413,7 @@ static int jfs_remount(struct super_block *sb, int *flags, char *data) int flag = JFS_SBI(sb)->flag; int ret; + sync_filesystem(sb); if (!parse_options(data, sb, &newLVSize, &flag)) { return -EINVAL; } diff --git a/fs/minix/inode.c b/fs/minix/inode.c index df122496f328..a54d08865e59 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -123,6 +123,7 @@ static int minix_remount (struct super_block * sb, int * flags, char * data) struct minix_sb_info * sbi = minix_sb(sb); struct minix_super_block * ms; + sync_filesystem(sb); ms = sbi->s_ms; if ((*flags & MS_RDONLY) == (sb->s_flags & MS_RDONLY)) return 0; diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index 26910c8154da..3f54348b926b 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -99,6 +99,7 @@ static void destroy_inodecache(void) static int ncp_remount(struct super_block *sb, int *flags, char* data) { + sync_filesystem(sb); *flags |= MS_NODIRATIME; return 0; } diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index e9be01b2cc5a..e2bb3012d025 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -79,7 +79,7 @@ int nfs_wait_bit_killable(void *word) { if (fatal_signal_pending(current)) return -ERESTARTSYS; - freezable_schedule(); + freezable_schedule_unsafe(); return 0; } EXPORT_SYMBOL_GPL(nfs_wait_bit_killable); diff --git a/fs/nfs/nfs3proc.c b/fs/nfs/nfs3proc.c index 43ea96ced28c..ce90eb4775c2 100644 --- a/fs/nfs/nfs3proc.c +++ b/fs/nfs/nfs3proc.c @@ -33,7 +33,7 @@ nfs3_rpc_wrapper(struct rpc_clnt *clnt, struct rpc_message *msg, int flags) res = rpc_call_sync(clnt, msg, flags); if (res != -EJUKEBOX) break; - freezable_schedule_timeout_killable(NFS_JUKEBOX_RETRY_TIME); + freezable_schedule_timeout_killable_unsafe(NFS_JUKEBOX_RETRY_TIME); res = -ERESTARTSYS; } while (!fatal_signal_pending(current)); return res; diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 20ebcfa3c92e..86390c3a95db 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -268,7 +268,7 @@ static int nfs4_delay(struct rpc_clnt *clnt, long *timeout) *timeout = NFS4_POLL_RETRY_MIN; if (*timeout > NFS4_POLL_RETRY_MAX) *timeout = NFS4_POLL_RETRY_MAX; - freezable_schedule_timeout_killable(*timeout); + freezable_schedule_timeout_killable_unsafe(*timeout); if (fatal_signal_pending(current)) res = -ERESTARTSYS; *timeout <<= 1; @@ -4558,7 +4558,7 @@ int nfs4_proc_delegreturn(struct inode *inode, struct rpc_cred *cred, const nfs4 static unsigned long nfs4_set_lock_task_retry(unsigned long timeout) { - freezable_schedule_timeout_killable(timeout); + freezable_schedule_timeout_killable_unsafe(timeout); timeout <<= 1; if (timeout > NFS4_LOCK_MAXTIMEOUT) return NFS4_LOCK_MAXTIMEOUT; diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 2d7525fbcf25..d85c1b819145 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -2133,6 +2133,8 @@ nfs_remount(struct super_block *sb, int *flags, char *raw_data) struct nfs4_mount_data *options4 = (struct nfs4_mount_data *)raw_data; u32 nfsvers = nfss->nfs_client->rpc_ops->version; + sync_filesystem(sb); + /* * Userspace mount programs that send binary options generally send * them populated with default values. We have no way to know which diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 4e9a21db867a..105a3b080d12 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -240,11 +240,16 @@ struct name_list { struct list_head list; }; +struct nfs4_dir_ctx { + struct dir_context ctx; + struct list_head names; +}; + static int nfsd4_build_namelist(void *arg, const char *name, int namlen, loff_t offset, u64 ino, unsigned int d_type) { - struct list_head *names = arg; + struct nfs4_dir_ctx *ctx = arg; struct name_list *entry; if (namlen != HEXDIR_LEN - 1) @@ -254,7 +259,7 @@ nfsd4_build_namelist(void *arg, const char *name, int namlen, return -ENOMEM; memcpy(entry->name, name, HEXDIR_LEN - 1); entry->name[HEXDIR_LEN - 1] = '\0'; - list_add(&entry->list, names); + list_add(&entry->list, &ctx->names); return 0; } @@ -263,7 +268,10 @@ nfsd4_list_rec_dir(recdir_func *f, struct nfsd_net *nn) { const struct cred *original_cred; struct dentry *dir = nn->rec_file->f_path.dentry; - LIST_HEAD(names); + struct nfs4_dir_ctx ctx = { + .ctx.actor = nfsd4_build_namelist, + .names = LIST_HEAD_INIT(ctx.names) + }; int status; status = nfs4_save_creds(&original_cred); @@ -276,11 +284,11 @@ nfsd4_list_rec_dir(recdir_func *f, struct nfsd_net *nn) return status; } - status = vfs_readdir(nn->rec_file, nfsd4_build_namelist, &names); + status = iterate_dir(nn->rec_file, &ctx.ctx); mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_PARENT); - while (!list_empty(&names)) { + while (!list_empty(&ctx.names)) { struct name_list *entry; - entry = list_entry(names.next, struct name_list, list); + entry = list_entry(ctx.names.next, struct name_list, list); if (!status) { struct dentry *dentry; dentry = lookup_one_len(entry->name, dir, HEXDIR_LEN-1); diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 81325ba8660a..11224fa5ff25 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1950,6 +1950,7 @@ struct buffered_dirent { }; struct readdir_data { + struct dir_context ctx; char *dirent; size_t used; int full; @@ -1981,13 +1982,15 @@ static int nfsd_buffered_filldir(void *__buf, const char *name, int namlen, static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func, struct readdir_cd *cdp, loff_t *offsetp) { - struct readdir_data buf; struct buffered_dirent *de; int host_err; int size; loff_t offset; + struct readdir_data buf = { + .ctx.actor = nfsd_buffered_filldir, + .dirent = (void *)__get_free_page(GFP_KERNEL) + }; - buf.dirent = (void *)__get_free_page(GFP_KERNEL); if (!buf.dirent) return nfserrno(-ENOMEM); @@ -2001,7 +2004,7 @@ static __be32 nfsd_buffered_readdir(struct file *file, filldir_t func, buf.used = 0; buf.full = 0; - host_err = vfs_readdir(file, nfsd_buffered_filldir, &buf); + host_err = iterate_dir(file, &buf.ctx); if (buf.full) host_err = 0; diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index c7d1f9f18b09..4b0a8d4a8345 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -1114,6 +1114,7 @@ static int nilfs_remount(struct super_block *sb, int *flags, char *data) unsigned long old_mount_opt; int err; + sync_filesystem(sb); old_sb_flags = sb->s_flags; old_mount_opt = nilfs->ns_mount_opt; diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index 82650d52d916..bd5610d48242 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -468,6 +468,8 @@ static int ntfs_remount(struct super_block *sb, int *flags, char *opt) ntfs_debug("Entering with remount options string: %s", opt); + sync_filesystem(sb); + #ifndef NTFS_RW /* For read-only compiled driver, enforce read-only flag. */ *flags |= MS_RDONLY; diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 01b85165552b..616bde667d8d 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -632,6 +632,8 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data) struct ocfs2_super *osb = OCFS2_SB(sb); u32 tmp; + sync_filesystem(sb); + if (!ocfs2_parse_options(sb, data, &parsed_options, 1) || !ocfs2_check_set_options(sb, &parsed_options)) { ret = -EINVAL; diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 75885ffde44e..f4026aba26f3 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -375,6 +375,7 @@ static struct inode *openprom_iget(struct super_block *sb, ino_t ino) static int openprom_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_NOATIME; return 0; } diff --git a/fs/proc/base.c b/fs/proc/base.c index 8fc784aef0b8..809d92a68a50 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -2749,8 +2749,8 @@ static const struct pid_entry tgid_base_stuff[] = { REG("cgroup", S_IRUGO, proc_cgroup_operations), #endif INF("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), @@ -3106,8 +3106,8 @@ static const struct pid_entry tid_base_stuff[] = { REG("cgroup", S_IRUGO, proc_cgroup_operations), #endif INF("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/root.c b/fs/proc/root.c index 9459710c55ae..8ead812d9897 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -92,6 +92,8 @@ static int proc_parse_options(char *options, struct pid_namespace *pid) int proc_remount(struct super_block *sb, int *flags, char *data) { struct pid_namespace *pid = sb->s_fs_info; + + sync_filesystem(sb); return !proc_parse_options(data, pid); } diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 9f285fb9bab3..0e2899492133 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -134,6 +134,56 @@ 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 vm_area_struct *vma) { if (vma && vma != priv->tail_vma) { @@ -335,6 +385,12 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid) pad_len_spaces(m, len); seq_printf(m, "[stack:%d]", tid); } + goto done; + } + + if (vma_get_anon_name(vma)) { + pad_len_spaces(m, len); + seq_print_vma_name(m, vma); } } @@ -634,6 +690,12 @@ static int show_smap(struct seq_file *m, void *v, int is_pid) show_smap_vma_flags(m, vma); + if (vma_get_anon_name(vma)) { + seq_puts(m, "Name: "); + seq_print_vma_name(m, vma); + seq_putc(m, '\n'); + } + if (m->count < m->size) /* vma is copied successfully */ m->version = (vma != get_gate_vma(task->mm)) ? vma->vm_start : 0; @@ -722,6 +784,7 @@ static int clear_refs_pte_range(pmd_t *pmd, unsigned long addr, #define CLEAR_REFS_ALL 1 #define CLEAR_REFS_ANON 2 #define CLEAR_REFS_MAPPED 3 +#define CLEAR_REFS_MM_HIWATER_RSS 5 static ssize_t clear_refs_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos) @@ -741,7 +804,8 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, rv = kstrtoint(strstrip(buffer), 10, &type); if (rv < 0) return rv; - if (type < CLEAR_REFS_ALL || type > CLEAR_REFS_MAPPED) + if ((type < CLEAR_REFS_ALL || type > CLEAR_REFS_MAPPED) && + type != CLEAR_REFS_MM_HIWATER_RSS) return -EINVAL; task = get_proc_task(file_inode(file)); if (!task) @@ -752,6 +816,18 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, .pmd_entry = clear_refs_pte_range, .mm = mm, }; + + if (type == CLEAR_REFS_MM_HIWATER_RSS) { + /* + * Writing 5 to /proc/pid/clear_refs resets the peak + * resident set size to this mm's current rss value. + */ + down_write(&mm->mmap_sem); + reset_mm_hiwater_rss(mm); + up_write(&mm->mmap_sem); + goto out_mm; + } + down_read(&mm->mmap_sem); for (vma = mm->mmap; vma; vma = vma->vm_next) { clear_refs_walk.private = vma; @@ -775,6 +851,7 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, } flush_tlb_mm(mm); up_read(&mm->mmap_sem); +out_mm: mmput(mm); } put_task_struct(task); diff --git a/fs/pstore/Kconfig b/fs/pstore/Kconfig index ca71db69da07..da12fd4e3d47 100644 --- a/fs/pstore/Kconfig +++ b/fs/pstore/Kconfig @@ -19,6 +19,16 @@ config PSTORE_CONSOLE When the option is enabled, pstore will log all kernel messages, even if no oops or panic happened. +config PSTORE_PMSG + bool "Log user space messages" + depends on PSTORE + help + When the option is enabled, pstore will export a character + interface /dev/pmsg0 to log user space messages. On reboot + data can be retrieved from /sys/fs/pstore/pmsg-ramoops-[ID]. + + If unsure, say N. + config PSTORE_FTRACE bool "Persistent function tracer" depends on PSTORE diff --git a/fs/pstore/Makefile b/fs/pstore/Makefile index 4c9095c2781e..e647d8e81712 100644 --- a/fs/pstore/Makefile +++ b/fs/pstore/Makefile @@ -7,5 +7,7 @@ obj-y += pstore.o pstore-objs += inode.o platform.o obj-$(CONFIG_PSTORE_FTRACE) += ftrace.o +obj-$(CONFIG_PSTORE_PMSG) += pmsg.o + ramoops-objs += ram.o ram_core.o obj-$(CONFIG_PSTORE_RAM) += ramoops.o diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c index 66c8c2fe86b7..a12e0829d8ac 100644 --- a/fs/pstore/inode.c +++ b/fs/pstore/inode.c @@ -249,6 +249,7 @@ static void parse_options(char *options) static int pstore_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); parse_options(data); return 0; @@ -315,22 +316,27 @@ int pstore_mkfile(enum pstore_type_id type, char *psname, u64 id, int count, switch (type) { case PSTORE_TYPE_DMESG: - sprintf(name, "dmesg-%s-%lld", psname, id); + scnprintf(name, sizeof(name), "dmesg-%s-%lld", + psname, id); break; case PSTORE_TYPE_CONSOLE: - sprintf(name, "console-%s-%lld", psname, id); + scnprintf(name, sizeof(name), "console-%s", psname); break; case PSTORE_TYPE_FTRACE: - sprintf(name, "ftrace-%s-%lld", psname, id); + scnprintf(name, sizeof(name), "ftrace-%s", psname); break; case PSTORE_TYPE_MCE: - sprintf(name, "mce-%s-%lld", psname, id); + scnprintf(name, sizeof(name), "mce-%s-%lld", psname, id); + break; + case PSTORE_TYPE_PMSG: + scnprintf(name, sizeof(name), "pmsg-%s-%lld", psname, id); break; case PSTORE_TYPE_UNKNOWN: - sprintf(name, "unknown-%s-%lld", psname, id); + scnprintf(name, sizeof(name), "unknown-%s-%lld", psname, id); break; default: - sprintf(name, "type%d-%s-%lld", type, psname, id); + scnprintf(name, sizeof(name), "type%d-%s-%lld", + type, psname, id); break; } diff --git a/fs/pstore/internal.h b/fs/pstore/internal.h index 937d820f273c..fd8d248c285c 100644 --- a/fs/pstore/internal.h +++ b/fs/pstore/internal.h @@ -45,6 +45,12 @@ extern void pstore_register_ftrace(void); static inline void pstore_register_ftrace(void) {} #endif +#ifdef CONFIG_PSTORE_PMSG +extern void pstore_register_pmsg(void); +#else +static inline void pstore_register_pmsg(void) {} +#endif + extern struct pstore_info *psinfo; extern void pstore_set_kmsg_bytes(int); diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index 86d1038b5a12..4f11f2382e24 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -267,6 +267,7 @@ int pstore_register(struct pstore_info *psi) kmsg_dump_register(&pstore_dumper); pstore_register_console(); pstore_register_ftrace(); + pstore_register_pmsg(); if (pstore_update_ms >= 0) { pstore_timer.expires = jiffies + diff --git a/fs/pstore/pmsg.c b/fs/pstore/pmsg.c new file mode 100644 index 000000000000..db47810a92d8 --- /dev/null +++ b/fs/pstore/pmsg.c @@ -0,0 +1,114 @@ +/* + * Copyright 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/cdev.h> +#include <linux/device.h> +#include <linux/fs.h> +#include <linux/uaccess.h> +#include <linux/vmalloc.h> +#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; + + if (!count) + return 0; + + 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); + + 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, c, + psinfo); + + i += c; + } + + mutex_unlock(&pmsg_lock); + vfree(buffer); + return count; +} + +static const struct file_operations pmsg_fops = { + .owner = THIS_MODULE, + .llseek = noop_llseek, + .write = write_pmsg, +}; + +static struct class *pmsg_class; +static int pmsg_major; +#define PMSG_NAME "pmsg" +#undef pr_fmt +#define pr_fmt(fmt) PMSG_NAME ": " fmt + +static char *pmsg_devnode(struct device *dev, umode_t *mode) +{ + if (mode) + *mode = 0220; + return NULL; +} + +void pstore_register_pmsg(void) +{ + struct device *pmsg_device; + + pmsg_major = register_chrdev(0, PMSG_NAME, &pmsg_fops); + if (pmsg_major < 0) { + pr_err("register_chrdev failed\n"); + goto err; + } + + pmsg_class = class_create(THIS_MODULE, PMSG_NAME); + if (IS_ERR(pmsg_class)) { + pr_err("device class file already in use\n"); + goto err_class; + } + pmsg_class->devnode = pmsg_devnode; + + pmsg_device = device_create(pmsg_class, NULL, MKDEV(pmsg_major, 0), + NULL, "%s%d", PMSG_NAME, 0); + if (IS_ERR(pmsg_device)) { + pr_err("failed to create device\n"); + goto err_device; + } + return; + +err_device: + class_destroy(pmsg_class); +err_class: + unregister_chrdev(pmsg_major, PMSG_NAME); +err: + return; +} diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c index d3d37142bd93..f721e47db8ad 100644 --- a/fs/pstore/ram.c +++ b/fs/pstore/ram.c @@ -51,6 +51,10 @@ static ulong ramoops_ftrace_size = MIN_MEM_SIZE; module_param_named(ftrace_size, ramoops_ftrace_size, ulong, 0400); MODULE_PARM_DESC(ftrace_size, "size of ftrace log"); +static ulong ramoops_pmsg_size = MIN_MEM_SIZE; +module_param_named(pmsg_size, ramoops_pmsg_size, ulong, 0400); +MODULE_PARM_DESC(pmsg_size, "size of user space message log"); + static ulong mem_address; module_param(mem_address, ulong, 0400); MODULE_PARM_DESC(mem_address, @@ -82,12 +86,14 @@ struct ramoops_context { struct persistent_ram_zone **przs; struct persistent_ram_zone *cprz; struct persistent_ram_zone *fprz; + struct persistent_ram_zone *mprz; phys_addr_t phys_addr; unsigned long size; unsigned int memtype; size_t record_size; size_t console_size; size_t ftrace_size; + size_t pmsg_size; int dump_oops; struct persistent_ram_ecc_info ecc_info; unsigned int max_dump_cnt; @@ -96,6 +102,7 @@ struct ramoops_context { unsigned int dump_read_cnt; unsigned int console_read_cnt; unsigned int ftrace_read_cnt; + unsigned int pmsg_read_cnt; struct pstore_info pstore; }; @@ -109,6 +116,7 @@ static int ramoops_pstore_open(struct pstore_info *psi) cxt->dump_read_cnt = 0; cxt->console_read_cnt = 0; cxt->ftrace_read_cnt = 0; + cxt->pmsg_read_cnt = 0; return 0; } @@ -141,6 +149,12 @@ ramoops_get_next_prz(struct persistent_ram_zone *przs[], uint *c, uint max, return prz; } +static bool prz_ok(struct persistent_ram_zone *prz) +{ + return !!prz && !!(persistent_ram_old_size(prz) + + persistent_ram_ecc_string(prz, NULL, 0)); +} + static ssize_t ramoops_pstore_read(u64 *id, enum pstore_type_id *type, int *count, struct timespec *time, char **buf, struct pstore_info *psi) @@ -153,13 +167,16 @@ static ssize_t ramoops_pstore_read(u64 *id, enum pstore_type_id *type, prz = ramoops_get_next_prz(cxt->przs, &cxt->dump_read_cnt, cxt->max_dump_cnt, id, type, PSTORE_TYPE_DMESG, 1); - if (!prz) + if (!prz_ok(prz)) prz = ramoops_get_next_prz(&cxt->cprz, &cxt->console_read_cnt, 1, id, type, PSTORE_TYPE_CONSOLE, 0); - if (!prz) + if (!prz_ok(prz)) prz = ramoops_get_next_prz(&cxt->fprz, &cxt->ftrace_read_cnt, 1, id, type, PSTORE_TYPE_FTRACE, 0); - if (!prz) + if (!prz_ok(prz)) + prz = ramoops_get_next_prz(&cxt->mprz, &cxt->pmsg_read_cnt, + 1, id, type, PSTORE_TYPE_PMSG, 0); + if (!prz_ok(prz)) return 0; /* TODO(kees): Bogus time for the moment. */ @@ -222,6 +239,11 @@ static int notrace ramoops_pstore_write_buf(enum pstore_type_id type, return -ENOMEM; persistent_ram_write(cxt->fprz, buf, size); return 0; + } else if (type == PSTORE_TYPE_PMSG) { + if (!cxt->mprz) + return -ENOMEM; + persistent_ram_write(cxt->mprz, buf, size); + return 0; } if (type != PSTORE_TYPE_DMESG) @@ -279,6 +301,9 @@ static int ramoops_pstore_erase(enum pstore_type_id type, u64 id, int count, case PSTORE_TYPE_FTRACE: prz = cxt->fprz; break; + case PSTORE_TYPE_PMSG: + prz = cxt->mprz; + break; default: return -EINVAL; } @@ -388,6 +413,12 @@ 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_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -404,7 +435,7 @@ static int ramoops_probe(struct platform_device *pdev) goto fail_out; if (!pdata->mem_size || (!pdata->record_size && !pdata->console_size && - !pdata->ftrace_size)) { + !pdata->ftrace_size && !pdata->pmsg_size)) { pr_err("The memory size and the record/console size must be " "non-zero\n"); goto fail_out; @@ -418,6 +449,8 @@ static int ramoops_probe(struct platform_device *pdev) pdata->console_size = rounddown_pow_of_two(pdata->console_size); if (!is_power_of_2(pdata->ftrace_size)) pdata->ftrace_size = rounddown_pow_of_two(pdata->ftrace_size); + if (pdata->pmsg_size && !is_power_of_2(pdata->pmsg_size)) + pdata->pmsg_size = rounddown_pow_of_two(pdata->pmsg_size); cxt->size = pdata->mem_size; cxt->phys_addr = pdata->mem_address; @@ -425,12 +458,14 @@ static int ramoops_probe(struct platform_device *pdev) cxt->record_size = pdata->record_size; cxt->console_size = pdata->console_size; cxt->ftrace_size = pdata->ftrace_size; + cxt->pmsg_size = pdata->pmsg_size; cxt->dump_oops = pdata->dump_oops; cxt->ecc_info = pdata->ecc_info; paddr = cxt->phys_addr; - dump_mem_sz = cxt->size - cxt->console_size - cxt->ftrace_size; + dump_mem_sz = cxt->size - cxt->console_size - cxt->ftrace_size + - cxt->pmsg_size; err = ramoops_init_przs(dev, cxt, &paddr, dump_mem_sz); if (err) goto fail_out; @@ -445,13 +480,9 @@ static int ramoops_probe(struct platform_device *pdev) if (err) goto fail_init_fprz; - if (!cxt->przs && !cxt->cprz && !cxt->fprz) { - pr_err("memory size too small, minimum is %zu\n", - cxt->console_size + cxt->record_size + - cxt->ftrace_size); - err = -EINVAL; - goto fail_cnt; - } + err = ramoops_init_prz(dev, cxt, &cxt->mprz, &paddr, cxt->pmsg_size, 0); + if (err) + goto fail_init_mprz; cxt->pstore.data = cxt; /* @@ -497,7 +528,8 @@ fail_buf: fail_clear: cxt->pstore.bufsize = 0; cxt->max_dump_cnt = 0; -fail_cnt: + kfree(cxt->mprz); +fail_init_mprz: kfree(cxt->fprz); fail_init_fprz: kfree(cxt->cprz); @@ -556,6 +588,7 @@ static void ramoops_register_dummy(void) dummy_data->record_size = record_size; dummy_data->console_size = ramoops_console_size; dummy_data->ftrace_size = ramoops_ftrace_size; + dummy_data->pmsg_size = ramoops_pmsg_size; dummy_data->dump_oops = dump_oops; /* * For backwards compatibility ramoops.ecc=1 means 16 bytes ECC diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 2e8caa62da78..3410e9f657ca 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -46,6 +46,7 @@ static int qnx4_remount(struct super_block *sb, int *flags, char *data) { struct qnx4_sb_info *qs; + sync_filesystem(sb); qs = qnx4_sb(sb); qs->Version = QNX4_VERSION; *flags |= MS_RDONLY; diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c index 8d941edfefa1..65cdaab3ed49 100644 --- a/fs/qnx6/inode.c +++ b/fs/qnx6/inode.c @@ -55,6 +55,7 @@ static int qnx6_show_options(struct seq_file *seq, struct dentry *root) static int qnx6_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_RDONLY; return 0; } diff --git a/fs/readdir.c b/fs/readdir.c index fee38e04fae4..d46eca8567a4 100644 --- a/fs/readdir.c +++ b/fs/readdir.c @@ -20,11 +20,11 @@ #include <asm/uaccess.h> -int vfs_readdir(struct file *file, filldir_t filler, void *buf) +int iterate_dir(struct file *file, struct dir_context *ctx) { struct inode *inode = file_inode(file); int res = -ENOTDIR; - if (!file->f_op || !file->f_op->readdir) + if (!file->f_op || (!file->f_op->readdir && !file->f_op->iterate)) goto out; res = security_file_permission(file, MAY_READ); @@ -37,15 +37,21 @@ int vfs_readdir(struct file *file, filldir_t filler, void *buf) res = -ENOENT; if (!IS_DEADDIR(inode)) { - res = file->f_op->readdir(file, buf, filler); + if (file->f_op->iterate) { + ctx->pos = file->f_pos; + res = file->f_op->iterate(file, ctx); + file->f_pos = ctx->pos; + } else { + res = file->f_op->readdir(file, ctx, ctx->actor); + ctx->pos = file->f_pos; + } file_accessed(file); } mutex_unlock(&inode->i_mutex); out: return res; } - -EXPORT_SYMBOL(vfs_readdir); +EXPORT_SYMBOL(iterate_dir); /* * Traditional linux readdir() handling.. @@ -66,6 +72,7 @@ struct old_linux_dirent { }; struct readdir_callback { + struct dir_context ctx; struct old_linux_dirent __user * dirent; int result; }; @@ -73,7 +80,7 @@ struct readdir_callback { static int fillonedir(void * __buf, const char * name, int namlen, loff_t offset, u64 ino, unsigned int d_type) { - struct readdir_callback * buf = (struct readdir_callback *) __buf; + struct readdir_callback *buf = (struct readdir_callback *) __buf; struct old_linux_dirent __user * dirent; unsigned long d_ino; @@ -107,15 +114,15 @@ SYSCALL_DEFINE3(old_readdir, unsigned int, fd, { int error; struct fd f = fdget(fd); - struct readdir_callback buf; + struct readdir_callback buf = { + .ctx.actor = fillonedir, + .dirent = dirent + }; if (!f.file) return -EBADF; - buf.result = 0; - buf.dirent = dirent; - - error = vfs_readdir(f.file, fillonedir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (buf.result) error = buf.result; @@ -137,6 +144,7 @@ struct linux_dirent { }; struct getdents_callback { + struct dir_context ctx; struct linux_dirent __user * current_dir; struct linux_dirent __user * previous; int count; @@ -191,7 +199,11 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd, { struct fd f; struct linux_dirent __user * lastdirent; - struct getdents_callback buf; + struct getdents_callback buf = { + .ctx.actor = filldir, + .count = count, + .current_dir = dirent + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -201,17 +213,12 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, filldir, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(f.file->f_pos, &lastdirent->d_off)) + if (put_user(buf.ctx.pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; @@ -221,6 +228,7 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd, } struct getdents_callback64 { + struct dir_context ctx; struct linux_dirent64 __user * current_dir; struct linux_dirent64 __user * previous; int count; @@ -271,7 +279,11 @@ SYSCALL_DEFINE3(getdents64, unsigned int, fd, { struct fd f; struct linux_dirent64 __user * lastdirent; - struct getdents_callback64 buf; + struct getdents_callback64 buf = { + .ctx.actor = filldir64, + .count = count, + .current_dir = dirent + }; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) @@ -281,17 +293,12 @@ SYSCALL_DEFINE3(getdents64, unsigned int, fd, if (!f.file) return -EBADF; - buf.current_dir = dirent; - buf.previous = NULL; - buf.count = count; - buf.error = 0; - - error = vfs_readdir(f.file, filldir64, &buf); + error = iterate_dir(f.file, &buf.ctx); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - typeof(lastdirent->d_off) d_off = f.file->f_pos; + typeof(lastdirent->d_off) d_off = buf.ctx.pos; if (__put_user(d_off, &lastdirent->d_off)) error = -EFAULT; else diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index e2e202a07b31..460f762a85c0 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -1317,6 +1317,7 @@ static int reiserfs_remount(struct super_block *s, int *mount_flags, char *arg) int i; #endif + sync_filesystem(s); reiserfs_write_lock(s); #ifdef CONFIG_QUOTA diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 15cbc41ee365..ae839482c341 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -435,6 +435,7 @@ static int romfs_statfs(struct dentry *dentry, struct kstatfs *buf) */ static int romfs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_RDONLY; return 0; } diff --git a/fs/select.c b/fs/select.c index 8c1c96c27062..6b14dc7df3a4 100644 --- a/fs/select.c +++ b/fs/select.c @@ -27,6 +27,7 @@ #include <linux/rcupdate.h> #include <linux/hrtimer.h> #include <linux/sched/rt.h> +#include <linux/freezer.h> #include <asm/uaccess.h> @@ -236,7 +237,8 @@ int poll_schedule_timeout(struct poll_wqueues *pwq, int state, set_current_state(state); if (!pwq->triggered) - rc = schedule_hrtimeout_range(expires, slack, HRTIMER_MODE_ABS); + rc = freezable_schedule_hrtimeout_range(expires, slack, + HRTIMER_MODE_ABS); __set_current_state(TASK_RUNNING); /* diff --git a/fs/seq_file.c b/fs/seq_file.c index 3dd44db1465e..d84ca5932c73 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -8,8 +8,10 @@ #include <linux/fs.h> #include <linux/export.h> #include <linux/seq_file.h> +#include <linux/vmalloc.h> #include <linux/slab.h> #include <linux/cred.h> +#include <linux/mm.h> #include <asm/uaccess.h> #include <asm/page.h> @@ -30,6 +32,16 @@ static void seq_set_overflow(struct seq_file *m) m->count = m->size; } +static void *seq_buf_alloc(unsigned long size) +{ + void *buf; + + buf = kmalloc(size, GFP_KERNEL | __GFP_NOWARN); + if (!buf && size > PAGE_SIZE) + buf = vmalloc(size); + return buf; +} + /** * seq_open - initialize sequential file * @file: file we initialize @@ -96,7 +108,7 @@ static int traverse(struct seq_file *m, loff_t offset) return 0; } if (!m->buf) { - m->buf = kmalloc(m->size = PAGE_SIZE, GFP_KERNEL); + m->buf = seq_buf_alloc(m->size = PAGE_SIZE); if (!m->buf) return -ENOMEM; } @@ -135,8 +147,8 @@ static int traverse(struct seq_file *m, loff_t offset) Eoverflow: m->op->stop(m, p); - kfree(m->buf); - m->buf = kmalloc(m->size <<= 1, GFP_KERNEL); + kvfree(m->buf); + m->buf = seq_buf_alloc(m->size <<= 1); return !m->buf ? -ENOMEM : -EAGAIN; } @@ -191,7 +203,7 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos) /* grab buffer if we didn't have one */ if (!m->buf) { - m->buf = kmalloc(m->size = PAGE_SIZE, GFP_KERNEL); + m->buf = seq_buf_alloc(m->size = PAGE_SIZE); if (!m->buf) goto Enomem; } @@ -231,8 +243,8 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos) if (m->count < m->size) goto Fill; m->op->stop(m, p); - kfree(m->buf); - m->buf = kmalloc(m->size <<= 1, GFP_KERNEL); + kvfree(m->buf); + m->buf = seq_buf_alloc(m->size <<= 1); if (!m->buf) goto Enomem; m->count = 0; @@ -349,7 +361,7 @@ EXPORT_SYMBOL(seq_lseek); int seq_release(struct inode *inode, struct file *file) { struct seq_file *m = file->private_data; - kfree(m->buf); + kvfree(m->buf); kfree(m); return 0; } @@ -604,13 +616,13 @@ EXPORT_SYMBOL(single_open); int single_open_size(struct file *file, int (*show)(struct seq_file *, void *), void *data, size_t size) { - char *buf = kmalloc(size, GFP_KERNEL); + char *buf = seq_buf_alloc(size); int ret; if (!buf) return -ENOMEM; ret = single_open(file, show, data); if (ret) { - kfree(buf); + kvfree(buf); return ret; } ((struct seq_file *)file->private_data)->buf = buf; diff --git a/fs/squashfs/Kconfig b/fs/squashfs/Kconfig index c70111ebefd4..ffb093e72b6c 100644 --- a/fs/squashfs/Kconfig +++ b/fs/squashfs/Kconfig @@ -25,6 +25,78 @@ 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 + Squashfs now supports three parallelisation options for + decompression. Each one exhibits various trade-offs between + decompression performance and CPU and memory usage. + + If in doubt, select "Single threaded compression" + +config SQUASHFS_DECOMP_SINGLE + bool "Single threaded compression" + help + Traditionally Squashfs has used single-threaded decompression. + Only one block (data or metadata) can be decompressed at any + one time. This limits CPU and memory usage to a minimum. + +config SQUASHFS_DECOMP_MULTI + bool "Use multiple decompressors for parallel I/O" + help + By default Squashfs uses a single decompressor but it gives + poor performance on parallel I/O workloads when using multiple CPU + machines due to waiting on decompressor availability. + + If you have a parallel I/O workload and your system has enough memory, + using this option may improve overall I/O performance. + + This decompressor implementation uses up to two parallel + decompressors per core. It dynamically allocates decompressors + on a demand basis. + +config SQUASHFS_DECOMP_MULTI_PERCPU + bool "Use percpu multiple decompressors for parallel I/O" + help + By default Squashfs uses a single decompressor but it gives + poor performance on parallel I/O workloads when using multiple CPU + machines due to waiting on decompressor availability. + + This decompressor implementation uses a maximum of one + decompressor per core. It uses percpu variables to ensure + decompression is load-balanced across the cores. + +endchoice + config SQUASHFS_XATTR bool "Squashfs XATTR support" depends on SQUASHFS @@ -48,6 +120,21 @@ config SQUASHFS_ZLIB If unsure, say Y. +config SQUASHFS_LZ4 + bool "Include support for LZ4 compressed file systems" + depends on SQUASHFS + select LZ4_DECOMPRESS + help + Saying Y here includes support for reading Squashfs file systems + compressed with LZ4 compression. LZ4 compression is mainly + aimed at embedded systems with slower CPUs where the overheads + of zlib are too high. + + LZ4 is not the standard compression used in Squashfs and so most + file systems will be readable without selecting this option. + + If unsure, say N. + config SQUASHFS_LZO bool "Include support for LZO compressed file systems" depends on SQUASHFS diff --git a/fs/squashfs/Makefile b/fs/squashfs/Makefile index 110b0476f3b4..246a6f329d89 100644 --- a/fs/squashfs/Makefile +++ b/fs/squashfs/Makefile @@ -5,7 +5,13 @@ 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-$(CONFIG_SQUASHFS_FILE_DIRECT) += file_direct.o page_actor.o +squashfs-$(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 squashfs-$(CONFIG_SQUASHFS_XATTR) += xattr.o xattr_id.o +squashfs-$(CONFIG_SQUASHFS_LZ4) += lz4_wrapper.o squashfs-$(CONFIG_SQUASHFS_LZO) += lzo_wrapper.o squashfs-$(CONFIG_SQUASHFS_XZ) += xz_wrapper.o squashfs-$(CONFIG_SQUASHFS_ZLIB) += zlib_wrapper.o diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c index fb50652e4e11..0cea9b9236d0 100644 --- a/fs/squashfs/block.c +++ b/fs/squashfs/block.c @@ -36,6 +36,7 @@ #include "squashfs_fs_sb.h" #include "squashfs.h" #include "decompressor.h" +#include "page_actor.h" /* * Read the metadata block length, this is stored in the first two @@ -86,16 +87,16 @@ static struct buffer_head *get_block_length(struct super_block *sb, * generated a larger block - this does occasionally happen with compression * algorithms). */ -int squashfs_read_data(struct super_block *sb, void **buffer, u64 index, - int length, u64 *next_index, int srclength, int pages) +int squashfs_read_data(struct super_block *sb, u64 index, int length, + u64 *next_index, struct squashfs_page_actor *output) { 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, page = 0, avail; + int bytes, compressed, b = 0, k = 0, avail, i; - bh = kcalloc(((srclength + msblk->devblksize - 1) + bh = kcalloc(((output->length + msblk->devblksize - 1) >> msblk->devblksize_log2) + 1, sizeof(*bh), GFP_KERNEL); if (bh == NULL) return -ENOMEM; @@ -111,9 +112,9 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index, *next_index = index + length; TRACE("Block @ 0x%llx, %scompressed size %d, src size %d\n", - index, compressed ? "" : "un", length, srclength); + index, compressed ? "" : "un", length, output->length); - if (length < 0 || length > srclength || + if (length < 0 || length > output->length || (index + length) > msblk->bytes_used) goto read_failure; @@ -145,7 +146,7 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index, TRACE("Block @ 0x%llx, %scompressed size %d\n", index, compressed ? "" : "un", length); - if (length < 0 || length > srclength || + if (length < 0 || length > output->length || (index + length) > msblk->bytes_used) goto block_release; @@ -158,35 +159,36 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index, ll_rw_block(READ, b - 1, bh + 1); } + for (i = 0; i < b; i++) { + wait_on_buffer(bh[i]); + if (!buffer_uptodate(bh[i])) + goto block_release; + } + if (compressed) { - length = squashfs_decompress(msblk, buffer, bh, b, offset, - length, srclength, pages); + length = squashfs_decompress(msblk, bh, b, offset, length, + output); if (length < 0) goto read_failure; } else { /* * Block is uncompressed. */ - int i, in, pg_offset = 0; - - for (i = 0; i < b; i++) { - wait_on_buffer(bh[i]); - if (!buffer_uptodate(bh[i])) - goto block_release; - } + 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) { - page++; + data = squashfs_next_page(output); pg_offset = 0; } avail = min_t(int, in, PAGE_CACHE_SIZE - pg_offset); - memcpy(buffer[page] + pg_offset, - bh[k]->b_data + offset, avail); + memcpy(data + pg_offset, bh[k]->b_data + offset, + avail); in -= avail; pg_offset += avail; offset += avail; @@ -194,6 +196,7 @@ int squashfs_read_data(struct super_block *sb, void **buffer, u64 index, offset = 0; put_bh(bh[k]); } + squashfs_finish_page(output); } kfree(bh); diff --git a/fs/squashfs/cache.c b/fs/squashfs/cache.c index af0b73802592..1cb70a0b2168 100644 --- a/fs/squashfs/cache.c +++ b/fs/squashfs/cache.c @@ -56,6 +56,7 @@ #include "squashfs_fs.h" #include "squashfs_fs_sb.h" #include "squashfs.h" +#include "page_actor.h" /* * Look-up block in cache, and increment usage count. If not in cache, read @@ -119,9 +120,8 @@ struct squashfs_cache_entry *squashfs_cache_get(struct super_block *sb, entry->error = 0; spin_unlock(&cache->lock); - entry->length = squashfs_read_data(sb, entry->data, - block, length, &entry->next_index, - cache->block_size, cache->pages); + entry->length = squashfs_read_data(sb, block, length, + &entry->next_index, entry->actor); spin_lock(&cache->lock); @@ -220,6 +220,7 @@ void squashfs_cache_delete(struct squashfs_cache *cache) kfree(cache->entry[i].data[j]); kfree(cache->entry[i].data); } + kfree(cache->entry[i].actor); } kfree(cache->entry); @@ -280,6 +281,13 @@ struct squashfs_cache *squashfs_cache_init(char *name, int entries, goto cleanup; } } + + entry->actor = squashfs_page_actor_init(entry->data, + cache->pages, 0); + if (entry->actor == NULL) { + ERROR("Failed to allocate %s cache entry\n", name); + goto cleanup; + } } return cache; @@ -410,6 +418,7 @@ 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 squashfs_page_actor *actor; table = buffer = kmalloc(length, GFP_KERNEL); if (table == NULL) @@ -421,19 +430,28 @@ void *squashfs_read_table(struct super_block *sb, u64 block, int length) goto failed; } + actor = squashfs_page_actor_init(data, pages, length); + if (actor == NULL) { + res = -ENOMEM; + goto failed2; + } + for (i = 0; i < pages; i++, buffer += PAGE_CACHE_SIZE) data[i] = buffer; - res = squashfs_read_data(sb, data, block, length | - SQUASHFS_COMPRESSED_BIT_BLOCK, NULL, length, pages); + res = squashfs_read_data(sb, block, length | + SQUASHFS_COMPRESSED_BIT_BLOCK, NULL, actor); kfree(data); + kfree(actor); if (res < 0) goto failed; return table; +failed2: + kfree(data); failed: kfree(table); return ERR_PTR(res); diff --git a/fs/squashfs/decompressor.c b/fs/squashfs/decompressor.c index 3f6271d86abc..e9034bf6e5ae 100644 --- a/fs/squashfs/decompressor.c +++ b/fs/squashfs/decompressor.c @@ -30,6 +30,7 @@ #include "squashfs_fs_sb.h" #include "decompressor.h" #include "squashfs.h" +#include "page_actor.h" /* * This file (and decompressor.h) implements a decompressor framework for @@ -37,33 +38,40 @@ */ static const struct squashfs_decompressor squashfs_lzma_unsupported_comp_ops = { - NULL, NULL, NULL, LZMA_COMPRESSION, "lzma", 0 + NULL, NULL, NULL, NULL, LZMA_COMPRESSION, "lzma", 0 }; +#ifndef CONFIG_SQUASHFS_LZ4 +static const struct squashfs_decompressor squashfs_lz4_comp_ops = { + NULL, NULL, NULL, NULL, LZ4_COMPRESSION, "lz4", 0 +}; +#endif + #ifndef CONFIG_SQUASHFS_LZO static const struct squashfs_decompressor squashfs_lzo_comp_ops = { - NULL, NULL, NULL, LZO_COMPRESSION, "lzo", 0 + NULL, NULL, NULL, NULL, LZO_COMPRESSION, "lzo", 0 }; #endif #ifndef CONFIG_SQUASHFS_XZ static const struct squashfs_decompressor squashfs_xz_comp_ops = { - NULL, NULL, NULL, XZ_COMPRESSION, "xz", 0 + NULL, NULL, NULL, NULL, XZ_COMPRESSION, "xz", 0 }; #endif #ifndef CONFIG_SQUASHFS_ZLIB static const struct squashfs_decompressor squashfs_zlib_comp_ops = { - NULL, NULL, NULL, ZLIB_COMPRESSION, "zlib", 0 + NULL, NULL, NULL, NULL, ZLIB_COMPRESSION, "zlib", 0 }; #endif static const struct squashfs_decompressor squashfs_unknown_comp_ops = { - NULL, NULL, NULL, 0, "unknown", 0 + NULL, NULL, NULL, NULL, 0, "unknown", 0 }; static const struct squashfs_decompressor *decompressor[] = { &squashfs_zlib_comp_ops, + &squashfs_lz4_comp_ops, &squashfs_lzo_comp_ops, &squashfs_xz_comp_ops, &squashfs_lzma_unsupported_comp_ops, @@ -83,10 +91,11 @@ const struct squashfs_decompressor *squashfs_lookup_decompressor(int id) } -void *squashfs_decompressor_init(struct super_block *sb, unsigned short flags) +static void *get_comp_opts(struct super_block *sb, unsigned short flags) { struct squashfs_sb_info *msblk = sb->s_fs_info; - void *strm, *buffer = NULL; + void *buffer = NULL, *comp_opts; + struct squashfs_page_actor *actor = NULL; int length = 0; /* @@ -94,23 +103,46 @@ void *squashfs_decompressor_init(struct super_block *sb, unsigned short flags) */ if (SQUASHFS_COMP_OPTS(flags)) { buffer = kmalloc(PAGE_CACHE_SIZE, GFP_KERNEL); - if (buffer == NULL) - return ERR_PTR(-ENOMEM); + if (buffer == NULL) { + comp_opts = ERR_PTR(-ENOMEM); + goto out; + } - length = squashfs_read_data(sb, &buffer, - sizeof(struct squashfs_super_block), 0, NULL, - PAGE_CACHE_SIZE, 1); + 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) { - strm = ERR_PTR(length); - goto finished; + comp_opts = ERR_PTR(length); + goto out; } } - strm = msblk->decompressor->init(msblk, buffer, length); + comp_opts = squashfs_comp_opts(msblk, buffer, length); -finished: +out: + kfree(actor); kfree(buffer); + return comp_opts; +} + + +void *squashfs_decompressor_setup(struct super_block *sb, unsigned short flags) +{ + struct squashfs_sb_info *msblk = sb->s_fs_info; + void *stream, *comp_opts = get_comp_opts(sb, flags); + + if (IS_ERR(comp_opts)) + return comp_opts; + + stream = squashfs_decompressor_create(msblk, comp_opts); + if (IS_ERR(stream)) + kfree(comp_opts); - return strm; + return stream; } diff --git a/fs/squashfs/decompressor.h b/fs/squashfs/decompressor.h index 330073e29029..a25713c031a5 100644 --- a/fs/squashfs/decompressor.h +++ b/fs/squashfs/decompressor.h @@ -24,34 +24,32 @@ */ struct squashfs_decompressor { - void *(*init)(struct squashfs_sb_info *, void *, int); + void *(*init)(struct squashfs_sb_info *, void *); + void *(*comp_opts)(struct squashfs_sb_info *, void *, int); void (*free)(void *); - int (*decompress)(struct squashfs_sb_info *, void **, - struct buffer_head **, int, int, int, int, int); + int (*decompress)(struct squashfs_sb_info *, void *, + struct buffer_head **, int, int, int, + struct squashfs_page_actor *); int id; char *name; int supported; }; -static inline void squashfs_decompressor_free(struct squashfs_sb_info *msblk, - void *s) +static inline void *squashfs_comp_opts(struct squashfs_sb_info *msblk, + void *buff, int length) { - if (msblk->decompressor) - msblk->decompressor->free(s); -} - -static inline int squashfs_decompress(struct squashfs_sb_info *msblk, - void **buffer, struct buffer_head **bh, int b, int offset, int length, - int srclength, int pages) -{ - return msblk->decompressor->decompress(msblk, buffer, bh, b, offset, - length, srclength, pages); + return msblk->decompressor->comp_opts ? + msblk->decompressor->comp_opts(msblk, buff, length) : NULL; } #ifdef CONFIG_SQUASHFS_XZ extern const struct squashfs_decompressor squashfs_xz_comp_ops; #endif +#ifdef CONFIG_SQUASHFS_LZ4 +extern const struct squashfs_decompressor squashfs_lz4_comp_ops; +#endif + #ifdef CONFIG_SQUASHFS_LZO extern const struct squashfs_decompressor squashfs_lzo_comp_ops; #endif diff --git a/fs/squashfs/decompressor_multi.c b/fs/squashfs/decompressor_multi.c new file mode 100644 index 000000000000..d6008a636479 --- /dev/null +++ b/fs/squashfs/decompressor_multi.c @@ -0,0 +1,198 @@ +/* + * Copyright (c) 2013 + * Minchan Kim <minchan@kernel.org> + * + * This work is licensed under the terms of the GNU GPL, version 2. See + * the COPYING file in the top-level directory. + */ +#include <linux/types.h> +#include <linux/mutex.h> +#include <linux/slab.h> +#include <linux/buffer_head.h> +#include <linux/sched.h> +#include <linux/wait.h> +#include <linux/cpumask.h> + +#include "squashfs_fs.h" +#include "squashfs_fs_sb.h" +#include "decompressor.h" +#include "squashfs.h" + +/* + * This file implements multi-threaded decompression in the + * decompressor framework + */ + + +/* + * The reason that multiply two is that a CPU can request new I/O + * while it is waiting previous request. + */ +#define MAX_DECOMPRESSOR (num_online_cpus() * 2) + + +int squashfs_max_decompressors(void) +{ + return MAX_DECOMPRESSOR; +} + + +struct squashfs_stream { + void *comp_opts; + struct list_head strm_list; + struct mutex mutex; + int avail_decomp; + wait_queue_head_t wait; +}; + + +struct decomp_stream { + void *stream; + struct list_head list; +}; + + +static void put_decomp_stream(struct decomp_stream *decomp_strm, + struct squashfs_stream *stream) +{ + mutex_lock(&stream->mutex); + list_add(&decomp_strm->list, &stream->strm_list); + mutex_unlock(&stream->mutex); + wake_up(&stream->wait); +} + +void *squashfs_decompressor_create(struct squashfs_sb_info *msblk, + void *comp_opts) +{ + struct squashfs_stream *stream; + struct decomp_stream *decomp_strm = NULL; + int err = -ENOMEM; + + stream = kzalloc(sizeof(*stream), GFP_KERNEL); + if (!stream) + goto out; + + stream->comp_opts = comp_opts; + mutex_init(&stream->mutex); + INIT_LIST_HEAD(&stream->strm_list); + init_waitqueue_head(&stream->wait); + + /* + * We should have a decompressor at least as default + * so if we fail to allocate new decompressor dynamically, + * we could always fall back to default decompressor and + * file system works. + */ + decomp_strm = kmalloc(sizeof(*decomp_strm), GFP_KERNEL); + if (!decomp_strm) + goto out; + + decomp_strm->stream = msblk->decompressor->init(msblk, + stream->comp_opts); + if (IS_ERR(decomp_strm->stream)) { + err = PTR_ERR(decomp_strm->stream); + goto out; + } + + list_add(&decomp_strm->list, &stream->strm_list); + stream->avail_decomp = 1; + return stream; + +out: + kfree(decomp_strm); + kfree(stream); + return ERR_PTR(err); +} + + +void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk) +{ + struct squashfs_stream *stream = msblk->stream; + if (stream) { + struct decomp_stream *decomp_strm; + + while (!list_empty(&stream->strm_list)) { + decomp_strm = list_entry(stream->strm_list.prev, + struct decomp_stream, list); + list_del(&decomp_strm->list); + msblk->decompressor->free(decomp_strm->stream); + kfree(decomp_strm); + stream->avail_decomp--; + } + WARN_ON(stream->avail_decomp); + kfree(stream->comp_opts); + kfree(stream); + } +} + + +static struct decomp_stream *get_decomp_stream(struct squashfs_sb_info *msblk, + struct squashfs_stream *stream) +{ + struct decomp_stream *decomp_strm; + + while (1) { + mutex_lock(&stream->mutex); + + /* There is available decomp_stream */ + if (!list_empty(&stream->strm_list)) { + decomp_strm = list_entry(stream->strm_list.prev, + struct decomp_stream, list); + list_del(&decomp_strm->list); + mutex_unlock(&stream->mutex); + break; + } + + /* + * If there is no available decomp and already full, + * let's wait for releasing decomp from other users. + */ + if (stream->avail_decomp >= MAX_DECOMPRESSOR) + goto wait; + + /* Let's allocate new decomp */ + decomp_strm = kmalloc(sizeof(*decomp_strm), GFP_KERNEL); + if (!decomp_strm) + goto wait; + + decomp_strm->stream = msblk->decompressor->init(msblk, + stream->comp_opts); + if (IS_ERR(decomp_strm->stream)) { + kfree(decomp_strm); + goto wait; + } + + stream->avail_decomp++; + WARN_ON(stream->avail_decomp > MAX_DECOMPRESSOR); + + mutex_unlock(&stream->mutex); + break; +wait: + /* + * If system memory is tough, let's for other's + * releasing instead of hurting VM because it could + * make page cache thrashing. + */ + mutex_unlock(&stream->mutex); + wait_event(stream->wait, + !list_empty(&stream->strm_list)); + } + + return decomp_strm; +} + + +int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh, + int b, int offset, int length, struct squashfs_page_actor *output) +{ + int res; + struct squashfs_stream *stream = msblk->stream; + struct decomp_stream *decomp_stream = get_decomp_stream(msblk, stream); + res = msblk->decompressor->decompress(msblk, decomp_stream->stream, + bh, b, offset, length, output); + put_decomp_stream(decomp_stream, stream); + if (res < 0) + ERROR("%s decompression failed, data probably corrupt\n", + msblk->decompressor->name); + return res; +} diff --git a/fs/squashfs/decompressor_multi_percpu.c b/fs/squashfs/decompressor_multi_percpu.c new file mode 100644 index 000000000000..23a9c28ad8ea --- /dev/null +++ b/fs/squashfs/decompressor_multi_percpu.c @@ -0,0 +1,97 @@ +/* + * 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/types.h> +#include <linux/slab.h> +#include <linux/percpu.h> +#include <linux/buffer_head.h> + +#include "squashfs_fs.h" +#include "squashfs_fs_sb.h" +#include "decompressor.h" +#include "squashfs.h" + +/* + * This file implements multi-threaded decompression using percpu + * variables, one thread per cpu core. + */ + +struct squashfs_stream { + void *stream; +}; + +void *squashfs_decompressor_create(struct squashfs_sb_info *msblk, + void *comp_opts) +{ + struct squashfs_stream *stream; + struct squashfs_stream __percpu *percpu; + int err, cpu; + + percpu = alloc_percpu(struct squashfs_stream); + if (percpu == NULL) + return ERR_PTR(-ENOMEM); + + for_each_possible_cpu(cpu) { + stream = per_cpu_ptr(percpu, cpu); + stream->stream = msblk->decompressor->init(msblk, comp_opts); + if (IS_ERR(stream->stream)) { + err = PTR_ERR(stream->stream); + goto out; + } + } + + kfree(comp_opts); + return (__force void *) percpu; + +out: + for_each_possible_cpu(cpu) { + stream = per_cpu_ptr(percpu, cpu); + if (!IS_ERR_OR_NULL(stream->stream)) + msblk->decompressor->free(stream->stream); + } + free_percpu(percpu); + return ERR_PTR(err); +} + +void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk) +{ + struct squashfs_stream __percpu *percpu = + (struct squashfs_stream __percpu *) msblk->stream; + struct squashfs_stream *stream; + int cpu; + + if (msblk->stream) { + for_each_possible_cpu(cpu) { + stream = per_cpu_ptr(percpu, cpu); + msblk->decompressor->free(stream->stream); + } + free_percpu(percpu); + } +} + +int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh, + int b, int offset, int length, struct squashfs_page_actor *output) +{ + struct squashfs_stream __percpu *percpu = + (struct squashfs_stream __percpu *) msblk->stream; + struct squashfs_stream *stream = get_cpu_ptr(percpu); + int res = msblk->decompressor->decompress(msblk, stream->stream, bh, b, + offset, length, output); + put_cpu_ptr(stream); + + if (res < 0) + ERROR("%s decompression failed, data probably corrupt\n", + msblk->decompressor->name); + + return res; +} + +int squashfs_max_decompressors(void) +{ + return num_possible_cpus(); +} diff --git a/fs/squashfs/decompressor_single.c b/fs/squashfs/decompressor_single.c new file mode 100644 index 000000000000..a6c75929a00e --- /dev/null +++ b/fs/squashfs/decompressor_single.c @@ -0,0 +1,85 @@ +/* + * 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/types.h> +#include <linux/mutex.h> +#include <linux/slab.h> +#include <linux/buffer_head.h> + +#include "squashfs_fs.h" +#include "squashfs_fs_sb.h" +#include "decompressor.h" +#include "squashfs.h" + +/* + * This file implements single-threaded decompression in the + * decompressor framework + */ + +struct squashfs_stream { + void *stream; + struct mutex mutex; +}; + +void *squashfs_decompressor_create(struct squashfs_sb_info *msblk, + void *comp_opts) +{ + struct squashfs_stream *stream; + int err = -ENOMEM; + + stream = kmalloc(sizeof(*stream), GFP_KERNEL); + if (stream == NULL) + goto out; + + stream->stream = msblk->decompressor->init(msblk, comp_opts); + if (IS_ERR(stream->stream)) { + err = PTR_ERR(stream->stream); + goto out; + } + + kfree(comp_opts); + mutex_init(&stream->mutex); + return stream; + +out: + kfree(stream); + return ERR_PTR(err); +} + +void squashfs_decompressor_destroy(struct squashfs_sb_info *msblk) +{ + struct squashfs_stream *stream = msblk->stream; + + if (stream) { + msblk->decompressor->free(stream->stream); + kfree(stream); + } +} + +int squashfs_decompress(struct squashfs_sb_info *msblk, struct buffer_head **bh, + int b, int offset, int length, struct squashfs_page_actor *output) +{ + int res; + struct squashfs_stream *stream = msblk->stream; + + mutex_lock(&stream->mutex); + res = msblk->decompressor->decompress(msblk, stream->stream, bh, b, + offset, length, output); + mutex_unlock(&stream->mutex); + + if (res < 0) + ERROR("%s decompression failed, data probably corrupt\n", + msblk->decompressor->name); + + return res; +} + +int squashfs_max_decompressors(void) +{ + return 1; +} diff --git a/fs/squashfs/dir.c b/fs/squashfs/dir.c index 57dc70ebbb19..d8c2d747be28 100644 --- a/fs/squashfs/dir.c +++ b/fs/squashfs/dir.c @@ -54,6 +54,7 @@ static int get_dir_index_using_offset(struct super_block *sb, { struct squashfs_sb_info *msblk = sb->s_fs_info; int err, i, index, length = 0; + unsigned int size; struct squashfs_dir_index dir_index; TRACE("Entered get_dir_index_using_offset, i_count %d, f_pos %lld\n", @@ -81,8 +82,14 @@ static int get_dir_index_using_offset(struct super_block *sb, */ break; + size = le32_to_cpu(dir_index.size) + 1; + + /* size should never be larger than SQUASHFS_NAME_LEN */ + if (size > SQUASHFS_NAME_LEN) + break; + err = squashfs_read_metadata(sb, NULL, &index_start, - &index_offset, le32_to_cpu(dir_index.size) + 1); + &index_offset, size); if (err < 0) break; @@ -100,14 +107,13 @@ static int get_dir_index_using_offset(struct super_block *sb, } -static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) +static int squashfs_readdir(struct file *file, struct dir_context *ctx) { struct inode *inode = file_inode(file); struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; u64 block = squashfs_i(inode)->start + msblk->directory_table; - int offset = squashfs_i(inode)->offset, length, dir_count, size, - type, err; - unsigned int inode_number; + int offset = squashfs_i(inode)->offset, length, err; + unsigned int inode_number, dir_count, size, type; struct squashfs_dir_header dirh; struct squashfs_dir_entry *dire; @@ -127,11 +133,11 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) * It also means that the external f_pos is offset by 3 from the * on-disk directory f_pos. */ - while (file->f_pos < 3) { + while (ctx->pos < 3) { char *name; int i_ino; - if (file->f_pos == 0) { + if (ctx->pos == 0) { name = "."; size = 1; i_ino = inode->i_ino; @@ -141,24 +147,18 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) i_ino = squashfs_i(inode)->parent; } - TRACE("Calling filldir(%p, %s, %d, %lld, %d, %d)\n", - dirent, name, size, file->f_pos, i_ino, - squashfs_filetype_table[1]); - - if (filldir(dirent, name, size, file->f_pos, i_ino, - squashfs_filetype_table[1]) < 0) { - TRACE("Filldir returned less than 0\n"); + if (!dir_emit(ctx, name, size, i_ino, + squashfs_filetype_table[1])) goto finish; - } - file->f_pos += size; + ctx->pos += size; } length = get_dir_index_using_offset(inode->i_sb, &block, &offset, squashfs_i(inode)->dir_idx_start, squashfs_i(inode)->dir_idx_offset, squashfs_i(inode)->dir_idx_cnt, - file->f_pos); + ctx->pos); while (length < i_size_read(inode)) { /* @@ -198,7 +198,7 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) length += sizeof(*dire) + size; - if (file->f_pos >= length) + if (ctx->pos >= length) continue; dire->name[size] = '\0'; @@ -206,22 +206,15 @@ static int squashfs_readdir(struct file *file, void *dirent, filldir_t filldir) ((short) le16_to_cpu(dire->inode_number)); type = le16_to_cpu(dire->type); - TRACE("Calling filldir(%p, %s, %d, %lld, %x:%x, %d, %d)" - "\n", dirent, dire->name, size, - file->f_pos, - le32_to_cpu(dirh.start_block), - le16_to_cpu(dire->offset), - inode_number, - squashfs_filetype_table[type]); + if (type > SQUASHFS_MAX_DIR_TYPE) + goto failed_read; - if (filldir(dirent, dire->name, size, file->f_pos, + if (!dir_emit(ctx, dire->name, size, inode_number, - squashfs_filetype_table[type]) < 0) { - TRACE("Filldir returned less than 0\n"); + squashfs_filetype_table[type])) goto finish; - } - file->f_pos = length; + ctx->pos = length; } } @@ -238,6 +231,6 @@ failed_read: const struct file_operations squashfs_dir_ops = { .read = generic_read_dir, - .readdir = squashfs_readdir, + .iterate = squashfs_readdir, .llseek = default_llseek, }; diff --git a/fs/squashfs/file.c b/fs/squashfs/file.c index 8ca62c28fe12..e5c9689062ba 100644 --- a/fs/squashfs/file.c +++ b/fs/squashfs/file.c @@ -370,77 +370,15 @@ static int read_blocklist(struct inode *inode, int index, u64 *block) return le32_to_cpu(size); } - -static int squashfs_readpage(struct file *file, struct page *page) +/* Copy data into page cache */ +void squashfs_copy_cache(struct page *page, struct squashfs_cache_entry *buffer, + int bytes, int offset) { struct inode *inode = page->mapping->host; struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; - int bytes, i, offset = 0, sparse = 0; - struct squashfs_cache_entry *buffer = NULL; void *pageaddr; - - int mask = (1 << (msblk->block_log - PAGE_CACHE_SHIFT)) - 1; - int index = page->index >> (msblk->block_log - PAGE_CACHE_SHIFT); - int start_index = page->index & ~mask; - int end_index = start_index | mask; - int file_end = i_size_read(inode) >> msblk->block_log; - - TRACE("Entered squashfs_readpage, page index %lx, start block %llx\n", - page->index, squashfs_i(inode)->start); - - if (page->index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> - PAGE_CACHE_SHIFT)) - goto out; - - if (index < file_end || squashfs_i(inode)->fragment_block == - SQUASHFS_INVALID_BLK) { - /* - * Reading a datablock from disk. Need to read block list - * to get location and block size. - */ - u64 block = 0; - int bsize = read_blocklist(inode, index, &block); - if (bsize < 0) - goto error_out; - - if (bsize == 0) { /* hole */ - bytes = index == file_end ? - (i_size_read(inode) & (msblk->block_size - 1)) : - msblk->block_size; - sparse = 1; - } else { - /* - * Read and decompress datablock. - */ - buffer = squashfs_get_datablock(inode->i_sb, - block, bsize); - if (buffer->error) { - ERROR("Unable to read page, block %llx, size %x" - "\n", block, bsize); - squashfs_cache_put(buffer); - goto error_out; - } - bytes = buffer->length; - } - } else { - /* - * Datablock is stored inside a fragment (tail-end packed - * block). - */ - buffer = squashfs_get_fragment(inode->i_sb, - squashfs_i(inode)->fragment_block, - squashfs_i(inode)->fragment_size); - - if (buffer->error) { - ERROR("Unable to read page, block %llx, size %x\n", - squashfs_i(inode)->fragment_block, - squashfs_i(inode)->fragment_size); - squashfs_cache_put(buffer); - goto error_out; - } - bytes = i_size_read(inode) & (msblk->block_size - 1); - offset = squashfs_i(inode)->fragment_offset; - } + int i, mask = (1 << (msblk->block_log - PAGE_CACHE_SHIFT)) - 1; + int start_index = page->index & ~mask, end_index = start_index | mask; /* * Loop copying datablock into pages. As the datablock likely covers @@ -451,7 +389,7 @@ static int squashfs_readpage(struct file *file, struct page *page) for (i = start_index; i <= end_index && bytes > 0; i++, bytes -= PAGE_CACHE_SIZE, offset += PAGE_CACHE_SIZE) { struct page *push_page; - int avail = sparse ? 0 : min_t(int, bytes, PAGE_CACHE_SIZE); + int avail = buffer ? min_t(int, bytes, PAGE_CACHE_SIZE) : 0; TRACE("bytes %d, i %d, available_bytes %d\n", bytes, i, avail); @@ -475,11 +413,75 @@ skip_page: if (i != page->index) page_cache_release(push_page); } +} + +/* Read datablock stored packed inside a fragment (tail-end packed block) */ +static int squashfs_readpage_fragment(struct page *page) +{ + struct inode *inode = page->mapping->host; + struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; + struct squashfs_cache_entry *buffer = squashfs_get_fragment(inode->i_sb, + squashfs_i(inode)->fragment_block, + squashfs_i(inode)->fragment_size); + int res = buffer->error; + + if (res) + ERROR("Unable to read page, block %llx, size %x\n", + squashfs_i(inode)->fragment_block, + squashfs_i(inode)->fragment_size); + else + squashfs_copy_cache(page, buffer, i_size_read(inode) & + (msblk->block_size - 1), + squashfs_i(inode)->fragment_offset); + + squashfs_cache_put(buffer); + return res; +} - if (!sparse) - squashfs_cache_put(buffer); +static int squashfs_readpage_sparse(struct page *page, int index, int file_end) +{ + struct inode *inode = page->mapping->host; + struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; + int bytes = index == file_end ? + (i_size_read(inode) & (msblk->block_size - 1)) : + msblk->block_size; + squashfs_copy_cache(page, NULL, bytes, 0); return 0; +} + +static int squashfs_readpage(struct file *file, struct page *page) +{ + struct inode *inode = page->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); + + if (page->index >= ((i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> + PAGE_CACHE_SHIFT)) + goto out; + + 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; + + if (bsize == 0) + res = squashfs_readpage_sparse(page, index, file_end); + else + res = squashfs_readpage_block(page, block, bsize); + } else + res = squashfs_readpage_fragment(page); + + if (!res) + return 0; error_out: SetPageError(page); diff --git a/fs/squashfs/file_cache.c b/fs/squashfs/file_cache.c new file mode 100644 index 000000000000..f2310d2a2019 --- /dev/null +++ b/fs/squashfs/file_cache.c @@ -0,0 +1,38 @@ +/* + * 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 new file mode 100644 index 000000000000..43e7a7eddac0 --- /dev/null +++ b/fs/squashfs/file_direct.c @@ -0,0 +1,176 @@ +/* + * 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" +#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) + +{ + struct inode *inode = target_page->mapping->host; + struct squashfs_sb_info *msblk = inode->i_sb->s_fs_info; + + 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; + 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); + + if (page[i] == NULL) { + missing_pages++; + continue; + } + + if (PageUptodate(page[i])) { + unlock_page(page[i]); + page_cache_release(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; + } + + /* 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; +} + + +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; + + 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]); + } + +out: + squashfs_cache_put(buffer); + return res; +} diff --git a/fs/squashfs/lz4_wrapper.c b/fs/squashfs/lz4_wrapper.c new file mode 100644 index 000000000000..c31e2bc9c081 --- /dev/null +++ b/fs/squashfs/lz4_wrapper.c @@ -0,0 +1,142 @@ +/* + * Copyright (c) 2013, 2014 + * 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/buffer_head.h> +#include <linux/mutex.h> +#include <linux/slab.h> +#include <linux/vmalloc.h> +#include <linux/lz4.h> + +#include "squashfs_fs.h" +#include "squashfs_fs_sb.h" +#include "squashfs.h" +#include "decompressor.h" +#include "page_actor.h" + +#define LZ4_LEGACY 1 + +struct lz4_comp_opts { + __le32 version; + __le32 flags; +}; + +struct squashfs_lz4 { + void *input; + void *output; +}; + + +static void *lz4_comp_opts(struct squashfs_sb_info *msblk, + void *buff, int len) +{ + struct lz4_comp_opts *comp_opts = buff; + + /* LZ4 compressed filesystems always have compression options */ + if (comp_opts == NULL || len < sizeof(*comp_opts)) + return ERR_PTR(-EIO); + + if (le32_to_cpu(comp_opts->version) != LZ4_LEGACY) { + /* LZ4 format currently used by the kernel is the 'legacy' + * format */ + ERROR("Unknown LZ4 version\n"); + return ERR_PTR(-EINVAL); + } + + return NULL; +} + + +static void *lz4_init(struct squashfs_sb_info *msblk, void *buff) +{ + int block_size = max_t(int, msblk->block_size, SQUASHFS_METADATA_SIZE); + struct squashfs_lz4 *stream; + + stream = kzalloc(sizeof(*stream), GFP_KERNEL); + if (stream == NULL) + goto failed; + stream->input = vmalloc(block_size); + if (stream->input == NULL) + goto failed2; + stream->output = vmalloc(block_size); + if (stream->output == NULL) + goto failed3; + + return stream; + +failed3: + vfree(stream->input); +failed2: + kfree(stream); +failed: + ERROR("Failed to initialise LZ4 decompressor\n"); + return ERR_PTR(-ENOMEM); +} + + +static void lz4_free(void *strm) +{ + struct squashfs_lz4 *stream = strm; + + if (stream) { + vfree(stream->input); + vfree(stream->output); + } + kfree(stream); +} + + +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; + size_t dest_len = output->length; + + 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]); + } + + 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); + + return dest_len; +} + +const struct squashfs_decompressor squashfs_lz4_comp_ops = { + .init = lz4_init, + .comp_opts = lz4_comp_opts, + .free = lz4_free, + .decompress = lz4_uncompress, + .id = LZ4_COMPRESSION, + .name = "lz4", + .supported = 1 +}; diff --git a/fs/squashfs/lzo_wrapper.c b/fs/squashfs/lzo_wrapper.c index 00f4dfc5f088..244b9fbfff7b 100644 --- a/fs/squashfs/lzo_wrapper.c +++ b/fs/squashfs/lzo_wrapper.c @@ -31,13 +31,14 @@ #include "squashfs_fs_sb.h" #include "squashfs.h" #include "decompressor.h" +#include "page_actor.h" struct squashfs_lzo { void *input; void *output; }; -static void *lzo_init(struct squashfs_sb_info *msblk, void *buff, int len) +static void *lzo_init(struct squashfs_sb_info *msblk, void *buff) { int block_size = max_t(int, msblk->block_size, SQUASHFS_METADATA_SIZE); @@ -74,22 +75,16 @@ static void lzo_free(void *strm) } -static int lzo_uncompress(struct squashfs_sb_info *msblk, void **buffer, - struct buffer_head **bh, int b, int offset, int length, int srclength, - int pages) +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 = msblk->stream; - void *buff = stream->input; + struct squashfs_lzo *stream = strm; + void *buff = stream->input, *data; int avail, i, bytes = length, res; - size_t out_len = srclength; - - mutex_lock(&msblk->read_data_mutex); + size_t out_len = output->length; for (i = 0; i < b; i++) { - wait_on_buffer(bh[i]); - if (!buffer_uptodate(bh[i])) - goto block_release; - avail = min(bytes, msblk->devblksize - offset); memcpy(buff, bh[i]->b_data + offset, avail); buff += avail; @@ -104,24 +99,24 @@ static int lzo_uncompress(struct squashfs_sb_info *msblk, void **buffer, goto failed; res = bytes = (int)out_len; - for (i = 0, buff = stream->output; bytes && i < pages; i++) { - avail = min_t(int, bytes, PAGE_CACHE_SIZE); - memcpy(buffer[i], buff, avail); - buff += avail; - bytes -= avail; + 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); - mutex_unlock(&msblk->read_data_mutex); return res; -block_release: - for (; i < b; i++) - put_bh(bh[i]); - failed: - mutex_unlock(&msblk->read_data_mutex); - - ERROR("lzo decompression failed, data probably corrupt\n"); return -EIO; } diff --git a/fs/squashfs/namei.c b/fs/squashfs/namei.c index 7834a517f7f4..67cad77fefb4 100644 --- a/fs/squashfs/namei.c +++ b/fs/squashfs/namei.c @@ -79,7 +79,8 @@ static int get_dir_index_using_name(struct super_block *sb, int len) { struct squashfs_sb_info *msblk = sb->s_fs_info; - int i, size, length = 0, err; + int i, length = 0, err; + unsigned int size; struct squashfs_dir_index *index; char *str; @@ -103,6 +104,8 @@ static int get_dir_index_using_name(struct super_block *sb, size = le32_to_cpu(index->size) + 1; + if (size > SQUASHFS_NAME_LEN) + break; err = squashfs_read_metadata(sb, index->name, &index_start, &index_offset, size); @@ -144,7 +147,8 @@ static struct dentry *squashfs_lookup(struct inode *dir, struct dentry *dentry, struct squashfs_dir_entry *dire; u64 block = squashfs_i(dir)->start + msblk->directory_table; int offset = squashfs_i(dir)->offset; - int err, length, dir_count, size; + int err, length; + unsigned int dir_count, size; TRACE("Entered squashfs_lookup [%llx:%x]\n", block, offset); diff --git a/fs/squashfs/page_actor.c b/fs/squashfs/page_actor.c new file mode 100644 index 000000000000..5a1c11f56441 --- /dev/null +++ b/fs/squashfs/page_actor.c @@ -0,0 +1,100 @@ +/* + * 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/kernel.h> +#include <linux/slab.h> +#include <linux/pagemap.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 *actor = kmalloc(sizeof(*actor), GFP_KERNEL); + + if (actor == NULL) + return NULL; + + actor->length = length ? : pages * PAGE_CACHE_SIZE; + actor->buffer = buffer; + 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; + return actor; +} + +/* Implementation of page_actor for decompressing directly into page cache. */ +static void *direct_first_page(struct squashfs_page_actor *actor) +{ + actor->next_page = 1; + return actor->pageaddr = kmap_atomic(actor->page[0]); +} + +static void *direct_next_page(struct squashfs_page_actor *actor) +{ + if (actor->pageaddr) + kunmap_atomic(actor->pageaddr); + + return actor->pageaddr = actor->next_page == actor->pages ? NULL : + kmap_atomic(actor->page[actor->next_page++]); +} + +static void direct_finish_page(struct squashfs_page_actor *actor) +{ + if (actor->pageaddr) + kunmap_atomic(actor->pageaddr); +} + +struct squashfs_page_actor *squashfs_page_actor_init_special(struct page **page, + int pages, int length) +{ + struct squashfs_page_actor *actor = kmalloc(sizeof(*actor), GFP_KERNEL); + + if (actor == NULL) + return NULL; + + 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; +} diff --git a/fs/squashfs/page_actor.h b/fs/squashfs/page_actor.h new file mode 100644 index 000000000000..26dd82008b82 --- /dev/null +++ b/fs/squashfs/page_actor.h @@ -0,0 +1,81 @@ +#ifndef PAGE_ACTOR_H +#define PAGE_ACTOR_H +/* + * 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. + */ + +#ifndef CONFIG_SQUASHFS_FILE_DIRECT +struct squashfs_page_actor { + void **page; + int pages; + int length; + int next_page; +}; + +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; + + actor->length = length ? : pages * PAGE_CACHE_SIZE; + actor->page = page; + actor->pages = pages; + actor->next_page = 0; + return actor; +} + +static inline void *squashfs_first_page(struct squashfs_page_actor *actor) +{ + actor->next_page = 1; + return actor->page[0]; +} + +static inline void *squashfs_next_page(struct squashfs_page_actor *actor) +{ + return actor->next_page == actor->pages ? NULL : + actor->page[actor->next_page++]; +} + +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; +}; + +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); +} +static inline void squashfs_finish_page(struct squashfs_page_actor *actor) +{ + actor->squashfs_finish_page(actor); +} +#endif +#endif diff --git a/fs/squashfs/squashfs.h b/fs/squashfs/squashfs.h index d1266516ed08..887d6d270080 100644 --- a/fs/squashfs/squashfs.h +++ b/fs/squashfs/squashfs.h @@ -25,11 +25,11 @@ #define ERROR(s, args...) pr_err("SQUASHFS error: "s, ## args) -#define WARNING(s, args...) pr_warning("SQUASHFS: "s, ## args) +#define WARNING(s, args...) pr_warn("SQUASHFS: "s, ## args) /* block.c */ -extern int squashfs_read_data(struct super_block *, void **, u64, int, u64 *, - int, int); +extern int squashfs_read_data(struct super_block *, u64, int, u64 *, + struct squashfs_page_actor *); /* cache.c */ extern struct squashfs_cache *squashfs_cache_init(char *, int, int); @@ -48,7 +48,14 @@ extern void *squashfs_read_table(struct super_block *, u64, int); /* decompressor.c */ extern const struct squashfs_decompressor *squashfs_lookup_decompressor(int); -extern void *squashfs_decompressor_init(struct super_block *, unsigned short); +extern void *squashfs_decompressor_setup(struct super_block *, unsigned short); + +/* decompressor_xxx.c */ +extern void *squashfs_decompressor_create(struct squashfs_sb_info *, void *); +extern void squashfs_decompressor_destroy(struct squashfs_sb_info *); +extern int squashfs_decompress(struct squashfs_sb_info *, struct buffer_head **, + int, int, int, struct squashfs_page_actor *); +extern int squashfs_max_decompressors(void); /* export.c */ extern __le64 *squashfs_read_inode_lookup_table(struct super_block *, u64, u64, @@ -59,6 +66,13 @@ extern int squashfs_frag_lookup(struct super_block *, unsigned int, u64 *); extern __le64 *squashfs_read_fragment_index_table(struct super_block *, u64, u64, unsigned int); +/* file.c */ +void squashfs_copy_cache(struct page *, struct squashfs_cache_entry *, int, + int); + +/* file_xxx.c */ +extern int squashfs_readpage_block(struct page *, u64, int); + /* id.c */ extern int squashfs_get_id(struct super_block *, unsigned int, unsigned int *); extern __le64 *squashfs_read_id_index_table(struct super_block *, u64, u64, diff --git a/fs/squashfs/squashfs_fs.h b/fs/squashfs/squashfs_fs.h index 9e2349d07cb1..506f4ba5b983 100644 --- a/fs/squashfs/squashfs_fs.h +++ b/fs/squashfs/squashfs_fs.h @@ -87,7 +87,7 @@ #define SQUASHFS_COMP_OPTS(flags) SQUASHFS_BIT(flags, \ SQUASHFS_COMP_OPT) -/* Max number of types and file types */ +/* Inode types including extended types */ #define SQUASHFS_DIR_TYPE 1 #define SQUASHFS_REG_TYPE 2 #define SQUASHFS_SYMLINK_TYPE 3 @@ -103,6 +103,9 @@ #define SQUASHFS_LFIFO_TYPE 13 #define SQUASHFS_LSOCKET_TYPE 14 +/* Max type value stored in directory entry */ +#define SQUASHFS_MAX_DIR_TYPE 7 + /* Xattr types */ #define SQUASHFS_XATTR_USER 0 #define SQUASHFS_XATTR_TRUSTED 1 @@ -237,6 +240,7 @@ struct meta_index { #define LZMA_COMPRESSION 2 #define LZO_COMPRESSION 3 #define XZ_COMPRESSION 4 +#define LZ4_COMPRESSION 5 struct squashfs_super_block { __le32 s_magic; diff --git a/fs/squashfs/squashfs_fs_sb.h b/fs/squashfs/squashfs_fs_sb.h index 52934a22f296..1da565cb50c3 100644 --- a/fs/squashfs/squashfs_fs_sb.h +++ b/fs/squashfs/squashfs_fs_sb.h @@ -50,6 +50,7 @@ struct squashfs_cache_entry { wait_queue_head_t wait_queue; struct squashfs_cache *cache; void **data; + struct squashfs_page_actor *actor; }; struct squashfs_sb_info { @@ -63,10 +64,9 @@ struct squashfs_sb_info { __le64 *id_table; __le64 *fragment_index; __le64 *xattr_id_table; - struct mutex read_data_mutex; struct mutex meta_index_mutex; struct meta_index *meta_index; - void *stream; + struct squashfs_stream *stream; __le64 *inode_lookup_table; u64 inode_table; u64 directory_table; diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 60553a9053ca..5056babe00df 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -27,6 +27,8 @@ * the filesystem. */ +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + #include <linux/fs.h> #include <linux/vfs.h> #include <linux/slab.h> @@ -98,7 +100,6 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) msblk->devblksize = sb_min_blocksize(sb, SQUASHFS_DEVBLK_SIZE); msblk->devblksize_log2 = ffz(~msblk->devblksize); - mutex_init(&msblk->read_data_mutex); mutex_init(&msblk->meta_index_mutex); /* @@ -206,13 +207,14 @@ static int squashfs_fill_super(struct super_block *sb, void *data, int silent) goto failed_mount; /* Allocate read_page block */ - msblk->read_page = squashfs_cache_init("data", 1, msblk->block_size); + msblk->read_page = squashfs_cache_init("data", + squashfs_max_decompressors(), msblk->block_size); if (msblk->read_page == NULL) { ERROR("Failed to allocate read_page block\n"); goto failed_mount; } - msblk->stream = squashfs_decompressor_init(sb, flags); + msblk->stream = squashfs_decompressor_setup(sb, flags); if (IS_ERR(msblk->stream)) { err = PTR_ERR(msblk->stream); msblk->stream = NULL; @@ -336,7 +338,7 @@ failed_mount: squashfs_cache_delete(msblk->block_cache); squashfs_cache_delete(msblk->fragment_cache); squashfs_cache_delete(msblk->read_page); - squashfs_decompressor_free(msblk, msblk->stream); + squashfs_decompressor_destroy(msblk); kfree(msblk->inode_lookup_table); kfree(msblk->fragment_index); kfree(msblk->id_table); @@ -371,6 +373,7 @@ static int squashfs_statfs(struct dentry *dentry, struct kstatfs *buf) static int squashfs_remount(struct super_block *sb, int *flags, char *data) { + sync_filesystem(sb); *flags |= MS_RDONLY; return 0; } @@ -383,7 +386,7 @@ static void squashfs_put_super(struct super_block *sb) squashfs_cache_delete(sbi->block_cache); squashfs_cache_delete(sbi->fragment_cache); squashfs_cache_delete(sbi->read_page); - squashfs_decompressor_free(sbi, sbi->stream); + squashfs_decompressor_destroy(sbi); kfree(sbi->id_table); kfree(sbi->fragment_index); kfree(sbi->meta_index); @@ -447,8 +450,7 @@ static int __init init_squashfs_fs(void) return err; } - printk(KERN_INFO "squashfs: version 4.0 (2009/01/31) " - "Phillip Lougher\n"); + pr_info("version 4.0 (2009/01/31) Phillip Lougher\n"); return 0; } diff --git a/fs/squashfs/xz_wrapper.c b/fs/squashfs/xz_wrapper.c index 1760b7d108f6..c609624e4b8a 100644 --- a/fs/squashfs/xz_wrapper.c +++ b/fs/squashfs/xz_wrapper.c @@ -32,44 +32,70 @@ #include "squashfs_fs_sb.h" #include "squashfs.h" #include "decompressor.h" +#include "page_actor.h" struct squashfs_xz { struct xz_dec *state; struct xz_buf buf; }; -struct comp_opts { +struct disk_comp_opts { __le32 dictionary_size; __le32 flags; }; -static void *squashfs_xz_init(struct squashfs_sb_info *msblk, void *buff, - int len) +struct comp_opts { + int dict_size; +}; + +static void *squashfs_xz_comp_opts(struct squashfs_sb_info *msblk, + void *buff, int len) { - struct comp_opts *comp_opts = buff; - struct squashfs_xz *stream; - int dict_size = msblk->block_size; - int err, n; + struct disk_comp_opts *comp_opts = buff; + struct comp_opts *opts; + int err = 0, n; + + opts = kmalloc(sizeof(*opts), GFP_KERNEL); + if (opts == NULL) { + err = -ENOMEM; + goto out2; + } if (comp_opts) { /* check compressor options are the expected length */ if (len < sizeof(*comp_opts)) { err = -EIO; - goto failed; + goto out; } - dict_size = le32_to_cpu(comp_opts->dictionary_size); + opts->dict_size = le32_to_cpu(comp_opts->dictionary_size); /* the dictionary size should be 2^n or 2^n+2^(n+1) */ - n = ffs(dict_size) - 1; - if (dict_size != (1 << n) && dict_size != (1 << n) + + n = ffs(opts->dict_size) - 1; + if (opts->dict_size != (1 << n) && opts->dict_size != (1 << n) + (1 << (n + 1))) { err = -EIO; - goto failed; + goto out; } - } + } else + /* use defaults */ + opts->dict_size = max_t(int, msblk->block_size, + SQUASHFS_METADATA_SIZE); + + return opts; + +out: + kfree(opts); +out2: + return ERR_PTR(err); +} + - dict_size = max_t(int, dict_size, SQUASHFS_METADATA_SIZE); +static void *squashfs_xz_init(struct squashfs_sb_info *msblk, void *buff) +{ + struct comp_opts *comp_opts = buff; + struct squashfs_xz *stream; + int err; stream = kmalloc(sizeof(*stream), GFP_KERNEL); if (stream == NULL) { @@ -77,7 +103,7 @@ static void *squashfs_xz_init(struct squashfs_sb_info *msblk, void *buff, goto failed; } - stream->state = xz_dec_init(XZ_PREALLOC, dict_size); + stream->state = xz_dec_init(XZ_PREALLOC, comp_opts->dict_size); if (stream->state == NULL) { kfree(stream); err = -ENOMEM; @@ -103,42 +129,37 @@ static void squashfs_xz_free(void *strm) } -static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void **buffer, - struct buffer_head **bh, int b, int offset, int length, int srclength, - int pages) +static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void *strm, + struct buffer_head **bh, int b, int offset, int length, + struct squashfs_page_actor *output) { enum xz_ret xz_err; - int avail, total = 0, k = 0, page = 0; - struct squashfs_xz *stream = msblk->stream; - - mutex_lock(&msblk->read_data_mutex); + int avail, total = 0, k = 0; + struct squashfs_xz *stream = strm; xz_dec_reset(stream->state); stream->buf.in_pos = 0; stream->buf.in_size = 0; stream->buf.out_pos = 0; stream->buf.out_size = PAGE_CACHE_SIZE; - stream->buf.out = buffer[page++]; + stream->buf.out = squashfs_first_page(output); do { if (stream->buf.in_pos == stream->buf.in_size && k < b) { avail = min(length, msblk->devblksize - offset); length -= avail; - wait_on_buffer(bh[k]); - if (!buffer_uptodate(bh[k])) - goto release_mutex; - stream->buf.in = bh[k]->b_data + offset; stream->buf.in_size = avail; stream->buf.in_pos = 0; offset = 0; } - if (stream->buf.out_pos == stream->buf.out_size - && page < pages) { - stream->buf.out = buffer[page++]; - stream->buf.out_pos = 0; - total += PAGE_CACHE_SIZE; + if (stream->buf.out_pos == stream->buf.out_size) { + stream->buf.out = squashfs_next_page(output); + if (stream->buf.out != NULL) { + stream->buf.out_pos = 0; + total += PAGE_CACHE_SIZE; + } } xz_err = xz_dec_run(stream->state, &stream->buf); @@ -147,23 +168,14 @@ static int squashfs_xz_uncompress(struct squashfs_sb_info *msblk, void **buffer, put_bh(bh[k++]); } while (xz_err == XZ_OK); - if (xz_err != XZ_STREAM_END) { - ERROR("xz_dec_run error, data probably corrupt\n"); - goto release_mutex; - } - - if (k < b) { - ERROR("xz_uncompress error, input remaining\n"); - goto release_mutex; - } + squashfs_finish_page(output); - total += stream->buf.out_pos; - mutex_unlock(&msblk->read_data_mutex); - return total; + if (xz_err != XZ_STREAM_END || k < b) + goto out; -release_mutex: - mutex_unlock(&msblk->read_data_mutex); + return total + stream->buf.out_pos; +out: for (; k < b; k++) put_bh(bh[k]); @@ -172,6 +184,7 @@ release_mutex: const struct squashfs_decompressor squashfs_xz_comp_ops = { .init = squashfs_xz_init, + .comp_opts = squashfs_xz_comp_opts, .free = squashfs_xz_free, .decompress = squashfs_xz_uncompress, .id = XZ_COMPRESSION, diff --git a/fs/squashfs/zlib_wrapper.c b/fs/squashfs/zlib_wrapper.c index 55d918fd2d86..8727caba6882 100644 --- a/fs/squashfs/zlib_wrapper.c +++ b/fs/squashfs/zlib_wrapper.c @@ -32,8 +32,9 @@ #include "squashfs_fs_sb.h" #include "squashfs.h" #include "decompressor.h" +#include "page_actor.h" -static void *zlib_init(struct squashfs_sb_info *dummy, void *buff, int len) +static void *zlib_init(struct squashfs_sb_info *dummy, void *buff) { z_stream *stream = kmalloc(sizeof(z_stream), GFP_KERNEL); if (stream == NULL) @@ -61,44 +62,37 @@ static void zlib_free(void *strm) } -static int zlib_uncompress(struct squashfs_sb_info *msblk, void **buffer, - struct buffer_head **bh, int b, int offset, int length, int srclength, - int pages) +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) { - int zlib_err, zlib_init = 0; - int k = 0, page = 0; - z_stream *stream = msblk->stream; - - mutex_lock(&msblk->read_data_mutex); + int zlib_err, zlib_init = 0, k = 0; + z_stream *stream = strm; - stream->avail_out = 0; + stream->avail_out = PAGE_CACHE_SIZE; + stream->next_out = squashfs_first_page(output); stream->avail_in = 0; do { if (stream->avail_in == 0 && k < b) { int avail = min(length, msblk->devblksize - offset); length -= avail; - wait_on_buffer(bh[k]); - if (!buffer_uptodate(bh[k])) - goto release_mutex; - stream->next_in = bh[k]->b_data + offset; stream->avail_in = avail; offset = 0; } - if (stream->avail_out == 0 && page < pages) { - stream->next_out = buffer[page++]; - stream->avail_out = PAGE_CACHE_SIZE; + if (stream->avail_out == 0) { + stream->next_out = squashfs_next_page(output); + if (stream->next_out != NULL) + stream->avail_out = PAGE_CACHE_SIZE; } if (!zlib_init) { zlib_err = zlib_inflateInit(stream); if (zlib_err != Z_OK) { - ERROR("zlib_inflateInit returned unexpected " - "result 0x%x, srclength %d\n", - zlib_err, srclength); - goto release_mutex; + squashfs_finish_page(output); + goto out; } zlib_init = 1; } @@ -109,29 +103,21 @@ static int zlib_uncompress(struct squashfs_sb_info *msblk, void **buffer, put_bh(bh[k++]); } while (zlib_err == Z_OK); - if (zlib_err != Z_STREAM_END) { - ERROR("zlib_inflate error, data probably corrupt\n"); - goto release_mutex; - } + squashfs_finish_page(output); - zlib_err = zlib_inflateEnd(stream); - if (zlib_err != Z_OK) { - ERROR("zlib_inflate error, data probably corrupt\n"); - goto release_mutex; - } + if (zlib_err != Z_STREAM_END) + goto out; - if (k < b) { - ERROR("zlib_uncompress error, data remaining\n"); - goto release_mutex; - } + zlib_err = zlib_inflateEnd(stream); + if (zlib_err != Z_OK) + goto out; - length = stream->total_out; - mutex_unlock(&msblk->read_data_mutex); - return length; + if (k < b) + goto out; -release_mutex: - mutex_unlock(&msblk->read_data_mutex); + return stream->total_out; +out: for (; k < b; k++) put_bh(bh[k]); diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index c327d4ee1235..4742e58f3fc5 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -60,6 +60,7 @@ static int sysv_remount(struct super_block *sb, int *flags, char *data) { struct sysv_sb_info *sbi = SYSV_SB(sb); + sync_filesystem(sb); if (sbi->s_forced_ro) *flags |= MS_RDONLY; return 0; diff --git a/fs/timerfd.c b/fs/timerfd.c index 32b644f03690..0013142c0475 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -8,6 +8,7 @@ * */ +#include <linux/alarmtimer.h> #include <linux/file.h> #include <linux/poll.h> #include <linux/init.h> @@ -26,7 +27,10 @@ #include <linux/rcupdate.h> struct timerfd_ctx { - struct hrtimer tmr; + union { + struct hrtimer tmr; + struct alarm alarm; + } t; ktime_t tintv; ktime_t moffs; wait_queue_head_t wqh; @@ -41,14 +45,19 @@ struct timerfd_ctx { static LIST_HEAD(cancel_list); static DEFINE_SPINLOCK(cancel_lock); +static inline bool isalarm(struct timerfd_ctx *ctx) +{ + return ctx->clockid == CLOCK_REALTIME_ALARM || + ctx->clockid == CLOCK_BOOTTIME_ALARM; +} + /* * This gets called when the timer event triggers. We set the "expired" * flag, but we do not re-arm the timer (in case it's necessary, * tintv.tv64 != 0) until the timer is accessed. */ -static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr) +static void timerfd_triggered(struct timerfd_ctx *ctx) { - struct timerfd_ctx *ctx = container_of(htmr, struct timerfd_ctx, tmr); unsigned long flags; spin_lock_irqsave(&ctx->wqh.lock, flags); @@ -56,10 +65,25 @@ static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr) ctx->ticks++; wake_up_locked(&ctx->wqh); spin_unlock_irqrestore(&ctx->wqh.lock, flags); +} +static enum hrtimer_restart timerfd_tmrproc(struct hrtimer *htmr) +{ + struct timerfd_ctx *ctx = container_of(htmr, struct timerfd_ctx, + t.tmr); + timerfd_triggered(ctx); return HRTIMER_NORESTART; } +static enum alarmtimer_restart timerfd_alarmproc(struct alarm *alarm, + ktime_t now) +{ + struct timerfd_ctx *ctx = container_of(alarm, struct timerfd_ctx, + t.alarm); + timerfd_triggered(ctx); + return ALARMTIMER_NORESTART; +} + /* * Called when the clock was set to cancel the timers in the cancel * list. This will wake up processes waiting on these timers. The @@ -107,8 +131,9 @@ static bool timerfd_canceled(struct timerfd_ctx *ctx) static void timerfd_setup_cancel(struct timerfd_ctx *ctx, int flags) { - if (ctx->clockid == CLOCK_REALTIME && (flags & TFD_TIMER_ABSTIME) && - (flags & TFD_TIMER_CANCEL_ON_SET)) { + if ((ctx->clockid == CLOCK_REALTIME || + ctx->clockid == CLOCK_REALTIME_ALARM) && + (flags & TFD_TIMER_ABSTIME) && (flags & TFD_TIMER_CANCEL_ON_SET)) { if (!ctx->might_cancel) { ctx->might_cancel = true; spin_lock(&cancel_lock); @@ -124,7 +149,11 @@ static ktime_t timerfd_get_remaining(struct timerfd_ctx *ctx) { ktime_t remaining; - remaining = hrtimer_expires_remaining(&ctx->tmr); + if (isalarm(ctx)) + remaining = alarm_expires_remaining(&ctx->t.alarm); + else + remaining = hrtimer_expires_remaining(&ctx->t.tmr); + return remaining.tv64 < 0 ? ktime_set(0, 0): remaining; } @@ -142,11 +171,28 @@ static int timerfd_setup(struct timerfd_ctx *ctx, int flags, ctx->expired = 0; ctx->ticks = 0; ctx->tintv = timespec_to_ktime(ktmr->it_interval); - hrtimer_init(&ctx->tmr, clockid, htmode); - hrtimer_set_expires(&ctx->tmr, texp); - ctx->tmr.function = timerfd_tmrproc; + + if (isalarm(ctx)) { + alarm_init(&ctx->t.alarm, + ctx->clockid == CLOCK_REALTIME_ALARM ? + ALARM_REALTIME : ALARM_BOOTTIME, + timerfd_alarmproc); + } else { + hrtimer_init(&ctx->t.tmr, clockid, htmode); + hrtimer_set_expires(&ctx->t.tmr, texp); + ctx->t.tmr.function = timerfd_tmrproc; + } + if (texp.tv64 != 0) { - hrtimer_start(&ctx->tmr, texp, htmode); + if (isalarm(ctx)) { + if (flags & TFD_TIMER_ABSTIME) + alarm_start(&ctx->t.alarm, texp); + else + alarm_start_relative(&ctx->t.alarm, texp); + } else { + hrtimer_start(&ctx->t.tmr, texp, htmode); + } + if (timerfd_canceled(ctx)) return -ECANCELED; } @@ -158,7 +204,11 @@ static int timerfd_release(struct inode *inode, struct file *file) struct timerfd_ctx *ctx = file->private_data; timerfd_remove_cancel(ctx); - hrtimer_cancel(&ctx->tmr); + + if (isalarm(ctx)) + alarm_cancel(&ctx->t.alarm); + else + hrtimer_cancel(&ctx->t.tmr); kfree_rcu(ctx, rcu); return 0; } @@ -215,9 +265,15 @@ static ssize_t timerfd_read(struct file *file, char __user *buf, size_t count, * callback to avoid DoS attacks specifying a very * short timer period. */ - ticks += hrtimer_forward_now(&ctx->tmr, - ctx->tintv) - 1; - hrtimer_restart(&ctx->tmr); + if (isalarm(ctx)) { + ticks += alarm_forward_now( + &ctx->t.alarm, ctx->tintv) - 1; + alarm_restart(&ctx->t.alarm); + } else { + ticks += hrtimer_forward_now(&ctx->t.tmr, + ctx->tintv) - 1; + hrtimer_restart(&ctx->t.tmr); + } } ctx->expired = 0; ctx->ticks = 0; @@ -259,7 +315,10 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) if ((flags & ~TFD_CREATE_FLAGS) || (clockid != CLOCK_MONOTONIC && - clockid != CLOCK_REALTIME)) + clockid != CLOCK_REALTIME && + clockid != CLOCK_REALTIME_ALARM && + clockid != CLOCK_BOOTTIME && + clockid != CLOCK_BOOTTIME_ALARM)) return -EINVAL; ctx = kzalloc(sizeof(*ctx), GFP_KERNEL); @@ -268,7 +327,15 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) init_waitqueue_head(&ctx->wqh); ctx->clockid = clockid; - hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); + + if (isalarm(ctx)) + alarm_init(&ctx->t.alarm, + ctx->clockid == CLOCK_REALTIME_ALARM ? + ALARM_REALTIME : ALARM_BOOTTIME, + timerfd_alarmproc); + else + hrtimer_init(&ctx->t.tmr, clockid, HRTIMER_MODE_ABS); + ctx->moffs = ktime_get_monotonic_offset(); ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, @@ -305,8 +372,14 @@ static int do_timerfd_settime(int ufd, int flags, */ for (;;) { spin_lock_irq(&ctx->wqh.lock); - if (hrtimer_try_to_cancel(&ctx->tmr) >= 0) - break; + + if (isalarm(ctx)) { + if (alarm_try_to_cancel(&ctx->t.alarm) >= 0) + break; + } else { + if (hrtimer_try_to_cancel(&ctx->t.tmr) >= 0) + break; + } spin_unlock_irq(&ctx->wqh.lock); cpu_relax(); } @@ -317,8 +390,12 @@ static int do_timerfd_settime(int ufd, int flags, * We do not update "ticks" and "expired" since the timer will be * re-programmed again in the following timerfd_setup() call. */ - if (ctx->expired && ctx->tintv.tv64) - hrtimer_forward_now(&ctx->tmr, ctx->tintv); + if (ctx->expired && ctx->tintv.tv64) { + if (isalarm(ctx)) + alarm_forward_now(&ctx->t.alarm, ctx->tintv); + else + hrtimer_forward_now(&ctx->t.tmr, ctx->tintv); + } old->it_value = ktime_to_timespec(timerfd_get_remaining(ctx)); old->it_interval = ktime_to_timespec(ctx->tintv); @@ -345,9 +422,18 @@ static int do_timerfd_gettime(int ufd, struct itimerspec *t) spin_lock_irq(&ctx->wqh.lock); if (ctx->expired && ctx->tintv.tv64) { ctx->expired = 0; - ctx->ticks += - hrtimer_forward_now(&ctx->tmr, ctx->tintv) - 1; - hrtimer_restart(&ctx->tmr); + + if (isalarm(ctx)) { + ctx->ticks += + alarm_forward_now( + &ctx->t.alarm, ctx->tintv) - 1; + alarm_restart(&ctx->t.alarm); + } else { + ctx->ticks += + hrtimer_forward_now(&ctx->t.tmr, ctx->tintv) + - 1; + hrtimer_restart(&ctx->t.tmr); + } } t->it_value = ktime_to_timespec(timerfd_get_remaining(ctx)); t->it_interval = ktime_to_timespec(ctx->tintv); diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 05115d719408..3bf098a82dac 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -1840,6 +1840,7 @@ static int ubifs_remount_fs(struct super_block *sb, int *flags, char *data) int err; struct ubifs_info *c = sb->s_fs_info; + sync_filesystem(sb); dbg_gen("old flags %#lx, new flags %#x", sb->s_flags, *flags); err = ubifs_parse_options(c, data, 1); diff --git a/fs/udf/super.c b/fs/udf/super.c index 839a2bad7f45..32f5297dfbab 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -629,6 +629,7 @@ static int udf_remount_fs(struct super_block *sb, int *flags, char *options) struct udf_options uopt; struct udf_sb_info *sbi = UDF_SB(sb); int error = 0; + sync_filesystem(sb); if (sbi->s_lvid_bh) { int write_rev = le16_to_cpu(udf_sb_lvidiu(sbi)->minUDFWriteRev); diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 329f2f53b7ed..b8c6791f046f 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1280,6 +1280,7 @@ static int ufs_remount (struct super_block *sb, int *mount_flags, char *data) unsigned new_mount_opt, ufstype; unsigned flags; + sync_filesystem(sb); lock_ufs(sb); mutex_lock(&UFS_SB(sb)->s_lock); uspi = UFS_SB(sb)->s_uspi; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 3033ba5e9762..478c0ad5e36f 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1218,6 +1218,7 @@ xfs_fs_remount( char *p; int error; + sync_filesystem(sb); while ((p = strsep(&options, ",")) != NULL) { int token; diff --git a/include/asm-generic/barrier.h b/include/asm-generic/barrier.h index 639d7a4d033b..01613b382b0e 100644 --- a/include/asm-generic/barrier.h +++ b/include/asm-generic/barrier.h @@ -46,5 +46,20 @@ #define read_barrier_depends() do {} while (0) #define smp_read_barrier_depends() do {} while (0) +#define smp_store_release(p, v) \ +do { \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ACCESS_ONCE(*p) = (v); \ +} while (0) + +#define smp_load_acquire(p) \ +({ \ + typeof(*p) ___p1 = ACCESS_ONCE(*p); \ + compiletime_assert_atomic_type(*p); \ + smp_mb(); \ + ___p1; \ +}) + #endif /* !__ASSEMBLY__ */ #endif /* __ASM_GENERIC_BARRIER_H */ diff --git a/include/asm-generic/seccomp.h b/include/asm-generic/seccomp.h new file mode 100644 index 000000000000..663ac3dc02ea --- /dev/null +++ b/include/asm-generic/seccomp.h @@ -0,0 +1,29 @@ +/* + * include/asm-generic/seccomp.h + * + * Copyright (C) 2014 Linaro Limited + * Author: AKASHI Takahiro <takahiro.akashi <at> 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. + */ +#ifndef _ASM_GENERIC_SECCOMP_H +#define _ASM_GENERIC_SECCOMP_H + +#include <asm-generic/unistd.h> + +#if defined(CONFIG_COMPAT) && !defined(__NR_seccomp_read_32) +#define __NR_seccomp_read_32 __NR_read +#define __NR_seccomp_write_32 __NR_write +#define __NR_seccomp_exit_32 __NR_exit +#define __NR_seccomp_sigreturn_32 __NR_rt_sigreturn +#endif /* CONFIG_COMPAT && ! already defined */ + +#define __NR_seccomp_read __NR_read +#define __NR_seccomp_write __NR_write +#define __NR_seccomp_exit __NR_exit +#define __NR_seccomp_sigreturn __NR_rt_sigreturn + +#endif /* _ASM_GENERIC_SECCOMP_H */ + diff --git a/include/asm-generic/syscall.h b/include/asm-generic/syscall.h index 5b09392db673..d401e5463fb0 100644 --- a/include/asm-generic/syscall.h +++ b/include/asm-generic/syscall.h @@ -144,8 +144,6 @@ void syscall_set_arguments(struct task_struct *task, struct pt_regs *regs, /** * syscall_get_arch - return the AUDIT_ARCH for the current system call - * @task: task of interest, must be in system call entry tracing - * @regs: task_pt_regs() of @task * * Returns the AUDIT_ARCH_* based on the system call convention in use. * @@ -155,5 +153,5 @@ void syscall_set_arguments(struct task_struct *task, struct pt_regs *regs, * Architectures which permit CONFIG_HAVE_ARCH_SECCOMP_FILTER must * provide an implementation of this. */ -int syscall_get_arch(struct task_struct *task, struct pt_regs *regs); +int syscall_get_arch(void); #endif /* _ASM_SYSCALL_H */ 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/alarmtimer.h b/include/linux/alarmtimer.h index 9069694e70eb..a899402a5a0e 100644 --- a/include/linux/alarmtimer.h +++ b/include/linux/alarmtimer.h @@ -44,10 +44,14 @@ struct alarm { void alarm_init(struct alarm *alarm, enum alarmtimer_type type, enum alarmtimer_restart (*function)(struct alarm *, ktime_t)); int alarm_start(struct alarm *alarm, ktime_t start); +int alarm_start_relative(struct alarm *alarm, ktime_t start); +void alarm_restart(struct alarm *alarm); int alarm_try_to_cancel(struct alarm *alarm); int alarm_cancel(struct alarm *alarm); u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval); +u64 alarm_forward_now(struct alarm *alarm, ktime_t interval); +ktime_t alarm_expires_remaining(const struct alarm *alarm); /* Provide way to access the rtc device being used by alarmtimers */ struct rtc_device *alarmtimer_get_rtcdev(void); diff --git a/include/linux/amba/mmci.h b/include/linux/amba/mmci.h index 32a89cf5ec45..9d1d9caf2611 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; +}; /* @@ -73,6 +82,9 @@ struct mmci_platform_data { bool (*dma_filter)(struct dma_chan *chan, void *filter_param); void *dma_rx_param; void *dma_tx_param; + unsigned int status_irq; + struct embedded_sdio_data *embedded_sdio; + int (*register_status_notify)(void (*callback)(int card_present, void *dev_id), void *dev_id); }; #endif diff --git a/include/linux/android_aid.h b/include/linux/android_aid.h new file mode 100644 index 000000000000..06264b8be5f0 --- /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 3001 /* was NET_BT_ADMIN */ +#define AID_OBSOLETE_001 3002 /* was NET_BT */ +#define AID_INET 3003 +#define AID_NET_RAW 3004 +#define AID_NET_ADMIN 3005 +#define AID_NET_BW_STATS 3006 /* read bandwidth statistics */ +#define AID_NET_BW_ACCT 3007 /* change bandwidth statistics accounting */ + +#endif diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index 8852d370c720..a3997d560114 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -578,6 +578,7 @@ struct cgroup_subsys { void (*css_offline)(struct cgroup *cgrp); void (*css_free)(struct cgroup *cgrp); + int (*allow_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); int (*can_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); void (*cancel_attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); void (*attach)(struct cgroup *cgrp, struct cgroup_taskset *tset); @@ -868,6 +869,17 @@ unsigned short css_id(struct cgroup_subsys_state *css); unsigned short css_depth(struct cgroup_subsys_state *css); struct cgroup_subsys_state *cgroup_css_from_dir(struct file *f, int id); +/* + * Default Android check for whether the current process is allowed to move a + * task across cgroups, either because CAP_SYS_NICE is set or because the uid + * of the calling process is the same as the moved task or because we are + * running as root. + * Returns 0 if this is allowed, or -EACCES otherwise. + */ +int subsys_cgroup_allow_attach(struct cgroup *cgrp, + struct cgroup_taskset *tset); + + #else /* !CONFIG_CGROUPS */ static inline int cgroup_init_early(void) { return 0; } diff --git a/include/linux/compiler.h b/include/linux/compiler.h index a2329c5e6206..2472740d7ab2 100644 --- a/include/linux/compiler.h +++ b/include/linux/compiler.h @@ -302,6 +302,11 @@ void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect); # define __same_type(a, b) __builtin_types_compatible_p(typeof(a), typeof(b)) #endif +/* Is this type a native word size -- useful for atomic operations */ +#ifndef __native_word +# define __native_word(t) (sizeof(t) == sizeof(int) || sizeof(t) == sizeof(long)) +#endif + /* Compile time object size, -1 for unknown */ #ifndef __compiletime_object_size # define __compiletime_object_size(obj) -1 @@ -341,6 +346,10 @@ void ftrace_likely_update(struct ftrace_branch_data *f, int val, int expect); #define compiletime_assert(condition, msg) \ _compiletime_assert(condition, msg, __compiletime_assert_, __LINE__) +#define compiletime_assert_atomic_type(t) \ + compiletime_assert(__native_word(t), \ + "Need native word sized stores/loads for atomicity.") + /* * Prevent the compiler from merging or refetching accesses. The compiler * is also forbidden from reordering successive instances of ACCESS_ONCE(), diff --git a/include/linux/cpu.h b/include/linux/cpu.h index 770c8047e998..a482811399a9 100644 --- a/include/linux/cpu.h +++ b/include/linux/cpu.h @@ -280,4 +280,11 @@ void arch_cpu_idle_enter(void); void arch_cpu_idle_exit(void); void arch_cpu_idle_dead(void); +#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 1a81b7470bc4..df55b769f550 100644 --- a/include/linux/cpufreq.h +++ b/include/linux/cpufreq.h @@ -338,9 +338,11 @@ const char *cpufreq_get_current_driver(void); /********************************************************************* * CPUFREQ 2.6. INTERFACE * *********************************************************************/ +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); +struct kobject *get_governor_parent_kobj(struct cpufreq_policy *policy); #ifdef CONFIG_CPU_FREQ /* query the current CPU frequency (in kHz). If zero, cpufreq couldn't detect it */ @@ -394,6 +396,9 @@ 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) #endif diff --git a/include/linux/debug_locks.h b/include/linux/debug_locks.h index 21ca773f77bf..822c1354f3a6 100644 --- a/include/linux/debug_locks.h +++ b/include/linux/debug_locks.h @@ -51,7 +51,7 @@ struct task_struct; extern void debug_show_all_locks(void); extern void debug_show_held_locks(struct task_struct *task); extern void debug_check_no_locks_freed(const void *from, unsigned long len); -extern void debug_check_no_locks_held(struct task_struct *task); +extern void debug_check_no_locks_held(void); #else static inline void debug_show_all_locks(void) { @@ -67,7 +67,7 @@ debug_check_no_locks_freed(const void *from, unsigned long len) } static inline void -debug_check_no_locks_held(struct task_struct *task) +debug_check_no_locks_held(void) { } #endif diff --git a/include/linux/freezer.h b/include/linux/freezer.h index e70df40d84f6..7fd81b8c4897 100644 --- a/include/linux/freezer.h +++ b/include/linux/freezer.h @@ -3,6 +3,7 @@ #ifndef FREEZER_H_INCLUDED #define FREEZER_H_INCLUDED +#include <linux/debug_locks.h> #include <linux/sched.h> #include <linux/wait.h> #include <linux/atomic.h> @@ -46,7 +47,11 @@ extern int freeze_kernel_threads(void); extern void thaw_processes(void); extern void thaw_kernel_threads(void); -static inline bool try_to_freeze(void) +/* + * DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION + * If try_to_freeze causes a lockdep warning it means the caller may deadlock + */ +static inline bool try_to_freeze_unsafe(void) { might_sleep(); if (likely(!freezing(current))) @@ -54,6 +59,13 @@ static inline bool try_to_freeze(void) return __refrigerator(false); } +static inline bool try_to_freeze(void) +{ + if (!(current->flags & PF_NOFREEZE)) + debug_check_no_locks_held(); + return try_to_freeze_unsafe(); +} + extern bool freeze_task(struct task_struct *p); extern bool set_freezable(void); @@ -115,6 +127,14 @@ static inline void freezer_count(void) try_to_freeze(); } +/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ +static inline void freezer_count_unsafe(void) +{ + current->flags &= ~PF_FREEZER_SKIP; + smp_mb(); + try_to_freeze_unsafe(); +} + /** * freezer_should_skip - whether to skip a task when determining frozen * state is reached @@ -139,28 +159,86 @@ static inline bool freezer_should_skip(struct task_struct *p) } /* - * These macros are intended to be used whenever you want allow a sleeping + * These functions are intended to be used whenever you want allow a sleeping * task to be frozen. Note that neither return any clear indication of * whether a freeze event happened while in this function. */ /* Like schedule(), but should not block the freezer. */ -#define freezable_schedule() \ -({ \ - freezer_do_not_count(); \ - schedule(); \ - freezer_count(); \ -}) +static inline void freezable_schedule(void) +{ + freezer_do_not_count(); + schedule(); + freezer_count(); +} + +/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ +static inline void freezable_schedule_unsafe(void) +{ + freezer_do_not_count(); + schedule(); + freezer_count_unsafe(); +} + +/* + * Like freezable_schedule_timeout(), but should not block the freezer. Do not + * call this with locks held. + */ +static inline long freezable_schedule_timeout(long timeout) +{ + long __retval; + freezer_do_not_count(); + __retval = schedule_timeout(timeout); + freezer_count(); + return __retval; +} + +/* + * Like schedule_timeout_interruptible(), but should not block the freezer. Do not + * call this with locks held. + */ +static inline long freezable_schedule_timeout_interruptible(long timeout) +{ + long __retval; + freezer_do_not_count(); + __retval = schedule_timeout_interruptible(timeout); + freezer_count(); + return __retval; +} /* Like schedule_timeout_killable(), but should not block the freezer. */ -#define freezable_schedule_timeout_killable(timeout) \ -({ \ - long __retval; \ - freezer_do_not_count(); \ - __retval = schedule_timeout_killable(timeout); \ - freezer_count(); \ - __retval; \ -}) +static inline long freezable_schedule_timeout_killable(long timeout) +{ + long __retval; + freezer_do_not_count(); + __retval = schedule_timeout_killable(timeout); + freezer_count(); + return __retval; +} + +/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ +static inline long freezable_schedule_timeout_killable_unsafe(long timeout) +{ + long __retval; + freezer_do_not_count(); + __retval = schedule_timeout_killable(timeout); + freezer_count_unsafe(); + return __retval; +} + +/* + * Like schedule_hrtimeout_range(), but should not block the freezer. Do not + * call this with locks held. + */ +static inline int freezable_schedule_hrtimeout_range(ktime_t *expires, + unsigned long delta, const enum hrtimer_mode mode) +{ + int __retval; + freezer_do_not_count(); + __retval = schedule_hrtimeout_range(expires, delta, mode); + freezer_count(); + return __retval; +} /* * Freezer-friendly wrappers around wait_event_interruptible(), @@ -177,33 +255,45 @@ static inline bool freezer_should_skip(struct task_struct *p) __retval; \ }) +/* DO NOT ADD ANY NEW CALLERS OF THIS FUNCTION */ +#define wait_event_freezekillable_unsafe(wq, condition) \ +({ \ + int __retval; \ + freezer_do_not_count(); \ + __retval = wait_event_killable(wq, (condition)); \ + freezer_count_unsafe(); \ + __retval; \ +}) + #define wait_event_freezable(wq, condition) \ ({ \ int __retval; \ - for (;;) { \ - __retval = wait_event_interruptible(wq, \ - (condition) || freezing(current)); \ - if (__retval || (condition)) \ - break; \ - try_to_freeze(); \ - } \ + freezer_do_not_count(); \ + __retval = wait_event_interruptible(wq, (condition)); \ + freezer_count(); \ __retval; \ }) #define wait_event_freezable_timeout(wq, condition, timeout) \ ({ \ long __retval = timeout; \ - for (;;) { \ - __retval = wait_event_interruptible_timeout(wq, \ - (condition) || freezing(current), \ - __retval); \ - if (__retval <= 0 || (condition)) \ - break; \ - try_to_freeze(); \ - } \ + freezer_do_not_count(); \ + __retval = wait_event_interruptible_timeout(wq, (condition), \ + __retval); \ + freezer_count(); \ __retval; \ }) +#define wait_event_freezable_exclusive(wq, condition) \ +({ \ + int __retval; \ + freezer_do_not_count(); \ + __retval = wait_event_interruptible_exclusive(wq, condition); \ + freezer_count(); \ + __retval; \ +}) + + #else /* !CONFIG_FREEZER */ static inline bool frozen(struct task_struct *p) { return false; } static inline bool freezing(struct task_struct *p) { return false; } @@ -225,18 +315,37 @@ static inline void set_freezable(void) {} #define freezable_schedule() schedule() +#define freezable_schedule_unsafe() schedule() + +#define freezable_schedule_timeout(timeout) schedule_timeout(timeout) + +#define freezable_schedule_timeout_interruptible(timeout) \ + schedule_timeout_interruptible(timeout) + #define freezable_schedule_timeout_killable(timeout) \ schedule_timeout_killable(timeout) +#define freezable_schedule_timeout_killable_unsafe(timeout) \ + schedule_timeout_killable(timeout) + +#define freezable_schedule_hrtimeout_range(expires, delta, mode) \ + schedule_hrtimeout_range(expires, delta, mode) + #define wait_event_freezable(wq, condition) \ wait_event_interruptible(wq, condition) #define wait_event_freezable_timeout(wq, condition, timeout) \ wait_event_interruptible_timeout(wq, condition, timeout) +#define wait_event_freezable_exclusive(wq, condition) \ + wait_event_interruptible_exclusive(wq, condition) + #define wait_event_freezekillable(wq, condition) \ wait_event_killable(wq, condition) +#define wait_event_freezekillable_unsafe(wq, condition) \ + wait_event_killable(wq, condition) + #endif /* !CONFIG_FREEZER */ #endif /* FREEZER_H_INCLUDED */ diff --git a/include/linux/fs.h b/include/linux/fs.h index d57bc5df7225..9a3437377f71 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1506,6 +1506,17 @@ int fiemap_check_flags(struct fiemap_extent_info *fieinfo, u32 fs_flags); * to have different dirent layouts depending on the binary type. */ typedef int (*filldir_t)(void *, const char *, int, loff_t, u64, unsigned); +struct dir_context { + const filldir_t actor; + loff_t pos; +}; + +static inline bool dir_emit(struct dir_context *ctx, + const char *name, int namelen, + u64 ino, unsigned type) +{ + return ctx->actor(ctx, name, namelen, ctx->pos, ino, type) == 0; +} struct block_device_operations; /* These macros are for out of kernel modules to test that @@ -1522,6 +1533,7 @@ struct file_operations { ssize_t (*aio_read) (struct kiocb *, const struct iovec *, unsigned long, loff_t); ssize_t (*aio_write) (struct kiocb *, const struct iovec *, unsigned long, loff_t); int (*readdir) (struct file *, void *, filldir_t); + int (*iterate) (struct file *, struct dir_context *); unsigned int (*poll) (struct file *, struct poll_table_struct *); long (*unlocked_ioctl) (struct file *, unsigned int, unsigned long); long (*compat_ioctl) (struct file *, unsigned int, unsigned long); @@ -2495,6 +2507,7 @@ loff_t inode_get_bytes(struct inode *inode); void inode_set_bytes(struct inode *inode, loff_t bytes); extern int vfs_readdir(struct file *, filldir_t, void *); +extern int iterate_dir(struct file *, struct dir_context *); extern int vfs_stat(const char __user *, struct kstat *); extern int vfs_lstat(const char __user *, struct kstat *); 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/hid.h b/include/linux/hid.h index 4f8aa4733fb6..40e9c9d76d1e 100644 --- a/include/linux/hid.h +++ b/include/linux/hid.h @@ -649,8 +649,8 @@ struct hid_driver { int (*input_mapped)(struct hid_device *hdev, struct hid_input *hidinput, struct hid_field *field, struct hid_usage *usage, unsigned long **bit, int *max); - void (*input_configured)(struct hid_device *hdev, - struct hid_input *hidinput); + int (*input_configured)(struct hid_device *hdev, + struct hid_input *hidinput); void (*feature_mapping)(struct hid_device *hdev, struct hid_field *field, struct hid_usage *usage); 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 aff7ad8a4ea3..dccd621d6377 100644 --- a/include/linux/if_pppox.h +++ b/include/linux/if_pppox.h @@ -41,6 +41,25 @@ 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 length); + int (*backlog_rcv)(struct sock *sk_raw, struct sk_buff *skb); +}; + #include <net/sock.h> struct pppox_sock { @@ -51,6 +70,8 @@ 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/ipv6.h b/include/linux/ipv6.h index b8b7dc755752..ebc6306661eb 100644 --- a/include/linux/ipv6.h +++ b/include/linux/ipv6.h @@ -36,10 +36,12 @@ struct ipv6_devconf { __s32 accept_ra_rt_info_max_plen; #endif #endif + __s32 accept_ra_rt_table; __s32 proxy_ndp; __s32 accept_source_route; #ifdef CONFIG_IPV6_OPTIMISTIC_DAD __s32 optimistic_dad; + __s32 use_optimistic; #endif #ifdef CONFIG_IPV6_MROUTE __s32 mc_forwarding; diff --git a/include/linux/kernel.h b/include/linux/kernel.h index 341551c7b4c8..5ff503bb94f6 100644 --- a/include/linux/kernel.h +++ b/include/linux/kernel.h @@ -805,4 +805,7 @@ static inline void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) { } # define REBUILD_DUE_TO_FTRACE_MCOUNT_RECORD #endif +/* To identify board information in panic logs, set this */ +extern char *mach_panic_string; + #endif 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/lsm_audit.h b/include/linux/lsm_audit.h index 1cc89e9df480..ffb9c9da4f39 100644 --- a/include/linux/lsm_audit.h +++ b/include/linux/lsm_audit.h @@ -40,6 +40,11 @@ struct lsm_network_audit { } fam; }; +struct lsm_ioctlop_audit { + struct path path; + u16 cmd; +}; + /* Auxiliary data to use in generating the audit record. */ struct common_audit_data { char type; @@ -53,6 +58,7 @@ struct common_audit_data { #define LSM_AUDIT_DATA_KMOD 8 #define LSM_AUDIT_DATA_INODE 9 #define LSM_AUDIT_DATA_DENTRY 10 +#define LSM_AUDIT_DATA_IOCTL_OP 11 union { struct path path; struct dentry *dentry; @@ -68,6 +74,7 @@ struct common_audit_data { } key_struct; #endif char *kmod_name; + struct lsm_ioctlop_audit *op; } u; /* this union contains LSM specific data */ union { diff --git a/include/linux/mm.h b/include/linux/mm.h index 0ea351df9e14..d06441c10e32 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -328,6 +328,8 @@ static inline int is_vmalloc_or_module_addr(const void *x) } #endif +extern void kvfree(const void *addr); + static inline void compound_lock(struct page *page) { #ifdef CONFIG_TRANSPARENT_HUGEPAGE @@ -926,6 +928,7 @@ 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 *); extern int can_do_mlock(void); @@ -1182,6 +1185,11 @@ static inline void update_hiwater_vm(struct mm_struct *mm) mm->hiwater_vm = mm->total_vm; } +static inline void reset_mm_hiwater_rss(struct mm_struct *mm) +{ + mm->hiwater_rss = get_mm_rss(mm); +} + static inline void setmax_mm_hiwater_rss(unsigned long *maxrss, struct mm_struct *mm) { @@ -1504,7 +1512,7 @@ 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 mempolicy *, 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 10a9a17342fc..c14df44a9567 100644 --- a/include/linux/mm_types.h +++ b/include/linux/mm_types.h @@ -255,6 +255,10 @@ struct vm_area_struct { * For areas with an address space and backing store, * linkage into the address_space->i_mmap interval tree, or * linkage of vma in the address_space->i_mmap_nonlinear list. + * + * 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. */ union { struct { @@ -262,6 +266,7 @@ struct vm_area_struct { unsigned long rb_subtree_last; } linear; struct list_head nonlinear; + const char __user *anon_name; } shared; /* @@ -465,6 +470,14 @@ static inline cpumask_t *mm_cpumask(struct mm_struct *mm) return mm->cpu_vm_mask_var; } +/* 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->shared.anon_name; +} #if defined(CONFIG_NUMA_BALANCING) || defined(CONFIG_COMPACTION) /* * Memory barriers to keep this state in sync are graciously provided by diff --git a/include/linux/mmc/host.h b/include/linux/mmc/host.h index e326ae2882a0..66b465927602 100644 --- a/include/linux/mmc/host.h +++ b/include/linux/mmc/host.h @@ -15,6 +15,7 @@ #include <linux/sched.h> #include <linux/device.h> #include <linux/fault-inject.h> +#include <linux/wakelock.h> #include <linux/mmc/core.h> #include <linux/mmc/pm.h> @@ -329,12 +330,17 @@ struct mmc_host { int claim_cnt; /* "claim" nesting count */ struct delayed_work detect; + struct wake_lock detect_wake_lock; int detect_change; /* card detect flag */ struct mmc_slot slot; const struct mmc_bus_ops *bus_ops; /* current bus driver */ unsigned int bus_refs; /* reference counter */ + unsigned int bus_resume_flags; +#define MMC_BUSRESUME_MANUAL_RESUME (1 << 0) +#define MMC_BUSRESUME_NEEDS_RESUME (1 << 1) + unsigned int sdio_irqs; struct task_struct *sdio_irq_thread; bool sdio_irq_pending; @@ -362,6 +368,15 @@ struct mmc_host { unsigned int slotno; /* used for sdio acpi binding */ +#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 + unsigned long private[0] ____cacheline_aligned; }; @@ -371,6 +386,14 @@ void mmc_remove_host(struct mmc_host *); void mmc_free_host(struct mmc_host *); void 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; @@ -381,6 +404,18 @@ static inline void *mmc_priv(struct mmc_host *host) #define mmc_dev(x) ((x)->parent) #define mmc_classdev(x) (&(x)->class_dev) #define mmc_hostname(x) (dev_name(&(x)->class_dev)) +#define mmc_bus_needs_resume(host) ((host)->bus_resume_flags & MMC_BUSRESUME_NEEDS_RESUME) +#define mmc_bus_manual_resume(host) ((host)->bus_resume_flags & MMC_BUSRESUME_MANUAL_RESUME) + +static inline void mmc_set_bus_resume_policy(struct mmc_host *host, int manual) +{ + if (manual) + host->bus_resume_flags |= MMC_BUSRESUME_MANUAL_RESUME; + else + host->bus_resume_flags &= ~MMC_BUSRESUME_MANUAL_RESUME; +} + +extern int mmc_resume_bus(struct mmc_host *host); int mmc_suspend_host(struct mmc_host *); int mmc_resume_host(struct mmc_host *); 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 50f0bc952328..dc680c4b50d4 100644..100755 --- 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 { @@ -130,6 +138,8 @@ 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/netfilter/xt_qtaguid.h b/include/linux/netfilter/xt_qtaguid.h new file mode 100644 index 000000000000..ca60fbdec2f3 --- /dev/null +++ b/include/linux/netfilter/xt_qtaguid.h @@ -0,0 +1,13 @@ +#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 + +#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 db50840e6355..c8f8aa0383e5 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 c9722fdf39c5..7c00ee271395 100644 --- a/include/linux/of_fdt.h +++ b/include/linux/of_fdt.h @@ -90,6 +90,27 @@ extern int of_flat_dt_is_compatible(unsigned long node, const char *name); extern int of_flat_dt_match(unsigned long node, const char *const *matches); extern unsigned long of_get_flat_dt_root(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 void early_init_dt_check_for_initrd(unsigned long node); 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/power_supply.h b/include/linux/power_supply.h index 3828cefb4f65..03d921feb094 100644 --- a/include/linux/power_supply.h +++ b/include/linux/power_supply.h @@ -15,6 +15,7 @@ #include <linux/workqueue.h> #include <linux/leds.h> +#include <linux/types.h> struct device; @@ -140,6 +141,12 @@ enum power_supply_property { POWER_SUPPLY_PROP_TIME_TO_FULL_AVG, POWER_SUPPLY_PROP_TYPE, /* use power_supply.type instead */ POWER_SUPPLY_PROP_SCOPE, + /* 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, @@ -160,6 +167,7 @@ enum power_supply_type { union power_supply_propval { int intval; const char *strval; + int64_t int64val; }; struct power_supply { @@ -194,6 +202,8 @@ struct power_supply { /* private */ struct device *dev; struct work_struct changed_work; + spinlock_t changed_lock; + bool changed; #ifdef CONFIG_THERMAL struct thermal_zone_device *tzd; struct thermal_cooling_device *tcd; diff --git a/include/linux/pstore.h b/include/linux/pstore.h index 75d01760c911..289884121d30 100644 --- a/include/linux/pstore.h +++ b/include/linux/pstore.h @@ -35,6 +35,7 @@ enum pstore_type_id { PSTORE_TYPE_MCE = 1, PSTORE_TYPE_CONSOLE = 2, PSTORE_TYPE_FTRACE = 3, + PSTORE_TYPE_PMSG = 4, /* Backport: 7 in upstream 3.19.0-rc3 */ PSTORE_TYPE_UNKNOWN = 255 }; diff --git a/include/linux/pstore_ram.h b/include/linux/pstore_ram.h index 4af3fdc85b01..712757f320a4 100644 --- a/include/linux/pstore_ram.h +++ b/include/linux/pstore_ram.h @@ -68,6 +68,8 @@ 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 @@ -81,6 +83,7 @@ struct ramoops_platform_data { unsigned long record_size; unsigned long console_size; unsigned long ftrace_size; + unsigned long pmsg_size; int dump_oops; struct persistent_ram_ecc_info ecc_info; }; diff --git a/include/linux/sched.h b/include/linux/sched.h index 6ca9630fde2d..5229df9d7107 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1143,13 +1143,12 @@ struct task_struct { * execve */ unsigned in_iowait:1; - /* task may not gain privileges */ - unsigned no_new_privs:1; - /* Revert to default priority/policy when forking */ unsigned sched_reset_on_fork:1; unsigned sched_contributes_to_load:1; + unsigned long atomic_flags; /* Flags needing atomic access. */ + pid_t pid; pid_t tgid; @@ -1639,6 +1638,9 @@ static inline cputime_t task_gtime(struct task_struct *t) extern void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st); extern void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st); +extern int task_free_register(struct notifier_block *n); +extern int task_free_unregister(struct notifier_block *n); + /* * Per process flags */ @@ -1719,6 +1721,19 @@ static inline void memalloc_noio_restore(unsigned int flags) current->flags = (current->flags & ~PF_MEMALLOC_NOIO) | flags; } +/* Per-process atomic flags. */ +#define PFA_NO_NEW_PRIVS 0x00000001 /* May not gain new privileges. */ + +static inline bool task_no_new_privs(struct task_struct *p) +{ + return test_bit(PFA_NO_NEW_PRIVS, &p->atomic_flags); +} + +static inline void task_set_no_new_privs(struct task_struct *p) +{ + set_bit(PFA_NO_NEW_PRIVS, &p->atomic_flags); +} + /* * task->jobctl flags */ diff --git a/include/linux/seccomp.h b/include/linux/seccomp.h index 6f19cfd1840e..9687691799ff 100644 --- a/include/linux/seccomp.h +++ b/include/linux/seccomp.h @@ -3,6 +3,8 @@ #include <uapi/linux/seccomp.h> +#define SECCOMP_FILTER_FLAG_MASK (SECCOMP_FILTER_FLAG_TSYNC) + #ifdef CONFIG_SECCOMP #include <linux/thread_info.h> @@ -14,11 +16,11 @@ struct seccomp_filter; * * @mode: indicates one of the valid values above for controlled * system calls available to a process. - * @filter: The metadata and ruleset for determining what system calls - * are allowed for a task. + * @filter: must always point to a valid seccomp-filter or NULL as it is + * accessed without locking during system call entry. * * @filter must only be accessed from the context of current as there - * is no locking. + * is no read locking. */ struct seccomp { int mode; diff --git a/include/linux/security.h b/include/linux/security.h index 4686491852a7..17e1888ff505 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -1402,6 +1402,11 @@ static inline void security_free_mnt_opts(struct security_mnt_opts *opts) struct security_operations { char name[SECURITY_NAME_MAX + 1]; + int (*binder_set_context_mgr) (struct task_struct *mgr); + int (*binder_transaction) (struct task_struct *from, struct task_struct *to); + int (*binder_transfer_binder) (struct task_struct *from, struct task_struct *to); + int (*binder_transfer_file) (struct task_struct *from, struct task_struct *to, struct file *file); + int (*ptrace_access_check) (struct task_struct *child, unsigned int mode); int (*ptrace_traceme) (struct task_struct *parent); int (*capget) (struct task_struct *target, @@ -1690,6 +1695,10 @@ extern void __init security_fixup_ops(struct security_operations *ops); /* Security operations */ +int security_binder_set_context_mgr(struct task_struct *mgr); +int security_binder_transaction(struct task_struct *from, struct task_struct *to); +int security_binder_transfer_binder(struct task_struct *from, struct task_struct *to); +int security_binder_transfer_file(struct task_struct *from, struct task_struct *to, struct file *file); int security_ptrace_access_check(struct task_struct *child, unsigned int mode); int security_ptrace_traceme(struct task_struct *parent); int security_capget(struct task_struct *target, @@ -1869,6 +1878,26 @@ static inline int security_init(void) return 0; } +static inline int security_binder_set_context_mgr(struct task_struct *mgr) +{ + return 0; +} + +static inline int security_binder_transaction(struct task_struct *from, struct task_struct *to) +{ + return 0; +} + +static inline int security_binder_transfer_binder(struct task_struct *from, struct task_struct *to) +{ + return 0; +} + +static inline int security_binder_transfer_file(struct task_struct *from, struct task_struct *to, struct file *file) +{ + return 0; +} + static inline int security_ptrace_access_check(struct task_struct *child, unsigned int mode) { diff --git a/include/linux/serial_core.h b/include/linux/serial_core.h index c2b355fd921a..21c074b2200e 100644 --- a/include/linux/serial_core.h +++ b/include/linux/serial_core.h @@ -60,6 +60,7 @@ struct uart_ops { void (*pm)(struct uart_port *, unsigned int state, unsigned int oldstate); int (*set_wake)(struct uart_port *, unsigned int state); + void (*wake_peer)(struct uart_port *); /* * Return a string describing the type of the port diff --git a/include/linux/suspend.h b/include/linux/suspend.h index d4e3f16d5e89..a34821358ae5 100644 --- a/include/linux/suspend.h +++ b/include/linux/suspend.h @@ -363,7 +363,7 @@ extern bool pm_wakeup_pending(void); 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_get_active_wakeup_sources(char *pending_sources, size_t max); static inline void lock_system_sleep(void) { current->flags |= PF_FREEZER_SKIP; diff --git a/include/linux/switch.h b/include/linux/switch.h new file mode 100644 index 000000000000..3e4c748e343a --- /dev/null +++ b/include/linux/switch.h @@ -0,0 +1,53 @@ +/* + * Switch class 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_SWITCH_H__ +#define __LINUX_SWITCH_H__ + +struct switch_dev { + const char *name; + struct device *dev; + int index; + int state; + + ssize_t (*print_name)(struct switch_dev *sdev, char *buf); + ssize_t (*print_state)(struct switch_dev *sdev, char *buf); +}; + +struct gpio_switch_platform_data { + const char *name; + unsigned gpio; + + /* if NULL, switch_dev.name will be printed */ + const char *name_on; + const char *name_off; + /* if NULL, "0" or "1" will be printed */ + const char *state_on; + const char *state_off; +}; + +extern int switch_dev_register(struct switch_dev *sdev); +extern void switch_dev_unregister(struct switch_dev *sdev); + +static inline int switch_get_state(struct switch_dev *sdev) +{ + return sdev->state; +} + +extern void switch_set_state(struct switch_dev *sdev, int state); + +#endif /* __LINUX_SWITCH_H__ */ diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h index 84662ecc7b51..4e98d7174134 100644 --- a/include/linux/syscalls.h +++ b/include/linux/syscalls.h @@ -846,4 +846,6 @@ asmlinkage long sys_process_vm_writev(pid_t pid, asmlinkage long sys_kcmp(pid_t pid1, pid_t pid2, int type, unsigned long idx1, unsigned long idx2); asmlinkage long sys_finit_module(int fd, const char __user *uargs, int flags); +asmlinkage long sys_seccomp(unsigned int op, unsigned int flags, + const char __user *uargs); #endif diff --git a/include/linux/uid_stat.h b/include/linux/uid_stat.h new file mode 100644 index 000000000000..6bd6c4e52d17 --- /dev/null +++ b/include/linux/uid_stat.h @@ -0,0 +1,29 @@ +/* include/linux/uid_stat.h + * + * Copyright (C) 2008-2009 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. + * + */ + +#ifndef __uid_stat_h +#define __uid_stat_h + +/* Contains definitions for resource tracking per uid. */ + +#ifdef CONFIG_UID_STAT +int uid_stat_tcp_snd(uid_t uid, int size); +int uid_stat_tcp_rcv(uid_t uid, int size); +#else +#define uid_stat_tcp_snd(uid, size) do {} while (0); +#define uid_stat_tcp_rcv(uid, size) do {} while (0); +#endif + +#endif /* _LINUX_UID_STAT_H */ 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/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..ad8b76936c7f --- /dev/null +++ b/include/linux/wakeup_reason.h @@ -0,0 +1,27 @@ +/* + * 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); +void log_suspend_abort_reason(const char *fmt, ...); +int check_wakeup_reason(int irq); + +#endif /* _LINUX_WAKEUP_REASON_H */ diff --git a/include/linux/wifi_tiwlan.h b/include/linux/wifi_tiwlan.h new file mode 100644 index 000000000000..f07e0679fb82 --- /dev/null +++ b/include/linux/wifi_tiwlan.h @@ -0,0 +1,27 @@ +/* include/linux/wifi_tiwlan.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_WIFI_TIWLAN_H_ +#define _LINUX_WIFI_TIWLAN_H_ + +#include <linux/wlan_plat.h> + +#define WMPA_NUMBER_OF_SECTIONS 3 +#define WMPA_NUMBER_OF_BUFFERS 160 +#define WMPA_SECTION_HEADER 24 +#define WMPA_SECTION_SIZE_0 (WMPA_NUMBER_OF_BUFFERS * 64) +#define WMPA_SECTION_SIZE_1 (WMPA_NUMBER_OF_BUFFERS * 256) +#define WMPA_SECTION_SIZE_2 (WMPA_NUMBER_OF_BUFFERS * 2048) + +#endif 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/activity_stats.h b/include/net/activity_stats.h new file mode 100644 index 000000000000..10e4c1506eeb --- /dev/null +++ b/include/net/activity_stats.h @@ -0,0 +1,25 @@ +/* + * 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) + */ + +#ifndef __activity_stats_h +#define __activity_stats_h + +#ifdef CONFIG_NET_ACTIVITY_STATS +void activity_stats_update(void); +#else +#define activity_stats_update(void) {} +#endif + +#endif /* _NET_ACTIVITY_STATS_H */ diff --git a/include/net/addrconf.h b/include/net/addrconf.h index 25100687babb..2cbf0baa6226 100644 --- a/include/net/addrconf.h +++ b/include/net/addrconf.h @@ -190,6 +190,8 @@ static inline bool ipv6_is_mld(struct sk_buff *skb, int nexthdr, int offset) extern 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) */ @@ -201,6 +203,7 @@ extern int ipv6_dev_ac_inc(struct net_device *dev, const struct in6_addr *addr); extern int __ipv6_dev_ac_dec(struct inet6_dev *idev, const struct in6_addr *addr); extern bool ipv6_chk_acast_addr(struct net *net, struct net_device *dev, const struct in6_addr *addr); +extern void ipv6_ac_destroy_dev(struct inet6_dev *idev); /* Device notifier */ diff --git a/include/net/bluetooth/hci.h b/include/net/bluetooth/hci.h index 0ef00066dae8..db43501b7599 100644 --- a/include/net/bluetooth/hci.h +++ b/include/net/bluetooth/hci.h @@ -199,8 +199,10 @@ enum { #define ESCO_2EV5 0x0100 #define ESCO_3EV5 0x0200 -#define SCO_ESCO_MASK (ESCO_HV1 | ESCO_HV2 | ESCO_HV3) -#define EDR_ESCO_MASK (ESCO_2EV3 | ESCO_3EV3 | ESCO_2EV5 | ESCO_3EV5) +#define SCO_ESCO_MASK (ESCO_HV1 | ESCO_HV2 | ESCO_HV3) +#define EDR_ESCO_MASK (ESCO_2EV3 | ESCO_3EV3 | ESCO_2EV5 | ESCO_3EV5) +#define ALL_ESCO_MASK (SCO_ESCO_MASK | ESCO_EV3 | ESCO_EV4 | ESCO_EV5 | \ + EDR_ESCO_MASK) /* ACL flags */ #define ACL_START_NO_FLUSH 0x00 @@ -1629,6 +1631,9 @@ struct hci_conn_info { __u8 out; __u16 state; __u32 link_mode; + __u32 mtu; + __u32 cnt; + __u32 pkts; }; struct hci_dev_req { diff --git a/include/net/bluetooth/hci_core.h b/include/net/bluetooth/hci_core.h index 7cb6d360d147..57123eeb21aa 100644 --- a/include/net/bluetooth/hci_core.h +++ b/include/net/bluetooth/hci_core.h @@ -581,7 +581,8 @@ void hci_disconnect(struct hci_conn *conn, __u8 reason); void hci_setup_sync(struct hci_conn *conn, __u16 handle); void hci_sco_setup(struct hci_conn *conn, __u8 status); -struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst); +struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, + __u16 pkt_type, bdaddr_t *dst); int hci_conn_del(struct hci_conn *conn); void hci_conn_hash_flush(struct hci_dev *hdev); void hci_conn_check_pending(struct hci_dev *hdev); @@ -591,7 +592,8 @@ void hci_chan_del(struct hci_chan *chan); void hci_chan_list_flush(struct hci_conn *conn); struct hci_chan *hci_chan_lookup_handle(struct hci_dev *hdev, __u16 handle); -struct hci_conn *hci_connect(struct hci_dev *hdev, int type, bdaddr_t *dst, +struct hci_conn *hci_connect(struct hci_dev *hdev, int type, + __u16 pkt_type, bdaddr_t *dst, __u8 dst_type, __u8 sec_level, __u8 auth_type); int hci_conn_check_link_mode(struct hci_conn *conn); int hci_conn_check_secure(struct hci_conn *conn, __u8 sec_level); @@ -654,7 +656,7 @@ static inline void hci_conn_drop(struct hci_conn *conn) if (conn->state == BT_CONNECTED) { timeo = conn->disc_timeout; if (!conn->out) - timeo *= 2; + timeo *= 20; } else { timeo = msecs_to_jiffies(10); } diff --git a/include/net/bluetooth/sco.h b/include/net/bluetooth/sco.h index 1e35c43657c8..6d1857ab8e5f 100644 --- a/include/net/bluetooth/sco.h +++ b/include/net/bluetooth/sco.h @@ -37,6 +37,7 @@ struct sockaddr_sco { sa_family_t sco_family; bdaddr_t sco_bdaddr; + __u16 sco_pkt_type; }; /* SCO socket options */ @@ -72,7 +73,8 @@ struct sco_conn { struct sco_pinfo { struct bt_sock bt; - __u32 flags; + __u16 pkt_type; + struct sco_conn *conn; }; diff --git a/include/net/cfg80211.h b/include/net/cfg80211.h index 26b5b692c22b..d9681a288ce6 100644 --- a/include/net/cfg80211.h +++ b/include/net/cfg80211.h @@ -2254,22 +2254,28 @@ struct cfg80211_ops { * enum wiphy_flags - wiphy capability flags * * @WIPHY_FLAG_CUSTOM_REGULATORY: tells us the driver for this device - * has its own custom regulatory domain and cannot identify the - * ISO / IEC 3166 alpha2 it belongs to. When this is enabled - * we will disregard the first regulatory hint (when the - * initiator is %REGDOM_SET_BY_CORE). - * @WIPHY_FLAG_STRICT_REGULATORY: tells us the driver for this device will - * ignore regulatory domain settings until it gets its own regulatory - * domain via its regulatory_hint() unless the regulatory hint is - * from a country IE. After its gets its own regulatory domain it will - * only allow further regulatory domain settings to further enhance - * compliance. For example if channel 13 and 14 are disabled by this - * regulatory domain no user regulatory domain can enable these channels - * at a later time. This can be used for devices which do not have - * calibration information guaranteed for frequencies or settings - * outside of its regulatory domain. If used in combination with - * WIPHY_FLAG_CUSTOM_REGULATORY the inspected country IE power settings - * will be followed. + * has its own custom regulatory domain and cannot identify the + * ISO / IEC 3166 alpha2 it belongs to. When this is enabled + * we will disregard the first regulatory hint (when the + * initiator is %REGDOM_SET_BY_CORE). wiphys can set the custom + * regulatory domain using wiphy_apply_custom_regulatory() + * prior to wiphy registration. + * @WIPHY_FLAG_STRICT_REGULATORY: tells us that the wiphy for this device + * has regulatory domain that it wishes to be considered as the + * superset for regulatory rules. After this device gets its regulatory + * domain programmed further regulatory hints shall only be considered + * for this device to enhance regulatory compliance, forcing the + * device to only possibly use subsets of the original regulatory + * rules. For example if channel 13 and 14 are disabled by this + * device's regulatory domain no user specified regulatory hint which + * has these channels enabled would enable them for this wiphy, + * the device's original regulatory domain will be trusted as the + * base. You can program the superset of regulatory rules for this + * wiphy with regulatory_hint() for cards programmed with an + * ISO3166-alpha2 country code. wiphys that use regulatory_hint() + * will have their wiphy->regd programmed once the regulatory + * domain is set, and all other regulatory hints will be ignored + * until their own regulatory domain gets programmed. * @WIPHY_FLAG_DISABLE_BEACON_HINTS: enable this if your driver needs to ensure * that passive scan flags and beaconing flags may not be lifted by * cfg80211 due to regulatory beacon hints. For more information on beacon @@ -2467,6 +2473,34 @@ struct wiphy_wowlan_support { }; /** + * enum wiphy_vendor_command_flags - validation flags for vendor commands + * @WIPHY_VENDOR_CMD_NEED_WDEV: vendor command requires wdev + * @WIPHY_VENDOR_CMD_NEED_NETDEV: vendor command requires netdev + * @WIPHY_VENDOR_CMD_NEED_RUNNING: interface/wdev must be up & running + * (must be combined with %_WDEV or %_NETDEV) + */ +enum wiphy_vendor_command_flags { + WIPHY_VENDOR_CMD_NEED_WDEV = BIT(0), + WIPHY_VENDOR_CMD_NEED_NETDEV = BIT(1), + WIPHY_VENDOR_CMD_NEED_RUNNING = BIT(2), +}; + +/** + * struct wiphy_vendor_command - vendor command definition + * @info: vendor command identifying information, as used in nl80211 + * @flags: flags, see &enum wiphy_vendor_command_flags + * @doit: callback for the operation, note that wdev is %NULL if the + * flags didn't ask for a wdev and non-%NULL otherwise; the data + * pointer may be %NULL if userspace provided no data at all + */ +struct wiphy_vendor_command { + struct nl80211_vendor_cmd_info info; + u32 flags; + int (*doit)(struct wiphy *wiphy, struct wireless_dev *wdev, + const void *data, int data_len); +}; + +/** * struct wiphy - wireless hardware description * @reg_notifier: the driver's regulatory notification callback, * note that if your driver uses wiphy_apply_custom_regulatory() @@ -2573,6 +2607,12 @@ struct wiphy_wowlan_support { * 802.11-2012 8.4.2.29 for the defined fields. * @extended_capabilities_mask: mask of the valid values * @extended_capabilities_len: length of the extended capabilities + * @country_ie_pref: country IE processing preferences specified + * by enum nl80211_country_ie_pref + * @vendor_commands: array of vendor commands supported by the hardware + * @n_vendor_commands: number of vendor commands + * @vendor_events: array of vendor events supported by the hardware + * @n_vendor_events: number of vendor events */ struct wiphy { /* assign these fields before you register the wiphy */ @@ -2642,6 +2682,8 @@ struct wiphy { const u8 *extended_capabilities, *extended_capabilities_mask; u8 extended_capabilities_len; + u8 country_ie_pref; + /* If multiple wiphys are registered and you're handed e.g. * a regular netdev with assigned ieee80211_ptr, you won't * know whether it points to a wiphy your driver has registered @@ -2681,6 +2723,10 @@ struct wiphy { const struct iw_handler_def *wext; #endif + const struct wiphy_vendor_command *vendor_commands; + const struct nl80211_vendor_cmd_info *vendor_events; + int n_vendor_commands, n_vendor_events; + char priv[0] __aligned(NETDEV_ALIGN); }; @@ -3591,6 +3637,121 @@ void wiphy_rfkill_start_polling(struct wiphy *wiphy); */ void wiphy_rfkill_stop_polling(struct wiphy *wiphy); +/** + * DOC: Vendor commands + * + * Occasionally, there are special protocol or firmware features that + * can't be implemented very openly. For this and similar cases, the + * vendor command functionality allows implementing the features with + * (typically closed-source) userspace and firmware, using nl80211 as + * the configuration mechanism. + * + * A driver supporting vendor commands must register them as an array + * in struct wiphy, with handlers for each one, each command has an + * OUI and sub command ID to identify it. + * + * Note that this feature should not be (ab)used to implement protocol + * features that could openly be shared across drivers. In particular, + * it must never be required to use vendor commands to implement any + * "normal" functionality that higher-level userspace like connection + * managers etc. need. + */ + +struct sk_buff *__cfg80211_alloc_reply_skb(struct wiphy *wiphy, + enum nl80211_commands cmd, + enum nl80211_attrs attr, + int approxlen); + +struct sk_buff *__cfg80211_alloc_event_skb(struct wiphy *wiphy, + enum nl80211_commands cmd, + enum nl80211_attrs attr, + int vendor_event_idx, + int approxlen, gfp_t gfp); + +void __cfg80211_send_event_skb(struct sk_buff *skb, gfp_t gfp); + +/** + * cfg80211_vendor_cmd_alloc_reply_skb - allocate vendor command reply + * @wiphy: the wiphy + * @approxlen: an upper bound of the length of the data that will + * be put into the skb + * + * This function allocates and pre-fills an skb for a reply to + * a vendor command. Since it is intended for a reply, calling + * it outside of a vendor command's doit() operation is invalid. + * + * The returned skb is pre-filled with some identifying data in + * a way that any data that is put into the skb (with skb_put(), + * nla_put() or similar) will end up being within the + * %NL80211_ATTR_VENDOR_DATA attribute, so all that needs to be done + * with the skb is adding data for the corresponding userspace tool + * which can then read that data out of the testdata attribute. You + * must not modify the skb in any other way. + * + * When done, call cfg80211_vendor_cmd_reply() with the skb and return + * its error code as the result of the doit() operation. + * + * Return: An allocated and pre-filled skb. %NULL if any errors happen. + */ +static inline struct sk_buff * +cfg80211_vendor_cmd_alloc_reply_skb(struct wiphy *wiphy, int approxlen) +{ + return __cfg80211_alloc_reply_skb(wiphy, NL80211_CMD_VENDOR, + NL80211_ATTR_VENDOR_DATA, approxlen); +} + +/** + * cfg80211_vendor_cmd_reply - send the reply skb + * @skb: The skb, must have been allocated with + * cfg80211_vendor_cmd_alloc_reply_skb() + * + * Since calling this function will usually be the last thing + * before returning from the vendor command doit() you should + * return the error code. Note that this function consumes the + * skb regardless of the return value. + * + * Return: An error code or 0 on success. + */ +int cfg80211_vendor_cmd_reply(struct sk_buff *skb); + +/** + * cfg80211_vendor_event_alloc - allocate vendor-specific event skb + * @wiphy: the wiphy + * @event_idx: index of the vendor event in the wiphy's vendor_events + * @approxlen: an upper bound of the length of the data that will + * be put into the skb + * @gfp: allocation flags + * + * This function allocates and pre-fills an skb for an event on the + * vendor-specific multicast group. + * + * When done filling the skb, call cfg80211_vendor_event() with the + * skb to send the event. + * + * Return: An allocated and pre-filled skb. %NULL if any errors happen. + */ +static inline struct sk_buff * +cfg80211_vendor_event_alloc(struct wiphy *wiphy, int approxlen, + int event_idx, gfp_t gfp) +{ + return __cfg80211_alloc_event_skb(wiphy, NL80211_CMD_VENDOR, + NL80211_ATTR_VENDOR_DATA, + event_idx, approxlen, gfp); +} + +/** + * cfg80211_vendor_event - send the event + * @skb: The skb, must have been allocated with cfg80211_vendor_event_alloc() + * @gfp: allocation flags + * + * This function sends the given @skb, which must have been allocated + * by cfg80211_vendor_event_alloc(), as an event. It always consumes it. + */ +static inline void cfg80211_vendor_event(struct sk_buff *skb, gfp_t gfp) +{ + __cfg80211_send_event_skb(skb, gfp); +} + #ifdef CONFIG_NL80211_TESTMODE /** * DOC: Test mode @@ -3626,8 +3787,12 @@ void wiphy_rfkill_stop_polling(struct wiphy *wiphy); * * Return: An allocated and pre-filled skb. %NULL if any errors happen. */ -struct sk_buff *cfg80211_testmode_alloc_reply_skb(struct wiphy *wiphy, - int approxlen); +static inline struct sk_buff * +cfg80211_testmode_alloc_reply_skb(struct wiphy *wiphy, int approxlen) +{ + return __cfg80211_alloc_reply_skb(wiphy, NL80211_CMD_TESTMODE, + NL80211_ATTR_TESTDATA, approxlen); +} /** * cfg80211_testmode_reply - send the reply skb @@ -3641,7 +3806,10 @@ struct sk_buff *cfg80211_testmode_alloc_reply_skb(struct wiphy *wiphy, * * Return: An error code or 0 on success. */ -int cfg80211_testmode_reply(struct sk_buff *skb); +static inline int cfg80211_testmode_reply(struct sk_buff *skb) +{ + return cfg80211_vendor_cmd_reply(skb); +} /** * cfg80211_testmode_alloc_event_skb - allocate testmode event @@ -3664,8 +3832,13 @@ int cfg80211_testmode_reply(struct sk_buff *skb); * * Return: An allocated and pre-filled skb. %NULL if any errors happen. */ -struct sk_buff *cfg80211_testmode_alloc_event_skb(struct wiphy *wiphy, - int approxlen, gfp_t gfp); +static inline struct sk_buff * +cfg80211_testmode_alloc_event_skb(struct wiphy *wiphy, int approxlen, gfp_t gfp) +{ + return __cfg80211_alloc_event_skb(wiphy, NL80211_CMD_TESTMODE, + NL80211_ATTR_TESTDATA, -1, + approxlen, gfp); +} /** * cfg80211_testmode_event - send the event @@ -3677,7 +3850,10 @@ struct sk_buff *cfg80211_testmode_alloc_event_skb(struct wiphy *wiphy, * by cfg80211_testmode_alloc_event_skb(), as an event. It always * consumes it. */ -void cfg80211_testmode_event(struct sk_buff *skb, gfp_t gfp); +static inline void cfg80211_testmode_event(struct sk_buff *skb, gfp_t gfp) +{ + __cfg80211_send_event_skb(skb, gfp); +} #define CFG80211_TESTMODE_CMD(cmd) .testmode_cmd = (cmd), #define CFG80211_TESTMODE_DUMP(cmd) .testmode_dump = (cmd), diff --git a/include/net/fib_rules.h b/include/net/fib_rules.h index e361f4882426..4ac12e14c6d9 100644 --- a/include/net/fib_rules.h +++ b/include/net/fib_rules.h @@ -23,6 +23,8 @@ struct fib_rule { struct fib_rule __rcu *ctarget; char iifname[IFNAMSIZ]; char oifname[IFNAMSIZ]; + kuid_t uid_start; + kuid_t uid_end; struct rcu_head rcu; struct net * fr_net; }; @@ -80,7 +82,9 @@ struct fib_rules_ops { [FRA_FWMARK] = { .type = NLA_U32 }, \ [FRA_FWMASK] = { .type = NLA_U32 }, \ [FRA_TABLE] = { .type = NLA_U32 }, \ - [FRA_GOTO] = { .type = NLA_U32 } + [FRA_GOTO] = { .type = NLA_U32 }, \ + [FRA_UID_START] = { .type = NLA_U32 }, \ + [FRA_UID_END] = { .type = NLA_U32 } static inline void fib_rule_get(struct fib_rule *rule) { diff --git a/include/net/flow.h b/include/net/flow.h index 628e11b98c58..1426681f7cf3 100644 --- a/include/net/flow.h +++ b/include/net/flow.h @@ -10,6 +10,15 @@ #include <linux/socket.h> #include <linux/in6.h> #include <linux/atomic.h> +#include <linux/uidgid.h> + +/* + * ifindex generation is per-net namespace, and loopback is + * always the 1st device in ns (see net_dev_init), thus any + * loopback device should get ifindex 1 + */ + +#define LOOPBACK_IFINDEX 1 struct flowi_common { int flowic_oif; @@ -23,6 +32,7 @@ struct flowi_common { #define FLOWI_FLAG_CAN_SLEEP 0x02 #define FLOWI_FLAG_KNOWN_NH 0x04 __u32 flowic_secid; + kuid_t flowic_uid; }; union flowi_uli { @@ -59,6 +69,7 @@ struct flowi4 { #define flowi4_proto __fl_common.flowic_proto #define flowi4_flags __fl_common.flowic_flags #define flowi4_secid __fl_common.flowic_secid +#define flowi4_uid __fl_common.flowic_uid /* (saddr,daddr) must be grouped, same order as in IP header */ __be32 saddr; @@ -78,16 +89,18 @@ 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 = 0; + fl4->flowi4_iif = LOOPBACK_IFINDEX; fl4->flowi4_mark = mark; fl4->flowi4_tos = tos; fl4->flowi4_scope = scope; fl4->flowi4_proto = proto; fl4->flowi4_flags = flags; fl4->flowi4_secid = 0; + fl4->flowi4_uid = uid; fl4->daddr = daddr; fl4->saddr = saddr; fl4->fl4_dport = dport; @@ -115,6 +128,7 @@ struct flowi6 { #define flowi6_proto __fl_common.flowic_proto #define flowi6_flags __fl_common.flowic_flags #define flowi6_secid __fl_common.flowic_secid +#define flowi6_uid __fl_common.flowic_uid struct in6_addr daddr; struct in6_addr saddr; __be32 flowlabel; @@ -158,6 +172,7 @@ struct flowi { #define flowi_proto u.__fl_common.flowic_proto #define flowi_flags u.__fl_common.flowic_flags #define flowi_secid u.__fl_common.flowic_secid +#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/inet_sock.h b/include/net/inet_sock.h index 7235ae73a1e8..9528e10fa0b4 100644 --- a/include/net/inet_sock.h +++ b/include/net/inet_sock.h @@ -88,6 +88,7 @@ struct inet_request_sock { acked : 1, no_srccheck: 1; kmemcheck_bitfield_end(flags); + u32 ir_mark; struct ip_options_rcu *opt; }; @@ -96,6 +97,14 @@ static inline struct inet_request_sock *inet_rsk(const struct request_sock *sk) return (struct inet_request_sock *)sk; } +static inline u32 inet_request_mark(struct sock *sk, struct sk_buff *skb) +{ + if (!sk->sk_mark && sock_net(sk)->ipv4.sysctl_tcp_fwmark_accept) + return skb->mark; + + return sk->sk_mark; +} + struct inet_cork { unsigned int flags; __be32 addr; diff --git a/include/net/ip.h b/include/net/ip.h index 0a62365149e2..fc62ae0a47d2 100644 --- a/include/net/ip.h +++ b/include/net/ip.h @@ -154,6 +154,7 @@ struct ip_reply_arg { /* -1 if not needed */ int bound_dev_if; u8 tos; + kuid_t uid; }; #define IP_REPLY_ARG_NOSRCCHECK 1 @@ -226,6 +227,9 @@ extern void ipfrag_init(void); extern void ip_static_sysctl_init(void); +#define IP4_REPLY_MARK(net, mark) \ + ((net)->ipv4.sysctl_fwmark_reflect ? (mark) : 0) + static inline bool ip_is_fragment(const struct iphdr *iph) { return (iph->frag_off & htons(IP_MF | IP_OFFSET)) != 0; diff --git a/include/net/ip6_route.h b/include/net/ip6_route.h index 8d977b343647..6be6debb5361 100644 --- a/include/net/ip6_route.h +++ b/include/net/ip6_route.h @@ -136,7 +136,7 @@ extern int rt6_route_rcv(struct net_device *dev, const struct in6_addr *gwaddr); extern void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu, - int oif, u32 mark); + int oif, u32 mark, kuid_t uid); extern void ip6_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, __be32 mtu); extern void ip6_redirect(struct sk_buff *skb, struct net *net, int oif, u32 mark); diff --git a/include/net/ipv6.h b/include/net/ipv6.h index 087370ff05f1..ac1d532965a2 100644 --- a/include/net/ipv6.h +++ b/include/net/ipv6.h @@ -111,6 +111,9 @@ struct frag_hdr { #define IP6_MF 0x0001 +#define IP6_REPLY_MARK(net, mark) \ + ((net)->ipv6.sysctl.fwmark_reflect ? (mark) : 0) + #include <net/sock.h> /* sysctls */ @@ -260,6 +263,12 @@ static inline void fl6_sock_release(struct ip6_flowlabel *fl) extern void icmpv6_notify(struct sk_buff *skb, u8 type, u8 code, __be32 info); +int icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, + struct icmp6hdr *thdr, int len); + +struct dst_entry *icmpv6_route_lookup(struct net *net, struct sk_buff *skb, + struct sock *sk, struct flowi6 *fl6); + extern int ip6_ra_control(struct sock *sk, int sel); extern int ipv6_parse_hopopts(struct sk_buff *skb); @@ -796,8 +805,7 @@ extern int compat_ipv6_getsockopt(struct sock *sk, extern int ip6_datagram_connect(struct sock *sk, struct sockaddr *addr, int addr_len); -extern int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len, - int *addr_len); +extern int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len); extern int ipv6_recv_rxpmtu(struct sock *sk, struct msghdr *msg, int len, int *addr_len); extern void ipv6_icmp_error(struct sock *sk, struct sk_buff *skb, int err, __be16 port, diff --git a/include/net/net_namespace.h b/include/net/net_namespace.h index b17697827482..b064d6dd14fb 100644 --- a/include/net/net_namespace.h +++ b/include/net/net_namespace.h @@ -9,6 +9,7 @@ #include <linux/list.h> #include <linux/sysctl.h> +#include <net/flow.h> #include <net/netns/core.h> #include <net/netns/mib.h> #include <net/netns/unix.h> @@ -120,14 +121,6 @@ struct net { atomic_t rt_genid; }; -/* - * ifindex generation is per-net namespace, and loopback is - * always the 1st device in ns (see net_dev_init), thus any - * loopback device should get ifindex 1 - */ - -#define LOOPBACK_IFINDEX 1 - #include <linux/seq_file_net.h> /* Init's network namespace */ diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h index 03e6378d5353..4f6c72095cf6 100644 --- a/include/net/netns/ipv4.h +++ b/include/net/netns/ipv4.h @@ -65,6 +65,9 @@ struct netns_ipv4 { int sysctl_tcp_ecn; + int sysctl_fwmark_reflect; + int sysctl_tcp_fwmark_accept; + kgid_t sysctl_ping_group_range[2]; long sysctl_tcp_mem[3]; diff --git a/include/net/netns/ipv6.h b/include/net/netns/ipv6.h index 005e2c2e39a9..4b9f99e3a91c 100644 --- a/include/net/netns/ipv6.h +++ b/include/net/netns/ipv6.h @@ -28,6 +28,7 @@ struct netns_sysctl_ipv6 { int ip6_rt_mtu_expires; int ip6_rt_min_advmss; int icmpv6_time; + int fwmark_reflect; }; struct netns_ipv6 { diff --git a/include/net/ping.h b/include/net/ping.h index 682b5ae9af51..2db4860e5848 100644 --- a/include/net/ping.h +++ b/include/net/ping.h @@ -13,6 +13,7 @@ #ifndef _PING_H #define _PING_H +#include <net/icmp.h> #include <net/netns/hash.h> /* PING_HTABLE_SIZE must be power of 2 */ @@ -28,6 +29,18 @@ */ #define GID_T_MAX (((gid_t)~0U) >> 1) +/* Compatibility glue so we can support IPv6 when it's compiled as a module */ +struct pingv6_ops { + int (*ipv6_recv_error)(struct sock *sk, struct msghdr *msg, int len); + int (*ip6_datagram_recv_ctl)(struct sock *sk, struct msghdr *msg, + struct sk_buff *skb); + int (*icmpv6_err_convert)(u8 type, u8 code, int *err); + void (*ipv6_icmp_error)(struct sock *sk, struct sk_buff *skb, int err, + __be16 port, u32 info, u8 *payload); + int (*ipv6_chk_addr)(struct net *net, const struct in6_addr *addr, + const struct net_device *dev, int strict); +}; + struct ping_table { struct hlist_nulls_head hash[PING_HTABLE_SIZE]; rwlock_t lock; @@ -39,10 +52,39 @@ struct ping_iter_state { }; extern struct proto ping_prot; +extern struct ping_table ping_table; +#if IS_ENABLED(CONFIG_IPV6) +extern struct pingv6_ops pingv6_ops; +#endif +struct pingfakehdr { + struct icmphdr icmph; + struct iovec *iov; + sa_family_t family; + __wsum wcheck; +}; -extern void ping_rcv(struct sk_buff *); -extern void ping_err(struct sk_buff *, u32 info); +int ping_get_port(struct sock *sk, unsigned short ident); +void ping_hash(struct sock *sk); +void ping_unhash(struct sock *sk); + +int ping_init_sock(struct sock *sk); +void ping_close(struct sock *sk, long timeout); +int ping_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len); +void ping_err(struct sk_buff *skb, int offset, u32 info); +int ping_getfrag(void *from, char *to, int offset, int fraglen, int odd, + struct sk_buff *); + +int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int noblock, int flags, int *addr_len); +int ping_common_sendmsg(int family, struct msghdr *msg, size_t len, + void *user_icmph, size_t icmph_len); +int ping_v4_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len); +int ping_v6_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len); +int ping_queue_rcv_skb(struct sock *sk, struct sk_buff *skb); +void ping_rcv(struct sk_buff *skb); #ifdef CONFIG_PROC_FS extern int __init ping_proc_init(void); @@ -50,6 +92,7 @@ extern void ping_proc_exit(void); #endif void __init ping_init(void); - +int __init pingv6_init(void); +void pingv6_exit(void); #endif /* _PING_H */ diff --git a/include/net/route.h b/include/net/route.h index 2ea40c1b5e00..647bb2adbffd 100644 --- a/include/net/route.h +++ b/include/net/route.h @@ -142,7 +142,7 @@ 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, sk ? sock_i_uid(sk) : 0); if (sk) security_sk_classify_flow(sk, flowi4_to_flowi(fl4)); return ip_route_output_flow(net, fl4, sk); @@ -253,7 +253,8 @@ static inline void ip_route_connect_init(struct flowi4 *fl4, __be32 dst, __be32 flow_flags |= FLOWI_FLAG_CAN_SLEEP; 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, + sock_i_uid(sk)); } static inline struct rtable *ip_route_connect(struct flowi4 *fl4, diff --git a/include/net/tcp.h b/include/net/tcp.h index 29a1a63cd303..e0fc21357582 100644 --- a/include/net/tcp.h +++ b/include/net/tcp.h @@ -288,6 +288,7 @@ extern int sysctl_tcp_early_retrans; extern int sysctl_tcp_limit_output_bytes; extern int sysctl_tcp_challenge_ack_limit; extern int sysctl_tcp_min_tso_segs; +extern int sysctl_tcp_default_init_rwnd; extern atomic_long_t tcp_memory_allocated; extern struct percpu_counter tcp_sockets_allocated; @@ -1553,6 +1554,8 @@ extern struct sk_buff **tcp4_gro_receive(struct sk_buff **head, extern int tcp_gro_complete(struct sk_buff *skb); extern int tcp4_gro_complete(struct sk_buff *skb); +extern int tcp_nuke_addr(struct net *net, struct sockaddr *addr); + #ifdef CONFIG_PROC_FS extern int tcp4_proc_init(void); extern void tcp4_proc_exit(void); diff --git a/include/net/transp_v6.h b/include/net/transp_v6.h index 938b7fd11204..eb40e71ff2ee 100644 --- a/include/net/transp_v6.h +++ b/include/net/transp_v6.h @@ -11,6 +11,7 @@ extern struct proto rawv6_prot; extern struct proto udpv6_prot; extern struct proto udplitev6_prot; extern struct proto tcpv6_prot; +extern struct proto pingv6_prot; struct flowi6; @@ -21,6 +22,8 @@ extern int ipv6_frag_init(void); extern void ipv6_frag_exit(void); /* transport protocols */ +extern int pingv6_init(void); +extern void pingv6_exit(void); extern int rawv6_init(void); extern void rawv6_exit(void); extern int udpv6_init(void); 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/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/power.h b/include/trace/events/power.h index 427acab5d69a..503fc2b870b9 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -146,6 +146,25 @@ 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/uapi/asm-generic/unistd.h b/include/uapi/asm-generic/unistd.h index 0cc74c4403e4..b422ad5d238b 100644 --- a/include/uapi/asm-generic/unistd.h +++ b/include/uapi/asm-generic/unistd.h @@ -692,9 +692,19 @@ __SC_COMP(__NR_process_vm_writev, sys_process_vm_writev, \ __SYSCALL(__NR_kcmp, sys_kcmp) #define __NR_finit_module 273 __SYSCALL(__NR_finit_module, sys_finit_module) +/* Backporting seccomp, skip a few ... + * #define __NR_sched_setattr 274 +__SYSCALL(__NR_sched_setattr, sys_sched_setattr) + * #define __NR_sched_getattr 275 +__SYSCALL(__NR_sched_getattr, sys_sched_getattr) + * #define __NR_renameat2 276 +__SYSCALL(__NR_renameat2, sys_renameat2) + */ +#define __NR_seccomp 277 +__SYSCALL(__NR_seccomp, sys_seccomp) #undef __NR_syscalls -#define __NR_syscalls 274 +#define __NR_syscalls 278 /* * All syscalls below here should go away really, diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild index 405887bec8b3..c5268778832e 100644 --- a/include/uapi/linux/Kbuild +++ b/include/uapi/linux/Kbuild @@ -1,4 +1,5 @@ # UAPI Header export list +header-y += android/ header-y += byteorder/ header-y += can/ header-y += caif/ diff --git a/include/uapi/linux/android/Kbuild b/include/uapi/linux/android/Kbuild new file mode 100644 index 000000000000..ca011eec252a --- /dev/null +++ b/include/uapi/linux/android/Kbuild @@ -0,0 +1,2 @@ +# UAPI Header export list +header-y += binder.h diff --git a/drivers/staging/android/binder.h b/include/uapi/linux/android/binder.h index dbe81ceca1bd..c653800b3a1a 100644 --- a/drivers/staging/android/binder.h +++ b/include/uapi/linux/android/binder.h @@ -17,8 +17,8 @@ * */ -#ifndef _LINUX_BINDER_H -#define _LINUX_BINDER_H +#ifndef _UAPI_LINUX_BINDER_H +#define _UAPI_LINUX_BINDER_H #include <linux/ioctl.h> @@ -39,6 +39,14 @@ enum { FLAT_BINDER_FLAG_ACCEPTS_FDS = 0x100, }; +#ifdef BINDER_IPC_32BIT +typedef __u32 binder_size_t; +typedef __u32 binder_uintptr_t; +#else +typedef __u64 binder_size_t; +typedef __u64 binder_uintptr_t; +#endif + /* * This is the flattened representation of a Binder object for transfer * between processes. The 'offsets' supplied as part of a binder transaction @@ -48,17 +56,17 @@ enum { */ struct flat_binder_object { /* 8 bytes for large_flat_header. */ - unsigned long type; - unsigned long flags; + __u32 type; + __u32 flags; /* 8 bytes of data. */ union { - void __user *binder; /* local object */ - signed long handle; /* remote object */ + binder_uintptr_t binder; /* local object */ + __u32 handle; /* remote object */ }; /* extra data associated with local object */ - void __user *cookie; + binder_uintptr_t cookie; }; /* @@ -67,26 +75,30 @@ struct flat_binder_object { */ struct binder_write_read { - signed long write_size; /* bytes to write */ - signed long write_consumed; /* bytes consumed by driver */ - unsigned long write_buffer; - signed long read_size; /* bytes to read */ - signed long read_consumed; /* bytes consumed by driver */ - unsigned long read_buffer; + binder_size_t write_size; /* bytes to write */ + binder_size_t write_consumed; /* bytes consumed by driver */ + binder_uintptr_t write_buffer; + binder_size_t read_size; /* bytes to read */ + binder_size_t read_consumed; /* bytes consumed by driver */ + binder_uintptr_t read_buffer; }; /* Use with BINDER_VERSION, driver fills in fields. */ struct binder_version { /* driver protocol version -- increment with incompatible change */ - signed long protocol_version; + __s32 protocol_version; }; /* This is the current protocol version. */ +#ifdef BINDER_IPC_32BIT #define BINDER_CURRENT_PROTOCOL_VERSION 7 +#else +#define BINDER_CURRENT_PROTOCOL_VERSION 8 +#endif #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, size_t) +#define BINDER_SET_MAX_THREADS _IOW('b', 5, __u32) #define BINDER_SET_IDLE_PRIORITY _IOW('b', 6, __s32) #define BINDER_SET_CONTEXT_MGR _IOW('b', 7, __s32) #define BINDER_THREAD_EXIT _IOW('b', 8, __s32) @@ -119,18 +131,20 @@ struct binder_transaction_data { * identifying the target and contents of the transaction. */ union { - size_t handle; /* target descriptor of command transaction */ - void *ptr; /* target descriptor of return transaction */ + /* target descriptor of command transaction */ + __u32 handle; + /* target descriptor of return transaction */ + binder_uintptr_t ptr; } target; - void *cookie; /* target object cookie */ - unsigned int code; /* transaction command */ + binder_uintptr_t cookie; /* target object cookie */ + __u32 code; /* transaction command */ /* General information about the transaction. */ - unsigned int flags; + __u32 flags; pid_t sender_pid; uid_t sender_euid; - size_t data_size; /* number of bytes of data */ - size_t offsets_size; /* number of bytes of offsets */ + binder_size_t data_size; /* number of bytes of data */ + binder_size_t offsets_size; /* number of bytes of offsets */ /* If this transaction is inline, the data immediately * follows here; otherwise, it ends with a pointer to @@ -139,32 +153,37 @@ struct binder_transaction_data { union { struct { /* transaction data */ - const void __user *buffer; + binder_uintptr_t buffer; /* offsets from buffer to flat_binder_object structs */ - const void __user *offsets; + binder_uintptr_t offsets; } ptr; - uint8_t buf[8]; + __u8 buf[8]; } data; }; struct binder_ptr_cookie { - void *ptr; - void *cookie; + binder_uintptr_t ptr; + binder_uintptr_t cookie; }; +struct binder_handle_cookie { + __u32 handle; + binder_uintptr_t cookie; +} __packed; + struct binder_pri_desc { - int priority; - int desc; + __s32 priority; + __u32 desc; }; struct binder_pri_ptr_cookie { - int priority; - void *ptr; - void *cookie; + __s32 priority; + binder_uintptr_t ptr; + binder_uintptr_t cookie; }; enum binder_driver_return_protocol { - BR_ERROR = _IOR('r', 0, int), + BR_ERROR = _IOR('r', 0, __s32), /* * int: error code */ @@ -178,7 +197,7 @@ enum binder_driver_return_protocol { * binder_transaction_data: the received command. */ - BR_ACQUIRE_RESULT = _IOR('r', 4, int), + BR_ACQUIRE_RESULT = _IOR('r', 4, __s32), /* * not currently supported * int: 0 if the last bcATTEMPT_ACQUIRE was not successful. @@ -235,11 +254,11 @@ enum binder_driver_return_protocol { * stop threadpool thread */ - BR_DEAD_BINDER = _IOR('r', 15, void *), + BR_DEAD_BINDER = _IOR('r', 15, binder_uintptr_t), /* * void *: cookie */ - BR_CLEAR_DEATH_NOTIFICATION_DONE = _IOR('r', 16, void *), + BR_CLEAR_DEATH_NOTIFICATION_DONE = _IOR('r', 16, binder_uintptr_t), /* * void *: cookie */ @@ -258,22 +277,22 @@ enum binder_driver_command_protocol { * binder_transaction_data: the sent command. */ - BC_ACQUIRE_RESULT = _IOW('c', 2, int), + BC_ACQUIRE_RESULT = _IOW('c', 2, __s32), /* * not currently supported * int: 0 if the last BR_ATTEMPT_ACQUIRE was not successful. * Else you have acquired a primary reference on the object. */ - BC_FREE_BUFFER = _IOW('c', 3, int), + BC_FREE_BUFFER = _IOW('c', 3, binder_uintptr_t), /* * void *: ptr to transaction data received on a read */ - BC_INCREFS = _IOW('c', 4, int), - BC_ACQUIRE = _IOW('c', 5, int), - BC_RELEASE = _IOW('c', 6, int), - BC_DECREFS = _IOW('c', 7, int), + BC_INCREFS = _IOW('c', 4, __u32), + BC_ACQUIRE = _IOW('c', 5, __u32), + BC_RELEASE = _IOW('c', 6, __u32), + BC_DECREFS = _IOW('c', 7, __u32), /* * int: descriptor */ @@ -308,23 +327,23 @@ enum binder_driver_command_protocol { * of looping threads it has available. */ - BC_REQUEST_DEATH_NOTIFICATION = _IOW('c', 14, struct binder_ptr_cookie), + BC_REQUEST_DEATH_NOTIFICATION = _IOW('c', 14, struct binder_handle_cookie), /* - * void *: ptr to binder + * int: handle * void *: cookie */ - BC_CLEAR_DEATH_NOTIFICATION = _IOW('c', 15, struct binder_ptr_cookie), + BC_CLEAR_DEATH_NOTIFICATION = _IOW('c', 15, struct binder_handle_cookie), /* - * void *: ptr to binder + * int: handle * void *: cookie */ - BC_DEAD_BINDER_DONE = _IOW('c', 16, void *), + BC_DEAD_BINDER_DONE = _IOW('c', 16, binder_uintptr_t), /* * void *: cookie */ }; -#endif /* _LINUX_BINDER_H */ +#endif /* _UAPI_LINUX_BINDER_H */ diff --git a/include/uapi/linux/audit.h b/include/uapi/linux/audit.h index 75cef3fd97ad..ce8750f8788a 100644 --- a/include/uapi/linux/audit.h +++ b/include/uapi/linux/audit.h @@ -324,6 +324,8 @@ enum { /* distinguish syscall tables */ #define __AUDIT_ARCH_64BIT 0x80000000 #define __AUDIT_ARCH_LE 0x40000000 + +#define AUDIT_ARCH_AARCH64 (EM_AARCH64|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) #define AUDIT_ARCH_ALPHA (EM_ALPHA|__AUDIT_ARCH_64BIT|__AUDIT_ARCH_LE) #define AUDIT_ARCH_ARM (EM_ARM|__AUDIT_ARCH_LE) #define AUDIT_ARCH_ARMEB (EM_ARM) diff --git a/include/uapi/linux/eventpoll.h b/include/uapi/linux/eventpoll.h index 2c267bcbb85c..bc81fb2e1f0e 100644 --- a/include/uapi/linux/eventpoll.h +++ b/include/uapi/linux/eventpoll.h @@ -61,5 +61,16 @@ struct epoll_event { __u64 data; } EPOLL_PACKED; - +#ifdef CONFIG_PM_SLEEP +static inline void ep_take_care_of_epollwakeup(struct epoll_event *epev) +{ + if ((epev->events & EPOLLWAKEUP) && !capable(CAP_BLOCK_SUSPEND)) + epev->events &= ~EPOLLWAKEUP; +} +#else +static inline void ep_take_care_of_epollwakeup(struct epoll_event *epev) +{ + epev->events &= ~EPOLLWAKEUP; +} +#endif #endif /* _UAPI_LINUX_EVENTPOLL_H */ diff --git a/include/uapi/linux/fib_rules.h b/include/uapi/linux/fib_rules.h index 51da65b68b85..9dcdb6251cb8 100644 --- a/include/uapi/linux/fib_rules.h +++ b/include/uapi/linux/fib_rules.h @@ -49,6 +49,8 @@ enum { FRA_TABLE, /* Extended table id */ FRA_FWMASK, /* mask for netfilter mark */ FRA_OIFNAME, + FRA_UID_START, /* UID range */ + FRA_UID_END, __FRA_MAX }; diff --git a/include/uapi/linux/fs.h b/include/uapi/linux/fs.h index a4ed56cf0eac..5014a5c472ed 100644 --- a/include/uapi/linux/fs.h +++ b/include/uapi/linux/fs.h @@ -154,6 +154,8 @@ 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) 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 e36a4aecd311..87f478b0ca6f 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/input.h b/include/uapi/linux/input.h index 4649ee35b605..a86f5301c609 100644 --- a/include/uapi/linux/input.h +++ b/include/uapi/linux/input.h @@ -153,6 +153,9 @@ struct input_keymap_entry { #define EVIOCGRAB _IOW('E', 0x90, int) /* Grab/Release device */ +#define EVIOCGSUSPENDBLOCK _IOR('E', 0x91, int) /* get suspend block enable */ +#define EVIOCSSUSPENDBLOCK _IOW('E', 0x91, int) /* set suspend block enable */ + #define EVIOCSCLOCKID _IOW('E', 0xa0, int) /* Set clockid to be used for timestamps */ /* @@ -458,10 +461,14 @@ struct input_keymap_entry { #define KEY_VIDEO_NEXT 241 /* drive next video source */ #define KEY_VIDEO_PREV 242 /* drive previous video source */ #define KEY_BRIGHTNESS_CYCLE 243 /* brightness up, after max is min */ -#define KEY_BRIGHTNESS_ZERO 244 /* brightness off, use ambient */ +#define KEY_BRIGHTNESS_AUTO 244 /* Set Auto Brightness: manual + brightness control is off, + rely on ambient */ +#define KEY_BRIGHTNESS_ZERO KEY_BRIGHTNESS_AUTO #define KEY_DISPLAY_OFF 245 /* display device to off state */ -#define KEY_WIMAX 246 +#define KEY_WWAN 246 /* Wireless WAN (LTE, UMTS, GSM, etc.) */ +#define KEY_WIMAX KEY_WWAN #define KEY_RFKILL 247 /* Key that controls all radios */ #define KEY_MICMUTE 248 /* Mute / unmute the microphone */ @@ -506,11 +513,15 @@ struct input_keymap_entry { #define BTN_DEAD 0x12f #define BTN_GAMEPAD 0x130 -#define BTN_A 0x130 -#define BTN_B 0x131 +#define BTN_SOUTH 0x130 +#define BTN_A BTN_SOUTH +#define BTN_EAST 0x131 +#define BTN_B BTN_EAST #define BTN_C 0x132 -#define BTN_X 0x133 -#define BTN_Y 0x134 +#define BTN_NORTH 0x133 +#define BTN_X BTN_NORTH +#define BTN_WEST 0x134 +#define BTN_Y BTN_WEST #define BTN_Z 0x135 #define BTN_TL 0x136 #define BTN_TR 0x137 @@ -623,6 +634,7 @@ struct input_keymap_entry { #define KEY_ADDRESSBOOK 0x1ad /* AL Contacts/Address Book */ #define KEY_MESSENGER 0x1ae /* AL Instant Messaging */ #define KEY_DISPLAYTOGGLE 0x1af /* Turn display (LCD) on and off */ +#define KEY_BRIGHTNESS_TOGGLE KEY_DISPLAYTOGGLE #define KEY_SPELLCHECK 0x1b0 /* AL Spell Check */ #define KEY_LOGOFF 0x1b1 /* AL Logoff */ @@ -707,6 +719,24 @@ struct input_keymap_entry { #define KEY_ATTENDANT_TOGGLE 0x21d /* Attendant call on or off */ #define KEY_LIGHTS_TOGGLE 0x21e /* Reading light on or off */ +#define BTN_DPAD_UP 0x220 +#define BTN_DPAD_DOWN 0x221 +#define BTN_DPAD_LEFT 0x222 +#define BTN_DPAD_RIGHT 0x223 + +#define KEY_ALS_TOGGLE 0x230 /* Ambient light sensor */ + +#define KEY_BUTTONCONFIG 0x240 /* AL Button Configuration */ +#define KEY_TASKMANAGER 0x241 /* AL Task/Project Manager */ +#define KEY_JOURNAL 0x242 /* AL Log/Journal/Timecard */ +#define KEY_CONTROLPANEL 0x243 /* AL Control Panel */ +#define KEY_APPSELECT 0x244 /* AL Select Task/Application */ +#define KEY_SCREENSAVER 0x245 /* AL Screen Saver */ +#define KEY_VOICECOMMAND 0x246 /* Listening Voice Command */ + +#define KEY_BRIGHTNESS_MIN 0x250 /* Set Brightness to Minimum */ +#define KEY_BRIGHTNESS_MAX 0x251 /* Set Brightness to Maximum */ + #define BTN_TRIGGER_HAPPY 0x2c0 #define BTN_TRIGGER_HAPPY1 0x2c0 #define BTN_TRIGGER_HAPPY2 0x2c1 @@ -844,6 +874,7 @@ struct input_keymap_entry { #define SW_FRONT_PROXIMITY 0x0b /* set = front proximity sensor active */ #define SW_ROTATE_LOCK 0x0c /* set = rotate locked/disabled */ #define SW_LINEIN_INSERT 0x0d /* set = inserted */ +#define SW_MUTE_DEVICE 0x0e /* set = device disabled */ #define SW_MAX 0x0f #define SW_CNT (SW_MAX+1) diff --git a/include/uapi/linux/ipv6.h b/include/uapi/linux/ipv6.h index 4bda4cf5b0f5..e9d0f7efde3b 100644 --- a/include/uapi/linux/ipv6.h +++ b/include/uapi/linux/ipv6.h @@ -160,6 +160,8 @@ enum { DEVCONF_ACCEPT_DAD, DEVCONF_FORCE_TLLAO, DEVCONF_NDISC_NOTIFY, + DEVCONF_ACCEPT_RA_RT_TABLE, + DEVCONF_USE_OPTIMISTIC, 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/msdos_fs.h b/include/uapi/linux/msdos_fs.h index f055e58b3147..db4ae0cd16c7 100644 --- a/include/uapi/linux/msdos_fs.h +++ b/include/uapi/linux/msdos_fs.h @@ -104,6 +104,7 @@ struct __fat_dirent { /* <linux/videotext.h> has used 0x72 ('r') in collision, so skip a few */ #define FAT_IOCTL_GET_ATTRIBUTES _IOR('r', 0x10, __u32) #define FAT_IOCTL_SET_ATTRIBUTES _IOW('r', 0x11, __u32) +#define VFAT_IOCTL_GET_VOLUME_ID _IOR('r', 0x12, __u32) struct fat_boot_sector { __u8 ignored[3]; /* Boot strap short or near jump */ @@ -161,6 +162,17 @@ struct fat_boot_fsinfo { __le32 reserved2[4]; }; +struct fat_boot_bsx { + __u8 drive; /* drive number */ + __u8 reserved1; + __u8 signature; /* extended boot signature */ + __u8 vol_id[4]; /* volume ID */ + __u8 vol_label[11]; /* volume label */ + __u8 type[8]; /* file system type */ +}; +#define FAT16_BSX_OFFSET 36 /* offset of fat_boot_bsx in FAT12 and FAT16 */ +#define FAT32_BSX_OFFSET 64 /* offset of fat_boot_bsx in FAT32 */ + struct msdos_dir_entry { __u8 name[MSDOS_NAME];/* name and extension */ __u8 attr; /* attribute bits */ 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 26d7217bd4f1..63594564831c 100644 --- a/include/uapi/linux/netfilter/xt_socket.h +++ b/include/uapi/linux/netfilter/xt_socket.h @@ -11,4 +11,10 @@ struct xt_socket_mtinfo1 { __u8 flags; }; +void xt_socket_put_sk(struct sock *sk); +struct sock *xt_socket_get4_sk(const struct sk_buff *skb, + struct xt_action_param *par); +struct sock *xt_socket_get6_sk(const struct sk_buff *skb, + struct xt_action_param *par); + #endif /* _XT_SOCKET_H */ diff --git a/include/uapi/linux/nl80211.h b/include/uapi/linux/nl80211.h index d1e48b5e348f..11c8c1707d23 100644 --- a/include/uapi/linux/nl80211.h +++ b/include/uapi/linux/nl80211.h @@ -646,6 +646,28 @@ * @NL80211_CMD_CRIT_PROTOCOL_STOP: Indicates the connection reliability can * return back to normal. * + * @NL80211_CMD_GET_COALESCE: Get currently supported coalesce rules. + * @NL80211_CMD_SET_COALESCE: Configure coalesce rules or clear existing rules. + * + * @NL80211_CMD_CHANNEL_SWITCH: Perform a channel switch by announcing the + * the new channel information (Channel Switch Announcement - CSA) + * in the beacon for some time (as defined in the + * %NL80211_ATTR_CH_SWITCH_COUNT parameter) and then change to the + * new channel. Userspace provides the new channel information (using + * %NL80211_ATTR_WIPHY_FREQ and the attributes determining channel + * width). %NL80211_ATTR_CH_SWITCH_BLOCK_TX may be supplied to inform + * other station that transmission must be blocked until the channel + * switch is complete. + * + * @NL80211_CMD_VENDOR: Vendor-specified command/event. The command is specified + * by the %NL80211_ATTR_VENDOR_ID attribute and a sub-command in + * %NL80211_ATTR_VENDOR_SUBCMD. Parameter(s) can be transported in + * %NL80211_ATTR_VENDOR_DATA. + * For feature advertisement, the %NL80211_ATTR_VENDOR_DATA attribute is + * used in the wiphy data as a nested attribute containing descriptions + * (&struct nl80211_vendor_cmd_info) of the supported vendor commands. + * This may also be sent as an event with the same attributes. + * * @NL80211_CMD_MAX: highest used command number * @__NL80211_CMD_AFTER_LAST: internal use */ @@ -808,6 +830,13 @@ enum nl80211_commands { NL80211_CMD_CRIT_PROTOCOL_START, NL80211_CMD_CRIT_PROTOCOL_STOP, + NL80211_CMD_GET_COALESCE, + NL80211_CMD_SET_COALESCE, + + NL80211_CMD_CHANNEL_SWITCH, + + NL80211_CMD_VENDOR, + /* add new commands above here */ /* used to define NL80211_CMD_MAX below */ @@ -1429,6 +1458,48 @@ enum nl80211_commands { * @NL80211_ATTR_MAX_CRIT_PROT_DURATION: duration in milliseconds in which * the connection should have increased reliability (u16). * + * @NL80211_ATTR_PEER_AID: Association ID for the peer TDLS station (u16). + * This is similar to @NL80211_ATTR_STA_AID but with a difference of being + * allowed to be used with the first @NL80211_CMD_SET_STATION command to + * update a TDLS peer STA entry. + * + * @NL80211_ATTR_COALESCE_RULE: Coalesce rule information. + * + * @NL80211_ATTR_CH_SWITCH_COUNT: u32 attribute specifying the number of TBTT's + * until the channel switch event. + * @NL80211_ATTR_CH_SWITCH_BLOCK_TX: flag attribute specifying that transmission + * must be blocked on the current channel (before the channel switch + * operation). + * @NL80211_ATTR_CSA_IES: Nested set of attributes containing the IE information + * for the time while performing a channel switch. + * @NL80211_ATTR_CSA_C_OFF_BEACON: Offset of the channel switch counter + * field in the beacons tail (%NL80211_ATTR_BEACON_TAIL). + * @NL80211_ATTR_CSA_C_OFF_PRESP: Offset of the channel switch counter + * field in the probe response (%NL80211_ATTR_PROBE_RESP). + * + * @NL80211_ATTR_RXMGMT_FLAGS: flags for nl80211_send_mgmt(), u32. + * As specified in the &enum nl80211_rxmgmt_flags. + * + * @NL80211_ATTR_STA_SUPPORTED_CHANNELS: array of supported channels. + * + * @NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES: array of supported + * supported operating classes. + * + * @NL80211_ATTR_HANDLE_DFS: A flag indicating whether user space + * controls DFS operation in IBSS mode. If the flag is included in + * %NL80211_CMD_JOIN_IBSS request, the driver will allow use of DFS + * channels and reports radar events to userspace. Userspace is required + * to react to radar events, e.g. initiate a channel switch or leave the + * IBSS network. + * + * @NL80211_ATTR_VENDOR_ID: The vendor ID, either a 24-bit OUI or, if + * %NL80211_VENDOR_ID_IS_LINUX is set, a special Linux ID (not used yet) + * @NL80211_ATTR_VENDOR_SUBCMD: vendor sub-command + * @NL80211_ATTR_VENDOR_DATA: data for the vendor command, if any; this + * attribute is also used for vendor command feature advertisement + * @NL80211_ATTR_VENDOR_EVENTS: used for event list advertising in the wiphy + * info, containing a nested array of possible events + * * @NL80211_ATTR_MAX: highest attribute number currently defined * @__NL80211_ATTR_AFTER_LAST: internal use */ @@ -1727,6 +1798,35 @@ enum nl80211_attrs { NL80211_ATTR_CRIT_PROT_ID, NL80211_ATTR_MAX_CRIT_PROT_DURATION, + NL80211_ATTR_PEER_AID, + + NL80211_ATTR_COALESCE_RULE, + + NL80211_ATTR_CH_SWITCH_COUNT, + NL80211_ATTR_CH_SWITCH_BLOCK_TX, + NL80211_ATTR_CSA_IES, + NL80211_ATTR_CSA_C_OFF_BEACON, + NL80211_ATTR_CSA_C_OFF_PRESP, + + NL80211_ATTR_RXMGMT_FLAGS, + + NL80211_ATTR_STA_SUPPORTED_CHANNELS, + + NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES, + + NL80211_ATTR_HANDLE_DFS, + + NL80211_ATTR_SUPPORT_5_MHZ, + NL80211_ATTR_SUPPORT_10_MHZ, + + NL80211_ATTR_OPMODE_NOTIF, + + NL80211_ATTR_VENDOR_ID, + NL80211_ATTR_VENDOR_SUBCMD, + NL80211_ATTR_VENDOR_DATA, + + NL80211_ATTR_VENDOR_EVENTS, + /* add attributes here, update the policy in nl80211.c */ __NL80211_ATTR_AFTER_LAST, @@ -2278,9 +2378,15 @@ enum nl80211_reg_rule_attr { * enum nl80211_sched_scan_match_attr - scheduled scan match attributes * @__NL80211_SCHED_SCAN_MATCH_ATTR_INVALID: attribute number 0 is reserved * @NL80211_SCHED_SCAN_MATCH_ATTR_SSID: SSID to be used for matching, - * only report BSS with matching SSID. + * only report BSS with matching SSID. * @NL80211_SCHED_SCAN_MATCH_ATTR_RSSI: RSSI threshold (in dBm) for reporting a - * BSS in scan results. Filtering is turned off if not specified. + * BSS in scan results. Filtering is turned off if not specified. Note that + * if this attribute is in a match set of its own, then it is treated as + * the default value for all matchsets with an SSID, rather than being a + * matchset of its own without an RSSI filter. This is due to problems with + * how this API was implemented in the past. Also, due to the same problem, + * the only way to create a matchset with only an RSSI filter (with this + * attribute) is if there's only a single matchset with the RSSI attribute. * @NL80211_SCHED_SCAN_MATCH_ATTR_MAX: highest scheduled scan filter * attribute number currently defined * @__NL80211_SCHED_SCAN_MATCH_ATTR_AFTER_LAST: internal use @@ -3721,4 +3827,24 @@ enum nl80211_crit_proto_id { /* maximum duration for critical protocol measures */ #define NL80211_CRIT_PROTO_MAX_DURATION 5000 /* msec */ +/* + * If this flag is unset, the lower 24 bits are an OUI, if set + * a Linux nl80211 vendor ID is used (no such IDs are allocated + * yet, so that's not valid so far) + */ +#define NL80211_VENDOR_ID_IS_LINUX 0x80000000 + +/** + * struct nl80211_vendor_cmd_info - vendor command data + * @vendor_id: If the %NL80211_VENDOR_ID_IS_LINUX flag is clear, then the + * value is a 24-bit OUI; if it is set then a separately allocated ID + * may be used, but no such IDs are allocated yet. New IDs should be + * added to this file when needed. + * @subcmd: sub-command ID for the command + */ +struct nl80211_vendor_cmd_info { + __u32 vendor_id; + __u32 subcmd; +}; + #endif /* __LINUX_NL80211_H */ diff --git a/include/uapi/linux/prctl.h b/include/uapi/linux/prctl.h index 289760f424aa..28bb0b3a08bf 100644 --- a/include/uapi/linux/prctl.h +++ b/include/uapi/linux/prctl.h @@ -149,4 +149,13 @@ #define PR_GET_TID_ADDRESS 40 +/* 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 41 + +#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 7a2144e1afae..07c1146c1f51 100644 --- a/include/uapi/linux/rtnetlink.h +++ b/include/uapi/linux/rtnetlink.h @@ -297,6 +297,7 @@ enum rtattr_type_t { RTA_TABLE, RTA_MARK, RTA_MFC_STATS, + RTA_UID, __RTA_MAX }; diff --git a/include/uapi/linux/seccomp.h b/include/uapi/linux/seccomp.h index ac2dc9f72973..0f238a43ff1e 100644 --- a/include/uapi/linux/seccomp.h +++ b/include/uapi/linux/seccomp.h @@ -10,6 +10,13 @@ #define SECCOMP_MODE_STRICT 1 /* uses hard-coded filter. */ #define SECCOMP_MODE_FILTER 2 /* uses user-supplied filter. */ +/* Valid operations for seccomp syscall. */ +#define SECCOMP_SET_MODE_STRICT 0 +#define SECCOMP_SET_MODE_FILTER 1 + +/* Valid flags for SECCOMP_SET_MODE_FILTER */ +#define SECCOMP_FILTER_FLAG_TSYNC 1 + /* * All BPF programs must return a 32-bit value. * The bottom 16-bits are for optional return data. diff --git a/include/uapi/linux/sockios.h b/include/uapi/linux/sockios.h index 7997a506ad41..f7ffe36db03c 100644 --- a/include/uapi/linux/sockios.h +++ b/include/uapi/linux/sockios.h @@ -65,6 +65,7 @@ #define SIOCDIFADDR 0x8936 /* delete PA address */ #define SIOCSIFHWBROADCAST 0x8937 /* set hardware broadcast addr */ #define SIOCGIFCOUNT 0x8938 /* get number of devices */ +#define SIOCKILLADDR 0x8939 /* kill sockets with this local addr */ #define SIOCGIFBR 0x8940 /* Bridging support */ #define SIOCSIFBR 0x8941 /* Set bridging options */ 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/sound/compress_params.h b/include/uapi/sound/compress_params.h index 602dc6c45d1a..165e7059de75 100644 --- a/include/uapi/sound/compress_params.h +++ b/include/uapi/sound/compress_params.h @@ -57,6 +57,7 @@ #define MAX_NUM_CODECS 32 #define MAX_NUM_CODEC_DESCRIPTORS 32 #define MAX_NUM_BITRATES 32 +#define MAX_NUM_SAMPLE_RATES 32 /* Codecs are listed linearly to allow for extensibility */ #define SND_AUDIOCODEC_PCM ((__u32) 0x00000001) @@ -324,7 +325,8 @@ union snd_codec_options { /** struct snd_codec_desc - description of codec capabilities * @max_ch: Maximum number of audio channels - * @sample_rates: Sampling rates in Hz, use SNDRV_PCM_RATE_xxx for this + * @sample_rates: Sampling rates in Hz, use values like 48000 for this + * @num_sample_rates: Number of valid values in sample_rates array * @bit_rate: Indexed array containing supported bit rates * @num_bitrates: Number of valid values in bit_rate array * @rate_control: value is specified by SND_RATECONTROLMODE defines. @@ -346,7 +348,8 @@ union snd_codec_options { struct snd_codec_desc { __u32 max_ch; - __u32 sample_rates; + __u32 sample_rates[MAX_NUM_SAMPLE_RATES]; + __u32 num_sample_rates; __u32 bit_rate[MAX_NUM_BITRATES]; __u32 num_bitrates; __u32 rate_control; @@ -364,7 +367,8 @@ struct snd_codec_desc { * @ch_out: Number of output channels. In case of contradiction between * this field and the channelMode field, the channelMode field * overrides. - * @sample_rate: Audio sample rate of input data + * @sample_rate: Audio sample rate of input data in Hz, use values like 48000 + * for this. * @bit_rate: Bitrate of encoded data. May be ignored by decoders * @rate_control: Encoding rate control. See SND_RATECONTROLMODE defines. * Encoders may rely on profiles for quality levels. 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 8fa4f758821a..79b356f892f9 100644 --- a/init/Kconfig +++ b/init/Kconfig @@ -1251,6 +1251,12 @@ config HOTPLUG config HAVE_PCSPKR_PLATFORM bool +config PANIC_TIMEOUT + int "Default panic timeout" + default 0 + help + Set default panic timeout. + menuconfig EXPERT bool "Configure standard kernel features (expert users)" # Unhide debug options, to make the on-by-default options visible diff --git a/kernel/cgroup.c b/kernel/cgroup.c index ef130605ac43..fcf0aeff45b4 100644 --- a/kernel/cgroup.c +++ b/kernel/cgroup.c @@ -2106,6 +2106,43 @@ out_free_group_list: return retval; } +static int cgroup_allow_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) +{ + struct cgroup_subsys *ss; + int ret; + + for_each_subsys(cgrp->root, ss) { + if (ss->allow_attach) { + ret = ss->allow_attach(cgrp, tset); + if (ret) + return ret; + } else { + return -EACCES; + } + } + + return 0; +} + +int subsys_cgroup_allow_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) +{ + const struct cred *cred = current_cred(), *tcred; + struct task_struct *task; + + if (capable(CAP_SYS_NICE)) + return 0; + + cgroup_taskset_for_each(task, cgrp, tset) { + tcred = __task_cred(task); + + if (current != task && cred->euid != tcred->uid && + cred->euid != tcred->suid) + return -EACCES; + } + + return 0; +} + /* * Find the task_struct of the task to attach by vpid and pass it along to the * function to attach either it or all tasks in its threadgroup. Will lock @@ -2137,9 +2174,18 @@ retry_find_task: if (!uid_eq(cred->euid, GLOBAL_ROOT_UID) && !uid_eq(cred->euid, tcred->uid) && !uid_eq(cred->euid, tcred->suid)) { - rcu_read_unlock(); - ret = -EACCES; - goto out_unlock_cgroup; + /* + * if the default permission check fails, give each + * cgroup a chance to extend the permission check + */ + struct cgroup_taskset tset = { }; + tset.single.task = tsk; + tset.single.cgrp = cgrp; + ret = cgroup_allow_attach(cgrp, &tset); + if (ret) { + rcu_read_unlock(); + goto out_unlock_cgroup; + } } } else tsk = current; diff --git a/kernel/cpu.c b/kernel/cpu.c index 5fbcbdd31fb3..f1cfa7367a92 100644 --- a/kernel/cpu.c +++ b/kernel/cpu.c @@ -745,3 +745,23 @@ 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/debug/debug_core.c b/kernel/debug/debug_core.c index 0506d447aed2..2d4438b14b42 100644 --- a/kernel/debug/debug_core.c +++ b/kernel/debug/debug_core.c @@ -86,6 +86,10 @@ static int kgdb_use_con; bool dbg_is_early = true; /* Next cpu to become the master debug core */ int dbg_switch_cpu; +/* Flag for entering kdb when a panic occurs */ +static bool break_on_panic = true; +/* Flag for entering kdb when an exception occurs */ +static bool break_on_exception = true; /* Use kdb or gdbserver mode */ int dbg_kdb_mode = 1; @@ -100,6 +104,8 @@ early_param("kgdbcon", opt_kgdb_con); module_param(kgdb_use_con, int, 0644); module_param(kgdbreboot, int, 0644); +module_param(break_on_panic, bool, 0644); +module_param(break_on_exception, bool, 0644); /* * Holds information about breakpoints in a kernel. These breakpoints are @@ -678,6 +684,9 @@ kgdb_handle_exception(int evector, int signo, int ecode, struct pt_regs *regs) if (arch_kgdb_ops.enable_nmi) arch_kgdb_ops.enable_nmi(0); + if (unlikely(signo != SIGTRAP && !break_on_exception)) + return 1; + ks->cpu = raw_smp_processor_id(); ks->ex_vector = evector; ks->signo = signo; @@ -784,6 +793,9 @@ static int kgdb_panic_event(struct notifier_block *self, unsigned long val, void *data) { + if (!break_on_panic) + return NOTIFY_DONE; + if (dbg_kdb_mode) kdb_printf("PANIC: %s\n", (char *)data); kgdb_breakpoint(); diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c index 14ff4849262c..4b0fb2fb779c 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/exit.c b/kernel/exit.c index 717efbd7cb72..33fde71b83d0 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -841,7 +841,7 @@ void do_exit(long code) /* * Make sure we are holding no locks: */ - debug_check_no_locks_held(tsk); + debug_check_no_locks_held(); /* * We can do this unlocked here. The futex code uses this flag * just to verify whether the pi state cleanup has been done diff --git a/kernel/fork.c b/kernel/fork.c index 514dbc40f98f..b0663950634b 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -198,6 +198,9 @@ struct kmem_cache *vm_area_cachep; /* SLAB cache for mm_struct structures (tsk->mm) */ static struct kmem_cache *mm_cachep; +/* Notifier list called when a task struct is freed */ +static ATOMIC_NOTIFIER_HEAD(task_free_notifier); + static void account_kernel_stack(struct thread_info *ti, int account) { struct zone *zone = page_zone(virt_to_page(ti)); @@ -231,6 +234,18 @@ static inline void put_signal_struct(struct signal_struct *sig) free_signal_struct(sig); } +int task_free_register(struct notifier_block *n) +{ + return atomic_notifier_chain_register(&task_free_notifier, n); +} +EXPORT_SYMBOL(task_free_register); + +int task_free_unregister(struct notifier_block *n) +{ + return atomic_notifier_chain_unregister(&task_free_notifier, n); +} +EXPORT_SYMBOL(task_free_unregister); + void __put_task_struct(struct task_struct *tsk) { WARN_ON(!tsk->exit_state); @@ -242,6 +257,7 @@ void __put_task_struct(struct task_struct *tsk) delayacct_tsk_free(tsk); put_signal_struct(tsk->signal); + atomic_notifier_call_chain(&task_free_notifier, 0, tsk); if (!profile_handoff_task(tsk)) free_task(tsk); } @@ -311,6 +327,15 @@ static struct task_struct *dup_task_struct(struct task_struct *orig) goto free_ti; tsk->stack = ti; +#ifdef CONFIG_SECCOMP + /* + * We must handle setting up seccomp filters once we're under + * the sighand lock in case orig has changed between now and + * then. Until then, filter must be NULL to avoid messing up + * the usage counts on the error path calling free_task. + */ + tsk->seccomp.filter = NULL; +#endif setup_thread_stack(tsk, orig); clear_user_return_notifier(tsk); @@ -697,7 +722,8 @@ struct mm_struct *mm_access(struct task_struct *task, unsigned int mode) mm = get_task_mm(task); if (mm && mm != current->mm && - !ptrace_may_access(task, mode)) { + !ptrace_may_access(task, mode) && + !capable(CAP_SYS_RESOURCE)) { mmput(mm); mm = ERR_PTR(-EACCES); } @@ -1091,6 +1117,39 @@ static void copy_flags(unsigned long clone_flags, struct task_struct *p) p->flags = new_flags; } +static void copy_seccomp(struct task_struct *p) +{ +#ifdef CONFIG_SECCOMP + /* + * Must be called with sighand->lock held, which is common to + * all threads in the group. Holding cred_guard_mutex is not + * needed because this new task is not yet running and cannot + * be racing exec. + */ + assert_spin_locked(¤t->sighand->siglock); + + /* Ref-count the new filter user, and assign it. */ + get_seccomp_filter(current); + p->seccomp = current->seccomp; + + /* + * Explicitly enable no_new_privs here in case it got set + * between the task_struct being duplicated and holding the + * sighand lock. The seccomp state and nnp must be in sync. + */ + if (task_no_new_privs(current)) + task_set_no_new_privs(p); + + /* + * If the parent gained a seccomp mode after copying thread + * flags and between before we held the sighand lock, we have + * to manually enable the seccomp thread flag here. + */ + if (p->seccomp.mode != SECCOMP_MODE_DISABLED) + set_tsk_thread_flag(p, TIF_SECCOMP); +#endif +} + SYSCALL_DEFINE1(set_tid_address, int __user *, tidptr) { current->clear_child_tid = tidptr; @@ -1195,7 +1254,6 @@ static struct task_struct *copy_process(unsigned long clone_flags, goto fork_out; ftrace_graph_init_task(p); - get_seccomp_filter(p); rt_mutex_init_task(p); @@ -1438,6 +1496,12 @@ static struct task_struct *copy_process(unsigned long clone_flags, spin_lock(¤t->sighand->siglock); /* + * Copy seccomp details explicitly here, in case they were changed + * before holding sighand lock. + */ + copy_seccomp(p); + + /* * Process group and session signals need to be delivered to just the * parent before the fork or both the parent and the child after the * fork. Restart if a signal comes in before we add the new process to diff --git a/kernel/freezer.c b/kernel/freezer.c index bd733f6e610d..4ada72f5f55a 100644 --- a/kernel/freezer.c +++ b/kernel/freezer.c @@ -119,6 +119,18 @@ bool freeze_task(struct task_struct *p) { unsigned long flags; + /* + * This check can race with freezer_do_not_count, but worst case that + * will result in an extra wakeup being sent to the task. It does not + * race with freezer_count(), the barriers in freezer_count() and + * freezer_should_skip() ensure that either freezer_count() sees + * freezing == true in try_to_freeze() and freezes, or + * freezer_should_skip() sees !PF_FREEZE_SKIP and freezes the task + * normally. + */ + if (freezer_should_skip(p)) + return false; + spin_lock_irqsave(&freezer_lock, flags); if (!freezing(p) || frozen(p)) { spin_unlock_irqrestore(&freezer_lock, flags); diff --git a/kernel/futex.c b/kernel/futex.c index 625a4e659e7a..ad971d0f0be0 100644 --- a/kernel/futex.c +++ b/kernel/futex.c @@ -62,6 +62,7 @@ #include <linux/ptrace.h> #include <linux/sched/rt.h> #include <linux/hugetlb.h> +#include <linux/freezer.h> #include <asm/futex.h> @@ -1935,7 +1936,7 @@ static void futex_wait_queue_me(struct futex_hash_bucket *hb, struct futex_q *q, * is no timeout, or if it has yet to expire. */ if (!timeout || timeout->task) - schedule(); + freezable_schedule(); } __set_current_state(TASK_RUNNING); } diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c index aadf4b7a607c..1e84be107436 100644 --- a/kernel/hrtimer.c +++ b/kernel/hrtimer.c @@ -47,6 +47,7 @@ #include <linux/sched/sysctl.h> #include <linux/sched/rt.h> #include <linux/timer.h> +#include <linux/freezer.h> #include <asm/uaccess.h> @@ -1565,7 +1566,7 @@ static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mod t->task = NULL; if (likely(t->task)) - schedule(); + freezable_schedule(); hrtimer_cancel(&t->timer); mode = HRTIMER_MODE_ABS; diff --git a/kernel/irq/pm.c b/kernel/irq/pm.c index abcd6ca86cb7..da5f2ffa7106 100644 --- a/kernel/irq/pm.c +++ b/kernel/irq/pm.c @@ -10,7 +10,7 @@ #include <linux/module.h> #include <linux/interrupt.h> #include <linux/syscore_ops.h> - +#include <linux/wakeup_reason.h> #include "internals.h" /** @@ -103,14 +103,18 @@ int check_wakeup_irqs(void) int irq; for_each_irq_desc(irq, desc) { - /* - * Only interrupts which are marked as wakeup source - * and have not been disabled before the suspend check - * can abort suspend. - */ if (irqd_is_wakeup_set(&desc->irq_data)) { - if (desc->depth == 1 && desc->istate & IRQS_PENDING) + if (desc->istate & IRQS_PENDING) { + log_suspend_abort_reason("Wakeup IRQ %d %s pending", + irq, + desc->action && desc->action->name ? + desc->action->name : ""); + pr_info("Wakeup IRQ %d %s pending, suspend aborted\n", + irq, + desc->action && desc->action->name ? + desc->action->name : ""); return -EBUSY; + } continue; } /* diff --git a/kernel/lockdep.c b/kernel/lockdep.c index 1f3186b37fd5..e16c45b9ee77 100644 --- a/kernel/lockdep.c +++ b/kernel/lockdep.c @@ -4090,7 +4090,7 @@ void debug_check_no_locks_freed(const void *mem_from, unsigned long mem_len) } EXPORT_SYMBOL_GPL(debug_check_no_locks_freed); -static void print_held_locks_bug(struct task_struct *curr) +static void print_held_locks_bug(void) { if (!debug_locks_off()) return; @@ -4099,22 +4099,21 @@ static void print_held_locks_bug(struct task_struct *curr) printk("\n"); printk("=====================================\n"); - printk("[ BUG: lock held at task exit time! ]\n"); + printk("[ BUG: %s/%d still has locks held! ]\n", + current->comm, task_pid_nr(current)); print_kernel_ident(); printk("-------------------------------------\n"); - printk("%s/%d is exiting with locks still held!\n", - curr->comm, task_pid_nr(curr)); - lockdep_print_held_locks(curr); - + lockdep_print_held_locks(current); printk("\nstack backtrace:\n"); dump_stack(); } -void debug_check_no_locks_held(struct task_struct *task) +void debug_check_no_locks_held(void) { - if (unlikely(task->lockdep_depth > 0)) - print_held_locks_bug(task); + if (unlikely(current->lockdep_depth > 0)) + print_held_locks_bug(); } +EXPORT_SYMBOL_GPL(debug_check_no_locks_held); void debug_show_all_locks(void) { diff --git a/kernel/panic.c b/kernel/panic.c index 167ec097ce8b..126b2ef2eb61 100644 --- a/kernel/panic.c +++ b/kernel/panic.c @@ -26,13 +26,19 @@ #define PANIC_TIMER_STEP 100 #define PANIC_BLINK_SPD 18 +/* Machine specific panic information string */ +char *mach_panic_string; + int panic_on_oops = CONFIG_PANIC_ON_OOPS_VALUE; static unsigned long tainted_mask; static int pause_on_oops; static int pause_on_oops_flag; static DEFINE_SPINLOCK(pause_on_oops_lock); -int panic_timeout; +#ifndef CONFIG_PANIC_TIMEOUT +#define CONFIG_PANIC_TIMEOUT 0 +#endif +int panic_timeout = CONFIG_PANIC_TIMEOUT; EXPORT_SYMBOL_GPL(panic_timeout); ATOMIC_NOTIFIER_HEAD(panic_notifier_list); @@ -375,6 +381,11 @@ late_initcall(init_oops_id); void print_oops_end_marker(void) { init_oops_id(); + + if (mach_panic_string) + printk(KERN_WARNING "Board Information: %s\n", + mach_panic_string); + printk(KERN_WARNING "---[ end trace %016llx ]---\n", (unsigned long long)oops_id); } diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index 52423b5d7ae9..e97c63994f6b 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -18,6 +18,14 @@ config SUSPEND_FREEZER Turning OFF this setting is NOT recommended! If in doubt, say Y. +config HAS_WAKELOCK + bool + default y + +config WAKELOCK + bool + default y + config HIBERNATE_CALLBACKS bool @@ -298,3 +306,10 @@ config PM_GENERIC_DOMAINS_OF config CPU_PM bool depends on SUSPEND || CPU_IDLE + +config SUSPEND_TIME + bool "Log time spent in suspend" + ---help--- + Prints the time spent in suspend in the kernel log, and + keeps statistics on the time spent in suspend in + /sys/kernel/debug/suspend_time diff --git a/kernel/power/Makefile b/kernel/power/Makefile index 29472bff11ef..74c713ba61b0 100644 --- a/kernel/power/Makefile +++ b/kernel/power/Makefile @@ -11,5 +11,8 @@ obj-$(CONFIG_HIBERNATION) += hibernate.o snapshot.o swap.o user.o \ block_io.o obj-$(CONFIG_PM_AUTOSLEEP) += autosleep.o obj-$(CONFIG_PM_WAKELOCKS) += wakelock.o +obj-$(CONFIG_SUSPEND_TIME) += suspend_time.o obj-$(CONFIG_MAGIC_SYSRQ) += poweroff.o + +obj-$(CONFIG_SUSPEND) += wakeup_reason.o diff --git a/kernel/power/process.c b/kernel/power/process.c index 0695319b5fde..d1ccce3cfee8 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -17,7 +17,7 @@ #include <linux/delay.h> #include <linux/workqueue.h> #include <linux/kmod.h> - +#include <linux/wakeup_reason.h> /* * Timeout for stopping processes */ @@ -30,9 +30,13 @@ static int try_to_freeze_tasks(bool user_only) unsigned int todo; bool wq_busy = false; struct timeval start, end; - u64 elapsed_csecs64; - unsigned int elapsed_csecs; + u64 elapsed_msecs64; + 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); @@ -62,42 +66,51 @@ 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; } /* * We need to retry, but first give the freezing tasks some - * time to enter the refrigerator. + * time to enter the refrigerator. Start with an initial + * 1 ms sleep followed by exponential backoff until 8 ms. */ - msleep(10); + usleep_range(sleep_usecs / 2, sleep_usecs); + if (sleep_usecs < 8 * USEC_PER_MSEC) + sleep_usecs *= 2; } do_gettimeofday(&end); - elapsed_csecs64 = timeval_to_ns(&end) - timeval_to_ns(&start); - do_div(elapsed_csecs64, NSEC_PER_SEC / 100); - elapsed_csecs = elapsed_csecs64; + elapsed_msecs64 = timeval_to_ns(&end) - timeval_to_ns(&start); + do_div(elapsed_msecs64, NSEC_PER_MSEC); + elapsed_msecs = elapsed_msecs64; - if (todo) { + if (wakeup) { + printk("\n"); + printk(KERN_ERR "Freezing of tasks aborted after %d.%03d seconds", + elapsed_msecs / 1000, elapsed_msecs % 1000); + } else if (todo) { printk("\n"); - printk(KERN_ERR "Freezing of tasks %s after %d.%02d seconds " - "(%d tasks refusing to freeze, wq_busy=%d):\n", - wakeup ? "aborted" : "failed", - elapsed_csecs / 100, elapsed_csecs % 100, + printk(KERN_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); - do_each_thread(g, p) { - if (p != current && !freezer_should_skip(p) - && freezing(p) && !frozen(p)) - sched_show_task(p); - } while_each_thread(g, p); - read_unlock(&tasklist_lock); - } + read_lock(&tasklist_lock); + do_each_thread(g, p) { + if (p != current && !freezer_should_skip(p) + && freezing(p) && !frozen(p)) + sched_show_task(p); + } while_each_thread(g, p); + read_unlock(&tasklist_lock); } else { - printk("(elapsed %d.%02d seconds) ", elapsed_csecs / 100, - elapsed_csecs % 100); + printk("(elapsed %d.%03d seconds) ", elapsed_msecs / 1000, + elapsed_msecs % 1000); } return todo ? -EBUSY : 0; diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c index 903c517b14da..82450c20875c 100644 --- a/kernel/power/suspend.c +++ b/kernel/power/suspend.c @@ -25,7 +25,9 @@ #include <linux/suspend.h> #include <linux/syscore_ops.h> #include <linux/ftrace.h> +#include <linux/rtc.h> #include <trace/events/power.h> +#include <linux/wakeup_reason.h> #include "power.h" @@ -138,7 +140,7 @@ static int suspend_prepare(suspend_state_t state) error = suspend_freeze_processes(); if (!error) return 0; - + log_suspend_abort_reason("One or more tasks refusing to freeze"); suspend_stats.failed_freeze++; dpm_save_failed_step(SUSPEND_FREEZE); Finish: @@ -168,7 +170,8 @@ void __attribute__ ((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; if (need_suspend_ops(state) && suspend_ops->prepare) { error = suspend_ops->prepare(); @@ -178,7 +181,11 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) error = dpm_suspend_end(PMSG_SUSPEND); if (error) { + last_dev = suspend_stats.last_failed_dev + REC_FAILED_NUM - 1; + last_dev %= REC_FAILED_NUM; printk(KERN_ERR "PM: Some devices failed to power down\n"); + log_suspend_abort_reason("%s device failed to power down", + suspend_stats.failed_devs[last_dev]); goto Platform_finish; } @@ -203,8 +210,10 @@ 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()); @@ -215,6 +224,10 @@ static int suspend_enter(suspend_state_t state, bool *wakeup) if (!(suspend_test(TEST_CORE) || *wakeup)) { error = suspend_ops->enter(state); events_check_enabled = false; + } else { + pm_get_active_wakeup_sources(suspend_abort, + MAX_SUSPEND_ABORT_LEN); + log_suspend_abort_reason(suspend_abort); } syscore_resume(); } @@ -262,6 +275,7 @@ int suspend_devices_and_enter(suspend_state_t state) error = dpm_suspend_start(PMSG_SUSPEND); if (error) { printk(KERN_ERR "PM: Some devices failed to suspend\n"); + log_suspend_abort_reason("Some devices failed to suspend"); goto Recover_platform; } suspend_test_finish("suspend devices"); @@ -358,6 +372,18 @@ 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. @@ -372,6 +398,7 @@ 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++; @@ -379,6 +406,7 @@ 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/suspend_time.c b/kernel/power/suspend_time.c new file mode 100644 index 000000000000..d2a65da9f22c --- /dev/null +++ b/kernel/power/suspend_time.c @@ -0,0 +1,111 @@ +/* + * debugfs file to track time spent in suspend + * + * Copyright (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 as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * 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/debugfs.h> +#include <linux/err.h> +#include <linux/init.h> +#include <linux/kernel.h> +#include <linux/seq_file.h> +#include <linux/syscore_ops.h> +#include <linux/time.h> + +static struct timespec suspend_time_before; +static unsigned int time_in_suspend_bins[32]; + +#ifdef CONFIG_DEBUG_FS +static int suspend_time_debug_show(struct seq_file *s, void *data) +{ + int bin; + seq_printf(s, "time (secs) count\n"); + seq_printf(s, "------------------\n"); + for (bin = 0; bin < 32; bin++) { + if (time_in_suspend_bins[bin] == 0) + continue; + seq_printf(s, "%4d - %4d %4u\n", + bin ? 1 << (bin - 1) : 0, 1 << bin, + time_in_suspend_bins[bin]); + } + return 0; +} + +static int suspend_time_debug_open(struct inode *inode, struct file *file) +{ + return single_open(file, suspend_time_debug_show, NULL); +} + +static const struct file_operations suspend_time_debug_fops = { + .open = suspend_time_debug_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int __init suspend_time_debug_init(void) +{ + struct dentry *d; + + d = debugfs_create_file("suspend_time", 0755, NULL, NULL, + &suspend_time_debug_fops); + if (!d) { + pr_err("Failed to create suspend_time debug file\n"); + return -ENOMEM; + } + + return 0; +} + +late_initcall(suspend_time_debug_init); +#endif + +static int suspend_time_syscore_suspend(void) +{ + read_persistent_clock(&suspend_time_before); + + return 0; +} + +static void suspend_time_syscore_resume(void) +{ + struct timespec after; + + read_persistent_clock(&after); + + after = timespec_sub(after, suspend_time_before); + + time_in_suspend_bins[fls(after.tv_sec)]++; + + pr_info("Suspended for %lu.%03lu seconds\n", after.tv_sec, + after.tv_nsec / NSEC_PER_MSEC); +} + +static struct syscore_ops suspend_time_syscore_ops = { + .suspend = suspend_time_syscore_suspend, + .resume = suspend_time_syscore_resume, +}; + +static int suspend_time_syscore_init(void) +{ + register_syscore_ops(&suspend_time_syscore_ops); + + return 0; +} + +static void suspend_time_syscore_exit(void) +{ + unregister_syscore_ops(&suspend_time_syscore_ops); +} +module_init(suspend_time_syscore_init); +module_exit(suspend_time_syscore_exit); diff --git a/kernel/power/wakelock.c b/kernel/power/wakelock.c index 8f50de394d22..c8fba3380076 100644 --- a/kernel/power/wakelock.c +++ b/kernel/power/wakelock.c @@ -9,7 +9,6 @@ * manipulate wakelocks on Android. */ -#include <linux/capability.h> #include <linux/ctype.h> #include <linux/device.h> #include <linux/err.h> @@ -189,9 +188,6 @@ int pm_wake_lock(const char *buf) size_t len; int ret = 0; - if (!capable(CAP_BLOCK_SUSPEND)) - return -EPERM; - while (*str && !isspace(*str)) str++; @@ -235,9 +231,6 @@ int pm_wake_unlock(const char *buf) size_t len; int ret = 0; - if (!capable(CAP_BLOCK_SUSPEND)) - return -EPERM; - len = strlen(buf); if (!len) return -EINVAL; diff --git a/kernel/power/wakeup_reason.c b/kernel/power/wakeup_reason.c new file mode 100644 index 000000000000..76b53400cb16 --- /dev/null +++ b/kernel/power/wakeup_reason.c @@ -0,0 +1,216 @@ +/* + * 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 struct timespec last_xtime; /* wall time before last suspend */ +static struct timespec curr_xtime; /* wall time after last suspend */ +static struct timespec last_stime; /* total_sleep_time before last suspend */ +static struct timespec curr_stime; /* total_sleep_time 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; + + sleep_time = timespec_sub(curr_stime, last_stime); + total_time = timespec_sub(curr_xtime, last_xtime); + suspend_resume_time = timespec_sub(total_time, sleep_time); + + /* + * suspend_resume_time is calculated from sleep_time. Userspace would + * always need both. Export them 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); + snprintf(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) +{ + struct timespec xtom; /* wall_to_monotonic, ignored */ + + switch (pm_event) { + case PM_SUSPEND_PREPARE: + spin_lock(&resume_reason_lock); + irqcount = 0; + suspend_abort = false; + spin_unlock(&resume_reason_lock); + + get_xtime_and_monotonic_and_sleep_offset(&last_xtime, &xtom, + &last_stime); + break; + case PM_POST_SUSPEND: + get_xtime_and_monotonic_and_sleep_offset(&curr_xtime, &xtom, + &curr_stime); + 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/sched/core.c b/kernel/sched/core.c index aa08f6419beb..ea4e780697b4 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7138,13 +7138,24 @@ 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) { static unsigned long prev_jiffy; /* ratelimiting */ rcu_sleep_check(); /* WARN_ON_ONCE() by default, no rate limit reqd. */ if ((preempt_count_equals(preempt_offset) && !irqs_disabled()) || - system_state != SYSTEM_RUNNING || oops_in_progress) + 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; @@ -8116,6 +8127,7 @@ struct cgroup_subsys cpu_cgroup_subsys = { .css_offline = cpu_cgroup_css_offline, .can_attach = cpu_cgroup_can_attach, .attach = cpu_cgroup_attach, + .allow_attach = subsys_cgroup_allow_attach, .exit = cpu_cgroup_exit, .subsys_id = cpu_cgroup_subsys_id, .base_cftypes = cpu_files, diff --git a/kernel/seccomp.c b/kernel/seccomp.c index b7a10048a32c..1fbb1a2bc459 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -18,15 +18,17 @@ #include <linux/compat.h> #include <linux/sched.h> #include <linux/seccomp.h> +#include <linux/slab.h> +#include <linux/syscalls.h> /* #define SECCOMP_DEBUG 1 */ #ifdef CONFIG_SECCOMP_FILTER #include <asm/syscall.h> #include <linux/filter.h> +#include <linux/pid.h> #include <linux/ptrace.h> #include <linux/security.h> -#include <linux/slab.h> #include <linux/tracehook.h> #include <linux/uaccess.h> @@ -95,7 +97,7 @@ u32 seccomp_bpf_load(int off) if (off == BPF_DATA(nr)) return syscall_get_nr(current, regs); if (off == BPF_DATA(arch)) - return syscall_get_arch(current, regs); + return syscall_get_arch(); if (off >= BPF_DATA(args[0]) && off < BPF_DATA(args[6])) { unsigned long value; int arg = (off - BPF_DATA(args[0])) / sizeof(u64); @@ -201,32 +203,170 @@ static int seccomp_check_filter(struct sock_filter *filter, unsigned int flen) */ static u32 seccomp_run_filters(int syscall) { - struct seccomp_filter *f; + struct seccomp_filter *f = ACCESS_ONCE(current->seccomp.filter); u32 ret = SECCOMP_RET_ALLOW; /* Ensure unexpected behavior doesn't result in failing open. */ - if (WARN_ON(current->seccomp.filter == NULL)) + if (unlikely(WARN_ON(f == NULL))) return SECCOMP_RET_KILL; + /* Make sure cross-thread synced filter points somewhere sane. */ + smp_read_barrier_depends(); + /* * All filters in the list are evaluated and the lowest BPF return * value always takes priority (ignoring the DATA). */ - for (f = current->seccomp.filter; f; f = f->prev) { + for (; f; f = f->prev) { u32 cur_ret = sk_run_filter(NULL, f->insns); + if ((cur_ret & SECCOMP_RET_ACTION) < (ret & SECCOMP_RET_ACTION)) ret = cur_ret; } return ret; } +#endif /* CONFIG_SECCOMP_FILTER */ + +static inline bool seccomp_may_assign_mode(unsigned long seccomp_mode) +{ + assert_spin_locked(¤t->sighand->siglock); + + if (current->seccomp.mode && current->seccomp.mode != seccomp_mode) + return false; + + return true; +} + +static inline void seccomp_assign_mode(struct task_struct *task, + unsigned long seccomp_mode) +{ + assert_spin_locked(&task->sighand->siglock); + + task->seccomp.mode = seccomp_mode; + /* + * Make sure TIF_SECCOMP cannot be set before the mode (and + * filter) is set. + */ + smp_mb(); + set_tsk_thread_flag(task, TIF_SECCOMP); +} + +#ifdef CONFIG_SECCOMP_FILTER +/* Returns 1 if the parent is an ancestor of the child. */ +static int is_ancestor(struct seccomp_filter *parent, + struct seccomp_filter *child) +{ + /* NULL is the root ancestor. */ + if (parent == NULL) + return 1; + for (; child; child = child->prev) + if (child == parent) + return 1; + return 0; +} /** - * seccomp_attach_filter: Attaches a seccomp filter to current. + * seccomp_can_sync_threads: checks if all threads can be synchronized + * + * Expects sighand and cred_guard_mutex locks to be held. + * + * Returns 0 on success, -ve on error, or the pid of a thread which was + * either not in the correct seccomp mode or it did not have an ancestral + * seccomp filter. + */ +static inline pid_t seccomp_can_sync_threads(void) +{ + struct task_struct *thread, *caller; + + BUG_ON(!mutex_is_locked(¤t->signal->cred_guard_mutex)); + assert_spin_locked(¤t->sighand->siglock); + + /* Validate all threads being eligible for synchronization. */ + caller = current; + for_each_thread(caller, thread) { + pid_t failed; + + /* Skip current, since it is initiating the sync. */ + if (thread == caller) + continue; + + if (thread->seccomp.mode == SECCOMP_MODE_DISABLED || + (thread->seccomp.mode == SECCOMP_MODE_FILTER && + is_ancestor(thread->seccomp.filter, + caller->seccomp.filter))) + continue; + + /* Return the first thread that cannot be synchronized. */ + failed = task_pid_vnr(thread); + /* If the pid cannot be resolved, then return -ESRCH */ + if (unlikely(WARN_ON(failed == 0))) + failed = -ESRCH; + return failed; + } + + return 0; +} + +/** + * seccomp_sync_threads: sets all threads to use current's filter + * + * Expects sighand and cred_guard_mutex locks to be held, and for + * seccomp_can_sync_threads() to have returned success already + * without dropping the locks. + * + */ +static inline void seccomp_sync_threads(void) +{ + struct task_struct *thread, *caller; + + BUG_ON(!mutex_is_locked(¤t->signal->cred_guard_mutex)); + assert_spin_locked(¤t->sighand->siglock); + + /* Synchronize all threads. */ + caller = current; + for_each_thread(caller, thread) { + /* Skip current, since it needs no changes. */ + if (thread == caller) + continue; + + /* Get a task reference for the new leaf node. */ + get_seccomp_filter(caller); + /* + * Drop the task reference to the shared ancestor since + * current's path will hold a reference. (This also + * allows a put before the assignment.) + */ + put_seccomp_filter(thread); + smp_store_release(&thread->seccomp.filter, + caller->seccomp.filter); + /* + * Opt the other thread into seccomp if needed. + * As threads are considered to be trust-realm + * equivalent (see ptrace_may_access), it is safe to + * allow one thread to transition the other. + */ + if (thread->seccomp.mode == SECCOMP_MODE_DISABLED) { + /* + * Don't let an unprivileged task work around + * the no_new_privs restriction by creating + * a thread that sets it up, enters seccomp, + * then dies. + */ + if (task_no_new_privs(caller)) + task_set_no_new_privs(thread); + + seccomp_assign_mode(thread, SECCOMP_MODE_FILTER); + } + } +} + +/** + * seccomp_prepare_filter: Prepares a seccomp filter for use. * @fprog: BPF program to install * - * Returns 0 on success or an errno on failure. + * Returns filter on success or an ERR_PTR on failure. */ -static long seccomp_attach_filter(struct sock_fprog *fprog) +static struct seccomp_filter *seccomp_prepare_filter(struct sock_fprog *fprog) { struct seccomp_filter *filter; unsigned long fp_size = fprog->len * sizeof(struct sock_filter); @@ -234,12 +374,13 @@ static long seccomp_attach_filter(struct sock_fprog *fprog) long ret; if (fprog->len == 0 || fprog->len > BPF_MAXINSNS) - return -EINVAL; + return ERR_PTR(-EINVAL); + BUG_ON(INT_MAX / fprog->len < sizeof(struct sock_filter)); for (filter = current->seccomp.filter; filter; filter = filter->prev) total_insns += filter->len + 4; /* include a 4 instr penalty */ if (total_insns > MAX_INSNS_PER_PATH) - return -ENOMEM; + return ERR_PTR(-ENOMEM); /* * Installing a seccomp filter requires that the task have @@ -247,16 +388,16 @@ static long seccomp_attach_filter(struct sock_fprog *fprog) * This avoids scenarios where unprivileged tasks can affect the * behavior of privileged children. */ - if (!current->no_new_privs && + if (!task_no_new_privs(current) && security_capable_noaudit(current_cred(), current_user_ns(), CAP_SYS_ADMIN) != 0) - return -EACCES; + return ERR_PTR(-EACCES); /* Allocate a new seccomp_filter */ filter = kzalloc(sizeof(struct seccomp_filter) + fp_size, GFP_KERNEL|__GFP_NOWARN); if (!filter) - return -ENOMEM; + return ERR_PTR(-ENOMEM);; atomic_set(&filter->usage, 1); filter->len = fprog->len; @@ -275,28 +416,24 @@ static long seccomp_attach_filter(struct sock_fprog *fprog) if (ret) goto fail; - /* - * If there is an existing filter, make it the prev and don't drop its - * task reference. - */ - filter->prev = current->seccomp.filter; - current->seccomp.filter = filter; - return 0; + return filter; + fail: kfree(filter); - return ret; + return ERR_PTR(ret); } /** - * seccomp_attach_user_filter - attaches a user-supplied sock_fprog + * seccomp_prepare_user_filter - prepares a user-supplied sock_fprog * @user_filter: pointer to the user data containing a sock_fprog. * * Returns 0 on success and non-zero otherwise. */ -long seccomp_attach_user_filter(char __user *user_filter) +static struct seccomp_filter * +seccomp_prepare_user_filter(const char __user *user_filter) { struct sock_fprog fprog; - long ret = -EFAULT; + struct seccomp_filter *filter = ERR_PTR(-EFAULT); #ifdef CONFIG_COMPAT if (is_compat_task()) { @@ -309,9 +446,56 @@ long seccomp_attach_user_filter(char __user *user_filter) #endif if (copy_from_user(&fprog, user_filter, sizeof(fprog))) goto out; - ret = seccomp_attach_filter(&fprog); + filter = seccomp_prepare_filter(&fprog); out: - return ret; + return filter; +} + +/** + * seccomp_attach_filter: validate and attach filter + * @flags: flags to change filter behavior + * @filter: seccomp filter to add to the current process + * + * Caller must be holding current->sighand->siglock lock. + * + * Returns 0 on success, -ve on error. + */ +static long seccomp_attach_filter(unsigned int flags, + struct seccomp_filter *filter) +{ + unsigned long total_insns; + struct seccomp_filter *walker; + + assert_spin_locked(¤t->sighand->siglock); + + /* Validate resulting filter length. */ + total_insns = filter->len; + for (walker = current->seccomp.filter; walker; walker = walker->prev) + total_insns += walker->len + 4; /* 4 instr penalty */ + if (total_insns > MAX_INSNS_PER_PATH) + return -ENOMEM; + + /* If thread sync has been requested, check that it is possible. */ + if (flags & SECCOMP_FILTER_FLAG_TSYNC) { + int ret; + + ret = seccomp_can_sync_threads(); + if (ret) + return ret; + } + + /* + * If there is an existing filter, make it the prev and don't drop its + * task reference. + */ + filter->prev = current->seccomp.filter; + current->seccomp.filter = filter; + + /* Now that the new filter is in place, synchronize to all threads. */ + if (flags & SECCOMP_FILTER_FLAG_TSYNC) + seccomp_sync_threads(); + + return 0; } /* get_seccomp_filter - increments the reference count of the filter on @tsk */ @@ -324,6 +508,13 @@ void get_seccomp_filter(struct task_struct *tsk) atomic_inc(&orig->usage); } +static inline void seccomp_filter_free(struct seccomp_filter *filter) +{ + if (filter) { + kfree(filter); + } +} + /* put_seccomp_filter - decrements the ref count of tsk->seccomp.filter */ void put_seccomp_filter(struct task_struct *tsk) { @@ -332,7 +523,7 @@ void put_seccomp_filter(struct task_struct *tsk) while (orig && atomic_dec_and_test(&orig->usage)) { struct seccomp_filter *freeme = orig; orig = orig->prev; - kfree(freeme); + seccomp_filter_free(freeme); } } @@ -351,7 +542,7 @@ static void seccomp_send_sigsys(int syscall, int reason) info.si_code = SYS_SECCOMP; info.si_call_addr = (void __user *)KSTK_EIP(current); info.si_errno = reason; - info.si_arch = syscall_get_arch(current, task_pt_regs(current)); + info.si_arch = syscall_get_arch(); info.si_syscall = syscall; force_sig_info(SIGSYS, &info, current); } @@ -376,12 +567,17 @@ static int mode1_syscalls_32[] = { int __secure_computing(int this_syscall) { - int mode = current->seccomp.mode; int exit_sig = 0; int *syscall; u32 ret; - switch (mode) { + /* + * Make sure that any changes to mode from another thread have + * been seen after TIF_SECCOMP was seen. + */ + rmb(); + + switch (current->seccomp.mode) { case SECCOMP_MODE_STRICT: syscall = mode1_syscalls; #ifdef CONFIG_COMPAT @@ -467,47 +663,152 @@ long prctl_get_seccomp(void) } /** - * prctl_set_seccomp: configures current->seccomp.mode - * @seccomp_mode: requested mode to use - * @filter: optional struct sock_fprog for use with SECCOMP_MODE_FILTER + * seccomp_set_mode_strict: internal function for setting strict seccomp + * + * Once current->seccomp.mode is non-zero, it may not be changed. + * + * Returns 0 on success or -EINVAL on failure. + */ +static long seccomp_set_mode_strict(void) +{ + const unsigned long seccomp_mode = SECCOMP_MODE_STRICT; + long ret = -EINVAL; + + spin_lock_irq(¤t->sighand->siglock); + + if (!seccomp_may_assign_mode(seccomp_mode)) + goto out; + +#ifdef TIF_NOTSC + disable_TSC(); +#endif + seccomp_assign_mode(current, seccomp_mode); + ret = 0; + +out: + spin_unlock_irq(¤t->sighand->siglock); + + return ret; +} + +#ifdef CONFIG_SECCOMP_FILTER +/** + * seccomp_set_mode_filter: internal function for setting seccomp filter + * @flags: flags to change filter behavior + * @filter: struct sock_fprog containing filter * - * This function may be called repeatedly with a @seccomp_mode of - * SECCOMP_MODE_FILTER to install additional filters. Every filter - * successfully installed will be evaluated (in reverse order) for each system - * call the task makes. + * This function may be called repeatedly to install additional filters. + * Every filter successfully installed will be evaluated (in reverse order) + * for each system call the task makes. * * Once current->seccomp.mode is non-zero, it may not be changed. * * Returns 0 on success or -EINVAL on failure. */ -long prctl_set_seccomp(unsigned long seccomp_mode, char __user *filter) +static long seccomp_set_mode_filter(unsigned int flags, + const char __user *filter) { + const unsigned long seccomp_mode = SECCOMP_MODE_FILTER; + struct seccomp_filter *prepared = NULL; long ret = -EINVAL; - if (current->seccomp.mode && - current->seccomp.mode != seccomp_mode) + /* Validate flags. */ + if (flags & ~SECCOMP_FILTER_FLAG_MASK) + return -EINVAL; + + /* Prepare the new filter before holding any locks. */ + prepared = seccomp_prepare_user_filter(filter); + if (IS_ERR(prepared)) + return PTR_ERR(prepared); + + /* + * Make sure we cannot change seccomp or nnp state via TSYNC + * while another thread is in the middle of calling exec. + */ + if (flags & SECCOMP_FILTER_FLAG_TSYNC && + mutex_lock_killable(¤t->signal->cred_guard_mutex)) + goto out_free; + + spin_lock_irq(¤t->sighand->siglock); + + if (!seccomp_may_assign_mode(seccomp_mode)) + goto out; + + ret = seccomp_attach_filter(flags, prepared); + if (ret) goto out; + /* Do not free the successfully attached filter. */ + prepared = NULL; + + seccomp_assign_mode(current, seccomp_mode); +out: + spin_unlock_irq(¤t->sighand->siglock); + if (flags & SECCOMP_FILTER_FLAG_TSYNC) + mutex_unlock(¤t->signal->cred_guard_mutex); +out_free: + seccomp_filter_free(prepared); + return ret; +} +#else +static inline long seccomp_set_mode_filter(unsigned int flags, + const char __user *filter) +{ + return -EINVAL; +} +#endif + +/* Common entry point for both prctl and syscall. */ +static long do_seccomp(unsigned int op, unsigned int flags, + const char __user *uargs) +{ + switch (op) { + case SECCOMP_SET_MODE_STRICT: + if (flags != 0 || uargs != NULL) + return -EINVAL; + return seccomp_set_mode_strict(); + case SECCOMP_SET_MODE_FILTER: + return seccomp_set_mode_filter(flags, uargs); + default: + return -EINVAL; + } +} + +SYSCALL_DEFINE3(seccomp, unsigned int, op, unsigned int, flags, + const char __user *, uargs) +{ + return do_seccomp(op, flags, uargs); +} + +/** + * prctl_set_seccomp: configures current->seccomp.mode + * @seccomp_mode: requested mode to use + * @filter: optional struct sock_fprog for use with SECCOMP_MODE_FILTER + * + * Returns 0 on success or -EINVAL on failure. + */ +long prctl_set_seccomp(unsigned long seccomp_mode, char __user *filter) +{ + unsigned int op; + char __user *uargs; switch (seccomp_mode) { case SECCOMP_MODE_STRICT: - ret = 0; -#ifdef TIF_NOTSC - disable_TSC(); -#endif + op = SECCOMP_SET_MODE_STRICT; + /* + * Setting strict mode through prctl always ignored filter, + * so make sure it is always NULL here to pass the internal + * check in do_seccomp(). + */ + uargs = NULL; break; -#ifdef CONFIG_SECCOMP_FILTER case SECCOMP_MODE_FILTER: - ret = seccomp_attach_user_filter(filter); - if (ret) - goto out; + op = SECCOMP_SET_MODE_FILTER; + uargs = filter; break; -#endif default: - goto out; + return -EINVAL; } - current->seccomp.mode = seccomp_mode; - set_thread_flag(TIF_SECCOMP); -out: - return ret; + /* prctl interface doesn't have flags, so they are always zero. */ + return do_seccomp(op, 0, uargs); } diff --git a/kernel/signal.c b/kernel/signal.c index 113411bfe8b1..50e41075ac77 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -2848,7 +2848,7 @@ int do_sigtimedwait(const sigset_t *which, siginfo_t *info, recalc_sigpending(); spin_unlock_irq(&tsk->sighand->siglock); - timeout = schedule_timeout_interruptible(timeout); + timeout = freezable_schedule_timeout_interruptible(timeout); spin_lock_irq(&tsk->sighand->siglock); __set_task_blocked(tsk, &tsk->real_blocked); diff --git a/kernel/sys.c b/kernel/sys.c index 2bbd9a73b54c..a3bef5bd452b 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -42,6 +42,9 @@ #include <linux/syscore_ops.h> #include <linux/version.h> #include <linux/ctype.h> +#include <linux/mm.h> +#include <linux/mempolicy.h> +#include <linux/sched.h> #include <linux/compat.h> #include <linux/syscalls.h> @@ -2099,10 +2102,158 @@ 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), + 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->shared.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; @@ -2245,6 +2396,26 @@ 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; @@ -2256,12 +2427,15 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3, if (arg2 != 1 || arg3 || arg4 || arg5) return -EINVAL; - current->no_new_privs = 1; + task_set_no_new_privs(current); break; case PR_GET_NO_NEW_PRIVS: if (arg2 || arg3 || arg4 || arg5) return -EINVAL; - return current->no_new_privs ? 1 : 0; + return task_no_new_privs(current) ? 1 : 0; + case PR_SET_VMA: + error = prctl_set_vma(arg2, arg3, arg4, arg5); + break; default: error = -EINVAL; break; diff --git a/kernel/sys_ni.c b/kernel/sys_ni.c index 7078052284fd..7e7fc0a082c4 100644 --- a/kernel/sys_ni.c +++ b/kernel/sys_ni.c @@ -209,3 +209,6 @@ cond_syscall(compat_sys_open_by_handle_at); /* compare kernel pointers */ cond_syscall(sys_kcmp); + +/* operate on Secure Computing state */ +cond_syscall(sys_seccomp); diff --git a/kernel/sysctl.c b/kernel/sysctl.c index 9469f4c61a30..34b8f77b4c02 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -105,6 +105,8 @@ extern char core_pattern[]; extern unsigned int core_pipe_limit; #endif extern int pid_max; +extern int extra_free_kbytes; +extern int min_free_order_shift; extern int pid_max_min, pid_max_max; extern int percpu_pagelist_fraction; extern int compat_log; @@ -1282,6 +1284,21 @@ 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 = "min_free_order_shift", + .data = &min_free_order_shift, + .maxlen = sizeof(min_free_order_shift), + .mode = 0644, + .proc_handler = &proc_dointvec + }, + { .procname = "percpu_pagelist_fraction", .data = &percpu_pagelist_fraction, .maxlen = sizeof(percpu_pagelist_fraction), diff --git a/kernel/time/alarmtimer.c b/kernel/time/alarmtimer.c index 7d19fca0617e..bb08d0961f80 100644 --- a/kernel/time/alarmtimer.c +++ b/kernel/time/alarmtimer.c @@ -199,6 +199,12 @@ static enum hrtimer_restart alarmtimer_fired(struct hrtimer *timer) } +ktime_t alarm_expires_remaining(const struct alarm *alarm) +{ + struct alarm_base *base = &alarm_bases[alarm->type]; + return ktime_sub(alarm->node.expires, base->gettime()); +} + #ifdef CONFIG_RTC_CLASS /** * alarmtimer_suspend - Suspend time callback @@ -305,7 +311,7 @@ void alarm_init(struct alarm *alarm, enum alarmtimer_type type, } /** - * alarm_start - Sets an alarm to fire + * alarm_start - Sets an absolute alarm to fire * @alarm: ptr to alarm to set * @start: time to run the alarm */ @@ -325,6 +331,31 @@ int alarm_start(struct alarm *alarm, ktime_t start) } /** + * alarm_start_relative - Sets a relative alarm to fire + * @alarm: ptr to alarm to set + * @start: time relative to now to run the alarm + */ +int alarm_start_relative(struct alarm *alarm, ktime_t start) +{ + struct alarm_base *base = &alarm_bases[alarm->type]; + + start = ktime_add(start, base->gettime()); + return alarm_start(alarm, start); +} + +void alarm_restart(struct alarm *alarm) +{ + struct alarm_base *base = &alarm_bases[alarm->type]; + unsigned long flags; + + spin_lock_irqsave(&base->lock, flags); + hrtimer_set_expires(&alarm->timer, alarm->node.expires); + hrtimer_restart(&alarm->timer); + alarmtimer_enqueue(base, alarm); + spin_unlock_irqrestore(&base->lock, flags); +} + +/** * alarm_try_to_cancel - Tries to cancel an alarm timer * @alarm: ptr to alarm to be canceled * @@ -394,6 +425,12 @@ u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval) return overrun; } +u64 alarm_forward_now(struct alarm *alarm, ktime_t interval) +{ + struct alarm_base *base = &alarm_bases[alarm->type]; + + return alarm_forward(alarm, base->gettime(), interval); +} diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index 015f85aaca08..e24c188cbbcc 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -82,6 +82,9 @@ config EVENT_TRACING select CONTEXT_SWITCH_TRACER bool +config GPU_TRACEPOINTS + bool + config CONTEXT_SWITCH_TRACER bool diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index d7e2068e4b71..45012122fbb6 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -60,5 +60,6 @@ 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 libftrace-y := ftrace.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 640e4c44b170..249e32b5aa0d 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -730,6 +730,7 @@ static const char *trace_options[] = { "irq-info", "markers", "function-trace", + "print-tgid", NULL }; @@ -1242,6 +1243,7 @@ void tracing_reset_all_online_cpus(void) static unsigned map_pid_to_cmdline[PID_MAX_DEFAULT+1]; static unsigned map_cmdline_to_pid[SAVED_CMDLINES]; static char saved_cmdlines[SAVED_CMDLINES][TASK_COMM_LEN]; +static unsigned saved_tgids[SAVED_CMDLINES]; static int cmdline_idx; static arch_spinlock_t trace_cmdline_lock = __ARCH_SPIN_LOCK_UNLOCKED; @@ -1443,6 +1445,7 @@ static int trace_save_cmdline(struct task_struct *tsk) } memcpy(&saved_cmdlines[idx], tsk->comm, TASK_COMM_LEN); + saved_tgids[idx] = tsk->tgid; arch_spin_unlock(&trace_cmdline_lock); @@ -1480,6 +1483,25 @@ 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 = 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()) @@ -2435,6 +2457,13 @@ static void print_func_help_header(struct trace_buffer *buf, struct seq_file *m) seq_puts(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); @@ -2447,6 +2476,18 @@ static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file seq_puts(m, "# | | | |||| | |\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) { @@ -2747,9 +2788,15 @@ 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); } } } @@ -3601,9 +3648,53 @@ tracing_saved_cmdlines_read(struct file *file, char __user *ubuf, } static const struct file_operations tracing_saved_cmdlines_fops = { - .open = tracing_open_generic, - .read = tracing_saved_cmdlines_read, - .llseek = generic_file_llseek, + .open = tracing_open_generic, + .read = tracing_saved_cmdlines_read, + .llseek = generic_file_llseek, +}; + +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*(16+1+16), GFP_KERNEL); + if (!file_buf) + return -ENOMEM; + + buf = file_buf; + + for (i = 0; i < SAVED_CMDLINES; i++) { + int tgid; + int r; + + pid = 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 @@ -6157,6 +6248,9 @@ init_tracer_debugfs(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 aa0e736b72ac..79462443658d 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -653,6 +653,7 @@ 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; @@ -866,6 +867,7 @@ enum trace_iterator_flags { TRACE_ITER_IRQ_INFO = 0x800000, TRACE_ITER_MARKERS = 0x1000000, TRACE_ITER_FUNCTION = 0x2000000, + TRACE_ITER_TGID = 0x4000000, }; /* diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index 8388bc99f2ee..28dd40c2c423 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -46,6 +46,8 @@ struct fgraph_data { #define TRACE_GRAPH_PRINT_DURATION 0x10 #define TRACE_GRAPH_PRINT_ABS_TIME 0x20 #define TRACE_GRAPH_PRINT_IRQS 0x40 +#define TRACE_GRAPH_PRINT_FLAT 0x80 + static unsigned int max_depth; @@ -64,6 +66,8 @@ static struct tracer_opt trace_opts[] = { { TRACER_OPT(funcgraph-abstime, TRACE_GRAPH_PRINT_ABS_TIME) }, /* Display interrupts */ { TRACER_OPT(funcgraph-irqs, TRACE_GRAPH_PRINT_IRQS) }, + /* Use standard trace formatting rather than hierarchical */ + { TRACER_OPT(funcgraph-flat, TRACE_GRAPH_PRINT_FLAT) }, { } /* Empty entry */ }; @@ -1234,6 +1238,9 @@ 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; @@ -1291,13 +1298,6 @@ 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 */ @@ -1364,6 +1364,11 @@ void print_graph_headers_flags(struct seq_file *s, u32 flags) { struct trace_iterator *iter = s->private; + if (flags & TRACE_GRAPH_PRINT_FLAT) { + trace_default_header(s); + return; + } + if (!(trace_flags & TRACE_ITER_CONTEXT_INFO)) return; @@ -1434,20 +1439,6 @@ static int func_graph_set_flag(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 __read_mostly = { .name = "function_graph", .open = graph_trace_open, @@ -1523,16 +1514,6 @@ static __init int init_graph_trace(void) { max_bytes_for_cpu = snprintf(NULL, 0, "%d", nr_cpu_ids - 1); - if (!register_ftrace_event(&graph_trace_entry_event)) { - pr_warning("Warning: could not register graph trace events\n"); - return 1; - } - - if (!register_ftrace_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_output.c b/kernel/trace/trace_output.c index bb922d9ee51b..a68e5e34c00b 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -702,11 +702,25 @@ int trace_print_context(struct trace_iterator *iter) unsigned long secs, usec_rem; char comm[TASK_COMM_LEN]; int ret; + int tgid; trace_find_cmdline(entry->pid, comm); - ret = trace_seq_printf(s, "%16s-%-5d [%03d] ", - comm, entry->pid, iter->cpu); + ret = trace_seq_printf(s, "%16s-%-5d ", comm, entry->pid); + if (!ret) + return 0; + + if (trace_flags & TRACE_ITER_TGID) { + tgid = trace_find_tgid(entry->pid); + if (tgid < 0) + ret = trace_seq_puts(s, "(-----) "); + else + ret = trace_seq_printf(s, "(%5d) ", tgid); + if (!ret) + return 0; + } + + ret = trace_seq_printf(s, "[%03d] ", iter->cpu); if (!ret) return 0; @@ -1035,6 +1049,168 @@ 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); + + if (!trace_seq_puts(s, "graph_ent: func=")) + return TRACE_TYPE_PARTIAL_LINE; + + if (!seq_print_ip_sym(s, field->graph_ent.func, flags)) + return TRACE_TYPE_PARTIAL_LINE; + + if (!trace_seq_puts(s, "\n")) + 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); + + if (!trace_seq_printf(&iter->seq, "%lx %d\n", + field->graph_ent.func, + field->graph_ent.depth)) + 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_RET(s, field->graph_ent.func); + SEQ_PUT_HEX_FIELD_RET(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_RET(s, field->graph_ent.func); + SEQ_PUT_FIELD_RET(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); + + if (!trace_seq_puts(s, "graph_ret: func=")) + return TRACE_TYPE_PARTIAL_LINE; + + if (!seq_print_ip_sym(s, field->ret.func, flags)) + return TRACE_TYPE_PARTIAL_LINE; + + if (!trace_seq_puts(s, "\n")) + 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); + + if (!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)); + 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_RET(s, field->ret.func); + SEQ_PUT_HEX_FIELD_RET(s, field->ret.calltime); + SEQ_PUT_HEX_FIELD_RET(s, field->ret.rettime); + SEQ_PUT_HEX_FIELD_RET(s, field->ret.overrun); + SEQ_PUT_HEX_FIELD_RET(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_RET(s, field->ret.func); + SEQ_PUT_FIELD_RET(s, field->ret.calltime); + SEQ_PUT_FIELD_RET(s, field->ret.rettime); + SEQ_PUT_FIELD_RET(s, field->ret.overrun); + SEQ_PUT_FIELD_RET(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) @@ -1425,6 +1601,8 @@ 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 05039e348f07..e092e5a6cdd7 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -45,6 +45,11 @@ static DEFINE_PER_CPU(unsigned long, soft_lockup_hrtimer_cnt); 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 @@ -178,7 +183,7 @@ void touch_softlockup_watchdog_sync(void) __raw_get_cpu_var(watchdog_touch_ts) = 0; } -#ifdef CONFIG_HARDLOCKUP_DETECTOR +#ifdef CONFIG_HARDLOCKUP_DETECTOR_NMI /* watchdog detector functions */ static int is_hardlockup(void) { @@ -192,6 +197,76 @@ static int 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(); @@ -203,7 +278,7 @@ 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, @@ -251,7 +326,7 @@ 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) { @@ -271,6 +346,9 @@ 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)); @@ -395,7 +473,7 @@ static void watchdog(unsigned int cpu) __touch_watchdog(); } -#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 @@ -471,9 +549,44 @@ static void watchdog_nmi_disable(unsigned int cpu) return; } #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 */ /* prepare/enable/disable routines */ /* sysctl functions */ diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 74fdc5cf4adc..d317c1ad62ab 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -191,15 +191,27 @@ 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 @@ -669,8 +681,9 @@ config DEBUG_LOCKING_API_SELFTESTS mutexes and rwsems. config STACKTRACE - bool + bool "Stacktrace" depends on STACKTRACE_SUPPORT + default y config DEBUG_STACK_USAGE bool "Stack utilization instrumentation" diff --git a/mm/madvise.c b/mm/madvise.c index 7055883e6e25..85d1a5427c27 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -102,7 +102,8 @@ 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_file, pgoff, vma_policy(vma), + vma_get_anon_name(vma)); if (*prev) { vma = *prev; goto success; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index eaa3accb01e7..81bf16fd2266 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -6795,6 +6795,12 @@ static int mem_cgroup_can_attach(struct cgroup *cgroup, return ret; } +static int mem_cgroup_allow_attach(struct cgroup *cgroup, + struct cgroup_taskset *tset) +{ + return subsys_cgroup_allow_attach(cgroup, tset); +} + static void mem_cgroup_cancel_attach(struct cgroup *cgroup, struct cgroup_taskset *tset) { @@ -6963,6 +6969,11 @@ static int mem_cgroup_can_attach(struct cgroup *cgroup, { return 0; } +static int mem_cgroup_allow_attach(struct cgroup *cgroup, + struct cgroup_taskset *tset) +{ + return 0; +} static void mem_cgroup_cancel_attach(struct cgroup *cgroup, struct cgroup_taskset *tset) { @@ -6998,6 +7009,7 @@ struct cgroup_subsys mem_cgroup_subsys = { .can_attach = mem_cgroup_can_attach, .cancel_attach = mem_cgroup_cancel_attach, .attach = mem_cgroup_move_task, + .allow_attach = mem_cgroup_allow_attach, .bind = mem_cgroup_bind, .base_cftypes = mem_cgroup_files, .early_init = 0, diff --git a/mm/mempolicy.c b/mm/mempolicy.c index b2061bb5af73..420e4b97ffab 100644 --- a/mm/mempolicy.c +++ b/mm/mempolicy.c @@ -725,7 +725,7 @@ 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); + new_pol, vma_get_anon_name(vma)); if (prev) { vma = prev; next = vma->vm_next; diff --git a/mm/mlock.c b/mm/mlock.c index 713e462c0776..b680dfd7dbf2 100644 --- a/mm/mlock.c +++ b/mm/mlock.c @@ -23,10 +23,10 @@ int can_do_mlock(void) { - if (capable(CAP_IPC_LOCK)) - return 1; if (rlimit(RLIMIT_MEMLOCK) != 0) return 1; + if (capable(CAP_IPC_LOCK)) + return 1; return 0; } EXPORT_SYMBOL(can_do_mlock); @@ -289,7 +289,8 @@ 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_file, pgoff, vma_policy(vma), + vma_get_anon_name(vma)); if (*prev) { vma = *prev; goto success; diff --git a/mm/mmap.c b/mm/mmap.c index 70ff9b41c970..84a1c44242a8 100644 --- a/mm/mmap.c +++ b/mm/mmap.c @@ -893,7 +893,8 @@ again: remove_next = 1 + (end > next->vm_end); * per-vma resources, so we don't attempt to merge those. */ static inline int is_mergeable_vma(struct vm_area_struct *vma, - struct file *file, unsigned long vm_flags) + struct file *file, unsigned long vm_flags, + const char __user *anon_name) { if (vma->vm_flags ^ vm_flags) return 0; @@ -901,6 +902,8 @@ static inline int is_mergeable_vma(struct vm_area_struct *vma, return 0; if (vma->vm_ops && vma->vm_ops->close) return 0; + if (vma_get_anon_name(vma) != anon_name) + return 0; return 1; } @@ -931,9 +934,10 @@ static inline int is_mergeable_anon_vma(struct anon_vma *anon_vma1, */ 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 anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, + const char __user *anon_name) { - if (is_mergeable_vma(vma, file, vm_flags) && + if (is_mergeable_vma(vma, file, vm_flags, anon_name) && is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { if (vma->vm_pgoff == vm_pgoff) return 1; @@ -950,9 +954,10 @@ can_vma_merge_before(struct vm_area_struct *vma, unsigned long vm_flags, */ 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 anon_vma *anon_vma, struct file *file, pgoff_t vm_pgoff, + const char __user *anon_name) { - if (is_mergeable_vma(vma, file, vm_flags) && + if (is_mergeable_vma(vma, file, vm_flags, anon_name) && is_mergeable_anon_vma(anon_vma, vma->anon_vma, vma)) { pgoff_t vm_pglen; vm_pglen = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT; @@ -963,9 +968,9 @@ 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 @@ -995,7 +1000,8 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm, struct vm_area_struct *prev, unsigned long addr, unsigned long end, unsigned long vm_flags, struct anon_vma *anon_vma, struct file *file, - pgoff_t pgoff, struct mempolicy *policy) + pgoff_t pgoff, struct mempolicy *policy, + const char __user *anon_name) { pgoff_t pglen = (end - addr) >> PAGE_SHIFT; struct vm_area_struct *area, *next; @@ -1021,15 +1027,15 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm, */ if (prev && prev->vm_end == addr && mpol_equal(vma_policy(prev), policy) && - can_vma_merge_after(prev, vm_flags, - anon_vma, file, pgoff)) { + can_vma_merge_after(prev, vm_flags, anon_vma, + file, pgoff, anon_name)) { /* * OK, it can. Can we now merge in the successor as well? */ if (next && end == next->vm_start && mpol_equal(policy, vma_policy(next)) && - can_vma_merge_before(next, vm_flags, - anon_vma, file, pgoff+pglen) && + can_vma_merge_before(next, vm_flags, anon_vma, + file, pgoff+pglen, anon_name) && is_mergeable_anon_vma(prev->anon_vma, next->anon_vma, NULL)) { /* cases 1, 6 */ @@ -1049,8 +1055,8 @@ struct vm_area_struct *vma_merge(struct mm_struct *mm, */ if (next && end == next->vm_start && mpol_equal(policy, vma_policy(next)) && - can_vma_merge_before(next, vm_flags, - anon_vma, file, pgoff+pglen)) { + can_vma_merge_before(next, vm_flags, anon_vma, + file, pgoff+pglen, anon_name)) { if (prev && addr < prev->vm_end) /* case 4 */ err = vma_adjust(prev, prev->vm_start, addr, prev->vm_pgoff, NULL); @@ -1519,7 +1525,8 @@ munmap_back: /* * Can we just expand an old mapping? */ - vma = vma_merge(mm, prev, addr, addr + len, vm_flags, NULL, file, pgoff, NULL); + vma = vma_merge(mm, prev, addr, addr + len, vm_flags, NULL, file, pgoff, + NULL, NULL); if (vma) goto out; @@ -2666,7 +2673,7 @@ 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, NULL, pgoff, NULL, NULL); if (vma) goto out; @@ -2824,7 +2831,8 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap, if (find_vma_links(mm, addr, addr + len, &prev, &rb_link, &rb_parent)) 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->anon_vma, vma->vm_file, pgoff, vma_policy(vma), + 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 e9f65aaa3182..3c44c971f9f3 100644 --- a/mm/mprotect.c +++ b/mm/mprotect.c @@ -278,7 +278,8 @@ 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->anon_vma, vma->vm_file, pgoff, vma_policy(vma), + vma_get_anon_name(vma)); if (*pprev) { vma = *pprev; goto success; diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 494a081ec5e4..025ebfc6d5a7 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -196,7 +196,20 @@ static char * const zone_names[MAX_NR_ZONES] = { "Movable", }; +/* + * 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 min_free_order_shift = 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; @@ -1650,7 +1663,7 @@ static bool __zone_watermark_ok(struct zone *z, int order, unsigned long mark, free_pages -= z->free_area[o].nr_free << o; /* Require fewer higher order pages to be free */ - min >>= 1; + min >>= min_free_order_shift; if (free_pages <= min) return false; @@ -5330,6 +5343,7 @@ 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; @@ -5341,11 +5355,14 @@ 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 @@ -5366,11 +5383,13 @@ 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); setup_zone_migrate_reserve(zone); spin_unlock_irqrestore(&zone->lock, flags); @@ -5483,7 +5502,7 @@ module_init(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(ctl_table *table, int write, void __user *buffer, size_t *length, loff_t *ppos) diff --git a/mm/page_io.c b/mm/page_io.c index a8a3ef45fed7..ba05b64e5d8d 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -21,6 +21,7 @@ #include <linux/writeback.h> #include <linux/frontswap.h> #include <linux/aio.h> +#include <linux/blkdev.h> #include <asm/pgtable.h> static struct bio *get_swap_bio(gfp_t gfp_flags, @@ -80,9 +81,54 @@ void end_swap_bio_read(struct bio *bio, int err) imajor(bio->bi_bdev->bd_inode), iminor(bio->bi_bdev->bd_inode), (unsigned long long)bio->bi_sector); - } else { - SetPageUptodate(page); + goto out; } + + SetPageUptodate(page); + + /* + * There is no guarantee that the page is in swap cache - the software + * suspend code (at least) uses end_swap_bio_read() against a non- + * swapcache page. So we must check PG_swapcache before proceeding with + * this optimization. + */ + if (likely(PageSwapCache(page))) { + struct swap_info_struct *sis; + + sis = page_swap_info(page); + if (sis->flags & SWP_BLKDEV) { + /* + * The swap subsystem performs lazy swap slot freeing, + * expecting that the page will be swapped out again. + * So we can avoid an unnecessary write if the page + * isn't redirtied. + * This is good for real swap storage because we can + * reduce unnecessary I/O and enhance wear-leveling + * if an SSD is used as the as swap device. + * But if in-memory swap device (eg zram) is used, + * this causes a duplicated copy between uncompressed + * data in VM-owned memory and compressed data in + * zram-owned memory. So let's free zram-owned memory + * and make the VM-owned decompressed page *dirty*, + * so the page should be swapped out somewhere again if + * we again wish to reclaim it. + */ + struct gendisk *disk = sis->bdev->bd_disk; + if (disk->fops->swap_slot_free_notify) { + swp_entry_t entry; + unsigned long offset; + + entry.val = page_private(page); + offset = swp_offset(entry); + + SetPageDirty(page); + disk->fops->swap_slot_free_notify(sis->bdev, + offset); + } + } + } + +out: unlock_page(page); bio_put(bio); } diff --git a/mm/shmem.c b/mm/shmem.c index 4e4a7349c5cd..5373c7fffd99 100644 --- a/mm/shmem.c +++ b/mm/shmem.c @@ -3026,6 +3026,14 @@ put_memory: } 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 @@ -3039,10 +3047,7 @@ 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/util.c b/mm/util.c index 0b1725254ff1..f68b2db27e2b 100644 --- a/mm/util.c +++ b/mm/util.c @@ -7,6 +7,7 @@ #include <linux/security.h> #include <linux/swap.h> #include <linux/swapops.h> +#include <linux/vmalloc.h> #include <asm/uaccess.h> #include "internal.h" @@ -381,6 +382,15 @@ unsigned long vm_mmap(struct file *file, unsigned long addr, } EXPORT_SYMBOL(vm_mmap); +void kvfree(const void *addr) +{ + if (is_vmalloc_addr(addr)) + vfree(addr); + else + kfree(addr); +} +EXPORT_SYMBOL(kvfree); + struct address_space *page_mapping(struct page *page) { struct address_space *mapping = page->mapping; diff --git a/mm/vmscan.c b/mm/vmscan.c index a2fd7e759cb7..13a4a39a1f4b 100644 --- a/mm/vmscan.c +++ b/mm/vmscan.c @@ -43,6 +43,7 @@ #include <linux/sysctl.h> #include <linux/oom.h> #include <linux/prefetch.h> +#include <linux/debugfs.h> #include <asm/tlbflush.h> #include <asm/div64.h> @@ -155,6 +156,39 @@ static unsigned long get_lru_size(struct lruvec *lruvec, enum lru_list lru) return zone_page_state(lruvec_zone(lruvec), NR_LRU_BASE + lru); } +struct dentry *debug_file; + +static int debug_shrinker_show(struct seq_file *s, void *unused) +{ + struct shrinker *shrinker; + struct shrink_control sc; + + sc.gfp_mask = -1; + sc.nr_to_scan = 0; + + down_read(&shrinker_rwsem); + list_for_each_entry(shrinker, &shrinker_list, list) { + int num_objs; + + num_objs = shrinker->shrink(shrinker, &sc); + seq_printf(s, "%pf %d\n", shrinker->shrink, num_objs); + } + up_read(&shrinker_rwsem); + return 0; +} + +static int debug_shrinker_open(struct inode *inode, struct file *file) +{ + return single_open(file, debug_shrinker_show, inode->i_private); +} + +static const struct file_operations debug_shrinker_fops = { + .open = debug_shrinker_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + /* * Add a shrinker callback to be called from the vm */ @@ -167,6 +201,15 @@ void register_shrinker(struct shrinker *shrinker) } EXPORT_SYMBOL(register_shrinker); +static int __init add_shrinker_debug(void) +{ + debugfs_create_file("shrinker", 0644, NULL, NULL, + &debug_shrinker_fops); + return 0; +} + +late_initcall(add_shrinker_debug); + /* * Remove one */ diff --git a/net/Kconfig b/net/Kconfig index 2ddc9046868e..2a680dadfd6c 100644 --- a/net/Kconfig +++ b/net/Kconfig @@ -81,6 +81,20 @@ source "net/netlabel/Kconfig" endif # if INET +config ANDROID_PARANOID_NETWORK + bool "Only allow certain groups to create sockets" + default ANDROID + help + none + +config NET_ACTIVITY_STATS + bool "Network activity statistics tracking" + default ANDROID + help + Network activity statistics are useful for tracking wireless + modem activity on 2G, 3G, 4G wireless networks. Counts number of + transmissions and groups them in specified time buckets. + config NETWORK_SECMARK bool "Security Marking" help @@ -220,7 +234,7 @@ source "net/vmw_vsock/Kconfig" source "net/netlink/Kconfig" config RPS - boolean + boolean "RPS" depends on SMP && SYSFS && USE_GENERIC_SMP_HELPERS default y diff --git a/net/Makefile b/net/Makefile index 091e7b04f301..67d460aa1c48 100644 --- a/net/Makefile +++ b/net/Makefile @@ -70,3 +70,4 @@ obj-$(CONFIG_BATMAN_ADV) += batman-adv/ obj-$(CONFIG_NFC) += nfc/ obj-$(CONFIG_OPENVSWITCH) += openvswitch/ obj-$(CONFIG_VSOCKETS) += vmw_vsock/ +obj-$(CONFIG_NET_ACTIVITY_STATS) += activity_stats.o diff --git a/net/activity_stats.c b/net/activity_stats.c new file mode 100644 index 000000000000..4609ce2043eb --- /dev/null +++ b/net/activity_stats.c @@ -0,0 +1,119 @@ +/* net/activity_stats.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/proc_fs.h> +#include <linux/seq_file.h> +#include <linux/suspend.h> +#include <net/net_namespace.h> + +/* + * Track transmission rates in buckets (power of 2). + * 1,2,4,8...512 seconds. + * + * Buckets represent the count of network transmissions at least + * N seconds apart, where N is 1 << bucket index. + */ +#define BUCKET_MAX 10 + +/* Track network activity frequency */ +static unsigned long activity_stats[BUCKET_MAX]; +static ktime_t last_transmit; +static ktime_t suspend_time; +static DEFINE_SPINLOCK(activity_lock); + +void activity_stats_update(void) +{ + int i; + unsigned long flags; + ktime_t now; + s64 delta; + + spin_lock_irqsave(&activity_lock, flags); + now = ktime_get(); + delta = ktime_to_ns(ktime_sub(now, last_transmit)); + + for (i = BUCKET_MAX - 1; i >= 0; i--) { + /* + * Check if the time delta between network activity is within the + * minimum bucket range. + */ + if (delta < (1000000000ULL << i)) + continue; + + activity_stats[i]++; + last_transmit = now; + break; + } + spin_unlock_irqrestore(&activity_lock, flags); +} + +static int activity_stats_show(struct seq_file *m, void *v) +{ + int i; + int ret; + + seq_printf(m, "Min Bucket(sec) Count\n"); + + for (i = 0; i < BUCKET_MAX; i++) { + ret = seq_printf(m, "%15d %lu\n", 1 << i, activity_stats[i]); + if (ret) + return ret; + } + + return 0; +} + +static int activity_stats_notifier(struct notifier_block *nb, + unsigned long event, void *dummy) +{ + switch (event) { + case PM_SUSPEND_PREPARE: + suspend_time = ktime_get_real(); + break; + + case PM_POST_SUSPEND: + suspend_time = ktime_sub(ktime_get_real(), suspend_time); + last_transmit = ktime_sub(last_transmit, suspend_time); + } + + return 0; +} + +static int activity_stats_open(struct inode *inode, struct file *file) +{ + return single_open(file, activity_stats_show, PDE_DATA(inode)); +} + +static const struct file_operations activity_stats_fops = { + .open = activity_stats_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; + +static struct notifier_block activity_stats_notifier_block = { + .notifier_call = activity_stats_notifier, +}; + +static int __init activity_stats_init(void) +{ + proc_create("activity", S_IRUGO, + init_net.proc_net_stat, &activity_stats_fops); + return register_pm_notifier(&activity_stats_notifier_block); +} + +subsys_initcall(activity_stats_init); + diff --git a/net/bluetooth/af_bluetooth.c b/net/bluetooth/af_bluetooth.c index 6629cdc134dc..f7c36826f3f4 100644 --- a/net/bluetooth/af_bluetooth.c +++ b/net/bluetooth/af_bluetooth.c @@ -30,6 +30,11 @@ #include <net/bluetooth/bluetooth.h> #include <linux/proc_fs.h> +#ifndef CONFIG_BT_SOCK_DEBUG +#undef BT_DBG +#define BT_DBG(D...) +#endif + #define VERSION "2.16" /* Bluetooth sockets */ @@ -103,11 +108,40 @@ 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/bluetooth/amp.c b/net/bluetooth/amp.c index d459ed43c779..a3f3380c2095 100644 --- a/net/bluetooth/amp.c +++ b/net/bluetooth/amp.c @@ -113,7 +113,7 @@ struct hci_conn *phylink_add(struct hci_dev *hdev, struct amp_mgr *mgr, bdaddr_t *dst = mgr->l2cap_conn->dst; struct hci_conn *hcon; - hcon = hci_conn_add(hdev, AMP_LINK, dst); + hcon = hci_conn_add(hdev, AMP_LINK, 0, dst); if (!hcon) return NULL; diff --git a/net/bluetooth/hci_conn.c b/net/bluetooth/hci_conn.c index 8e7290aea8f8..c9b2d5011fe4 100644 --- a/net/bluetooth/hci_conn.c +++ b/net/bluetooth/hci_conn.c @@ -354,7 +354,8 @@ static void hci_conn_auto_accept(unsigned long arg) &conn->dst); } -struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst) +struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, + __u16 pkt_type, bdaddr_t *dst) { struct hci_conn *conn; @@ -382,14 +383,22 @@ struct hci_conn *hci_conn_add(struct hci_dev *hdev, int type, bdaddr_t *dst) conn->pkt_type = hdev->pkt_type & ACL_PTYPE_MASK; break; case SCO_LINK: - if (lmp_esco_capable(hdev)) - conn->pkt_type = (hdev->esco_type & SCO_ESCO_MASK) | - (hdev->esco_type & EDR_ESCO_MASK); - else - conn->pkt_type = hdev->pkt_type & SCO_PTYPE_MASK; - break; + if (!pkt_type) + pkt_type = SCO_ESCO_MASK; case ESCO_LINK: - conn->pkt_type = hdev->esco_type & ~EDR_ESCO_MASK; + if (!pkt_type) + pkt_type = ALL_ESCO_MASK; + if (lmp_esco_capable(hdev)) { + /* HCI Setup Synchronous Connection Command uses + reverse logic on the EDR_ESCO_MASK bits */ + conn->pkt_type = (pkt_type ^ EDR_ESCO_MASK) & + hdev->esco_type; + } else { + /* Legacy HCI Add Sco Connection Command uses a + shifted bitmask */ + conn->pkt_type = (pkt_type << 5) & hdev->pkt_type & + SCO_PTYPE_MASK; + } break; } @@ -520,7 +529,7 @@ static struct hci_conn *hci_connect_le(struct hci_dev *hdev, bdaddr_t *dst, if (le) return ERR_PTR(-EBUSY); - le = hci_conn_add(hdev, LE_LINK, dst); + le = hci_conn_add(hdev, LE_LINK, 0, dst); if (!le) return ERR_PTR(-ENOMEM); @@ -543,7 +552,7 @@ static struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst, acl = hci_conn_hash_lookup_ba(hdev, ACL_LINK, dst); if (!acl) { - acl = hci_conn_add(hdev, ACL_LINK, dst); + acl = hci_conn_add(hdev, ACL_LINK, 0, dst); if (!acl) return ERR_PTR(-ENOMEM); } @@ -561,7 +570,8 @@ static struct hci_conn *hci_connect_acl(struct hci_dev *hdev, bdaddr_t *dst, } static struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, - bdaddr_t *dst, u8 sec_level, u8 auth_type) + __u16 pkt_type, bdaddr_t *dst, + u8 sec_level, u8 auth_type) { struct hci_conn *acl; struct hci_conn *sco; @@ -572,7 +582,7 @@ static struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, sco = hci_conn_hash_lookup_ba(hdev, type, dst); if (!sco) { - sco = hci_conn_add(hdev, type, dst); + sco = hci_conn_add(hdev, type, pkt_type, dst); if (!sco) { hci_conn_drop(acl); return ERR_PTR(-ENOMEM); @@ -602,7 +612,8 @@ static struct hci_conn *hci_connect_sco(struct hci_dev *hdev, int type, } /* Create SCO, ACL or LE connection. */ -struct hci_conn *hci_connect(struct hci_dev *hdev, int type, bdaddr_t *dst, +struct hci_conn *hci_connect(struct hci_dev *hdev, int type, + __u16 pkt_type, bdaddr_t *dst, __u8 dst_type, __u8 sec_level, __u8 auth_type) { BT_DBG("%s dst %pMR type 0x%x", hdev->name, dst, type); @@ -614,7 +625,7 @@ struct hci_conn *hci_connect(struct hci_dev *hdev, int type, bdaddr_t *dst, return hci_connect_acl(hdev, dst, sec_level, auth_type); case SCO_LINK: case ESCO_LINK: - return hci_connect_sco(hdev, type, dst, sec_level, auth_type); + return hci_connect_sco(hdev, type, pkt_type, dst, sec_level, auth_type); } return ERR_PTR(-EINVAL); @@ -883,6 +894,15 @@ int hci_get_conn_list(void __user *arg) (ci + n)->out = c->out; (ci + n)->state = c->state; (ci + n)->link_mode = c->link_mode; + if (c->type == SCO_LINK) { + (ci + n)->mtu = hdev->sco_mtu; + (ci + n)->cnt = hdev->sco_cnt; + (ci + n)->pkts = hdev->sco_pkts; + } else { + (ci + n)->mtu = hdev->acl_mtu; + (ci + n)->cnt = hdev->acl_cnt; + (ci + n)->pkts = hdev->acl_pkts; + } if (++n >= req.conn_num) break; } @@ -919,6 +939,15 @@ int hci_get_conn_info(struct hci_dev *hdev, void __user *arg) ci.out = conn->out; ci.state = conn->state; ci.link_mode = conn->link_mode; + if (req.type == SCO_LINK) { + ci.mtu = hdev->sco_mtu; + ci.cnt = hdev->sco_cnt; + ci.pkts = hdev->sco_pkts; + } else { + ci.mtu = hdev->acl_mtu; + ci.cnt = hdev->acl_cnt; + ci.pkts = hdev->acl_pkts; + } } hci_dev_unlock(hdev); diff --git a/net/bluetooth/hci_event.c b/net/bluetooth/hci_event.c index 5daf7ab26710..1526fb232b3f 100644..100755 --- a/net/bluetooth/hci_event.c +++ b/net/bluetooth/hci_event.c @@ -1119,7 +1119,7 @@ static void hci_cs_create_conn(struct hci_dev *hdev, __u8 status) } } else { if (!conn) { - conn = hci_conn_add(hdev, ACL_LINK, &cp->bdaddr); + conn = hci_conn_add(hdev, ACL_LINK, 0, &cp->bdaddr); if (conn) { conn->out = true; conn->link_mode |= HCI_LM_MASTER; @@ -1748,6 +1748,15 @@ unlock: hci_conn_check_pending(hdev); } +static inline bool is_sco_active(struct hci_dev *hdev) +{ + if (hci_conn_hash_lookup_state(hdev, SCO_LINK, BT_CONNECTED) || + (hci_conn_hash_lookup_state(hdev, ESCO_LINK, + BT_CONNECTED))) + return true; + return false; +} + static void hci_conn_request_evt(struct hci_dev *hdev, struct sk_buff *skb) { struct hci_ev_conn_request *ev = (void *) skb->data; @@ -1775,7 +1784,8 @@ static void hci_conn_request_evt(struct hci_dev *hdev, struct sk_buff *skb) conn = hci_conn_hash_lookup_ba(hdev, ev->link_type, &ev->bdaddr); if (!conn) { - conn = hci_conn_add(hdev, ev->link_type, &ev->bdaddr); + /* pkt_type not yet used for incoming connections */ + conn = hci_conn_add(hdev, ev->link_type, 0, &ev->bdaddr); if (!conn) { BT_ERR("No memory for new connection"); hci_dev_unlock(hdev); @@ -1794,7 +1804,8 @@ static void hci_conn_request_evt(struct hci_dev *hdev, struct sk_buff *skb) bacpy(&cp.bdaddr, &ev->bdaddr); - if (lmp_rswitch_capable(hdev) && (mask & HCI_LM_MASTER)) + if (lmp_rswitch_capable(hdev) && ((mask & HCI_LM_MASTER) + || is_sco_active(hdev))) cp.role = 0x00; /* Become master */ else cp.role = 0x01; /* Remain slave */ @@ -2963,6 +2974,7 @@ static void hci_sync_conn_complete_evt(struct hci_dev *hdev, hci_conn_add_sysfs(conn); break; + case 0x10: /* Connection Accept Timeout */ case 0x11: /* Unsupported Feature or Parameter Value */ case 0x1c: /* SCO interval rejected */ case 0x1a: /* Unsupported Remote Feature */ @@ -3540,7 +3552,7 @@ static void hci_le_conn_complete_evt(struct hci_dev *hdev, struct sk_buff *skb) conn = hci_conn_hash_lookup_state(hdev, LE_LINK, BT_CONNECT); if (!conn) { - conn = hci_conn_add(hdev, LE_LINK, &ev->bdaddr); + conn = hci_conn_add(hdev, LE_LINK, 0, &ev->bdaddr); if (!conn) { BT_ERR("No memory for new connection"); goto unlock; diff --git a/net/bluetooth/l2cap_core.c b/net/bluetooth/l2cap_core.c index 68843a28a7af..c11a28bae844 100644 --- a/net/bluetooth/l2cap_core.c +++ b/net/bluetooth/l2cap_core.c @@ -1793,10 +1793,10 @@ int l2cap_chan_connect(struct l2cap_chan *chan, __le16 psm, u16 cid, auth_type = l2cap_get_auth_type(chan); if (chan->dcid == L2CAP_CID_LE_DATA) - hcon = hci_connect(hdev, LE_LINK, dst, dst_type, + hcon = hci_connect(hdev, LE_LINK, 0, dst, dst_type, chan->sec_level, auth_type); else - hcon = hci_connect(hdev, ACL_LINK, dst, dst_type, + hcon = hci_connect(hdev, ACL_LINK, 0, dst, dst_type, chan->sec_level, auth_type); if (IS_ERR(hcon)) { diff --git a/net/bluetooth/mgmt.c b/net/bluetooth/mgmt.c index 8208a13a9837..3e574540b2c2 100644 --- a/net/bluetooth/mgmt.c +++ b/net/bluetooth/mgmt.c @@ -2205,10 +2205,10 @@ static int pair_device(struct sock *sk, struct hci_dev *hdev, void *data, auth_type = HCI_AT_DEDICATED_BONDING_MITM; if (cp->addr.type == BDADDR_BREDR) - conn = hci_connect(hdev, ACL_LINK, &cp->addr.bdaddr, + conn = hci_connect(hdev, ACL_LINK, 0, &cp->addr.bdaddr, cp->addr.type, sec_level, auth_type); else - conn = hci_connect(hdev, LE_LINK, &cp->addr.bdaddr, + conn = hci_connect(hdev, LE_LINK, 0, &cp->addr.bdaddr, cp->addr.type, sec_level, auth_type); if (IS_ERR(conn)) { diff --git a/net/bluetooth/rfcomm/core.c b/net/bluetooth/rfcomm/core.c index 19ba192e9dbf..3ca5e40fe390 100644 --- a/net/bluetooth/rfcomm/core.c +++ b/net/bluetooth/rfcomm/core.c @@ -436,7 +436,6 @@ static int __rfcomm_dlc_close(struct rfcomm_dlc *d, int err) switch (d->state) { case BT_CONNECT: - case BT_CONFIG: if (test_and_clear_bit(RFCOMM_DEFER_SETUP, &d->flags)) { set_bit(RFCOMM_AUTH_REJECT, &d->flags); rfcomm_schedule(); diff --git a/net/bluetooth/sco.c b/net/bluetooth/sco.c index c9ae6b703c13..de9c955b247a 100644 --- a/net/bluetooth/sco.c +++ b/net/bluetooth/sco.c @@ -158,6 +158,7 @@ static int sco_connect(struct sock *sk) { bdaddr_t *src = &bt_sk(sk)->src; bdaddr_t *dst = &bt_sk(sk)->dst; + __u16 pkt_type = sco_pi(sk)->pkt_type; struct sco_conn *conn; struct hci_conn *hcon; struct hci_dev *hdev; @@ -173,11 +174,13 @@ static int sco_connect(struct sock *sk) if (lmp_esco_capable(hdev) && !disable_esco) type = ESCO_LINK; - else + else { type = SCO_LINK; + pkt_type &= SCO_ESCO_MASK; + } - hcon = hci_connect(hdev, type, dst, BDADDR_BREDR, BT_SECURITY_LOW, - HCI_AT_NO_BONDING); + hcon = hci_connect(hdev, type, pkt_type, dst, BDADDR_BREDR, + BT_SECURITY_LOW, HCI_AT_NO_BONDING); if (IS_ERR(hcon)) { err = PTR_ERR(hcon); goto done; @@ -445,17 +448,21 @@ static int sco_sock_create(struct net *net, struct socket *sock, int protocol, return 0; } -static int sco_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_len) +static int sco_sock_bind(struct socket *sock, struct sockaddr *addr, int alen) { - struct sockaddr_sco *sa = (struct sockaddr_sco *) addr; + struct sockaddr_sco sa; struct sock *sk = sock->sk; - int err = 0; + int len, err = 0; - BT_DBG("sk %p %pMR", sk, &sa->sco_bdaddr); + BT_DBG("sk %p %pMR", sk, &sa.sco_bdaddr); if (!addr || addr->sa_family != AF_BLUETOOTH) return -EINVAL; + memset(&sa, 0, sizeof(sa)); + len = min_t(unsigned int, sizeof(sa), alen); + memcpy(&sa, addr, len); + lock_sock(sk); if (sk->sk_state != BT_OPEN) { @@ -468,7 +475,8 @@ static int sco_sock_bind(struct socket *sock, struct sockaddr *addr, int addr_le goto done; } - bacpy(&bt_sk(sk)->src, &sa->sco_bdaddr); + bacpy(&bt_sk(sk)->src, &sa.sco_bdaddr); + sco_pi(sk)->pkt_type = sa.sco_pkt_type; sk->sk_state = BT_BOUND; @@ -479,26 +487,34 @@ done: static int sco_sock_connect(struct socket *sock, struct sockaddr *addr, int alen, int flags) { - struct sockaddr_sco *sa = (struct sockaddr_sco *) addr; struct sock *sk = sock->sk; - int err; + struct sockaddr_sco sa; + int len, err; BT_DBG("sk %p", sk); - if (alen < sizeof(struct sockaddr_sco) || - addr->sa_family != AF_BLUETOOTH) + if (!addr || addr->sa_family != AF_BLUETOOTH) return -EINVAL; - if (sk->sk_state != BT_OPEN && sk->sk_state != BT_BOUND) - return -EBADFD; - - if (sk->sk_type != SOCK_SEQPACKET) - return -EINVAL; + memset(&sa, 0, sizeof(sa)); + len = min_t(unsigned int, sizeof(sa), alen); + memcpy(&sa, addr, len); lock_sock(sk); + if (sk->sk_type != SOCK_SEQPACKET) { + err = -EINVAL; + goto done; + } + + if (sk->sk_state != BT_OPEN && sk->sk_state != BT_BOUND) { + err = -EBADFD; + goto done; + } + /* Set destination address and psm */ - bacpy(&bt_sk(sk)->dst, &sa->sco_bdaddr); + bacpy(&bt_sk(sk)->dst, &sa.sco_bdaddr); + sco_pi(sk)->pkt_type = sa.sco_pkt_type; err = sco_connect(sk); if (err) @@ -622,6 +638,7 @@ static int sco_sock_getname(struct socket *sock, struct sockaddr *addr, int *len bacpy(&sa->sco_bdaddr, &bt_sk(sk)->dst); else bacpy(&sa->sco_bdaddr, &bt_sk(sk)->src); + sa->sco_pkt_type = sco_pi(sk)->pkt_type; return 0; } diff --git a/net/bridge/br_device.c b/net/bridge/br_device.c index 967312803e41..239e0e84f9e6 100644 --- a/net/bridge/br_device.c +++ b/net/bridge/br_device.c @@ -41,11 +41,6 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *dev) } #endif - u64_stats_update_begin(&brstats->syncp); - brstats->tx_packets++; - brstats->tx_bytes += skb->len; - u64_stats_update_end(&brstats->syncp); - if (!br_allowed_ingress(br, br_get_vlan_info(br), skb, &vid)) goto out; @@ -54,6 +49,12 @@ netdev_tx_t br_dev_xmit(struct sk_buff *skb, struct net_device *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 (is_broadcast_ether_addr(dest)) br_flood_deliver(br, skb); else if (is_multicast_ether_addr(dest)) { diff --git a/net/core/fib_rules.c b/net/core/fib_rules.c index 55e08e2de3a1..da78f5c6d295 100644 --- a/net/core/fib_rules.c +++ b/net/core/fib_rules.c @@ -31,6 +31,8 @@ int fib_default_rule_add(struct fib_rules_ops *ops, r->pref = pref; r->table = table; r->flags = flags; + r->uid_start = INVALID_UID; + r->uid_end = INVALID_UID; r->fr_net = hold_net(ops->fro_net); /* The lock is not required here, the list in unreacheable @@ -179,6 +181,23 @@ void fib_rules_unregister(struct fib_rules_ops *ops) } EXPORT_SYMBOL_GPL(fib_rules_unregister); +static inline kuid_t fib_nl_uid(struct nlattr *nla) +{ + return make_kuid(current_user_ns(), nla_get_u32(nla)); +} + +static int nla_put_uid(struct sk_buff *skb, int idx, kuid_t uid) +{ + return nla_put_u32(skb, idx, from_kuid_munged(current_user_ns(), uid)); +} + +static int fib_uid_range_match(struct flowi *fl, struct fib_rule *rule) +{ + return (!uid_valid(rule->uid_start) && !uid_valid(rule->uid_end)) || + (uid_gte(fl->flowi_uid, rule->uid_start) && + uid_lte(fl->flowi_uid, rule->uid_end)); +} + static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, struct flowi *fl, int flags) { @@ -193,6 +212,9 @@ static int fib_rule_match(struct fib_rule *rule, struct fib_rules_ops *ops, if ((rule->mark ^ fl->flowi_mark) & rule->mark_mask) goto out; + if (!fib_uid_range_match(fl, rule)) + goto out; + ret = ops->match(rule, fl, flags); out: return (rule->flags & FIB_RULE_INVERT) ? !ret : ret; @@ -363,6 +385,19 @@ static int fib_nl_newrule(struct sk_buff *skb, struct nlmsghdr* nlh) } else if (rule->action == FR_ACT_GOTO) goto errout_free; + /* UID start and end must either both be valid or both unspecified. */ + rule->uid_start = rule->uid_end = INVALID_UID; + if (tb[FRA_UID_START] || tb[FRA_UID_END]) { + if (tb[FRA_UID_START] && tb[FRA_UID_END]) { + rule->uid_start = fib_nl_uid(tb[FRA_UID_START]); + rule->uid_end = fib_nl_uid(tb[FRA_UID_END]); + } + if (!uid_valid(rule->uid_start) || + !uid_valid(rule->uid_end) || + !uid_lte(rule->uid_start, rule->uid_end)) + goto errout_free; + } + err = ops->configure(rule, skb, frh, tb); if (err < 0) goto errout_free; @@ -469,6 +504,14 @@ static int fib_nl_delrule(struct sk_buff *skb, struct nlmsghdr* nlh) (rule->mark_mask != nla_get_u32(tb[FRA_FWMASK]))) continue; + if (tb[FRA_UID_START] && + !uid_eq(rule->uid_start, fib_nl_uid(tb[FRA_UID_START]))) + continue; + + if (tb[FRA_UID_END] && + !uid_eq(rule->uid_end, fib_nl_uid(tb[FRA_UID_END]))) + continue; + if (!ops->compare(rule, frh, tb)) continue; @@ -525,7 +568,9 @@ static inline size_t fib_rule_nlmsg_size(struct fib_rules_ops *ops, + nla_total_size(4) /* FRA_PRIORITY */ + nla_total_size(4) /* FRA_TABLE */ + nla_total_size(4) /* FRA_FWMARK */ - + nla_total_size(4); /* FRA_FWMASK */ + + nla_total_size(4) /* FRA_FWMASK */ + + nla_total_size(4) /* FRA_UID_START */ + + nla_total_size(4); /* FRA_UID_END */ if (ops->nlmsg_payload) payload += ops->nlmsg_payload(rule); @@ -579,7 +624,11 @@ static int fib_nl_fill_rule(struct sk_buff *skb, struct fib_rule *rule, ((rule->mark_mask || rule->mark) && nla_put_u32(skb, FRA_FWMASK, rule->mark_mask)) || (rule->target && - nla_put_u32(skb, FRA_GOTO, rule->target))) + nla_put_u32(skb, FRA_GOTO, rule->target)) || + (uid_valid(rule->uid_start) && + nla_put_uid(skb, FRA_UID_START, rule->uid_start)) || + (uid_valid(rule->uid_end) && + nla_put_uid(skb, FRA_UID_END, rule->uid_end))) goto nla_put_failure; if (ops->fill(rule, skb, frh) < 0) goto nla_put_failure; diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile index 089cb9f36387..5a9af0a9b0e6 100644 --- a/net/ipv4/Makefile +++ b/net/ipv4/Makefile @@ -15,6 +15,7 @@ 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 c4adc319cc2e..f022e0e97dc5 100644 --- a/net/ipv4/af_inet.c +++ b/net/ipv4/af_inet.c @@ -119,6 +119,19 @@ #include <linux/mroute.h> #endif +#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. @@ -284,6 +297,9 @@ static int inet_create(struct net *net, struct socket *sock, int protocol, int try_loading_module = 0; int err; + if (!current_has_network()) + return -EACCES; + if (unlikely(!inet_ehash_secret)) if (sock->type != SOCK_RAW && sock->type != SOCK_DGRAM) build_ehash_secret(); @@ -336,8 +352,7 @@ 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; @@ -905,6 +920,7 @@ int inet_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) case SIOCSIFPFLAGS: case SIOCGIFPFLAGS: case SIOCSIFFLAGS: + case SIOCKILLADDR: err = devinet_ioctl(net, cmd, (void __user *)arg); break; default: diff --git a/net/ipv4/devinet.c b/net/ipv4/devinet.c index e40eef4ac697..b151e0ac7f27 100644 --- a/net/ipv4/devinet.c +++ b/net/ipv4/devinet.c @@ -59,6 +59,7 @@ #include <net/arp.h> #include <net/ip.h> +#include <net/tcp.h> #include <net/route.h> #include <net/ip_fib.h> #include <net/rtnetlink.h> @@ -918,6 +919,7 @@ int devinet_ioctl(struct net *net, unsigned int cmd, void __user *arg) case SIOCSIFBRDADDR: /* Set the broadcast address */ case SIOCSIFDSTADDR: /* Set the destination address */ case SIOCSIFNETMASK: /* Set the netmask for the interface */ + case SIOCKILLADDR: /* Nuke all sockets on this address */ ret = -EPERM; if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) goto out; @@ -969,7 +971,8 @@ int devinet_ioctl(struct net *net, unsigned int cmd, void __user *arg) } ret = -EADDRNOTAVAIL; - if (!ifa && cmd != SIOCSIFADDR && cmd != SIOCSIFFLAGS) + if (!ifa && cmd != SIOCSIFADDR && cmd != SIOCSIFFLAGS + && cmd != SIOCKILLADDR) goto done; switch (cmd) { @@ -1096,6 +1099,9 @@ int devinet_ioctl(struct net *net, unsigned int cmd, void __user *arg) inet_insert_ifa(ifa); } break; + case SIOCKILLADDR: /* Nuke all connections on this address */ + ret = tcp_nuke_addr(net, (struct sockaddr *) sin); + break; } done: rtnl_unlock(); diff --git a/net/ipv4/fib_frontend.c b/net/ipv4/fib_frontend.c index 4556cd25acde..4a7cf03a192c 100644 --- a/net/ipv4/fib_frontend.c +++ b/net/ipv4/fib_frontend.c @@ -250,7 +250,7 @@ static int __fib_validate_source(struct sk_buff *skb, __be32 src, __be32 dst, bool dev_match; fl4.flowi4_oif = 0; - fl4.flowi4_iif = oif; + fl4.flowi4_iif = oif ? : LOOPBACK_IFINDEX; fl4.daddr = src; fl4.saddr = dst; fl4.flowi4_tos = tos; @@ -531,6 +531,7 @@ const struct nla_policy rtm_ipv4_policy[RTA_MAX + 1] = { [RTA_METRICS] = { .type = NLA_NESTED }, [RTA_MULTIPATH] = { .len = sizeof(struct rtnexthop) }, [RTA_FLOW] = { .type = NLA_U32 }, + [RTA_UID] = { .type = NLA_U32 }, }; static int rtm_to_fib_config(struct net *net, struct sk_buff *skb, diff --git a/net/ipv4/fib_semantics.c b/net/ipv4/fib_semantics.c index bc773a10dca6..d00d8f9650e9 100644 --- a/net/ipv4/fib_semantics.c +++ b/net/ipv4/fib_semantics.c @@ -629,6 +629,7 @@ static int fib_check_nh(struct fib_config *cfg, struct fib_info *fi, .daddr = nh->nh_gw, .flowi4_scope = cfg->fc_scope + 1, .flowi4_oif = nh->nh_oif, + .flowi4_iif = LOOPBACK_IFINDEX, }; /* It is not necessary, but requires a bit of thinking */ diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c index ea78ef5ac352..5af8781b65e1 100644 --- a/net/ipv4/icmp.c +++ b/net/ipv4/icmp.c @@ -337,6 +337,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) struct sock *sk; struct inet_sock *inet; __be32 daddr, saddr; + u32 mark = IP4_REPLY_MARK(net, skb->mark); if (ip_options_echo(&icmp_param->replyopts.opt.opt, skb)) return; @@ -349,6 +350,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) icmp_param->data.icmph.checksum = 0; inet->tos = ip_hdr(skb)->tos; + sk->sk_mark = mark; daddr = ipc.addr = ip_hdr(skb)->saddr; saddr = fib_compute_spec_dst(skb); ipc.opt = NULL; @@ -361,6 +363,7 @@ static void icmp_reply(struct icmp_bxm *icmp_param, struct sk_buff *skb) memset(&fl4, 0, sizeof(fl4)); fl4.daddr = daddr; fl4.saddr = saddr; + fl4.flowi4_mark = mark; fl4.flowi4_tos = RT_TOS(ip_hdr(skb)->tos); fl4.flowi4_proto = IPPROTO_ICMP; security_skb_classify_flow(skb, flowi4_to_flowi(&fl4)); @@ -379,7 +382,7 @@ static struct rtable *icmp_route_lookup(struct net *net, struct flowi4 *fl4, struct sk_buff *skb_in, const struct iphdr *iph, - __be32 saddr, u8 tos, + __be32 saddr, u8 tos, u32 mark, int type, int code, struct icmp_bxm *param) { @@ -391,6 +394,7 @@ static struct rtable *icmp_route_lookup(struct net *net, fl4->daddr = (param->replyopts.opt.opt.srr ? param->replyopts.opt.opt.faddr : iph->saddr); fl4->saddr = saddr; + fl4->flowi4_mark = mark; fl4->flowi4_tos = RT_TOS(tos); fl4->flowi4_proto = IPPROTO_ICMP; fl4->fl4_icmp_type = type; @@ -488,6 +492,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) struct flowi4 fl4; __be32 saddr; u8 tos; + u32 mark; struct net *net; struct sock *sk; @@ -584,6 +589,7 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) tos = icmp_pointers[type].error ? ((iph->tos & IPTOS_TOS_MASK) | IPTOS_PREC_INTERNETCONTROL) : iph->tos; + mark = IP4_REPLY_MARK(net, skb_in->mark); if (ip_options_echo(&icmp_param.replyopts.opt.opt, skb_in)) goto out_unlock; @@ -600,11 +606,12 @@ void icmp_send(struct sk_buff *skb_in, int type, int code, __be32 info) icmp_param.skb = skb_in; icmp_param.offset = skb_network_offset(skb_in); inet_sk(sk)->tos = tos; + sk->sk_mark = mark; ipc.addr = iph->saddr; ipc.opt = &icmp_param.replyopts.opt; ipc.tx_flags = 0; - rt = icmp_route_lookup(net, &fl4, skb_in, iph, saddr, tos, + rt = icmp_route_lookup(net, &fl4, skb_in, iph, saddr, tos, mark, type, code, &icmp_param); if (IS_ERR(rt)) goto out_unlock; @@ -937,7 +944,8 @@ error: void icmp_err(struct sk_buff *skb, u32 info) { struct iphdr *iph = (struct iphdr *)skb->data; - struct icmphdr *icmph = (struct icmphdr *)(skb->data+(iph->ihl<<2)); + int offset = iph->ihl<<2; + struct icmphdr *icmph = (struct icmphdr *)(skb->data + offset); int type = icmp_hdr(skb)->type; int code = icmp_hdr(skb)->code; struct net *net = dev_net(skb->dev); @@ -947,7 +955,7 @@ void icmp_err(struct sk_buff *skb, u32 info) * triggered by ICMP_ECHOREPLY which sent from kernel. */ if (icmph->type != ICMP_ECHOREPLY) { - ping_err(skb, info); + ping_err(skb, offset, info); return; } diff --git a/net/ipv4/inet_connection_sock.c b/net/ipv4/inet_connection_sock.c index 6acb541c9091..6dfec2f18214 100644 --- a/net/ipv4/inet_connection_sock.c +++ b/net/ipv4/inet_connection_sock.c @@ -417,12 +417,13 @@ struct dst_entry *inet_csk_route_req(struct sock *sk, struct net *net = sock_net(sk); int flags = inet_sk_flowi_flags(sk); - flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark, + flowi4_init_output(fl4, sk->sk_bound_dev_if, ireq->ir_mark, RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, sk->sk_protocol, flags, (opt && opt->opt.srr) ? opt->opt.faddr : ireq->rmt_addr, - ireq->loc_addr, ireq->rmt_port, inet_sk(sk)->inet_sport); + ireq->loc_addr, ireq->rmt_port, inet_sk(sk)->inet_sport, + sock_i_uid(sk)); security_req_classify_flow(req, flowi4_to_flowi(fl4)); rt = ip_route_output_flow(net, fl4, sk); if (IS_ERR(rt)) @@ -454,11 +455,12 @@ struct dst_entry *inet_csk_route_child_sock(struct sock *sk, rcu_read_lock(); opt = rcu_dereference(newinet->inet_opt); - flowi4_init_output(fl4, sk->sk_bound_dev_if, sk->sk_mark, + flowi4_init_output(fl4, sk->sk_bound_dev_if, inet_rsk(req)->ir_mark, RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, sk->sk_protocol, inet_sk_flowi_flags(sk), (opt && opt->opt.srr) ? opt->opt.faddr : ireq->rmt_addr, - ireq->loc_addr, ireq->rmt_port, inet_sk(sk)->inet_sport); + ireq->loc_addr, ireq->rmt_port, inet_sk(sk)->inet_sport, + sock_i_uid(sk)); security_req_classify_flow(req, flowi4_to_flowi(fl4)); rt = ip_route_output_flow(net, fl4, sk); if (IS_ERR(rt)) @@ -688,6 +690,8 @@ struct sock *inet_csk_clone_lock(const struct sock *sk, inet_sk(newsk)->inet_sport = inet_rsk(req)->loc_port; newsk->sk_write_space = sk_stream_write_space; + newsk->sk_mark = inet_rsk(req)->ir_mark; + newicsk->icsk_retransmits = 0; newicsk->icsk_backoff = 0; newicsk->icsk_probes_out = 0; diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c index 57e745086302..47d236156f0f 100644 --- a/net/ipv4/ip_output.c +++ b/net/ipv4/ip_output.c @@ -1482,12 +1482,14 @@ void ip_send_unicast_reply(struct sock *sk, struct sk_buff *skb, __be32 daddr, daddr = replyopts.opt.opt.faddr; } - flowi4_init_output(&fl4, arg->bound_dev_if, 0, + flowi4_init_output(&fl4, arg->bound_dev_if, + IP4_REPLY_MARK(net, skb->mark), RT_TOS(arg->tos), 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/ipmr.c b/net/ipv4/ipmr.c index 56d079b63ad3..9fc74ec238de 100644 --- a/net/ipv4/ipmr.c +++ b/net/ipv4/ipmr.c @@ -454,7 +454,7 @@ static netdev_tx_t reg_vif_xmit(struct sk_buff *skb, struct net_device *dev) struct mr_table *mrt; struct flowi4 fl4 = { .flowi4_oif = dev->ifindex, - .flowi4_iif = skb->skb_iif, + .flowi4_iif = skb->skb_iif ? : LOOPBACK_IFINDEX, .flowi4_mark = skb->mark, }; int err; diff --git a/net/ipv4/netfilter/Kconfig b/net/ipv4/netfilter/Kconfig index e7916c193932..23dfd4a312ce 100644 --- a/net/ipv4/netfilter/Kconfig +++ b/net/ipv4/netfilter/Kconfig @@ -110,6 +110,18 @@ config IP_NF_TARGET_REJECT To compile it as a module, choose M here. If unsure, say N. +config IP_NF_TARGET_REJECT_SKERR + bool "Force socket error when rejecting with icmp*" + depends on IP_NF_TARGET_REJECT + default n + help + This option enables turning a "--reject-with icmp*" into a matching + socket error also. + The REJECT target normally allows sending an ICMP message. But it + leaves the local socket unaware of any ingress rejects. + + If unsure, say N. + config IP_NF_TARGET_ULOG tristate "ULOG target support" default m if NETFILTER_ADVANCED=n diff --git a/net/ipv4/netfilter/ipt_REJECT.c b/net/ipv4/netfilter/ipt_REJECT.c index 04b18c1ac345..452e8a587c34 100644 --- a/net/ipv4/netfilter/ipt_REJECT.c +++ b/net/ipv4/netfilter/ipt_REJECT.c @@ -129,6 +129,14 @@ static void send_reset(struct sk_buff *oldskb, int hook) static inline void send_unreach(struct sk_buff *skb_in, int code) { icmp_send(skb_in, ICMP_DEST_UNREACH, code, 0); +#ifdef CONFIG_IP_NF_TARGET_REJECT_SKERR + if (skb_in->sk) { + skb_in->sk->sk_err = icmp_err_convert[code].errno; + skb_in->sk->sk_error_report(skb_in->sk); + pr_debug("ipt_REJECT: sk_err=%d for skb=%p sk=%p\n", + skb_in->sk->sk_err, skb_in, skb_in->sk); + } +#endif } static unsigned int diff --git a/net/ipv4/netfilter/ipt_rpfilter.c b/net/ipv4/netfilter/ipt_rpfilter.c index c49dcd0284a0..4bfaedf9b34e 100644 --- a/net/ipv4/netfilter/ipt_rpfilter.c +++ b/net/ipv4/netfilter/ipt_rpfilter.c @@ -89,11 +89,8 @@ static bool rpfilter_mt(const struct sk_buff *skb, struct xt_action_param *par) if (ipv4_is_multicast(iph->daddr)) { if (ipv4_is_zeronet(iph->saddr)) return ipv4_is_local_multicast(iph->daddr) ^ invert; - flow.flowi4_iif = 0; - } else { - flow.flowi4_iif = LOOPBACK_IFINDEX; } - + flow.flowi4_iif = LOOPBACK_IFINDEX; flow.daddr = iph->saddr; flow.saddr = rpfilter_get_saddr(iph->daddr); flow.flowi4_oif = 0; diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c index 459b957104a8..47d1c7e10d54 100644 --- a/net/ipv4/ping.c +++ b/net/ipv4/ping.c @@ -33,7 +33,6 @@ #include <linux/netdevice.h> #include <net/snmp.h> #include <net/ip.h> -#include <net/ipv6.h> #include <net/icmp.h> #include <net/protocol.h> #include <linux/skbuff.h> @@ -46,8 +45,18 @@ #include <net/inet_common.h> #include <net/checksum.h> +#if IS_ENABLED(CONFIG_IPV6) +#include <linux/in6.h> +#include <linux/icmpv6.h> +#include <net/addrconf.h> +#include <net/ipv6.h> +#include <net/transp_v6.h> +#endif -static struct ping_table ping_table; + +struct ping_table ping_table; +struct pingv6_ops pingv6_ops; +EXPORT_SYMBOL_GPL(pingv6_ops); static u16 ping_port_rover; @@ -58,6 +67,7 @@ static inline int ping_hashfn(struct net *net, unsigned int num, unsigned int ma pr_debug("hash(%d) = %d\n", num, res); return res; } +EXPORT_SYMBOL_GPL(ping_hash); static inline struct hlist_nulls_head *ping_hashslot(struct ping_table *table, struct net *net, unsigned int num) @@ -65,7 +75,7 @@ static inline struct hlist_nulls_head *ping_hashslot(struct ping_table *table, return &table->hash[ping_hashfn(net, num, PING_HTABLE_MASK)]; } -static int ping_v4_get_port(struct sock *sk, unsigned short ident) +int ping_get_port(struct sock *sk, unsigned short ident) { struct hlist_nulls_node *node; struct hlist_nulls_head *hlist; @@ -103,6 +113,10 @@ next_port: ping_portaddr_for_each_entry(sk2, node, hlist) { isk2 = inet_sk(sk2); + /* BUG? Why is this reuse and not reuseaddr? ping.c + * doesn't turn off SO_REUSEADDR, and it doesn't expect + * that other ping processes can steal its packets. + */ if ((isk2->inet_num == ident) && (sk2 != sk) && (!sk2->sk_reuse || !sk->sk_reuse)) @@ -125,17 +139,18 @@ fail: write_unlock_bh(&ping_table.lock); return 1; } +EXPORT_SYMBOL_GPL(ping_get_port); -static void ping_v4_hash(struct sock *sk) +void ping_hash(struct sock *sk) { - pr_debug("ping_v4_hash(sk->port=%u)\n", inet_sk(sk)->inet_num); + pr_debug("ping_hash(sk->port=%u)\n", inet_sk(sk)->inet_num); BUG(); /* "Please do not press this button again." */ } -static void ping_v4_unhash(struct sock *sk) +void ping_unhash(struct sock *sk) { struct inet_sock *isk = inet_sk(sk); - pr_debug("ping_v4_unhash(isk=%p,isk->num=%u)\n", isk, isk->inet_num); + pr_debug("ping_unhash(isk=%p,isk->num=%u)\n", isk, isk->inet_num); if (sk_hashed(sk)) { write_lock_bh(&ping_table.lock); hlist_nulls_del(&sk->sk_nulls_node); @@ -147,31 +162,63 @@ static void ping_v4_unhash(struct sock *sk) write_unlock_bh(&ping_table.lock); } } +EXPORT_SYMBOL_GPL(ping_unhash); -static struct sock *ping_v4_lookup(struct net *net, __be32 saddr, __be32 daddr, - u16 ident, int dif) +static struct sock *ping_lookup(struct net *net, struct sk_buff *skb, u16 ident) { struct hlist_nulls_head *hslot = ping_hashslot(&ping_table, net, ident); struct sock *sk = NULL; struct inet_sock *isk; struct hlist_nulls_node *hnode; + int dif = skb->dev->ifindex; + + if (skb->protocol == htons(ETH_P_IP)) { + pr_debug("try to find: num = %d, daddr = %pI4, dif = %d\n", + (int)ident, &ip_hdr(skb)->daddr, dif); +#if IS_ENABLED(CONFIG_IPV6) + } else if (skb->protocol == htons(ETH_P_IPV6)) { + pr_debug("try to find: num = %d, daddr = %pI6c, dif = %d\n", + (int)ident, &ipv6_hdr(skb)->daddr, dif); +#endif + } - pr_debug("try to find: num = %d, daddr = %pI4, dif = %d\n", - (int)ident, &daddr, dif); read_lock_bh(&ping_table.lock); ping_portaddr_for_each_entry(sk, hnode, hslot) { isk = inet_sk(sk); - pr_debug("found: %p: num = %d, daddr = %pI4, dif = %d\n", sk, - (int)isk->inet_num, &isk->inet_rcv_saddr, - sk->sk_bound_dev_if); - pr_debug("iterate\n"); if (isk->inet_num != ident) continue; - if (isk->inet_rcv_saddr && isk->inet_rcv_saddr != daddr) + + if (skb->protocol == htons(ETH_P_IP) && + sk->sk_family == AF_INET) { + pr_debug("found: %p: num=%d, daddr=%pI4, dif=%d\n", sk, + (int) isk->inet_num, &isk->inet_rcv_saddr, + sk->sk_bound_dev_if); + + if (isk->inet_rcv_saddr && + isk->inet_rcv_saddr != ip_hdr(skb)->daddr) + continue; +#if IS_ENABLED(CONFIG_IPV6) + } else if (skb->protocol == htons(ETH_P_IPV6) && + sk->sk_family == AF_INET6) { + struct ipv6_pinfo *np = inet6_sk(sk); + + pr_debug("found: %p: num=%d, daddr=%pI6c, dif=%d\n", sk, + (int) isk->inet_num, + &inet6_sk(sk)->rcv_saddr, + sk->sk_bound_dev_if); + + if (!ipv6_addr_any(&np->rcv_saddr) && + !ipv6_addr_equal(&np->rcv_saddr, + &ipv6_hdr(skb)->daddr)) + continue; +#endif + } else { continue; + } + if (sk->sk_bound_dev_if && sk->sk_bound_dev_if != dif) continue; @@ -201,7 +248,7 @@ static void inet_get_ping_group_range_net(struct net *net, kgid_t *low, } -static int ping_init_sock(struct sock *sk) +int ping_init_sock(struct sock *sk) { struct net *net = sock_net(sk); kgid_t group = current_egid(); @@ -210,6 +257,9 @@ static int ping_init_sock(struct sock *sk) kgid_t low, high; int ret = 0; + if (sk->sk_family == AF_INET6) + inet6_sk(sk)->ipv6only = 1; + inet_get_ping_group_range_net(net, &low, &high); if (gid_lte(low, group) && gid_lte(group, high)) return 0; @@ -233,8 +283,9 @@ out_release_group: put_group_info(group_info); return ret; } +EXPORT_SYMBOL_GPL(ping_init_sock); -static void ping_close(struct sock *sk, long timeout) +void ping_close(struct sock *sk, long timeout) { pr_debug("ping_close(sk=%p,sk->num=%u)\n", inet_sk(sk), inet_sk(sk)->inet_num); @@ -242,36 +293,130 @@ static void ping_close(struct sock *sk, long timeout) sk_common_release(sk); } +EXPORT_SYMBOL_GPL(ping_close); + +/* Checks the bind address and possibly modifies sk->sk_bound_dev_if. */ +int ping_check_bind_addr(struct sock *sk, struct inet_sock *isk, + struct sockaddr *uaddr, int addr_len) { + struct net *net = sock_net(sk); + if (sk->sk_family == AF_INET) { + struct sockaddr_in *addr = (struct sockaddr_in *) uaddr; + int chk_addr_ret; + + if (addr_len < sizeof(*addr)) + return -EINVAL; + + if (addr->sin_family != AF_INET && + !(addr->sin_family == AF_UNSPEC && + addr->sin_addr.s_addr == htonl(INADDR_ANY))) + return -EAFNOSUPPORT; + + pr_debug("ping_check_bind_addr(sk=%p,addr=%pI4,port=%d)\n", + sk, &addr->sin_addr.s_addr, ntohs(addr->sin_port)); + chk_addr_ret = inet_addr_type(net, addr->sin_addr.s_addr); + + if (addr->sin_addr.s_addr == htonl(INADDR_ANY)) + chk_addr_ret = RTN_LOCAL; + + if ((sysctl_ip_nonlocal_bind == 0 && + isk->freebind == 0 && isk->transparent == 0 && + chk_addr_ret != RTN_LOCAL) || + chk_addr_ret == RTN_MULTICAST || + chk_addr_ret == RTN_BROADCAST) + return -EADDRNOTAVAIL; + +#if IS_ENABLED(CONFIG_IPV6) + } else if (sk->sk_family == AF_INET6) { + struct sockaddr_in6 *addr = (struct sockaddr_in6 *) uaddr; + int addr_type, scoped, has_addr; + struct net_device *dev = NULL; + + if (addr_len < sizeof(*addr)) + return -EINVAL; + + if (addr->sin6_family != AF_INET6) + return -EAFNOSUPPORT; + + pr_debug("ping_check_bind_addr(sk=%p,addr=%pI6c,port=%d)\n", + sk, addr->sin6_addr.s6_addr, ntohs(addr->sin6_port)); + + addr_type = ipv6_addr_type(&addr->sin6_addr); + scoped = __ipv6_addr_needs_scope_id(addr_type); + if ((addr_type != IPV6_ADDR_ANY && + !(addr_type & IPV6_ADDR_UNICAST)) || + (scoped && !addr->sin6_scope_id)) + return -EINVAL; + + rcu_read_lock(); + if (addr->sin6_scope_id) { + dev = dev_get_by_index_rcu(net, addr->sin6_scope_id); + if (!dev) { + rcu_read_unlock(); + return -ENODEV; + } + } + has_addr = pingv6_ops.ipv6_chk_addr(net, &addr->sin6_addr, dev, + scoped); + rcu_read_unlock(); + + if (!(isk->freebind || isk->transparent || has_addr || + addr_type == IPV6_ADDR_ANY)) + return -EADDRNOTAVAIL; + + if (scoped) + sk->sk_bound_dev_if = addr->sin6_scope_id; +#endif + } else { + return -EAFNOSUPPORT; + } + return 0; +} + +void ping_set_saddr(struct sock *sk, struct sockaddr *saddr) +{ + if (saddr->sa_family == AF_INET) { + struct inet_sock *isk = inet_sk(sk); + struct sockaddr_in *addr = (struct sockaddr_in *) saddr; + isk->inet_rcv_saddr = isk->inet_saddr = addr->sin_addr.s_addr; +#if IS_ENABLED(CONFIG_IPV6) + } else if (saddr->sa_family == AF_INET6) { + struct sockaddr_in6 *addr = (struct sockaddr_in6 *) saddr; + struct ipv6_pinfo *np = inet6_sk(sk); + np->rcv_saddr = np->saddr = addr->sin6_addr; +#endif + } +} + +void ping_clear_saddr(struct sock *sk, int dif) +{ + sk->sk_bound_dev_if = dif; + if (sk->sk_family == AF_INET) { + struct inet_sock *isk = inet_sk(sk); + isk->inet_rcv_saddr = isk->inet_saddr = 0; +#if IS_ENABLED(CONFIG_IPV6) + } else if (sk->sk_family == AF_INET6) { + struct ipv6_pinfo *np = inet6_sk(sk); + memset(&np->rcv_saddr, 0, sizeof(np->rcv_saddr)); + memset(&np->saddr, 0, sizeof(np->saddr)); +#endif + } +} /* * We need our own bind because there are no privileged id's == local ports. * Moreover, we don't allow binding to multi- and broadcast addresses. */ -static int ping_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) +int ping_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) { - struct sockaddr_in *addr = (struct sockaddr_in *)uaddr; struct inet_sock *isk = inet_sk(sk); unsigned short snum; - int chk_addr_ret; int err; + int dif = sk->sk_bound_dev_if; - if (addr_len < sizeof(struct sockaddr_in)) - return -EINVAL; - - pr_debug("ping_v4_bind(sk=%p,sa_addr=%08x,sa_port=%d)\n", - sk, addr->sin_addr.s_addr, ntohs(addr->sin_port)); - - chk_addr_ret = inet_addr_type(sock_net(sk), addr->sin_addr.s_addr); - if (addr->sin_addr.s_addr == htonl(INADDR_ANY)) - chk_addr_ret = RTN_LOCAL; - - if ((sysctl_ip_nonlocal_bind == 0 && - isk->freebind == 0 && isk->transparent == 0 && - chk_addr_ret != RTN_LOCAL) || - chk_addr_ret == RTN_MULTICAST || - chk_addr_ret == RTN_BROADCAST) - return -EADDRNOTAVAIL; + err = ping_check_bind_addr(sk, isk, uaddr, addr_len); + if (err) + return err; lock_sock(sk); @@ -280,42 +425,50 @@ static int ping_bind(struct sock *sk, struct sockaddr *uaddr, int addr_len) goto out; err = -EADDRINUSE; - isk->inet_rcv_saddr = isk->inet_saddr = addr->sin_addr.s_addr; - snum = ntohs(addr->sin_port); - if (ping_v4_get_port(sk, snum) != 0) { - isk->inet_saddr = isk->inet_rcv_saddr = 0; + ping_set_saddr(sk, uaddr); + snum = ntohs(((struct sockaddr_in *)uaddr)->sin_port); + if (ping_get_port(sk, snum) != 0) { + ping_clear_saddr(sk, dif); goto out; } - pr_debug("after bind(): num = %d, daddr = %pI4, dif = %d\n", + pr_debug("after bind(): num = %d, dif = %d\n", (int)isk->inet_num, - &isk->inet_rcv_saddr, (int)sk->sk_bound_dev_if); err = 0; - if (isk->inet_rcv_saddr) + if ((sk->sk_family == AF_INET && isk->inet_rcv_saddr) || + (sk->sk_family == AF_INET6 && + !ipv6_addr_any(&inet6_sk(sk)->rcv_saddr))) sk->sk_userlocks |= SOCK_BINDADDR_LOCK; + if (snum) sk->sk_userlocks |= SOCK_BINDPORT_LOCK; isk->inet_sport = htons(isk->inet_num); isk->inet_daddr = 0; isk->inet_dport = 0; + +#if IS_ENABLED(CONFIG_IPV6) + if (sk->sk_family == AF_INET6) + memset(&inet6_sk(sk)->daddr, 0, sizeof(inet6_sk(sk)->daddr)); +#endif + sk_dst_reset(sk); out: release_sock(sk); pr_debug("ping_v4_bind -> %d\n", err); return err; } +EXPORT_SYMBOL_GPL(ping_bind); /* * Is this a supported type of ICMP message? */ -static inline int ping_supported(int type, int code) +static inline int ping_supported(int family, int type, int code) { - if (type == ICMP_ECHO && code == 0) - return 1; - return 0; + return (family == AF_INET && type == ICMP_ECHO && code == 0) || + (family == AF_INET6 && type == ICMPV6_ECHO_REQUEST && code == 0); } /* @@ -323,30 +476,42 @@ static inline int ping_supported(int type, int code) * sort of error condition. */ -static int ping_queue_rcv_skb(struct sock *sk, struct sk_buff *skb); - -void ping_err(struct sk_buff *skb, u32 info) +void ping_err(struct sk_buff *skb, int offset, u32 info) { - struct iphdr *iph = (struct iphdr *)skb->data; - struct icmphdr *icmph = (struct icmphdr *)(skb->data+(iph->ihl<<2)); + int family; + struct icmphdr *icmph; struct inet_sock *inet_sock; - int type = icmp_hdr(skb)->type; - int code = icmp_hdr(skb)->code; + int type; + int code; struct net *net = dev_net(skb->dev); struct sock *sk; int harderr; int err; + if (skb->protocol == htons(ETH_P_IP)) { + family = AF_INET; + type = icmp_hdr(skb)->type; + code = icmp_hdr(skb)->code; + icmph = (struct icmphdr *)(skb->data + offset); + } else if (skb->protocol == htons(ETH_P_IPV6)) { + family = AF_INET6; + type = icmp6_hdr(skb)->icmp6_type; + code = icmp6_hdr(skb)->icmp6_code; + icmph = (struct icmphdr *) (skb->data + offset); + } else { + BUG(); + } + /* We assume the packet has already been checked by icmp_unreach */ - if (!ping_supported(icmph->type, icmph->code)) + if (!ping_supported(family, icmph->type, icmph->code)) return; - pr_debug("ping_err(type=%04x,code=%04x,id=%04x,seq=%04x)\n", type, - code, ntohs(icmph->un.echo.id), ntohs(icmph->un.echo.sequence)); + pr_debug("ping_err(proto=0x%x,type=%d,code=%d,id=%04x,seq=%04x)\n", + skb->protocol, type, code, ntohs(icmph->un.echo.id), + ntohs(icmph->un.echo.sequence)); - sk = ping_v4_lookup(net, iph->daddr, iph->saddr, - ntohs(icmph->un.echo.id), skb->dev->ifindex); + sk = ping_lookup(net, skb, ntohs(icmph->un.echo.id)); if (sk == NULL) { pr_debug("no socket, dropping\n"); return; /* No socket for error */ @@ -357,72 +522,83 @@ void ping_err(struct sk_buff *skb, u32 info) harderr = 0; inet_sock = inet_sk(sk); - switch (type) { - default: - case ICMP_TIME_EXCEEDED: - err = EHOSTUNREACH; - break; - case ICMP_SOURCE_QUENCH: - /* This is not a real error but ping wants to see it. - * Report it with some fake errno. */ - err = EREMOTEIO; - break; - case ICMP_PARAMETERPROB: - err = EPROTO; - harderr = 1; - break; - case ICMP_DEST_UNREACH: - if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */ - ipv4_sk_update_pmtu(skb, sk, info); - if (inet_sock->pmtudisc != IP_PMTUDISC_DONT) { - err = EMSGSIZE; - harderr = 1; - break; + if (skb->protocol == htons(ETH_P_IP)) { + switch (type) { + default: + case ICMP_TIME_EXCEEDED: + err = EHOSTUNREACH; + break; + case ICMP_SOURCE_QUENCH: + /* This is not a real error but ping wants to see it. + * Report it with some fake errno. + */ + err = EREMOTEIO; + break; + case ICMP_PARAMETERPROB: + err = EPROTO; + harderr = 1; + break; + case ICMP_DEST_UNREACH: + if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */ + ipv4_sk_update_pmtu(skb, sk, info); + if (inet_sock->pmtudisc != IP_PMTUDISC_DONT) { + err = EMSGSIZE; + harderr = 1; + break; + } + goto out; } - goto out; - } - err = EHOSTUNREACH; - if (code <= NR_ICMP_UNREACH) { - harderr = icmp_err_convert[code].fatal; - err = icmp_err_convert[code].errno; + err = EHOSTUNREACH; + if (code <= NR_ICMP_UNREACH) { + harderr = icmp_err_convert[code].fatal; + err = icmp_err_convert[code].errno; + } + break; + case ICMP_REDIRECT: + /* See ICMP_SOURCE_QUENCH */ + ipv4_sk_redirect(skb, sk); + err = EREMOTEIO; + break; } - break; - case ICMP_REDIRECT: - /* See ICMP_SOURCE_QUENCH */ - ipv4_sk_redirect(skb, sk); - err = EREMOTEIO; - break; +#if IS_ENABLED(CONFIG_IPV6) + } else if (skb->protocol == htons(ETH_P_IPV6)) { + harderr = pingv6_ops.icmpv6_err_convert(type, code, &err); +#endif } /* * RFC1122: OK. Passes ICMP errors back to application, as per * 4.1.3.3. */ - if (!inet_sock->recverr) { + if ((family == AF_INET && !inet_sock->recverr) || + (family == AF_INET6 && !inet6_sk(sk)->recverr)) { if (!harderr || sk->sk_state != TCP_ESTABLISHED) goto out; } else { - ip_icmp_error(sk, skb, err, 0 /* no remote port */, - info, (u8 *)icmph); + if (family == AF_INET) { + ip_icmp_error(sk, skb, err, 0 /* no remote port */, + info, (u8 *)icmph); +#if IS_ENABLED(CONFIG_IPV6) + } else if (family == AF_INET6) { + pingv6_ops.ipv6_icmp_error(sk, skb, err, 0, + info, (u8 *)icmph); +#endif + } } sk->sk_err = err; sk->sk_error_report(sk); out: sock_put(sk); } +EXPORT_SYMBOL_GPL(ping_err); /* - * Copy and checksum an ICMP Echo packet from user space into a buffer. + * Copy and checksum an ICMP Echo packet from user space into a buffer + * starting from the payload. */ -struct pingfakehdr { - struct icmphdr icmph; - struct iovec *iov; - __wsum wcheck; -}; - -static int ping_getfrag(void *from, char *to, - int offset, int fraglen, int odd, struct sk_buff *skb) +int ping_getfrag(void *from, char *to, + int offset, int fraglen, int odd, struct sk_buff *skb) { struct pingfakehdr *pfh = (struct pingfakehdr *)from; @@ -433,20 +609,33 @@ static int ping_getfrag(void *from, char *to, pfh->iov, 0, fraglen - sizeof(struct icmphdr), &pfh->wcheck)) return -EFAULT; + } else if (offset < sizeof(struct icmphdr)) { + BUG(); + } else { + if (csum_partial_copy_fromiovecend + (to, pfh->iov, offset - sizeof(struct icmphdr), + fraglen, &pfh->wcheck)) + return -EFAULT; + } - return 0; +#if IS_ENABLED(CONFIG_IPV6) + /* For IPv6, checksum each skb as we go along, as expected by + * icmpv6_push_pending_frames. For IPv4, accumulate the checksum in + * wcheck, it will be finalized in ping_v4_push_pending_frames. + */ + if (pfh->family == AF_INET6) { + skb->csum = pfh->wcheck; + skb->ip_summed = CHECKSUM_NONE; + pfh->wcheck = 0; } - if (offset < sizeof(struct icmphdr)) - BUG(); - if (csum_partial_copy_fromiovecend - (to, pfh->iov, offset - sizeof(struct icmphdr), - fraglen, &pfh->wcheck)) - return -EFAULT; +#endif + return 0; } +EXPORT_SYMBOL_GPL(ping_getfrag); -static int ping_push_pending_frames(struct sock *sk, struct pingfakehdr *pfh, - struct flowi4 *fl4) +static int ping_v4_push_pending_frames(struct sock *sk, struct pingfakehdr *pfh, + struct flowi4 *fl4) { struct sk_buff *skb = skb_peek(&sk->sk_write_queue); @@ -458,24 +647,9 @@ static int ping_push_pending_frames(struct sock *sk, struct pingfakehdr *pfh, return ip_push_pending_frames(sk, fl4); } -static int ping_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, - size_t len) -{ - struct net *net = sock_net(sk); - struct flowi4 fl4; - struct inet_sock *inet = inet_sk(sk); - struct ipcm_cookie ipc; - struct icmphdr user_icmph; - struct pingfakehdr pfh; - struct rtable *rt = NULL; - struct ip_options_data opt_copy; - int free = 0; - __be32 saddr, daddr, faddr; - u8 tos; - int err; - - pr_debug("ping_sendmsg(sk=%p,sk->num=%u)\n", inet, inet->inet_num); - +int ping_common_sendmsg(int family, struct msghdr *msg, size_t len, + void *user_icmph, size_t icmph_len) { + u8 type, code; if (len > 0xFFFF) return -EMSGSIZE; @@ -490,15 +664,53 @@ static int ping_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, /* * Fetch the ICMP header provided by the userland. - * iovec is modified! + * iovec is modified! The ICMP header is consumed. */ - - if (memcpy_fromiovec((u8 *)&user_icmph, msg->msg_iov, - sizeof(struct icmphdr))) + if (memcpy_fromiovec(user_icmph, msg->msg_iov, icmph_len)) return -EFAULT; - if (!ping_supported(user_icmph.type, user_icmph.code)) + + if (family == AF_INET) { + type = ((struct icmphdr *) user_icmph)->type; + code = ((struct icmphdr *) user_icmph)->code; +#if IS_ENABLED(CONFIG_IPV6) + } else if (family == AF_INET6) { + type = ((struct icmp6hdr *) user_icmph)->icmp6_type; + code = ((struct icmp6hdr *) user_icmph)->icmp6_code; +#endif + } else { + BUG(); + } + + if (!ping_supported(family, type, code)) return -EINVAL; + return 0; +} +EXPORT_SYMBOL_GPL(ping_common_sendmsg); + +int ping_v4_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len) +{ + struct net *net = sock_net(sk); + struct flowi4 fl4; + struct inet_sock *inet = inet_sk(sk); + struct ipcm_cookie ipc; + struct icmphdr user_icmph; + struct pingfakehdr pfh; + struct rtable *rt = NULL; + struct ip_options_data opt_copy; + int free = 0; + __be32 saddr, daddr, faddr; + u8 tos; + int err; + + pr_debug("ping_v4_sendmsg(sk=%p,sk->num=%u)\n", inet, inet->inet_num); + + err = ping_common_sendmsg(AF_INET, msg, len, &user_icmph, + sizeof(user_icmph)); + if (err) + return err; + /* * Get and verify the address. */ @@ -508,7 +720,7 @@ static int ping_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, if (msg->msg_namelen < sizeof(*usin)) return -EINVAL; if (usin->sin_family != AF_INET) - return -EINVAL; + return -EAFNOSUPPORT; daddr = usin->sin_addr.s_addr; /* no remote port */ } else { @@ -570,7 +782,8 @@ static int ping_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, 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, + sock_i_uid(sk)); security_sk_classify_flow(sk, flowi4_to_flowi(&fl4)); rt = ip_route_output_flow(net, &fl4, sk); @@ -603,13 +816,14 @@ back_from_confirm: pfh.icmph.un.echo.sequence = user_icmph.un.echo.sequence; pfh.iov = msg->msg_iov; pfh.wcheck = 0; + pfh.family = AF_INET; err = ip_append_data(sk, &fl4, ping_getfrag, &pfh, len, 0, &ipc, &rt, msg->msg_flags); if (err) ip_flush_pending_frames(sk); else - err = ping_push_pending_frames(sk, &pfh, &fl4); + err = ping_v4_push_pending_frames(sk, &pfh, &fl4); release_sock(sk); out: @@ -630,10 +844,13 @@ do_confirm: goto out; } -static int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, - size_t len, int noblock, int flags, int *addr_len) +int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len, int noblock, int flags, int *addr_len) { struct inet_sock *isk = inet_sk(sk); + int family = sk->sk_family; + struct sockaddr_in *sin; + struct sockaddr_in6 *sin6; struct sk_buff *skb; int copied, err; @@ -643,8 +860,22 @@ static int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, if (flags & MSG_OOB) goto out; - if (flags & MSG_ERRQUEUE) - return ip_recv_error(sk, msg, len, addr_len); + if (addr_len) { + if (family == AF_INET) + *addr_len = sizeof(*sin); + else if (family == AF_INET6 && addr_len) + *addr_len = sizeof(*sin6); + } + + if (flags & MSG_ERRQUEUE) { + if (family == AF_INET) { + return ip_recv_error(sk, msg, len, addr_len); +#if IS_ENABLED(CONFIG_IPV6) + } else if (family == AF_INET6) { + return pingv6_ops.ipv6_recv_error(sk, msg, len); +#endif + } + } skb = skb_recv_datagram(sk, flags, noblock, &err); if (!skb) @@ -663,18 +894,44 @@ static int ping_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, sock_recv_timestamp(msg, sk, skb); - /* Copy the address. */ - if (msg->msg_name) { - struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name; + /* Copy the address and add cmsg data. */ + if (family == AF_INET) { + sin = (struct sockaddr_in *) msg->msg_name; + if (sin) { + sin->sin_family = AF_INET; + sin->sin_port = 0 /* skb->h.uh->source */; + sin->sin_addr.s_addr = ip_hdr(skb)->saddr; + memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); + } + + if (isk->cmsg_flags) + ip_cmsg_recv(msg, skb); + +#if IS_ENABLED(CONFIG_IPV6) + } else if (family == AF_INET6) { + struct ipv6_pinfo *np = inet6_sk(sk); + struct ipv6hdr *ip6 = ipv6_hdr(skb); + sin6 = (struct sockaddr_in6 *) msg->msg_name; + + if (sin6) { + sin6->sin6_family = AF_INET6; + sin6->sin6_port = 0; + sin6->sin6_addr = ip6->saddr; + sin6->sin6_flowinfo = 0; + if (np->sndflow) + sin6->sin6_flowinfo = ip6_flowinfo(ip6); + sin6->sin6_scope_id = + ipv6_iface_scope_id(&sin6->sin6_addr, + IP6CB(skb)->iif); + } - sin->sin_family = AF_INET; - sin->sin_port = 0 /* skb->h.uh->source */; - sin->sin_addr.s_addr = ip_hdr(skb)->saddr; - memset(sin->sin_zero, 0, sizeof(sin->sin_zero)); - *addr_len = sizeof(*sin); + if (inet6_sk(sk)->rxopt.all) + pingv6_ops.ip6_datagram_recv_ctl(sk, msg, skb); +#endif + } else { + BUG(); } - if (isk->cmsg_flags) - ip_cmsg_recv(msg, skb); + err = copied; done: @@ -683,8 +940,9 @@ out: pr_debug("ping_recvmsg -> %d\n", err); return err; } +EXPORT_SYMBOL_GPL(ping_recvmsg); -static int ping_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) +int ping_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) { pr_debug("ping_queue_rcv_skb(sk=%p,sk->num=%d,skb=%p)\n", inet_sk(sk), inet_sk(sk)->inet_num, skb); @@ -695,6 +953,7 @@ static int ping_queue_rcv_skb(struct sock *sk, struct sk_buff *skb) } return 0; } +EXPORT_SYMBOL_GPL(ping_queue_rcv_skb); /* @@ -705,10 +964,7 @@ void ping_rcv(struct sk_buff *skb) { struct sock *sk; struct net *net = dev_net(skb->dev); - struct iphdr *iph = ip_hdr(skb); struct icmphdr *icmph = icmp_hdr(skb); - __be32 saddr = iph->saddr; - __be32 daddr = iph->daddr; /* We assume the packet has already been checked by icmp_rcv */ @@ -718,8 +974,7 @@ void ping_rcv(struct sk_buff *skb) /* Push ICMP header back */ skb_push(skb, skb->data - (u8 *)icmph); - sk = ping_v4_lookup(net, saddr, daddr, ntohs(icmph->un.echo.id), - skb->dev->ifindex); + sk = ping_lookup(net, skb, ntohs(icmph->un.echo.id)); if (sk != NULL) { struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC); @@ -733,6 +988,7 @@ void ping_rcv(struct sk_buff *skb) /* We're called from icmp_rcv(). kfree_skb() is done there. */ } +EXPORT_SYMBOL_GPL(ping_rcv); struct proto ping_prot = { .name = "PING", @@ -743,14 +999,14 @@ struct proto ping_prot = { .disconnect = udp_disconnect, .setsockopt = ip_setsockopt, .getsockopt = ip_getsockopt, - .sendmsg = ping_sendmsg, + .sendmsg = ping_v4_sendmsg, .recvmsg = ping_recvmsg, .bind = ping_bind, .backlog_rcv = ping_queue_rcv_skb, .release_cb = ip4_datagram_release_cb, - .hash = ping_v4_hash, - .unhash = ping_v4_unhash, - .get_port = ping_v4_get_port, + .hash = ping_hash, + .unhash = ping_unhash, + .get_port = ping_get_port, .obj_size = sizeof(struct inet_sock), }; EXPORT_SYMBOL(ping_prot); diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c index b4a1c42a627f..2dfe804fd663 100644 --- a/net/ipv4/raw.c +++ b/net/ipv4/raw.c @@ -571,9 +571,9 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, flowi4_init_output(&fl4, ipc.oif, sk->sk_mark, tos, RT_SCOPE_UNIVERSE, inet->hdrincl ? IPPROTO_RAW : sk->sk_protocol, - inet_sk_flowi_flags(sk) | FLOWI_FLAG_CAN_SLEEP | - (inet->hdrincl ? FLOWI_FLAG_KNOWN_NH : 0), - daddr, saddr, 0, 0); + inet_sk_flowi_flags(sk) | FLOWI_FLAG_CAN_SLEEP, + daddr, saddr, 0, 0, + sock_i_uid(sk)); if (!inet->hdrincl) { err = raw_probe_proto_opt(&fl4, msg); diff --git a/net/ipv4/route.c b/net/ipv4/route.c index e23c5f64286b..577801102a70 100644 --- a/net/ipv4/route.c +++ b/net/ipv4/route.c @@ -515,7 +515,7 @@ void __ip_select_ident(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(struct flowi4 *fl4, struct sock *sk, const struct iphdr *iph, int oif, u8 tos, u8 prot, u32 mark, int flow_flags) @@ -531,11 +531,12 @@ 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, + sk ? sock_i_uid(sk) : 0); } static void build_skb_flow_key(struct flowi4 *fl4, const struct sk_buff *skb, - const struct sock *sk) + struct sock *sk) { const struct iphdr *iph = ip_hdr(skb); int oif = skb->dev->ifindex; @@ -546,7 +547,7 @@ static void build_skb_flow_key(struct flowi4 *fl4, const struct sk_buff *skb, __build_flow_key(fl4, sk, iph, oif, tos, prot, mark, 0); } -static void build_sk_flow_key(struct flowi4 *fl4, const struct sock *sk) +static void build_sk_flow_key(struct flowi4 *fl4, struct sock *sk) { const struct inet_sock *inet = inet_sk(sk); const struct ip_options_rcu *inet_opt; @@ -560,11 +561,12 @@ 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, + sock_i_uid(sk)); rcu_read_unlock(); } -static void ip_rt_build_flow_key(struct flowi4 *fl4, const struct sock *sk, +static void ip_rt_build_flow_key(struct flowi4 *fl4, struct sock *sk, const struct sk_buff *skb) { if (skb) @@ -971,6 +973,9 @@ void ipv4_update_pmtu(struct sk_buff *skb, struct net *net, u32 mtu, struct flowi4 fl4; struct rtable *rt; + if (!mark) + mark = IP4_REPLY_MARK(net, skb->mark); + __build_flow_key(&fl4, NULL, iph, oif, RT_TOS(iph->tos), protocol, mark, flow_flags); rt = __ip_route_output_key(net, &fl4); @@ -988,6 +993,10 @@ static void __ipv4_sk_update_pmtu(struct sk_buff *skb, struct sock *sk, u32 mtu) struct rtable *rt; __build_flow_key(&fl4, sk, iph, 0, 0, 0, 0, 0); + + if (!fl4.flowi4_mark) + fl4.flowi4_mark = IP4_REPLY_MARK(sock_net(sk), skb->mark); + rt = __ip_route_output_key(sock_net(sk), &fl4); if (!IS_ERR(rt)) { __ip_rt_update_pmtu(rt, &fl4, mtu); @@ -2301,6 +2310,11 @@ static int rt_fill_info(struct net *net, __be32 dst, __be32 src, 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)) { @@ -2350,6 +2364,7 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh) int err; int mark; struct sk_buff *skb; + kuid_t uid; err = nlmsg_parse(nlh, sizeof(*rtm), tb, RTA_MAX, rtm_ipv4_policy); if (err < 0) @@ -2377,6 +2392,10 @@ static int inet_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh) dst = tb[RTA_DST] ? nla_get_be32(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; @@ -2384,6 +2403,7 @@ 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 (iif) { struct net_device *dev; diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c index b05c96e7af8b..c94032b95c60 100644 --- a/net/ipv4/syncookies.c +++ b/net/ipv4/syncookies.c @@ -312,6 +312,7 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, ireq->rmt_port = th->source; ireq->loc_addr = ip_hdr(skb)->daddr; ireq->rmt_addr = ip_hdr(skb)->saddr; + ireq->ir_mark = inet_request_mark(sk, skb); ireq->ecn_ok = ecn_ok; ireq->snd_wscale = tcp_opt.snd_wscale; ireq->sack_ok = tcp_opt.sack_ok; @@ -348,11 +349,12 @@ struct sock *cookie_v4_check(struct sock *sk, struct sk_buff *skb, * hasn't changed since we received the original syn, but I see * no easy way to do this. */ - flowi4_init_output(&fl4, sk->sk_bound_dev_if, sk->sk_mark, + flowi4_init_output(&fl4, sk->sk_bound_dev_if, ireq->ir_mark, RT_CONN_FLAGS(sk), RT_SCOPE_UNIVERSE, IPPROTO_TCP, inet_sk_flowi_flags(sk), (opt && opt->srr) ? opt->faddr : ireq->rmt_addr, - ireq->loc_addr, th->source, th->dest); + ireq->loc_addr, th->source, th->dest, + sock_i_uid(sk)); 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 90b26beb84d4..2c707a91e04c 100644 --- a/net/ipv4/sysctl_net_ipv4.c +++ b/net/ipv4/sysctl_net_ipv4.c @@ -138,6 +138,21 @@ static int ipv4_ping_group_range(ctl_table *table, int write, return ret; } +/* Validate changes from /proc interface. */ +static int proc_tcp_default_init_rwnd(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(ctl_table *ctl, int write, void __user *buffer, size_t *lenp, loff_t *ppos) { @@ -737,7 +752,7 @@ static struct ctl_table ipv4_table[] = { .mode = 0644, .proc_handler = proc_dointvec }, - { + { .procname = "tcp_thin_dupack", .data = &sysctl_tcp_thin_dupack, .maxlen = sizeof(int), @@ -763,6 +778,13 @@ static struct ctl_table ipv4_table[] = { .extra2 = &gso_max_segs, }, { + .procname = "tcp_default_init_rwnd", + .data = &sysctl_tcp_default_init_rwnd, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_tcp_default_init_rwnd + }, + { .procname = "udp_mem", .data = &sysctl_udp_mem, .maxlen = sizeof(sysctl_udp_mem), @@ -851,6 +873,20 @@ static struct ctl_table ipv4_net_table[] = { .mode = 0644, .proc_handler = ipv4_tcp_mem, }, + { + .procname = "fwmark_reflect", + .data = &init_net.ipv4.sysctl_fwmark_reflect, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "tcp_fwmark_accept", + .data = &init_net.ipv4.sysctl_tcp_fwmark_accept, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, { } }; 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 5d4bd6ca3ab1..8cc9b5499013 100644 --- a/net/ipv4/tcp.c +++ b/net/ipv4/tcp.c @@ -268,12 +268,16 @@ #include <linux/crypto.h> #include <linux/time.h> #include <linux/slab.h> +#include <linux/uid_stat.h> #include <net/icmp.h> #include <net/inet_common.h> #include <net/tcp.h> #include <net/xfrm.h> #include <net/ip.h> +#include <net/ip6_route.h> +#include <net/ipv6.h> +#include <net/transp_v6.h> #include <net/netdma.h> #include <net/sock.h> @@ -1240,6 +1244,9 @@ out: tcp_push(sk, flags, mss_now, tp->nonagle); out_nopush: release_sock(sk); + + if (copied + copied_syn) + uid_stat_tcp_snd(current_uid(), copied + copied_syn); return copied + copied_syn; do_fault: @@ -1544,6 +1551,7 @@ int tcp_read_sock(struct sock *sk, read_descriptor_t *desc, if (copied > 0) { tcp_recv_skb(sk, seq, &offset); tcp_cleanup_rbuf(sk, copied); + uid_stat_tcp_rcv(current_uid(), copied); } return copied; } @@ -1948,6 +1956,9 @@ skip_copy: tcp_cleanup_rbuf(sk, copied); release_sock(sk); + + if (copied > 0) + uid_stat_tcp_rcv(current_uid(), copied); return copied; out: @@ -1956,6 +1967,8 @@ out: recv_urg: err = tcp_recv_urg(sk, msg, len, flags); + if (err > 0) + uid_stat_tcp_rcv(current_uid(), err); goto out; recv_sndq: @@ -3468,3 +3481,108 @@ void __init tcp_init(void) tcp_tasklet_init(); } + +static int tcp_is_local(struct net *net, __be32 addr) { + struct rtable *rt; + struct flowi4 fl4 = { .daddr = addr }; + rt = ip_route_output_key(net, &fl4); + if (IS_ERR_OR_NULL(rt)) + return 0; + return rt->dst.dev && (rt->dst.dev->flags & IFF_LOOPBACK); +} + +#if defined(CONFIG_IPV6) +static int tcp_is_local6(struct net *net, struct in6_addr *addr) { + struct rt6_info *rt6 = rt6_lookup(net, addr, addr, 0, 0); + return rt6 && rt6->dst.dev && (rt6->dst.dev->flags & IFF_LOOPBACK); +} +#endif + +/* + * tcp_nuke_addr - destroy all sockets on the given local address + * if local address is the unspecified address (0.0.0.0 or ::), destroy all + * sockets with local addresses that are not configured. + */ +int tcp_nuke_addr(struct net *net, struct sockaddr *addr) +{ + int family = addr->sa_family; + unsigned int bucket; + + struct in_addr *in; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + struct in6_addr *in6 = NULL; +#endif + if (family == AF_INET) { + in = &((struct sockaddr_in *)addr)->sin_addr; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + } else if (family == AF_INET6) { + in6 = &((struct sockaddr_in6 *)addr)->sin6_addr; +#endif + } else { + return -EAFNOSUPPORT; + } + + for (bucket = 0; bucket < tcp_hashinfo.ehash_mask; bucket++) { + struct hlist_nulls_node *node; + struct sock *sk; + spinlock_t *lock = inet_ehash_lockp(&tcp_hashinfo, bucket); + +restart: + spin_lock_bh(lock); + sk_nulls_for_each(sk, node, &tcp_hashinfo.ehash[bucket].chain) { + struct inet_sock *inet = inet_sk(sk); + + if (sysctl_ip_dynaddr && sk->sk_state == TCP_SYN_SENT) + continue; + if (sock_flag(sk, SOCK_DEAD)) + continue; + + if (family == AF_INET) { + __be32 s4 = inet->inet_rcv_saddr; + if (s4 == LOOPBACK4_IPV6) + continue; + + if (in->s_addr != s4 && + !(in->s_addr == INADDR_ANY && + !tcp_is_local(net, s4))) + continue; + } + +#if defined(CONFIG_IPV6) + if (family == AF_INET6) { + struct in6_addr *s6; + if (!inet->pinet6) + continue; + + s6 = &inet->pinet6->rcv_saddr; + if (ipv6_addr_type(s6) == IPV6_ADDR_MAPPED) + continue; + + if (!ipv6_addr_equal(in6, s6) && + !(ipv6_addr_equal(in6, &in6addr_any) && + !tcp_is_local6(net, s6))) + continue; + } +#endif + + sock_hold(sk); + spin_unlock_bh(lock); + + local_bh_disable(); + bh_lock_sock(sk); + sk->sk_err = ETIMEDOUT; + sk->sk_error_report(sk); + + tcp_done(sk); + bh_unlock_sock(sk); + local_bh_enable(); + sock_put(sk); + + goto restart; + } + spin_unlock_bh(lock); + } + + return 0; +} +EXPORT_SYMBOL_GPL(tcp_nuke_addr); diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c index a8be45e4d34f..22fc869465e8 100644 --- a/net/ipv4/tcp_input.c +++ b/net/ipv4/tcp_input.c @@ -98,6 +98,7 @@ 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_default_init_rwnd __read_mostly = TCP_DEFAULT_INIT_RCVWND; #define FLAG_DATA 0x01 /* Incoming frame contained data. */ #define FLAG_WIN_UPDATE 0x02 /* Incoming ACK was a window update. */ @@ -351,14 +352,14 @@ static void tcp_grow_window(struct sock *sk, const struct sk_buff *skb) static void tcp_fixup_rcvbuf(struct sock *sk) { u32 mss = tcp_sk(sk)->advmss; - u32 icwnd = TCP_DEFAULT_INIT_RCVWND; + u32 icwnd = sysctl_tcp_default_init_rwnd; int rcvmem; /* Limit to 10 segments if mss <= 1460, * or 14600/mss segments, with a minimum of two segments. */ if (mss > 1460) - icwnd = max_t(u32, (1460 * TCP_DEFAULT_INIT_RCVWND) / mss, 2); + icwnd = max_t(u32, (1460 * icwnd) / mss, 2); rcvmem = SKB_TRUESIZE(mss + MAX_TCP_HEADER); while (tcp_win_from_space(rcvmem) < mss) diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c index 7c3eec386a4b..d64453a81028 100644 --- a/net/ipv4/tcp_ipv4.c +++ b/net/ipv4/tcp_ipv4.c @@ -1530,6 +1530,7 @@ int tcp_v4_conn_request(struct sock *sk, struct sk_buff *skb) ireq->rmt_addr = saddr; ireq->no_srccheck = inet_sk(sk)->transparent; ireq->opt = tcp_v4_save_options(skb); + ireq->ir_mark = inet_request_mark(sk, skb); if (security_inet_conn_request(sk, skb, req)) goto drop_and_free; diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c index 76c80b59e80f..5653febcb2ac 100644 --- a/net/ipv4/tcp_output.c +++ b/net/ipv4/tcp_output.c @@ -231,14 +231,13 @@ void tcp_select_initial_window(int __space, __u32 mss, } /* Set initial window to a value enough for senders starting with - * initial congestion window of TCP_DEFAULT_INIT_RCVWND. Place + * initial congestion window of sysctl_tcp_default_init_rwnd. Place * a limit on the initial window when mss is larger than 1460. */ if (mss > (1 << *rcv_wscale)) { - int init_cwnd = TCP_DEFAULT_INIT_RCVWND; + int init_cwnd = sysctl_tcp_default_init_rwnd; if (mss > 1460) - init_cwnd = - max_t(u32, (1460 * TCP_DEFAULT_INIT_RCVWND) / mss, 2); + init_cwnd = max_t(u32, (1460 * init_cwnd) / mss, 2); /* when initializing use the value from init_rcv_wnd * rather than the default from above */ diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c index c3075b552248..58c5dbd5fb94 100644 --- a/net/ipv4/udp.c +++ b/net/ipv4/udp.c @@ -963,7 +963,8 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, flowi4_init_output(fl4, ipc.oif, sk->sk_mark, tos, RT_SCOPE_UNIVERSE, sk->sk_protocol, inet_sk_flowi_flags(sk)|FLOWI_FLAG_CAN_SLEEP, - faddr, saddr, dport, inet->inet_sport); + faddr, saddr, dport, inet->inet_sport, + sock_i_uid(sk)); security_sk_classify_flow(sk, flowi4_to_flowi(fl4)); rt = ip_route_output_flow(net, fl4, sk); diff --git a/net/ipv6/Makefile b/net/ipv6/Makefile index 9af088d2cdaa..470a9c008e9b 100644 --- a/net/ipv6/Makefile +++ b/net/ipv6/Makefile @@ -7,7 +7,7 @@ obj-$(CONFIG_IPV6) += ipv6.o ipv6-objs := af_inet6.o anycast.o ip6_output.o ip6_input.o addrconf.o \ addrlabel.o \ route.o ip6_fib.o ipv6_sockglue.o ndisc.o udp.o udplite.o \ - raw.o icmp.o mcast.o reassembly.o tcp_ipv6.o \ + raw.o icmp.o mcast.o reassembly.o tcp_ipv6.o ping.o \ exthdrs.o datagram.o ip6_flowlabel.o inet6_connection_sock.o ipv6-offload := ip6_offload.o tcpv6_offload.o udp_offload.o exthdrs_offload.o diff --git a/net/ipv6/addrconf.c b/net/ipv6/addrconf.c index d0912acd9522..69e8b50a45bd 100644 --- a/net/ipv6/addrconf.c +++ b/net/ipv6/addrconf.c @@ -198,6 +198,7 @@ static struct ipv6_devconf ipv6_devconf __read_mostly = { .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, @@ -232,6 +233,7 @@ static struct ipv6_devconf ipv6_devconf_dflt __read_mostly = { .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, @@ -1173,6 +1175,9 @@ enum { #endif IPV6_SADDR_RULE_ORCHID, IPV6_SADDR_RULE_PREFIX, +#ifdef CONFIG_IPV6_OPTIMISTIC_DAD + IPV6_SADDR_RULE_NOT_OPTIMISTIC, +#endif IPV6_SADDR_RULE_MAX }; @@ -1200,6 +1205,15 @@ static inline int ipv6_saddr_preferred(int type) return 0; } +static inline bool ipv6_use_optimistic_addr(struct inet6_dev *idev) +{ +#ifdef CONFIG_IPV6_OPTIMISTIC_DAD + return idev && idev->cnf.optimistic_dad && idev->cnf.use_optimistic; +#else + return false; +#endif +} + static int ipv6_get_saddr_eval(struct net *net, struct ipv6_saddr_score *score, struct ipv6_saddr_dst *dst, @@ -1260,10 +1274,16 @@ static int ipv6_get_saddr_eval(struct net *net, score->scopedist = ret; break; case IPV6_SADDR_RULE_PREFERRED: + { /* Rule 3: Avoid deprecated and optimistic addresses */ + u8 avoid = IFA_F_DEPRECATED; + + if (!ipv6_use_optimistic_addr(score->ifa->idev)) + avoid |= IFA_F_OPTIMISTIC; ret = ipv6_saddr_preferred(score->addr_type) || - !(score->ifa->flags & (IFA_F_DEPRECATED|IFA_F_OPTIMISTIC)); + !(score->ifa->flags & avoid); break; + } #ifdef CONFIG_IPV6_MIP6 case IPV6_SADDR_RULE_HOA: { @@ -1311,6 +1331,14 @@ static int ipv6_get_saddr_eval(struct net *net, ret = score->ifa->prefix_len; score->matchlen = ret; break; +#ifdef CONFIG_IPV6_OPTIMISTIC_DAD + case IPV6_SADDR_RULE_NOT_OPTIMISTIC: + /* Optimistic addresses still have lower precedence than other + * preferred addresses. + */ + ret = !(score->ifa->flags & IFA_F_OPTIMISTIC); + break; +#endif default: ret = 0; } @@ -1506,7 +1534,9 @@ int ipv6_chk_addr(struct net *net, const struct in6_addr *addr, if (!net_eq(dev_net(ifp->idev->dev), net)) continue; if (ipv6_addr_equal(&ifp->addr, addr) && - !(ifp->flags&IFA_F_TENTATIVE) && + (!(ifp->flags&IFA_F_TENTATIVE) || + (ipv6_use_optimistic_addr(ifp->idev) && + ifp->flags&IFA_F_OPTIMISTIC)) && (dev == NULL || ifp->idev->dev == dev || !(ifp->scope&(IFA_LINK|IFA_HOST) || strict))) { rcu_read_unlock_bh(); @@ -1946,6 +1976,31 @@ static void __ipv6_try_regen_rndid(struct inet6_dev *idev, struct in6_addr *tmp } #endif +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. */ @@ -1955,7 +2010,7 @@ addrconf_prefix_route(struct in6_addr *pfx, int plen, struct net_device *dev, unsigned long expires, u32 flags) { struct fib6_config cfg = { - .fc_table = RT6_TABLE_PREFIX, + .fc_table = addrconf_rt_table(dev, RT6_TABLE_PREFIX), .fc_metric = IP6_RT_PRIO_ADDRCONF, .fc_ifindex = dev->ifindex, .fc_expires = expires, @@ -1989,7 +2044,8 @@ static struct rt6_info *addrconf_get_prefix_route(const struct in6_addr *pfx, struct rt6_info *rt = NULL; struct fib6_table *table; - table = fib6_get_table(dev_net(dev), RT6_TABLE_PREFIX); + table = fib6_get_table(dev_net(dev), + addrconf_rt_table(dev, RT6_TABLE_PREFIX)); if (table == NULL) return NULL; @@ -3145,11 +3201,13 @@ static int addrconf_ifdown(struct net_device *dev, int how) write_unlock_bh(&idev->lock); - /* Step 5: Discard multicast list */ - if (how) + /* Step 5: Discard anycast and multicast list */ + if (how) { + ipv6_ac_destroy_dev(idev); ipv6_mc_destroy_dev(idev); - else + } else { ipv6_mc_down(idev); + } idev->tstamp = jiffies; @@ -3263,8 +3321,15 @@ static void addrconf_dad_start(struct inet6_ifaddr *ifp) * Optimistic nodes can start receiving * Frames right away */ - if (ifp->flags & IFA_F_OPTIMISTIC) + if (ifp->flags & IFA_F_OPTIMISTIC) { ip6_ins_rt(ifp->rt); + if (ipv6_use_optimistic_addr(idev)) { + /* Because optimistic nodes can use this address, + * notify listeners. If DAD fails, RTM_DELADDR is sent. + */ + ipv6_ifa_notify(RTM_NEWADDR, ifp); + } + } addrconf_dad_kick(ifp); out: @@ -4205,10 +4270,12 @@ static inline void ipv6_store_devconf(struct ipv6_devconf *cnf, 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 array[DEVCONF_OPTIMISTIC_DAD] = cnf->optimistic_dad; + array[DEVCONF_USE_OPTIMISTIC] = cnf->use_optimistic; #endif #ifdef CONFIG_IPV6_MROUTE array[DEVCONF_MC_FORWARDING] = cnf->mc_forwarding; @@ -4914,6 +4981,13 @@ 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), @@ -4936,6 +5010,14 @@ static struct addrconf_sysctl_table .proc_handler = proc_dointvec, }, + { + .procname = "use_optimistic", + .data = &ipv6_devconf.use_optimistic, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + + }, #endif #ifdef CONFIG_IPV6_MROUTE { diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c index ab5c7ad482cd..d29ae19ae698 100644 --- a/net/ipv6/af_inet6.c +++ b/net/ipv6/af_inet6.c @@ -49,6 +49,7 @@ #include <net/udp.h> #include <net/udplite.h> #include <net/tcp.h> +#include <net/ping.h> #include <net/protocol.h> #include <net/inet_common.h> #include <net/route.h> @@ -62,6 +63,20 @@ #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"); @@ -108,6 +123,9 @@ static int inet6_create(struct net *net, struct socket *sock, int protocol, int try_loading_module = 0; int err; + if (!current_has_network()) + return -EACCES; + if (sock->type != SOCK_RAW && sock->type != SOCK_DGRAM && !inet_ehash_secret) @@ -159,8 +177,7 @@ 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; @@ -477,6 +494,21 @@ int inet6_getname(struct socket *sock, struct sockaddr *uaddr, } EXPORT_SYMBOL(inet6_getname); +int inet6_killaddr_ioctl(struct net *net, void __user *arg) { + struct in6_ifreq ireq; + struct sockaddr_in6 sin6; + + if (!capable(CAP_NET_ADMIN)) + return -EACCES; + + if (copy_from_user(&ireq, arg, sizeof(struct in6_ifreq))) + return -EFAULT; + + sin6.sin6_family = AF_INET6; + sin6.sin6_addr = ireq.ifr6_addr; + return tcp_nuke_addr(net, (struct sockaddr *) &sin6); +} + int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) { struct sock *sk = sock->sk; @@ -500,6 +532,8 @@ int inet6_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg) return addrconf_del_ifaddr(net, (void __user *) arg); case SIOCSIFDSTADDR: return addrconf_set_dstaddr(net, (void __user *) arg); + case SIOCKILLADDR: + return inet6_killaddr_ioctl(net, (void __user *) arg); default: if (!sk->sk_prot->ioctl) return -ENOIOCTLCMD; @@ -660,6 +694,7 @@ 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 = sock_i_uid(sk); security_sk_classify_flow(sk, flowi6_to_flowi(&fl6)); final_p = fl6_update_dst(&fl6, np->opt, &final); @@ -840,6 +875,9 @@ static int __init inet6_init(void) if (err) goto out_unregister_udplite_proto; + err = proto_register(&pingv6_prot, 1); + if (err) + goto out_unregister_ping_proto; /* We MUST register RAW sockets before we create the ICMP6, * IGMP6, or NDISC control sockets. @@ -930,6 +968,10 @@ static int __init inet6_init(void) if (err) goto ipv6_packet_fail; + err = pingv6_init(); + if (err) + goto pingv6_fail; + #ifdef CONFIG_SYSCTL err = ipv6_sysctl_register(); if (err) @@ -942,6 +984,8 @@ out: sysctl_fail: ipv6_packet_cleanup(); #endif +pingv6_fail: + pingv6_exit(); ipv6_packet_fail: tcpv6_exit(); tcpv6_fail: @@ -985,6 +1029,8 @@ register_pernet_fail: rtnl_unregister_all(PF_INET6); out_sock_register_fail: rawv6_exit(); +out_unregister_ping_proto: + proto_unregister(&pingv6_prot); out_unregister_raw_proto: proto_unregister(&rawv6_prot); out_unregister_udplite_proto: diff --git a/net/ipv6/ah6.c b/net/ipv6/ah6.c index bb02e176cb70..b903e19463c9 100644 --- a/net/ipv6/ah6.c +++ b/net/ipv6/ah6.c @@ -630,7 +630,7 @@ static void ah6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (type == NDISC_REDIRECT) ip6_redirect(skb, net, 0, 0); else - ip6_update_pmtu(skb, net, info, 0, 0); + ip6_update_pmtu(skb, net, info, 0, 0, INVALID_UID); xfrm_state_put(x); } diff --git a/net/ipv6/anycast.c b/net/ipv6/anycast.c index 5a80f15a9de2..1e97d0886597 100644 --- a/net/ipv6/anycast.c +++ b/net/ipv6/anycast.c @@ -341,6 +341,27 @@ static int ipv6_dev_ac_dec(struct net_device *dev, const struct in6_addr *addr) return __ipv6_dev_ac_dec(idev, addr); } +void ipv6_ac_destroy_dev(struct inet6_dev *idev) +{ + struct ifacaddr6 *aca; + + write_lock_bh(&idev->lock); + while ((aca = idev->ac_list) != NULL) { + idev->ac_list = aca->aca_next; + write_unlock_bh(&idev->lock); + + addrconf_leave_solict(idev, &aca->aca_addr); + + dst_hold(&aca->aca_rt->dst); + ip6_del_rt(aca->aca_rt); + + aca_put(aca); + + write_lock_bh(&idev->lock); + } + write_unlock_bh(&idev->lock); +} + /* * check if the interface has this anycast address * called with rcu_read_lock() diff --git a/net/ipv6/datagram.c b/net/ipv6/datagram.c index ce17d3da9b2b..ae52918a3367 100644 --- a/net/ipv6/datagram.c +++ b/net/ipv6/datagram.c @@ -162,6 +162,7 @@ ipv4_connected: fl6.flowi6_mark = sk->sk_mark; fl6.fl6_dport = inet->inet_dport; fl6.fl6_sport = inet->inet_sport; + fl6.flowi6_uid = sock_i_uid(sk); if (!fl6.flowi6_oif && (addr_type&IPV6_ADDR_MULTICAST)) fl6.flowi6_oif = np->mcast_oif; @@ -318,7 +319,7 @@ void ipv6_local_rxpmtu(struct sock *sk, struct flowi6 *fl6, u32 mtu) /* * Handle MSG_ERRQUEUE */ -int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len) +int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len) { struct ipv6_pinfo *np = inet6_sk(sk); struct sock_exterr_skb *serr; @@ -369,7 +370,6 @@ int ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len, int *addr_len) &sin->sin6_addr); sin->sin6_scope_id = 0; } - *addr_len = sizeof(*sin); } memcpy(&errhdr.ee, &serr->ee, sizeof(struct sock_extended_err)); diff --git a/net/ipv6/esp6.c b/net/ipv6/esp6.c index 40ffd72243a4..fdc81cb29e80 100644 --- a/net/ipv6/esp6.c +++ b/net/ipv6/esp6.c @@ -449,7 +449,7 @@ static void esp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (type == NDISC_REDIRECT) ip6_redirect(skb, net, 0, 0); else - ip6_update_pmtu(skb, net, info, 0, 0); + ip6_update_pmtu(skb, net, info, 0, 0, INVALID_UID); xfrm_state_put(x); } diff --git a/net/ipv6/exthdrs_core.c b/net/ipv6/exthdrs_core.c index 51af9d0d019a..11de7379fb9a 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; + } } return -ENOENT; } diff --git a/net/ipv6/icmp.c b/net/ipv6/icmp.c index 2dee1d9d7305..c1b611cc55ae 100644 --- a/net/ipv6/icmp.c +++ b/net/ipv6/icmp.c @@ -57,6 +57,7 @@ #include <net/ipv6.h> #include <net/ip6_checksum.h> +#include <net/ping.h> #include <net/protocol.h> #include <net/raw.h> #include <net/rawv6.h> @@ -84,12 +85,18 @@ static inline struct sock *icmpv6_sk(struct net *net) static void icmpv6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, u8 type, u8 code, int offset, __be32 info) { + /* icmpv6_notify checks 8 bytes can be pulled, icmp6hdr is 8 bytes */ + struct icmp6hdr *icmp6 = (struct icmp6hdr *) (skb->data + offset); 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, INVALID_UID); else if (type == NDISC_REDIRECT) ip6_redirect(skb, net, 0, 0); + + if (!(type & ICMPV6_INFOMSG_MASK)) + if (icmp6->icmp6_type == ICMPV6_ECHO_REQUEST) + ping_err(skb, offset, info); } static int icmpv6_rcv(struct sk_buff *skb); @@ -224,7 +231,8 @@ static bool opt_unrec(struct sk_buff *skb, __u32 offset) return (*op & 0xC0) == 0x80; } -static int icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, struct icmp6hdr *thdr, int len) +int icmpv6_push_pending_frames(struct sock *sk, struct flowi6 *fl6, + struct icmp6hdr *thdr, int len) { struct sk_buff *skb; struct icmp6hdr *icmp6h; @@ -307,8 +315,8 @@ static void mip6_addr_swap(struct sk_buff *skb) static inline void mip6_addr_swap(struct sk_buff *skb) {} #endif -static struct dst_entry *icmpv6_route_lookup(struct net *net, struct sk_buff *skb, - struct sock *sk, struct flowi6 *fl6) +struct dst_entry *icmpv6_route_lookup(struct net *net, struct sk_buff *skb, + struct sock *sk, struct flowi6 *fl6) { struct dst_entry *dst, *dst2; struct flowi6 fl2; @@ -389,6 +397,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) int len; int hlimit; int err = 0; + u32 mark = IP6_REPLY_MARK(net, skb->mark); if ((u8 *)hdr < skb->head || (skb->network_header + sizeof(*hdr)) > skb->tail) @@ -454,6 +463,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) fl6.daddr = hdr->saddr; if (saddr) fl6.saddr = *saddr; + fl6.flowi6_mark = mark; fl6.flowi6_oif = iif; fl6.fl6_icmp_type = type; fl6.fl6_icmp_code = code; @@ -462,6 +472,7 @@ static void icmp6_send(struct sk_buff *skb, u8 type, u8 code, __u32 info) sk = icmpv6_xmit_lock(net); if (sk == NULL) return; + sk->sk_mark = mark; np = inet6_sk(sk); if (!icmpv6_xrlim_allow(sk, type, &fl6)) @@ -543,6 +554,7 @@ static void icmpv6_echo_reply(struct sk_buff *skb) struct dst_entry *dst; int err = 0; int hlimit; + u32 mark = IP6_REPLY_MARK(net, skb->mark); saddr = &ipv6_hdr(skb)->daddr; @@ -559,11 +571,13 @@ static void icmpv6_echo_reply(struct sk_buff *skb) fl6.saddr = *saddr; fl6.flowi6_oif = skb->dev->ifindex; fl6.fl6_icmp_type = ICMPV6_ECHO_REPLY; + fl6.flowi6_mark = mark; security_skb_classify_flow(skb, flowi6_to_flowi(&fl6)); sk = icmpv6_xmit_lock(net); if (sk == NULL) return; + sk->sk_mark = mark; np = inet6_sk(sk); if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr)) @@ -697,7 +711,8 @@ static int icmpv6_rcv(struct sk_buff *skb) skb->csum = ~csum_unfold(csum_ipv6_magic(saddr, daddr, skb->len, IPPROTO_ICMPV6, 0)); if (__skb_checksum_complete(skb)) { - LIMIT_NETDEBUG(KERN_DEBUG "ICMPv6 checksum failed [%pI6 > %pI6]\n", + LIMIT_NETDEBUG(KERN_DEBUG + "ICMPv6 checksum failed [%pI6c > %pI6c]\n", saddr, daddr); goto csum_error; } @@ -718,7 +733,7 @@ static int icmpv6_rcv(struct sk_buff *skb) break; case ICMPV6_ECHO_REPLY: - /* we couldn't care less */ + ping_rcv(skb); break; case ICMPV6_PKT_TOOBIG: diff --git a/net/ipv6/inet6_connection_sock.c b/net/ipv6/inet6_connection_sock.c index e4311cbc8b4e..65a46058c854 100644 --- a/net/ipv6/inet6_connection_sock.c +++ b/net/ipv6/inet6_connection_sock.c @@ -81,9 +81,10 @@ struct dst_entry *inet6_csk_route_req(struct sock *sk, final_p = fl6_update_dst(fl6, np->opt, &final); fl6->saddr = treq->loc_addr; fl6->flowi6_oif = treq->iif; - fl6->flowi6_mark = sk->sk_mark; + fl6->flowi6_mark = inet_rsk(req)->ir_mark; fl6->fl6_dport = inet_rsk(req)->rmt_port; fl6->fl6_sport = inet_rsk(req)->loc_port; + fl6->flowi6_uid = sock_i_uid(sk); security_req_classify_flow(req, flowi6_to_flowi(fl6)); dst = ip6_dst_lookup_flow(sk, fl6, final_p, false); @@ -211,6 +212,7 @@ 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 = sock_i_uid(sk); security_sk_classify_flow(sk, flowi6_to_flowi(fl6)); final_p = fl6_update_dst(fl6, np->opt, &final); diff --git a/net/ipv6/ip6mr.c b/net/ipv6/ip6mr.c index 2c84072b1da7..63cf7cded36d 100644 --- a/net/ipv6/ip6mr.c +++ b/net/ipv6/ip6mr.c @@ -701,7 +701,7 @@ static netdev_tx_t reg_vif_xmit(struct sk_buff *skb, struct mr6_table *mrt; struct flowi6 fl6 = { .flowi6_oif = dev->ifindex, - .flowi6_iif = skb->skb_iif, + .flowi6_iif = skb->skb_iif ? : LOOPBACK_IFINDEX, .flowi6_mark = skb->mark, }; int err; diff --git a/net/ipv6/ipcomp6.c b/net/ipv6/ipcomp6.c index 7af5aee75d98..a1beb59a841e 100644 --- a/net/ipv6/ipcomp6.c +++ b/net/ipv6/ipcomp6.c @@ -78,7 +78,7 @@ static void ipcomp6_err(struct sk_buff *skb, struct inet6_skb_parm *opt, if (type == NDISC_REDIRECT) ip6_redirect(skb, net, 0, 0); else - ip6_update_pmtu(skb, net, info, 0, 0); + ip6_update_pmtu(skb, net, info, 0, 0, INVALID_UID); xfrm_state_put(x); } diff --git a/net/ipv6/netfilter/Kconfig b/net/ipv6/netfilter/Kconfig index 4433ab40e7de..7f45af3e8128 100644 --- a/net/ipv6/netfilter/Kconfig +++ b/net/ipv6/netfilter/Kconfig @@ -153,6 +153,18 @@ config IP6_NF_TARGET_REJECT To compile it as a module, choose M here. If unsure, say N. +config IP6_NF_TARGET_REJECT_SKERR + bool "Force socket error when rejecting with icmp*" + depends on IP6_NF_TARGET_REJECT + default n + help + This option enables turning a "--reject-with icmp*" into a matching + socket error also. + The REJECT target normally allows sending an ICMP message. But it + leaves the local socket unaware of any ingress rejects. + + If unsure, say N. + config IP6_NF_MANGLE tristate "Packet mangling" default m if NETFILTER_ADVANCED=n diff --git a/net/ipv6/netfilter/ip6t_REJECT.c b/net/ipv6/netfilter/ip6t_REJECT.c index 70f9abc0efe9..573c232239b0 100644 --- a/net/ipv6/netfilter/ip6t_REJECT.c +++ b/net/ipv6/netfilter/ip6t_REJECT.c @@ -180,6 +180,15 @@ send_unreach(struct net *net, struct sk_buff *skb_in, unsigned char code, skb_in->dev = net->loopback_dev; icmpv6_send(skb_in, ICMPV6_DEST_UNREACH, code, 0); +#ifdef CONFIG_IP6_NF_TARGET_REJECT_SKERR + if (skb_in->sk) { + icmpv6_err_convert(ICMPV6_DEST_UNREACH, code, + &skb_in->sk->sk_err); + skb_in->sk->sk_error_report(skb_in->sk); + pr_debug("ip6t_REJECT: sk_err=%d for skb=%p sk=%p\n", + skb_in->sk->sk_err, skb_in, skb_in->sk); + } +#endif } static unsigned int diff --git a/net/ipv6/ping.c b/net/ipv6/ping.c new file mode 100644 index 000000000000..857201f4936f --- /dev/null +++ b/net/ipv6/ping.c @@ -0,0 +1,224 @@ +/* + * INET An implementation of the TCP/IP protocol suite for the LINUX + * operating system. INET is implemented using the BSD Socket + * interface as the means of communication with the user level. + * + * "Ping" sockets + * + * 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. + * + * Based on ipv4/ping.c code. + * + * Authors: Lorenzo Colitti (IPv6 support) + * Vasiliy Kulikov / Openwall (IPv4 implementation, for Linux 2.6), + * Pavel Kankovsky (IPv4 implementation, for Linux 2.4.32) + * + */ + +#include <net/addrconf.h> +#include <net/ipv6.h> +#include <net/ip6_route.h> +#include <net/protocol.h> +#include <net/udp.h> +#include <net/transp_v6.h> +#include <net/ping.h> + +struct proto pingv6_prot = { + .name = "PINGv6", + .owner = THIS_MODULE, + .init = ping_init_sock, + .close = ping_close, + .connect = ip6_datagram_connect, + .disconnect = udp_disconnect, + .setsockopt = ipv6_setsockopt, + .getsockopt = ipv6_getsockopt, + .sendmsg = ping_v6_sendmsg, + .recvmsg = ping_recvmsg, + .bind = ping_bind, + .backlog_rcv = ping_queue_rcv_skb, + .hash = ping_hash, + .unhash = ping_unhash, + .get_port = ping_get_port, + .obj_size = sizeof(struct raw6_sock), +}; +EXPORT_SYMBOL_GPL(pingv6_prot); + +static struct inet_protosw pingv6_protosw = { + .type = SOCK_DGRAM, + .protocol = IPPROTO_ICMPV6, + .prot = &pingv6_prot, + .ops = &inet6_dgram_ops, + .no_check = UDP_CSUM_DEFAULT, + .flags = INET_PROTOSW_REUSE, +}; + + +/* Compatibility glue so we can support IPv6 when it's compiled as a module */ +int dummy_ipv6_recv_error(struct sock *sk, struct msghdr *msg, int len) +{ + return -EAFNOSUPPORT; +} +int dummy_ip6_datagram_recv_ctl(struct sock *sk, struct msghdr *msg, + struct sk_buff *skb) +{ + return -EAFNOSUPPORT; +} +int dummy_icmpv6_err_convert(u8 type, u8 code, int *err) +{ + return -EAFNOSUPPORT; +} +void dummy_ipv6_icmp_error(struct sock *sk, struct sk_buff *skb, int err, + __be16 port, u32 info, u8 *payload) {} +int dummy_ipv6_chk_addr(struct net *net, const struct in6_addr *addr, + const struct net_device *dev, int strict) +{ + return 0; +} + +int __init pingv6_init(void) +{ + pingv6_ops.ipv6_recv_error = ipv6_recv_error; + pingv6_ops.ip6_datagram_recv_ctl = ip6_datagram_recv_ctl; + pingv6_ops.icmpv6_err_convert = icmpv6_err_convert; + pingv6_ops.ipv6_icmp_error = ipv6_icmp_error; + pingv6_ops.ipv6_chk_addr = ipv6_chk_addr; + return inet6_register_protosw(&pingv6_protosw); +} + +/* This never gets called because it's not possible to unload the ipv6 module, + * but just in case. + */ +void pingv6_exit(void) +{ + pingv6_ops.ipv6_recv_error = dummy_ipv6_recv_error; + pingv6_ops.ip6_datagram_recv_ctl = dummy_ip6_datagram_recv_ctl; + pingv6_ops.icmpv6_err_convert = dummy_icmpv6_err_convert; + pingv6_ops.ipv6_icmp_error = dummy_ipv6_icmp_error; + pingv6_ops.ipv6_chk_addr = dummy_ipv6_chk_addr; + inet6_unregister_protosw(&pingv6_protosw); +} + +int ping_v6_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, + size_t len) +{ + struct inet_sock *inet = inet_sk(sk); + struct ipv6_pinfo *np = inet6_sk(sk); + struct icmp6hdr user_icmph; + int addr_type; + struct in6_addr *daddr; + int iif = 0; + struct flowi6 fl6; + int err; + int hlimit; + struct dst_entry *dst; + struct rt6_info *rt; + struct pingfakehdr pfh; + + pr_debug("ping_v6_sendmsg(sk=%p,sk->num=%u)\n", inet, inet->inet_num); + + err = ping_common_sendmsg(AF_INET6, msg, len, &user_icmph, + sizeof(user_icmph)); + if (err) + return err; + + if (msg->msg_name) { + struct sockaddr_in6 *u = (struct sockaddr_in6 *) msg->msg_name; + if (msg->msg_namelen < sizeof(*u)) + return -EINVAL; + 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; + } else { + if (sk->sk_state != TCP_ESTABLISHED) + return -EDESTADDRREQ; + daddr = &np->daddr; + } + + if (!iif) + iif = sk->sk_bound_dev_if; + + addr_type = ipv6_addr_type(daddr); + if (__ipv6_addr_needs_scope_id(addr_type) && !iif) + return -EINVAL; + if (addr_type & IPV6_ADDR_MAPPED) + return -EINVAL; + + /* TODO: use ip6_datagram_send_ctl to get options from cmsg */ + + memset(&fl6, 0, sizeof(fl6)); + + fl6.flowi6_proto = IPPROTO_ICMPV6; + fl6.saddr = np->saddr; + fl6.daddr = *daddr; + fl6.flowi6_mark = sk->sk_mark; + fl6.flowi6_uid = sock_i_uid(sk); + 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, 1); + if (IS_ERR(dst)) + return PTR_ERR(dst); + rt = (struct rt6_info *) dst; + + np = inet6_sk(sk); + if (!np) + return -EBADF; + + 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; + pfh.icmph.un.echo.id = inet->inet_sport; + pfh.icmph.un.echo.sequence = user_icmph.icmp6_sequence; + pfh.iov = msg->msg_iov; + pfh.wcheck = 0; + pfh.family = AF_INET6; + + if (ipv6_addr_is_multicast(&fl6.daddr)) + hlimit = np->mcast_hops; + else + hlimit = np->hop_limit; + if (hlimit < 0) + hlimit = ip6_dst_hoplimit(dst); + + lock_sock(sk); + err = ip6_append_data(sk, ping_getfrag, &pfh, len, + 0, hlimit, + np->tclass, NULL, &fl6, rt, + MSG_DONTWAIT, np->dontfrag); + + if (err) { + ICMP6_INC_STATS_BH(sock_net(sk), rt->rt6i_idev, + ICMP6_MIB_OUTERRORS); + ip6_flush_pending_frames(sk); + } else { + err = icmpv6_push_pending_frames(sk, &fl6, + (struct icmp6hdr *) &pfh.icmph, + len); + } + release_sock(sk); + + if (err) + return err; + + return len; +} diff --git a/net/ipv6/raw.c b/net/ipv6/raw.c index 464b1c9c08e4..a4bf16d0d3d0 100644 --- a/net/ipv6/raw.c +++ b/net/ipv6/raw.c @@ -459,8 +459,10 @@ static int rawv6_recvmsg(struct kiocb *iocb, struct sock *sk, if (flags & MSG_OOB) return -EOPNOTSUPP; - if (flags & MSG_ERRQUEUE) - return ipv6_recv_error(sk, msg, len, addr_len); + if (flags & MSG_ERRQUEUE) { + *addr_len = sizeof(struct sockaddr_in6); + return ipv6_recv_error(sk, msg, len); + } if (np->rxpmtu && np->rxopt.bits.rxpmtu) return ipv6_recv_rxpmtu(sk, msg, len, addr_len); @@ -759,6 +761,7 @@ static int rawv6_sendmsg(struct kiocb *iocb, struct sock *sk, memset(&fl6, 0, sizeof(fl6)); fl6.flowi6_mark = sk->sk_mark; + fl6.flowi6_uid = sock_i_uid(sk); if (sin6) { if (addr_len < SIN6_LEN_RFC2133) diff --git a/net/ipv6/route.c b/net/ipv6/route.c index d94d224f7e68..2f127bd67a5d 100644 --- a/net/ipv6/route.c +++ b/net/ipv6/route.c @@ -93,13 +93,12 @@ static void rt6_do_redirect(struct dst_entry *dst, struct sock *sk, struct sk_buff *skb); #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 static u32 *ipv6_cow_metrics(struct dst_entry *dst, unsigned long old) @@ -685,7 +684,6 @@ static struct rt6_info *rt6_select(struct fib6_node *fn, int oif, int strict) 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; @@ -730,8 +728,7 @@ 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); @@ -739,8 +736,7 @@ 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); @@ -1150,7 +1146,7 @@ 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; @@ -1158,11 +1154,12 @@ void ip6_update_pmtu(struct sk_buff *skb, struct net *net, __be32 mtu, memset(&fl6, 0, sizeof(fl6)); fl6.flowi6_oif = oif; - fl6.flowi6_mark = mark; + fl6.flowi6_mark = mark ? mark : IP6_REPLY_MARK(net, skb->mark); fl6.flowi6_flags = 0; 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) @@ -1174,7 +1171,7 @@ 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, sock_i_uid(sk)); } EXPORT_SYMBOL_GPL(ip6_sk_update_pmtu); @@ -1849,15 +1846,16 @@ static struct rt6_info *ip6_rt_copy(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; @@ -1867,7 +1865,7 @@ 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; @@ -1881,21 +1879,20 @@ 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_table = RT6_TABLE_INFO, + .fc_table = addrconf_rt_table(dev, RT6_TABLE_INFO), .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_dst = *prefix; @@ -1907,7 +1904,7 @@ 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 @@ -1916,7 +1913,8 @@ 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; @@ -1938,7 +1936,7 @@ struct rt6_info *rt6_add_dflt_router(const struct in6_addr *gwaddr, unsigned int pref) { struct fib6_config cfg = { - .fc_table = RT6_TABLE_DFLT, + .fc_table = 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 | @@ -1955,28 +1953,17 @@ 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, 0, NULL); } static void rtmsg_to_fib6_config(struct net *net, @@ -2261,6 +2248,7 @@ static const struct nla_policy rtm_ipv6_policy[RTA_MAX+1] = { [RTA_PRIORITY] = { .type = NLA_U32 }, [RTA_METRICS] = { .type = NLA_NESTED }, [RTA_MULTIPATH] = { .len = sizeof(struct rtnexthop) }, + [RTA_UID] = { .type = NLA_U32 }, }; static int rtm_to_fib6_config(struct sk_buff *skb, struct nlmsghdr *nlh, @@ -2647,6 +2635,15 @@ static int inet6_rtm_getroute(struct sk_buff *in_skb, struct nlmsghdr* nlh) if (tb[RTA_OIF]) oif = nla_get_u32(tb[RTA_OIF]); + 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 d5dda20bd717..ba8622daffd7 100644 --- a/net/ipv6/syncookies.c +++ b/net/ipv6/syncookies.c @@ -212,6 +212,8 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) ipv6_addr_type(&ireq6->rmt_addr) & IPV6_ADDR_LINKLOCAL) ireq6->iif = inet6_iif(skb); + ireq->ir_mark = inet_request_mark(sk, skb); + req->expires = 0UL; req->num_retrans = 0; ireq->ecn_ok = ecn_ok; @@ -238,9 +240,10 @@ struct sock *cookie_v6_check(struct sock *sk, struct sk_buff *skb) final_p = fl6_update_dst(&fl6, np->opt, &final); fl6.saddr = ireq6->loc_addr; fl6.flowi6_oif = sk->sk_bound_dev_if; - fl6.flowi6_mark = sk->sk_mark; + fl6.flowi6_mark = ireq->ir_mark; fl6.fl6_dport = inet_rsk(req)->rmt_port; fl6.fl6_sport = inet_sk(sk)->inet_sport; + fl6.flowi6_uid = sock_i_uid(sk); security_req_classify_flow(req, flowi6_to_flowi(&fl6)); dst = ip6_dst_lookup_flow(sk, &fl6, final_p, false); diff --git a/net/ipv6/sysctl_net_ipv6.c b/net/ipv6/sysctl_net_ipv6.c index e85c48bd404f..53a9f5a64536 100644 --- a/net/ipv6/sysctl_net_ipv6.c +++ b/net/ipv6/sysctl_net_ipv6.c @@ -24,6 +24,13 @@ static ctl_table ipv6_table_template[] = { .mode = 0644, .proc_handler = proc_dointvec }, + { + .procname = "fwmark_reflect", + .data = &init_net.ipv6.sysctl.fwmark_reflect, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec + }, { } }; diff --git a/net/ipv6/tcp_ipv6.c b/net/ipv6/tcp_ipv6.c index 4659b8ab55d9..a4738211c601 100644 --- a/net/ipv6/tcp_ipv6.c +++ b/net/ipv6/tcp_ipv6.c @@ -252,6 +252,7 @@ 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 = sock_i_uid(sk); final_p = fl6_update_dst(&fl6, np->opt, &final); @@ -791,6 +792,7 @@ static void tcp_v6_send_response(struct sk_buff *skb, u32 seq, u32 ack, u32 win, fl6.flowi6_proto = IPPROTO_TCP; if (ipv6_addr_type(&fl6.daddr) & IPV6_ADDR_LINKLOCAL) fl6.flowi6_oif = inet6_iif(skb); + fl6.flowi6_mark = IP6_REPLY_MARK(net, skb->mark); fl6.fl6_dport = t1->dest; fl6.fl6_sport = t1->source; security_skb_classify_flow(skb, flowi6_to_flowi(&fl6)); @@ -999,6 +1001,7 @@ static int tcp_v6_conn_request(struct sock *sk, struct sk_buff *skb) TCP_ECN_create_request(req, skb, sock_net(sk)); treq->iif = sk->sk_bound_dev_if; + inet_rsk(req)->ir_mark = inet_request_mark(sk, skb); /* So that link locals have meaning */ if (!sk->sk_bound_dev_if && diff --git a/net/ipv6/udp.c b/net/ipv6/udp.c index 6b298dc614e3..e06772c4bccb 100644 --- a/net/ipv6/udp.c +++ b/net/ipv6/udp.c @@ -373,8 +373,10 @@ int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk, int is_udp4; bool slow; - if (flags & MSG_ERRQUEUE) - return ipv6_recv_error(sk, msg, len, addr_len); + if (flags & MSG_ERRQUEUE) { + *addr_len = sizeof(struct sockaddr_in6); + return ipv6_recv_error(sk, msg, len); + } if (np->rxpmtu && np->rxopt.bits.rxpmtu) return ipv6_recv_rxpmtu(sk, msg, len, addr_len); @@ -1149,6 +1151,7 @@ do_udp_sendmsg: fl6.flowi6_oif = np->sticky_pktinfo.ipi6_ifindex; fl6.flowi6_mark = sk->sk_mark; + fl6.flowi6_uid = sock_i_uid(sk); if (msg->msg_controllen) { opt = &opt_space; diff --git a/net/netfilter/Kconfig b/net/netfilter/Kconfig index 56d22cae5906..bdf9c60b296a 100644 --- a/net/netfilter/Kconfig +++ b/net/netfilter/Kconfig @@ -1089,6 +1089,8 @@ 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 @@ -1122,6 +1124,22 @@ 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 @@ -1132,6 +1150,30 @@ 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 + depends on IP_NF_TARGET_ULOG=n # not yes, not module, just no + 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 a1abf87d43bf..d9655f6f3466 100644 --- a/net/netfilter/Makefile +++ b/net/netfilter/Makefile @@ -127,7 +127,9 @@ obj-$(CONFIG_NETFILTER_XT_MATCH_OWNER) += xt_owner.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 f407ebc13481..ddf77f7fbe24 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 +39,16 @@ #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> struct idletimer_tg_attr { struct attribute attr; @@ -55,14 +64,110 @@ 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(×tamp_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(×tamp_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 +188,7 @@ 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 +198,15 @@ 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 +215,9 @@ 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 +225,55 @@ 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(×tamp_lock); + timer->active = false; + timer->work_pending = true; schedule_work(&timer->work); + spin_unlock_bh(×tamp_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(×tamp_lock); + if (!timer->active) { + spin_unlock_bh(×tamp_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(×tamp_lock); + break; + default: + break; + } + return NOTIFY_DONE; } static int idletimer_tg_create(struct idletimer_tg_info *info) @@ -145,6 +305,21 @@ 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); @@ -161,6 +336,46 @@ 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(×tamp_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)) { + struct sock *sk = skb->sk; + read_lock_bh(&sk->sk_callback_lock); + if ((sk->sk_socket) && (sk->sk_socket->file) && + (sk->sk_socket->file->f_cred)) + timer->uid = sk->sk_socket->file->f_cred->uid; + read_unlock_bh(&sk->sk_callback_lock); + } + + /* 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(×tamp_lock); +} + /* * The actual xt_tables plugin. */ @@ -168,15 +383,23 @@ 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; } @@ -185,7 +408,7 @@ 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"); @@ -204,9 +427,7 @@ 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 { @@ -219,6 +440,7 @@ static int idletimer_tg_checkentry(const struct xt_tgchk_param *par) } mutex_unlock(&list_mutex); + return 0; } @@ -236,11 +458,12 @@ 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); 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); @@ -248,6 +471,7 @@ 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), @@ -313,3 +537,4 @@ 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..1f576dab9472 --- /dev/null +++ b/net/netfilter/xt_qtaguid.c @@ -0,0 +1,3017 @@ +/* + * 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/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(!current_fsuid()) || unlikely(!proc_ctrl_write_limited) + || unlikely(current_fsuid() == xt_qtaguid_ctrl_file->uid); +} + +static bool can_impersonate_uid(uid_t uid) +{ + return uid == current_fsuid() || can_manipulate_uids(); +} + +static bool can_read_other_uid_stats(uid_t uid) +{ + /* root pwnd */ + return in_egroup_p(xt_qtaguid_stats_file->gid) + || unlikely(!current_fsuid()) || uid == current_fsuid() + || unlikely(!proc_stats_readall_limited) + || unlikely(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); + sockfd_put(st_entry->socket); + 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, 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 { + long 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, 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); +} + +/* + * 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 = par->in ? IFS_RX : IFS_TX; + int bytes = skb->len; + int proto; + + if (!skb->dev) { + MT_DEBUG("qtaguid[%d]: no skb->dev\n", par->hooknum); + el_dev = par->in ? : par->out; + } else { + const struct net_device *other_dev; + el_dev = skb->dev; + other_dev = par->in ? : par->out; + if (el_dev != other_dev) { + MT_DEBUG("qtaguid[%d]: skb->dev=%p %s vs " + "par->(in/out)=%p %s\n", + par->hooknum, el_dev, el_dev->name, other_dev, + other_dev->name); + } + } + + if (unlikely(!el_dev)) { + pr_err_ratelimited("qtaguid[%d]: %s(): no par->in/out?!!\n", + par->hooknum, __func__); + BUG(); + } else if (unlikely(!el_dev->name)) { + pr_err_ratelimited("qtaguid[%d]: %s(): no dev->name?!!\n", + par->hooknum, __func__); + BUG(); + } else { + proto = ipx_proto(skb, par); + MT_DEBUG("qtaguid[%d]: dev name=%s type=%d fam=%d proto=%d\n", + par->hooknum, el_dev->name, el_dev->type, + par->family, proto); + } + + spin_lock_bh(&iface_stat_list_lock); + entry = get_iface_entry(el_dev->name); + if (entry == NULL) { + IF_DEBUG("qtaguid: iface_stat: %s(%s): not tracked\n", + __func__, el_dev->name); + spin_unlock_bh(&iface_stat_list_lock); + return; + } + + IF_DEBUG("qtaguid: %s(%s): entry=%p\n", __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); + + + iface_entry = get_iface_entry(ifname); + if (!iface_entry) { + pr_err_ratelimited("qtaguid: iface_stat: stat_update() " + "%s not found\n", ifname); + return; + } + /* It is ok to process data when an iface_entry is inactive */ + + MT_DEBUG("qtaguid: iface_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: iface_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); + spin_unlock_bh(&iface_entry->tag_stat_list_lock); + return; + } + + /* 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); +} + +static int iface_netdev_event_handler(struct notifier_block *nb, + unsigned long event, void *ptr) { + struct net_device *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: find_sk(skb=%p) hooknum=%d family=%d\n", skb, + par->hooknum, 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_get6_sk(skb, par); + break; + case NFPROTO_IPV4: + sk = xt_socket_get4_sk(skb, par); + break; + default: + return NULL; + } + + if (sk) { + MT_DEBUG("qtaguid: %p->sk_proto=%u " + "->sk_state=%d\n", 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->sk_state == TCP_TIME_WAIT) { + xt_socket_put_sk(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; + + if (!skb->dev) { + MT_DEBUG("qtaguid[%d]: no skb->dev\n", par->hooknum); + el_dev = par->in ? : par->out; + } else { + const struct net_device *other_dev; + el_dev = skb->dev; + other_dev = par->in ? : par->out; + if (el_dev != other_dev) { + MT_DEBUG("qtaguid[%d]: skb->dev=%p %s vs " + "par->(in/out)=%p %s\n", + par->hooknum, el_dev, el_dev->name, other_dev, + other_dev->name); + } + } + + if (unlikely(!el_dev)) { + pr_info("qtaguid[%d]: no par->in/out?!!\n", par->hooknum); + } else if (unlikely(!el_dev->name)) { + pr_info("qtaguid[%d]: no dev->name?!!\n", par->hooknum); + } else { + int proto = ipx_proto(skb, par); + MT_DEBUG("qtaguid[%d]: dev name=%s type=%d fam=%d proto=%d\n", + par->hooknum, el_dev->name, el_dev->type, + par->family, proto); + + if_tag_stat_update(el_dev->name, uid, + skb->sk ? skb->sk : alternate_sk, + par->in ? IFS_RX : IFS_TX, + 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; + uid_t sock_uid; + bool res; + bool set_sk_callback_lock = false; + + 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->sk; + /* + * 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 != NULL) { + 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; + MT_DEBUG("qtaguid[%d]: filp...uid=%u\n", + par->hooknum, filp ? filp->f_cred->fsuid : -1); + } + + if (sk == NULL || sk->sk_socket == NULL) { + /* + * Here, the qtaguid_find_sk() using connection tracking + * couldn't find the owner, so for now we just count them + * against the system. + */ + /* + * TODO: unhack how to force just accounting. + * For now we only do iface stats when the uid-owner is not + * requested. + */ + if (!(info->match & XT_QTAGUID_UID)) + account_for_uid(skb, sk, 0, par); + MT_DEBUG("qtaguid[%d]: leaving (sk?sk->sk_socket)=%p\n", + par->hooknum, + sk ? sk->sk_socket : NULL); + 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; + } + filp = sk->sk_socket->file; + if (filp == NULL) { + MT_DEBUG("qtaguid[%d]: leaving filp=NULL\n", par->hooknum); + account_for_uid(skb, sk, 0, par); + res = ((info->match ^ info->invert) & + (XT_QTAGUID_UID | XT_QTAGUID_GID)) == 0; + atomic64_inc(&qtu_events.match_no_sk_file); + goto put_sock_ret_res; + } + sock_uid = filp->f_cred->fsuid; + /* + * TODO: unhack how to force just accounting. + * For now we only do iface stats when the uid-owner is not requested + */ + if (!(info->match & XT_QTAGUID_UID)) + account_for_uid(skb, sk, 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) + if ((filp->f_cred->fsuid >= info->uid_min && + filp->f_cred->fsuid <= info->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) + if ((filp->f_cred->fsgid >= info->gid_min && + filp->f_cred->fsgid <= info->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) + xt_socket_put_sk(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 */ +static void prdebug_full_state(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); + + spin_lock_bh(&sock_tag_list_lock); + prdebug_sock_tag_tree(indent_level, &sock_tag_tree); + spin_unlock_bh(&sock_tag_list_lock); + + spin_lock_bh(&sock_tag_list_lock); + 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_unlock_bh(&sock_tag_list_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(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; + long f_count; + + CT_DEBUG("qtaguid: proc ctrl pid=%u tgid=%u uid=%u\n", + current->pid, current->tgid, current_fsuid()); + + if (sock_tag_entry != SEQ_START_TOKEN) { + 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 + ); + f_count = atomic_long_read( + &sock_tag_entry->socket->file->f_count); + seq_printf(m, "sock=%p tag=0x%llx (uid=%u) pid=%u " + "f_count=%lu\n", + sock_tag_entry->sk, + sock_tag_entry->tag, uid, + sock_tag_entry->pid, f_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_file=%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_file)); + + /* Count the following as part of the last item_index */ + prdebug_full_state(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; + uid_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); + CT_DEBUG("qtaguid: ctrl_delete(%s): argc=%d cmd=%c " + "user_tag=0x%llx uid=%u\n", input, argc, cmd, + acct_tag, uid); + 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(); + } 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, current_fsuid()); + res = -EPERM; + goto err; + } + + tag = combine_atag_with_uid(acct_tag, uid); + CT_DEBUG("qtaguid: ctrl_delete(%s): " + "looking for tag=0x%llx (uid=%u)\n", + input, tag, uid); + + /* 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) + 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) + 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) + 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, 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; + uid_t uid = 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); + 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); + 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, + current_fsuid()); + goto err; + } + CT_DEBUG("qtaguid: ctrl_tag(%s): socket->...->f_count=%ld ->sk=%p\n", + input, atomic_long_read(&el_socket->file->f_count), + 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, current_uid(), + current_euid(), current_fsuid(), + 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(); + } 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, current_fsuid()); + res = -EPERM; + goto err_put; + } + full_tag = combine_atag_with_uid(acct_tag, uid); + + 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 ...->f_count=%ld\n", + input, el_socket->sk, sock_tag_entry, + atomic_long_read(&el_socket->file->f_count)); + /* + * This is a re-tagging, so release the sock_fd that was + * locked at the time of the 1st tagging. + * There is still the ref from this call's sockfd_lookup() so + * it can be done within the spinlock. + */ + sockfd_put(sock_tag_entry->socket); + 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; + } + sock_tag_entry->sk = el_socket->sk; + sock_tag_entry->socket = el_socket; + sock_tag_entry->pid = current->tgid; + sock_tag_entry->tag = combine_atag_with_uid(acct_tag, + uid); + 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, + 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 socket (file) until it is untagged */ + CT_DEBUG("qtaguid: ctrl_tag(%s): done st@%p ...->f_count=%ld\n", + input, sock_tag_entry, + atomic_long_read(&el_socket->file->f_count)); + 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. ...->f_count=%ld\n", + input, atomic_long_read(&el_socket->file->f_count) - 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; + struct sock_tag *sock_tag_entry; + struct tag_ref *tag_ref_entry; + struct uid_tag_data *utd_entry; + struct proc_qtu_data *pqd_entry; + + 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; + goto err; + } + 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, + current_fsuid()); + goto err; + } + 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); + 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; + goto err_put; + } + /* + * 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); + 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, 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, + * and once more for the sockfd_lookup() here. + */ + sockfd_put(sock_tag_entry->socket); + CT_DEBUG("qtaguid: ctrl_untag(%s): done. st@%p ...->f_count=%ld\n", + input, sock_tag_entry, + atomic_long_read(&el_socket->file->f_count) - 1); + sockfd_put(el_socket); + + kfree(sock_tag_entry); + atomic64_inc(&qtu_events.sockets_untagged); + + return 0; + +err_put: + CT_DEBUG("qtaguid: ctrl_untag(%s): done. socket->...->f_count=%ld\n", + input, atomic_long_read(&el_socket->file->f_count) - 1); + /* Release the sock_fd that was grabbed by sockfd_lookup(). */ + sockfd_put(el_socket); + return res; + +err: + CT_DEBUG("qtaguid: ctrl_untag(%s): done.\n", input); + return res; +} + +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, 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) +{ + int ret; + 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 (get_atag_from_tag(tag) && !can_read_other_uid_stats(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, current_fsuid(), + xt_qtaguid_stats_file->gid); + return 0; + } + ppi->item_index++; + cnts = &ts_entry->counters; + ret = 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 ret ?: 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, current_fsuid()); + + spin_lock_bh(&uid_tag_data_tree_lock); + + /* Look for existing uid data, or alloc one. */ + utd_entry = get_uid_data(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, 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, 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", + 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); + + prdebug_full_state(0, "%s(): pid=%u tgid=%u", __func__, + current->pid, current->tgid); + 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..6dc14a9c6889 --- /dev/null +++ b/net/netfilter/xt_qtaguid_internal.h @@ -0,0 +1,352 @@ +/* + * 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 */ + /* The socket is needed for sockfd_put() */ + struct socket *socket; + /* 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. + * This might happen for traffic while the socket is being closed. + */ + atomic64_t match_no_sk_file; +}; + +/* 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..f6a00a3520ed --- /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 "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 socket=%p (f_count=%lu), list=list_head{...}, " + "pid=%u, tag=%s}", + st, st->sk, st->socket, atomic_long_read( + &st->socket->file->f_count), + 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..4328562572f6 --- /dev/null +++ b/net/netfilter/xt_quota2.c @@ -0,0 +1,385 @@ +/* + * 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 +#include <linux/netfilter_ipv4/ipt_ULOG.h> +#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 unsigned int quota_list_uid = 0; +static unsigned int quota_list_gid = 0; +module_param_named(perms, quota_list_perms, uint, S_IRUGO | S_IWUSR); +module_param_named(uid, quota_list_uid, uint, S_IRUGO | S_IWUSR); +module_param_named(gid, quota_list_gid, 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 63b2bdb59e95..030f33cd9ee9 100644 --- a/net/netfilter/xt_socket.c +++ b/net/netfilter/xt_socket.c @@ -35,7 +35,7 @@ #include <net/netfilter/nf_conntrack.h> #endif -static void +void xt_socket_put_sk(struct sock *sk) { if (sk->sk_state == TCP_TIME_WAIT) @@ -43,6 +43,7 @@ xt_socket_put_sk(struct sock *sk) else sock_put(sk); } +EXPORT_SYMBOL(xt_socket_put_sk); static int extract_icmp4_fields(const struct sk_buff *skb, @@ -101,9 +102,8 @@ extract_icmp4_fields(const struct sk_buff *skb, return 0; } -static bool -socket_match(const struct sk_buff *skb, struct xt_action_param *par, - const struct xt_socket_mtinfo1 *info) +struct sock* +xt_socket_get4_sk(const struct sk_buff *skb, struct xt_action_param *par) { const struct iphdr *iph = ip_hdr(skb); struct udphdr _hdr, *hp = NULL; @@ -120,7 +120,7 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par, hp = skb_header_pointer(skb, ip_hdrlen(skb), sizeof(_hdr), &_hdr); if (hp == NULL) - return false; + return NULL; protocol = iph->protocol; saddr = iph->saddr; @@ -131,9 +131,9 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par, } else if (iph->protocol == IPPROTO_ICMP) { if (extract_icmp4_fields(skb, &protocol, &saddr, &daddr, &sport, &dport)) - return false; + return NULL; } else { - return false; + return NULL; } #ifdef XT_SOCKET_HAVE_CONNTRACK @@ -157,6 +157,23 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par, sk = nf_tproxy_get_sock_v4(dev_net(skb->dev), protocol, saddr, daddr, sport, dport, par->in, NFT_LOOKUP_ANY); + + pr_debug("proto %hhu %pI4:%hu -> %pI4:%hu (orig %pI4:%hu) sock %p\n", + protocol, &saddr, ntohs(sport), + &daddr, ntohs(dport), + &iph->daddr, hp ? ntohs(hp->dest) : 0, sk); + + return sk; +} +EXPORT_SYMBOL(xt_socket_get4_sk); + +static bool +socket_match(const struct sk_buff *skb, struct xt_action_param *par, + const struct xt_socket_mtinfo1 *info) +{ + struct sock *sk; + + sk = xt_socket_get4_sk(skb, par); if (sk != NULL) { bool wildcard; bool transparent = true; @@ -179,11 +196,6 @@ socket_match(const struct sk_buff *skb, struct xt_action_param *par, sk = NULL; } - pr_debug("proto %hhu %pI4:%hu -> %pI4:%hu (orig %pI4:%hu) sock %p\n", - protocol, &saddr, ntohs(sport), - &daddr, ntohs(dport), - &iph->daddr, hp ? ntohs(hp->dest) : 0, sk); - return (sk != NULL); } @@ -255,8 +267,8 @@ extract_icmp6_fields(const struct sk_buff *skb, return 0; } -static bool -socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par) +struct sock* +xt_socket_get6_sk(const struct sk_buff *skb, struct xt_action_param *par) { struct ipv6hdr *iph = ipv6_hdr(skb); struct udphdr _hdr, *hp = NULL; @@ -264,7 +276,6 @@ socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par) struct in6_addr *daddr = NULL, *saddr = NULL; __be16 uninitialized_var(dport), uninitialized_var(sport); int thoff = 0, uninitialized_var(tproto); - const struct xt_socket_mtinfo1 *info = (struct xt_socket_mtinfo1 *) par->matchinfo; tproto = ipv6_find_hdr(skb, &thoff, -1, NULL, NULL); if (tproto < 0) { @@ -276,7 +287,7 @@ socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par) hp = skb_header_pointer(skb, thoff, sizeof(_hdr), &_hdr); if (hp == NULL) - return false; + return NULL; saddr = &iph->saddr; sport = hp->source; @@ -286,13 +297,30 @@ socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par) } else if (tproto == IPPROTO_ICMPV6) { if (extract_icmp6_fields(skb, thoff, &tproto, &saddr, &daddr, &sport, &dport)) - return false; + return NULL; } else { - return false; + return NULL; } sk = nf_tproxy_get_sock_v6(dev_net(skb->dev), tproto, saddr, daddr, sport, dport, par->in, NFT_LOOKUP_ANY); + pr_debug("proto %hhd %pI6:%hu -> %pI6:%hu " + "(orig %pI6:%hu) sock %p\n", + tproto, saddr, ntohs(sport), + daddr, ntohs(dport), + &iph->daddr, hp ? ntohs(hp->dest) : 0, sk); + return sk; +} +EXPORT_SYMBOL(xt_socket_get6_sk); + +static bool +socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par) +{ + struct sock *sk; + const struct xt_socket_mtinfo1 *info; + + info = (struct xt_socket_mtinfo1 *) par->matchinfo; + sk = xt_socket_get6_sk(skb, par); if (sk != NULL) { bool wildcard; bool transparent = true; @@ -315,12 +343,6 @@ socket_mt6_v1(const struct sk_buff *skb, struct xt_action_param *par) sk = NULL; } - pr_debug("proto %hhd %pI6:%hu -> %pI6:%hu " - "(orig %pI6:%hu) sock %p\n", - tproto, saddr, ntohs(sport), - daddr, ntohs(dport), - &iph->daddr, hp ? ntohs(hp->dest) : 0, sk); - return (sk != NULL); } #endif diff --git a/net/rfkill/Kconfig b/net/rfkill/Kconfig index 78efe895b663..8e12c8a2b82b 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 1cec5e4f3a5e..c099b4fffd93 100644 --- a/net/rfkill/core.c +++ b/net/rfkill/core.c @@ -792,6 +792,7 @@ void rfkill_pause_polling(struct rfkill *rfkill) } EXPORT_SYMBOL(rfkill_pause_polling); +#ifdef CONFIG_RFKILL_PM void rfkill_resume_polling(struct rfkill *rfkill) { BUG_ON(!rfkill); @@ -826,14 +827,17 @@ static int rfkill_resume(struct device *dev) return 0; } +#endif static struct class rfkill_class = { .name = "rfkill", .dev_release = rfkill_release, .dev_attrs = rfkill_dev_attrs, .dev_uevent = rfkill_dev_uevent, +#ifdef CONFIG_RFKILL_PM .suspend = rfkill_suspend, .resume = rfkill_resume, +#endif }; bool rfkill_blocked(struct rfkill *rfkill) diff --git a/net/sunrpc/sched.c b/net/sunrpc/sched.c index 5356b120dbf8..77d251e02593 100644 --- a/net/sunrpc/sched.c +++ b/net/sunrpc/sched.c @@ -254,7 +254,7 @@ static int rpc_wait_bit_killable(void *word) { if (fatal_signal_pending(current)) return -ERESTARTSYS; - freezable_schedule(); + freezable_schedule_unsafe(); return 0; } diff --git a/net/unix/af_unix.c b/net/unix/af_unix.c index 75e198d029d2..c80c107139f1 100644 --- a/net/unix/af_unix.c +++ b/net/unix/af_unix.c @@ -114,6 +114,7 @@ #include <linux/mount.h> #include <net/checksum.h> #include <linux/security.h> +#include <linux/freezer.h> struct hlist_head unix_socket_table[2 * UNIX_HASH_SIZE]; EXPORT_SYMBOL_GPL(unix_socket_table); @@ -1896,7 +1897,7 @@ static long unix_stream_data_wait(struct sock *sk, long timeo, set_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); unix_state_unlock(sk); - timeo = schedule_timeout(timeo); + timeo = freezable_schedule_timeout(timeo); unix_state_lock(sk); clear_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags); } diff --git a/net/wireless/Kconfig b/net/wireless/Kconfig index 16d08b399210..4c602d100480 100644 --- a/net/wireless/Kconfig +++ b/net/wireless/Kconfig @@ -166,3 +166,14 @@ config LIB80211_DEBUG from lib80211. If unsure, say N. + +config CFG80211_ALLOW_RECONNECT + bool "Allow reconnect while already connected" + depends on CFG80211 + default n + help + cfg80211 stack doesn't allow to connect if you are already + connected. This option allows to make a connection in this case. + + Select this option ONLY for wlan drivers that are specifically + built for such purposes. diff --git a/net/wireless/core.h b/net/wireless/core.h index fd35dae547c4..15b4bb7ac046 100644 --- a/net/wireless/core.h +++ b/net/wireless/core.h @@ -77,9 +77,7 @@ struct cfg80211_registered_device { struct mutex sched_scan_mtx; -#ifdef CONFIG_NL80211_TESTMODE - struct genl_info *testmode_info; -#endif + struct genl_info *cur_cmd_info; struct work_struct conn_work; struct work_struct event_work; diff --git a/net/wireless/nl80211.c b/net/wireless/nl80211.c index 90a0e552cb32..e63d881d31b7 100644 --- a/net/wireless/nl80211.c +++ b/net/wireless/nl80211.c @@ -378,6 +378,12 @@ static const struct nla_policy nl80211_policy[NL80211_ATTR_MAX+1] = { [NL80211_ATTR_MDID] = { .type = NLA_U16 }, [NL80211_ATTR_IE_RIC] = { .type = NLA_BINARY, .len = IEEE80211_MAX_DATA_LEN }, + [NL80211_ATTR_PEER_AID] = { .type = NLA_U16 }, + [NL80211_ATTR_STA_SUPPORTED_CHANNELS] = { .type = NLA_BINARY }, + [NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES] = { .type = NLA_BINARY }, + [NL80211_ATTR_VENDOR_ID] = { .type = NLA_U32 }, + [NL80211_ATTR_VENDOR_SUBCMD] = { .type = NLA_U32 }, + [NL80211_ATTR_VENDOR_DATA] = { .type = NLA_BINARY }, }; /* policy for the key attributes */ @@ -1520,6 +1526,39 @@ static int nl80211_send_wiphy(struct cfg80211_registered_device *dev, dev->wiphy.max_acl_mac_addrs)) goto nla_put_failure; + if (dev->wiphy.n_vendor_commands) { + const struct nl80211_vendor_cmd_info *info; + struct nlattr *nested; + + nested = nla_nest_start(msg, NL80211_ATTR_VENDOR_DATA); + if (!nested) + goto nla_put_failure; + + for (i = 0; i < dev->wiphy.n_vendor_commands; i++) { + info = &dev->wiphy.vendor_commands[i].info; + if (nla_put(msg, i + 1, sizeof(*info), info)) + goto nla_put_failure; + } + nla_nest_end(msg, nested); + } + + if (dev->wiphy.n_vendor_events) { + const struct nl80211_vendor_cmd_info *info; + struct nlattr *nested; + + nested = nla_nest_start(msg, + NL80211_ATTR_VENDOR_EVENTS); + if (!nested) + goto nla_put_failure; + + for (i = 0; i < dev->wiphy.n_vendor_events; i++) { + info = &dev->wiphy.vendor_events[i]; + if (nla_put(msg, i + 1, sizeof(*info), info)) + goto nla_put_failure; + } + nla_nest_end(msg, nested); + } + /* * Any information below this point is only available to * applications that can deal with it being split. This @@ -2638,8 +2677,8 @@ static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info) hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0, NL80211_CMD_NEW_KEY); - if (IS_ERR(hdr)) - return PTR_ERR(hdr); + if (!hdr) + goto nla_put_failure; cookie.msg = msg; cookie.idx = key_idx; @@ -6402,12 +6441,62 @@ static int nl80211_set_mcast_rate(struct sk_buff *skb, struct genl_info *info) return err; } +static struct sk_buff * +__cfg80211_alloc_vendor_skb(struct cfg80211_registered_device *rdev, + int approxlen, u32 portid, u32 seq, + enum nl80211_commands cmd, + enum nl80211_attrs attr, + const struct nl80211_vendor_cmd_info *info, + gfp_t gfp) +{ + struct sk_buff *skb; + void *hdr; + struct nlattr *data; + + skb = nlmsg_new(approxlen + 100, gfp); + if (!skb) + return NULL; + + hdr = nl80211hdr_put(skb, portid, seq, 0, cmd); + if (!hdr) { + kfree_skb(skb); + return NULL; + } + + if (nla_put_u32(skb, NL80211_ATTR_WIPHY, rdev->wiphy_idx)) + goto nla_put_failure; + + if (info) { + if (nla_put_u32(skb, NL80211_ATTR_VENDOR_ID, + info->vendor_id)) + goto nla_put_failure; + if (nla_put_u32(skb, NL80211_ATTR_VENDOR_SUBCMD, + info->subcmd)) + goto nla_put_failure; + } + + data = nla_nest_start(skb, attr); + + ((void **)skb->cb)[0] = rdev; + ((void **)skb->cb)[1] = hdr; + ((void **)skb->cb)[2] = data; + + return skb; + + nla_put_failure: + kfree_skb(skb); + return NULL; +} #ifdef CONFIG_NL80211_TESTMODE static struct genl_multicast_group nl80211_testmode_mcgrp = { .name = "testmode", }; +static struct genl_multicast_group nl80211_vendor_mcgrp = { + .name = "vendor", +}; + static int nl80211_testmode_do(struct sk_buff *skb, struct genl_info *info) { struct cfg80211_registered_device *rdev = info->user_ptr[0]; @@ -6418,11 +6507,11 @@ static int nl80211_testmode_do(struct sk_buff *skb, struct genl_info *info) err = -EOPNOTSUPP; if (rdev->ops->testmode_cmd) { - rdev->testmode_info = info; + rdev->cur_cmd_info = info; err = rdev_testmode_cmd(rdev, nla_data(info->attrs[NL80211_ATTR_TESTDATA]), nla_len(info->attrs[NL80211_ATTR_TESTDATA])); - rdev->testmode_info = NULL; + rdev->cur_cmd_info = NULL; } return err; @@ -6491,6 +6580,9 @@ static int nl80211_testmode_dump(struct sk_buff *skb, NL80211_CMD_TESTMODE); struct nlattr *tmdata; + if (!hdr) + break; + if (nla_put_u32(skb, NL80211_ATTR_WIPHY, phy_idx)) { genlmsg_cancel(skb, hdr); break; @@ -6523,84 +6615,37 @@ static int nl80211_testmode_dump(struct sk_buff *skb, return err; } -static struct sk_buff * -__cfg80211_testmode_alloc_skb(struct cfg80211_registered_device *rdev, - int approxlen, u32 portid, u32 seq, gfp_t gfp) -{ - struct sk_buff *skb; - void *hdr; - struct nlattr *data; - - skb = nlmsg_new(approxlen + 100, gfp); - if (!skb) - return NULL; - - hdr = nl80211hdr_put(skb, portid, seq, 0, NL80211_CMD_TESTMODE); - if (!hdr) { - kfree_skb(skb); - return NULL; - } - - if (nla_put_u32(skb, NL80211_ATTR_WIPHY, rdev->wiphy_idx)) - goto nla_put_failure; - data = nla_nest_start(skb, NL80211_ATTR_TESTDATA); - - ((void **)skb->cb)[0] = rdev; - ((void **)skb->cb)[1] = hdr; - ((void **)skb->cb)[2] = data; - - return skb; - - nla_put_failure: - kfree_skb(skb); - return NULL; -} - -struct sk_buff *cfg80211_testmode_alloc_reply_skb(struct wiphy *wiphy, - int approxlen) +struct sk_buff *__cfg80211_alloc_event_skb(struct wiphy *wiphy, + enum nl80211_commands cmd, + enum nl80211_attrs attr, + int vendor_event_idx, + int approxlen, gfp_t gfp) { struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + const struct nl80211_vendor_cmd_info *info; - if (WARN_ON(!rdev->testmode_info)) + switch (cmd) { + case NL80211_CMD_TESTMODE: + if (WARN_ON(vendor_event_idx != -1)) + return NULL; + info = NULL; + break; + case NL80211_CMD_VENDOR: + if (WARN_ON(vendor_event_idx < 0 || + vendor_event_idx >= wiphy->n_vendor_events)) + return NULL; + info = &wiphy->vendor_events[vendor_event_idx]; + break; + default: + WARN_ON(1); return NULL; - - return __cfg80211_testmode_alloc_skb(rdev, approxlen, - rdev->testmode_info->snd_portid, - rdev->testmode_info->snd_seq, - GFP_KERNEL); -} -EXPORT_SYMBOL(cfg80211_testmode_alloc_reply_skb); - -int cfg80211_testmode_reply(struct sk_buff *skb) -{ - struct cfg80211_registered_device *rdev = ((void **)skb->cb)[0]; - void *hdr = ((void **)skb->cb)[1]; - struct nlattr *data = ((void **)skb->cb)[2]; - - /* clear CB data for netlink core to own from now on */ - memset(skb->cb, 0, sizeof(skb->cb)); - - if (WARN_ON(!rdev->testmode_info)) { - kfree_skb(skb); - return -EINVAL; } - - nla_nest_end(skb, data); - genlmsg_end(skb, hdr); - return genlmsg_reply(skb, rdev->testmode_info); -} -EXPORT_SYMBOL(cfg80211_testmode_reply); - -struct sk_buff *cfg80211_testmode_alloc_event_skb(struct wiphy *wiphy, - int approxlen, gfp_t gfp) -{ - struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); - - return __cfg80211_testmode_alloc_skb(rdev, approxlen, 0, 0, gfp); + return __cfg80211_alloc_vendor_skb(rdev, approxlen, 0, 0, + cmd, attr, info, gfp); } -EXPORT_SYMBOL(cfg80211_testmode_alloc_event_skb); +EXPORT_SYMBOL(__cfg80211_alloc_event_skb); -void cfg80211_testmode_event(struct sk_buff *skb, gfp_t gfp) +void __cfg80211_send_event_skb(struct sk_buff *skb, gfp_t gfp) { struct cfg80211_registered_device *rdev = ((void **)skb->cb)[0]; void *hdr = ((void **)skb->cb)[1]; @@ -6611,10 +6656,15 @@ void cfg80211_testmode_event(struct sk_buff *skb, gfp_t gfp) nla_nest_end(skb, data); genlmsg_end(skb, hdr); - genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), skb, 0, - nl80211_testmode_mcgrp.id, gfp); + + if (data->nla_type == NL80211_ATTR_VENDOR_DATA) + genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), skb, 0, + nl80211_vendor_mcgrp.id, gfp); + else + genlmsg_multicast_netns(wiphy_net(&rdev->wiphy), skb, 0, + nl80211_testmode_mcgrp.id, gfp); } -EXPORT_SYMBOL(cfg80211_testmode_event); +EXPORT_SYMBOL(__cfg80211_send_event_skb); #endif static int nl80211_connect(struct sk_buff *skb, struct genl_info *info) @@ -6935,9 +6985,8 @@ static int nl80211_remain_on_channel(struct sk_buff *skb, hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0, NL80211_CMD_REMAIN_ON_CHANNEL); - - if (IS_ERR(hdr)) { - err = PTR_ERR(hdr); + if (!hdr) { + err = -ENOBUFS; goto free_msg; } @@ -7224,9 +7273,8 @@ static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info) hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0, NL80211_CMD_FRAME); - - if (IS_ERR(hdr)) { - err = PTR_ERR(hdr); + if (!hdr) { + err = -ENOBUFS; goto free_msg; } } @@ -8087,9 +8135,8 @@ static int nl80211_probe_client(struct sk_buff *skb, hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0, NL80211_CMD_PROBE_CLIENT); - - if (IS_ERR(hdr)) { - err = PTR_ERR(hdr); + if (!hdr) { + err = -ENOBUFS; goto free_msg; } @@ -8306,6 +8353,108 @@ static int nl80211_crit_protocol_stop(struct sk_buff *skb, return 0; } +static int nl80211_vendor_cmd(struct sk_buff *skb, struct genl_info *info) +{ + struct cfg80211_registered_device *rdev = info->user_ptr[0]; + struct wireless_dev *wdev = + __cfg80211_wdev_from_attrs(genl_info_net(info), info->attrs); + int i, err; + u32 vid, subcmd; + + if (!rdev || !rdev->wiphy.vendor_commands) + return -EOPNOTSUPP; + + if (IS_ERR(wdev)) { + err = PTR_ERR(wdev); + if (err != -EINVAL) + return err; + wdev = NULL; + } else if (wdev->wiphy != &rdev->wiphy) { + return -EINVAL; + } + + if (!info->attrs[NL80211_ATTR_VENDOR_ID] || + !info->attrs[NL80211_ATTR_VENDOR_SUBCMD]) + return -EINVAL; + + vid = nla_get_u32(info->attrs[NL80211_ATTR_VENDOR_ID]); + subcmd = nla_get_u32(info->attrs[NL80211_ATTR_VENDOR_SUBCMD]); + for (i = 0; i < rdev->wiphy.n_vendor_commands; i++) { + const struct wiphy_vendor_command *vcmd; + void *data = NULL; + int len = 0; + + vcmd = &rdev->wiphy.vendor_commands[i]; + + if (vcmd->info.vendor_id != vid || vcmd->info.subcmd != subcmd) + continue; + + if (vcmd->flags & (WIPHY_VENDOR_CMD_NEED_WDEV | + WIPHY_VENDOR_CMD_NEED_NETDEV)) { + if (!wdev) + return -EINVAL; + if (vcmd->flags & WIPHY_VENDOR_CMD_NEED_NETDEV && + !wdev->netdev) + return -EINVAL; + + if (vcmd->flags & WIPHY_VENDOR_CMD_NEED_RUNNING) { + if (!wdev->netdev || + !netif_running(wdev->netdev)) + return -ENETDOWN; + } + } else { + wdev = NULL; + } + + if (info->attrs[NL80211_ATTR_VENDOR_DATA]) { + data = nla_data(info->attrs[NL80211_ATTR_VENDOR_DATA]); + len = nla_len(info->attrs[NL80211_ATTR_VENDOR_DATA]); + } + + rdev->cur_cmd_info = info; + err = rdev->wiphy.vendor_commands[i].doit(&rdev->wiphy, wdev, + data, len); + rdev->cur_cmd_info = NULL; + return err; + } + + return -EOPNOTSUPP; +} + +struct sk_buff *__cfg80211_alloc_reply_skb(struct wiphy *wiphy, + enum nl80211_commands cmd, + enum nl80211_attrs attr, + int approxlen) +{ + struct cfg80211_registered_device *rdev = wiphy_to_dev(wiphy); + + if (WARN_ON(!rdev->cur_cmd_info)) + return NULL; + + return __cfg80211_alloc_vendor_skb(rdev, approxlen, + 0, + 0, + cmd, attr, NULL, GFP_KERNEL); +} +EXPORT_SYMBOL(__cfg80211_alloc_reply_skb); + +int cfg80211_vendor_cmd_reply(struct sk_buff *skb) +{ + struct cfg80211_registered_device *rdev = ((void **)skb->cb)[0]; + void *hdr = ((void **)skb->cb)[1]; + struct nlattr *data = ((void **)skb->cb)[2]; + + if (WARN_ON(!rdev->cur_cmd_info)) { + kfree_skb(skb); + return -EINVAL; + } + + nla_nest_end(skb, data); + genlmsg_end(skb, hdr); + return genlmsg_reply(skb, rdev->cur_cmd_info); +} +EXPORT_SYMBOL(cfg80211_vendor_cmd_reply); + #define NL80211_FLAG_NEED_WIPHY 0x01 #define NL80211_FLAG_NEED_NETDEV 0x02 #define NL80211_FLAG_NEED_RTNL 0x04 @@ -9010,7 +9159,15 @@ static struct genl_ops nl80211_ops[] = { .flags = GENL_ADMIN_PERM, .internal_flags = NL80211_FLAG_NEED_WDEV_UP | NL80211_FLAG_NEED_RTNL, - } + }, + { + .cmd = NL80211_CMD_VENDOR, + .doit = nl80211_vendor_cmd, + .policy = nl80211_policy, + .flags = GENL_ADMIN_PERM, + .internal_flags = NL80211_FLAG_NEED_WIPHY | + NL80211_FLAG_NEED_RTNL, + }, }; static struct genl_multicast_group nl80211_mlme_mcgrp = { @@ -10829,6 +10986,10 @@ int nl80211_init(void) err = genl_register_mc_group(&nl80211_fam, &nl80211_testmode_mcgrp); if (err) goto err_out; + + err = genl_register_mc_group(&nl80211_fam, &nl80211_vendor_mcgrp); + if (err) + goto err_out; #endif err = netlink_register_notifier(&nl80211_netlink_notifier); diff --git a/net/wireless/scan.c b/net/wireless/scan.c index 81019ee3ddc8..ab406d0462b7 100644 --- a/net/wireless/scan.c +++ b/net/wireless/scan.c @@ -55,7 +55,7 @@ * also linked into the probe response struct. */ -#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/wireless/sme.c b/net/wireless/sme.c index 3ed35c345cae..e2f74e66a169 100644 --- a/net/wireless/sme.c +++ b/net/wireless/sme.c @@ -707,8 +707,10 @@ void __cfg80211_disconnected(struct net_device *dev, const u8 *ie, wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)) return; +#ifndef CONFIG_CFG80211_ALLOW_RECONNECT if (wdev->sme_state != CFG80211_SME_CONNECTED) return; +#endif if (wdev->current_bss) { cfg80211_unhold_bss(wdev->current_bss); @@ -785,10 +787,14 @@ int __cfg80211_connect(struct cfg80211_registered_device *rdev, ASSERT_WDEV_LOCK(wdev); +#ifndef CONFIG_CFG80211_ALLOW_RECONNECT if (wdev->sme_state != CFG80211_SME_IDLE) return -EALREADY; if (WARN_ON(wdev->connect_keys)) { +#else + if (wdev->connect_keys) { +#endif kfree(wdev->connect_keys); wdev->connect_keys = NULL; } diff --git a/scripts/Makefile.lib b/scripts/Makefile.lib index c4b37f6b5478..76fec0940545 100644 --- a/scripts/Makefile.lib +++ b/scripts/Makefile.lib @@ -274,6 +274,18 @@ $(obj)/%.dtb: $(src)/%.dts FORCE dtc-tmp = $(subst $(comma),_,$(dot-target).dts.tmp) +$(obj)/%.dtb: $(src)/%.dts FORCE + $(call if_changed_dep,dtc) + +dtc-tmp = $(subst $(comma),_,$(dot-target).dts) + +quiet_cmd_dtc_cpp = DTC+CPP $@ +cmd_dtc_cpp = $(CPP) $(dtc_cpp_flags) -x assembler-with-cpp -o $(dtc-tmp) $< ; \ + $(objtree)/scripts/dtc/dtc -O dtb -o $@ -b 0 $(DTC_FLAGS) $(dtc-tmp) + +$(obj)/%.dtb: $(src)/%.dtsp FORCE + $(call if_changed_dep,dtc_cpp) + # Bzip2 # --------------------------------------------------------------------------- diff --git a/scripts/Makefile.modinst b/scripts/Makefile.modinst index ecbb44797e28..8900c5c5390d 100644 --- a/scripts/Makefile.modinst +++ b/scripts/Makefile.modinst @@ -24,7 +24,7 @@ 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/kconfig/confdata.c b/scripts/kconfig/confdata.c index 43eda40c3838..35e0f164ef81 100644 --- a/scripts/kconfig/confdata.c +++ b/scripts/kconfig/confdata.c @@ -1083,7 +1083,7 @@ static void randomize_choice_values(struct symbol *csym) csym->flags &= ~(SYMBOL_VALID); } -static void set_all_choice_values(struct symbol *csym) +void set_all_choice_values(struct symbol *csym) { struct property *prop; struct symbol *sym; @@ -1100,7 +1100,7 @@ static void set_all_choice_values(struct symbol *csym) } csym->flags |= SYMBOL_DEF_USER; /* clear VALID to get value calculated */ - csym->flags &= ~(SYMBOL_VALID); + csym->flags &= ~(SYMBOL_VALID | SYMBOL_NEED_SET_CHOICE_VALUES); } void conf_set_all_new_symbols(enum conf_def_mode mode) @@ -1202,6 +1202,14 @@ void conf_set_all_new_symbols(enum conf_def_mode mode) * selected in a choice block and we set it to yes, * and the rest to no. */ + if (mode != def_random) { + for_all_symbols(i, csym) { + if ((sym_is_choice(csym) && !sym_has_value(csym)) || + sym_is_choice_value(csym)) + csym->flags |= SYMBOL_NEED_SET_CHOICE_VALUES; + } + } + for_all_symbols(i, csym) { if (sym_has_value(csym) || !sym_is_choice(csym)) continue; @@ -1209,7 +1217,5 @@ void conf_set_all_new_symbols(enum conf_def_mode mode) sym_calc_value(csym); if (mode == def_random) randomize_choice_values(csym); - else - set_all_choice_values(csym); } } diff --git a/scripts/kconfig/expr.h b/scripts/kconfig/expr.h index cdd48600e02a..df198a5f4822 100644 --- a/scripts/kconfig/expr.h +++ b/scripts/kconfig/expr.h @@ -106,6 +106,9 @@ struct symbol { #define SYMBOL_DEF3 0x40000 /* symbol.def[S_DEF_3] is valid */ #define SYMBOL_DEF4 0x80000 /* symbol.def[S_DEF_4] is valid */ +/* choice values need to be set before calculating this symbol value */ +#define SYMBOL_NEED_SET_CHOICE_VALUES 0x100000 + #define SYMBOL_MAXLENGTH 256 #define SYMBOL_HASHSIZE 9973 diff --git a/scripts/kconfig/lkc.h b/scripts/kconfig/lkc.h index f8aee5fc6d5e..0c8d4191ca87 100644 --- a/scripts/kconfig/lkc.h +++ b/scripts/kconfig/lkc.h @@ -87,6 +87,7 @@ char *conf_get_default_confname(void); void sym_set_change_count(int count); void sym_add_change_count(int count); void conf_set_all_new_symbols(enum conf_def_mode mode); +void set_all_choice_values(struct symbol *csym); struct conf_printer { void (*print_symbol)(FILE *, struct symbol *, const char *, void *); diff --git a/scripts/kconfig/symbol.c b/scripts/kconfig/symbol.c index ecc5aa5f865d..ab8f4c835933 100644 --- a/scripts/kconfig/symbol.c +++ b/scripts/kconfig/symbol.c @@ -300,6 +300,14 @@ void sym_calc_value(struct symbol *sym) if (sym->flags & SYMBOL_VALID) return; + + if (sym_is_choice_value(sym) && + sym->flags & SYMBOL_NEED_SET_CHOICE_VALUES) { + sym->flags &= ~SYMBOL_NEED_SET_CHOICE_VALUES; + prop = sym_get_choice_prop(sym); + sym_calc_value(prop_get_symbol(prop)); + } + sym->flags |= SYMBOL_VALID; oldval = sym->curr; @@ -425,6 +433,9 @@ void sym_calc_value(struct symbol *sym) if (sym->flags & SYMBOL_AUTO) sym->flags &= ~SYMBOL_WRITE; + + if (sym->flags & SYMBOL_NEED_SET_CHOICE_VALUES) + set_all_choice_values(sym); } void sym_clear_all_valid(void) diff --git a/security/Kconfig b/security/Kconfig index e9c6ac724fef..beb86b500adf 100644 --- a/security/Kconfig +++ b/security/Kconfig @@ -103,7 +103,7 @@ config INTEL_TXT config LSM_MMAP_MIN_ADDR int "Low address space for LSM to protect from user allocation" depends on SECURITY && SECURITY_SELINUX - default 32768 if ARM + default 32768 if ARM || (ARM64 && COMPAT) default 65536 help This is the portion of low virtual memory which should be protected diff --git a/security/apparmor/domain.c b/security/apparmor/domain.c index 859abdaac1ea..9aaa4e72cc1f 100644 --- a/security/apparmor/domain.c +++ b/security/apparmor/domain.c @@ -629,7 +629,7 @@ int aa_change_hat(const char *hats[], int count, u64 token, bool permtest) * There is no exception for unconfined as change_hat is not * available. */ - if (current->no_new_privs) + if (task_no_new_privs(current)) return -EPERM; /* released below */ @@ -780,7 +780,7 @@ int aa_change_profile(const char *ns_name, const char *hname, bool onexec, * no_new_privs is set because this aways results in a reduction * of permissions. */ - if (current->no_new_privs && !unconfined(profile)) { + if (task_no_new_privs(current) && !unconfined(profile)) { put_cred(cred); return -EPERM; } diff --git a/security/apparmor/include/apparmor.h b/security/apparmor/include/apparmor.h index 40aedd9f73ea..4a8cbfeef8b3 100644 --- a/security/apparmor/include/apparmor.h +++ b/security/apparmor/include/apparmor.h @@ -65,7 +65,6 @@ extern int apparmor_initialized __initdata; char *aa_split_fqname(char *args, char **ns_name); void aa_info_message(const char *str); void *kvmalloc(size_t size); -void kvfree(void *buffer); /** diff --git a/security/apparmor/lib.c b/security/apparmor/lib.c index 7430298116d6..ce8d9a84ab2d 100644 --- a/security/apparmor/lib.c +++ b/security/apparmor/lib.c @@ -103,35 +103,3 @@ void *kvmalloc(size_t size) } return buffer; } - -/** - * do_vfree - workqueue routine for freeing vmalloced memory - * @work: data to be freed - * - * The work_struct is overlaid to the data being freed, as at the point - * the work is scheduled the data is no longer valid, be its freeing - * needs to be delayed until safe. - */ -static void do_vfree(struct work_struct *work) -{ - vfree(work); -} - -/** - * kvfree - free an allocation do by kvmalloc - * @buffer: buffer to free (MAYBE_NULL) - * - * Free a buffer allocated by kvmalloc - */ -void kvfree(void *buffer) -{ - if (is_vmalloc_addr(buffer)) { - /* Data is no longer valid so just use the allocated space - * as the work_struct - */ - struct work_struct *work = (struct work_struct *) buffer; - INIT_WORK(work, do_vfree); - schedule_work(work); - } else - kfree(buffer); -} diff --git a/security/capability.c b/security/capability.c index 1728d4e375db..6e4fc776badf 100644 --- a/security/capability.c +++ b/security/capability.c @@ -12,6 +12,26 @@ #include <linux/security.h> +static int cap_binder_set_context_mgr(struct task_struct *mgr) +{ + return 0; +} + +static int cap_binder_transaction(struct task_struct *from, struct task_struct *to) +{ + return 0; +} + +static int cap_binder_transfer_binder(struct task_struct *from, struct task_struct *to) +{ + return 0; +} + +static int cap_binder_transfer_file(struct task_struct *from, struct task_struct *to, struct file *file) +{ + return 0; +} + static int cap_syslog(int type) { return 0; @@ -903,6 +923,10 @@ static void cap_audit_rule_free(void *lsmrule) void __init security_fixup_ops(struct security_operations *ops) { + set_to_cap_if_null(ops, binder_set_context_mgr); + set_to_cap_if_null(ops, binder_transaction); + set_to_cap_if_null(ops, binder_transfer_binder); + set_to_cap_if_null(ops, binder_transfer_file); set_to_cap_if_null(ops, ptrace_access_check); set_to_cap_if_null(ops, ptrace_traceme); set_to_cap_if_null(ops, capget); diff --git a/security/commoncap.c b/security/commoncap.c index c9219a66b7c6..0405522995c5 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. @@ -78,6 +82,13 @@ 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/lsm_audit.c b/security/lsm_audit.c index 8d8d97dbb389..8f6b8e8a4cc8 100644 --- a/security/lsm_audit.c +++ b/security/lsm_audit.c @@ -245,6 +245,21 @@ static void dump_common_audit_data(struct audit_buffer *ab, } break; } + case LSM_AUDIT_DATA_IOCTL_OP: { + struct inode *inode; + + audit_log_d_path(ab, " path=", &a->u.op->path); + + inode = a->u.op->path.dentry->d_inode; + if (inode) { + audit_log_format(ab, " dev="); + audit_log_untrustedstring(ab, inode->i_sb->s_id); + audit_log_format(ab, " ino=%lu", inode->i_ino); + } + + audit_log_format(ab, " ioctlcmd=%hx", a->u.op->cmd); + break; + } case LSM_AUDIT_DATA_DENTRY: { struct inode *inode; diff --git a/security/security.c b/security/security.c index a3dce87d1aef..d6ab2d2b363d 100644 --- a/security/security.c +++ b/security/security.c @@ -134,6 +134,26 @@ int __init register_security(struct security_operations *ops) /* Security operations */ +int security_binder_set_context_mgr(struct task_struct *mgr) +{ + return security_ops->binder_set_context_mgr(mgr); +} + +int security_binder_transaction(struct task_struct *from, struct task_struct *to) +{ + return security_ops->binder_transaction(from, to); +} + +int security_binder_transfer_binder(struct task_struct *from, struct task_struct *to) +{ + return security_ops->binder_transfer_binder(from, to); +} + +int security_binder_transfer_file(struct task_struct *from, struct task_struct *to, struct file *file) +{ + return security_ops->binder_transfer_file(from, to, file); +} + int security_ptrace_access_check(struct task_struct *child, unsigned int mode) { #ifdef CONFIG_SECURITY_YAMA_STACKED diff --git a/security/selinux/avc.c b/security/selinux/avc.c index dad36a6ab45f..f3dbbc0f15dd 100644 --- a/security/selinux/avc.c +++ b/security/selinux/avc.c @@ -22,6 +22,7 @@ #include <linux/init.h> #include <linux/skbuff.h> #include <linux/percpu.h> +#include <linux/list.h> #include <net/sock.h> #include <linux/un.h> #include <net/af_unix.h> @@ -48,6 +49,7 @@ struct avc_entry { u32 tsid; u16 tclass; struct av_decision avd; + struct avc_operation_node *ops_node; }; struct avc_node { @@ -64,6 +66,16 @@ struct avc_cache { u32 latest_notif; /* latest revocation notification */ }; +struct avc_operation_decision_node { + struct operation_decision od; + struct list_head od_list; +}; + +struct avc_operation_node { + struct operation ops; + struct list_head od_head; /* list of operation_decision_node */ +}; + struct avc_callback_node { int (*callback) (u32 event); u32 events; @@ -80,6 +92,9 @@ DEFINE_PER_CPU(struct avc_cache_stats, avc_cache_stats) = { 0 }; static struct avc_cache avc_cache; static struct avc_callback_node *avc_callbacks; static struct kmem_cache *avc_node_cachep; +static struct kmem_cache *avc_operation_decision_node_cachep; +static struct kmem_cache *avc_operation_node_cachep; +static struct kmem_cache *avc_operation_perm_cachep; static inline int avc_hash(u32 ssid, u32 tsid, u16 tclass) { @@ -171,6 +186,16 @@ void __init avc_init(void) avc_node_cachep = kmem_cache_create("avc_node", sizeof(struct avc_node), 0, SLAB_PANIC, NULL); + avc_operation_node_cachep = kmem_cache_create("avc_operation_node", + sizeof(struct avc_operation_node), + 0, SLAB_PANIC, NULL); + avc_operation_decision_node_cachep = kmem_cache_create( + "avc_operation_decision_node", + sizeof(struct avc_operation_decision_node), + 0, SLAB_PANIC, NULL); + avc_operation_perm_cachep = kmem_cache_create("avc_operation_perm", + sizeof(struct operation_perm), + 0, SLAB_PANIC, NULL); audit_log(current->audit_context, GFP_KERNEL, AUDIT_KERNEL, "AVC INITIALIZED\n"); } @@ -205,9 +230,271 @@ int avc_get_hash_stats(char *page) slots_used, AVC_CACHE_SLOTS, max_chain_len); } +/* + * using a linked list for operation_decision lookup because the list is + * always small. i.e. less than 5, typically 1 + */ +static struct operation_decision *avc_operation_lookup(u8 type, + struct avc_operation_node *ops_node) +{ + struct avc_operation_decision_node *od_node; + struct operation_decision *od = NULL; + + list_for_each_entry(od_node, &ops_node->od_head, od_list) { + if (od_node->od.type != type) + continue; + od = &od_node->od; + break; + } + return od; +} + +static inline unsigned int avc_operation_has_perm(struct operation_decision *od, + u16 cmd, u8 specified) +{ + unsigned int rc = 0; + u8 num = cmd & 0xff; + + if ((specified == OPERATION_ALLOWED) && + (od->specified & OPERATION_ALLOWED)) + rc = security_operation_test(od->allowed->perms, num); + else if ((specified == OPERATION_AUDITALLOW) && + (od->specified & OPERATION_AUDITALLOW)) + rc = security_operation_test(od->auditallow->perms, num); + else if ((specified == OPERATION_DONTAUDIT) && + (od->specified & OPERATION_DONTAUDIT)) + rc = security_operation_test(od->dontaudit->perms, num); + return rc; +} + +static void avc_operation_allow_perm(struct avc_operation_node *node, u16 cmd) +{ + struct operation_decision *od; + u8 type; + u8 num; + + type = cmd >> 8; + num = cmd & 0xff; + security_operation_set(node->ops.type, type); + od = avc_operation_lookup(type, node); + if (od && od->allowed) + security_operation_set(od->allowed->perms, num); +} + +static void avc_operation_decision_free( + struct avc_operation_decision_node *od_node) +{ + struct operation_decision *od; + + od = &od_node->od; + if (od->allowed) + kmem_cache_free(avc_operation_perm_cachep, od->allowed); + if (od->auditallow) + kmem_cache_free(avc_operation_perm_cachep, od->auditallow); + if (od->dontaudit) + kmem_cache_free(avc_operation_perm_cachep, od->dontaudit); + kmem_cache_free(avc_operation_decision_node_cachep, od_node); +} + +static void avc_operation_free(struct avc_operation_node *ops_node) +{ + struct avc_operation_decision_node *od_node, *tmp; + + if (!ops_node) + return; + + list_for_each_entry_safe(od_node, tmp, &ops_node->od_head, od_list) { + list_del(&od_node->od_list); + avc_operation_decision_free(od_node); + } + kmem_cache_free(avc_operation_node_cachep, ops_node); +} + +static void avc_copy_operation_decision(struct operation_decision *dest, + struct operation_decision *src) +{ + dest->type = src->type; + dest->specified = src->specified; + if (dest->specified & OPERATION_ALLOWED) + memcpy(dest->allowed->perms, src->allowed->perms, + sizeof(src->allowed->perms)); + if (dest->specified & OPERATION_AUDITALLOW) + memcpy(dest->auditallow->perms, src->auditallow->perms, + sizeof(src->auditallow->perms)); + if (dest->specified & OPERATION_DONTAUDIT) + memcpy(dest->dontaudit->perms, src->dontaudit->perms, + sizeof(src->dontaudit->perms)); +} + +/* + * similar to avc_copy_operation_decision, but only copy decision + * information relevant to this command + */ +static inline void avc_quick_copy_operation_decision(u16 cmd, + struct operation_decision *dest, + struct operation_decision *src) +{ + /* + * compute index of the u32 of the 256 bits (8 u32s) that contain this + * command permission + */ + u8 i = (0xff & cmd) >> 5; + + dest->specified = src->specified; + if (dest->specified & OPERATION_ALLOWED) + dest->allowed->perms[i] = src->allowed->perms[i]; + if (dest->specified & OPERATION_AUDITALLOW) + dest->auditallow->perms[i] = src->auditallow->perms[i]; + if (dest->specified & OPERATION_DONTAUDIT) + dest->dontaudit->perms[i] = src->dontaudit->perms[i]; +} + +static struct avc_operation_decision_node + *avc_operation_decision_alloc(u8 specified) +{ + struct avc_operation_decision_node *node; + struct operation_decision *od; + + node = kmem_cache_zalloc(avc_operation_decision_node_cachep, + GFP_ATOMIC | __GFP_NOMEMALLOC); + if (!node) + return NULL; + + od = &node->od; + if (specified & OPERATION_ALLOWED) { + od->allowed = kmem_cache_zalloc(avc_operation_perm_cachep, + GFP_ATOMIC | __GFP_NOMEMALLOC); + if (!od->allowed) + goto error; + } + if (specified & OPERATION_AUDITALLOW) { + od->auditallow = kmem_cache_zalloc(avc_operation_perm_cachep, + GFP_ATOMIC | __GFP_NOMEMALLOC); + if (!od->auditallow) + goto error; + } + if (specified & OPERATION_DONTAUDIT) { + od->dontaudit = kmem_cache_zalloc(avc_operation_perm_cachep, + GFP_ATOMIC | __GFP_NOMEMALLOC); + if (!od->dontaudit) + goto error; + } + return node; +error: + avc_operation_decision_free(node); + return NULL; +} + +static int avc_add_operation(struct avc_node *node, + struct operation_decision *od) +{ + struct avc_operation_decision_node *dest_od; + + node->ae.ops_node->ops.len++; + dest_od = avc_operation_decision_alloc(od->specified); + if (!dest_od) + return -ENOMEM; + avc_copy_operation_decision(&dest_od->od, od); + list_add(&dest_od->od_list, &node->ae.ops_node->od_head); + return 0; +} + +static struct avc_operation_node *avc_operation_alloc(void) +{ + struct avc_operation_node *ops; + + ops = kmem_cache_zalloc(avc_operation_node_cachep, + GFP_ATOMIC|__GFP_NOMEMALLOC); + if (!ops) + return ops; + INIT_LIST_HEAD(&ops->od_head); + return ops; +} + +static int avc_operation_populate(struct avc_node *node, + struct avc_operation_node *src) +{ + struct avc_operation_node *dest; + struct avc_operation_decision_node *dest_od; + struct avc_operation_decision_node *src_od; + + if (src->ops.len == 0) + return 0; + dest = avc_operation_alloc(); + if (!dest) + return -ENOMEM; + + memcpy(dest->ops.type, &src->ops.type, sizeof(dest->ops.type)); + dest->ops.len = src->ops.len; + + /* for each source od allocate a destination od and copy */ + list_for_each_entry(src_od, &src->od_head, od_list) { + dest_od = avc_operation_decision_alloc(src_od->od.specified); + if (!dest_od) + goto error; + avc_copy_operation_decision(&dest_od->od, &src_od->od); + list_add(&dest_od->od_list, &dest->od_head); + } + node->ae.ops_node = dest; + return 0; +error: + avc_operation_free(dest); + return -ENOMEM; + +} + +static inline u32 avc_operation_audit_required(u32 requested, + struct av_decision *avd, + struct operation_decision *od, + u16 cmd, + int result, + u32 *deniedp) +{ + u32 denied, audited; + + denied = requested & ~avd->allowed; + if (unlikely(denied)) { + audited = denied & avd->auditdeny; + if (audited && od) { + if (avc_operation_has_perm(od, cmd, + OPERATION_DONTAUDIT)) + audited &= ~requested; + } + } else if (result) { + audited = denied = requested; + } else { + audited = requested & avd->auditallow; + if (audited && od) { + if (!avc_operation_has_perm(od, cmd, + OPERATION_AUDITALLOW)) + audited &= ~requested; + } + } + + *deniedp = denied; + return audited; +} + +static inline int avc_operation_audit(u32 ssid, u32 tsid, u16 tclass, + u32 requested, struct av_decision *avd, + struct operation_decision *od, + u16 cmd, int result, + struct common_audit_data *ad) +{ + u32 audited, denied; + + audited = avc_operation_audit_required( + requested, avd, od, cmd, result, &denied); + if (likely(!audited)) + return 0; + return slow_avc_audit(ssid, tsid, tclass, requested, + audited, denied, result, ad, 0); +} + static void avc_node_free(struct rcu_head *rhead) { struct avc_node *node = container_of(rhead, struct avc_node, rhead); + avc_operation_free(node->ae.ops_node); kmem_cache_free(avc_node_cachep, node); avc_cache_stats_incr(frees); } @@ -221,6 +508,7 @@ static void avc_node_delete(struct avc_node *node) static void avc_node_kill(struct avc_node *node) { + avc_operation_free(node->ae.ops_node); kmem_cache_free(avc_node_cachep, node); avc_cache_stats_incr(frees); atomic_dec(&avc_cache.active_nodes); @@ -367,6 +655,7 @@ static int avc_latest_notif_update(int seqno, int is_insert) * @tsid: target security identifier * @tclass: target security class * @avd: resulting av decision + * @ops: resulting operation decisions * * Insert an AVC entry for the SID pair * (@ssid, @tsid) and class @tclass. @@ -378,7 +667,9 @@ static int avc_latest_notif_update(int seqno, int is_insert) * the access vectors into a cache entry, returns * avc_node inserted. Otherwise, this function returns NULL. */ -static struct avc_node *avc_insert(u32 ssid, u32 tsid, u16 tclass, struct av_decision *avd) +static struct avc_node *avc_insert(u32 ssid, u32 tsid, u16 tclass, + struct av_decision *avd, + struct avc_operation_node *ops_node) { struct avc_node *pos, *node = NULL; int hvalue; @@ -391,10 +682,15 @@ static struct avc_node *avc_insert(u32 ssid, u32 tsid, u16 tclass, struct av_dec if (node) { struct hlist_head *head; spinlock_t *lock; + int rc = 0; hvalue = avc_hash(ssid, tsid, tclass); avc_node_populate(node, ssid, tsid, tclass, avd); - + rc = avc_operation_populate(node, ops_node); + if (rc) { + kmem_cache_free(avc_node_cachep, node); + return NULL; + } head = &avc_cache.slots[hvalue]; lock = &avc_cache.slots_lock[hvalue]; @@ -444,11 +740,15 @@ static void avc_audit_post_callback(struct audit_buffer *ab, void *a) avc_dump_query(ab, ad->selinux_audit_data->ssid, ad->selinux_audit_data->tsid, ad->selinux_audit_data->tclass); + if (ad->selinux_audit_data->denied) { + audit_log_format(ab, " permissive=%u", + ad->selinux_audit_data->result ? 0 : 1); + } } /* This is the slow part of avc audit with big stack footprint */ noinline int slow_avc_audit(u32 ssid, u32 tsid, u16 tclass, - u32 requested, u32 audited, u32 denied, + u32 requested, u32 audited, u32 denied, int result, struct common_audit_data *a, unsigned flags) { @@ -477,6 +777,7 @@ noinline int slow_avc_audit(u32 ssid, u32 tsid, u16 tclass, sad.tsid = tsid; sad.audited = audited; sad.denied = denied; + sad.result = result; a->selinux_audit_data = &sad; @@ -523,14 +824,17 @@ static inline int avc_sidcmp(u32 x, u32 y) * @perms : Permission mask bits * @ssid,@tsid,@tclass : identifier of an AVC entry * @seqno : sequence number when decision was made + * @od: operation_decision to be added to the node * * if a valid AVC entry doesn't exist,this function returns -ENOENT. * if kmalloc() called internal returns NULL, this function returns -ENOMEM. * otherwise, this function updates the AVC entry. The original AVC-entry object * will release later by RCU. */ -static int avc_update_node(u32 event, u32 perms, u32 ssid, u32 tsid, u16 tclass, - u32 seqno) +static int avc_update_node(u32 event, u32 perms, u16 cmd, u32 ssid, u32 tsid, + u16 tclass, u32 seqno, + struct operation_decision *od, + u32 flags) { int hvalue, rc = 0; unsigned long flag; @@ -574,9 +878,19 @@ static int avc_update_node(u32 event, u32 perms, u32 ssid, u32 tsid, u16 tclass, avc_node_populate(node, ssid, tsid, tclass, &orig->ae.avd); + if (orig->ae.ops_node) { + rc = avc_operation_populate(node, orig->ae.ops_node); + if (rc) { + kmem_cache_free(avc_node_cachep, node); + goto out_unlock; + } + } + switch (event) { case AVC_CALLBACK_GRANT: node->ae.avd.allowed |= perms; + if (node->ae.ops_node && (flags & AVC_OPERATION_CMD)) + avc_operation_allow_perm(node->ae.ops_node, cmd); break; case AVC_CALLBACK_TRY_REVOKE: case AVC_CALLBACK_REVOKE: @@ -594,6 +908,9 @@ static int avc_update_node(u32 event, u32 perms, u32 ssid, u32 tsid, u16 tclass, case AVC_CALLBACK_AUDITDENY_DISABLE: node->ae.avd.auditdeny &= ~perms; break; + case AVC_CALLBACK_ADD_OPERATION: + avc_add_operation(node, od); + break; } avc_node_replace(node, orig); out_unlock: @@ -665,18 +982,20 @@ int avc_ss_reset(u32 seqno) * results in a bigger stack frame. */ static noinline struct avc_node *avc_compute_av(u32 ssid, u32 tsid, - u16 tclass, struct av_decision *avd) + u16 tclass, struct av_decision *avd, + struct avc_operation_node *ops_node) { rcu_read_unlock(); - security_compute_av(ssid, tsid, tclass, avd); + INIT_LIST_HEAD(&ops_node->od_head); + security_compute_av(ssid, tsid, tclass, avd, &ops_node->ops); rcu_read_lock(); - return avc_insert(ssid, tsid, tclass, avd); + return avc_insert(ssid, tsid, tclass, avd, ops_node); } static noinline int avc_denied(u32 ssid, u32 tsid, - u16 tclass, u32 requested, - unsigned flags, - struct av_decision *avd) + u16 tclass, u32 requested, + u16 cmd, unsigned flags, + struct av_decision *avd) { if (flags & AVC_STRICT) return -EACCES; @@ -684,11 +1003,92 @@ static noinline int avc_denied(u32 ssid, u32 tsid, if (selinux_enforcing && !(avd->flags & AVD_FLAGS_PERMISSIVE)) return -EACCES; - avc_update_node(AVC_CALLBACK_GRANT, requested, ssid, - tsid, tclass, avd->seqno); + avc_update_node(AVC_CALLBACK_GRANT, requested, cmd, ssid, + tsid, tclass, avd->seqno, NULL, flags); return 0; } +/* + * ioctl commands are comprised of four fields, direction, size, type, and + * number. The avc operation logic filters based on two of them: + * + * type: or code, typically unique to each driver + * number: or function + * + * For example, 0x89 is a socket type, and number 0x27 is the get hardware + * address function. + */ +int avc_has_operation(u32 ssid, u32 tsid, u16 tclass, u32 requested, + u16 cmd, struct common_audit_data *ad) +{ + struct avc_node *node; + struct av_decision avd; + u32 denied; + struct operation_decision *od = NULL; + struct operation_decision od_local; + struct operation_perm allowed; + struct operation_perm auditallow; + struct operation_perm dontaudit; + struct avc_operation_node local_ops_node; + struct avc_operation_node *ops_node; + u8 type = cmd >> 8; + int rc = 0, rc2; + + ops_node = &local_ops_node; + BUG_ON(!requested); + + rcu_read_lock(); + + node = avc_lookup(ssid, tsid, tclass); + if (unlikely(!node)) { + node = avc_compute_av(ssid, tsid, tclass, &avd, ops_node); + } else { + memcpy(&avd, &node->ae.avd, sizeof(avd)); + ops_node = node->ae.ops_node; + } + /* if operations are not defined, only consider av_decision */ + if (!ops_node || !ops_node->ops.len) + goto decision; + + od_local.allowed = &allowed; + od_local.auditallow = &auditallow; + od_local.dontaudit = &dontaudit; + + /* lookup operation decision */ + od = avc_operation_lookup(type, ops_node); + if (unlikely(!od)) { + /* Compute operation decision if type is flagged */ + if (!security_operation_test(ops_node->ops.type, type)) { + avd.allowed &= ~requested; + goto decision; + } + rcu_read_unlock(); + security_compute_operation(ssid, tsid, tclass, type, &od_local); + rcu_read_lock(); + avc_update_node(AVC_CALLBACK_ADD_OPERATION, requested, cmd, + ssid, tsid, tclass, avd.seqno, &od_local, 0); + } else { + avc_quick_copy_operation_decision(cmd, &od_local, od); + } + od = &od_local; + + if (!avc_operation_has_perm(od, cmd, OPERATION_ALLOWED)) + avd.allowed &= ~requested; + +decision: + denied = requested & ~(avd.allowed); + if (unlikely(denied)) + rc = avc_denied(ssid, tsid, tclass, requested, cmd, + AVC_OPERATION_CMD, &avd); + + rcu_read_unlock(); + + rc2 = avc_operation_audit(ssid, tsid, tclass, requested, + &avd, od, cmd, rc, ad); + if (rc2) + return rc2; + return rc; +} /** * avc_has_perm_noaudit - Check permissions but perform no auditing. @@ -716,6 +1116,7 @@ inline int avc_has_perm_noaudit(u32 ssid, u32 tsid, struct av_decision *avd) { struct avc_node *node; + struct avc_operation_node ops_node; int rc = 0; u32 denied; @@ -724,16 +1125,14 @@ inline int avc_has_perm_noaudit(u32 ssid, u32 tsid, rcu_read_lock(); node = avc_lookup(ssid, tsid, tclass); - if (unlikely(!node)) { - node = avc_compute_av(ssid, tsid, tclass, avd); - } else { + if (unlikely(!node)) + node = avc_compute_av(ssid, tsid, tclass, avd, &ops_node); + else memcpy(avd, &node->ae.avd, sizeof(*avd)); - avd = &node->ae.avd; - } denied = requested & ~(avd->allowed); if (unlikely(denied)) - rc = avc_denied(ssid, tsid, tclass, requested, flags, avd); + rc = avc_denied(ssid, tsid, tclass, requested, 0, flags, avd); rcu_read_unlock(); return rc; diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index fdd6e4f8be39..607c3d58a691 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -147,6 +147,17 @@ static int selinux_secmark_enabled(void) return (atomic_read(&selinux_secmark_refcount) > 0); } +static int selinux_netcache_avc_callback(u32 event) +{ + if (event == AVC_CALLBACK_RESET) { + sel_netif_flush(); + sel_netnode_flush(); + sel_netport_flush(); + synchronize_net(); + } + return 0; +} + /* * initialise the security for the init task */ @@ -419,8 +430,11 @@ static int sb_finish_set_opts(struct super_block *sb) sbsec->behavior > ARRAY_SIZE(labeling_behaviors)) sbsec->flags &= ~SE_SBLABELSUPP; - /* Special handling for sysfs. Is genfs but also has setxattr handler*/ - if (strncmp(sb->s_type->name, "sysfs", sizeof("sysfs")) == 0) + /* Special handling. Is genfs but also has in-core setxattr handler*/ + if (!strcmp(sb->s_type->name, "sysfs") || + !strcmp(sb->s_type->name, "pstore") || + !strcmp(sb->s_type->name, "debugfs") || + !strcmp(sb->s_type->name, "rootfs")) sbsec->flags |= SE_SBLABELSUPP; /* Initialize the root inode. */ @@ -1856,6 +1870,65 @@ static inline u32 open_file_to_av(struct file *file) /* Hook functions begin here. */ +static int selinux_binder_set_context_mgr(struct task_struct *mgr) +{ + u32 mysid = current_sid(); + u32 mgrsid = task_sid(mgr); + + return avc_has_perm(mysid, mgrsid, SECCLASS_BINDER, BINDER__SET_CONTEXT_MGR, NULL); +} + +static int selinux_binder_transaction(struct task_struct *from, struct task_struct *to) +{ + u32 mysid = current_sid(); + u32 fromsid = task_sid(from); + u32 tosid = task_sid(to); + int rc; + + if (mysid != fromsid) { + rc = avc_has_perm(mysid, fromsid, SECCLASS_BINDER, BINDER__IMPERSONATE, NULL); + if (rc) + return rc; + } + + return avc_has_perm(fromsid, tosid, SECCLASS_BINDER, BINDER__CALL, NULL); +} + +static int selinux_binder_transfer_binder(struct task_struct *from, struct task_struct *to) +{ + u32 fromsid = task_sid(from); + u32 tosid = task_sid(to); + return avc_has_perm(fromsid, tosid, SECCLASS_BINDER, BINDER__TRANSFER, NULL); +} + +static int selinux_binder_transfer_file(struct task_struct *from, struct task_struct *to, struct file *file) +{ + u32 sid = task_sid(to); + struct file_security_struct *fsec = file->f_security; + struct inode *inode = file->f_path.dentry->d_inode; + struct inode_security_struct *isec = inode->i_security; + struct common_audit_data ad; + int rc; + + ad.type = LSM_AUDIT_DATA_PATH; + ad.u.path = file->f_path; + + if (sid != fsec->sid) { + rc = avc_has_perm(sid, fsec->sid, + SECCLASS_FD, + FD__USE, + &ad); + if (rc) + return rc; + } + + if (unlikely(IS_PRIVATE(inode))) + return 0; + + return avc_has_perm(sid, isec->sid, isec->sclass, file_to_av(file), + &ad); +} + static int selinux_ptrace_access_check(struct task_struct *child, unsigned int mode) { @@ -2672,6 +2745,7 @@ static int selinux_inode_follow_link(struct dentry *dentry, struct nameidata *na static noinline int audit_inode_permission(struct inode *inode, u32 perms, u32 audited, u32 denied, + int result, unsigned flags) { struct common_audit_data ad; @@ -2682,7 +2756,7 @@ static noinline int audit_inode_permission(struct inode *inode, ad.u.inode = inode; rc = slow_avc_audit(current_sid(), isec->sid, isec->sclass, perms, - audited, denied, &ad, flags); + audited, denied, result, &ad, flags); if (rc) return rc; return 0; @@ -2724,7 +2798,7 @@ static int selinux_inode_permission(struct inode *inode, int mask) if (likely(!audited)) return rc; - rc2 = audit_inode_permission(inode, perms, audited, denied, flags); + rc2 = audit_inode_permission(inode, perms, audited, denied, rc, flags); if (rc2) return rc2; return rc; @@ -3035,6 +3109,44 @@ static void selinux_file_free_security(struct file *file) file_free_security(file); } +/* + * Check whether a task has the ioctl permission and cmd + * operation to an inode. + */ +int ioctl_has_perm(const struct cred *cred, struct file *file, + u32 requested, u16 cmd) +{ + struct common_audit_data ad; + struct file_security_struct *fsec = file->f_security; + struct inode *inode = file_inode(file); + struct inode_security_struct *isec = inode->i_security; + struct lsm_ioctlop_audit ioctl; + u32 ssid = cred_sid(cred); + int rc; + + ad.type = LSM_AUDIT_DATA_IOCTL_OP; + ad.u.op = &ioctl; + ad.u.op->cmd = cmd; + ad.u.op->path = file->f_path; + + if (ssid != fsec->sid) { + rc = avc_has_perm(ssid, fsec->sid, + SECCLASS_FD, + FD__USE, + &ad); + if (rc) + goto out; + } + + if (unlikely(IS_PRIVATE(inode))) + return 0; + + rc = avc_has_operation(ssid, isec->sid, isec->sclass, + requested, cmd, &ad); +out: + return rc; +} + static int selinux_file_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { @@ -3077,7 +3189,7 @@ static int selinux_file_ioctl(struct file *file, unsigned int cmd, * to the file's ioctl() function. */ default: - error = file_has_perm(cred, file, FILE__IOCTL); + error = ioctl_has_perm(cred, file, FILE__IOCTL, (u16) cmd); } return error; } @@ -3121,24 +3233,20 @@ error: static int selinux_mmap_addr(unsigned long addr) { - int rc = 0; - u32 sid = current_sid(); + int rc; + + /* do DAC check on address space usage */ + rc = cap_mmap_addr(addr); + if (rc) + return rc; - /* - * notice that we are intentionally putting the SELinux check before - * the secondary cap_file_mmap check. This is such a likely attempt - * at bad behaviour/exploit that we always want to get the AVC, even - * if DAC would have also denied the operation. - */ if (addr < CONFIG_LSM_MMAP_MIN_ADDR) { + u32 sid = current_sid(); rc = avc_has_perm(sid, sid, SECCLASS_MEMPROTECT, MEMPROTECT__MMAP_ZERO, NULL); - if (rc) - return rc; } - /* do DAC check on address space usage */ - return cap_mmap_addr(addr); + return rc; } static int selinux_mmap_file(struct file *file, unsigned long reqprot, @@ -4153,15 +4261,15 @@ static int selinux_socket_unix_may_send(struct socket *sock, &ad); } -static int selinux_inet_sys_rcv_skb(int ifindex, char *addrp, u16 family, - u32 peer_sid, +static int selinux_inet_sys_rcv_skb(struct net *ns, int ifindex, + char *addrp, u16 family, u32 peer_sid, struct common_audit_data *ad) { int err; u32 if_sid; u32 node_sid; - err = sel_netif_sid(ifindex, &if_sid); + err = sel_netif_sid(ns, ifindex, &if_sid); if (err) return err; err = avc_has_perm(peer_sid, if_sid, @@ -4254,8 +4362,8 @@ static int selinux_socket_sock_rcv_skb(struct sock *sk, struct sk_buff *skb) err = selinux_skb_peerlbl_sid(skb, family, &peer_sid); if (err) return err; - err = selinux_inet_sys_rcv_skb(skb->skb_iif, addrp, family, - peer_sid, &ad); + err = selinux_inet_sys_rcv_skb(sock_net(sk), skb->skb_iif, + addrp, family, peer_sid, &ad); if (err) { selinux_netlbl_err(skb, err, 0); return err; @@ -4598,7 +4706,8 @@ out: #ifdef CONFIG_NETFILTER -static unsigned int selinux_ip_forward(struct sk_buff *skb, int ifindex, +static unsigned int selinux_ip_forward(struct sk_buff *skb, + const struct net_device *indev, u16 family) { int err; @@ -4624,14 +4733,14 @@ static unsigned int selinux_ip_forward(struct sk_buff *skb, int ifindex, ad.type = LSM_AUDIT_DATA_NET; ad.u.net = &net; - ad.u.net->netif = ifindex; + ad.u.net->netif = indev->ifindex; ad.u.net->family = family; if (selinux_parse_skb(skb, &ad, &addrp, 1, NULL) != 0) return NF_DROP; if (peerlbl_active) { - err = selinux_inet_sys_rcv_skb(ifindex, addrp, family, - peer_sid, &ad); + err = selinux_inet_sys_rcv_skb(dev_net(indev), indev->ifindex, + addrp, family, peer_sid, &ad); if (err) { selinux_netlbl_err(skb, err, 1); return NF_DROP; @@ -4660,7 +4769,7 @@ static unsigned int selinux_ipv4_forward(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - return selinux_ip_forward(skb, in->ifindex, PF_INET); + return selinux_ip_forward(skb, in, PF_INET); } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) @@ -4670,7 +4779,7 @@ static unsigned int selinux_ipv6_forward(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - return selinux_ip_forward(skb, in->ifindex, PF_INET6); + return selinux_ip_forward(skb, in, PF_INET6); } #endif /* IPV6 */ @@ -4758,11 +4867,13 @@ static unsigned int selinux_ip_postroute_compat(struct sk_buff *skb, return NF_ACCEPT; } -static unsigned int selinux_ip_postroute(struct sk_buff *skb, int ifindex, +static unsigned int selinux_ip_postroute(struct sk_buff *skb, + const struct net_device *outdev, u16 family) { u32 secmark_perm; u32 peer_sid; + int ifindex = outdev->ifindex; struct sock *sk; struct common_audit_data ad; struct lsm_network_audit net = {0,}; @@ -4875,7 +4986,7 @@ static unsigned int selinux_ip_postroute(struct sk_buff *skb, int ifindex, u32 if_sid; u32 node_sid; - if (sel_netif_sid(ifindex, &if_sid)) + if (sel_netif_sid(dev_net(outdev), ifindex, &if_sid)) return NF_DROP; if (avc_has_perm(peer_sid, if_sid, SECCLASS_NETIF, NETIF__EGRESS, &ad)) @@ -4897,7 +5008,7 @@ static unsigned int selinux_ipv4_postroute(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - return selinux_ip_postroute(skb, out->ifindex, PF_INET); + return selinux_ip_postroute(skb, out, PF_INET); } #if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) @@ -4907,7 +5018,7 @@ static unsigned int selinux_ipv6_postroute(unsigned int hooknum, const struct net_device *out, int (*okfn)(struct sk_buff *)) { - return selinux_ip_postroute(skb, out->ifindex, PF_INET6); + return selinux_ip_postroute(skb, out, PF_INET6); } #endif /* IPV6 */ @@ -5656,6 +5767,11 @@ static int selinux_key_getsecurity(struct key *key, char **_buffer) static struct security_operations selinux_ops = { .name = "selinux", + .binder_set_context_mgr = selinux_binder_set_context_mgr, + .binder_transaction = selinux_binder_transaction, + .binder_transfer_binder = selinux_binder_transfer_binder, + .binder_transfer_file = selinux_binder_transfer_file, + .ptrace_access_check = selinux_ptrace_access_check, .ptrace_traceme = selinux_ptrace_traceme, .capget = selinux_capget, @@ -5882,6 +5998,9 @@ static __init int selinux_init(void) if (register_security(&selinux_ops)) panic("SELinux: Unable to register with kernel.\n"); + if (avc_add_callback(selinux_netcache_avc_callback, AVC_CALLBACK_RESET)) + panic("SELinux: Unable to register AVC netcache callback\n"); + if (selinux_enforcing) printk(KERN_DEBUG "SELinux: Starting in enforcing mode\n"); else diff --git a/security/selinux/include/avc.h b/security/selinux/include/avc.h index 92d0ab561db8..8109ad846e99 100644 --- a/security/selinux/include/avc.h +++ b/security/selinux/include/avc.h @@ -102,7 +102,7 @@ static inline u32 avc_audit_required(u32 requested, } int slow_avc_audit(u32 ssid, u32 tsid, u16 tclass, - u32 requested, u32 audited, u32 denied, + u32 requested, u32 audited, u32 denied, int result, struct common_audit_data *a, unsigned flags); @@ -137,16 +137,20 @@ static inline int avc_audit(u32 ssid, u32 tsid, if (likely(!audited)) return 0; return slow_avc_audit(ssid, tsid, tclass, - requested, audited, denied, + requested, audited, denied, result, a, flags); } #define AVC_STRICT 1 /* Ignore permissive mode. */ +#define AVC_OPERATION_CMD 2 /* ignore command when updating operations */ int avc_has_perm_noaudit(u32 ssid, u32 tsid, u16 tclass, u32 requested, unsigned flags, struct av_decision *avd); +int avc_has_operation(u32 ssid, u32 tsid, u16 tclass, u32 requested, + u16 cmd, struct common_audit_data *ad); + int avc_has_perm_flags(u32 ssid, u32 tsid, u16 tclass, u32 requested, struct common_audit_data *auditdata, @@ -169,6 +173,7 @@ u32 avc_policy_seqno(void); #define AVC_CALLBACK_AUDITALLOW_DISABLE 32 #define AVC_CALLBACK_AUDITDENY_ENABLE 64 #define AVC_CALLBACK_AUDITDENY_DISABLE 128 +#define AVC_CALLBACK_ADD_OPERATION 256 int avc_add_callback(int (*callback)(u32 event), u32 events); diff --git a/security/selinux/include/classmap.h b/security/selinux/include/classmap.h index 14d04e63b1f0..c32ff7bde81a 100644 --- a/security/selinux/include/classmap.h +++ b/security/selinux/include/classmap.h @@ -151,5 +151,6 @@ struct security_class_mapping secclass_map[] = { { "kernel_service", { "use_as_override", "create_files_as", NULL } }, { "tun_socket", { COMMON_SOCK_PERMS, "attach_queue", NULL } }, + { "binder", { "impersonate", "call", "set_context_mgr", "transfer", NULL } }, { NULL } }; diff --git a/security/selinux/include/netif.h b/security/selinux/include/netif.h index 43d507242b42..c72145444090 100644 --- a/security/selinux/include/netif.h +++ b/security/selinux/include/netif.h @@ -17,7 +17,11 @@ #ifndef _SELINUX_NETIF_H_ #define _SELINUX_NETIF_H_ -int sel_netif_sid(int ifindex, u32 *sid); +#include <net/net_namespace.h> + +void sel_netif_flush(void); + +int sel_netif_sid(struct net *ns, int ifindex, u32 *sid); #endif /* _SELINUX_NETIF_H_ */ diff --git a/security/selinux/include/netnode.h b/security/selinux/include/netnode.h index df7a5ed6c694..937668dd3024 100644 --- a/security/selinux/include/netnode.h +++ b/security/selinux/include/netnode.h @@ -27,6 +27,8 @@ #ifndef _SELINUX_NETNODE_H #define _SELINUX_NETNODE_H +void sel_netnode_flush(void); + int sel_netnode_sid(void *addr, u16 family, u32 *sid); #endif diff --git a/security/selinux/include/netport.h b/security/selinux/include/netport.h index 4d965b83d735..d1ce896b2cb0 100644 --- a/security/selinux/include/netport.h +++ b/security/selinux/include/netport.h @@ -26,6 +26,8 @@ #ifndef _SELINUX_NETPORT_H #define _SELINUX_NETPORT_H +void sel_netport_flush(void); + int sel_netport_sid(u8 protocol, u16 pnum, u32 *sid); #endif diff --git a/security/selinux/include/objsec.h b/security/selinux/include/objsec.h index 6fd9dd256a62..2a3be7adc071 100644 --- a/security/selinux/include/objsec.h +++ b/security/selinux/include/objsec.h @@ -24,6 +24,7 @@ #include <linux/binfmts.h> #include <linux/in.h> #include <linux/spinlock.h> +#include <net/net_namespace.h> #include "flask.h" #include "avc.h" @@ -78,6 +79,7 @@ struct ipc_security_struct { }; struct netif_security_struct { + struct net *ns; /* network namespace */ int ifindex; /* device index */ u32 sid; /* SID for this interface */ }; diff --git a/security/selinux/include/security.h b/security/selinux/include/security.h index 6d3885165d14..e72d8de93725 100644 --- a/security/selinux/include/security.h +++ b/security/selinux/include/security.h @@ -33,13 +33,15 @@ #define POLICYDB_VERSION_ROLETRANS 26 #define POLICYDB_VERSION_NEW_OBJECT_DEFAULTS 27 #define POLICYDB_VERSION_DEFAULT_TYPE 28 +#define POLICYDB_VERSION_CONSTRAINT_NAMES 29 +#define POLICYDB_VERSION_IOCTL_OPERATIONS 30 /* Range of policy versions we understand*/ #define POLICYDB_VERSION_MIN POLICYDB_VERSION_BASE #ifdef CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX #define POLICYDB_VERSION_MAX CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX_VALUE #else -#define POLICYDB_VERSION_MAX POLICYDB_VERSION_DEFAULT_TYPE +#define POLICYDB_VERSION_MAX POLICYDB_VERSION_IOCTL_OPERATIONS #endif /* Mask for just the mount related flags */ @@ -102,11 +104,40 @@ struct av_decision { u32 flags; }; +#define security_operation_set(perms, x) (perms[x >> 5] |= 1 << (x & 0x1f)) +#define security_operation_test(perms, x) (1 & (perms[x >> 5] >> (x & 0x1f))) + +struct operation_perm { + u32 perms[8]; +}; + +struct operation_decision { + u8 type; + u8 specified; + struct operation_perm *allowed; + struct operation_perm *auditallow; + struct operation_perm *dontaudit; +}; + +#define OPERATION_ALLOWED 1 +#define OPERATION_AUDITALLOW 2 +#define OPERATION_DONTAUDIT 4 +#define OPERATION_ALL (OPERATION_ALLOWED | OPERATION_AUDITALLOW |\ + OPERATION_DONTAUDIT) +struct operation { + u16 len; /* length of operation decision chain */ + u32 type[8]; /* 256 types */ +}; + /* definitions of av_decision.flags */ #define AVD_FLAGS_PERMISSIVE 0x0001 void security_compute_av(u32 ssid, u32 tsid, - u16 tclass, struct av_decision *avd); + u16 tclass, struct av_decision *avd, + struct operation *ops); + +void security_compute_operation(u32 ssid, u32 tsid, u16 tclass, + u8 type, struct operation_decision *od); void security_compute_av_user(u32 ssid, u32 tsid, u16 tclass, struct av_decision *avd); diff --git a/security/selinux/netif.c b/security/selinux/netif.c index 47a49d1a6f6a..548651a73876 100644 --- a/security/selinux/netif.c +++ b/security/selinux/netif.c @@ -45,6 +45,7 @@ static struct list_head sel_netif_hash[SEL_NETIF_HASH_SIZE]; /** * sel_netif_hashfn - Hashing function for the interface table + * @ns: the network namespace * @ifindex: the network interface * * Description: @@ -52,13 +53,14 @@ static struct list_head sel_netif_hash[SEL_NETIF_HASH_SIZE]; * bucket number for the given interface. * */ -static inline u32 sel_netif_hashfn(int ifindex) +static inline u32 sel_netif_hashfn(const struct net *ns, int ifindex) { - return (ifindex & (SEL_NETIF_HASH_SIZE - 1)); + return (((uintptr_t)ns + ifindex) & (SEL_NETIF_HASH_SIZE - 1)); } /** * sel_netif_find - Search for an interface record + * @ns: the network namespace * @ifindex: the network interface * * Description: @@ -66,15 +68,15 @@ static inline u32 sel_netif_hashfn(int ifindex) * If an entry can not be found in the table return NULL. * */ -static inline struct sel_netif *sel_netif_find(int ifindex) +static inline struct sel_netif *sel_netif_find(const struct net *ns, + int ifindex) { - int idx = sel_netif_hashfn(ifindex); + int idx = sel_netif_hashfn(ns, ifindex); struct sel_netif *netif; list_for_each_entry_rcu(netif, &sel_netif_hash[idx], list) - /* all of the devices should normally fit in the hash, so we - * optimize for that case */ - if (likely(netif->nsec.ifindex == ifindex)) + if (net_eq(netif->nsec.ns, ns) && + netif->nsec.ifindex == ifindex) return netif; return NULL; @@ -96,7 +98,7 @@ static int sel_netif_insert(struct sel_netif *netif) if (sel_netif_total >= SEL_NETIF_HASH_MAX) return -ENOSPC; - idx = sel_netif_hashfn(netif->nsec.ifindex); + idx = sel_netif_hashfn(netif->nsec.ns, netif->nsec.ifindex); list_add_rcu(&netif->list, &sel_netif_hash[idx]); sel_netif_total++; @@ -120,6 +122,7 @@ static void sel_netif_destroy(struct sel_netif *netif) /** * sel_netif_sid_slow - Lookup the SID of a network interface using the policy + * @ns: the network namespace * @ifindex: the network interface * @sid: interface SID * @@ -130,7 +133,7 @@ static void sel_netif_destroy(struct sel_netif *netif) * failure. * */ -static int sel_netif_sid_slow(int ifindex, u32 *sid) +static int sel_netif_sid_slow(struct net *ns, int ifindex, u32 *sid) { int ret; struct sel_netif *netif; @@ -140,7 +143,7 @@ static int sel_netif_sid_slow(int ifindex, u32 *sid) /* NOTE: we always use init's network namespace since we don't * currently support containers */ - dev = dev_get_by_index(&init_net, ifindex); + dev = dev_get_by_index(ns, ifindex); if (unlikely(dev == NULL)) { printk(KERN_WARNING "SELinux: failure in sel_netif_sid_slow()," @@ -149,7 +152,7 @@ static int sel_netif_sid_slow(int ifindex, u32 *sid) } spin_lock_bh(&sel_netif_lock); - netif = sel_netif_find(ifindex); + netif = sel_netif_find(ns, ifindex); if (netif != NULL) { *sid = netif->nsec.sid; ret = 0; @@ -163,6 +166,7 @@ static int sel_netif_sid_slow(int ifindex, u32 *sid) ret = security_netif_sid(dev->name, &new->nsec.sid); if (ret != 0) goto out; + new->nsec.ns = ns; new->nsec.ifindex = ifindex; ret = sel_netif_insert(new); if (ret != 0) @@ -184,6 +188,7 @@ out: /** * sel_netif_sid - Lookup the SID of a network interface + * @ns: the network namespace * @ifindex: the network interface * @sid: interface SID * @@ -195,12 +200,12 @@ out: * on failure. * */ -int sel_netif_sid(int ifindex, u32 *sid) +int sel_netif_sid(struct net *ns, int ifindex, u32 *sid) { struct sel_netif *netif; rcu_read_lock(); - netif = sel_netif_find(ifindex); + netif = sel_netif_find(ns, ifindex); if (likely(netif != NULL)) { *sid = netif->nsec.sid; rcu_read_unlock(); @@ -208,11 +213,12 @@ int sel_netif_sid(int ifindex, u32 *sid) } rcu_read_unlock(); - return sel_netif_sid_slow(ifindex, sid); + return sel_netif_sid_slow(ns, ifindex, sid); } /** * sel_netif_kill - Remove an entry from the network interface table + * @ns: the network namespace * @ifindex: the network interface * * Description: @@ -220,13 +226,13 @@ int sel_netif_sid(int ifindex, u32 *sid) * table if it exists. * */ -static void sel_netif_kill(int ifindex) +static void sel_netif_kill(const struct net *ns, int ifindex) { struct sel_netif *netif; rcu_read_lock(); spin_lock_bh(&sel_netif_lock); - netif = sel_netif_find(ifindex); + netif = sel_netif_find(ns, ifindex); if (netif) sel_netif_destroy(netif); spin_unlock_bh(&sel_netif_lock); @@ -240,7 +246,7 @@ static void sel_netif_kill(int ifindex) * Remove all entries from the network interface table. * */ -static void sel_netif_flush(void) +void sel_netif_flush(void) { int idx; struct sel_netif *netif; @@ -252,25 +258,13 @@ static void sel_netif_flush(void) spin_unlock_bh(&sel_netif_lock); } -static int sel_netif_avc_callback(u32 event) -{ - if (event == AVC_CALLBACK_RESET) { - sel_netif_flush(); - synchronize_net(); - } - return 0; -} - static int sel_netif_netdev_notifier_handler(struct notifier_block *this, unsigned long event, void *ptr) { struct net_device *dev = ptr; - if (dev_net(dev) != &init_net) - return NOTIFY_DONE; - if (event == NETDEV_DOWN) - sel_netif_kill(dev->ifindex); + sel_netif_kill(dev_net(dev), dev->ifindex); return NOTIFY_DONE; } @@ -291,10 +285,6 @@ static __init int sel_netif_init(void) register_netdevice_notifier(&sel_netif_netdev_notifier); - err = avc_add_callback(sel_netif_avc_callback, AVC_CALLBACK_RESET); - if (err) - panic("avc_add_callback() failed, error %d\n", err); - return err; } diff --git a/security/selinux/netnode.c b/security/selinux/netnode.c index c5454c0477c3..bb8de9d9b429 100644 --- a/security/selinux/netnode.c +++ b/security/selinux/netnode.c @@ -281,7 +281,7 @@ int sel_netnode_sid(void *addr, u16 family, u32 *sid) * Remove all entries from the network address table. * */ -static void sel_netnode_flush(void) +void sel_netnode_flush(void) { unsigned int idx; struct sel_netnode *node, *node_tmp; @@ -298,15 +298,6 @@ static void sel_netnode_flush(void) spin_unlock_bh(&sel_netnode_lock); } -static int sel_netnode_avc_callback(u32 event) -{ - if (event == AVC_CALLBACK_RESET) { - sel_netnode_flush(); - synchronize_net(); - } - return 0; -} - static __init int sel_netnode_init(void) { int iter; @@ -320,10 +311,6 @@ static __init int sel_netnode_init(void) sel_netnode_hash[iter].size = 0; } - ret = avc_add_callback(sel_netnode_avc_callback, AVC_CALLBACK_RESET); - if (ret != 0) - panic("avc_add_callback() failed, error %d\n", ret); - return ret; } diff --git a/security/selinux/netport.c b/security/selinux/netport.c index d35379781c2c..73ac6784d091 100644 --- a/security/selinux/netport.c +++ b/security/selinux/netport.c @@ -217,7 +217,7 @@ int sel_netport_sid(u8 protocol, u16 pnum, u32 *sid) * Remove all entries from the network address table. * */ -static void sel_netport_flush(void) +void sel_netport_flush(void) { unsigned int idx; struct sel_netport *port, *port_tmp; @@ -234,15 +234,6 @@ static void sel_netport_flush(void) spin_unlock_bh(&sel_netport_lock); } -static int sel_netport_avc_callback(u32 event) -{ - if (event == AVC_CALLBACK_RESET) { - sel_netport_flush(); - synchronize_net(); - } - return 0; -} - static __init int sel_netport_init(void) { int iter; @@ -256,10 +247,6 @@ static __init int sel_netport_init(void) sel_netport_hash[iter].size = 0; } - ret = avc_add_callback(sel_netport_avc_callback, AVC_CALLBACK_RESET); - if (ret != 0) - panic("avc_add_callback() failed, error %d\n", ret); - return ret; } diff --git a/security/selinux/nlmsgtab.c b/security/selinux/nlmsgtab.c index 855e464e92ef..4c29bcc7fad2 100644 --- a/security/selinux/nlmsgtab.c +++ b/security/selinux/nlmsgtab.c @@ -17,6 +17,7 @@ #include <linux/inet_diag.h> #include <linux/xfrm.h> #include <linux/audit.h> +#include <linux/sock_diag.h> #include "flask.h" #include "av_permissions.h" @@ -78,6 +79,7 @@ static struct nlmsg_perm nlmsg_tcpdiag_perms[] = { { TCPDIAG_GETSOCK, NETLINK_TCPDIAG_SOCKET__NLMSG_READ }, { DCCPDIAG_GETSOCK, NETLINK_TCPDIAG_SOCKET__NLMSG_READ }, + { SOCK_DIAG_BY_FAMILY, NETLINK_TCPDIAG_SOCKET__NLMSG_READ }, }; static struct nlmsg_perm nlmsg_xfrm_perms[] = @@ -98,6 +100,13 @@ static struct nlmsg_perm nlmsg_xfrm_perms[] = { XFRM_MSG_FLUSHPOLICY, NETLINK_XFRM_SOCKET__NLMSG_WRITE }, { XFRM_MSG_NEWAE, NETLINK_XFRM_SOCKET__NLMSG_WRITE }, { XFRM_MSG_GETAE, NETLINK_XFRM_SOCKET__NLMSG_READ }, + { XFRM_MSG_REPORT, NETLINK_XFRM_SOCKET__NLMSG_READ }, + { XFRM_MSG_MIGRATE, NETLINK_XFRM_SOCKET__NLMSG_WRITE }, + { XFRM_MSG_NEWSADINFO, NETLINK_XFRM_SOCKET__NLMSG_READ }, + { XFRM_MSG_GETSADINFO, NETLINK_XFRM_SOCKET__NLMSG_READ }, + { XFRM_MSG_NEWSPDINFO, NETLINK_XFRM_SOCKET__NLMSG_WRITE }, + { XFRM_MSG_GETSPDINFO, NETLINK_XFRM_SOCKET__NLMSG_READ }, + { XFRM_MSG_MAPPING, NETLINK_XFRM_SOCKET__NLMSG_READ }, }; static struct nlmsg_perm nlmsg_audit_perms[] = diff --git a/security/selinux/ss/avtab.c b/security/selinux/ss/avtab.c index a3dd9faa19c0..dd7466cb2021 100644 --- a/security/selinux/ss/avtab.c +++ b/security/selinux/ss/avtab.c @@ -24,6 +24,7 @@ #include "policydb.h" static struct kmem_cache *avtab_node_cachep; +static struct kmem_cache *avtab_operation_cachep; static inline int avtab_hash(struct avtab_key *keyp, u16 mask) { @@ -37,11 +38,24 @@ avtab_insert_node(struct avtab *h, int hvalue, struct avtab_key *key, struct avtab_datum *datum) { struct avtab_node *newnode; + struct avtab_operation *ops; newnode = kmem_cache_zalloc(avtab_node_cachep, GFP_KERNEL); if (newnode == NULL) return NULL; newnode->key = *key; - newnode->datum = *datum; + + if (key->specified & AVTAB_OP) { + ops = kmem_cache_zalloc(avtab_operation_cachep, GFP_KERNEL); + if (ops == NULL) { + kmem_cache_free(avtab_node_cachep, newnode); + return NULL; + } + *ops = *(datum->u.ops); + newnode->datum.u.ops = ops; + } else { + newnode->datum.u.data = datum->u.data; + } + if (prev) { newnode->next = prev->next; prev->next = newnode; @@ -70,8 +84,11 @@ static int avtab_insert(struct avtab *h, struct avtab_key *key, struct avtab_dat if (key->source_type == cur->key.source_type && key->target_type == cur->key.target_type && key->target_class == cur->key.target_class && - (specified & cur->key.specified)) + (specified & cur->key.specified)) { + if (specified & AVTAB_OPNUM) + break; return -EEXIST; + } if (key->source_type < cur->key.source_type) break; if (key->source_type == cur->key.source_type && @@ -232,6 +249,9 @@ void avtab_destroy(struct avtab *h) while (cur) { temp = cur; cur = cur->next; + if (temp->key.specified & AVTAB_OP) + kmem_cache_free(avtab_operation_cachep, + temp->datum.u.ops); kmem_cache_free(avtab_node_cachep, temp); } h->htable[i] = NULL; @@ -320,7 +340,13 @@ static uint16_t spec_order[] = { AVTAB_AUDITALLOW, AVTAB_TRANSITION, AVTAB_CHANGE, - AVTAB_MEMBER + AVTAB_MEMBER, + AVTAB_OPNUM_ALLOWED, + AVTAB_OPNUM_AUDITALLOW, + AVTAB_OPNUM_DONTAUDIT, + AVTAB_OPTYPE_ALLOWED, + AVTAB_OPTYPE_AUDITALLOW, + AVTAB_OPTYPE_DONTAUDIT }; int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol, @@ -330,10 +356,11 @@ int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol, { __le16 buf16[4]; u16 enabled; - __le32 buf32[7]; u32 items, items2, val, vers = pol->policyvers; struct avtab_key key; struct avtab_datum datum; + struct avtab_operation ops; + __le32 buf32[ARRAY_SIZE(ops.op.perms)]; int i, rc; unsigned set; @@ -390,11 +417,15 @@ int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol, printk(KERN_ERR "SELinux: avtab: entry has both access vectors and types\n"); return -EINVAL; } + if (val & AVTAB_OP) { + printk(KERN_ERR "SELinux: avtab: entry has operations\n"); + return -EINVAL; + } for (i = 0; i < ARRAY_SIZE(spec_order); i++) { if (val & spec_order[i]) { key.specified = spec_order[i] | enabled; - datum.data = le32_to_cpu(buf32[items++]); + datum.u.data = le32_to_cpu(buf32[items++]); rc = insertf(a, &key, &datum, p); if (rc) return rc; @@ -413,7 +444,6 @@ int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol, printk(KERN_ERR "SELinux: avtab: truncated entry\n"); return rc; } - items = 0; key.source_type = le16_to_cpu(buf16[items++]); key.target_type = le16_to_cpu(buf16[items++]); @@ -437,14 +467,32 @@ int avtab_read_item(struct avtab *a, void *fp, struct policydb *pol, return -EINVAL; } - rc = next_entry(buf32, fp, sizeof(u32)); - if (rc) { - printk(KERN_ERR "SELinux: avtab: truncated entry\n"); - return rc; + if ((vers < POLICYDB_VERSION_IOCTL_OPERATIONS) + || !(key.specified & AVTAB_OP)) { + rc = next_entry(buf32, fp, sizeof(u32)); + if (rc) { + printk(KERN_ERR "SELinux: avtab: truncated entry\n"); + return rc; + } + datum.u.data = le32_to_cpu(*buf32); + } else { + memset(&ops, 0, sizeof(struct avtab_operation)); + rc = next_entry(&ops.type, fp, sizeof(u8)); + if (rc) { + printk(KERN_ERR "SELinux: avtab: truncated entry\n"); + return rc; + } + rc = next_entry(buf32, fp, sizeof(u32)*ARRAY_SIZE(ops.op.perms)); + if (rc) { + printk(KERN_ERR "SELinux: avtab: truncated entry\n"); + return rc; + } + for (i = 0; i < ARRAY_SIZE(ops.op.perms); i++) + ops.op.perms[i] = le32_to_cpu(buf32[i]); + datum.u.ops = &ops; } - datum.data = le32_to_cpu(*buf32); if ((key.specified & AVTAB_TYPE) && - !policydb_type_isvalid(pol, datum.data)) { + !policydb_type_isvalid(pol, datum.u.data)) { printk(KERN_ERR "SELinux: avtab: invalid type\n"); return -EINVAL; } @@ -504,8 +552,9 @@ bad: int avtab_write_item(struct policydb *p, struct avtab_node *cur, void *fp) { __le16 buf16[4]; - __le32 buf32[1]; + __le32 buf32[ARRAY_SIZE(cur->datum.u.ops->op.perms)]; int rc; + unsigned int i; buf16[0] = cpu_to_le16(cur->key.source_type); buf16[1] = cpu_to_le16(cur->key.target_type); @@ -514,8 +563,19 @@ int avtab_write_item(struct policydb *p, struct avtab_node *cur, void *fp) rc = put_entry(buf16, sizeof(u16), 4, fp); if (rc) return rc; - buf32[0] = cpu_to_le32(cur->datum.data); - rc = put_entry(buf32, sizeof(u32), 1, fp); + + if (cur->key.specified & AVTAB_OP) { + rc = put_entry(&cur->datum.u.ops->type, sizeof(u8), 1, fp); + if (rc) + return rc; + for (i = 0; i < ARRAY_SIZE(cur->datum.u.ops->op.perms); i++) + buf32[i] = cpu_to_le32(cur->datum.u.ops->op.perms[i]); + rc = put_entry(buf32, sizeof(u32), + ARRAY_SIZE(cur->datum.u.ops->op.perms), fp); + } else { + buf32[0] = cpu_to_le32(cur->datum.u.data); + rc = put_entry(buf32, sizeof(u32), 1, fp); + } if (rc) return rc; return 0; @@ -548,9 +608,13 @@ void avtab_cache_init(void) avtab_node_cachep = kmem_cache_create("avtab_node", sizeof(struct avtab_node), 0, SLAB_PANIC, NULL); + avtab_operation_cachep = kmem_cache_create("avtab_operation", + sizeof(struct avtab_operation), + 0, SLAB_PANIC, NULL); } void avtab_cache_destroy(void) { kmem_cache_destroy(avtab_node_cachep); + kmem_cache_destroy(avtab_operation_cachep); } diff --git a/security/selinux/ss/avtab.h b/security/selinux/ss/avtab.h index 63ce2f9e441d..97acd6fa705e 100644 --- a/security/selinux/ss/avtab.h +++ b/security/selinux/ss/avtab.h @@ -23,6 +23,8 @@ #ifndef _SS_AVTAB_H_ #define _SS_AVTAB_H_ +#include "security.h" + struct avtab_key { u16 source_type; /* source type */ u16 target_type; /* target type */ @@ -35,13 +37,34 @@ struct avtab_key { #define AVTAB_MEMBER 0x0020 #define AVTAB_CHANGE 0x0040 #define AVTAB_TYPE (AVTAB_TRANSITION | AVTAB_MEMBER | AVTAB_CHANGE) +#define AVTAB_OPNUM_ALLOWED 0x0100 +#define AVTAB_OPNUM_AUDITALLOW 0x0200 +#define AVTAB_OPNUM_DONTAUDIT 0x0400 +#define AVTAB_OPNUM (AVTAB_OPNUM_ALLOWED | \ + AVTAB_OPNUM_AUDITALLOW | \ + AVTAB_OPNUM_DONTAUDIT) +#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_OP (AVTAB_OPNUM | AVTAB_OPTYPE) #define AVTAB_ENABLED_OLD 0x80000000 /* reserved for used in cond_avtab */ #define AVTAB_ENABLED 0x8000 /* reserved for used in cond_avtab */ u16 specified; /* what field is specified */ }; +struct avtab_operation { + u8 type; + struct operation_perm op; +}; + struct avtab_datum { - u32 data; /* access vector or type value */ + union { + u32 data; /* access vector or type value */ + struct avtab_operation *ops; /* ioctl operations */ + } u; }; struct avtab_node { diff --git a/security/selinux/ss/conditional.c b/security/selinux/ss/conditional.c index 377d148e7157..16651c7a1541 100644 --- a/security/selinux/ss/conditional.c +++ b/security/selinux/ss/conditional.c @@ -15,6 +15,7 @@ #include "security.h" #include "conditional.h" +#include "services.h" /* * cond_evaluate_expr evaluates a conditional expr @@ -617,21 +618,39 @@ int cond_write_list(struct policydb *p, struct cond_node *list, void *fp) return 0; } + +void cond_compute_operation(struct avtab *ctab, struct avtab_key *key, + struct operation_decision *od) +{ + struct avtab_node *node; + + if (!ctab || !key || !od) + return; + + for (node = avtab_search_node(ctab, key); node; + node = avtab_search_node_next(node, key->specified)) { + if (node->key.specified & AVTAB_ENABLED) + services_compute_operation_num(od, node); + } + return; + +} /* Determine whether additional permissions are granted by the conditional * av table, and if so, add them to the result */ -void cond_compute_av(struct avtab *ctab, struct avtab_key *key, struct av_decision *avd) +void cond_compute_av(struct avtab *ctab, struct avtab_key *key, + struct av_decision *avd, struct operation *ops) { struct avtab_node *node; - if (!ctab || !key || !avd) + if (!ctab || !key || !avd || !ops) return; for (node = avtab_search_node(ctab, key); node; node = avtab_search_node_next(node, key->specified)) { if ((u16)(AVTAB_ALLOWED|AVTAB_ENABLED) == (node->key.specified & (AVTAB_ALLOWED|AVTAB_ENABLED))) - avd->allowed |= node->datum.data; + avd->allowed |= node->datum.u.data; if ((u16)(AVTAB_AUDITDENY|AVTAB_ENABLED) == (node->key.specified & (AVTAB_AUDITDENY|AVTAB_ENABLED))) /* Since a '0' in an auditdeny mask represents a @@ -639,10 +658,13 @@ void cond_compute_av(struct avtab *ctab, struct avtab_key *key, struct av_decisi * the '&' operand to ensure that all '0's in the mask * are retained (much unlike the allow and auditallow cases). */ - avd->auditdeny &= node->datum.data; + avd->auditdeny &= node->datum.u.data; if ((u16)(AVTAB_AUDITALLOW|AVTAB_ENABLED) == (node->key.specified & (AVTAB_AUDITALLOW|AVTAB_ENABLED))) - avd->auditallow |= node->datum.data; + avd->auditallow |= node->datum.u.data; + if ((node->key.specified & AVTAB_ENABLED) && + (node->key.specified & AVTAB_OP)) + services_compute_operation_type(ops, node); } return; } diff --git a/security/selinux/ss/conditional.h b/security/selinux/ss/conditional.h index 4d1f87466508..80ee2bb20eee 100644 --- a/security/selinux/ss/conditional.h +++ b/security/selinux/ss/conditional.h @@ -73,8 +73,10 @@ int cond_read_list(struct policydb *p, void *fp); int cond_write_bool(void *key, void *datum, void *ptr); int cond_write_list(struct policydb *p, struct cond_node *list, void *fp); -void cond_compute_av(struct avtab *ctab, struct avtab_key *key, struct av_decision *avd); - +void cond_compute_av(struct avtab *ctab, struct avtab_key *key, + struct av_decision *avd, struct operation *ops); +void cond_compute_operation(struct avtab *ctab, struct avtab_key *key, + struct operation_decision *od); int evaluate_cond_node(struct policydb *p, struct cond_node *node); #endif /* _CONDITIONAL_H_ */ diff --git a/security/selinux/ss/constraint.h b/security/selinux/ss/constraint.h index 149dda731fd3..96fd947c494b 100644 --- a/security/selinux/ss/constraint.h +++ b/security/selinux/ss/constraint.h @@ -48,6 +48,7 @@ struct constraint_expr { u32 op; /* operator */ struct ebitmap names; /* names */ + struct type_set *type_names; struct constraint_expr *next; /* next expression */ }; diff --git a/security/selinux/ss/policydb.c b/security/selinux/ss/policydb.c index bcdca73033f3..a683475e1635 100644 --- a/security/selinux/ss/policydb.c +++ b/security/selinux/ss/policydb.c @@ -143,6 +143,16 @@ static struct policydb_compat_info policydb_compat[] = { .sym_num = SYM_NUM, .ocon_num = OCON_NUM, }, + { + .version = POLICYDB_VERSION_CONSTRAINT_NAMES, + .sym_num = SYM_NUM, + .ocon_num = OCON_NUM, + }, + { + .version = POLICYDB_VERSION_IOCTL_OPERATIONS, + .sym_num = SYM_NUM, + .ocon_num = OCON_NUM, + }, }; static struct policydb_compat_info *policydb_lookup_compat(int version) @@ -613,6 +623,19 @@ static int common_destroy(void *key, void *datum, void *p) return 0; } +static void constraint_expr_destroy(struct constraint_expr *expr) +{ + if (expr) { + ebitmap_destroy(&expr->names); + if (expr->type_names) { + ebitmap_destroy(&expr->type_names->types); + ebitmap_destroy(&expr->type_names->negset); + kfree(expr->type_names); + } + kfree(expr); + } +} + static int cls_destroy(void *key, void *datum, void *p) { struct class_datum *cladatum; @@ -628,10 +651,9 @@ static int cls_destroy(void *key, void *datum, void *p) while (constraint) { e = constraint->expr; while (e) { - ebitmap_destroy(&e->names); etmp = e; e = e->next; - kfree(etmp); + constraint_expr_destroy(etmp); } ctemp = constraint; constraint = constraint->next; @@ -642,16 +664,14 @@ static int cls_destroy(void *key, void *datum, void *p) while (constraint) { e = constraint->expr; while (e) { - ebitmap_destroy(&e->names); etmp = e; e = e->next; - kfree(etmp); + constraint_expr_destroy(etmp); } ctemp = constraint; constraint = constraint->next; kfree(ctemp); } - kfree(cladatum->comkey); } kfree(datum); @@ -1156,8 +1176,34 @@ bad: return rc; } -static int read_cons_helper(struct constraint_node **nodep, int ncons, - int allowxtarget, void *fp) +static void type_set_init(struct type_set *t) +{ + ebitmap_init(&t->types); + ebitmap_init(&t->negset); +} + +static int type_set_read(struct type_set *t, void *fp) +{ + __le32 buf[1]; + int rc; + + if (ebitmap_read(&t->types, fp)) + return -EINVAL; + if (ebitmap_read(&t->negset, fp)) + return -EINVAL; + + rc = next_entry(buf, fp, sizeof(u32)); + if (rc < 0) + return -EINVAL; + t->flags = le32_to_cpu(buf[0]); + + return 0; +} + + +static int read_cons_helper(struct policydb *p, + struct constraint_node **nodep, + int ncons, int allowxtarget, void *fp) { struct constraint_node *c, *lc; struct constraint_expr *e, *le; @@ -1225,6 +1271,18 @@ static int read_cons_helper(struct constraint_node **nodep, int ncons, rc = ebitmap_read(&e->names, fp); if (rc) return rc; + if (p->policyvers >= + POLICYDB_VERSION_CONSTRAINT_NAMES) { + e->type_names = kzalloc(sizeof + (*e->type_names), + GFP_KERNEL); + if (!e->type_names) + return -ENOMEM; + type_set_init(e->type_names); + rc = type_set_read(e->type_names, fp); + if (rc) + return rc; + } break; default: return -EINVAL; @@ -1301,7 +1359,7 @@ static int class_read(struct policydb *p, struct hashtab *h, void *fp) goto bad; } - rc = read_cons_helper(&cladatum->constraints, ncons, 0, fp); + rc = read_cons_helper(p, &cladatum->constraints, ncons, 0, fp); if (rc) goto bad; @@ -1311,7 +1369,8 @@ static int class_read(struct policydb *p, struct hashtab *h, void *fp) if (rc) goto bad; ncons = le32_to_cpu(buf[0]); - rc = read_cons_helper(&cladatum->validatetrans, ncons, 1, fp); + rc = read_cons_helper(p, &cladatum->validatetrans, + ncons, 1, fp); if (rc) goto bad; } @@ -2762,6 +2821,24 @@ static int common_write(void *vkey, void *datum, void *ptr) return 0; } +static int type_set_write(struct type_set *t, void *fp) +{ + int rc; + __le32 buf[1]; + + if (ebitmap_write(&t->types, fp)) + return -EINVAL; + if (ebitmap_write(&t->negset, fp)) + return -EINVAL; + + buf[0] = cpu_to_le32(t->flags); + rc = put_entry(buf, sizeof(u32), 1, fp); + if (rc) + return -EINVAL; + + return 0; +} + static int write_cons_helper(struct policydb *p, struct constraint_node *node, void *fp) { @@ -2793,6 +2870,12 @@ static int write_cons_helper(struct policydb *p, struct constraint_node *node, rc = ebitmap_write(&e->names, fp); if (rc) return rc; + if (p->policyvers >= + POLICYDB_VERSION_CONSTRAINT_NAMES) { + rc = type_set_write(e->type_names, fp); + if (rc) + return rc; + } break; default: break; diff --git a/security/selinux/ss/policydb.h b/security/selinux/ss/policydb.h index da637471d4ce..725d5945a97e 100644 --- a/security/selinux/ss/policydb.h +++ b/security/selinux/ss/policydb.h @@ -154,6 +154,17 @@ struct cond_bool_datum { struct cond_node; /* + * type set preserves data needed to determine constraint info from + * policy source. This is not used by the kernel policy but allows + * utilities such as audit2allow to determine constraint denials. + */ +struct type_set { + struct ebitmap types; + struct ebitmap negset; + u32 flags; +}; + +/* * The configuration data includes security contexts for * initial SIDs, unlabeled file systems, TCP and UDP port numbers, * network interfaces, and nodes. This structure stores the diff --git a/security/selinux/ss/services.c b/security/selinux/ss/services.c index 18caa16de27b..88178da2b834 100644 --- a/security/selinux/ss/services.c +++ b/security/selinux/ss/services.c @@ -92,9 +92,10 @@ static int context_struct_to_string(struct context *context, char **scontext, u32 *scontext_len); static void context_struct_compute_av(struct context *scontext, - struct context *tcontext, - u16 tclass, - struct av_decision *avd); + struct context *tcontext, + u16 tclass, + struct av_decision *avd, + struct operation *ops); struct selinux_mapping { u16 value; /* policy value */ @@ -564,7 +565,8 @@ static void type_attribute_bounds_av(struct context *scontext, context_struct_compute_av(&lo_scontext, tcontext, tclass, - &lo_avd); + &lo_avd, + NULL); if ((lo_avd.allowed & avd->allowed) == avd->allowed) return; /* no masked permission */ masked = ~lo_avd.allowed & avd->allowed; @@ -579,7 +581,8 @@ static void type_attribute_bounds_av(struct context *scontext, context_struct_compute_av(scontext, &lo_tcontext, tclass, - &lo_avd); + &lo_avd, + NULL); if ((lo_avd.allowed & avd->allowed) == avd->allowed) return; /* no masked permission */ masked = ~lo_avd.allowed & avd->allowed; @@ -595,7 +598,8 @@ static void type_attribute_bounds_av(struct context *scontext, context_struct_compute_av(&lo_scontext, &lo_tcontext, tclass, - &lo_avd); + &lo_avd, + NULL); if ((lo_avd.allowed & avd->allowed) == avd->allowed) return; /* no masked permission */ masked = ~lo_avd.allowed & avd->allowed; @@ -611,14 +615,39 @@ static void type_attribute_bounds_av(struct context *scontext, } } +/* flag ioctl types that have operation permissions */ +void services_compute_operation_type( + struct operation *ops, + struct avtab_node *node) +{ + u8 type; + unsigned int i; + + if (node->key.specified & AVTAB_OPTYPE) { + /* if allowing one or more complete types */ + for (i = 0; i < ARRAY_SIZE(ops->type); i++) + ops->type[i] |= node->datum.u.ops->op.perms[i]; + } else { + /* if allowing operations within a type */ + type = node->datum.u.ops->type; + security_operation_set(ops->type, type); + } + + /* If no ioctl commands are allowed, ignore auditallow and auditdeny */ + if (node->key.specified & AVTAB_OPTYPE_ALLOWED || + node->key.specified & AVTAB_OPNUM_ALLOWED) + ops->len = 1; +} + /* - * Compute access vectors based on a context structure pair for - * the permissions in a particular class. + * Compute access vectors and operations ranges based on a context + * structure pair for the permissions in a particular class. */ static void context_struct_compute_av(struct context *scontext, - struct context *tcontext, - u16 tclass, - struct av_decision *avd) + struct context *tcontext, + u16 tclass, + struct av_decision *avd, + struct operation *ops) { struct constraint_node *constraint; struct role_allow *ra; @@ -632,6 +661,10 @@ static void context_struct_compute_av(struct context *scontext, avd->allowed = 0; avd->auditallow = 0; avd->auditdeny = 0xffffffff; + if (ops) { + memset(&ops->type, 0, sizeof(ops->type)); + ops->len = 0; + } if (unlikely(!tclass || tclass > policydb.p_classes.nprim)) { if (printk_ratelimit()) @@ -646,7 +679,7 @@ static void context_struct_compute_av(struct context *scontext, * this permission check, then use it. */ avkey.target_class = tclass; - avkey.specified = AVTAB_AV; + avkey.specified = AVTAB_AV | AVTAB_OP; sattr = flex_array_get(policydb.type_attr_map_array, scontext->type - 1); BUG_ON(!sattr); tattr = flex_array_get(policydb.type_attr_map_array, tcontext->type - 1); @@ -659,15 +692,17 @@ static void context_struct_compute_av(struct context *scontext, node; node = avtab_search_node_next(node, avkey.specified)) { if (node->key.specified == AVTAB_ALLOWED) - avd->allowed |= node->datum.data; + avd->allowed |= node->datum.u.data; else if (node->key.specified == AVTAB_AUDITALLOW) - avd->auditallow |= node->datum.data; + avd->auditallow |= node->datum.u.data; else if (node->key.specified == AVTAB_AUDITDENY) - avd->auditdeny &= node->datum.data; + avd->auditdeny &= node->datum.u.data; + else if (ops && (node->key.specified & AVTAB_OP)) + services_compute_operation_type(ops, node); } /* Check conditional av table for additional permissions */ - cond_compute_av(&policydb.te_cond_avtab, &avkey, avd); + cond_compute_av(&policydb.te_cond_avtab, &avkey, avd, ops); } } @@ -898,13 +933,138 @@ static void avd_init(struct av_decision *avd) avd->flags = 0; } +void services_compute_operation_num(struct operation_decision *od, + struct avtab_node *node) +{ + unsigned int i; + if (node->key.specified & AVTAB_OPNUM) { + if (od->type != node->datum.u.ops->type) + return; + } else { + if (!security_operation_test(node->datum.u.ops->op.perms, + od->type)) + return; + } + + if (node->key.specified == AVTAB_OPTYPE_ALLOWED) { + od->specified |= OPERATION_ALLOWED; + memset(od->allowed->perms, 0xff, + sizeof(od->allowed->perms)); + } else if (node->key.specified == AVTAB_OPTYPE_AUDITALLOW) { + od->specified |= OPERATION_AUDITALLOW; + memset(od->auditallow->perms, 0xff, + sizeof(od->auditallow->perms)); + } else if (node->key.specified == AVTAB_OPTYPE_DONTAUDIT) { + od->specified |= OPERATION_DONTAUDIT; + memset(od->dontaudit->perms, 0xff, + sizeof(od->dontaudit->perms)); + } else if (node->key.specified == AVTAB_OPNUM_ALLOWED) { + od->specified |= OPERATION_ALLOWED; + for (i = 0; i < ARRAY_SIZE(od->allowed->perms); i++) + od->allowed->perms[i] |= + node->datum.u.ops->op.perms[i]; + } else if (node->key.specified == AVTAB_OPNUM_AUDITALLOW) { + od->specified |= OPERATION_AUDITALLOW; + for (i = 0; i < ARRAY_SIZE(od->auditallow->perms); i++) + od->auditallow->perms[i] |= + node->datum.u.ops->op.perms[i]; + } else if (node->key.specified == AVTAB_OPNUM_DONTAUDIT) { + od->specified |= OPERATION_DONTAUDIT; + for (i = 0; i < ARRAY_SIZE(od->dontaudit->perms); i++) + od->dontaudit->perms[i] |= + node->datum.u.ops->op.perms[i]; + } else { + BUG(); + } +} + +void security_compute_operation(u32 ssid, + u32 tsid, + u16 orig_tclass, + u8 type, + struct operation_decision *od) +{ + u16 tclass; + struct context *scontext, *tcontext; + struct avtab_key avkey; + struct avtab_node *node; + struct ebitmap *sattr, *tattr; + struct ebitmap_node *snode, *tnode; + unsigned int i, j; + + od->type = type; + od->specified = 0; + memset(od->allowed->perms, 0, sizeof(od->allowed->perms)); + memset(od->auditallow->perms, 0, sizeof(od->auditallow->perms)); + memset(od->dontaudit->perms, 0, sizeof(od->dontaudit->perms)); + + read_lock(&policy_rwlock); + if (!ss_initialized) + goto allow; + + scontext = sidtab_search(&sidtab, ssid); + if (!scontext) { + printk(KERN_ERR "SELinux: %s: unrecognized SID %d\n", + __func__, ssid); + goto out; + } + + tcontext = sidtab_search(&sidtab, tsid); + if (!tcontext) { + printk(KERN_ERR "SELinux: %s: unrecognized SID %d\n", + __func__, tsid); + goto out; + } + + tclass = unmap_class(orig_tclass); + if (unlikely(orig_tclass && !tclass)) { + if (policydb.allow_unknown) + goto allow; + goto out; + } + + + if (unlikely(!tclass || tclass > policydb.p_classes.nprim)) { + pr_warn_ratelimited("SELinux: Invalid class %hu\n", tclass); + goto out; + } + + avkey.target_class = tclass; + avkey.specified = AVTAB_OP; + sattr = flex_array_get(policydb.type_attr_map_array, + scontext->type - 1); + BUG_ON(!sattr); + tattr = flex_array_get(policydb.type_attr_map_array, + tcontext->type - 1); + BUG_ON(!tattr); + ebitmap_for_each_positive_bit(sattr, snode, i) { + ebitmap_for_each_positive_bit(tattr, tnode, j) { + avkey.source_type = i + 1; + avkey.target_type = j + 1; + for (node = avtab_search_node(&policydb.te_avtab, &avkey); + node; + node = avtab_search_node_next(node, avkey.specified)) + services_compute_operation_num(od, node); + + cond_compute_operation(&policydb.te_cond_avtab, + &avkey, od); + } + } +out: + read_unlock(&policy_rwlock); + return; +allow: + memset(od->allowed->perms, 0xff, sizeof(od->allowed->perms)); + goto out; +} /** * security_compute_av - Compute access vector decisions. * @ssid: source security identifier * @tsid: target security identifier * @tclass: target security class * @avd: access vector decisions + * @od: operation decisions * * Compute a set of access vector decisions based on the * SID pair (@ssid, @tsid) for the permissions in @tclass. @@ -912,13 +1072,15 @@ static void avd_init(struct av_decision *avd) void security_compute_av(u32 ssid, u32 tsid, u16 orig_tclass, - struct av_decision *avd) + struct av_decision *avd, + struct operation *ops) { u16 tclass; struct context *scontext = NULL, *tcontext = NULL; read_lock(&policy_rwlock); avd_init(avd); + ops->len = 0; if (!ss_initialized) goto allow; @@ -946,7 +1108,7 @@ void security_compute_av(u32 ssid, goto allow; goto out; } - context_struct_compute_av(scontext, tcontext, tclass, avd); + context_struct_compute_av(scontext, tcontext, tclass, avd, ops); map_decision(orig_tclass, avd, policydb.allow_unknown); out: read_unlock(&policy_rwlock); @@ -992,7 +1154,7 @@ void security_compute_av_user(u32 ssid, goto out; } - context_struct_compute_av(scontext, tcontext, tclass, avd); + context_struct_compute_av(scontext, tcontext, tclass, avd, NULL); out: read_unlock(&policy_rwlock); return; @@ -1512,7 +1674,7 @@ static int security_compute_sid(u32 ssid, if (avdatum) { /* Use the type from the type transition/member/change rule. */ - newcontext.type = avdatum->data; + newcontext.type = avdatum->u.data; } /* if we have a objname this is a file trans check so check those rules */ diff --git a/security/selinux/ss/services.h b/security/selinux/ss/services.h index e8d907e903cd..569757484d05 100644 --- a/security/selinux/ss/services.h +++ b/security/selinux/ss/services.h @@ -11,5 +11,11 @@ extern struct policydb policydb; +void services_compute_operation_type(struct operation *ops, + struct avtab_node *node); + +void services_compute_operation_num(struct operation_decision *od, + struct avtab_node *node); + #endif /* _SS_SERVICES_H_ */ diff --git a/sound/core/compress_offload.c b/sound/core/compress_offload.c index 3fdf998ad057..0d430adee88d 100644 --- a/sound/core/compress_offload.c +++ b/sound/core/compress_offload.c @@ -490,9 +490,6 @@ static int snd_compress_check_input(struct snd_compr_params *params) if (params->codec.ch_in == 0 || params->codec.ch_out == 0) return -EINVAL; - if (!(params->codec.sample_rate & SNDRV_PCM_RATE_8000_192000)) - return -EINVAL; - return 0; } |